Introduction

Top Abstract Introduction Materials & Methods Results Discussion References

It has been shown that alveolar Na⁺,K⁺-ATPase plays an important role in active Na⁺ transport in the lung (1). Na⁺,K⁺-ATPase is a ubiquitous transmembrane heterodimer composed of a catalytic  subunit and a glycosylated  subunit (5). The ouabain-inhibitable  subunit is a transmembrane protein that exchanges intracellular Na⁺ for extracellular K⁺. The smaller  subunit is a glycosylated transmembrane molecule that appears to control / heterodimer assembly and insertion into the plasma membrane. Studies using baculovirus vectors to overexpress ₁ subunits in insect cells show that ₁ expression, in the absence of ₁ expression, can generate ATPase activity (6). However, normal Na⁺,K⁺-ATPase function requires the concomitant expression of the ₁ subunit. In the alveolar epithelium, Na⁺,K⁺-ATPase is located in the basolateral membrane of AT2 cells (7). This multimeric "pump" works in concert with other epithelial transport proteins, including apical Na⁺ and water channels, to effect alveolar edema clearance (10). AT2 cells express mRNA transcripts and protein for the ₁ and ₁ subunits of this multigene family and the levels of expression change in response to edemagenic stimuli such as hyperoxia (3, 8).

We have previously reported increased active Na⁺ transport and edema clearance in the lungs of rats exposed to hyperoxia (3, 11, 12). These findings were associated with changes in Na⁺,K⁺-ATPase function and numbers in AT2 cells (3). Specifically, increased Na⁺,K⁺-ATPase expression was associated with increased active transport. Similar changes in expression of an amiloride-sensitive, low- affinity Na⁺ channel have also been reported (13, 14).

A549 cells are an immortalized, human lung adenocarcinoma cell line that is frequently employed in the study of alveolar epithelial cell physiology. Replication-deficient, recombinant adenoviruses can be used to transfer genes to eukaryotic cells and organs in vitro, ex vivo, or in vivo (15). These vectors are tropic for respiratory epithelium; can be grown in large, pure quantities; do not replicate or insert into the host genome; and are capable of producing high levels of gene transduction and expression (15). Using viral promoters, they can generate high level, transient expression of transgene mRNA and protein (15).

We engineered recombinant, replication-deficient, human type 5 adenoviruses containing cDNAs for the rat Na⁺,K⁺-ATPase ₁ and ₁ subunits driven by the immediate-early promoter of cytomegalovirus (CMV). We tested whether these viruses could overexpress rat Na⁺,K⁺-ATPase in human A549 cells and increase Na⁺,K⁺-ATPase activity. The results of this study demonstrate for the first time that Na⁺,K⁺-ATPase function can be increased by transfer and overexpression of the ₁ cDNA in vitro.

	Materials and Methods

Top Abstract Introduction Materials & Methods Results Discussion References

Shuttle Vector Construction

The expression cassette of pCMV (Clontech, San Francisco, CA) was inserted into the Xba 1 site of pXCX2, a pBR322-based plasmid containing the left end (map units 0-2 and 9.24-17.24) of the human adenovirus type 5 (a gift from F. Graham, McMaster University) to produce pMRCMV-gal (19). This expression vector contains the immediate-early promoter and enhancer from CMV, a cDNA for Escherichia coli lac Z and the SV40 t intron polyadenylation signal. The -galactosidase cDNA was excised from pMRCMV-gal and replaced with full-length cDNAs for the rat ₁ and ₁ Na⁺,K⁺-ATPase subunits to produce pMRCMV₁ and pMRCMV₁, respectively.

Adenovirus Construction

A 40.3-kb plasmid containing a human type 5 adenovirus (dl 309) genome (pJM17) (20) without the E1a gene was co-transfected (Lipofectin^®; Gibco BRL, Bethesda, MD) with the above-described shuttle vectors into human embryonal kidney cells (ATCC 293; American Type Culture Collection, Bethesda, MD) (20). Homologous recombination, viral assembly, and replication were detected by the development of cytopathologic effect (CPE). Cells from plates showing CPE were collected and disrupted by six cycles of freezing and thawing. This crude viral lysate was expanded in 293 cells. Following repeat development of CPE, polymerase chain reaction (PCR) using CMV, E. coli lac Z, ₁, and ₁ Na⁺,K⁺-ATPase subunit-specific primers were used to confirm the presence of the ₁ and ₁ cDNAs and CMV promoter in the lysate. PCR-positive cultures were plaque-purified 3 times in 293 cells prior to large-scale amplification (21). The viruses thus produced were AdMRCMV₁, AdMRCMV₁ and AdMRCMV-gal.

