Na-K-ATPase is a fundamental component of ion transport. Four α isoforms of the Na-K-ATPase catalytic α subunit are expressed in human cells. The ubiquitous Na-K-ATPase α1 was recently discovered to also mediate signal transduction through Src kinase. In contrast, α2 expression is limited to a few cell types including myocytes, where it is coupled to the Na+/Ca2+ exchanger. To test whether rat Na-K-ATPase α2 is capable of cellular signaling like its α1 counterpart in a recipient mammalian system, we used an α1 knockdown pig renal epithelial cell (PY-17) to create an α2-expressing cell line with no detectable level of α1 expression. These cells exhibited normal ouabain-sensitive ATPase, but failed to effectively regulate Src. In contrast to α1-expressing cells, ouabain did not stimulate Src kinase or downstream effectors such as ERK and Akt in α2 cells, although their signaling apparatus was intact as evidenced by EGF-mediated signal transduction. Additionally, α2 cells were unable to rescue caveolin-1. Unlike the NaKtide sequence derived from Na-K-ATPase α1, which downregulates basal Src activity, the corresponding α2 NaKtide was unable to inhibit Src in vitro. Finally, coimmunoprecipitation of cellular Src was diminished in α2 cells. These findings indicate that Na-K-ATPase α2 does not regulate Src and, therefore, may not serve the same role in signal transduction as α1. This further implies that the signaling mechanism of Na-K-ATPase is isoform specific, thereby supporting a model where α1 and α2 isoforms play distinct roles in mediating contraction and signaling in myocytes.
- rat Na-K-ATPase isoforms
- Src tyrosine kinase
- membrane transporters
- signal transduction
the na-k-atpase, discovered in 1957 by Jens Skou, is a member of the P-type ATPase family and an integral membrane protein maintaining cellular ion homeostasis by pumping Na+ and K+ across the cell membrane (32). The protein consists of two noncovalently linked subunits, α and β. The α-subunit contains the binding sites for substrates (e.g., ions and ATP) and ligands (e.g., ouabain) as it undergoes E1 and E2 conformational changes during a transport cycle.
Four isoforms of the Na-K-ATPase α-subunit are expressed in humans. While α1 is expressed ubiquitously, α2 and α3 are primarily found in myocytes and neurons, respectively, and α4 is detected in sperm (3, 4, 31, 45, 50). Recent studies have revealed major differences in the physiological functions of each isoform. For example, the α2 isoform appears to possess the unique ability to regulate intracellular Ca2+ levels in myocytes and coresides with the Na+/Ca2+ exchanger (5, 24). In addition, recent experiments using SWAP mice (ouabain-sensitive α1 Na-K-ATPase mutant and ouabain-resistant α2 Na-K-ATPase mutant) suggest that the α2 isoform plays a more prominent role in calcium release in cardiac and smooth muscle myocytes than α1 (12).
Over the last decade, we have also come to realize that the Na-K-ATPase may have many regulatory functions other than pumping ions across cell membranes. Studies from various laboratories have documented an important signaling function of the Na-K-ATPase (2, 27). For example, we and others have shown that Na-K-ATPase α1 is capable of keeping the Src kinase in an inactive state in many different types of cell (28, 30). This interaction allows the formation of a functional receptor complex, which becomes a target for cardiotonic steroids such as ouabain to stimulate protein and lipid kinase cascades and subsequently regulate many cellular activities including cell growth and migration. Interestingly, this newly appreciated signaling function of Na-K-ATPase α1 has also been implicated in the development of cardiac hypertrophy and fibrosis (14, 20).
