Biochemical evidence for the presence of an amiloride binding protein in adult alveolar type II pneumocytes

Modulation of sodium transport in fetal alveolar epithelial cells by oxygen and corticosterone

Regulation of active Na(+) transport across fetal distal lung epithelial cells (FDLE) by corticosterone (CST), corticotropin-releasing hormone (CRH), and oxygen tension may be crucial for postnatal adaptation. FDLE isolated from 19-day rat fetuses (term: 22 days) were grown on permeable supports to confluent monolayers (duration 3 days) in 2.5, 5, 12, or 20% O(2) with 5% CO(2)-balance N(2) and mounted in Ussing chambers for measurement of short-circuit currents (I(sc)). FDLE monolayers grown in 20% O(2) had significantly higher levels of total I(sc) and of their amiloride-sensitive (I(amil)) and ouabain-sensitive (I(ouab)) components than hypoxic cells. Values (microA/cm(2) +/- SE) for 2.5-5% O(2) and 20% O(2) were, respectively, I(sc) 5.3 +/- 0.2 vs. 8.4 +/- 0.3 (P < 0.001), I(amil) 3.4 +/- 0.2 vs. 4.3 +/- 0.2 (P < 0.01), and I(ouab) 3.4 +/- 0.6 vs. 9.1 +/- 0.6 (P < 0.001). Addition of CST but not CRH to the culture medium at any O(2) concentration increased I(amil). FDLE cells grown at 5% O(2) expressed significantly lower levels of alpha-, beta-, and gamma-epithelial Na(+) channel (ENaC), and of the alpha(1)-Na(+)-K(+)-ATPase, as determined by Western blotting. We conclude that higher O(2) concentrations increased total vectorial Na(+) transport, and the function of Na(+)-K(+)-ATPase and apical amiloride-sensitive Na(+) conductance, whereas CST only increased ENaC function.

Antisense oligonucleotides against the alpha-subunit of ENaC decrease lung epithelial cation-channel activity

Amiloride-sensitive Na+ transport by lung epithelia plays a critical role in maintaining alveolar Na+ and water balance. It has been generally assumed that Na+ transport is mediated by the amiloride-sensitive epithelial Na+ channel (ENaC) because molecular biology studies have confirmed the presence of ENaC subunits alpha, beta, and gamma in lung epithelia. However, the predominant Na+-transporting channel reported from electrophysiological studies by most laboratories is a nonselective, high-conductance channel that is very different from the highly selective, low-conductance ENaC reported in other tissues. In our laboratory, single-channel recordings from apical membrane patches from rat alveolar type II (ATII) cells in primary culture reveal a nonselective cation channel with a conductance of 20.6 +/- 1.1 pS and an Na+-to-K+ selectivity of 0.97 +/- 0.07. This channel is inhibited by submicromolar concentrations of amiloride. Thus there is some question about the relationship between the gene product observed with single-channel methods and the cloned ENaC subunits. We have employed antisense oligonucleotide methods to block the synthesis of individual ENaC subunit proteins (alpha, beta, and gamma) and determined the effect of a reduction in the subunit expression on the density of the nonselective cation channel observed in apical membrane patches on ATII cells. Treatment of ATII cells with antisense oligonucleotides inhibited the production of each subunit protein; however, single-channel recordings showed that only the antisense oligonucleotide targeting the alpha-subunit resulted in a significant decrease in the density of nonselective cation channels. Inhibition of the beta- and gamma-subunit proteins alone or together did not cause any changes in the observed channel density. There were no changes in open probability or other channel characteristics. These results support the hypothesis that the alpha-subunit of ENaC alone or in combination with some protein other than the beta- or gamma-subunit protein is the major component of lung alveolar epithelial cation channels.