Adenovirus Propagation and Purification

Subconfluent 15-cm tissue culture plates of 293 cells were infected with 3 plaque-forming units (pfu) of adenoviral vectors per cell. Following development of CPE the cells were harvested, concentrated, and disrupted with six cycles of freezing and thawing. The resultant cell lysate was cleared of cellular debris by centrifugation prior to purification through serial CsCl density gradient centrifugations. The resultant virus was dialyzed against 10 mM Tris HCl pH 7.4/1 mM MgCl/10% glycerol to remove CsCl prior to storage in 10% glycerol at 70°C. Viral titers were ascertained by the enumeration of plaques produced by adenovirus in 293 cells grown under agarose (21).

Adenovirus Infection Protocol

A549 cells (ATCC CCL 185; American Type Culture Collection) were plated on tissue culture-treated plastic dishes and maintained in a humidified atmosphere of 5% CO₂/ 95% air at 37°C using Dulbecco's modified Eagle's medium (DMEM) (Irvine Scientific, Irvine, CA) containing 10% fetal bovine serum (FBS) (Hyclone Inc., Logan, UT) with 2 mM l-glutamine (Irvine Scientific), 40 mg/ml gentamicin, 100 U/ml penicillin, and 100 µg/ml streptomycin (Sigma, St. Louis, MO). For studies of epithelial cell ion transport and assessment of cytotoxicity, 3.5 × 10⁶ cells in 2 ml of serum-containing medium were plated into each well of six-well tissue culture plates (Falcon, Franklin Lakes, NJ). For Northern and Western blot studies, 1 × 10⁷ cells were plated on 10-cm tissue culture dishes (Corning Glassworks, Corning, NY). In all experiments, cells were incubated for 24 h prior to use.

A549 cells were washed 3 times with DMEM/2% FBS (infection medium) prior to application of 1-200 pfu/cell of adenovirus in 1-2 ml of infection medium. Plates were intermittently rocked for 2 h following infection, whereupon 3-7 ml of DMEM/10% FBS with antibiotics was added (21). Preliminary experiments were conducted using multiplicities of infection (MOI, e.g., pfu/cell) of 0 to 500 of AdMRCMV₁. The results of these studies revealed diminished cell viability at 72 h with concentrations above an MOI of 200.

-Galactosidase Expression

At 24 h after infection with an MOI of 1-200 of AdMRCMV-gal, the cells were washed with phosphate-buffered saline (PBS), pH 7.4, and fixed for 15 min at 4°C with 2% formaldehyde/0.2% glutaraldehyde in PBS. A mixture of 5 mM K₄Fe(CN)₆-3H₂O, 5 mM K₃Fe(CN)₆, and 2 mM MgCl₂ in PBS with 0.5 mg/ml of X-gal solution (Sigma) was applied to the cells. The reaction was allowed to proceed overnight at 37°C prior to a final wash with PBS. Cells were visualized directly in the tissue culture dish with an inverted phase microscope. Transfection efficiency (percent of cells infected) was determined, from 10 randomly selected microscopic fields at ×100, as the number of cells with blue cytoplasm/100 cells chosen from each experimental condition (22, 23). Data represent mean values of triplicate experiments.

