Fetal Lung Epithelial Ion Channels Relocate in the Cell Membrane During Late Gestation
Article first published online: 1 AUG 2011
Copyright © 2011 Wiley-Liss, Inc.
The Anatomical Record
Volume 294, Issue 9, pages 1461–1471, September 2011
How to Cite
Beard, L. L., Li, T., Hu, Y. and Folkesson, H. G. (2011), Fetal Lung Epithelial Ion Channels Relocate in the Cell Membrane During Late Gestation. Anat Rec, 294: 1461–1471. doi: 10.1002/ar.21363
- Issue published online: 17 AUG 2011
- Article first published online: 1 AUG 2011
- Manuscript Accepted: 22 OCT 2010
- Manuscript Revised: 8 SEP 2010
- Manuscript Received: 8 JUN 2010
- March of Dimes Birth Defects Foundation. Grant Number: 6-FY03-64
- Cl transport;
- distal lung fluid absorption;
- fetal lung development;
- Na transport
Near the end of gestation, the direction of ion and fluid flow across the alveolar epithelium rapidly changes from secretion to absorption. Thus, the relative cell membrane location of epithelial Na channels (ENaCs) and cystic fibrosis transmembrane regulator (CFTR) Cl channels during late fetal lung development and after maternal interleukin-1β (IL-1β) pretreatment was the focus of our study. Western blot analysis after sucrose gradient separation of caveolin-1-(Cav-1)-rich membrane regions (CRR) and Cav-1-poor membrane (non-CRR) revealed primary CRR ENaC localization at gestation day (GD) 61 in guinea pigs. Correlating with the natural induction of distal lung fluid absorption, ENaC appeared in the non-CRR cell membrane regions at GD68. Conversely, CFTR was present in the non-CRR cell membrane regions at GD61 and in the CRRs at GD68. IL-1β-induced conversion to distal lung fluid absorption at GD61 was associated with ENaC non-CRR presence and CFTR CRR presence, suggesting that relative ENaC and CFTR locations induced distal lung fluid absorption and decreased fluid secretion. Instilling fetal lungs with the CRR-disrupting agent methyl-β-cyclodextrin resulted in the conversion from lung fluid secretion to absorption and ENaC non-CRR presence at GD61. Coimmunoprecipitation of Cav-1 with α- and β-ENaC demonstrated reduced coimmunoprecipitation with increased GD and after IL-1β pretreatment. On the other hand, coimmunoprecipitation of Cav-1 with CFTR demonstrated increased coimmunoprecipitation with increasing GD and after IL-1β pretreatment. This concept may provide novel molecular mechanisms for the rapid transition from fetal distal lung fluid secretion to absorption in near-term lungs. Anat Rec, 2011. © 2011 Wiley-Liss, Inc.
Preterm infants may suffer from infections and respiratory distress syndrome (RDS); both are important causes of preterm mortality. Experimental data suggest that cytokines, such as interleukin-1 (IL-1), released during infections and RDS, may accelerate lung maturation (Bry et al., 1997; Willet et al., 2002; Ye et al., 2004). Amniotic fluid IL-1α increases surfactant protein A and B mRNA (Bry et al., 1997) and alters postnatal lung adaptation to air-breathing life without eliciting systemic inflammation (Willet et al., 2002). IL-1β stimulates lung maturation via maternal hypothalamus–pituitary–adrenal gland axis activation by increasing fetal cortisol synthesis and affecting β-adrenoceptor regulation of preterm distal lung fluid absorption (Ye et al., 2004).
In fluid-absorbing fetal and adult lungs, Na absorption via apical Na channels (Hummler et al., 1996; Matalon et al., 2002; Matthay et al., 2002) and basolateral Na,K-ATPases (Ingbar et al., 1996) generates the driving force for fluid absorption (Olver et al., 1986; Finley et al., 1998). It has been suggested that Na transport by epithelial Na channels (ENaCs) requires Cl transport (i.e., the cystic fibrosis transmembrane regulator (CFTR) and/or the Na,2Cl,K cotransporter) (Jiang et al., 1998; Reddy et al., 1999; Fang et al., 2002). Regulation of Na transport by β-adrenoceptor activation and/or cortisol stimulation can occur by at least two mechanisms, increasing ENaC open probability (Po) and/or ENaC membrane density (Walters and Olver, 1978; Olver et al., 1986; Finley et al., 1998; Norlin and Folkesson, 2001; Matalon et al., 2002; Matthay et al., 2002; Ye et al., 2004). Changes in Po have been studied by patch-clamp techniques (Matalon et al., 2002), whereas mechanisms responsible for changed membrane ENaC density have been less studied. It has been shown both in the heart and the lungs that β-adrenoceptor stimulation of Na transport may result from an increased number of functional membrane Na channels (Lu et al., 1999; Matalon et al., 2002; Matthay et al., 2002; Yarbrough et al., 2002).
Cellular location of Na and Cl channels during fetal lung development and after IL-1β pretreatment was the focus of this study. We propose that a large fraction of functional ENaCs comes from caveolin-1-(Cav-1)-rich membrane regions (CRR) where channel proteins are stored until mobilized near term or by IL-1β-induced cortisol and/or β-adrenoceptor activation by labor-induced plasma epinephrine. We also propose that as term comes closer more CFTR Cl channels can be found within the Cav-1-poor membrane regions (non-CRR). A membrane integral component is the so-called caveoli (present within the CRRs), which are small flask-shaped, nonclathrin-coated membrane invaginations in epithelial, for example, alveolar epithelium (Newman et al., 1999), and nonepithelial, for example, endothelium (Lisanti et al., 1994; Guthmann et al., 2004) cells. Caveoli can internalize molecules by potocytosis (Anderson et al., 1992) and can function as storage and signaling membrane regions (Lisanti et al., 1994; Yarbrough et al., 2002). Cytoplasmic surfaces of caveoli are coated with Cav-1, a 21- to 24-kDa integral membrane protein, a protein that can be used as a CRR marker (Lisanti et al., 1994; Yarbrough et al., 2002). To test our hypotheses, we designed our studies to determine (1) if distal lung fluid absorption at GD61 and/or GD68 correlated to relative changes in ENaC and/or CFTR membrane location during normal fetal lung development, (2) if IL-1β-induced/stimulated distal lung fluid absorption correlated to increased ENaC and/or CFTR mobilization from prestimulation ENaC and CFTR cellular locations, (3) if disrupting CRR with methyl-β-cyclodextrin (MβCD) affected fetal distal lung fluid absorption and ENaC location at GD61 and/or GD68, and (4) if coimmunoprecipitation of Cav-1 with α-, β-ENaC, and CFTR changed with development and after IL-1β pretreatment.
