lung cancer is the leading cause of cancer deaths in the United States, accounting for roughly 30% and 26% of deaths in men and women, respectively (22). Lung neoplasms are categorized into two distinct histological subtypes that include small cell lung cancer (SCLC), which represents 15–25% of cases, and non-small cell lung cancer (NSCLC), which accounts for 75–85%. NSCLC is further morphologically divided into three histotypes: large cell, squamous, and adenocarcinoma, which is the most prevalent (44, 48). Despite the use of targeted therapies that inhibit commonly dysregulated pathways in lung cancer, the 5-yr survival rate remains low (7, 12). Identification of new markers may be useful for early diagnosis, determining prognosis, and aiding the selection of appropriate treatments.

The large surface area of the lung facilitates the efficient exchange of oxygen and carbon dioxide but also makes it vulnerable to environmental toxicants, such as cigarette smoke (CS). Numerous studies have shown that ∼80–90% of patients with lung cancer are known smokers (19, 26). Furthermore, patients who continue smoking after early diagnosis of lung cancer have a higher risk of mortality, recurrence, and development of secondary tumors (32). CS contains over 4,000 constituents including carcinogens, toxins, and free radicals that contribute to the development and progression of lung carcinogenesis (17, 43). The formation of covalent bonds between CS-derived carcinogens and DNA creates somatic mutations in several genes, such as K-ras and p53, which regulate cellular proliferation and cell death (23, 29, 33). Oxidative stress induced by CS, which contains a high concentration of oxidants (10¹⁴-10¹⁶ molecules/puff) (10), has been implicated in the development of lung cancer (15). The mechanisms by which reactive oxygen species (ROS) contribute to lung cancer are not completely clear but include DNA damage (15, 46) and activation of oncogenic signaling pathways, such as epidermal growth factor receptor activation (24), that promote unchecked growth.

Na,K-ATPase plays a highly conserved role in establishing a sodium gradient across the membrane and relies on the energy derived from ATP hydrolysis to catalyze the export of three sodium ions and import of two potassium ions into the cell. Functional Na,K-ATPase consists of an α-subunit (NaK-α) and β-subunit (NaK-β). Several reports have noted reduced expression in NaK-β₁ in a number of cancer tissues derived from kidney, bladder, breast, colon, and pancreas (13, 14, 35), and the NaK-β₁ isoform has emerged as an important suppressor of tumor development and advancement. Our laboratory and others have shown that NaK-β₁ regulates tumor progression by maintaining epithelial polarity (37, 39), enhancing cell-cell aggregation (5, 21, 25, 39, 42, 47), reducing cell motility and invasiveness (4, 38), and suppressing tumor growth in vivo (21). We have shown that patients with high NaK-α₁ expression coupled with low levels of NaK-β₁ in bladder cancer exhibited a higher risk for early recurrence and those with low levels of NaK-α₁ and high levels of NaK-β₁ had a significantly longer recurrence-free period (14, 41). In this study, we investigated the effect of CS on Na,K-ATPase function and showed that Na,K-ATPase activity was attenuated after CS exposure and that loss of pump activity was due to the generation of ROS. We established a correlation between Na,K-ATPase expression levels in NSCLC tumor tissue obtained from patients and showed that NaK-α₁ expression in NSCLC patients decreased slightly with increasing grade. Furthermore, higher levels of NaK-β₁ expression predicted a significant survival advantage in patients with stage I/II NSCLC with a current or former history of cigarette smoking.

MATERIALS AND METHODS

Cell culture.

A549 cells were maintained in RPMI-1640 (Invitrogen, Carlsbad, CA) supplemented with 10% fetal bovine serum, 2 mM L-glutamine, 25 U/ml penicillin, and 25 μg/ml streptomycin. Human bronchial epithelial cells (HBECs) were maintained in Keratinocyte serum-free medium (Invitrogen).

Antibodies.

Horseradish peroxidase-conjugated secondary antibodies were purchased from Cell Signaling (Danvers, MA). β-Actin antibody was purchased from Sigma Chemical (St. Louis, MO). Antibodies against NaK-α₁ (M7-PB-E9) and NaK-β₁ (M17-P5-F11) were obtained as described previously (39).

Cigarette smoking treatment.