Northern Blot Analyses

To demonstrate the steady-state levels of mRNA transcripts of ₁ and ₁ Na⁺,K⁺-ATPase subunits in rat A549 cells after infection with replication-deficient adenovirus (MOI = 25), Northern blot analyses were performed as described elsewhere (22, 24). Five micrograms of total RNA was size-fractionated through 1% agarose/Mops/ 1.7 M formaldehyde gels by electrophoresis, transferred to nylon membranes (Nytran; Schleicher & Schuell, Keene, NH) by capillary action and bound by ultraviolet crosslinking. Rat subunit-specific ³²P-labeled cDNA probes for hybridization were generated by random priming. Nylon membranes were then hybridized at 65°C for 36 h in 0.5 M Na₂HPO₄ (pH 7.0), 1 mM EDTA, 0.5% bovine serum albumin (BSA), and 7% sodium dodecyl sufate (SDS). Membranes were washed twice for 15 min at room temperature with 2× standard saline citrate (SSC)/0.1% SDS followed once with 0.5× SSC/0.1% SDS. Two 20-min final washes at 65°C were performed using 0.1× SSC/0.1% SDS. Membranes were then exposed to X-ray film (Biomax MR; Eastman Kodak Co., Rochester, NY) at 80°C for 16-24 h prior to development. To demonstrate similarity of lane loading, the membranes were stripped of Na⁺,K⁺-ATPase probes and rehybridized with a ³²P-labeled rat glyceraldehyde-3-phosphate dehydrogenase (GAPDH) cDNA probe and re-exposed to X-ray film. Message expression was quantified by laser scanning densitometry of the resultant autoradiograms (Molecular Dynamics, Sunnyvale, CA).

Western Blot Analyses

Abundance of Na⁺,K⁺-ATPase subunit proteins was determined by Western blot analysis of A549 membrane fractions obtained 24 h after sham infection or infection with an MOI of 25 of AdMRCMV₁ or AdMRCMV-gal. All manipulations and solutions were at 4°C. Cells were washed with PBS prior to lysis with 10 mM Tris-HEPES/3 mM EGTA/1 mM EDTA/2 mM DTT/10 mM Mannitol with 0.01 mg/ml N-tosyl-L-phenylalanine chlorylmethyl ketone, 0.1 mM PMSF, and 0.01 mg/ml Leupeptin (all from Sigma). Cells were scraped, collected, and homogenized with a Potter Ehvehjem homogenizer. The cell homogenate was centrifuged at 1,500 × g for 15 min and the supernatant was collected and centrifuged at 100,000 × g for 1 h. The subsequent supernatant was rehomogenized and centrifuged at 1,500 × g, prior to repeat centrifugation at 100,000 × g. The pellet thus obtained was combined with the first pellet. The final pellet was resuspended in 100 µl of homogenization buffer and quantified using a Bradford Assay (Bio-Rad protein assay; Bio-Rad, Hercules, CA). A total of 25 µg of protein of homogenate was fractionated on 7.5% SDS-polyacrylamide gel electrophoresis and transferred to nitrocellulose membranes (Optitran^TM; Schleicher and Schuell). The nitrocellulose membranes were blocked for 2 h in blotto (7.5% dry milk, 0.05% Tween 20 in TBS), followed by incubation with anti-Na⁺,K⁺- ATPase subunit antibodies in TBS, 0.1% BSA, and 0.1% Na⁺ Azide for 12 h. The anti-₁ antibody used was anti-NASE, a rabbit antihuman polyclonal antibody (T. Pressley, Texas Tech University, Lubbock, TX). A polyclonal rabbit antidog ₁ subunit antibody (A. Askari, Medical College of Ohio, Toledo, OH) was used to detect ₁ protein expression. Following incubation with primary antibody, blots were rinsed for 2 h with wash buffer (TBS, 0.05% Tween 20) and then incubated with horseradish peroxide-conjugated goat antirabbit secondary antibody (Bio-Rad) for 1 h at room temperature. Blots were rinsed for 2-4 h with wash buffer prior to chemiluminescent detection (ECL^®; Amersham, Arlington Heights, IL). Protein expression was quantified by laser scanning densitometry.

Immunocytochemistry

To demonstrate ₁ gene transfer efficiency and transgene expression, A549 cells were immunostained for the presence of rat ₁ protein. A549 cells were plated on 24-well dishes (Falcon) at a density of 3.5 × 10⁵ cells/well for 24 h prior to infection with an MOI of 25 of AdMRCMV₁, as described above. Twenty-four hours after infection the cell culture media was aspirated and the cells were washed 3 times with PBS prior to overnight fixation at 4°C with Cytospin Collection Fluid (Shandon, Pittsburgh, PA). Fixative was removed and the cells were repeatedly washed with PBS prior to application of 0.3% H₂O₂ for 15 min at room temperature to diminish endogenous peroxidase activity. Cells were again washed 3 times with PBS prior to incubation with 1% non-immune rabbit serum in PBS for 1 h at room temperature to block nonspecific immunoreactivity. The blocking solution was then removed and replaced with primary antibody at a 1:500 dilution (A. Askari). Incubation was continued for 2 h at room temperature. Immunodetection was accomplished via immunoperoxidase staining using a commercially available kit (Vector Elite ABC kit; Vector Laboratories, New Castle-upon-Tyne, UK). Cells were photographed in situ to allow estimation of ₁ protein expression and gene transfer efficiency. These experiments were performed in duplicate.