MATERIALS AND METHODS
Preterm Dunkin-Hartley guinea pigs (Hilltop Lab Animals, Scottdale, PA) were used (N = 76 delivered from 30 litters). Timed-pregnant guinea pigs (N = 30) were kept on a 12:12-hr day–night rhythm with free access to food (Standard guinea pig chow; Purina; Copley Feed, Copley, OH) and tap water. The Institutional Animal Care and Use Committee at the Northeastern Ohio Universities Colleges of Medicine and Pharmacy has reviewed and approved this study.
The IL-1β pretreatment solution was prepared by dissolving 10 μg rat recombinant IL-1β (Sigma, St. Louis, MO) in 10 mL 0.9% NaCl containing 0.1% BSA. Dissolved IL-1β was aliquoted into tubes each containing 500 ng and stored frozen at −20°C until used. Timed-pregnant guinea pigs of GD59 (N = 8) and GD66 (N = 8) were injected subcutaneously in the dorsal neck once daily with 250 ng IL-β/kg body wt for 3 days. Control timed-pregnant guinea pigs (N = 8 at each GD) were given 0.9% NaCl injections. Surgery, fluid absorption studies, and lung harvest were carried out on the morning of the third pretreatment day.
Abdominal Hysterotomy and Fetal Surgery
Timed-pregnant guinea pigs were anesthetized by intraperitoneal (i.p.) heparinized (1,000 U; Elkins-Sinn, Cherry Hill, NJ) pentobarbital sodium injection (120 mg/kg body wt; Abbott Laboratories, Chicago, IL) and euthanized by direct intracardiac injection of 60 mg pentobarbital sodium. A laparotomy was rapidly done and fetuses were carefully delivered. The umbilical cord was ligated to prevent bleeding.
Fetuses were euthanized by i.p. heparinized (500 U) pentobarbital sodium (12 mg). The timed-pregnant guinea pig fetuses were immediately weighed. Fetuses at GD61 weighed 70 ± 10 g, whereas GD68 fetuses weighed 98 ± 23 g, weights similar to previously published fetal weights for these gestation ages in guinea pigs (Lechner and Banchero, 1982; Ye et al., 2004). After euthanasia, an endotracheal tube (PE-190; Clay Adams, Becton Dickinson, Parsippany, NJ) was inserted through a tracheotomy. The fetuses were immediately connected to a constant O2 flow (O2 fraction 1.0; Praxair, Akron, OH), and the lungs were expanded by adjusting the O2 flow to a constant positive airway pressure (CPAP) of 5 cm H2O. The entire surgical procedure after euthanasia required 5 min. The fetuses were placed between two heating pads to maintain body temperature. A temperature probe measured body temperature and heating was adjusted to maintain the temperature at 37–38°C. Airway pressure was continuously monitored by calibrated pressure transducers, analog-to-digital converters, and amplifiers (ADInstruments, Grand Junction, CO). Distal lung fluid absorption was measured as described below.
Distal Lung Fluid Absorption and Secretion
After connection to CPAP, a 5% albumin instillate (10 mL/kg body wt; prepared as 50 mg/mL bovine serum albumin (Calbiochem-Novabiochem, La Jolla, CA) in 0.9% NaCl) was instilled into the lungs through the endotracheal tube as follows. First, the fetuses were briefly disconnected from the CPAP circuit and the lungs were deflated by gently aspirating residual air with the instillation syringe. The instillate was instilled into the lungs and withdrawn again. This procedure was repeated four times to allow thorough and adequate mixing of instillate and pre-existing fetal lung fluid, and then the fluid was finally instilled and the CPAP was reinitiated and maintained for the 1-hr study. A 0.1-mL sample of the instillate–lung fluid mixture (initial solution) was retained in the syringe for protein measurement. After 1 hr, the lungs and heart were carefully removed en bloc through a midline sternotomy, and a sample of the remaining lung fluid was collected. The lungs were after fluid collection snap frozen in liquid nitrogen and stored at −80°C until used for further analyses. Protein concentrations in instillates, initial, and final solutions were determined by the Lowry method (Lowry et al., 1951) adapted for microtiter plates.
Distal lung fluid absorption or fluid secretion was calculated from the change in protein concentration over 1 hr. This mass balance technique is possible because the lung epithelium is relatively impermeable to large molecules, such as albumin (MW 67,000). Therefore, water movement (absorption or secretion) will change air space protein concentrations. Because fetal lungs are fluid filled in utero (Bland and Nielson, 1992), we expected that a certain fluid volume would still be present in the lungs. This fluid is virtually free of protein and will not add protein to the instilled albumin concentration. In contrast, it will dilute the instilled protein concentration and influence distal lung fluid absorption and fluid secretion calculations differently depending on fluid volume present at different developmental stages. Distal lung fluid absorption (AFC) or fluid secretion (AFS) was calculated from the following equations:
where Vinstilled, Vinitial, and Vfinal are volumes of instillate, initial, and final solutions, respectively. Cinstilled, Cinitial, and Cfinal are protein concentrations in corresponding solutions.
Selection of Gestation Ages for the Studies
Histological examination of lung tissue from a variety of newborn mammals has led to the conclusion that lung development parallels overall level of development at birth (Engel, 1953). This view has gained support from morphometric and morphological analyses of lung growth in rats (Burri, 1974) and guinea pigs (Lechner and Banchero, 1982). Thus, we obtained prepartum guinea pigs at GD61 and GD68, average full-term gestation = 69 days. At GD68, the lungs are near-fully mature, because they can survive without ventilatory support (Baines et al., 2000). As human full-term gestation is 270 days, a direct extrapolation of these two gestation lengths for the guinea pig would be for the GD61 guinea pig fetuses around 240 days in the human situation and for the GD68 guinea pig fetuses around 265 days in the human situation. Thus, in both the human situation and in the guinea pig, the fetuses would be in the third trimester.
