Amiloride-sensitive epithelial sodium channel (ENaC) is a major sodium channel in the lung facilitating fluid absorption. ENaC is composed of α-, β-, and γ-subunits, and the α-subunit is indispensable for ENaC function in the lung. In human lungs, the α-subunit is expressed as various splice variants. Among them, α1- and α2-subunits are two major variants with different upstream regulatory sequences that possess similar channel characteristics when tested in Xenopus oocytes. Despite the importance of α-ENaC, little was known about the relative abundance of its variants in lung epithelial cells. Furthermore, lung infection and inflammation are often accompanied by reduced α-ENaC expression, oxidative stress, and pulmonary edema. However, it was not clear how oxidative stress affects expression of α-ENaC variants. In this study, we examined relative expression levels of α-subunit variants in four human lung epithelial cell lines. We also tested the hypothesis that oxidative stress inhibits α-ENaC expression. Our results show that both α1- and α2-ENaC variants are expressed in the cells we tested, but relative abundance varies. In the two monolayer-forming cell lines, H441 and Calu-3, α2-ENaC is the predominant variant. We also show that H2O2 specifically suppresses α1- and α2-ENaC variant expression in H441 and Calu-3 cells in a dose-dependent fashion. This suppression is achieved by inhibition of their promoters and is attenuated by dexamethasone. These data demonstrate the importance of the α2-subunit variant and suggest that glucocorticoids and antioxidants may be useful in correcting infection/inflammation-induced lung fluid imbalance.
- ion channel
- splice variant
amiloride-sensitive epithelial sodium channel (ENaC) is a major sodium channel mediating transepithelial sodium movement in the lung. ENaC plays an important role in alveolar fluid clearance under physiological and pathological conditions. In patients with systemic pseudohypoaldosteronism, a genetic disease caused by mutations resulting in inactive ENaC, frequent lung infections and increased airway fluid accumulation occur (12, 15). In contrast, patients with Little's syndrome, a disorder caused by gain-of-function mutations of ENaC, display increased alveolar fluid clearance (22). In normal individuals, the role of ENaC in lung fluid clearance has been proposed during perinatal lung fluid absorption in infants and the recovery of pulmonary edema in adults (9, 20).
ENaC is composed of three subunits, α-, β-, and γ- (2). In the lung, the α-subunit is indispensable in the maintenance of fluid homeostasis. Animals with inactivated α-subunit but not other subunits die shortly after birth with fluid-filled lungs (13). Decreased expression of the α-subunit predisposes animals to pulmonary edema (8). In human lung and kidney, ENaC α-subunit mRNA is expressed as several 5′ splice variants. The most predominant variants are α1- and α2-subunits (28). Additional in-frame translation start sites in the α2-ENaC transcript predict an additional NH2-terminal sequence of 59 amino acid residues compared with the α1-ENaC protein sequence. Studies in Xenopus oocyte show that α1-ENaC and α2-ENaC proteins have similar channel characteristics (28), suggesting that both variants may contribute to ENaC channel activity in vivo. In the initial report using 5′ rapid amplification of cDNA ends (RACE), the α2-ENaC transcript accounts for ∼74% of the total clones obtained (28), indicating the relative abundance of this subunit variant. Later studies using RNase protection assay (RPA) and in situ hybridization demonstrated that both α1- and α2-subunits are expressed in the lung and lung epithelial cells (1, 24). Because expression of these variants is controlled by different promoters, which are regulated through different mechanisms (4, 24), their expression may be differentially regulated. In addition, most data currently available about α-ENaC expression are based on the α1-ENaC variant, which may or may not be the predominant form of α-ENaC expressed in human lung epithelial cells.
Oxidative stress is a condition that often occurs in the lung under certain conditions such as infection and inflammation (3, 19). In experimental animal lungs, hydrogen peroxide induces pulmonary edema and acute lung injury (11, 26). In lung epithelial cells, oxidative stress alters epithelial ion transport mechanisms (16), which could contribute to the formation of acute lung injury or hinder its recovery process. An earlier study using lung epithelial cell line A549 shows that H2O2 inhibits dexamethasone-induced activation of α-ENaC transcription (30). It was not clear how the expression of α-ENaC variants is affected by oxidative stress without added glucocorticoids in lung epithelial cells.