Na⁺,K⁺-ATPase function

Ouabain-sensitive ⁸⁶Rb⁺ uptake was used to estimate the rate of K⁺ transport by Na⁺,K⁺-ATPase in A549. Cells in six-well plates were incubated with and without 5 mM ouabain (ICN, Aurora, OH) for 5 min at 37°C in a reciprocating water bath at 100 rpm. This medium was removed, and otherwise-identical fresh medium containing 1 µCi/ml ⁸⁶Rb⁺ (Amersham) was added. Five minutes later the assay medium was removed by aspiration followed by the addition of ice-cold 150 mM MgCl₂. Plates were allowed to dry and cells were solubilized in 0.2% SDS. ⁸⁶Rb⁺ influx was quantitated from aliquots of the SDS extract by a liquid scintillation counter. Protein was quantitated using the Lowry method. Initial influx, expressed as µM K⁺/g of protein/min, was calculated from:

where dRc/dt represents the slope of the linear phase of the uptake curve (counts per minute [cpm] ⁸⁶Rb⁺ taken up per g of protein in 1 min) and SAex is the specific activity of the extracellular phase (cpm/µmol K⁺). The uptake of ⁸⁶Rb⁺ added to the culture medium was linear for at least 10 min (data not shown); therefore, ⁸⁶Rb⁺ uptake was measured over a 5-min period (25, 26). Three data points were obtained for each condition during each experiment. All experiments were done in triplicate.

Ouabain Sensitivity

Ouabain-sensitive ⁸⁶R⁺ uptake was used to estimate the rate of K⁺ transport by Na,K-ATPase. Cells were incubated for 5 min at 37°C in a reciprocating water bath at 100 rpm. A solution of 50 mM NaCl and 25 mM Hepes with varying concentrations of ouabain was added to some of these cells (final concentrations of ouabain: 10¹¹-10³ M). An otherwise-identical solution without ouabain was used for control. The ouabain or control solutions were then removed, and otherwise-identical fresh medium containing 1 µCi/ml ⁸⁶Rb⁺ was added. Five minutes later, uptake was terminated by aspirating the assay medium and washing the plates in ice-cold MgCl₂. Plates were then processed as described above. Ouabain sensitivity was assessed by determining the concentration of ouabain that reduced ⁸⁶Rb⁺ uptake by 50% (IC₅₀). Triplicate sets of data were obtained for each ouabain concentration tested. Data were normalized to simultaneously processed cells not exposed to ouabain or to control for interexperimental variation and processed using a computerized nonlinear least squares regression analysis function designed to test for two receptors of differing affinities for a ligand (GraphPad Prism^TM, GraphPad Software, San Diego, CA).

Assessment of Cytotoxicity

Culture media concentrations of K⁺ and lactate dehydrogenase (LDH) were used as indicators of cytotoxicity. A549 cells were plated in 6-cm dishes for 24 h prior to infection with an MOI of 5 or 10 of AdMRCMV₁ or AdMRCMV-gal. Cells were maintained in 7 ml of complete medium for 24 h prior to aspiration and re-measurement of medium volume. Specimens were centrifuged at 600 × g to remove cells and cellular debris; K⁺ and LDH concentrations were measured in the resultant supernatant. K⁺ was determined using an ABL620-100EML electrolyte analyzer (Radiometer Medical A/S, Copenhagen, Denmark). LDH concentrations were measured with a Hitachi 747 analyzer (Boehringer Mannheim, Indianapolis, IN). Cells from these plates were trypsinized and counted to assure uniformity of cell number.

