• We are sorry, but NCBI web applications do not support your browser and may not function properly. More information
Logo of ajplungPublished ArticleArchivesSubscriptionsSubmissionsContact UsAJP - Lung Cellular and Molecular PhysiologyAmerican Physiological Society
Am J Physiol Lung Cell Mol Physiol. Apr 2010; 298(4): L509–L520.
Published online Jan 22, 2010. doi:  10.1152/ajplung.00230.2009
PMCID: PMC2853346

Rac1-mediated NADPH oxidase release of O2 regulates epithelial sodium channel activity in the alveolar epithelium


We examine whether alveolar cells can control release of O2 through regulated NADPH oxidase (NOX) 2 (NOX2) activity to maintain lung fluid homeostasis. Using FACS to purify alveolar epithelial cells, we show that type 1 cells robustly express each of the critical NOX components that catalyze the production of O2 (NOX2 or gp91phox, p22phox, p67phox, p47phox, and p40phox subunits) as well as Rac1 at substantially higher levels than type 2 cells. Immunohistochemical labeling of lung tissue shows that Rac1 expression is cytoplasmic and resides near the apical surface of type 1 cells, whereas NOX2 coimmunoprecipitates with epithelial sodium channel (ENaC). Since Rac1 is a known regulator of NOX2, and hence O2 release, we tested whether inhibition or activation of Rac1 influenced ENaC activity. Indeed, 1 μM NSC23766 inhibition of Rac1 decreased O2 output in lung cells and significantly decreased ENaC activity from 0.87 ± 0.16 to 0.52 ± 0.16 [mean number of channels (N) and single-channel open probability (Po) (NPo) ± SE, n = 6; P < 0.05] in type 2 cells. NSC23766 (10 μM) decreased ENaC NPo from 1.16 ± 0.27 to 0.38 ± 0.10 (n = 6 in type 1 cells). Conversely, 10 ng/ml EGF (a known stimulator of both Rac1 and O2 release) increased ENaC NPo values in both type 1 and 2 cells. NPo values increased from 0.48 ± 0.21 to 0.91 ± 0.28 in type 2 cells (P < 0.05; n = 10). In type 1 cells, ENaC activity also significantly increased from 0.40 ± 0.15 to 0.60 ± 0.23 following EGF treatment (n = 7). Sequestering O2 using 2,2,6,6-tetramethylpiperidine-N-oxyl (TEMPO) compound prevented EGF activation of ENaC in both type 1 and 2 cells. In conclusion, we report that Rac1-mediated NOX2 activity is an important component in O2 regulation of ENaC.

Keywords: redox, NOX2, lung slice, fluorescence-activated cell sorting, single-channel patch

reactive oxygen species, such as superoxide anions (O2), are important secondary messengers involved in maintaining normal cell function. Although release of O2 typically occurs in the mitochondria (as an incidental by-product of metabolic respiration), the identification of novel NADPH oxidase (NOX) isoforms in nonphagocytic cells has piqued new interest in the study of O2 signaling. In general, NOX is responsible for transporting electrons across biological membranes to reduce molecular oxygen to O2 (reviewed in Ref. 1). Of the seven homologs of NOX characterized thus far (18), NOX2, also known as gp91phox, is the best studied member of the NOX family. In Fig. 1, we diagram the general mechanism of NOX2 activation by the small G protein, Rac1, as it is understood to occur in phagocytic and nonphagocytic cells. Essentially, NOX2 production of O2 requires Rac1 activation of cytoplasmic subunits (p47phox, p67phox, and p40phox). The translocation and assembly of these regulatory subunits to the catalytic domain, also termed NOX2, stimulates electron transfer from NADPH to oxygen. Following electron transfer, O2 are released into the extracellular space. In the schematic, we depict NOX2 in the apical membrane, near epithelial sodium channels (ENaCs), to reflect our hypothesis that Rac1-mediated NOX2 release of O2 regulates ENaC activity. Moreover, we explore the signal transduction cascade of Rac1 in lung cells, as it relates to ion channel function. Although we (12) have previously reported that O2 plays an important role in regulating lung ENaC, the signaling mechanisms responsible for altered sodium transport remain unclear. Additionally, the physiological role of nonphagocytic NOX2 in the lung has not been established.

Fig. 1.
Schematic of NADPH oxidase (NOX) complex signaling and epithelial sodium channel (ENaC) in apical membrane. NOX2 is a multiprotein enzyme complex with the catalytic subunit (NOX2) stabilized at the cellular membrane via association with p22phox subunit. ...

In the current study, we examine the role of NOX in the signal transduction cascade leading to normal sodium channel activity in the alveolar epithelium, which has not been previously described. ENaCs play a key role in maintaining alveolar fluid balance in the lung by creating the osmotic driving force needed to move water out of the air space. Although the precise relationship between O2 release and ENaC function has not been clearly defined, it is certain that both NOX and ENaC activity are reliant on small monomeric G protein signaling. Our studies described below provide a plausible link between small G protein signaling, O2 production, and ENaC regulation in alveolar cells. The interrelatedness of these signaling proteins presents a novel mechanism of lung ENaC regulation.

The alveolar epithelium is made up of alveolar type 1 and type 2 cells. Both cell types express functional ENaCs; however, our recent studies indicate that type 1 and 2 cells may differ significantly in response to oxidative stress (12). Additionally, fundamental comparisons between type 1 and 2 cells, such as differences in the level of NOX expression, have not been made in these neighboring cells. Moreover, a functional role for alveolar NOX has not been clearly described. There are, however, several correlations between changes in oxidative state and the ability of the lung to maintain salt and water homeostasis, indicating that regulated NOX output of O2 must be important in ENaC regulation. For instance, high-altitude pulmonary edema (HAPE) is a serious condition affecting the ability of the lung to exchange CO2 for oxygen due to excessive fluid accumulation in low Po2 environments. The low and limited availability of molecular oxygen slows the rate of O2 production and, presumably, alters the normal function of ENaC and other important processes related to maintaining homeostasis. Additionally, patients with chronic granulomatous disease (CGD) inherit mutations in the normal expression of NOX enzyme, are immunocompromised, and also suffer from severe pneumonia. Clearly, in CGD, the compromised ability to release O2 impacts the ability of the lung to clear fluid. Because sodium channels are the rate-limiting factor in net fluid reuptake, there must be an important relationship between O2 signaling and the ENaC regulatory pathways, particularly in the lungs.