In the current study, we examine expression of α-ENaC variants in a number of human lung epithelial cell lines and investigate whether the expression is affected by hydrogen peroxide. Our results demonstrate that both α1- and α2-ENaC are expressed in lung epithelial cells, but α2-ENaC is the predominant variant in the two monolayer-forming cell lines we examined. We report for the first time that hydrogen peroxide suppresses baseline expression of α1- and α2-subunit variants at the transcriptional level via inhibition of their promoters. Simultaneous treatment with dexamethasone reverses this suppression. Our results suggest that fluid transport imbalance in lung parenchyma during infection and inflammation could be the result of oxidative stress. Understanding of regulations of transport molecules by oxidative stress may reveal new mechanisms applicable in the treatment of acute lung injury and other lung diseases.
MATERIALS AND METHODS
H441 cell line is derived from the pericardial fluid of a patient with papillary adenocarcinoma of the lung [American Type Culture Collection (ATCC) no. HTB-174] and was cultured in RPMI 1640 medium with 10% fetal bovine serum. Calu-3 is a lung adenocarcinoma cell line (ATCC no. HTB-55) and was cultured in MEM supplemented with 10% fetal bovine serum. BEAS-2B cells were isolated from normal human bronchial epithelium (ATCC no. CRL-9609) and were cultured in modified LHC-9 growth medium. HAE is a human lung epithelial cell line derived from alveolar cell carcinoma of a 44-yr-old female (ATCC no. CRL-2170) and was cultured in Leibovitz's L-15 medium supplemented with 10% fetal bovine serum.
RNA and cDNA preparation.
Total RNA was prepared from cultured cells using the SV Total RNA Isolation System (Promega, Madison, WI) following the manufacturer's protocol. The quality and quantity of the total RNA were examined by agarose gel electrophoresis and UV spectrophotometry. cDNA was prepared from 2 μg of total RNA using MMLV (Moloney murine leukemia virus) RT (Invitrogen, San Diego, CA). For subsequent PCR amplification, 10% of the prepared cDNA was used in each 50-μl reaction as the template.
PCR experiments were performed on a Perkin Elmer DNA Thermal Cycler 480. GoTaq Green Master Mix (Promega) was used. Primers used in PCR reactions are listed in Table 1. Amplifications were performed between 25–31 cycles. In each cycle, the reactions were denatured at 94°C for 1 min, annealed at 62°C for 40 s, and extended at 72°C for 40 s. PCR products were analyzed on 2% agarose gel. After electrophoresis, the amplified products were digitized and quantified by Aida Image Analyzer (Raytest Isotopenmeßgeräte). Each experiment was repeated three times (n = 3).
SDS-PAGE was performed using 10% gel according to the protocol previously described (4). The same amount of protein in the gel was then transblotted onto polyvinylidene difluoride membrane (Bio-Rad, Hercules, CA). After a 1-h incubation in a blocking solution containing 20 mM Tris·HCl, pH 7.5, 0.5 M NaCl, and 5% nonfat dried milk, the membrane was incubated with first antibody in an antibody buffer containing 20 mM Tris·HCl, pH 7.5, 0.5 M NaCl, 0.1% Tween 20, and 0.2% nonfat dried milk at 4°C overnight. Incubation at room temperature for 1 h was used for secondary antibodies. Three 10-min washes in washing buffer (20 mM Tris·HCl, pH 7.5, 0.5 M NaCl, and 0.1% Tween 20) were applied after each antibody reaction. Antibodies used in this study include a goat polyclonal antibody against a region in the COOH terminus of human α-ENaC, a mouse monoclonal antibody against amino acids 271–460 of human β-ENaC, a goat antibody against a peptide derived from human γ-ENaC (all purchased from Santa Cruz Biotechnology, Santa Cruz, CA), and a monoclonal antibody against human β-actin (Sigma). ECL kit (Amersham, Piscataway, NJ) was used to develop the image. Antibodies against ENaC subunits were diluted 1:500–1,000 when used, and β-actin antibody was diluted 1:10,000.