Cell Volume Determination

Measurements of cell volumes following infection with an MOI of 25 were made to assess for changes in cell volume due to altered Na⁺,K⁺-ATPase function. Adherent A549 cells were trypsinized from culture dishes, centrifuged at 300 × g, and resuspended in 0.5 ml PBS. The resuspended cells were fixed by dropwise addition of 3 ml 70% ethanol (4°C) while vortexing. Aliquots of the cell/ethanol suspension were diluted 1:1 with 1% methylene blue in methanol, incubated at 22°C for 20 min, and centrifuged onto 96-well culture plates. Video images of stained cells were acquired with MOCHA image analysis software (Jandel Scientific, San Rafael, CA) through the ×20 objective lens of a Ziess/ Jena phase-contrast inverted microscope. Cell-diameter measurements were obtained using the object measurement routine of MOCHA on a minimum of 100 cells per sample. Raw data from quadruplicate experiments were analyzed with INSTAT statistical software (Graphpad Software) and reported as mean diameter ± SEM. These values were converted to cell volume using data transformations based on the formula for the volume of a sphere (27). Data are represented as volume in cubic micrometers (µm³).

Statistical Analysis

Differences between groups were assessed using ANOVA (Excel; Microsoft Corp., Seattle, WA). Nonlinear regression analysis using a function designed to test for the presence of two receptors with different affinities for the same ligand was used to analyze the ouabain sensitivity data (Graphpad Prism^TM, Graphpad Software). Data are presented as mean standard deviation. A P value of < 0.05 was used as the cutoff for statistical significance.

	Results

Top Abstract Introduction Materials & Methods Results Discussion References

Transfection Efficiency

At 24 h after infection, A549 cells infected with MOIs of 1-200 of AdMRCMV-gal demonstrated transfection efficiencies that exceeded 85% at an MOI of 200 (Figure 1). Sham or AdMRCMV₁-infected A549 cells showed no evidence of -galactosidase activity (data not shown).

View larger version (167K): [in this window] [in a new window]   Figure 1.   Transfection efficiency. The efficiency of gene transfer into A549 cells was assessed using X-Gal staining of cells exposed to varying concentrations of AdMRCMV-gal. Transfection efficiencies ranged from 0 to  85% at an MOI of 200.

Na⁺,K⁺-ATPase Function (Ouabain-Sensitive ⁸⁶Rb⁺ Uptake)

Significant increases in ouabain-sensitive ⁸⁶Rb⁺ uptake were noted in cells infected with MOIs of 5 or more of AdMRCMV₁. Cells exposed to an MOI of 1 did not demonstrate any change in ouabain-sensitive ⁸⁶Rb⁺ uptake. Cells infected with similar concentrations of AdMRCMV-gal or AdMRCMV₁ showed no change in ouabain-sensitive ⁸⁶Rb⁺ uptake (Figure 2).

View larger version (18K): [in this window] [in a new window]   Figure 2.   Na+,K+-ATPase function. As compared with sham infected controls, ouabain-inhibitable 86Rb+ uptake was increased by up to 2.5-fold in A549 cells infected with MOIs of 5 or more of AdMRCMV1 (upper graph). No increase in activity is noted in this cell line following infection with AdMRCMV-gal or AdMRCMV1 (middle and lower graphs). *P < 0.001.

Northern Blot Analysis

As shown in Figure 3, hybridization of 5 µg/lane of total RNA using a 1-kb cDNA probe complementary to the 5' portions of rat ₁ subunit mRNA revealed the presence of a 3.7-kb ₁ message only in A549 cells infected with an MOI of 25 of AdMRCMV₁. No ₁ message was detected in the sham or AdMRCMV-gal-infected cells using this rat-specific probe. This message was of the same size as that noted in control rat tissues. Duplicate blots hybridized using a full-length cDNA probe complementary to rat ₁ subunit mRNA showed no change among any of the experimental groups. Because of the observed lack of functional changes noted following AdMRCMV₁ infection, we did not test what changes AdMRCMV₁ had on ₁ and ₁ mRNA expression. Hybridization of these blots with a rat GAPDH cDNA demonstrated equivalent lane loading of total RNA among all experimental samples.