ENaCs are located in the apical membrane of polarized cells and serve primarily to transport Na from the lumen to the interstitial space and ultimately back into the bloodstream. The net movement of Na in this direction creates the osmotic driving force needed for lung fluid clearance. The importance of ENaC in maintaining homeostasis and viability is best appreciated in ENaC knockout mice: low expression of β-ENaC significantly impairs lung fluid clearance (25), and α-ENaC knockout animals die within 40 h of birth due to an inability to clear lung fluid (14). Although it is clear that normal ENaC function is critical, the precise mechanism of sodium channel regulation remains unknown.

Recently, we (28) reported that steroid hormones, such as aldosterone and dexamethasone, regulate O2 production. This established a rudimentary connection between ENaC and O2 signaling, given that corticosteroids are the principal hormonal regulators of ENaC function. We provided a stronger link, using 2,2,6,6-tetramethylpiperidine-N-oxyl (TEMPO) compound, which scavenged cellular O2 and led to notable decreases in ENaC activity (29). Conversely, increasing local O2 levels in both A6 and primary lung cell models resulted in significant increases in single-channel measurements (12, 29). Together, these studies indicate that regulated O2 release cross talks with ENaC signal transduction pathways. In the present study, we examine NOX-mediated O2 release, following small G protein activation, as a new and important signaling mechanism that overlaps with canonical ENaC regulatory pathways.


Animal care.

Male Sprague-Dawley rats were housed with access to standard rat diet and water ad libitum. Between weeks 8 and 12, animals were anesthetized and killed for experimentation in accordance with Institutional Animal Care and Use Committee (IACUC) guidelines. All animal protocols conform to National Institutes of Health animal care and use guidelines and were approved by Emory University IACUC.

FACS of type 1 and 2 cells.

Rat lungs were lavaged with 2 mg/ml DNase and 1.5 mg/ml elastase dissolved in RPMI/HEPES solution. The lungs were then minced into small 1.5- to 2-mm pieces before filtering through a 40-μm filter. Cells were incubated with 1 mg/ml rat IgG antibody and goat anti-rat IgG magnetic beads before elution through an LC column (Miltenyi Biotec) to remove macrophages (leaving mostly type 1 and 2 cells in the eluted buffer). Eluted cells were centrifuged and resuspended in flow sort buffer (2% FBS in Ca- and Mg-free PBS) at a concentration of 1 × 106 cells/ml. Single cell suspension of type 1 and 2 cells were labeled using a 1:1,000 dilution of LysoTracker Red (Invitrogen) and Fluorescein-labeled Erythrina crista-galli Lectin (ECL; Vector Laboratories). In the flow cytometer (FACSVantage SE; Becton Dickinson), ECL-bound type 1 cells are extrapolated based on their unique forward (FSC) and fluorescent (FL1) scatter profiles following argon laser 488-nm excitation. From the same sort sample, type 2 cells with distinct FSC and FL2 scattering (on 633-nm laser excitation) were simultaneously sorted from the mix population of cells.

Western blot analysis.

Flow-sorted pneumocytes were rinsed three times with ice-cold PBS supplemented with 1× protease inhibitors. Cells were pelleted and then lysed in 600-μl RIPA buffer (150 mM NaCl, 10 mM NaPO4, pH 7.4, 0.1% SDS, 1% Nonidet P-40, 0.25% Na+-deoxycholate). All protein were electrophoresed on 7.5 or 15% acrylamide gels (where appropriate) under denaturing condition. Protein lysates were then transferred to Protran nitrocellulose membrane (Schleicher & Schuell) for immunolabeling. The membrane was blocked in TBST buffer (10 mM Tris, pH 7.5, 70 mM NaCl, and 0.1% Tween) with 5% dry milk and then incubated with 1 μg/ml NOX2 (Upstate Biotechnology), p67phox (Millipore), p47phox (Millipore), p40phox (Millipore), p22phox (Santa Cruz Biotechnology, Santa Cruz, CA), or Rac1 (Santa Cruz Biotechnology) antibodies for 1 h at room temperature. IgG-alkaline phosphatase (AP)-labeled secondary antibody (KPL, Gaithersburg, MD) was added at a concentration of 1 μg/10 ml TBST and incubated for another 1 h at room temperature. After five TBST washes, AP signal was detected using Nitro-Block chemiluminescence enhancer and CDP-Star substrate (Tropix) in combination with Kodak Image Station 2000MM and Carestream MI software.


A heterogeneous mixture of primary rat type 1 and 2 cells was used in coimmunoprecipitation (co-IP) studies. Briefly, 20 μg of rabbit anti-NOX antibody (Upstate Biotechnology) or goat anti-α-ENaC COOH-terminal antibody (Santa Cruz Biotechnology) was incubated in 3-mg lung-cell RIPA buffer-lysate overnight. Antibody-bound protein complexes were pulled down using 300 μl of IgG resin (UltraLink Immobilized Protein A Plus beads; Pierce) in 500-μl lysate volume. Following IP, standard Western blot procedures (described above) were used to detect NOX2 co-IP with α-ENaC subunit.

Lung slice preparation.

Lung slices that were 250-μm thin were prepared from rat lungs as previously described (13). Briefly, after pulmonary perfusion, low-melting-point agarose was instilled via the trachea. Excised lungs were removed en bloc, iced, and then mounted onto a vibrating microtome to prepare tissue slices. Tissue slices were transferred to 50:50 ice-cold DMEM/F-12 media (containing 10% FBS, 2 mM l-glutamine, 1 μM dexamethasone, 84 μM gentamicin, and 20 U/ml penicillin-streptomycin) and used within 6 h for patch-clamp analysis.

Confocal imaging.

Lung tissue, fixed in 2% paraformaldehyde solution, was immunolabeled with Rac1 or NOX2 antibodies (listed above). Primary antibodies were diluted 1:100, whereas secondary antibodies (as indicated in figure legends) were diluted 1:50,000 in PBS containing 1% BSA and 1× sodium azide (PBS/BSA). All washes were performed with PBS/BSA before mounting with VECTASHIELD HardSet mounting medium with 4′,6′-diamidino-2-phenylindole (DAPI; Vector Laboratories), although DAPI signals are not shown. Thin optical sections (1-μm z-stack images) were obtained using an Olympus BX61WI microscope designed for confocal fluorescence observations alongside Fluoview FV10-ASW 1.7 software.