CS extract was prepared using an airtight modulator incubator chamber (Billups-Rothenberg, Del Mar, CA). Briefly, cells (in 4 ml of medium) were placed in the chamber and exposed to CS from 2 Marlboro 100 cigarettes using a syringe-driven apparatus. Smoke was collected from one cigarette into a syringe and delivered to the cells through a three-way connector controlled by clamps in 35-ml puff volumes at a rate of 0.05 puff/s according to the International Standards Organization guidelines. Each cigarette yielded an average of 10–12 puffs using this procedure. Sham-treated cells were subjected to the same conditions but were exposed to air instead. After treatment, the chamber was incubated at 37°C for 15 min following which cells were removed and kept in a 37°C incubator with 5% CO₂ and harvested at various time points to measure the cellular effects of CS on Na,K-ATPase activity.

Immunoblot analysis.

Protein lysates for NaK-α₁, NaK-β₁, and β-actin immunoblots were prepared as described previously (39) in buffer containing 95 mM NaCl, 25 mM Tris, pH 7.4, 0.5 mM EDTA, 2% SDS, and protease inhibitors (1 mM phenylmethylsulfonyl fluoride, 5 μg/ml each of antipain, leupeptin, and pepstatin). Protein concentration was determined using the DC protein assay according to manufacturer's protocol (Bio-Rad, Hercules, CA). Equal amounts of protein were separated by SDS-PAGE and immunoblotted as described previously (39). β-Actin was the loading control.

Cell surface biotinylation assay.

Sham or CS-treated cells were washed four times in cold PBS containing 1 mM CaCl₂ and 1 mM MgCl₂ (PBS-CM). The EZ-Link Sulfo-NHS-LC-Biotin (ThermoFisher, Hudson, NH) was prepared in DMSO, diluted into TEA buffer (150 mM NaCl, 10 mM Triethanolamine pH 9, 1 mM CaCl₂, and 1 mM MgCl₂) to a final concentration of 4 mg/ml, and incubated for 20 min on ice. The biotinylation reaction was repeated twice. Excess biotin was quenched with 50 mM NH₄Cl in PBS-CM for 10 min. Cells were washed in PBS-CM before lysis in buffer containing 20 mM Tris·HCl, pH 7.4, 100 mM NaCl, 1% Triton X-100, 1 mM EDTA, 1 mM EGTA, 1 mM sodium glycerolphosphate, 1 mM sodium orthovanadate, 1 mM PMSF, and 5 μg/ml each of antipain, leupeptin, and pepstatin. Lysates were precleared with Protein A agarose beads for 30 min at 4°C. Equal amounts of protein were incubated with streptavidin beads (ThermoFisher) overnight at 4°C. Samples were centrifuged, and beads were washed three times with cold lysis buffer. Samples were processed for SDS-PAGE and immunoblotting as described above.

Lactate dehydrogenase assay.

The Cytotox96 nonradioactive assay was performed according to manufacturer's instructions (Promega, Madison, WI). Briefly, cells were seeded into six-well plates in phenol-free medium and exposed to CS. At various time points, medium was collected and assayed for lactate dehydrogenase (LDH) released from the dead cells (A₄₉₀ from medium). An equivalent volume of medium was replaced in the plate and incubated at −80°C for 30 min to lyse cells. The medium was harvested and assayed for LDH released from the viable cells. Fractions were clarified by centrifugation, and 50-μl aliquots were added (in triplicate) to 50 μl LDH substrate in a 96-well plate. The reaction was terminated using 50 μl stop solution, and absorbance values were obtained at 490 nm using a microtiter plate reader. Readings were corrected for background absorbance using the absorbance value of buffer only. %LDH released = (A₄₉₀ from medium)/(A₄₉₀ from plate + A₄₉₀ from medium) × 100%. Data represent the mean ± SE from three independent experiments.

Rubidium uptake assay.