Cell survival assay.
Cell viability after treatment with H2O2 was determined by acridine orange (AO)-propidium iodide (PI) assay. The staining solution was prepared by adding 100 μl of 1 mg/ml PI (Sigma) and 100 μl of 1 mg/ml AO (Sigma) to 10 ml of PBS. Cell suspension was mixed 1:1 with the staining solution in a microcentrifuge tube. Stained cells were counted in a hemocytometer under an Olympus fluorescence microscope. PI-excluded cells were counted as viable, and PI-stained cells were counted as dead cells. The experiments were repeated 3 times.
Construction of α1-ENaC and α2-ENaC promoter-reporter gene plasmids.
The promoter-luciferase gene constructs were generated using PCR cloning from HAE cell genomic DNA and subsequent cloning of the restriction fragments from the initial clones into the pGL3Basic vector (Promega). All clones were verified by DNA sequencing. Four constructs of α1-ENaC (pGLa1931, pGLa791, pGLa332, and pGLa100) and four constructs for α2-ENaC (pGLa1421, pGLa925, pGLa284, and pGLa194) were used in this study (Fig. 2).
α1-ENaC and α2-ENaC promoter-firefly luciferase gene constructs were introduced by FuGENE HD to cultured cells at ∼80% confluence. Each promoter plasmid was cotransfected with the Renilla luciferase expression vector pRL-TK (Promega) for calibration of transfection efficiency. A 2:1 ratio of FuGENE HD (microliters) to DNA (micrograms) was used. Following transfection, cells were incubated for 2 more days before measuring luciferase activity. The Dual-Glo Luciferase Assay System (Promega) was used for measuring firefly and Renilla luciferase activities according to the manufacturer's protocol. Samples were measured on a TopCount NXT Microplate Scintillation and Luminescence Counter (Packard, Meriden, CT). The firefly luciferase-to-Renilla luciferase activity ratio was used as relative luciferase activity. The experiments were repeated three to five times.
Real-time PCR was performed using SYBR Green on ABI 7500 Fast Real-Time PCR System (Applied Biosystems). The final reaction mixture contained 40 ng of cDNA, 200 nM each primer, 10 μl of 2× SYBR Green PCR Master Mix (Applied Biosystems), and RNase-free water. All reactions were performed in triplicate. The PCR was performed with a hot-start denaturation step at 95°C for 10 min and then was carried out for 40 cycles of 95°C (15 s) and 60°C (1 min). The fluorescence was read during the reaction, allowing a continuous monitoring of the amount of PCR product. The data were normalized to internal control GAPDH mRNA. The sequences of primers used in real-time PCR are shown in Table 2.
Statistical analyses of comparisons among groups were undertaken using the one-way ANOVA test. All analyses were performed using the SigmaStat 3.1 software package (Systat Software, San Jose, CA). P values less than 0.05 are considered significant.
α-ENaC variants are expressed in lung epithelial cells.
Conventional RT-PCR was used to quantify α1- and α2-ENaC expression in four human lung epithelial cell lines. We selected an area in the α-ENaC-transcribed region for PCR amplification where α1- and α2-ENaC genes share the same 3′ sequence in exon 2 (Fig. 1). This enabled us to use a common 3′ primer for both amplifications. The 5′ primers for α1- and α2-ENaC sequences derived from areas where no other variants are transcribed, which ensures that only the variants of interest will be amplified. As expected, both PCR products appear to be the right molecular weights, confirming the presence of α1-/α2-mRNA and the accuracy of their predicted structure. In addition, the difference of amplicon lengths were designed to be minimal (α1, 337 bp; α2, 263 bp) so that the variations in amplification efficiency caused by the difference in length would be limited. Amplifications were stopped at various cycle numbers. Monitoring of band intensity of the PCR products revealed increasing changes of product level with increasing cycle numbers. This avoided reaching amplification plateaus where the ability of quantification is diminished. This strategy has been successfully used in our previous studies (5).