Guinea pig fetuses at GD61 and GD68 (term = GD69) were used. Day of conception was set to the day when timed-pregnant guinea pigs gave birth to a previous litter, because guinea pigs enter estrus immediately after birth. All groups contained fetuses from at least two litters, and all fetuses were delivered by abdominal hysterotomy and surgically prepared as described (Fetal Surgery section). All fetuses were studied on 5 cm H2O CPAP for 1 hr after fluid instillation. Distal lung fluid absorption or fluid secretion was measured.
Guinea pig fetuses of GD61 (N = 6) and GD68 (N = 6) were given daily 0.9% NaCl injections. On the third treatment day, the guinea pig fetuses were prepared as described above, instilled with 5% bovine serum albumin in 0.9% NaCl (10 mL/kg body wt), and distal lung fluid absorption or fluid secretion was measured over 1 hr.
Guinea pig fetuses of GD61 (N = 6) and GD68 (N = 6) were pretreated with daily IL-1β (250 ng/kg body wt) injections for 3 days. On the third treatment day, the guinea pig fetuses were prepared as described above, instilled with 5% bovine serum albumin in 0.9% NaCl (10 mL/kg body wt), and distal lung fluid absorption or fluid secretion was measured over 1 hr.
Guinea pig fetuses of GD61 and GD68 (N = 6 at each GD) received the 5% albumin instillate with 30 mg/mL MβCD added (10 mL/kg body wt), and distal lung fluid absorption or secretion was measured over 1 hr. The MβCD dose used was obtained from a prior study by Klein et al. (1995).
Distal Lung Epithelial Cells
To determine if distal lung epithelial (DLE) cells express Cav-1 and PMCA, DLE cells were isolated from guinea pig lungs after enzyme digestion principally according to Monaghan et al. (1997). Timed-pregnant guinea pigs of GD61 (N = 2) and GD68 (N = 2) were anesthetized with heparinized pentobarbital sodium (40 mg/kg body wt). Fetuses were obtained by abdominal hysterotomy and euthanized immediately by heparinized (500 IU) pentobarbital sodium (40 mg per fetus). The fetuses were tracheotomized and the chest opened via a sternotomy. The lungs were perfused blood free in situ via the pulmonary artery and lavaged free of alveolar macrophages. The lungs were then excised and digested with elastase (3 U/mL, Calbiochem) and DNase I (MP Biochemicals, Aurora, OH) for 20 min at 37°C. After digestion, the lungs were minced, and the tissue solution was serially filtered through nylon mesh of decreasing sizes (100–70 μm; Becton Dickinson Labware, Franklin Lakes, NJ). The cells were centrifuged and the pellet resuspended in adequate cell densities using Dulbecco's modified Eagle medium/F-12 (DMEM/F-12 50/50; Cellgro, Herndon, VA). DLE cell purity was measured by modified Papanicolaou stain using Harris hematoxylin, and DLE cell viability was assessed using vital dye exclusion (Trypan blue). The DLE cells were then immediately harvested for Western blot analysis of Cav-1 and PMCA expression.
CRRs were prepared from lung tissue from four guinea pig fetuses at each gestation age with and without MβCD instillation and with and without maternal IL-1β pretreatment using a modification of the method of Song et al. (1996). In summary, lung tissue (25 mg) was placed in 2-mL ice-cold 500 mM sodium carbonate (pH 11) containing protease inhibitors, aprotinin (30 μg/mL; Sigma), and leupeptin (1 μg/mL; Sigma). The lung tissue was homogenized with 12 strokes of a Dounce homogenizer (VWR, Bridgeport, NJ), followed by three 10-sec bursts with a Polytron tissue grinder, and four 20-sec bursts with a sonicator. The 2-mL lung homogenate was placed at the bottom of an ultracentrifuge tube (Beckman Instruments, Palo Alto, CA), and the sucrose concentration was adjusted to 45% by addition of 2 mL 90% sucrose prepared in MBS (25 mM Mes, pH 6.5; 0.15 M NaCl). A 5–35% discontinuous sucrose gradient was formed above the lung homogenate by carefully layering 4-mL 35% sucrose and 4-mL 5% sucrose; both prepared in MBS containing 250 mM sodium carbonate. The gradient was formed by centrifugation at 250,000g for 16 hr, +4°C. A light-scattering band was found confined to the 5–35% sucrose interface after the centrifugation. From the gradient top, 12 1-mL fractions (F) were collected and aliquoted (200-μL each), snap frozen in liquid nitrogen, and stored at −80°C. One aliquot was used to determine protein concentration to ensure equal loading of electrophoresis gels by the Lowry method (Lowry et al., 1951) modified for microtiter plates.
Polyacrylamide gel electrophoresis and transfer to nitrocellulose membrane (Pierce, Rockford, IL) were carried out using standard protocols. After transfer, the nitrocellulose membranes were blocked (dry milk in tris-buffered saline (TBS) over 1 hr.
Cav-1, ENaC, CFTR, α1-Na,K-ATPase, and PMCA
Anti-Cav-1 monoclonal antibodies were obtained from BD Transduction Laboratories (San Diego, CA) and generated from RSV-CEF Cav-1 (FEDVIAEP). The antibodies recognize membrane proteins of appropriate size (22 kDa) in humans. Anti-ENaC antibodies were purchased from Alpha Diagnostics International (San Antonio, TX) and were directed against a segment on the extracellular domain near the N-terminus of α-ENaC and β-ENaC. These antibodies recognize membrane proteins of appropriate sizes (α-ENaC: 85–90 kDa and β-ENaC: 90–95 kDa) in rats. Anti-CFTR polyclonal antibodies were obtained from Santa Cruz Biotechnologies (Santa Cruz, CA) and directed against the N-terminus of the CFTR channel. These antibodies recognize a protein of appropriate size (˜150 kDa) in rats. Anti-α1-Na,K-ATPase antibodies were obtained from Upstate Cell Signaling Solutions (Waltham, MA) and were directed against residues 338–518. These antibodies recognize membrane proteins of appropriate size (˜95 kDa) in rats. Anti-PMCA monoclonal antibodies were obtained from Affinity BioReagents (Golden, CO) and generated from purified human erythrocyte calcium ATPase. These antibodies recognize membrane proteins of appropriate size (˜140 kDa) in multiple animal species. We tested all antibodies for cross-reactivity with the guinea pig and found similar bands specifically labeled for all molecules. After blocking, membranes were incubated with primary monoclonal or polyclonal antibodies (anti-Cav-1, anti-α-ENaC, anti-β-ENaC, anti-CFTR, or anti-PMCA; dilution 1:1,000) on an orbital shaker over night at +4°C. After incubation, membranes were washed five times, 10-min each (pH = 7.5; TBS with 0.1% Tween-20). After washing, membranes were incubated with enzyme-conjugated secondary antibodies, goat-anti-mouse IgG (Cav-1 and PMCA; 1:1,000) and goat-anti-rabbit (ENaC, α1-Na,K-ATPase, and CFTR; 1:1,000), for 1 hr at room temperature. After incubation, the membranes were washed again. Then, the substrate (SuperSignal® West Femto substrate; Pierce) was added and incubated for 5 min. The luminescence signal was detected using a Kodak image analyzer and analyzed densitometrically using TotalLab software (Nonlinear Dynamics, Newcastle upon Tyne, UK).