The ouabain-sensitive ion transport was measured by determining the uptake of ⁸⁶Rb⁺ as described previously (27). Cells were washed once with 1.5 ml of ice-cold wash solution (144 mM NaCl, 0.5 mM CaCl₂, 10 mM HEPES, pH 7.4), incubated for 10 min at 37°C with 1 ml of incubation buffer (144 mM NaCl, 10 mM HEPES pH 7.4, 0.5 mM MgCl₂, 0.5 mM CaCl₂, 1 mM RbCl, 1 mg/ml glucose, 1 μCi ⁸⁶Rb⁺), and washed three times with 1.5 ml of ice-cold wash solution. To measure the ouabain-sensitive ⁸⁶Rb⁺ transport, the cells were incubated with 50 μM ouabain for 30 min at 37°C before the experiment. The cells were dissolved with 500 μl of 0.5 M NaOH for 1 h at room temperature, and solubilized ⁸⁶Rb⁺ was counted using a LSC5000 liquid scintillation counter (Beckman Instruments, Mississauga, ON, Canada). The samples were normalized to the protein content, and the ouabain-sensitive ⁸⁶Rb⁺ flux was calculated as the difference between Rb uptake in ouabain and untreated cells.

Intracellular Na⁺ ion determination.

Either Sham or CS-treated cells from three 10-cm dishes were pooled in 0.25 M sucrose, digested with HNO₃ (Ultrex II; J.T. Baker, Phillipsburg, NJ) at a final concentration of 40% at 65°C for 15 h, and diluted 1:2 with Millipore Milli-Q UF plus filtered water. Intracellular ion concentrations were measured with an inductively coupled plasma atomic emission spectrometer (Vista Axial 730; Varian, Walnut Creek, CA) as described previously (39). The concentrations for Na⁺ (588.995 nm), K⁺ (766.941 nm), and Mg²⁺ (285.213 nm) were determined. Na⁺ and K⁺ concentrations were normalized to the total Mg²⁺ content (internal control).

Detection of ROS.

Cells were seeded into a black poly-lysine-coated 96-well plate in serum-free medium (Corning, Union City, CA). Dichlorofluorescein diacetate (Sigma) was added to a final concentration of 10 μM. Cells were incubated at 37°C and 5% CO₂ for 60 min to permit dye uptake. Excess dye was removed by washing with PBS. Serum-free medium (50 μl) containing 10 mM HEPES was added, following which cells were exposed to CS as described above. Fluorescence was measured in a Victor4 plate reader (Perkin-Elmer, Waltham, MA) at an excitation of 480 nm and emission of 535 nm. All data represent the mean of at least three independent measurements ± SE. P values were calculated using Student's t-test.

Immunohistochemistry.

A standard two-step indirect avidin-biotin complex (ABC) method was used for immunohistochemical studies (Vector Laboratories, Burlingame, CA) using sections of lung tissues from patients with NSCLC patients. Slides were heated to 60°C for 15 min, followed by deparaffinization in xylene. Sections were rehydrated in graded alcohol and endogenous peroxidase quenched with 10% hydrogen peroxide in PBS at room temperature for 20 min. Slides were placed in 95°C solution of 0.01 M sodium citrate buffer and then blocked using 1% normal horse serum and 5% BSA for 30 min. Either anti-Na,K-α- or β-subunit antibodies was added for 60 min at 37°C followed by biotinylated horse anti-mouse IgG for 30 min at room temperature. The ABC complex was added for 25 min, and diaminobenzidine was used as the chromogen. The sections were counterstained with Harris hematoxylin, dehydrated, and mounted. Negative controls minus the primary antibodies were performed for each slide. Sections were viewed with an Olympus BX-41 bright-field microscope, and semiquantitative analysis of chromogen/antibody staining of tissue sections was performed. Plasma membrane-associated staining intensity was categorized into four levels (scale 0–3, with 0 = below the level of detection, 1 = weak, 2 = moderate, and 3 = strong), and percentage of cells staining at each level was ascertained for all cases. A final integrated value [(3 × % cells staining at intensity 3) + (2 × % cells staining at intensity 2) + (1 × % cells staining at intensity 1)]/100 was used for comparing quantitative staining differences between nonneoplastic lung tissues and different grades of NSCLC.

Lung tissue microarray.