Cell-attached patch-clamp.

Lung slices were rinsed and then incubated in patch-clamp solution containing 140 mM NaCl, 5 mM KCl, 1 mM CaCl2, 1 mM MgCl2, and 10 mM HEPES, pH 7.4. Gigaohm seals were formed on alveolar epithelial cells using a glass microelectrode back-filled with patch solution. An Axopatch-1D (Molecular Devices) amplifier interfaced through an analog-to-digital board to a personal computer collected single-channel data. Channel currents were recorded at 5 kHz and filtered at 1 kHz with a low-pass Bessel filter. We used the product of the number of channels (N) and the single-channel open probability (Po) as a measure of ENaC activity within a patch. NPo was calculated using Fetchan and Clampfit 10.1 software (Molecular Devices).

Pharmacological regulators of Rac1.

EGF and NSC23766, reagents that activate and inhibit Rac1, respectively, were purchased from Calbiochem. The superoxide dismutase mimetic TEMPO was purchased from Sigma-Aldrich.

HPLC analysis and O2 measurements.

The reaction product of O2 with dihydroethidium (DHE) yields a fluorescent product, 2-hydroxyethidium (2-OH-E+), measureable both by its fluorescent properties (excitation 520 nm/emission 610 nm) and elution time in HPLC assays. HPLC measurements of O2 release were performed in the Free Radicals in Medicine Core, Division of Cardiology, Department of Medicine, Emory University, as previously described (8). Briefly, A6 model cells were grown on 100-mm plastic dishes (Corning) and washed with Krebs-HEPES buffer, pH 7.35 (in mM: 99 NaCl, 4.69 KCl, 25 NaHCO3, 1 KH2PO4, 5.6 d-glucose, 20 Na-HEPES, 2.5 CaCl2·2H2O, and 1.2 MgSO4·7H2O) following NSC23766 or vehicle treatment. DHE (10 μM) incubated with cells at 37°C for 20 min shielded from light. The cells were harvested for HPLC analysis (using the Beckman HPLC System Gold) in 300 μl of methanol, homogenized, and then passed through a 0.22-μm filter. As an alternative to HPLC detection of O2 release, the fluorescence property of oxyethidium was additionally measured using a Synergy 4 (Biotek) microplate reader with the appropriate 520-/610-nm filter sets.

Statistical evaluation.

Statistical determinations were made by paired t-test analysis with P < 0.05 considered significant. Standard errors of the means (SE) were reported.


FACS of type 1 and 2 cells.

Detailed study of type 1 cell property and function has been limited, largely due to the technical difficulties associated with isolating a pure preparation of primary type 1 cells (described further in Refs. 3, 16). Furthermore, an ideal cell line for studying type 1 cells has not been firmly established. As such, a “type 1-like” model system, in which primary isolated type 2 cells remain in culture past day 7 (and presumably transform into a type 1 phenotype), is commonly employed in biochemical assays (4). It is not clear how representative this type 1-like model system can be of alveolar cells in vivo. In the current study, we isolate primary type 1 and 2 cells simultaneously, and with certain purity, using standard flow cytometric techniques. Essentially, type 1 and 2 cells were FACS-sorted based on their differential binding to vital dyes (previously characterized in Ref. 12) and innate differences in size (10). Because type 1 and 2 cells differentially bind vital markers fluorescein-labeled ECL and LysoTracker Red, respectively, with specificity, and because type 1 cells are significantly larger than cuboidal type 2 cells, FACS sorting is a quick and reliable approach to isolating each cell type.

Figure 2A is a grayscale image of a surfactant labeled type 2 cell (inset) and a type 1 cell labeled with RTI40 antibody. This figure illustrates the morphological differences between the two cell types that make up the alveolar surface area. Macrophages were removed before cell sorting using standard immunoadsorption techniques described above in methods. Figure 2B shows the side light scatter (SSC) and FSC dot blots of single suspension lung cells as they pass through the flow cytometer. The magnitude of the forward scatter is roughly proportional to cell size. Thus the smaller type 2 cells are located in the bottom left corner [gate 2 (G2)], and type 1 cells are in G1. Side scatter profiles show granularity; cells in the top left and right quadrants are either dead cells or neutrophils, respectively, and were not gated for further analysis. Very small particles were ignored in threshold settings. Figure 2C shows the two-color dot-plot experiments performed using compatible fluorescent dyes. We (12) have recently reported that LysoTracker Red binds to lamellar bodies found in type 2 cells (and not in type 1 cells). Additionally, we (12, 13) routinely report the use of fluorescein-labeled ECL, which fluoresces green, as a reliable marker of type 1 cell surface protein. The major population of cells in region 1 represents the type 1 cells bound to ECL, with discrete 488-nm (green) emission. Type 2 cells with bright red fluorescence are in region 2. Multiparametric analysis of cells, obtainable via FACS sorting, is necessary for the precise physical separation of type 1 and 2 cells from a mix lung cell preparation. In the cytometer, ECL-bound type 1 cells are extrapolated based on both their far forward profile and fluorescence following argon laser 488-nm excitation. From the same sort sample, type 2 cells with distinct forward scatter profiles and fluorescent scatter (on 633-nm laser excitation) are simultaneously sorted. The FACS sort rejects cells that are double positive for ECL and LysoTracker Red bound cells, which are visible in the top right quadrant of Fig. 2C. These multiparametric analyses make FACS an efficient and reliable method for isolating type 1 and 2 cells for further biochemical studies. Cells were subjected to the FACS again to verify that only a single population of cell was indeed collected in postsort analysis. For example, FL1_GFP vs. FSC histograms were generated for cells collected from G1, to verify only cells with type 1 characteristics had been collected (data not shown).

Fig. 2.
FACS of alveolar type 1 and 2 cells. A: flow cytometry sorts heterogeneous lung cells based on size and fluorescent marker binding. The inset (top left) shows an type 2 cell, which are small and cuboidal cells with basal surface area averaging 180 μm ...