View larger version (70K): [in this window] [in a new window]   Figure 3.   Na+,K+-ATPase mRNA expression. Hybridization of 5 µg of total RNA harvested from A549 cells using a cDNA probe complementary to rat 1 subunit mRNA shows expression of an mRNA that migrates between the 2.35- and 4.40-kb molecular weight markers in cells infected with an MOI of 25 of AdMRCMV1 (upper blot). Sham and AdMRCMV-gal (MOI = 25)-infected cells do not express this message. This message is the same size as that seen in control rat tissues on the left side of this Northern blot. An otherwise identical Northern blot was probed with a cDNA complementary to rat 1 mRNA (lower blot). A 2.7/2.35-kb doublet, consistent with human 1, is evident in all samples. The first four lanes of these Northern blots contain size-fractionated total RNA isolated from uninfected control rat tissues; the right-hand lane is a molecular weight ladder. Numbers are in kilobars (kb).

Western Blot Analysis

The membrane fractions from A549 cells infected with AdMRCMV₁ demonstrated ₁ subunit immunoreactivity at approximately 100 kD. Laser scanning densitometry (Molecular Dynamics) showed a 3-fold increase in ₁ protein only in the cells infected with AdMRCMV₁. The observed band was of the same size as that noted in control rat tissues. Expression of ₁ protein was not different among any of the experimental groups (Figure 4). All experiments were performed in triplicate.

View larger version (36K): [in this window] [in a new window]   Figure 4.   Na+,K+-ATPase protein expression. Western blot analysis of 25 µg of protein from A549 cell membrane fractions reveals a 3-fold increase in 1 immunoreactivity in specimens obtained from cells infected with an MOI of 25 of AdMRCMV1 (upper blot). This protein was of the same size as that seen in rat kidney. The 1 subunit protein levels were not changed following infection with AdMRCMV1 (lower blot). Molecular weights are in kilodaltons (kD).

Immunocytochemistry

Immunocytochemistry using a antirat ₁ subunit antibody was employed to demonstrate that AdMRCMV₁ was able to infect A549 cells and express its transgene. As compared with sham and AdMRCMV-gal-infected A549 cells, cells infected with AdMRCMV₁ at an MOI of 25 demonstrated dense peroxidase activity in more than 50% of cells (Figure 5). This finding suggests that AdMRCMV₁ is able to infect these cells in an efficient manner and that it can express its transgene at a high level.

View larger version (108K): [in this window] [in a new window]   Figure 5.   1 subunit expression. Immunostaining using a polyclonal rabbit antirat antibody to the 1 Na+,K+-ATPase subunit was used to demonstrate gene transfer efficiency and transgene expression in A549 cells infected with an MOI of 25 of AdMRCMV1. A large proportion of the A549 cells infected with AdMRCMV1 show significant amounts of peroxidase activity indicative of increased 1 subunit protein expression. No similar activity was noted in the sham or AdMRCMV-gal-infected controls.

Ouabain Sensitivity

Ouabain sensitivity was determined by measuring ⁸⁶Rb⁺ uptake in the presence of different concentrations of ouabain. A549 cells infected with an MOI of 25 demonstrated the presence of two IC₅₀s (Figure 6). The first ₁ IC₅₀ was not different from that noted in sham or AdMRCMV-gal-infected controls. A second IC₅₀, 2 logs greater than the first, was noted only in the cells infected with AdMRCMV₁. Regression analysis indicated that the relative contributions to total ⁸⁶Rb⁺ uptake were 41% for the high-affinity isozyme and 59% for the low-affinity isozyme.

View larger version (19K): [in this window] [in a new window]   Figure 6.   Ouabain sensitivity. 86Rb+ uptake by A549 cells infected with an MOI of 25 of AdMRCMV1 was measured following treatment with 1 × 1011 to 1 × 103 M ouabain. Nonlinear regression analysis of the data demonstrated two IC50s for these cells (inset), suggesting the presence of two distinct Na+,K+-ATPase isozymes of different sensitivities to ouabain. The ouabain-insensitive isozyme contributed ~ 60% to total Na+,K+-ATPase activity under these experimental conditions. Sham infected or AdMRCMV-gal (MOI = 25) each had only a single IC50 that was similar to the first IC50 of the AdMRCMV1-infected cells.