Sodium channels in alveolar epithelial cells: molecular characterization, biophysical properties, and physiological significance

At birth, fetal distal lung epithelial (FDLE) cells switch from active chloride secretion to active sodium (Na+) reabsorption. Sodium ions enter the FDLE and alveolar type II (ATII) cells mainly through apical nonselective cation and Na(+)-selective channels, with conductances of 4-26 pS (picoSiemens) in FDLE and 20-25 pS in ATII cells. All these channels are inhibited by amiloride with a 50% inhibitory concentration of < 1 microM, and some are also inhibited by [N-ethyl-N-isopropyl]-2'-4'-amiloride (50% inhibitory concentration of < 1 microM). Both FDLE and ATII cells contain the alpha-, beta-, and gamma-rENaC (rat epithelial Na+ channels) mRNAs; reconstitution of an ATII cell Na(+)-channel protein into lipid bilayers revealed the presence of 25-pS Na+ single channels, inhibited by amiloride and [N-ethyl-N-isopropyl]-2'-4'-amiloride. A variety of agents, including cAMP, oxygen, glucocorticoids, and in some cases Ca2+, increased the activity and/or rENaC mRNA levels. The phenotypic properties of these channels differ from those observed in other Na(+)-absorbing epithelia. Pharmacological blockade of alveolar Na+ transport in vivo, as well as experiments with newborn alpha-rENaC knock-out mice, demonstrate the importance of active Na+ transport in the reabsorption of fluid from the fetal lung and in reabsorbing alveolar fluid in the injured adult lung. Indeed, in a number of inflammatory diseases, increased production of reactive oxygen-nitrogen intermediates, such as peroxynitrite (ONOO-), may damage ATII and FDLE Na+ channels, decrease Na+ reabsorption in vivo, and thus contribute to the formation of alveolar edema.

An aldosterone regulated chicken intestine protein with high affinity to amiloride

The pattern of chicken intestine amiloride-binding proteins was determined using the photoreactive amiloride analogue 2'-methoxy-5'-nitrobenzamil (NMBA) and a polyclonal anti-amiloride antibody. At 10(-7)M, NMBA inhibits approximately 62% of the Na+ channel activity. At this concentration the amiloride analogue labels a number of membrane proteins, and in particular a 40-45 kDa polypeptide denoted ABP40. Incorporation of NMBA into ABP40 could be prevented by a 100-fold excess of benzamil, but not by a 1000-fold excess of 5-(N-ethyl-N-isopropyl)-amiloride. Labeling of ABP40 was intense in membranes derived from salt-deprived chickens and approximately 5-fold weaker in membranes from salt-repleted animals. Because of its small size, ABP40 is not likely to be an avian Na+ channel subunit, yet this amiloride-binding protein could be involved in the response to aldosterone.

Role of actin in regulation of epithelial sodium channels by CFTR

Cystic fibrosis (CF) airway epithelia exhibit enhanced Na+ reabsorption in parallel with diminished Cl- secretion. We tested the hypothesis that actin plays a role in the regulation of a cloned epithelial Na+ channel (ENaC) by the cystic fibrosis transmembrane conductance regulator (CFTR). We found that immunopurified bovine tracheal CFTR coreconstituted into a planar lipid bilayer with alpha,beta,gamma-rat ENaC (rENaC) decreased single-channel open probability (Po) of rENaC in the presence of actin by over 60%, a significantly greater effect than was observed in the absence of actin (approximately 20%). In the presence of actin, protein kinase A plus ATP activated both CFTR and rENaC, but CFTR was activated in a sustained manner, whereas the activation of rENaC was transitory. ATP alone could also activate ENaC transiently in the presence ofactin but had no effect on CFTR. Stabilizing short actin filaments at a fixed length with gelsolin (at a ratio to actin of 2:1) produced a sustained activation of alpha,beta,gamma-rENaC in both the presence or absence of CFTR. Gelsolin alone (i.e., in the absence of actin) had no effect on the conductance or Po of either CFTR or rENaC. We have also found that short actin filaments produced their modulatory action on alpha-rENaC independent of the presence of the beta- or gamma-rENaC subunits. In contrast, CFTR did not affect any properties of the channel formed by alpha-rENaC alone, i.e., in the absence of beta- or gamma-rENaC. These results indicate that CFTR can directly downregulate single Na+ channel activity, which may account for the observed differences between Na+ transport in normal and CF-affected airway epithelia. Moreover, the presence of actin confers an enhanced modulatory ability of CFTR on Na+ channels.