Western blot analysis of FACS-isolated cells served as another important determinant of sample purity. In Fig. 3, equal cell number of type 1 and 2 cells were loaded onto polyacrylamide gels for detection of rat type 1 protein (40 kDa) and surfactant protein C (SP-C; 21 kDa). The RTI40 antibody was developed by Dobbs et al. as a specific type 1 cell marker, and SP-C is the most specific label for type 2 cells (15). As expected, FACS-sorted type 1 cells (Fig. 3A; lane 1) were immunoreactive with rat type 1 specific antibody, and type 2 cells in lane 2 were not. Flow-sorted type 2 cells, however, were immunoreactive with anti-SP-C antibody in Fig. 3B; type 1 cells in lane 1 were negative. Because there is no cross-contamination of cell types, we were then able to make novel biochemical comparisons between type 1 and 2 cells.

Fig. 3.
Western blot analysis of type 1 (T1) and 2 (T2) cell purity. A: 200,000 type 1 and 2 cells, obtained using FACS, were lysed and analyzed using standard polyacrylamide gel electrophoresis. In both A and B, lane 1 = FACS-isolated type 1 cells, and lane ...

Type 1 cells robustly express NOX2.

To establish a physiological role for NOX2 release of O2 in lung ENaC regulation, we first determined where and how much NOX2 was expressed in the alveoli. In Fig. 4A, equal number FACS-sorted type 1 and 2 cells were lysed and immunoblotted for each NOX subunit and Rac1. We found that type 1 cells express NOX2 and Rac1 at substantially higher levels than type 2 cells; the relative values are shown in Fig. 4B (quantified from 3 independent cell sorts and Western blot experiments). In each study, Rac1, NOX2, p67phox, and p47phox expression in type 1 cells are ≥8-fold greater than in type 2 cells, whereas p40phox and p21phox levels are ≥20-fold greater in type 1 cells compared with type 2. Below, we investigated the subcellular localization of Rac1 and NOX2 (catalytic domain) using intact lung tissue slices.

Fig. 4.
Type 1 cells robustly express Rac1 and NOX2. A: equal number FACS-sorted type 1 (lane 1) and type 2 (lane 2) cells were lysed and analyzed. Blots show robust Rac1 and NOX2 subunit expression in type 1 cells; NOX2 and Rac1 expression levels in equal number ...

NOX2 and Rac1 labeling in lung tissue.

Immunohistochemical labeling of lung tissue slices show predominant Rac1 and NOX2 localization in type 1 cells. The left panels in Fig. 5 are all bright field images of intact alveoli under ×40 magnification. In Fig. 5, A and B, type 1 cells were positively identified using fluorescein-labeled ECL (green cells; right); Rac1 (Fig. 5A) and NOX2 (Fig. 5B) were colabeled with ECL using anti-rabbit secondary antibody conjugated to Alexa Fluor 568 (red signal). In Fig. 5, C and D, alveolar cells were labeled with LysoTracker Red to highlight type 2 cells, with Rac1 (Fig. 5C) and NOX2 (Fig. 5D) counter-labeled using rabbit secondary antibody conjugated to Alexa Fluor 488 (pseudolabeled cyan for clarity and in no way altering data output). Regardless of the detection antibodies used, these images show that Rac1 and NOX2 are more readily detectable in type 1 cells, over type 2 cells, in intact alveoli. Seemingly, NOX2 expression is near the luminal surface of the alveoli, where, interestingly, ligand-gated Na channels are known to reside.

Fig. 5.
Immunohistochemical detection of Rac1 and NOX2 in paraformaldehyde-fixed lung slices. In each figure, the left and right are of the same cell. The left shows bright field illumination, and right shows fluorescent excitation of antibodies/vital dyes as ...

NOX2 co-IP with ENaC.

We confirmed that the catalytic domain of NOX2 and ENaC resides proximal to each other in the apical membrane using standard IP studies. In Fig. 6, lane 1, α-ENaC antibody (generated in goat against the COOH-terminal domain of α-subunit) was used to pull down ENaCs. The second lane served as a signal control, where anti-NOX2 antibody was used to IP the NOX catalytic domain. Standard Western blot analysis, using rabbit anti-NOX antibody, shows that ENaC associates with NOX2 in primary rat alveolar epithelial cells. Figure 6 is representative of four independent studies wherein NOX2 pulled down with ENaC. To be certain the signal generated is not from nonspecific binding of the conjugated secondary antibody, we subjected IP samples to Western blot analysis in which the blotted membrane was incubated with horseradish peroxidase-labeled anti-rabbit secondary antibody only, the result of which was negative (data not shown).

Fig. 6.
ENaC coimmunoprecipitates with NOX2 catalytic domain. Western blot (WB) analysis of immunoprecipitated (IP) protein, derived from rat primary alveolar epithelial cells, using either goat anti-α-ENaC subunit antibody (lane 1) or rabbit anti-NOX2 ...

A new role for Rac1-mediated NOX signaling in maintaining lung fluid homeostasis.

Monomeric G proteins have long been implicated in regulating ENaC (7), however, the precise mechanism of G protein activation remains unclear. Using cell-attached patch-clamp analysis and compounds that alter Rac1 activity, we show acute regulation of ENaC via Rac1-mediated NOX2 signaling in type 1 and 2 cells in situ.

In all patch-clamp recordings, we sample control recording periods for 5 min before drug application. As such, the same patch-clamp recording can be used as its own control. This sampling time also adequately reflects channel activity (NPo) without making assumptions about the total N present in a patch or the Po of a single channel (previously described in Ref. 22). Figure 7A shows a continuous current trace obtained from a representative type 1 cell recording that was accessed from live lung tissue. The effects of inhibiting Rac1 (using 10 μM NSC23766) can be seen immediately following drug application in this recording; Fig. 7B enlarges channel activity from the control recording period as well as during NSC23766 treatment to show details (see figure legend). In Fig. 7C, we show a summary of results from 6 independent patch-clamp recordings, both before and after Rac inhibition. On average, inhibition of Rac1 in type 1 cells significantly decreased ENaC NPo values from 1.16 ± 0.27 to 0.38 ± 0.10; P < 0.05. Rac inhibitor (1 μM) caused a similar decrease in ENaC activity in type 2 cells. Again, in Fig. 8A, we show a representative continuous recording of a cell-attached patch of a type 2 cell accessed from a lung slice preparation, with portions enlarged in Fig. 8B to highlight single-channel characteristics. Figure 8C shows that in each independent cell observation, ENaC NPo in type 2 cells decreased from 0.87 ± 0.16 to 0.52 ± 0.16 (mean NPo values ± SE, n = 6; P < 0.05) immediately following Rac inhibition.