Studying signaling by Na-K-ATPase isoforms other than α1 is challenging. A major obstacle is that the various α-polypeptides are coexpressed with the ubiquitous α1 (3). In the past, we and others have circumvented this problem by using heterologous expression systems in mammalian cells, where a ouabain-resistant isoform is expressed in a ouabain-sensitive recipient cell (39). Selection with a low concentration of ouabain in the media conveniently inhibits the endogenous form while forcing expression of the resistant isoform, an approach that is suitable for studies of the enzymatic properties, but not signaling. Indeed, in such model, both residual endogenous expression of α1 and a low concentration of ouabain would interfere with signaling. Consequently, earlier studies of isoform-specific signaling capabilities were conducted in a nonmammalian system with no endogenous expression of Na-K-ATPase α1, but the limitations inherent to such a model have left us in need of a more suitable system (40). Here, we report how we developed the first mammalian cell model that expresses the Na-K-ATPase α2 isoform in the absence of a detectable amount of α1 polypeptide, and the subsequent studies in that system to determine whether Na-K-ATPase α2 is also capable of mediating ouabain-induced activation of protein kinases. This may provide a better understanding of the interplay between ion pumping and signal transduction in myocytes, and ultimately lay the foundation for developing new isoform-specific therapeutics.
MATERIALS AND METHODS
Fetal bovine serum and trypsin were purchased from Invitrogen (Carlsbad, CA). Monoclonal anti-Src antibody (B12), polyclonal anti-ERK1/2 antibody, monoclonal anti-pERK1/2 antibody, monoclonal anti-caveolin-1 antibody, goat anti-rabbit and goat anti-mouse secondary antibodies were obtained from Santa Cruz Biotechnology (Santa Cruz, CA). Polyclonal anti-pY418Src was purchased from Invitrogen. Polyclonal anti-pY 1173 EGF receptor, polyclonal anti-pAkt, and polyclonal anti-Akt were obtained from Cell Signaling (Danvers, MA). Polyclonal anti-Na-K-ATPase β1 was from Upstate (Lake Placid, NY). Polyclonal anti-Na-K-ATPase α2 was from Millipore. Monoclonal anti-α1 antibody (α6F) was obtained from the Developmental Studies Hybridoma Bank at the University of Iowa. Polyclonal anti-rat α1-specific antibody [anti-NASE (42)] was provided by Dr. Thomas Pressley (Texas Tech University, Lubbock, TX). Radioactive 3H-ouabain was obtained from Perkin Elmer.
LLC-PK1 cells (ATCC CL-101), a well-characterized model of pig renal epithelial cells (10), were purchased from American Type Culture Collection (Manassas, VA). PY-17 and AAC-19 cells were generated from LLC-PK1 cells (29). Cells were cultured in DMEM in 10% FBS, 100 U/ml penicillin, and 100 μg/ml streptomycin in a 5% CO2 humidified incubator. After the cells reached 95–100% confluence, to eliminate the confounding effect of growth factors in the serum, the cells were serum starved before experiments unless otherwise indicated.
Generation of α2 cell lines.
To generate stable cell lines, Na-K-ATPase α1 knockdown PY-17 cells (29) were transfected with a ouabain-insensitive rat α2 mutant cDNA using Lipofectamine 2000 as previously described (29). Cells were then selected with ouabain (3 μM) starting 24 h after transfection. Ouabain-resistant colonies were isolated and expanded into stable cell lines. Cells were then cultured in the absence of ouabain for three generations before use in experiments.
Western blot analysis.
Western blot analysis was performed as previously described (16). Cells were washed with PBS and solubilized in ice-cold radioimmune precipitation assay buffer containing 1% Nonidet P-40, 1% sodium deoxycholate, 150 mM NaCl, 1 mM EDTA, 1 mM PMSF, 1 mM sodium orthovanadate, 1 mM NaF, 10 μg/ml aprotinin, 10 μg/ml leupeptin, and 50 mM Tris·HCl, pH 7.4. The cell lysates were then centrifuged at 14,000 rpm, and the supernatants were used for protein assay and subjected to Western blot analysis. The samples were separated on SDS-PAGE gel (50 μg/lane), transferred to Optitran membrane, and blotted with specific antibodies. The protein signals were detected using an ECL kit.
Cell growth curve.
Cells were cultured in 12-well plates at 50,000/well for 24 h, serum starved for 24 h, and then treated with different concentrations of ouabain for the indicated time periods. At each indicated time point, three wells of control or treated cells were trypsinized and counted.