Among the four cell lines, α1-ENaC transcript was found to be at higher levels in HAE and BEAS-2B cells, and α2-ENaC transcript was found to be more abundant in H441 and Calu-3 cells (Fig. 3).
Differential expression of α-ENaC variants is regulated by their promoters.
We investigated α1- and α2-promoter activities in HAE and H441 cells that express high levels of α1- and α2-ENaC variants, respectively. Previous studies have shown that α1- and α2-promoters are regulated by different elements. A glucocorticoid response element (GRE) in the α1-promoter is essential for its activity (18). As expected, the 332-bp construct, containing the GRE motif, exhibited the highest promoter activity among all α1-promoter constructs (Figs. 2 and 4). The baseline activity of the α2-promoter is dependent on transcription factor Sp1, which binds to a GC box motif located at the 5′ end of the 284-bp construct (4). In both HAE and H441 cell lines, the 284-bp α2-promoter construct displayed the highest promoter activity (Figs. 2 and 4). Despite these similarities between HAE and H441 cells, a marked difference exists in α2-promoter activity. In HAE cells, the highest activities of the α1- and α2-promoters are in a narrow range. In contrast, in H441 cells, the highest α2-promoter activity is ∼5 times that of the α1-promoter (Fig. 4). This is in line with the PCR result in which α2-ENaC is the more abundant variant of α-ENaC in H441 cells. However, the similar promoter activities in HAE cells do not explain the marked difference between α1- and α2-ENaC at transcript levels in the cell (Fig. 3). This apparent inconsistency is not entirely unexpected because promoter activity assays only predict the rate of transcription initiation. How efficient transcription can be completed and how fast the transcripts degrade will vary depending on the transcript and will contribute to the intracellular pool of mRNA. Similarly, transcript level may not exactly correspond to the levels of protein expression and function because of multiple steps involved.
Expression of α-ENaC variants is suppressed by H2O2.
To investigate the effect of oxidative stress on α-ENaC expression, we treated H441 cells with H2O2. We chose to use H441 because H441 cells form monolayers on porous membranes, which might indicate a better preservation of lung epithelial characteristics and might be convenient for future functional studies (25). In addition, reasonable expression levels of both transcripts in H441 cells will allow reliable detection of changes in both transcripts.
Immunoblotting using an antibody directed at the COOH terminus of the α-subunit demonstrates two bands with similar molecular weights as reported earlier for the α1- and α2- ENaC variants (28). Both bands showed marked decline in intensity in response to the 24-h H2O2 treatment in a dose-dependent fashion (Fig. 5A). β-ENaC was not significantly changed. Interestingly, γ-ENaC protein was increased along with increasing H2O2 concentration. Subsequent cell survival test using AO-PI assay showed that survival rate of H441 cells treated with H2O2 for 24 h stayed in a narrow range between 93–96% (Fig. 5B). This result suggests that H2O2 in this concentration range does not result in significant cell death and that therefore the decline of α-ENaC protein is not associated with cell death.
H2O2 inhibits expression of α-ENaC variants at the transcriptional level through inhibition of their promoters.
The change in α-ENaC expression at the protein level could be a result of regulation at the transcriptional level. This could be made clear by measuring transcript levels using real-time quantitative RT-PCR. In addition, by using specific primers, we would be able to examine changes in α1- and α2-ENaC transcripts individually. H441 cells were treated with H2O2 at different concentrations for up to 24 h. Real-time quantitative RT-PCR results showed that H2O2 treatment led to a steady decline of α1-ENaC transcript in a dose- and time-dependent fashion (Fig. 6; time-dependent response data not shown). H2O2 at 1 mM resulted in a decline of more than 50% in α1-ENaC transcript level in 24 h. A similar trend was seen in α2-ENaC transcript. These findings support the result at the protein level and indicate that the two major α-ENaC variants are both suppressed by H2O2 in H441 cells. To ensure that this response to H2O2 treatment is not only present in H441 cells, we tested Calu-3 cells that also form monolayers on porous membranes and express many lung epithelium-specific proteins such as CFTR (14). Our results show that transcripts of both variants are decreased in a dose-dependent fashion in Calu-3 cells (Fig. 6), suggesting a response similar to H441.