Fig. 7.
NSC23766, a specific Rac1 inhibitor, decreases ENaC activity in type 1 cells. A: representative patch-clamp recording obtained from a type 1 cell accessed in situ. #Point of 10 μM NSC23766 application to continuous patch following control recording ...
Fig. 8.
NSC23766 inhibits ENaC in type 2 cells. A: representative recording of a type 2 cell accessed in situ. NSC23766 Rac1 inhibitor (1 μM) applied to cell-attached patch (near #) following a control recording period. B: segment of continuous recording ...

DHE measurements of O2 verify that NSC23766 inhibition indeed decreases O2 release in sodium transporting epithelia (Fig. 9). As we have alluded to above, the reaction product of O2 with DHE yields a fluorescent product, 2-OH-E+, and is measureable both by its elution time in HPLC and its fluorescent properties (excitation 520 nm/emission 610 nm) using a monochromatic plate reader. Importantly, 2-OH-E+ is a product generated specifically by O2 oxidation of DHE, whereas ethidium formation is attributed primary to pathways involving H2O2 (19). In Fig. 9, left, we assessed 2-OH-E+ levels of A6 cells treated with Rac inhibitor and found that inhibiting this small G protein decreased O2 levels ~7-fold (gray bars). We can be confident that Rac inhibition does not inadvertently alter other reactive oxygen species, such as H2O2, since the ethidium levels remained unchanged ± Rac inhibitor (black bars). After verifying the specificity of DHE as an O2 sensor, we repeated studies using primary lung cells and detected the fluorescent properties of DHE using a microplate reader in Fig. 9, right,. Once again, we confirmed that NSC23766 inhibition of Rac1 indeed decreases O2 in alveolar epithelial cells: relative light units (an alternative fluorescent measure of 2-OH-E+) were significantly higher in vehicle-treated cells. Rac inhibition significantly decreased the fluorescent detection of O2 product.

Fig. 9.
Inhibiting Rac1 activity significantly decreases O2 production. Left: NSC23766 significantly decreases the production of the O2-specific product 2-hydroxyethidium (2-OH-E+; gray bars) without altering levels of other reactive oxygen ...

EGF-stimulated Rac1 acutely regulates lung ENaC.

Several studies have shown that EGF stimulates Rac1 (5, 27) and that EGF receptor inhibition (using AG1478 compound) significantly decreases Rac1 activity with associated decreases in reactive oxygen species generation (27). In the present study, we use EGF to acutely increase Rac1 in lung slices and then measured single-channel activity. Figures 10 and and1111 show that 10 ng/ml EGF increases ENaC NPo values in both type 1 and 2 cells, respectively. Continuous traces of cell-attached patches before and after drug treatment, with enlarged portions to show detail, are presented for each cell type (Fig. 10, A and B, and Fig. 11, A and B). In type 1 cells, ENaC activity significantly increased from 0.40 ± 0.15 to 0.60 ± 0.23 following EGF treatment in 7 separate observations with P < 0.05 (Fig. 10C). In 10 independent type 2 cell observations, NPo values also increased, on average, from 0.48 ± 0.21 to 0.91 ± 0.28 (P < 0.05; Fig. 11C).

Fig. 10.
EGF acutely increases ENaC activity in type 1 cells. A: representative cell-attached patch-clamp recording obtained from type 1 cell accessed in situ. EGF (10 ng/ml) added to patch solution following control recording period (#). B: portion of trace from ...
Fig. 11.
Acute increases in ENaC following EGF treatment in type 2 cells. A: representative cell-attached patch-clamp recording of type 2 cell accessed in situ. EGF was added to bath media (as indicated near # in continuous patch), which led to immediate increases ...

In Fig. 10, D and E, and Fig. 11, D and E, we further scrutinized the effect of EGF on NPo separately in both type 1 and 2 cell recordings. In type 1 cells, the significant increase in sodium channel activity following EGF treatment is largely attributed to increases in N in the surface membrane; Po is not altered by EGF treatment in type 1 cells (Fig. 10, D and E). EGF-treated type 2 cells, however, responded with significant increases in Po and no change in N measured (Fig. 11, D and E). These differences in the effect of EGF on N in type 1 cells vs. Po in type 2 cells allude to differences in the inherent regulatory mechanisms governing sodium readsorption in neighboring type 1 and 2 cells.

In Fig. 12, the superoxide dismutase mimetic TEMPO was used to scavenge endogenous O2 before EGF treatment and single-channel analysis. By sequestering endogenous levels of oxygen radicals before EGF activation, we gain additional insight into O2 signaling mechanisms in the lung. These experiments were performed in both type 1 (Fig. 12A) and type 2 cells (Fig. 12B) and show that O2 is an important signaling molecule in EGF activation of ENaCs in the lung. Sequestering O2 before EGF treatment in both type 1 and 2 cells prevented EGF activation of ENaC NPo measured by single-channel patch-clamp recordings.

Fig. 12.
Sequestering O2 abrogates EGF-induced changes in ENaC activity. In continuous patch-clamp recordings, type 1 (A) and type 2 cells (B) were treated with 10 ng/ml EGF after 250 μM 2,2,6,6-tetramethylpiperidine-N-oxyl (TEMPO) treatment. ...


FACS reliably purifies alveolar epithelial cells.