Ouabain-sensitive Na-K-ATPase activity.
The Na-K-ATPase activity was assayed according to the protocol previously described (29), with modification. Briefly, cells were harvested in Skou C buffer (30 mM histidine, 250 mM sucrose, 1 mM EDTA, pH 7.4) and sonicated briefly. After initial centrifugation (800 g for 10 min), the postnuclear fraction was further centrifuged (100,000 g for 45 min) to obtain crude membrane. The crude membrane pellet was resuspended in Skou C buffer and treated with alamethicin (0.1 mg/mg of protein) for 10 min at room temperature. The preparation was then incubated in the buffer containing 50 mM Tris (pH 7.4), 1 mM EGTA, 3 mM MgCl2, 25 mM KCl, 100 mM NaCl, 5 mM NaN3 and 2 mM ATP. Phosphate generated during the ATP hydrolysis was measured using BIOMOL GREEN Reagent (Enzo Life Science). Ouabain-sensitive Na-K-ATPase activities were calculated as the difference between the presence and absence of 1 mM ouabain.
3H-ouabain binding assay.
To determine the residual surface expression of the (endogenous) pig Na-K-ATPase α1 in PY-17 and LX-α2 cells, 3H-ouabain binding assays were performed as described (47). Briefly, 90% confluent cells were serum starved overnight. Cells were washed with warm K+-free Krebs buffer (142.4 mM NaCl, 2.8 mM CaCl2, 0.6 mM NaH2PO4, 1.2 mM MgSO4, 10 mM glucose, 15 mM Tris, 37°C and pH 7.4) and incubated with 3H-ouabain for 30 min at 37°C. The reaction was stopped by three washes with ice-cold K+-free Krebs buffer, and proteins were solubilized in a 0.1 N NaOH-0.2% SDS solution for 30 min at 37°C.
Src autophosphorylation assays.
Indicated amounts of peptide were incubated with 1 unit of purified Src at 37°C in PBS for 15 min. The reaction was initiated by adding 2 mM Mg2+-ATP and stopped by adding the SDS sample buffer after 15 min. Src activity was determined by phosphorylation of Src at Tyr418 using immunoblot analysis.
To assay for Na-K-ATPase α1 or α2 binding to Src, a coimmunoprecipitation assay was performed as previously described (16). Briefly, cell lysates were incubated with monoclonal anti-Src antibody overnight and then protein G agarose for 2 h. After extensive washes, immunoprecipitates were subjected to Western blot analysis.
Data are given as means ± SE. When more than two groups were compared, one-way ANOVA was performed prior to post hoc analysis. Statistical significance was accepted at P < 0.05.
Generation and characterization of Na-K-ATPase α2-expressing cell lines.
To characterize the pumping and signaling properties of Na-K-ATPase α2, we employed a newly developed knockdown and knock-in protocol to generate α2-expressing stable cell lines. Specifically, we transfected Na-K-ATPase α1 knockdown PY-17 cells with a ouabain-resistant rat α2 cDNA (18). As reported in the initial description of the PY-17 cell line, Na-K-ATPase α1-specific siRNA targeting reduces the expression of endogenous Na-K-ATPase α1 to ∼10% of that of the parent pig kidney LLC-PK1 cells (29). Subsequently, we have demonstrated that knock-in of rat α1 and other ouabain-resistant Na-K-ATPase mutants into PY-17 cells further reduces the expression of the residual endogenous pig α1, producing stable cell lines that express over 95% of exogenous Na-K-ATPase, and therefore making it possible to study the expressed mutant without significant interference from endogenous α1 Na-K-ATPase (23, 52). Ouabain selection of α2 cDNA-transfected PY-17 cells yielded numerous clones. Six clones were randomly selected and expanded in the absence of ouabain for three generations. Western blot analyses revealed varying levels of α2 expression in these clones. Three clones named LX-α2-2, LX-α2–4, and LX-α2–5 were further expanded and analyzed. The rat α1-rescued PY-17, called AAC-19 cells, were used as a control. As expected, no α2 signal was detected in AAC-19 cells (not shown), but variable levels of α2 expression were detected in the selected clones. As