We next examined α1- and α2-ENaC promoters in H441 cells treated with H2O2. In these experiments, promoter constructs with the highest promoter activities, the 332-bp α1-promoter and the 284-bp α2-promoter, were used. In response to increasing concentrations of H2O2, both promoter constructs showed dose-dependent decrease in activity. At 1 mM, H2O2 suppressed activities of the α1- and α2-promoter constructs by 65% and 60%, respectively (Fig. 7). These results suggest that the promoters play an important role in the H2O2-induced inhibition of α-ENaC expression in H441 cells.
Effects of H2O2 on ENaC subunit mRNA are markedly different.
Because the two major variants of α-ENaC are both suppressed by H2O2, we were concerned that this treatment might result in a nonspecific deteriorating condition and might elicit changes of a broad range of molecules in a degrading manner. We therefore examined the other two ENaC subunit mRNA for their expression using the same H2O2 treatment. We found that the β-subunit transcript was little affected, and the γ-subunit transcript level was increased (Fig. 8). These results, together with the protein data and cell survival data in Fig. 5, again support the notion that H2O2-induced inhibition of α-ENaC variants is not a nonspecific cell response related to cell death.
Dexamethasone attenuates H2O2-induced inhibition of α-ENaC expression.
Dexamethasone activates α1-ENaC through the GRE motif in the promoter. Because the α2-promoter is ∼0.7 kb downstream from the α1-promoter, dexamethasone may also activate α2-promoter but to a lesser extent (24). Because of this activating effect of dexamethasone, we postulated that dexamethasone might be able to limit or reverse the inhibitory effect of H2O2. To test this possibility, H441 cells were treated with H2O2, dexamethasone, or both. mRNA levels of α1- and α2-ENaC were determined by real-time RT-PCR. Both α1- and α2-ENaC transcripts were activated by dexamethasone. The α1-ENaC mRNA was increased more than six times whereas α2-ENaC mRNA was nearly doubled. When cells were treated with H2O2 in the presence of dexamethasone, rather than showing an inhibitory effect by H2O2, the net effect on α1-ENaC mRNA was an increase of ∼4 times, and the α2-ENaC mRNA was almost unchanged compared with its control (Fig. 9). These data suggest that dexamethasone activates expression of both α1- and α2-ENaC with a much higher efficiency in α1-ENaC activation. In the presence of H2O2, this activating effect is reduced but still produces a sizable net activation of α1-ENaC and a complete reversal of H2O2-induced inhibition of α2-ENaC under the experimental conditions used.
Certain cell lines such as H441 and Calu-3 are often used in cell physiology studies for ENaC channel activity because they form monolayers on porous membranes. Because both α1- and α2-ENaC transcripts code for functional ENaC α-subunits, it is not clear whether α1- or α2-subunits or both are being measured in these cell physiology experiments. Because of different regulatory sequences 5′ to the α1- and α2-ENaC transcription initiation sites (Fig. 2), α1- and α2-ENaC expression could be regulated by different mechanisms or by the same mechanisms with different efficiency. We therefore examined expression of the two major splice variants of α-ENaC in four lung epithelial cell lines. The relative abundance of α-ENaC variants was studied previously using RPA, but the results are not clear (28). This could be due to the complex RNA splice patterns and multiple overlaps of transcripts in a short region, which would make design of specific RPA probes difficult. We used conventional RT-PCR to study the relative expression levels of α-ENaC variants. The amplification of α1- and α2-ENaC products was performed both in a single tube and in separate tubes. When single tube amplification was used, two 5′ primers and one 3′ primer were added. The α1-ENaC served as the internal control for α2-ENaC and vice versa (Fig. 3). When separate tubes were used, a pair of primers for the α-ENaC variant and a pair of primers for β-actin were added into each tube. The latter was used as internal control for the α-ENaC variant. The same trend in the α1-to-α2-ratio was observed from both single tube and separate tube experiments in all four cell lines. With this experimental design and the use of four different cell lines, we believe the relative abundance is reliable. Our results reveal that α1- and α2-ENaC variants are expressed in all four cell lines tested, but the relative abundance varies. The two cell lines that form monolayers (H441 and Calu-3; Refs. 17, 29) both express higher levels of α2, whereas the two cell lines with scattered growth patterns have higher levels of α1-ENaC transcript. This may also indicate that the relative abundance in lung epithelia is cell type-specific, e.g., α2-ENaC is more abundant in Clara cells (H441) and airway epithelial cells (Calu-3; Refs. 21, 23).