Coimmunoprecipitation of Cav-1 With α-ENaC, β-ENaC, and CFTR
Lung tissue was homogenized and sucrose gradient fractionated as described above. Then, the equivalent of 10-μg protein of sucrose gradient F5 was used, and PBS was added to a total incubation volume of 0.5 mL. Four microliters (0.5–1.0 μg) of precipitation antibody (anti-Cav-1) was added, and samples were incubated for 1 hr. After this incubation, 20-μL protein A/G PLUS-Agarose was added, and the samples were incubated overnight at +4°C on a rotating shaker. The next morning, immunoprecipitates were collected by 1-min centrifugation at 1,000g. The pellet was washed three times with 1-mL PBS. After the final wash, the pellets were resuspended in electrophoresis sample buffer, boiled, electrophoresed, and analyzed by Western blot for the precipitating protein (Cav-1) and either α-ENaC, β-ENaC, or CFTR.
Values are presented as mean ± standard deviation (SD). Statistical analysis was carried out by one-way analysis of variance (ANOVA) with Tukey's test as post hoc or Student's t-test as appropriate. Differences were considered statistically significant when P < 0.05.
Cav-1, α-ENaC, β-ENaC, CFTR, α1-Na,K-ATPase, and PMCA Expression
We used Cav-1 as a marker for separating CRR and non-CRR cell membrane regions as this protein has been demonstrated in lung epithelial cells (Lisanti et al., 1994; Guthmann et al., 2004) and tested for membrane colocalization with ENaC after sucrose gradient fractionation. Plasma membrane Ca2+ ATPase (PMCA) was used as non-CRR marker. First, we analyzed if these two proteins could be detected in whole lung homogenate and freshly isolated DLE cells and found significant expression of both Cav-1 and PMCA in both whole lung homogenate and in freshly isolated DLE cells (Fig. 1A). Isolation of CRRs was carried out by a detergent-free method that exploits the buoyant density of CRRs by separation on a sucrose density gradient (Song et al., 1996). Lung tissue homogenates were separated using discontinuous sucrose density gradient fractionation, yielding 12 fractions, with distinct protein regionalization. CRRs were in sucrose gradients F5–8 and Cav-1 was absent from F9–12 (Fig. 1B). PMCA, the non-CRR marker, demonstrated an inverse profile, being expressed in F9–12 and being virtually absent from F5–8 (Fig. 1B). Depletion of cell membrane cholesterol by MβCD instillation disrupted the preferential location of Cav-1 to the sucrose gradient F5–8 (Fig. 1B) but did not affect PMCA location to sucrose gradient F9–12 (Fig. 1B).
Expression of the α1-Na,K-ATPase subunit was consistently found in sucrose gradient F9–12 (Fig. 2), that is, the non-CRR fractions, and did not change with gestation age or with IL-1β pretreatment.
We then tested for location of α- and β-ENaC in the sucrose gradient to determine if ENaC subunits predominantly were in the CRR or non-CRR cell membrane regions and if this changed during development and after IL-1β-pretreatment. Both α- and β-ENaC were predominantly found in CRR sucrose gradient F5–8 in GD61 fetuses (CRR: α-ENaC: 72% ± 8.4%; β-ENaC: 58% ± 4.6%; non-CRR: α-ENaC: 28% ± 12.4%; β-ENaC: 42% ± 2.6%;); at GD68, both α- and β-ENaC were clearly present in all fractions (CRR: α-ENaC: 50% ± 2.6%; β-ENaC: 49% ± 5.2%; non-CRR: α-ENaC: 50% ± 6.6%; β-ENaC: 51% ± 8.2%) (Fig. 3B,C). This shift from ENaC CRR localization at GD61 to ENaC non-CRR localization at GD68 correlated with the switch of the alveolar epithelium from fluid secretion to fluid absorption late gestation (Fig. 3A). Maternal IL-1β treatment shifted ENaC α- and β-subunit expression from CRR sucrose gradient fractions to non-CRR sucrose gradient fractions at GD61 (CRR: α-ENaC: 15% ± 2.1%; β-ENaC: 29% ± 4.2%; non-CRR: α-ENaC: 85% ± 9.1%; β-ENaC: 71% ± 3.2%) (Fig. 4B,C), a shift that correlated with induction of distal lung fluid absorption (Fig. 4A). At GD68, after IL-1β there was also a shift in ENaC presence, albeit less pronounced (CRR: α-ENaC: 35% ± 5.9%; β-ENaC: 34% ± 2.5%; non-CRR: α-ENaC: 65% ± 2.9%; β-ENaC: 66% ± 2.5%) (Fig. 4B,C).
We also tested for CFTR location in the sucrose gradient to determine if it predominantly was in the CRR or non-CRR sucrose gradient fractions and if this changed during development and/or after IL-1β-pretreatment. CFTR was predominantly found in the non-CRR sucrose gradient fractions F9–12 in GD61 fetuses and in CRR sucrose gradient fractions F5–8 at GD68 (CRR: GD61: 33% ± 7.9%; GD68: 84% ± 6.7%; non-CRR: GD61: 67% ± 11.9%; GD68: 16% ± 11.7%) (Fig. 5A). Maternal IL-1β pretreatment shifted CFTR expression from the non-CRR sucrose gradient fractions F9–12 to the CRR gradient fractions F5–8 in the GD61 fetal lungs (CRR: GD61: 81% ± 9.9%; non-CRR: GD61: 19% ± 1.9%) (Fig. 5B), a shift in CFTR presence that also was consistent with the induction of distal lung fluid absorption (Fig. 4A). At GD68, after the IL-1β pretreatment, there was little shift in CFTR presence (CRR: GD68: 77% ± 14.4%; non-CRR: GD68: 23% ± 7.4%) (Fig. 5B).