Although type 1 cells make up the vast majority of the surface area in the alveoli, little is known about the biochemical properties of this cell type. Advancement in our understanding of type 1 cells, particularly compared with type 2 cells, has progressed slowly due to limitations in establishing an appropriate model system for type 1 cells. In the current study, we physically separated type 1 cells from all other lung cells using FACS. The flow cytometer simultaneously separated cells based on fluorescence and size profiles (described in results); as such, the type 1 cells obtained in this manner are of certain purity and number. We (17) and others (11) have recently used FACS as a useful approach to obtaining alveolar epithelial cells. In a similar approach, Gonzalez et al. (11) fluorescently labeled type 1 and 2 cells (using anti-RTI40 and RTII70 primary antibodies and the appropriate fluorescent secondary conjugated antibodies) to purify alveolar epithelial and Clara cells. Moreover, Gonzalez et al. (11) showed that 85–90% of type 1 cells obtained using FACS maintained viability, were of essential purity, and proliferated under in vitro culture conditions. In our study, we used Western blot analysis to confirm that there are not contaminating populations of cells in the sorted type 1 and 2 samples (Fig. 1). Combined, these studies show that flow cytometry is an efficient and reproducible methodology to isolate alveolar epithelial cells of interest. Coupled with our recent development of a live lung tissue preparation that allows access to intact type 1 and 2 cells for single-channel patch-clamp analysis (6), we now have the appropriate tools to compare the biochemical and biophysical properties of all native alveolar epithelial cells.

A new role for NOX2 in lung ENaC regulation.

To learn more about O2 signaling in the lung, we first compared the level of NOX expression in type 1 and 2 cells to gain a better understanding of where O2 signaling would most likely exert a physiological effect. First, we showed that the NOX2 multiunit complex, and small G protein Rac1, are expressed at greater levels in type 1 cells compared with neighboring type 2 cells. In our study, Western blot analysis was normalized to cell number, since type 1 and 2 cells are not uniform in thickness nor protein composition. Because type 1 cells are reportedly only 0.35-μm thick, and type 2 cells are cuboidal with a uniform cell thickness of 10 μm (yet considerably smaller than type 1 cells), the protein content of each cell type must vary significantly (in terms of membrane vs. cytoplasmic protein expression). Hence, comparison of NOX2 subunit expression between equal number type 1 and 2 cell is most indicative of the relative contribution each cell type makes to the immediate redox environment. Based on our Bradford protein concentration measurements, type 1 cells typically have twice the protein content of a type 2 cell. Our findings, however, were that NOX2 subunit and monomeric G protein expression was ≥8-fold higher in type 1 cells compared with type 2 cells. These findings were consistent in three independent observations of flow-sorted type 1 and 2 cells performed in parallel. Of particular interest, however, is that these findings are in line with our recent report that type 1 cells generate significantly higher amounts of reactive oxygen species than type 2 cells (12), resulting in different responses to nitric oxide and ENaC activity. Seemingly, greater NOX2 expression in the type 1 cells contributes to the overall oxidative state, and cellular responses, in the alveoli.

To confirm the finding that type 1 and 2 cells may differ significantly in NOX2 release of O2, we performed additional immunohistochemical studies. Figure 5 reveals that NOX2 catalytic subunit is predominantly localized in type 1 cells and to a lesser extent in type 2 cells. Although we could not demonstrate colocalization of NOX2 with sodium channels, co-IP studies circumvented the detection limit of native ENaC using commercially available antibodies and standard confocal microscopy. In Fig. 6, we detected NOX2 from α-ENaC-bound immune complexes (i.e., ENaC subunits were immunoprecipitated and subsequently subjected to Western blot analysis using anti-NOX2 antibody). The implication is that O2 molecules released by NOX2 could act locally and immediately on ENaC before reactive species become inactivated by antioxidants. This is an important consideration, given that reactive species are notorious for short half-lives and quick reactivity. It is important to note that we performed a related study in which NOX2 protein was immunoprecipitated from lung cell lysate and subsequently subjected to Western blot analysis using anti-α-ENaC antibody (data not shown). When IP-ing with NOX2 antibody, α-ENaC could not be detected from the NOX2-bound immune complexes. One simple interpretation is that the α-ENaC antibody works effectively to pull-down sodium channel subunits from cell lysate but may be somewhat less efficient in detecting protein in Western blots. Alternatively, however, we can make inferences to the presence and quantity of protein-protein interactions between ENaC and NOX2 in the lung. Given that NOX2 could be detected from α-ENaC immune complexes using standard Western blot (Fig. 6), we can infer that a large proportion of α-ENaC subunits associate with NOX2 in lung cells (and NOX2 is hence easily detected in Western blot). Conversely, however, given that α-ENaC was not detected from NOX2-immunoprecipitated protein complexes, it may be the case that of all the NOX2 pulled down from lung cell lysate, a smaller percentage of NOX2 associates with ENaC (compared with the percentage of ENaCs that associate with NOX2). ENaC is therefore undetected in Western blot analysis when the NOX2 immune complex is transferred to nitrocellulose membrane and probed with anti-α-ENaC antibody (data not shown).

Using single-channel patch analysis and access to type 1 and 2 cells via live tissue slice preparations, we determined the acute effects of activating or inhibiting small Rac1 GTPase activity (and hence, subsequent NOX2 function). We examined the immediate effects of Rac1-mediated O2 signaling and its impact on sodium channel function, since the immediate early effect of O2 release on ENaC activity (within minutes) may be the most important time point to scrutinize in single-channel recordings given that reactive oxygen species can be quickly inactivated by antioxidants. Inhibition of Rac1, using NSC23766, led to significant decreases in ENaC activity in both type 1 and 2 cells that were detectable within ~5 min of treatment. NSC23766 has been characterized as a specific Rac1 inhibitor (30) that works by competing with guanine exchange factor (GEF) binding in the surface groove (centering Trp56) of Rac1. In our studies, we found that 1 μM Rac1 inhibitor effectively turned off ENaC activity in type 2 cells (Fig. 8), whereas 10 μM NSC23766 was needed to observe a significant decrease in ENaC activity in type 1 cells. This may be an important observation to further investigate, as it may reveal additional fundamental differences in the two cell types that make up the alveoli. Because Rac1 is regulated, in part, by GEFs that catalyze nucleotide exchange, the observed differences in effective NSC23766 concentrations in Fig. 8 may possibly be due to differences in GEFs or GEF action in each cell type examined. Alternatively, activated Rac1 may bind to different downstream target proteins in type 1 and 2 cells that may require different concentrations of Rac1 inhibitor to elicit the same inhibitory effect on sodium channel activity. Again, these speculative explanations for the different effective concentration of NSC23766 needed to block ENaC function in type 1 and 2 cells require further investigation.