Reconstitution of immunopurified alveolar type II cell Na+ channel protein into planar lipid bilayers

Low-amiloride-affinity (L-type) Na+ channels have been functionally and immunologically localized to alveolar type II (ATII) cells. Purified rabbit ATII epithelial cells were isolated by elastase digestion and solubilized with 3-[(3-cholamidopropyl)dimethyl-ammonio]-1-propanesulfonate. The solubilized proteins were purified by ion-exchange chromatography, followed by immunoaffinity purification over a column to which rabbit polyclonal antibodies raised against purified bovine renal Na+ channel protein were bound. The proteins eluted from the immunoaffinity column were assayed for specific binding of [3H]Br-benzamil and reconstituted into planar lipid bilayers. Sequential purification steps gave a final enrichment in specific [3H]Br-benzamil binding of > 2,000 compared with the homogenate. Single-channel currents of 25 pS were recorded from the immunopurified rabbit ATII cell protein. Addition of the catalytic subunit of protein kinase A (PKA) plus ATP to the presumed cytoplasmic side of the bilayer resulted in a significant increase in the single-channel open probability (Po), from 0.40 +/- 0.14 to 0.8 +/- 0.12, without altering single-channel conductance. The addition of amiloride or ethylisopropyl amiloride (EIPA) to the side opposite that in which PKA acts reduced Po with no change in single-channel conductance. Rabbit ATII Na+ channels in bilayers had an inhibitory constant for amiloride of 8 microM and 1 microM for EIPA. These data confirm the presence of L-type Na+ channels in adult mammalian ATII cells.

Conductive cation transport in apical membrane vesicles prepared from fetal lung

In order to characterise the apically-located conductive cation pathway of the type II pneumocyte, apical plasma membranes were prepared from mature fetal guinea pig lung. The protocol yielded purified apical membranes that enriched 19-fold with the brush border enzyme marker alkaline phosphatase; there was no significant contamination with other cellular membranes. A technique for imposing an outwardly-directed electrochemical Na+ gradient was used to amplify conductive 22Na+ uptake into vesicles. Uptake of 22Na+ was time-dependent, proportional to the magnitude of the Na+ gradient, specific and sensitive to the amiloride analogues phenamil and EIPA (apparent minimum Ki values of 50 nM and 10 microM, respectively, with maximum uptake inhibition of 42% and 39% at 100 microM). Uptake experiments in which the outwardly-directed Na+ gradient was replaced by outwardly-directed gradients of small monovalent cations and molecular cations were performed. The Na+/K+ permeability ratio was 1.2:1, and over the extended range of small monovalent cations, a permeability sequence of Na+ > K+ > Li+ > Rb+ > Cs+ was observed, indicating the presence of fixed negative charge in or spatially close to the pore. The molecular cation permeability sequence of NH4+ > methylamine+ > dimethylamine+ > choline+ > N-methyl-D-glucamine+ > tetraethylammonium+ > tetramethylammonium+, after transformation, gives an estimate of 8 A for the conducting pore diameter. These data are consistent with the presence in the apical membrane of fetal type II pneumocytes of a cation specific channel with low Na+ selectivity and amiloride sensitivity.

