Extracellular potassium dependence of the Na⁺-K⁺-ATPase in cardiac myocytes: isoform specificity and effect of phospholemman

Abstract

Cardiac Na⁺-K⁺-ATPase (NKA) regulates intracellular Na⁺, which in turn affects intracellular Ca²⁺ and contractility via the Na⁺/Ca²⁺ exchanger. Extracellular K⁺ concentration ([K⁺]) is a central regulator of NKA activity. Phospholemman (PLM) has recently been recognized as a critical regulator of NKA in the heart. PLM reduces the intracellular Na⁺ affinity of NKA, an effect relieved by PLM phosphorylation. Here we tested whether the NKA α₁- vs. α₂- isoforms have different external K⁺ sensitivity and whether PLM and PKA activation affects the NKA affinity for K⁺ in mouse cardiac myocytes. We measured the external [K⁺] dependence of the pump current generated by the ouabain-resistant NKA isoform in myocytes from wild-type (WT) mice (i.e., current due to NKA-α₁) and mice in which the NKA isoforms have swapped ouabain affinities (α₁ is ouabain sensitive and α₂ is ouabain resistant) to assess current due to NKA-α₂. We found that NKA-α₁ has a higher affinity for external K⁺ than NKA-α₂ [half-maximal pump activation (K_0.5) = 1.5 ± 0.1 vs. 2.9 ± 0.3 mM]. The apparent external K⁺ affinity of NKA was significantly lower in myocytes from WT vs. PLM-knockout mice (K_0.5 = 2.0 ± 0.2 vs. 1.05 ± 0.08 mM). However, PKA activation by isoproterenol (1 μM) did not alter the K_0.5 of NKA for external K⁺ in WT myocytes. We conclude that 1) NKA-α₁ has higher affinity for K⁺ than NKA-α₂ in cardiac myocytes, 2) PLM decreases the apparent external K⁺ affinity of NKA, and 3) phosphorylation of PLM at the cytosolic domain does not alter apparent extracellular K⁺ affinity of NKA.

the na⁺-k⁺-atpase (nka) is the ubiquitous transmembrane protein that establishes and maintains the Na⁺ and K⁺ gradients across the plasma membrane in most animal cells. The enzyme transports three Na⁺ ions out in exchange for two K⁺ ions brought into the cell, using the energy from the hydrolysis of one ATP molecule. The established electrochemical gradient is critical in maintaining the osmotic balance and the resting membrane potential of most cells, as well as the excitability of muscle cells and neurons (3, 20, 38). In the heart, NKA can greatly affect cardiac contractility and arrythmogenesis, since small changes in intracellular Na⁺ concentration ([Na⁺]_i) can have major effects on intracellular Ca²⁺, and thus contractility, via the Na⁺/Ca²⁺ exchanger (3).

NKA is activated by Na⁺ and ATP at cytoplasmic sites and by K⁺ at extracellular sites. The [Na⁺]_i for half-maximal pump activation (K_0.5) in the heart varies widely with the internal and external ionic conditions and is in the range of 8–22 mM and near the resting [Na⁺]_i in most cells (20, 35). The activating apparent K_0.5 for extracellular K⁺, in the presence of normal external Na⁺, is 1–2 mM (20). Therefore the pump is ∼70% saturated with respect to external K⁺ at a normal concentration of 4 mM.

The NKA is composed of two essential subunits, α, which contains the binding sites for Na⁺, K⁺, and ATP, and β, which plays a role in the processing and the proper membrane insertion of the pump. There are multiple NKA α-subunit isoforms in the heart and they may function differently in cardiac myocytes depending on differential membrane localization and/or regulation. Thus, the first aim of this study was to investigate the external K⁺ dependence of NKA α₁- and α₂-subunit isoforms in cardiac myocytes. In rodents, the dominant NKA α₁-isoform has an affinity for ouabain that is ∼100 times lower than that of NKA-α₂. This difference in the ouabain affinity is dictated by two residues (111 and 122) in the extracellular loop between the first and second transmembrane segments (24). Dostanic et al. (13) mutated these two residues and generated transgenic mice that have swapped ouabain affinities (i.e., NKA-α₁ is ouabain sensitive and NKA-α₂ is ouabain resistant; SWAP mice). Thus, we determined here the K⁺ dependence of the Na⁺-K⁺-ATPase current (I_Pump) generated by the ouabain-resistant isoform [α₁ in wild-type (WT) and α₂ in SWAP mice] while the ouabain-sensitive isoform was blocked with a low concentration of ouabain.