Coimmunoprecipitation of Cav-1 With α-ENaC, β-ENaC, and CFTR
A coimmunoprecipitation of Cav-1 and α- and β-ENaC subunits and CFTR was carried out to confirm that these proteins existed in the same membrane region at a given time and condition. Coimmunoprecipitation of Cav-1 with α- and β-ENaC demonstrated reduced colocalization with increasing gestation age (Fig. 6A,B). Maternal IL-1β pretreatment further reduced the α- and β-ENaC colocalization with Cav-1 at both GD61 and GD68 (Fig. 6A,B). Coimmunoprecipitation of Cav-1 with CFTR demonstrated the inverse relationship, where CFTR colocalization with Cav-1 increased with gestation age and after maternal IL-1β pretreatment (Fig. 6C).
MβCD, Distal Lung Fluid Absorption or Secretion, and ENaC Distribution
To further demonstrate CRR and non-CRR association of ENaC, we decided to chemically disrupt the CRRs. It has earlier been demonstrated that MβCD administration disrupts the CRRs in cell membranes (Klein et al., 1995; Graziani et al., 2004). We therefore used MβCD to investigate if the presence of CRRs in lung epithelial cells affected distal lung fluid absorption during development and if this was associated with an altered distal lung fluid absorption. In GD61 guinea pig fetuses, distal lung fluid absorption was induced by MβCD addition to the instillate (Fig. 7A). At GD68, distal lung fluid absorption was stimulated by MβCD instillation (Fig. 7A).
Once CRRs were disrupted by MβCD instillation, the preferential CRR presence of both α- and β-ENaC subunits at GD61 was lost (CRR: α-ENaC: 25% ± 5.3%; β-ENaC: 36% ± 6.0%; non-CRR: α-ENaC: 75% ± 5.3%; β-ENaC: 64% ± 3.0%) (Fig. 7B,C). Similar, though less pronounced, observations were made in the GD68 fetal lungs, where a more non-CRR-specific organization was apparent as shown by the distribution between F5–8 and F9–12 (CRR: α-ENaC: 39% ± 4.7%; β-ENaC: 43% ± 4.4%; non-CRR: α-ENaC: 61% ± 7.7%; β-ENaC: 57% ± 3.4%) (Fig. 7B,C). When the CRRs were disrupted by MβCD CFTR remained in the non-CRR cell membrane regions at all gestation ages studied (control: CRR: GD61: 33% ± 7.9%; GD68: 84% ± 6.7%; control: non-CRR: GD61: 67% ± 11.9%; GD68: 16% ± 11.7%; MβCD: CRR: GD61: 20% ± 2.1%; GD68: 23% ± 7.0%; MβCD: non-CRR: GD61: 80% ± 12.5%; GD68: 77% ± 10.0%) (Fig. 7D).
Because β-adrenoceptor and cortisol stimulation appears to increase functional Na channels in cell membranes (Matalon et al., 2002; Matthay et al., 2002; Ye et al., 2004), we hypothesized that functional ENaC channels were being recruited from a cellular pool, allowing for the rapid and controlled presentation of ENaCs to cell membranes upon β-adrenoceptor and cortisol stimulation near term. A limited number of reports are suggesting the presence of ion channels in lipid rafts and caveoli (CRRs). Previous work (Martens et al., 2001; Yarbrough et al., 2002) in the heart localized Na and K channels to lipid rafts and caveoli (CRRs). Blockers of vesicular transport attenuate Na currents in various cells (Ito et al., 1997; Zhou et al., 2000) by preventing stimulated increases in plasma membrane channel numbers. Furthermore, activated rat cardiac adenosine A1 receptors have been shown to relocate out of the caveoli (CRRs) and into the plasma membrane (Lasley et al., 2000). In this study, the results suggest that cortisol, IL-1β exerts its actions via cortisol synthesis induction (Ye et al., 2004); stimulation of lung maturation caused a release of ENaC channels from peripheral lung CRRs. We demonstrate, with both biochemical and functional evidence, that the CRRs may be involved in stimulated ENaC presentation at the cell membrane both during fetal lung development and after IL-1β induction of fetal lung maturation. The detergent-free buoyant density method of CRR isolation (Song et al., 1996) routinely yielded membrane regions cross-reactive with Cav-1. Western blot analysis of density gradient fractions demonstrated that both α-ENaC and β-ENaC subunits were associated with CRRs, that is, existed in the same sucrose gradient fractions as Cav-1, during unstimulated conditions. Our coimmunoprecipitation studies further confirmed these observations, where Cav-1 coimmunoprecipitated with α- and β-ENaC during unstimulated conditions. In contrast, Cav-1 did not coimmunoprecipitate with CFTR during unstimulated conditions. The membrane marker, PMCA, was not detectable in the CRR fractions, suggesting that the CRRs were relatively free from membrane contamination. After stimulation by gestation age or IL-1β, both α-ENaC and β-ENaC were detected in the PMCA-positive sucrose gradient fractions, that is, ENaC appeared in the non-CRR cell membrane regions. Moreover, after stimulation by gestation age or IL-1β, colocalization, as assessed by coimmunoprecipitation, of Cav-1 and α- and β-ENaC decreased significantly, whereas the coimmunoprecipitation of Cav-1 with CFTR increased significantly. This evidence supports the hypothesis that ENaC subunits are present in CRRs under unstimulated conditions and in the non-CRR after stress hormone stimulation during late gestation and/or after IL-1β-induced cortisol synthesis and release. It also suggests that CFTR is present in the cell non-CRR during unstimulated conditions and is present within the CRRs late during gestation and after IL-1β-induced lung maturation. In summary, our results provide novel molecular evidence of how the conversion of the near-term fetal lung epithelium from Cl-driven fluid secretion to Na-driven fluid absorption may take place.