A lung tissue microarray (TMA) was constructed under appropriate review of Institutional Review Board (IRB) and Health Insurance Portability and Accountability Act (HIPPA) regulations, using archival formalin-fixed, paraffin-embedded lung tissue samples from the Department of Pathology and Laboratory Medicine at the University of California at Los Angeles Medical Center as previously described (30). The TMA contained tissue for 396 marker-informative lung cancer cases linked to outcome information (survival vs. death due to disease). Of these patients, 330 were either former or current smokers and 51 were nonsmokers. Of the nonsmokers, 10 had known exposure to second-hand smoke. No smoking history could be obtained on the remaining 15 patients; 248 of the 330 current/former smokers had stage I or stage II NSCLC. Sampled tissues included primary lung tumor, matched nonneoplastic lung parenchyma, and metastatic lung carcinoma to lymph nodes and distant sites. Sections of all blocks that were used were reviewed by a board-certified pathologist to confirm the diagnosis. At least three core tissue biopsies were taken from select, morphologically representative regions of each paraffin-embedded lung tumor and precisely arrayed using a custom-built instrument, as previously described (30, 40). The demographics, histopathological distribution, grade, stage, clinical variables, smoking history, and outcome (survival and death due to disease) of the population studied here were similar to those reported in the United States and have been described elsewhere in detail (30). Individuals who were classified as nonsmokers had no history of ever smoking cigarettes. Individuals classified as smokers had all smoked >100 cigarettes at one point in their life. Immunohistochemistry for NaK-α₁ and NaK-β₁ on TMAs were done and scored as described above by two pathologists and then compared. Spearman correlation coefficients were 0.954 for NaK-α₁ and 0.998 for NaK-β₁, with P value < 1.0e-14 for both.

Statistical analysis.

Analyses were done using R software (including survival and part packages). Na,K-α₁ and Na,K-β₁ expression differences among various subgroups were determined using the Mann-Whitney U-test or Kruskal-Wallis rank-sum test. For analyses for which samples were dichotomized, robust cut-points were determined to minimize potential overfitting. High- and low-protein expression was found to be significant in the range of the upper 70^th to 75^th percentiles of patient expression levels. For dichotomized Na,K-ATPase subunit expression, the Fisher exact test was used for analysis with categorical variables such as stage, grade, and smoking history. Survival curves were calculated using the Kaplan-Meier method, and comparisons were done using the log-rank test. The Cox proportional hazards model (univariate and multivariate) was used to determine the significance of various factors related to survival. Log-rank and Fisher exact P values were two-sided and, a P value <0.05 was considered significant.

RESULTS

CS exposure impairs Na,K-ATPase activity.

Because cigarette smoking is the primary contributing factor to the development, progression, and recurrence of lung cancer (32), we examined the effects of mainstream CS on Na,K-ATPase function. The effects of CS exposure on Na,K-ATPase activity were measured in lung adenocarcinoma cells (A549) and normal HBECs using the Rb uptake assay. A reduction in pump activity was observed in A549 cells following CS exposure with 71.5 ± 5.3% (P = 0.0028) activity at 30 min, 44.7 ± 6.2% at 4 h (P = 0.00086), and almost complete inhibition after 8 h (P < 0.0001) compared with Sham-treated cells (Fig. 1A). For the HBEC cell line, reduction in Na,K-ATPase activity to 83.8 ± 4.5% occurred at 15 min (P = 0.04), to 37.6 ± 14.8% at 30 min (P = 0.017), and almost complete inhibition occurred at 4 h (P < 0.0001) post-CS exposure (Fig. 1B).

The inhibition of Na,K-ATPase function by CS exposure was accompanied by the accumulation of intracellular sodium. In the absence of CS, for A549 cells the Na/Mg ratio was 1.15 ± 0.17 presmoking treatment and 8.33 ± 0.68 (P < 0.0001) post-smoking treatment (Fig. 1C). Similarly, compared with Sham-treated cells (1.57 ± 0.062), the Na/Mg ratio was increased in HBEC cells (3.39 ± 0.29, P < 0.0001) after CS exposure, (Fig. 1C). The average LDH release of CS-treated cells was ∼18.6 ± 3.3% compared with control at 20.0 ± 5.5% for A549 cells (P = 0.89) after 4 h post-CS exposure. In HBECs, the LDH release was 27.0 ± 5.7% at 30 min post-CS-exposed cells compared with the control at 20.0 ± 14.1% (P = 0.58) (Fig. 1D). The low LDH release under these experimental conditions indicate that 73–80% of the cells were viable post-CS exposure. These results demonstrate that Na,K-ATPase pump function is inhibited by CS and not due to reduced cell viability.

Induction of intracellular ROS by CS correlates to reduced Na,K-ATPase activity.