Conversely, EGF (a known stimulator of both Rac1 and O2) led to significant increases in ENaC activity in both type 1 and 2 cells, albeit via different mechanisms. In general, increases in sodium channel activity can be attributed to either N and/or Po. We report in Fig. 10, D and E, that the 50% increase in sodium transport following EGF stimulation in type 1 cells is due to an increase in the number of active channels. However, EGF stimulation of type 2 cell activity (from an average NPo value of 0.48 ± 0.21 to 0.91 ± 0.28; Fig. 11, D and E) is due to significant changes in the Po of channels residing in the membrane. Based on our current finding that primary type 1 cells have a greater propensity for releasing O2 (because they express greater levels of NOX) over type 2 cells and our previous report (12) that cultured type 1 cells release more O2 than type 2 cells, we speculate that the amount of O2 release may be accountable for the differences in N and Po effect. Seemingly, high levels of O2 release in type 1 cells led to significant increases in the number of active channels in the cell membrane (as shown in Fig. 10E). Perhaps the large NOX-mediated increase in O2 release stabilizes sodium channel subunits in the membrane. This is a plausible explanation to investigate further, given that oxidation of cysteine residues form disulfide bridges that can act to stabilize proteins. Each ENaC subunit, indeed, expresses conserved cysteine residues in the large extracellular loops of the α-, β-, and γ-subunits (9). In type 2 cells, however, changes in the gating properties (i.e., Po) were observed following EGF treatment (Fig. 11). In type 2 cells, perhaps low levels of O2 release indirectly modulate regulatory proteins that have known effects on regulating ENaC Po (reviewed and discussed in Refs. 23, 24, 26). Additional evidence for direct O2 regulation of ENaC in type 1 cells, as opposed to indirect redox signaling in type 2 cells, can be gleaned from results in Fig. 12. In Fig. 12, sequestering endogenous O2 levels led to significant decreases in ENaC activity within minutes, even in the presence of EGF stimulation, in type 1 cells only. This observation supports the contention that O2 signaling is necessary and required for direct sodium channel activity in type 1 cells, whereas type 2 cells have compensatory and/or indirect ENaC regulatory pathways that are not redox-sensitive (TEMPO and EGF cotreatment neither increased nor decreased sodium current in type 2 cells). Given that the redox state of the alveolar microenvironment can change quickly and drastically from moment to moment, having multiple regulatory pathways controlling net salt and water balance is crucial for maintaining normal lung function.

In summary, we show that robust expression of Rac1 and NOX2 subunits in type 1 cells accounts for the higher oxidative state that these cells exist in, compared with type 2 cells. Single-channel analysis of intact alveoli indicates that Rac1-mediated NOX2 release of O2 plays an important role in regulating normal sodium channel activity. Absent Rac1 activation of NOX2, ENaC activity is significantly decreased in both type 1 and 2 cells. Conversely, immediate activation of Rac1 by EGF led to associated increases in ENaC activity measured by patch-clamp analysis. This study also indicates that small G protein signaling (Rac1) and regulated NOX release of O2 cross talks with signals that ultimately merge to regulate ENaC activity. Although the study makes evident the physiological role that NOX2 plays in maintaining lung fluid homeostasis via impacting sodium channel regulation, the precise molecular mechanism by which O2 could regulate the immediate increase in ENaC activity requires further investigation. Although speculative, we allude above to the possibility that oxidized cysteines on sodium channel subunits may form disulfide linkages that would increase the number of active channels in type 1 membrane.

CFTR and correlations with O2 regulation of ENaC.

The interrelatedness of the cystic fibrosis transmembrane regulator (CFTR) and normal ENaC regulation (reviewed in Ref. 2) is evident in cystic fibrosis lung disease. Interestingly, CFTR activity influences the oxidative state of the cell. Specifically, wild-type CFTR conducts glutathione (21). As such, cells with mutations in CFTR would have a decrease in the ability to secrete glutathione and buffer antioxidants in cells. The resultant increase in reactive oxygen species could account for the observed increase in salt readsorption in cystic fibrosis disease. Our experimental evidence supports the notion that the rise in radical oxygen species, caused by a decrease in glutathione buffering, could contribute to inappropriate sodium channel activation observed in CFTR. A better understanding of redox regulation of lung ENaC can lead to possible therapeutic insight into CFTR lung disease.


This work was supported by National Institutes of Health Grant K99-HL-09222601 awarded to M. N. Helms. This research was also supported, in part, by the Parker B. Francis Foundation, Emory University's Research Committee Grant, and the Flow Cytometry Core Facility of the Emory University School of Medicine.


No conflicts of interest are declared by the author(s).


We acknowledge the technical assistance of Julie L. Self in Western blot procedures.