depicted in Fig. 1A, LX-α2–4 cells expressed the highest amount of α2 compared with LX-α2-2 and LX-α2–5 cells. No detectable expression of endogenous α1 was observed in LX-α2–4 cells, whereas the other two cell lines expressed varying amounts of α1 compared with control AAC-19 cells (Fig. 1B). To further assess the residual level of expression of endogenous α1 in various cell lines, we repeated the above experiments using the parent PY-17 cell line as a control. As depicted in Fig. 1C, endogenous expression of α1 in LX-α2–5 was comparable to that in PY-17 cells, whereas LX-α2-2 cells expressed much more α1 than PY-17 cells. Strikingly, the level of endogenous α1 expression in LX-α2–4 cells was so low that it could not be detected in conditions that were sufficient to detect the residual amount of expression in PY-17 (Fig. 1D). Taking advantage of the fact that surface expression of the pig (endogenous) ouabain-sensitive Na-K-ATPase α1, but not that of the modified ouabain-insensitive rat α2, could be detected by 3H-ouabain binding studies, we next opted for this sensitive approach to determine whether any residual endogenous α1 in LX-α2–4 persisted. The results indicated that 3H-ouabain binding was 24 ± 2% of that detected in PY-17 (data not shown). By inference to the 3H-ouabain binding study of Liang et al. (30), which determined that PY-17 cells express 14% of the total amount of α1, the level of residual expression of endogenous Na-K-ATPase α1 in LX-α2–4 was therefore estimated to be about 3%.
Assembly of α and β subunits is crucial for normal ion pumping function of the Na-K-ATPase. We showed that knockdown of α1 results in downregulation of β1 expression and glycosylation in PY-17 cells (29) and that heterologous expression of rat α1 is sufficient to restore the expression and glycosylation of β1 in AAC-19 cells (52). Here, we observed that exogenous expression of rat α2 restored β1 expression and glycosylation as effectively as rat α1 expression did in the earlier study (Fig. 2), suggesting that the expressed α2 was able to assemble with the endogenous β1 subunit into a fully functional Na-K-ATPase α2 in the established cell lines.
To directly assess the ion-pumping function of the expressed α2 Na-K-ATPases, the difference in ATPase activity observed in the absence and in the presence of a concentration of 1 mM ouabain was measured. This standard protocol to measure both high-affinity and low-affinity forms of Na-K-ATPase isozymes (detailed in materials and methods) allowed us to compare the catalytic capacity of the relatively insensitive forms expressed in ACC-19 and in LX-α2 lines (rat α1 and modified rat α2, respectively). As shown in Fig. 3, Na-K-ATPase activity was comparable in AAC-19 and LX-α2–4 cells, but significant differences were noted between LX-α2–4 and the two other α2-expressing cell lines.
As depicted in Fig. 1, LX-α2–4 cells expressed no endogenous Na-K-ATPase α1 detectable by Western blot under the same experimental conditions as PY-17 cells. It is clear that these cells have much less α1 Na-K-ATPase, if any, than that in PY-17 cells. Because we have previously demonstrated that the expression level of Na-K-ATPase α1 in PY-17 cells is too low to allow ouabain-mediated signal transduction (29), we considered it reasonable to assume that the minimal level of α1 Na-K-ATPase remaining in LX-α2–4 cells detected by 3H-ouabain binding would not be sufficient to mediate ouabain-induced signal transduction. Moreover, LX-α2–4 cells have ouabain-sensitive ATPase activity comparable to that of control AAC-19 cells. Therefore, we decided to use these cells in the following experiments to characterize the signaling function of Na-K-ATPase α2. Rat α1 rescued AAC-19 cells were used as positive control for comparison. Other cell lines were used as additional controls in selected experiments.
Src and ERK kinases in Na-K-ATPase α2-expressing cell lines.