It has been reported that CS contains a large concentration of ROS (10). To determine whether the release of intracellular ROS affects the Na,K-ATPase pump function, we quantified ROS after CS exposure using dichlorofluorescein diacetate, a cell-permeable probe that exhibits fluorescence after oxidation. There was rapid induction of intracellular ROS with CS exposure in both A549 and HBEC cells. In A549 cells exposed to CS, the emitted fluorescence was 300 ± 54% higher than the control cells (Fig. 2A). Treatment with ROS scavenger, N-acetyl-cysteine, (NAC) (1, 20) completely quenched the fluorescence (Fig. 2A). HBEC cells also showed similar increases in ROS production and enhanced fluorescence (330 ± 17%) compared with control cells at 1 min post-CS exposure (Fig. 2B). In these cells, fluorescence levels decreased to untreated levels after 30 min (Fig. 2B). As observed with A549 cells, treatment with NAC also quenched the initial fluorescence in this cell line. Taken together, these data indicate that intracellular ROS is rapidly released in response to CS.

The Na,K-ATPase pump activity was measured as Rb uptake in the presence of NAC. The addition of NAC did not affect Na,K-ATPase activity in either control A549 and control HBEC cells or control cells treated with CS (Fig. 2, C and D) as similar activity levels were obtained. As previously observed (Fig. 1), after CS exposure in both cell lines, there was a reduction in pump activity by more than 50% at 4 h and 30 min, respectively. Strikingly, in the presence of 10 mM NAC, almost full restoration of the activity was achieved (Fig. 2, C and D). Collectively, these results indicate that ROS generated during CS exposure is most likely involved in the inhibition of Na,K-ATPase pump function.

Loss of Na,K-ATPase activity upon CS exposure does not correlate to changes in protein expression.

Because we observed reduced pump activity in the presence of CS, we examined the expression of Na,K-ATPase subunits in cells exposed to CS. Immunoblot analysis after CS exposure showed that, although there was a significant decline in pump function, a corresponding change in either NaK-α₁ or NaK-β₁ protein expression in both A549 and HBEC cells (Fig. 3A) was not observed. These results suggest that the downregulation of pump activity is not attributable to reduced protein levels. Dada et al. (11) have shown that cell internalization of the pump from the membrane impairs pump activity without affecting protein levels. To investigate this possibility, we performed a cell-surface biotinylation assay to determine Na,K-ATPase levels at the cell surface. We observed reduced expression of NaK-α₁ levels at the cell surface with CS exposure in both A549 and HBEC cells. Interestingly, in A549 cells the NaK-α₁ levels increased after 2 h post-CS. The NaK-β₁ expression did not change significantly (Fig. 3B). Taken together, these data suggest that the reduction of Na,K-ATPase activity during CS exposure is partially associated with reduced expression of NaK-α₁ at the cell surface. Furthermore, the increased surface expression after 2 h post-CS exposure is not sufficient to recover pump activity in A549 cells.

Na,K-α₁ expression is reduced in neoplastic respiratory tissues with higher tumor grade.

Survey of Na,K-ATPase expression level in tissues from patients with NSCLC showed fairly strong correlation between levels of Na,K-α₁ and levels of Na,K-β₁ (Spearman rho = 0.43, P < 0.00001). Figure 4 shows representative images of Na,K-α₁ and Na,K-β₁ protein staining. In general, there was slightly stronger expression of Na,K-α₁ in well-differentiated and moderately differentiated adenocarcinomas (Fig. 4, A and C) compared with poorly differentiated tumors (Figs. 4E and 5, left; Kruskal-Wallis P < 0.0001). Staining for Na,K-β₁ (Fig. 4, B, D, and F) was generally weaker than for Na,K-α₁ (Fig. 4, A, C, and E) but did not show a trend toward decreasing expression with increasing grade (Fig. 5, right; Kruskal-Wallis P = 0.10). However, there was considerable variability of expression among tumors of all grades.

Loss of Na,K-ATPase expression in smokers correlates with poorer survival.