Recent evidence indicates that NKA is regulated by phospholemman (PLM) in the heart. PLM (FXYD1) is a member of the FXYD family of proteins that interact with and regulate the NKA in a tissue-specific manner (7, 16–19, 36). The conserved FXYD motif and residues (two transmembrane glycines and a serine residue at the membrane-cytoplasm interface) may be involved in an essential function common to all FXYD proteins, such as the structural interaction with the NKA. Indeed, it has been shown that the FXYD motif is required for this structural interaction (1). Mutational analysis and modeling indicate that transmembrane (TM) domain 9 (TM9) of the NKA α-subunit exposes one face of the helix that interacts with FXYD proteins (specifically FXYD-2, -4, and -7) and contributes to the stable interaction with FXYD proteins, as well as mediating the effect of FXYD proteins on the apparent K⁺ affinity of NKA (31).

PLM is the only FXYD protein present in cardiac myocytes, where it is a major substrate for phosphorylation by both PKA (at site Ser68) and PKC (at sites Ser63 and Ser68). In Xenopus oocytes when PLM was coexpressed with rat NKA-α₁ and α₂ (6), PLM reduced the affinity for intracellular Na⁺ (K_0.5 increased from 9.3 to 16.5 mM for NKA-α₁ and from 13.6 to 20.5 mM for NKA-α₂) and external K⁺ (from 0.49 to 0.67 mM for α₁ and from 0.8 to 1.1 mM for α₂). In intact cardiac myocytes we have previously shown that PLM decreases the Na⁺ affinity of NKA and that PLM phosphorylation mediates the PKA/PKC-dependent effects on NKA (10, 21). Thus, the second aim of this paper was to investigate whether PLM regulates the external K⁺ dependence of NKA and whether this is affected by PLM phosphorylation in intact cardiac myocytes.

MATERIALS AND METHODS

Generation of PLM-knockout and SWAP mice.

PLM-knockout (KO) mice were generated as previously described (25) except that they are now congenic on a pure C57B/6 background. Heterozygous breeding pairs were used to generate PLM-KO and WT littermates. Mice 3–4 mo of age were used, before PLM-KO mice developed hyperthrophy.

Mice expressing ouabain-sensitive NKA-α₁ and ouabain-resistant NKA-α₂ (α₁^SSα₂^RR; SWAP) were developed by mating mice with a ouabain-sensitive NKA α₁-isoform (α₁^SSα₂^SS) (13) with mice having a ouabain-resistant NKA α₂-isoform (α₁^RRα₂^RR) (12), as previously described (13). The expression and tissue distribution of NKA α₁- and α₂-isoforms are normal in the SWAP animals, and the mutations did not alter the enzymatic activity of the two isoforms (13).

All animal protocols were approved by the Animal Care and Use Committee at University of California Davis, University of Virginia or University of Cincinnati.

Myocyte isolation.

Mouse ventricular myocytes were isolated as previously described (8). Briefly, hearts were excised quickly after animals were anesthetized (inhalation of 5% isoflurane) and mounted on a gravity-driven Langendorff perfusion apparatus. Hearts were perfused for 5 min at 37°C with nominally Ca²⁺-free solution with 24 mmol/l NaHCO₃ bubbled with 95% O₂-5% CO₂ (pH 7.4). Perfusion was then switched to DMEM solution containing 1.27 mg/ml collagenase (type B, Boehringer Mannheim, Indianapolis, IN). When hearts became flaccid (about 6–8 min), ventricular tissue was removed, dispersed, and filtered, and myocyte suspensions were rinsed several times.

Na⁺-K⁺-ATPase current measurement.