Previous studies have shown that Na-K-ATPase α1 interacts and regulates Src, thereby modulating Src effectors such as ERK (23, 26, 29, 52). Thus we measured total Src and active Src in LX-α2–4 and AAC-19 cells. As shown in Fig. 4A, no difference was noted in total Src expression between these two cell lines. However, a modest yet significant increase in Src kinase activity was observed in LX-α2–4 cells compared with AAC-19 cells as measured by Western blot analysis of Src phosphorylation at Y418 (pY418; Fig. 4B). Given this modest yet statistically significant difference, the downstream target in the signaling cascade, ERK, was investigated. A large increase in ERK kinase activity was found in LX-α2–4 cells compared with control AAC-19 cells (Fig. 4C). To confirm that this effect was due to the expression of α2, we compared ERK activity in LX-α2–4 and LX-α2-2 cells, since the latter expressed less α2 and more α1. As depicted in Fig. 4D, ERK kinase activity in LX-α2-2 cells was much lower than that in LX-α2–4 cells.
Ouabain-induced signal transduction in α2-expressing cell lines.
To assess whether Na-K-ATPase α2 is capable of functioning as a receptor for cardiotonic steroids to regulate protein kinases, we treated both control α1-expressing AAC-19 cells and LX-α2–4 cells with ouabain for different times. As previously reported (23, 29, 52), ouabain stimulated Src and ERK in the α1-expressing AAC-19 cells in a time-dependent manner (Fig. 5A). In contrast, ouabain was unable to activate Src or ERK in LX-α2–4 cells (Fig. 5B). Ouabain is also known to propagate lipid kinase cascades such as the PI3K/Akt pathway in many types of cells. Again, in contrast to AAC-19 cells, ouabain failed to stimulate Akt in LX-α2–4 cells (Fig. 5, C and D).
Src/Na-K-ATPase interaction in Na-K-ATPase α2-expressing cell lines.
We next immunoprecipitated Src from AAC-19 and LX-α2–4 cells, and then probed for α1 and α2, respectively. As previously reported (29, 52), anti-Src antibody coprecipitated α1 from AAC-19 cells (Fig. 6). The average yield calculated from Src signal in the eluate vs. Src signal in the input was similar for all cell types and was estimated at 40%. After adjusting for the input, ∼8% of Na-K-ATPase α1 was coprecipitated. When the same experiments were repeated in LX-α2–4 cells, anti-Src antibody was also able to coprecipitate α2. However, after adjustment for the input, only 3% of Na-K-ATPase α2 was coprecipitated, which was significantly less than that of coprecipitated Na-K-ATPase α1 (Fig. 6). Taken together, these findings provide further support for the notion that Na-K-ATPase α2 is unable to constitute a functional receptor for cardiotonic steroids to regulate protein and lipid kinase cascades.
EGF signaling in Na-K-ATPase α2-expressing cell lines.
To verify that the above-mentioned defects in ouabain signaling were not due to intrinsic signaling impairment in the cell, we also measured EGF-mediated activation of ERK. Both control AAC-19 and LX-α2–4 cells were exposed to EGF, and cell lysates were subjected to Western blot analyses of EGFR phosphorylation and ERK activation. Phosphorylation of Tyr1173 was increased by EGF in both control AAC-19 and LX-α2–4 cells (Fig. 7), indicating that the EGFR was fully functional in LX-α2–4 cells. Additionally, EGF was able to stimulate ERK in both LX-α2–4 and control AAC-19 cells (Fig. 7).
Caveolin-1 expression in Na-K-ATPase α2-expressing cell lines.
To further assess whether Na-K-ATPase α2 is capable of signal transduction, we examined the expression of caveolin-1. Caveolin-1 is a scaffolding protein and the main component of caveolae that are critical in ouabain-induced signal transduction (7, 34, 48). We have shown that knockdown of α1 expression increases endocytosis and degradation of caveolin-1, resulting in a significant decrease in total cellular caveolin-1 in PY-17 cells (6). Expression of rat α1 restored caveolin-1 expression in AAC-19 cells (52). In contrast, we here observe that caveolin-1 expression in LX-α2–4 cells remained low (Fig. 8). Interestingly, we also observed that caveolin-1 expression was fully restored in LX-α2-2 cells where α1 expression was much higher and α2 expression was much lower.