There may of course be a multitude of downstream regulatory pathways that may come into play in regulating ion transporter membrane presence besides β-adrenoceptor and cortisol-mediated upregulation of ion transporters. One such system that has recently been studied is the Nedd4-2-based degradation-promoting pathway. By using specific siRNA for Nedd4-2 in the rat, Li et al. (2007) demonstrated that downregulating Nedd4-2 increased fluid absorption from the distal airspaces by increasing ENaC membrane presence. This was later also demonstrated in physiologic studies in guinea pigs during lung development as GD68 fetuses had lower Nedd4-2 levels than GD61 fetuses and GD68 fetuses had higher lung fluid absorption than GD61 fetuses (Li et al., 2009). It has also been demonstrated that stretch-induced release of MAP kinases may be involved in regulating fetal lung fluid absorption (Koshy et al., in press). MAP kinases have, in several studies, been proposed to regulate lung fluid absorption in both positive and negative ways (Pesce et al., 2000; Guerrero et al., 2001; Frank et al., 2003; Upadhyay et al., 2003; Lin et al., 2005; Roux et al., 2005; Bhattacharjee et al., 2007).
The observation that ENaC existed in CRRs prestimulation and appeared to move to the non-CRR cell membrane regions upon stimulation may suggest that the ENaC that is mobilized resides in alveolar epithelial type I cells, as these presumably are the cells with CRRs (Newman et al., 1999; Williams, 2003). Our results demonstrated that Cav-1, our CRR marker, was present in the freshly isolated DLE cells. ENaC has recently been demonstrated in adult alveolar epithelial type I cells (Borok et al., 2002; Johnson et al., 2002, 2006), and CRRs have, in multiple studies, been found in alveolar epithelial type I cells (Newman et al., 1999; Williams, 2003). Our results thus suggest that IL-1β may activate ENaC in the type I cell, which in turn leads to induced distal lung fluid absorption in GD61 fetal lungs. The results may also imply that ENaC membrane appearance in alveolar epithelial type I cells may be responsible, in part, for the conversion of the alveolar epithelium from distal lung fluid secretion to fluid absorption.
Whether ENaC is present as multimeric channels or unique subunits within the CRR under unstimulated conditions cannot be conclusively determined from these studies. However, the non-CRR presence of both subunits simultaneously during development and after IL-1β pretreatment suggests that ENaC may exist as a multimeric channel in the CRRs. In fact, recent data have shown that all three ENaC subunits were present in CRRs both intracellularly and on the cell surface in cultured A6 cells (Hill et al., 2002).
Although not directly tested in this study, circumstantial evidence may actually have suggested a role for β-adrenoceptors in the relocation of ENaC from the CRR at GD61 to the non-CRR cell membrane regions at GD68. Based on the fact that β-adrenoceptors increase in density between GD61 and GD68 (Ye et al., 2004) and that endogenous plasma epinephrine levels increase dramatically during the same time (Norlin and Folkesson, 2001), it is possible that this relocation of ENaC channels may depend, at least partly, on the increased β-adrenoceptor stimulation as term approaches.
To confirm CRR and non-CRR association of ENaC, we decided to chemically disrupt the CRRs by cholesterol depletion after MβCD administration. It has earlier been shown that MβCD disrupts CRRs in cell membranes (Klein et al., 1995; Graziani et al., 2004). In our studies, MβCD instillation resulted in a conversion of GD61 lungs from distal lung fluid secretion to fluid absorption. This result supported our original hypothesis that ENaC subunits when in the non-CRR cell membrane regions become integral and activate parts of the distal lung fluid absorption mechanism in fetal lungs during development. As confirmation that MβCD was effective, Cav-1 after sucrose fractionation was found in all sucrose fractions (F5–12), that is, its distribution was nonpreferential. Similar observations were also made in the A6 cell line as well as in mouse cortical collecting duct cells after MβCD (Hill et al., 2002, 2007). With respect to ENaC in our studies, both subunits lacked preferential CRR presence after MβCD instillation. Thus, all evidence further strengthened our hypothesis that ENaC was present in the non-CRR cell membrane regions in GD68 fetal lungs and after IL-1β in GD61 fetal lungs, simultaneously with the appearance of distal lung fluid absorption.
However, there were some interesting differences between our current studies and those carried out by Hill et al (2007). In that study, a loss of lipid rafts in the mouse collecting duct resulted in a decreased sodium transport, while in our studies, in the lung at GD61 a loss of lipid rafts increased sodium transport. What this may be caused by, however, only be speculated upon. In the study by Hill et al. (2007), MβCD affected the function of the Na,K-ATPase, something we did not observe in the lung as net fluid absorption was increased after MβCD treatment with loss of lipid rafts. This could have been related to differences in way of administration. Hill et al. (2007) administered the MβCD on the basolateral side of the cell layer, whereas we in this study administered it on the apical side. Thus, they might have affected the Na,K-ATPase primarily, whereas we affected the apical ion channels more.
Recently, it has been proposed that ENaC subunits may be translated differentially during lung development and that this may provide a role in regulating ENaC expression to be able to respond rapidly to physiological changes, such as induction of distal lung fluid absorption near birth in preparation for air-breathing life (Otulakowski et al., 2004). Although ENaC subunit mRNA may be translated at different rates, presence at the cell membrane of the final product is essential for removal of fetal lung fluid at birth. Our data suggest that in spite of a potential differential translation, sequestration of ENaC subunits to CRRs may provide an opportunity for ENaC subunits to “catch up with one another” to form the complete ENaC channel that is in the cell membrane upon β-adrenoceptor stimulation by labor-released epinephrine (Walters and Olver, 1978; Norlin and Folkesson, 2001) and/or IL-1β, that is, cortisol (Ye et al., 2004).
The Na,K-ATPase seemed to exist in the cell membrane at either gestation age as well as when fluid absorption was induced by IL-1β. This suggests that the Na transport machinery is present and functioning in the alveolar epithelium as soon as ENaC relocates into the non-CRR cell membrane regions from its prestimulation CRR location. It also provides evidence that induction of distal lung fluid absorption during late gestation and after IL-1β may be primarily an apical membrane event.