Assessment of Cytotoxicity

No gross cytologic difference in appearance was noted among any of the experimental groups 24 h following infection with MOIs of 1 to 200. Media concentrations of K⁺ and LDH were not different from sham infected controls following infection with MOIs of 1 to 200 of AdMRCMV₁ or AdMRCMV-gal. A549 cells infected with an MOI of 25 had cell volumes of 3,156.8 ± 679.7, 2,194 ± 379.5, and 1,937.7 ± 697.1 for AdMRCMV₁, AdMRCMV-gal, and sham infected cells, respectively. Cell volumes were not statistically different among any of the experimental groups.

	Discussion

Top Abstract Introduction Materials & Methods Results Discussion References

It has been reported that in animal models of lung injury the development of pulmonary edema is associated with changes in Na⁺,K⁺-ATPase expression and function and that these changes parallel pulmonary edema clearance (3, 8, 11, 12). These studies suggest that alveolar Na⁺,K⁺-ATPase is an important contributor to the processes that keep the alveoli free of edema. We hypothesized that overexpression, by gene transfer, of Na⁺,K⁺-ATPases may increase Na⁺,K⁺-ATPase function in vitro. The results of this study show that high transfection efficiencies into A549 cells can be achieved using recombinant, replication-deficient adenoviruses. These cells tolerated infection well and showed no signs of cytotoxicity.

Studies in renal tubular cells and more recent data in lung epithelial cells indicate that Na⁺,K⁺-ATPases play an important role in transcellular (vectorial) Na⁺ and water transport (1, 2, 4, 8, 10). Na⁺,K⁺-ATPases work with other membrane-bound transport systems, such as water and apical Na⁺ channels, to effect net vectorial movement of Na⁺ and water out of the airspace and into the alveolar interstitium (10, 13, 14, 28). In the alveolar epithelium, Na⁺,K⁺-ATPases have been localized to the basolateral portion of AT2 cell membranes (7), although they may also be expressed in alveolar type 1 cells (9). Inhibition of Na⁺,K⁺-ATPase with ouabain has been shown to impair transcellular ion transport processes in AT2 cells in culture and edema clearance in isolated lungs (2, 11).

Hormones have been shown to increase Na⁺,K⁺-ATPase expression and/or activity via transcriptional, post-transcriptional, translational, and post-translational mechanisms. Regulation of Na⁺,K⁺-ATPase by hormones such as aldosterone, thyroid hormone, insulin, insulin-like growth factor, and glucocorticoids has been reported (29). Control of activity and life-span of assembled heterodimers in the plasma membrane, presumably via protein kinase-mediated phosphorylation and altered cytoskeletal interactions, has also been reported (30). Similarly, catecholamines have been shown to nonspecifically augment Na⁺,K⁺-ATPase activity in the lungs (4, 10). We chose gene transfer as a method of specifically overexpressing Na⁺,K⁺-ATPase and increasing its function in A549 cells (15, 31).

A549 cells are derived from a human lung adenocarcinoma explant. These cells have some phenotypic characteristics of AT2 cells and have been used previously for studies of alveolar epithelial cell biology (31). To demonstrate the function of our rat transgene in these human cells we measured ouabain-inhibitable ⁸⁶Rb⁺ uptake. As compared with controls, these cells showed significant increases in ⁸⁶Rb⁺ uptake following infection with MOIs of 5 or more of AdMRCMV₁ (Figure 2). We also observed rat ₁ message and increased ₁ protein only in the cells infected with AdMRCMV₁ (Figure 3). The absence of similar finding in control cells indicates that these findings are a specific response to adenoviral-mediated ₁ subunit gene transfer.

It has been previously established that the human ₁ subunit is 2 logs more sensitive to ouabain than is the rat ₁ isozyme (32). We utilized this difference to detect transgene expression and function. Control A549 cells demonstrated a pattern of ouabain sensitivity consistent with the presence of a single ouabain isozyme (Figure 6). Nonlinear regression analysis of ⁸⁶Rb⁺ uptake measured with the same concentrations of ouabain following infection with AdMRCMV₁ suggested the presence of two distinct IC₅₀s. The second IC₅₀ was observed at an ouabain concentration 2 logs greater than the first, suggesting the presence of a second functional ₁ isozyme; i.e., the rat ouabain-resistant ₁ isozyme. Previous studies have used similar methods to demonstrate a shift between ₁ isozyme expression in differentiating myotubes and alveolar epithelial cells (9, 33).