Effect of Na-K-ATPase α2 expression on cell growth.
Because Na-K-ATPase α1-mediated Src regulation plays an important role in the regulation of cell growth (23, 52), cell growth curves were compared, to assess the functional impact of the defect of α2 cells in Src regulation. As depicted in Fig. 9, growth of LX-α2–4 cells was much slower than that of the control AAC-19 cells. To verify that this inhibition was indeed due to the expression of α2, we also measured cell growth in LX-α2-2 cells. Consistently, we observed that those cells grew faster than LX-α2–4, but slower than AAC-19 (Fig. 9).
The α2 NaKtide sequence has less effect on Src in vitro than the NaKtide sequence derived from the α1 polypeptide sequence.
To understand the structural basis for these observed differences between α1 and α2 in signal transduction, we focused on the NaKtide sequence. We have previously demonstrated that the regulation of Src by Na-K-ATPase α1 depends on the interaction between the NaKtide sequence of the α1 N domain and Src kinase domain (23, 26). As depicted in the sequence alignment shown in Fig. 10A, the NaKtide sequence from human α2 subunit is different from that of α1. Consistently, in vitro kinase assays indicated that α2 NaKtide is much less effective in inhibiting Src than α1 NaKtide (Fig. 10B). Moreover, when the first two different amino acids of α2 NaKtide were mutated to correspond to the α1 sequence (Fig. 10A), the inhibitory effect of this mutant peptide on Src was significantly increased (Fig. 10B). Finally, sequence comparisons revealed that the observed sequence differences between human α2 and α1 are conserved in all mammalian α2 subunits (e.g., as listed in Fig. 10C).
Here we report the generation of a novel mammalian cell line that expresses Na-K-ATPase α2 (LX-α2–4) in the absence of detectable expression of Na-K-ATPase α1. Functional studies reveal that total Na-K-ATPase in LX-α2–4 cells is similar to that in AAC-19 control cells, an α1-cell line. However, LX-α2–4 cells exhibit increased basal Src/ERK activity and fail to support ouabain-induced signal transduction.
Since the initial discovery of the α2 isoform in rodents (46), researchers have attempted to determine whether the α1 and α2 isoforms serve different roles (4). Initial studies using various expression systems have revealed subtle kinetic differences between these two isoforms in ion transport (3). Studies to explore the signaling aspect of the Na-K-ATPase have primarily been conducted in cells where α1 is the only α isoform expressed (e.g., renal epithelial cells) or is predominantly expressed (e.g., cardiac myocytes) (2, 27). A α2-specific mammalian cell did not previously exist, making it very difficult to assess the specific signaling function of Na-K-ATPase α2.
The first indication that Na-K-ATPase α2 may not possess signaling capabilities similar to α1 came from the studies performed in insect Sf-9 cells (40). However, Sf-9 cells do not contain functional Na-K-ATPase and most likely have different signal transduction pathways than mammalian cells. They also grow in acidic medium containing high K+, with most of the Na-K-ATPase expressed via baculoviral infection being inactive (13). We were also unable to detect Src kinase or caveolin-1 using the commercially available antibodies in these cells (40). Thus, it became important to verify the findings made in Sf-9 cells in a mammalian cell line. Because of recent success in generating α1-expressing cell lines that lack signaling function (23, 52), we tested whether we could use the protocol used for α1 knockdown/knock-in cells to create a stable cell line expressing only α2 Na-K-ATPase. It is important to note a major difference between our expression system and commonly used mammalian expression systems. For example, studies that have specifically investigated this issue have reported that the level of expression of endogenous α1 in COS-1, HEK-293, or MDCK cells is about 50% of total Na-K-ATPase α1 (8, 43). Given the high level of endogenous α1, most assays have to be conducted in the presence of ouabain (e.g., 10 μM) to eliminate the contribution from endogenous α1 Na-K-ATPase. Although this is appropriate for analyses of membrane preparations, it would compromise the interpretation of experimental findings conducted in live cells, because the binding of ouabain to the endogenous α1 not only inhibits ion pumping but also activates signaling in this pool of Na-K-ATPases. As shown in Fig. 1, although we have generated several cell lines that predominantly express Na-K-ATPase α2 with residual endogenous α1, the LX-α2–4 clone did not contain any detectable α1, which was critical to minimize inherent cofounders and more accurately assess α2-specific properties.