Because CS has a profound effect on lung carcinogenesis, we used a lung tissue microarray to examine effects of smoking on Na,K-ATPase expression from lung cancer tissues by immunocytochemistry and correlated expression data to overall survival. No differences in either Na,K-α₁ or Na,K-β₁ levels were seen between nonsmokers, former smokers, and current smokers (Kruskal-Wallis P = 0.66 for Na,K-α₁ and P = 0.85 for Na,K-β₁). However, when we looked at the group of smokers (both current and former), Na,K-β₁ was found to be a significant predictor of survival in those patients with stage I or II NSCLC. By the Cox proportional hazards model, for Na,K-β₁ as a continuous variable, higher levels predicted better survival (log rank P = 0.033, hazard ratio = 0.65). Higher Na,K-α₁ expression did confer a slight but not significant survival benefit (log rank P = 0.298, hazard ratio = 0.65). A Kaplan-Meier curve illustrates improved survival with Na,K-β₁ as a dichotomized variable (Fig. 6A; P = 0.017). The predictive power was slightly greater when considering Na,K-α₁ and Na,K-β₁ together. In smokers with early stage (I and II) lung cancer, patients expressing higher Na,K-α₁ and Na,K-β₁ had an overall greater survival probability compared with individuals with tumors that expressed lower levels of Na,K-α₁ and Na,K-β₁ (Fig. 6B, P = 0.007). These findings were not explained by specific tumor stage, grade, or sex in the dichotomized groups. Finally, in a multivariate Cox model, which also included stage, grade and age, Na,K-β₁ remained an independent predictor of survival (P = 0.048, Tables 1, 2, and 3).

DISCUSSION

The previous decade has illuminated new insights on the functions of the Na,K-ATPase enzyme beyond its traditionally defined role as an ion pump (34). In this study, we investigated whether there was a correlation between CS, a major inducer of lung carcinogenesis, and Na,K-ATPase activity. We showed a significant loss of Na,K-ATPase activity with CS exposure in lung cell lines. We further demonstrated that this was attributable to the CS-induced ROS production. We have also provided evidence that the reduction in pump activity with CS exposure correlated with the internalization of the Na,K-α₁ and not changes in total protein expression. Using samples from patients with NSCLC, we further showed that NaK-α₁ expression diminished with increasing grades of tumor. High levels of Na,K-β₁ alone or expression of Na,K-α₁ along with Na,K-β₁ predicted a significant survival advantage. Thus these results indicate for the first time that diminished Na,K-ATPase function and/or expression may be an important predictor of tumor progression and patient survival in lung cancer.

CS contains a high concentration of ROS (3) and has been recognized to induce oxidative stress in inflammatory and epithelial cells (6). We investigated the contribution of ROS to the loss of Na,K-ATPase activity. First, we tested for the presence of intracellular ROS and found that peak ROS levels were detected very early after CS exposure. Secondly, we neutralized ROS with the general scavenger, NAC, and observed a near complete restoration of pump activity after CS exposure. These results suggest that CS-induced formation of ROS in lung cells is involved in diminishing Na,K-ATPase activity. In addition, it also indicates that inhibition of pump activity post-CS does not affect the viability of cells because Na,K-ATPase activity is completely restored by NAC treatment. The introduction of oxidative stress by other inducers, such as alcohol (2), airborne particulate matter (31), and addition of hydrogen peroxide (16), has been demonstrated to attenuate Na,K-ATPase activity. The precise ROS species and the mechanisms by which CS induced ROS attenuate Na,K-ATPase activity remain to be studied.

Because we observed a loss of Na,K-ATPase pump function with CS exposure, we examined whether this was attributable to loss of Na,K-ATPase protein. However, total Na,K-α₁ and Na,K-β₁ protein expression were unaltered. One explanation for this result is that ROS may be promoting endocytosis of Na,K-ATPase. This form of Na,K-ATPase regulation has been previously recognized in lung alveolar epithelial cells (9, 11, 18, 28). It was demonstrated that hypoxia-mediated generation of ROS in lung alveolar cells caused endocytosis of Na,K-ATPase, whereas overexpression of the ROS scavenger catalase prevented the internalization of Na,K-ATPase (18). Previously, it was shown that hypoxia-mediated endocytosis of Na,K-ATPase leads to its phosphorylation, ubiquitination, and subsequent degradation (11). Thus it is plausible that the formation of ROS during CS exposure reduced pump activity by causing a retrieval of the Na,K-ATPase from the cell surface. Consistent with this hypothesis, we observed reduced Na,K-α₁ at the cell surface, which may explain in part the reduction of pump activity during CS exposure. However, because there was a significant level of Na,K-α₁ at the cell surface, the dramatic reduction of pump activity cannot be explained solely by the internalization of Na,K-α₁ subunit. In addition, even though the surface expression of Na,K-α₁ increased in A549 cells after 2 h post-CS exposure, it was not sufficient to restore the pump activity. Other intracellular factors induced by ROS might contribute to pump inhibition in cells exposed to CS. We speculate that long-term or chronic exposure to CS may constitutively inhibit Na,K-ATPase activity and ultimately lead to downregulation of its expression by ROS-mediated endocytosis and subsequent degradation.