The differential expression of α1- and α2-ENaC transcripts in HAE and H441 cells appears to be related to a difference in their promoter activity. The 284-bp α2-promoter fragment in H441 cells is markedly more active than all other promoter constructs (Fig. 4), which might explain higher α2-ENaC transcript level in H441 cells. The 284-bp promoter fragment contains an Sp1 binding site, which is essential for the promoter activity (Fig. 2; Ref. 4). However, the role of Sp1 appears to be important in both HAE and H441 cells because the 284-bp sequence is the most active α2-promoter fragment in both cell lines (Fig. 4). Therefore, the difference in α2-ENaC expression between HAE and H441 cells is likely due to other non-Sp1 regulatory mechanisms. In HAE cells, although the activities of the 332-bp α1-promoter and the 284-bp α2-promoter are similar (Fig. 4), longer α1-promoter constructs (332-bp and longer) present much higher promoter activities than most of the α2-promoter constructs. The higher activities of the longer promoter fragments may represent true α1-promoter activities in vivo, contributing to the higher α1-ENaC expression in HAE. The α1-ENaC transcript is also more abundant in BEAS-2B cells, but the difference between α1- and α2-ENaC transcript levels is not as great as that in HAE. Therefore, the very high ratio of α1-to-α2-ENaC transcript in HAE cells could be unique because it is different from all three other cell lines examined. The high α1-to-α2-ENaC transcript ratio in HAE may not be entirely the result of a more active α1-promoter. Other mechanisms, such as transcript stability, may also contribute to the high α1- or low α2-ENaC transcript level in HAE cells.
Oxidative stress is a condition present in the lung under certain physiological and pathological conditions. Some of these conditions such as infection may trigger pulmonary edema. ENaC is the major epithelial sodium channel facilitating fluid absorption in the lung. Insufficient ENaC function may result in reduced lung fluid absorption contributing to the formation and delayed resolution of pulmonary edema. Previous studies have shown that α-ENaC expression is suppressed during lung infection and inflammation (6, 7). It is, however, not clear how α-ENaC variants are regulated under these conditions. Because infection and inflammation induce oxidative stress (3, 19), we suspected that H2O2 may inhibit α-subunit expression in lung epithelial cells. It was previously published that H2O2 inhibits glucocorticoid-dependent transcription of α-ENaC in A549 cells. It appears that H2O2 by itself does not affect α-ENaC promoter activity (30). It was not clear whether H2O2 regulates α-ENaC variants at the promoter, mRNA, and protein levels in other human lung epithelial cells without added glucocorticoids.
Our results show that H2O2 suppresses α-ENaC expression at the protein level (Fig. 5A). The antibody was a goat polyclonal antibody developed against the COOH terminus of human α-ENaC. Therefore, it should react with both α1- and α2-ENaC variants. Because α1- and α2-ENaC proteins are of different molecular weights, these variants should appear as two bands, as shown by Thomas et al. (28) using cDNAs in in vitro expression experiments, or appear as multiple bands due to multiple translation initiation sites at the 5′ end. The immunoblotting results demonstrate a ∼85-kDa band and a ∼70-kDa band, and both are suppressed by H2O2. Because the antibody recognizes a COOH terminal sequence, the identity of these bands would be difficult to determine. Based on the facts that α1- and α2-ENaC variants are more abundantly expressed in the lung, both are expressed in H441 cells, and mRNA of both variants are suppressed by H2O2, these bands are likely α1- and α2-ENaC variants. In H441 cells, α2- is the predominant form at the mRNA level. Although it is likely that the top band is α2-ENaC because of its higher molecular weight and higher band density in the control lane, there is also a possibility that this band represents both α1- and α2-ENaC, and the lower band represents another variant or an unknown protein. This is because if α1-ENaC was to be posttranslationally modified, its molecular weight may increase and become very close to that of α2. The best approach to clarify this issue would be to develop antibodies recognizing specifically individual variants.