While LX-α2–4 cells exhibited ouabain-sensitive ATPase activity comparable to AAC-19 cells, the α2 cells were unable to participate in signal transduction. Several lines of evidence support this conclusion. First, they failed to effectively regulate basal Src activity, which consequently lead to a large increase in ERK activity (Fig. 4). This is reminiscent of previous findings in cells expressing the α1A420P mutant defective in Src regulation (23). Additionally, coimmunoprecipitation data indicated that α2 was not as effective in the formation of the Na-K-ATPase/Src complex as the α1 isoform (Fig. 6). This is likely due to the sequence difference between these two isoforms in the NaKtide region of the N domain, which is supported by data from in vitro kinase assays (Fig. 10B). Interestingly, the same amino acid differences are also carried by α2 sequences of other mammalian species, suggesting that these differences in isoform signaling were likely acquired long ago. Second, ouabain was unable to stimulate Src/ERK pathways or Akt in LX-α2–4 cells (Fig. 5). This is most likely due to the inability of α2 to bind and inhibit Src (Fig. 10B). We have shown that knocking down Na-K-ATPase α1 reduces the expression and glycosylation of β1 in LLC-PK1 cells (29). It also reduces the expression of caveolin-1 because of increased endocytosis and degradation of caveolin-1 through Src-mediated phosphorylation (6). While the expression of α1 was able to rescue both β1 and caveolin-1, α2 expression restored β1 (Fig. 2) but not caveolin-1 (Fig. 8), comparable to previous experiments with A420P mutant α1-rescued cells (23). This again supports the notion that α2 is fully capable of acting as a functional ion transporter, but does not interact with Src in signal transduction. It is important to note that these deficiencies were not due to intrinsic defects in signal transduction since EGF remained effective (Fig. 7). Finally, LX-α2–4 cells exhibited much slower growth than AAC-19 cells (Fig. 9), which is consistent with what we have observed in PY-17 cells or other cells that express a mutant α1 incapable of Src regulation (23, 47, 52).
The findings presented here provide strong evidence that Na-K-ATPase α2 differs from α1 in its ability to act as a signal transducer, which carries several important implications. We and others have previously shown that activation of the signaling pool of Na-K-ATPase α1 by cardiotonic steroids is capable of protecting the heart from ischemic reperfusion injury (11, 37, 38, 41). On the other hand, chronic stimulation of the Na-K-ATPase/Src receptor complex by either endogenous or infused cardiotonic steroids increases the generation of reactive oxygen species and induces cardiac hypertrophy and fibrosis in vivo (14, 20, 21, 33, 35). Furthermore, replacement of endogenous ouabain-resistant α1 with a ouabain-sensitive α1 mutant increases both cardiac hypertrophy and fibrosis in pressure-overload models (49). In addition, transgenic overexpression of Na-K-ATPase α2 but nor α1 attenuated pressure-overload-induced cardiac hypertrophy in a recent study by Correll et al. (9). Retrospectively, these in vivo findings are consistent with our new data, indicating that α1 is most likely responsible for cardiotonic steroid-induced fibrosis of the heart whereas α2, predominantly involved in ion transport, may protect the heart from pressure-overload-induced hypertrophy. Indeed, overexpression studies strongly suggested that α2-mediated protection was promoted through its ability to promote Ca2+ clearance by NCX1 more efficiently than its α1 counterpart, rather than through a modulation of the hypertrophic signaling (9). Second, it is important to note that the α1 isoform is expressed ubiquitously whereas the α2 isoform is primarily found in myocytes. This may imply that the α1 isoform functions as a housekeeping enzyme, given a dual role in ion transport and signaling transduction. Finally, α2 may be a more proficient ion pump in restoring cellular Na+ and K+ gradients during muscle contraction. This notion is supported by in vitro and in vivo data (12, 53). Overall, the expression of α1 and α2 may allow myocytes to differentially regulate growth and contractility via two distinct and separate pathways. Needless to say, further investigation is needed to test this hypothesis.