Isolation of a rat amiloride-binding protein cDNA clone: tissue distribution and regulation of expression

Sodium transport across high-resistance epithelia involves both an apical amiloride-sensitive sodium channel and the basal Na(+)-K(+)-ATPase pump. Aldosterone regulates sodium transport by increasing the sodium permeability of the sodium channels. To study further the regulation of gene expression in sodium-transporting epithelia by corticosteroids, we have cloned an amiloride-binding protein (ABP) cDNA from rat descending colon and kidney. Identical 311 nucleotide cDNAs were amplified from both rat descending colon and kidney, and the predicted amino acid sequence exhibited 83% homology to the equivalent region of the human peptide sequence. Use of this cDNA as a probe resulted in detection of a transcript in both the small and large bowel, thymus, and seminal vesicle. The latter tissue exhibited the highest level of rat ABP expression. Low to undetectable levels of rat ABP were expressed in the descending colon and kidney. No regulation of rat ABP by either class of corticosteroids was observed. Levels of ABP were low at birth and increased gradually to adult levels just before weaning in the bowel. The distribution of rat ABP is not as would be predicted for an aldosterone-induced gene and is thus unlikely to be a component of the amiloride-sensitive electrogenic sodium channel.

Molecular cloning and functional expression of different molecular forms of rat amiloride-binding proteins

The colon and lung amiloride-binding proteins were cloned from rat tissues. Two sizes of transcripts were identified. The 2.7-kb transcript codes for an 85-kDa protein, whereas the 1.2-kb transcript codes for a 25-kDa polypeptide. The 2.7-kb transcript was detected in the proximal and distal colon and in duodenum, liver, placenta and thymus. The 1.2-kb transcript was the only form present in lung and spleen, and it was also detected in placenta and colon. The short form corresponds to the 3' terminus of the longer one. It is formed by alternative transcription under the control of an internal promoter. Cells stably transfected with cDNAs encoding these two proteins were used for binding studies using [3H]phenamil, a potent blocker of the epithelial Na+ channel, derived from amiloride. Both the long and short forms of the protein bind amiloride and some of its derivatives, but they have distinct pharmacologies. The order of potency of the different amiloride derivatives to inhibit [3H]phenamil binding was phenamil (K0.5 = 10 nM) > benzamil (K0.5 = 43 nM) > amiloride (K0.5 = 1.4 microM) approximately ethylisopropylamiloride (K0.5 = 1.6 microM) for the long form, whereas it was phenamil (K0.5 = 68 nM) > amiloride (K0.5 = 3.2 microM) approximately ethylisopropylamiloride (K0.5 = 4 microM) approximately benzamil (K0.5 = 6.3 microM) for the short form. Although the binding proteins described here are distinct from the pore-forming protein of the epithelial Na+ channel, the pharmacological profile of the long form of the ABP is identical to that described previously in pig and human kidney, and similar to that expected for an epithelial Na+ channel. The pharmacological profile of the short form resembles that previously described for an amiloride-binding protein in pneumocytes. Results presented in this paper suggest that previously purified preparations showing Na+ channel activity contain different forms of the amiloride-binding protein, possibly associated with other proteins. The similarity between amiloride-binding proteins and a protein identified in seminal vesicles suggests that amiloride-binding proteins are the first members of a new family of epithelia-specific proteins.

Single-channel characteristics of a purified bovine renal amiloride-sensitive Na+ channel in planar lipid bilayers

We have purified an amiloride-inhibitable Na+ channel protein from bovine renal papillae using ion-exchange and immunoaffinity chromatography. In the present study, these purified Na+ channels were reconstituted into planar lipid bilayers, and their single-channel characteristics were studied. We observed both large- and small-conductance Na(+)-selective ion channels in planar lipid bilayers. Single-channel conductance for the large- and small-conductance channels saturated as a function of Na+ concentration. These relations could be fitted by a simple Langmuir isotherm with a Michaelis constant of 55 and 45 mM and a maximum open-state conductance of 56 or 8.4 pS, respectively. Both channels were perfectly cation selective, with a Na(+)-to-K+ permeability ratio of 6.7:1 for the large channel and 7.8:1 for the small channel, and their open single-channel current-voltage relations were linear when bathed with symmetrical Na+ solutions. The percent open time of the reconstituted large or small channels varied between 10 and 50% or 1 and 20%, respectively. After application of amiloride, both the large- and small-conductance Na+ channels were inhibited in a dose-dependent manner.