CFTR involvement in distal lung fluid absorption is still controversial but has been implicated in multiple investigations. Some investigations suggest that CFTR activation is needed for ENaC activation (Reddy et al., 1999; O'Grady et al., 2000), whereas other studies imply that only β-adrenoceptor-stimulated distal lung fluid absorption is CFTR dependent (Jiang et al., 1998; O'Grady et al., 2000; Fang et al., 2002). Interestingly, when our ENaC and CFTR data are considered together, the results suggest that CFTR is present in the CRRs at the time ENaC is in the non-CRR cell membrane regions. This observation holds true for both the sucrose gradient fractionation studies and the coimmunoprecipitation between Cav-1 and CFTR, where colocation increased during gestation and after maternal IL-1β pretreatment. This may thus provide a clue to the molecular mechanism for the transition in the fetal lung from distal lung fluid secretion to fluid absorption near term. The fact that this would occur is important in that our data suggest that ENaC becomes more active in the non-CRR cell membrane regions after leaving the CRRs and thus conversely, CFTR may become less active upon entering the CRRs. This would lead to a net Na-absorptive lung instead of a predominantly Cl-secreting lung, event which is necessary for the transition from placental to pulmonary oxygenation. To further study the role of CFTR in this process, we determined what happened to the distribution of CFTR in the cell membrane when MβCD was added to the instillate. After MβCD addition, CFTR remained in the cell membrane at all gestation ages studied. This observation might suggest CFTR involvement in regulating ENaC because MβCD seemed to stimulate lung fluid absorption to a higher degree than gestation age alone. There are some earlier studies suggesting that the CFTR can regulate ENaC activity (O'Grady et al., 2000; Fang et al., 2002; Matthay et al., 2002). For example, Fang et al. (2002) demonstrated that β-adrenoceptor stimulation contained a CFTR-dependent component in that lung fluid absorption in CFTR-deficient mice could not be stimulated. For the main point of our studies, this may not have had much influence because CFTR was situated in the CRR regions of the cell membrane when there was fluid absorption.
In this study, we demonstrate that increased gestation age appeared to result in a relocation of ENaC subunits from their prestimulation CRR presence to the non-CRR cell membrane regions where they became active functional ENaC channels and a relocation of CFTR Cl channels into the CRRs where these became less conducive. Maternal IL-1β treatment more significantly increased the degree of ENaC presence in the non-CRR cell membrane regions as well as simultaneously locating CFTR to the CRRs. These observations were confirmed in our coimmunoprecipitation studies of Cav-1 and α- and β-ENaC as well as CFTR colocation. This is, to our knowledge, a novel and potentially important finding that may add to the understanding of how the vitally important and rapid shift in the direction of fluid flow may occur in the perinatal period. We suggest that IL-1β-induced, mediated via elevated plasma cortisol levels (Ye et al., 2004), β-adrenoceptor expression sensitizes the lung to the surge in plasma epinephrine that occurs near term and that this stimulation may relocate the ENaC subunits from prestimulation CRR presence to poststimulation non-CRR membrane presence and CFTR from the non-CRR cell membrane regions and into the CRRs. This shift in ENaC and CFTR locations thus generates an amiloride-sensitive more mature fetal lung. This concept provides a novel molecular model for how the rapid transition from distal lung fluid secretion in fetal lungs to distal lung fluid absorption in near-term and newborn lungs may occur. Moreover, and perhaps more importantly, a better understanding of the molecular mechanisms at the membrane level responsible for the very rapid transition from a fluid-secreting lung to a fluid-absorbing lung at birth could lead to development of new drugs that can assist in the treatment of preterm babies suffering from RDS.
The authors express their thanks to their Research Assistant Stephanie R. Kuzenko and all of the summer students for their hard and dedicated works on this project.
- 1992. Potocytosis: sequestration and transport of small molecules by caveolae. Science 255: 410–411. , , , .
- 2000. The influence of mode of delivery, hormonal status and postnatal O2 environment on epithelial sodium channel (ENaC) expression in perinatal guinea-pig lung. J Physiol (London) 522: 147–157. , , , , , .
- 2007. Stimulation of MAP kinase pathways after maternal IL-1β exposure induces lung fluid absorption in fetal guinea pig lungs. Respir Res 8: 1–8. , , , , , , , .
- 1992. Developmental changes in lung epithelial ion transport and liquid movement. Annu Rev Physiol 54: 373–394. , .
- 2002. Alveolar epithelial ion and fluid transport: Na transport proteins are expressed by rat alveolar epithelial type I cells. Am J Physiol Lung Cell Mol Physiol 282: L599–L608. , , , , , , , , .
- 1997. Intraamniotic interleukin-1 accelerates surfactant protein synthesis in fetal rabbits and improves lung stability after premature birth. J Clin Invest 99: 2992–2999. , , .
- 1974. The postnatal growth of the rat lung. III. Morphology. Anat Rec 180: 77–98. .
- 1953. The structure of the respiratory tissue in the newly-born. Acta Anat 19: 353–365. .
- 2002. Novel role for CFTR in fluid absorption from the distal airspaces of the lung. J Gen Physiol 119: 199–208. , , , , , .
- 1998. Alveolar epithelial fluid clearance is mediated by endogenous catecholamines at birth in guinea pigs. J Clin Invest 101: 972–981. , , , .
- 2003. Transforming growth factor-β1 decreases expression of the epithelial sodium channel αENaC and alveolar epithelial vectorial sodium and fluid transport via an ERK1/2-dependent mechanism. J Biol Chem 278: 43939–43950. , , , , , , , , , , , .
- 2004. Cholesterol- and caveolin-rich membrane domains are essential for phospholipase A2-dependent EDHF formation. Cardiovasc Res 64: 234–242. , , , , .
- 2001. Dopamine regulates Na-K-ATPase in alveolar epithelial cells via MAPK-ERK-dependent mechanisms. Am J Physiol Lung Cell Mol Physiol 281: L79–L85. , , , , .
- 2004. Phenotype of palmitic acid transport and of signalling in alveolar type II cells from E/H-FABP double-knockout mice: contribution of caveolin-1 and PPARγ. Biochim Biophys Acta 1636: 196–204. , , , , , , , , .
- 2002. Endogenously expressed epithelial sodium channel is present in lipid rafts in A6 cells. J Biol Chem 277: 33541–33544. , , .
- 2007. The epithelial sodium channel (ENaC) traffics to apical membrane in lipid rafts in mouse cortical collecting duct cells. J Biol Chem 282: 37402–37411. , , , , , , , .