1. Bedard K, Krause KH. The NOX family of ROS-generating NADPH oxidases: physiology and pathophysiology. Physiol Rev 87: 245–313, 2007. [PubMed]
2. Berdiev BK, Qadri YJ, Benos DJ. Assessment of the CFTR and ENaC association. Mol Biosyst 5: 123–127, 2009. [PMC free article] [PubMed]
3. Borok Z, Liebler JM, Lubman RL, Foster MJ, Zhou B, Li X, Zabski SM, Kim KJ, Crandall ED. Na transport proteins are expressed by rat alveolar epithelial type I cells. Am J Physiol Lung Cell Mol Physiol 282: L599–L608, 2002. [PubMed]
4. Borok Z, Danto SI, Lubman RL, Cao Y, Williams MC, Crandall ED. Modulation of T1α expression with alveolar epithelial cell phenotype in vitro. Am J Physiol Lung Cell Mol Physiol 275: L155–L164, 1998 [PubMed]
5. Dise RS, Frey MR, Whitehead RH, Polk DB. Epidermal growth factor stimulates Rac activation through Src and phosphatidylinositol 3-kinase to promote colonic epithelial cell migration. Am J Physiol Gastrointest Liver Physiol 294: G276–G285, 2008. [PubMed]
6. Eaton DC, Helms MN, Koval M, Bao HF, Jain L. The contribution of epithelial sodium channels to alveolar function in health and disease. Annu Rev Physiol 71: 403–423, 2009. [PubMed]
7. Eaton DC, Malik B, Saxena NC, Al-Khalili OK, Yue G. Mechanisms of aldosterone's action on epithelial Na+ transport. J Membr Biol 184: 313–319, 2001. [PubMed]
8. Fink B, Laude K, McCann L, Doughan A, Harrison DG, Dikalov S. Detection of intracellular superoxide formation in endothelial cells and intact tissues using dihydroethidium and an HPLC-based assay. Am J Physiol Cell Physiol 287: C895–C902, 2004. [PubMed]
9. Firsov D, Robert-Nicoud M, Gruender S, Schild L, Rossier BC. Mutational analysis of cysteine-rich domains of the epithelium sodium channel (ENaC). Identification of cysteines essential for channel expression at the cell surface. J Biol Chem 274: 2743–2749, 1999. [PubMed]
10. Fuchs S, Gumbleton M.I, Schaefer U, Lehr CM. Models of the alveolar epithelium. In: Cell Culture Models of Biological Barriers: In Vitro Test Systems for Drug Absorption and Drug Delivery, edited by Lehr CM. New York: Taylor & Francis, 2002, p. 189–210
11. Gonzalez RF, Allen L, Dobbs LG. Rat alveolar type I cells proliferate, express OCT-4, and exhibit phenotypic plasticity in vitro. Am J Physiol Lung Cell Mol Physiol 297: L1045–L1055, 2009. [PMC free article] [PubMed]
12. Helms MN, Jain L, Self JL, Eaton DC. Redox regulation of epithelial sodium channels examined in alveolar type 1 and 2 cells patch-clamped in lung slice tissue. J Biol Chem 283: 22875–22883, 2008. [PMC free article] [PubMed]
13. Helms MN, Self J, Bao HF, Job LC, Jain L, Eaton DC. Dopamine activates amiloride-sensitive sodium channels in alveolar type I cells in lung slice preparations. Am J Physiol Lung Cell Mol Physiol 291: L610–L618, 2006. [PubMed]
14. Hummler E, Barker P, Gatzy J, Beermann F, Verdumo C, Schmidt A, Boucher R, Rossier BC. Early death due to defective neonatal lung liquid clearance in αENaC-deficient mice. Nat Genet 12: 325–328, 1996. [PubMed]
15. Kalina M, Mason RJ, Shannon JM. Surfactant protein C is expressed in alveolar type II cells but not in Clara cells of rat lung. Am J Respir Cell Mol Biol 6: 594–600, 1992. [PubMed]
16. Koslowski R, Barth K, Augstein A, Tschernig T, Bargsten G, Aufderheide M, Kasper M. A new rat type I-like alveolar epithelial cell line R3/1: bleomycin effects on caveolin expression. Histochem Cell Biol 121: 509–519, 2004. [PubMed]
17. Koval M, Ward C, Findley MK, Roser-Page S, Helms MN, Roman J. Extracellular matrix influences alveolar epithelial claudin expression and barrier function. Am J Respir Cell Mol Biol 42: 172–180, 2010. [PMC free article] [PubMed]
18. Lambeth JD. NOX enzymes and the biology of reactive oxygen. Nat Rev Immunol 4: 181–189, 2004. [PubMed]
19. Laurindo FR, Fernandes DC, Santos CX. Assessment of superoxide production and NADPH oxidase activity by HPLC analysis of dihydroethidium oxidation products. In: Methods in Enzymology New York: Academic Press, 2008, p. 237–260 [PubMed]
20. Li JM, Shah AM. ROS generation by nonphagocytic NADPH oxidase: potential relevance in diabetic nephropathy. J Am Soc Nephrol 14: S221–S226, 2003. [PubMed]
21. Linsdell P, Hanrahan JW. Glutathione permeability of CFTR. Am J Physiol Cell Physiol 275: C323–C326, 1998 [PubMed]
22. Marunaka Y, Eaton DC. Effects of vasopressin and cAMP on single amiloride-blockable Na channels. Am J Physiol Cell Physiol 260: C1071–C1084, 1991 [PubMed]
23. Matalon S, Lazrak A, Jain L, Eaton DC. Invited review: biophysical properties of sodium channels in lung alveolar epithelial cells. J Appl Physiol 93: 1852–1859, 2002. [PubMed]
24. Palmer LG, Garty H. Epithelial Na channels. In: The Kidney, Physiology and Physiopathology Philadelphia: Lippincott Williams & Wilkins, 2000, p. 251–276
25. Randrianarison N, Clerici C, Ferreira C, Fontayne A, Pradervand S, Fowler-Jaeger N, Hummler E, Rossier BC, Planes C. Low expression of the β-ENaC subunit impairs lung fluid clearance in the mouse. Am J Physiol Lung Cell Mol Physiol 294: L409–L416, 2008. [PubMed]
26. Rossier BC. Hormonal regulation of the epithelial sodium channel ENaC: N or Po? J Gen Physiol 120: 67–70, 2002. [PMC free article] [PubMed]
27. Seshiah PN, Weber DS, Rocic P, Valppu L, Taniyama Y, Griendling KK. Angiotensin II stimulation of NAD(P)H oxidase activity: upstream mediators. Circ Res 91: 406–413, 2002. [PubMed]
28. Yu L, Bao HF, Self JL, Eaton DC, Helms MN. Aldosterone-induced increases in superoxide production counters nitric oxide inhibition of epithelial Na channel activity in A6 distal nephron cells. Am J Physiol Renal Physiol 293: F1666–F1677, 2007. [PubMed]
29. Yu L, Helms MN, Yue Q, Eaton DC. Single-channel analysis of functional epithelial sodium channel (ENaC) stability at the apical membrane of A6 distal kidney cells. Am J Physiol Renal Physiol 295: F1519–F1527, 2008. [PMC free article] [PubMed]
30. Yuan G, Dickerson JB, Guo F, Zheng J, Zheng Y. Rational design and characterization of a Rac GTPase-specific small molecule inhibitor. Proc Natl Acad Sci USA 101: 7618–7623, 2004. [PMC free article] [PubMed]

Articles from American Journal of Physiology - Lung Cellular and Molecular Physiology are provided here courtesy of American Physiological Society
PubReader format: click here to try


Related citations in PubMed

See reviews...See all...

Cited by other articles in PMC

See all...


Recent Activity

Your browsing activity is empty.

Activity recording is turned off.

Turn recording back on

See more...