Several limitations of the current investigation should also be discussed. First, we have only looked into the Src/ERK, caveolin-1, and Akt pathways. The expression of α2 may be responsible for other nonpumping functions such as interaction with the inositol 1,4,5-trisphosphate (IP3) receptor, caveolins, and other proteins (1, 15, 25, 36). For instance, the identified caveolin-1-binding motifs as well as the binding site for IP3 receptor are conserved in both α1 and α2 isoforms (51, 54), which may explain the ability of α2 to coprecipitate Src (4, 22). Furthermore, it is important to note the well-established role of α2 in the control of smooth muscle contraction and the development of hypertension via its interaction with the Na+/Ca2+ exchanger (17, 53). Moreover, recent studies from the University of Maryland highlighted the potential cross-talk between α1- and α2-induced signal transduction (44, 55). In the present study, we mutated the rat α2 into a ouabain-resistant form to use ouabain selection for the generation of the LX-α2–4 cell line. Although many in vitro and in vivo studies have demonstrated that the ouabain-resistant α2 functions similarly to the native α2 in cardiac myocytes in terms of its membrane distribution and interaction with other regulatory proteins such as the Na+/Ca2+ exchanger (9), this inherent difference should be noted. In terms of signaling, both ouabain-sensitive and ouabain-insensitive forms of α1 have been shown to mediate signaling (for example, both porcine α1 and rat α1 can signal). Conversely, neither ouabain-sensitive nor ouabain-insensitive forms of α2 are able to mediate ouabain-induced ERK phosphorylation, irrespective of the concentration (Ref. 40 and the present study). Taken together, these observations are consistent with a model whereby heterogeneity in isoform-specific structural determinants other than those involved in ouabain binding may be conferring isoform-specific signaling functions. In support of such a model, the sequence alignment and peptide-based experiments shown in Fig. 10 suggest that the NaKtide region of the α-polypeptide is one important determinant for the control of Src phosphorylation by Na-K-ATPase. This hypothesis warrants further investigation of the structure/signaling function of Na-K-ATPase α isoforms.
Finally, LLC-PK1 cells are derived from pig kidney proximal tubules and only express α1. Although we consider it unlikely, it would be prudent to test whether α2 could function as a signal transducer when a new experimental model is established in myocytes. Nevertheless, the novel findings reported here reveal molecular insights into the distinct roles of Na-K-ATPase α1 and α2 isoforms, and the interplay of signal transduction and ion transport in myocytes, as well as the potential structural basis that accounts for these differences.
This work was supported by National Heart, Lung, and Blood Institute Grant HL-109015.
No conflicts of interest, financial or otherwise, are declared by the author(s).
J.X., Q. Ye, and Z.X. conception and design of research; J.X., Q. Ye, X.C., N.M., and Q. Yi performed experiments; J.X., Q. Ye, X.C., N.M., and Q. Yi analyzed data; J.X., Q. Ye, X.C., N.M., Q. Yi, and S.V.P. interpreted results of experiments; J.X., Q. Ye, X.C., and N.M. prepared figures; J.X. and Q. Ye drafted manuscript; J.X., Q. Ye, X.C., S.V.P., and Z.X. edited and revised manuscript; S.V.P. and Z.X. approved final version of manuscript.
We thank Mano Tillekeratne for technical assistance in the generation of various cell lines used in the study.
- Copyright © 2015 the American Physiological Society