- 1996. Early death due to defective neonatal lung liquid clearance in αENaC-deficient mice. Nat Genet 12: 325–328. , , , , , , , .
- 1996. Developmental regulation of Na,K-ATPase in rat lung. Am J Physiol Lung Cell Mol Physiol 270: L619–L629. , , , , , , , .
- 1997. The effect of brefeldin A on terbutaline-induced sodium absorption in fetal rat distal lung epithelium. Pflügers Arch 434: 492–494. , , , .
- 1998. Adrenergic stimulation of Na+ transport across alveolar epithelial cells involves activation of apical Cl- channels. Am J Physiol Cell Physiol 275: C1610–C1620. , , .
- 2006. Functional ion channels in pulmonary alveolar type I cells support a role for type I cells in lung ion transport. Proc Natl Acad Sci USA 103: 4964–4969. , , , , , , , .
- 2002. Alveolar epithelial type I cells contain transport proteins and transport sodium, supporting an active role for type I cells in regulation of lung liquid homeostasis. Proc Natl Acad Sci USA 99: 1966–1971. , , , , .
- 1995. Alteration of the myometrial plasma membrane cholesterol content with β-cyclodextrin modulates the binding affinity of the oxytocin receptor. Biochemistry 34: 13784–13793. , , .
- 2011. Lung fluid absorption is induced in preterm guinea pigs ventilated with low tidal volumes. Exp Lung Res 37: 44–56. , , , , .
- 2000. Activated cardiac adenosine A1 receptors translocate out of caveolae. J Biol Chem 275: 4417–4421. , , , .
- 1982. Advanced pulmonary development in newborn guinea pigs (Cavia porcellus). Am J Anat 163: 235–246. , .
- 2007. Involvement of αENaC and Nedd4–2 in the conversion from lung fluid secretion to fluid absorption at birth in the rat as assayed by RNA interference analysis. Am J Physiol Lung Cell Mol Physiol 293: L1069–L1078. , , .
- 2009. IL-1β-induced cortisol stimulates lung fluid absorption in fetal guinea pigs via SGK-mediated Nedd4–2 inhibition. Am J Physiol Lung Cell Mol Physiol 296: L527–L533. , , .
- 2005. Involvement of p42/p44 MAPK, JNK, and NF-κB in IL-1β-induced ICAM-1 expression in human pulmonary epithelial cells. J Cell Physiol 202: 464–473. , , , , .
- 1994. Characterization of caveolin-rich membrane domains isolated from an endothelial-rich source: implications for human disease. J Cell Biol 126: 111–126. , , , , , , , .
- 1951. Protein measurement with the Folin phenol reagent. J Biol Chem 193: 265–275. , , , .
- 1999. Modulation of rat cardiac sodium channel by the stimulatory G protein α subunit. J Physiol (London) 518: 371–384. , , , .
- 2001. Isoform-specific localization of voltage-gated K+ channels to distinct lipid raft populations: targeting of Kv1.5 to caveolae. J Biol Chem 276: 8409–8414. , , , , .
- 2002. Biophysical properties of sodium channels in lung alveolar epithelial cells. J Appl Physiol 93: 1852–1859. , , , .
- 2002. Lung epithelial fluid transport and the resolution of pulmonary edema. Physiol Rev 82: 569–600. , , .
- 1997. Inwardly rectifying K+ currents of alveolar type II cells isolated from fetal guinea-pig lung: regulation by G protein- and Mg2+-dependent pathways. Pflügers Arch 433: 294–303. , , , .
- 1999. Caveolin and its cellular and subcellular immunolocalisation in lung alveolar epithelium: implications for alveolar epithelial type I cell function. Cell Tissue Res 295: 111–120. , , , , .
- 2001. Alveolar fluid clearance in late gestational guinea pigs after induction of labor: mechanisms and regulation. Am J Physiol Lung Cell Mol Physiol 280: L606–L616. , .
- 2000. Cl- channel activation is necessary for stimulation of Na transport in adult alveolar epithelial cells. Am J Physiol Lung Cell Mol Physiol 278: L239–L244. , , .
- 1986. The role of amiloride-blockable sodium transport in adrenaline-induced lung liquid reabsorption in the fetal lamb. J Physiol (London) 376: 321–340. , , , .
- 2004. Differential translational efficiency of ENaC subunits during lung development. Am J Respir Cell Mol Biol 30: 862–870. , , .
- 2000. β-Agonists regulate Na,K-ATPase via novel MAPK/ERK and rapamycin-sensitive pathways. FEBS Lett 486: 310–314. , , , , .
- 1999. Activation of the epithelial Na+ channel (ENaC) requires CFTR Cl− channel function. Nature 402: 301–304. , , .
- 2005. Interleukin-1β decreases expression of the epithelial sodium channel α-subunit in alveolar epithelial cells via a p38 MAPK-dependent signaling pathway. J Biol Chem 280: 18579–18589. , , , , , , , .
- 1996. Co-purification and direct interaction of Ras with caveolin, and integral membrane protein of caveolae microdomains. Detergent-free purification of caveolae membranes. J Biol Chem 271: 9690–9697. , , , , , .
- 2003. FGF-10 prevents mechanical stretch-induced alveolar epithelial cell DNA damage via MAPK activation. Am J Physiol Lung Cell Mol Physiol 284: L350–L359. , , , .
- 1978. The role of catecholamines in lung liquid absorption at birth. Pediatr Res 12: 239–242. , .
- 2002. Pre- and postnatal lung development, maturation, and plasticity: intra-amniotic injection of IL-1 induces inflammation and maturation in fetal sheep lung. Am J Physiol Lung Cell Mol Physiol 282: L411–L420. , , , , , , , .
- 2003. Alveolar type I cells: molecular phenotype and development. Annu Rev Physiol 65: 669–695. .
- 2002. Localization of cardiac sodium channels in caveolin-rich membrane domains. Regulation of sodium current amplitude. Circ Res 90: 443–449. , , , .
- 2004. IL-1β stimulates alveolar fluid clearance in fetal guinea pig lungs via the hypothalamus-pituitary-adrenal gland axis. Am J Physiol Lung Cell Mol Physiol 286: L756–L766. , , , , .
- 2000. Activation of protein kinase A modulates trafficking of the human cardiac sodium channel in Xenopus oocytes. Circ Res 87: 33–38. , , , , .