At the mRNA level, H2O2 inhibits expression of both α1- and α2-ENaC variants (Fig. 6). This inhibition appears to be the result of suppressed promoter activities (Fig. 7). By using both H441 and Calu-3 cells, we show that this inhibition is present in more than one cell line with well-preserved lung epithelial characteristics. H2O2-induced changes in α-subunit variants were not detected in HAE cells (32), suggesting that the response to H2O2 could be cell line-dependent. The finding that H2O2 inhibits α-subunit expression in H441 and Calu-3 cells will help us in the evaluation of cell physiology data derived from H441 and Calu-3 monolayers under oxidative stress.
Severe oxidative stress may cause oxidation of a broad range of macromolecules and result in cell death. To ensure that H2O2-induced suppression of α-subunit expression is not associated with cell death and general degradation of macromolecules, we examined cell survival and expression of other ENaC subunits under the same conditions. In H441 cells, only very small fluctuations in cell survival rate were seen when treated with H2O2 up to 1.5 mM. The β-subunit was not significantly affected by H2O2 treatment at 1 mM, and the γ-subunit transcript was increased in a dose-dependent fashion. Compared with the α1- and α2-ENaC transcripts, which were suppressed by 47% and 54%, respectively, when the cells were treated with 0.5 mM H2O2, these results suggest a specific inhibition of α1- and α2-ENaC variants by H2O2 not associated with cell death.
Glucocorticoids activate ENaC expression in lung epithelial cells and are used in certain types of pulmonary edema (10, 27, 31). Glucocorticoids activate gene expression through its receptor (GR), which translocates to nuclei upon ligand binding and activates promoters often through the conserved GRE. A GRE motif is present in the 332-bp α1-ENaC promoter fragment, which is the most active α1-promoter fragment tested in this study (Fig. 4). We thought glucocorticoid may reverse H2O2-induced inhibition of α-subunit transcription because of its activating effect on the promoter. As expected, cotreatment of H441 cells with dexamethasone and H2O2 resulted in a net activation of α1-ENaC transcription and cancellation of the inhibitory effect by H2O2 on α2-ENaC transcription. The stronger effect of dexamethasone on α1-promoter could be explained by the proximal location of the GRE motif in the α1-promoter compared with its location in the α2-promoter (Fig. 2). These results indicate that pulmonary edema as a result of oxidative stress could be prevented or treated with glucocorticoids in which ENaC-mediated fluid absorption would be maintained or enhanced. However, because of other regulatory effects, therapy using glucocorticoids has to be considered weighing all clinical indications and side effects.
Our results from this study demonstrate that different α-subunit variants are expressed in human lung epithelial cells and that the α2-ENaC variant is the more abundant variant in H441 and Calu-3 cells. Because α1- and α2-subunits are both functional proteins and are regulated differently, activation of either of them would effectively increase α-ENaC activity. Our results for the first time provide evidence that oxidative stress suppresses α-subunit expression in H441 and Calu-3 cells. This finding provides a possible explanation why α-ENaC is suppressed in the lung during infection and inflammation (6, 7). In addition to the effect of glucocorticoids in pulmonary edema via ENaC activation, our results also suggest that correction of oxidative stress in tissues may improve alveolar fluid balance and prevent the occurrence or facilitate clearance of pulmonary edema.
This work was supported by a Department of Veterans Affairs Merit Review Grant.
The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.
- Copyright © 2007 the American Physiological Society