We assessed Na,K-ATPase levels in NSCLC patient samples using TMA analysis. Interestingly, we noted that patients with a history of smoking in relatively early stages of the disease (stage I/II) had higher levels of Na,K-β₁ and a higher survival probability. The predictive power of Na,K-ATPase was even stronger in patients with stage I/II with a smoking history if we considered higher levels of both Na,K-β₁ plus Na,K-α₁; lower levels of Na,K-β₁ alone or lower levels of both Na,K-β₁ plus Na,K-α₁ predicted a reduced probability of survival. Future studies are required to determine how changes in Na,K-ATPase protein expression occur in lung neoplasms and how this loss may contribute to disease progression and outcome.

In the in vitro acute CS exposure, we did not observe changes in Na,K-ATPase expression, whereas the patient samples showed a reduction of Na,K-ATPase protein. It is possible that a longer time point may be needed to cause changes in the Na,K-ATPase expression. However, CS treatment beyond 24 h leads to significant apoptosis in cultured cells (our unpublished result). It is known that the average number of cigarettes smoked per day ranges from 17–22 cigarettes, with differences attributed to sex and age (8). As a result, we postulate that long-term changes involving the reduction of Na,K-ATPase expression and lung carcinogenesis are cumulative and result from prolonged exposure to CS. Such differences may account for the lack of diminished Na,K-ATPase expression observed in short-term cell culture compared with the significant differences we observed in patient tumor samples. Alternatively, the effects observed in the in vitro studies may be mechanistically distinct from the human tumor tissue studies. Further studies are required to determine whether the mechanism driving the loss of Na,K-ATPase activity after acute CS exposure also contributes to the reduced expression of Na,K-ATPase in the human lung tumor tissues. The mechanism by which Na,K-ATPase expression is reduced in patient samples and how it is associated with malignant lung cancer development and progression are not known. It is possible that CS-induced inhibition of Na,K-ATPase activity might affect tight junction permeability and polarity in lung epithelial cells because Na,K-ATPase activity is essential to maintain tight junction function and polarity (39). Furthermore, we have shown recently that reduced expression of NaK-β₁ is one of the early events during TGF-β-induced epithelial-to-mesenchymal transition (37). In addition, Na,K-β₁ is a target of Snail1 (13), which is a master molecule involved in the induction of epithelial-mesenchymal transition that facilitates tumor progression to metastatic disease (45). Long-term inhibition of Na,K-ATPase might lead to reduced levels of Na,K-ATPase β-subunit probably attributable to production of factors such as TGF-β and Snail. Reduced Na,K-β₁ levels might then lead to reduced Na,K-α₁ levels as well because Na,K-β₁ is necessary for maintaining the stability of Na,K-α₁ (36). Future studies aimed at elucidating the molecular mechanisms underlying the alterations in Na,K-ATPase expression would help determine its role in malignant progression of lung cancer.

GRANTS

This work was supported by the predoctoral National Ruth L. Kirschstein Research Service Award from the National Institutes of Health NIH-NCI-F31CA117050–01A1, a National Institutes of Health grant DK56216, the Nemours Foundation (A. Rajasekaran), and the Early Detection Research Network NCI CA-86366 (L. Goodglick and D. Chia).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the authors.

AUTHOR CONTRIBUTIONS

Author contributions: T.P.H., T.S., S.M.D., S.A.L., L.G., and A.K.R. conception and design of research; T.P.H., V.M., V.B.S., D.C.W., J.H., and T.S. performed experiments; T.P.H., V.M., V.B.S., D.C., M.C.F., S.H., M.A., J.H., L.G., and A.K.R. analyzed data; T.P.H., V.M., D.C., M.C.F., S.H., S.M.D., S.A.L., L.G., and A.K.R. interpreted results of experiments; T.P.H., V.M., V.B.S., and L.G. prepared figures; T.P.H., L.G., and A.K.R. drafted manuscript; T.P.H., V.B.S., S.H., S.A.L., L.G., and A.K.R. edited and revised manuscript; A.K.R. approved final version of manuscript.