Assuming that extracellular Na⁺ concentrations are only minimally changed during the 5-min experimental period during which we measure ⁸⁶Rb⁺ uptake, then the relative contributions of the human and rat isozymes can be estimated by nonlinear regression analysis. Our analysis suggests that 60% of total ouabain-inhibitable ⁸⁶Rb⁺/K⁺ uptake is due to the transgenic, ouabain-resistant rat ₁ isozyme (Figure 6). This is consistent with the 2.5-fold increase in ⁸⁶Rb⁺ uptake noted following infection with the same MOI of AdMRCMV₁. Thus it seems reasonable to conclude that with the transfection efficiency noted in this study, approximately 60% of the ⁸⁶Rb⁺ uptake is due to the rat ₁ transgene, and that the remainder is due to the endogenous human isozyme following infection of A549 cells with AdMRCMV₁. This would also suggest that endogenous ₁ activity remains unchanged following adenoviral transduction of the ₁ isoform.

Normal Na⁺,K⁺-ATPase function requires the synthesis, assembly, and transport of both  and  subunits to the cell membrane. However, subunits are not always synthesized in a coordinated fashion. Excess  and  units appear to exist in intracellular pools that can be used as reservoirs of subunits when availability of the rate-limiting subunit occurs. Lescale-Matys and colleagues (34) have shown increased Na⁺ pump activity in pig renal cells (LLC-Pk₁) in hypokalemic conditions. This increased activity was associated with transcription of ₁ mRNA and accumulation of newly synthesized  and  subunits, leading these investigators to conclude that the ₁ subunit was rate-limiting under hypokalemic conditions in these cells (35). In contrast Hensley and associates have reported that the ₁ subunit was rate-limiting in hyperthyroid cardiac myocytes (36). We noted increased Na⁺,K⁺-ATPase function only in the cells transduced with the ₁-expressing virus (see Figure 2). The absence of functional change following ₁ gene transfer, and the lack of change in ₁ message and protein levels following ₁ gene transfer, suggest that the ₁ subunit may be rate-limiting in these cells. While the possibility exists that ₁ protein stability may be altered, we reason that pre-exisiting pools of human ₁ protein may be available to form functional heterodimers with the rat ₁ transgene. Similar findings were noted by Takeyasu and coworkers (37), who reported that transfection of avian  subunits in mammalian cells already expressing avian  subunits increased cell membrane Na⁺,K⁺-ATPase. These investigators concluded that the  subunit played a key role in facilitating exiting of assembled heterodimers from the endoplasmic reticulum. They also suggested that there are unlikely to be significant pre-existing pools of unassembled subunits in the endoplasmic reticulum of skeletal myotubes due to continual degradation of unassembled subunits (38). Were this the case, altered stability of pre-existing  subunits could also explain our findings.

Concerns have arisen regarding possible cytotoxic and inflammatory effects of recombinant adenoviruses. Lung inflammation has been noted in rats, nonhuman primates, and humans (39). Multiple mechanisms have been postulated and studies have focused on identifying the responsible immune mechanisms. These mechanisms pertain to humoral and cellular immune processes in whole lungs and probably not to isolated alveolar cells. Our findings of normal cytologic appearance, cell volume, and K⁺ and LDH concentrations in culture media support the notion that these viruses do not produce direct cytotoxic responses. While current-generation adenoviruses clearly have limitations, they are capable of achieving the high transfection efficiencies and transgene expression levels required to produce biologically relevant levels of gene expression.

In summary, our study represents a novel approach of overexpressing Na⁺,K⁺-ATPase in mammalian cells. The increased Na⁺,K⁺-ATPase function noted in these studies following infection with AdMRCMV₁ demonstrates that Na⁺,K⁺-ATPase activity can be augmented via gene transfer. The vectors used in this study may prove useful for studying the effects of increased Na⁺,K⁺-ATPase function in vitro and in vivo. What impact overexpression of individual Na⁺,K⁺-ATPase subunits may have on other transport proteins or functions is as yet unknown and warrants further study. These vectors could also be useful for studying Na⁺,K⁺-ATPase subunit interdependency and regulation. Conceivably, gene transfer of Na⁺,K⁺-ATPase genes may prove useful for the treatment of pulmonary edema.