Na⁺-K⁺-ATPase current (I_Pump) was measured in whole cell voltage-clamp configuration at −20 mV with 100 mM Na⁺ (saturating concentration) in the pipette (low-resistance electrodes with initial resistance of ≤2 MΩ). I_Pump was identified as extracellular K⁺-activated (and ouabain sensitive) current at room temperature. Pipette solution contained (in mM) 10 NaCl, 20 KCl, 90 Na-aspartate, 20 TEA-Cl, 10 HEPES, 5 Mg-ATP, 0.7 MgCl₂ (∼1 mM free Mg), 3 BAPTA, and 1.15 CaCl₂ (∼100 nM free Ca), pH 7.2. External K⁺-free solution contained (in mM) 125 NaCl, 5 NiCl₂, 2 BaCl₂, 1 MgCl₂, 15 Tris-Cl, 5 HEPES, and 5 glucose (pH 7.4). Tris-Cl was replaced with KCl in the K⁺-containing solution. There is little change in I_Pump (<10%) over the duration of the experiments here, tested by repeated I_Pump activation with the same extracellular K⁺ concentration ([K⁺]_o) (not shown).

After whole cell configuration was achieved, myocytes were incubated in K⁺-free external solution for 10 min. I_Pump was then activated by adding back external K⁺ (15 mM when no ouabain was used and 8 mM when using ouabain to select the NKA α₁- or α₂-isoforms). Steady-state currents were recorded in the presence of decreasing external [K⁺] to determine the K⁺ affinity. Extracellular K⁺ reduces the NKA affinity for ouabain; however, most of the effect occurs at low [K⁺] (22). For our analysis here, where the same ouabain concentration was used to assess NKA-α₁ and -α₂ (in SWAP and WT mice), this effect is not expected to affect the conclusions concerning relative [K⁺] dependence of the isoforms.

Na⁺-K⁺-ATPase activity.

Sarcolemmal membranes were prepared as described (26) with modifications (10). Ouabain-sensitive Na⁺-K⁺-ATPase activity was measured at 37°C in a spectrophotometric enzyme-coupled assay as described (29). Sarcolemma (∼2 μg) was incubated in medium containing (in mmol/l) 30 Tris·HCl, 0 to 20 KCl, 3 MgCl₂, 1 EDTA, 1.5 phospho(enol)pyruvate, 1 NADH, 3 ATP, ±10 ouabain, and 4.6 units of lactic dehydrogenase and 3.3 units of pyruvate kinase at pH 7.2, with 120 mmol/l [NaCl]. The decline in NADH absorbance at 340 nm is used to calculate ATPase rate sensitive to 10 mmol/l ouabain.

Statistical analysis.

Data are expressed as means ± SE. Statistical discriminations were performed with Student's t-test (paired when appropriate), with P < 0.05 considered significant.

RESULTS

Measurement of external K⁺ affinity of NKA-α₁ and -α₂ in cardiac myocytes.

I_Pump was measured in ventricular myocytes from WT and SWAP mice using whole cell voltage clamp at high intracellular Na⁺ (100 mM Na⁺ in the pipette solution). Steady-state currents were recorded in the presence of decreasing external K⁺ concentration (Fig. 1). I_Pump was then plotted as a function of external K⁺, and the curve was fitted with a Hill expression to determine the maximum current (I_max), the K⁺ concentration where I_Pump is half-activated (K_0.5), and the Hill coefficient (n_Hill). When I_Pump was measured in WT mouse myocytes (containing both NKA α₁- and α₂-isoforms), I_max was 1.82 ± 0.07 pA/pF, K_0.5 = 2.0 ± 0.2 mM, and n_Hill = 1.10 ± 0.06.

Fig. 1.

Experimental protocol to measure external K⁺ affinity of Na⁺-K⁺-ATPase (NKA). After reaching the whole cell configuration, myocytes were maintained 10 min in K⁺-free external solution. Then, NKA current (I_Pump) was activated by adding back K⁺. Steady-state currents were recorded on stepwise changes in concentration of extracellular K⁺ (K_o).

WT mouse NKA α₁- and α₂-isoforms have distinct ouabain affinities: NKA-α₂ is ouabain sensitive, with a K_0.5 of 0.3 μM, and NKA-α₁ is ouabain resistant (K_0.5 = 105 μM) (2). This is reversed in the SWAP mice, where NKA-α₂ is ouabain insensitive while NKA-α₁ is ouabain sensitive. We inhibited the ouabain-sensitive NKA isoform in WT and SWAP mice (α₂ and α₁, respectively) with 20 μM ouabain and measured the K⁺ dependence of the I_Pump mediated by the ouabain-resistant isoform (NKA-α₁ in WT and -α₂ in SWAP mice) (Fig. 2A). The relatively high ouabain concentration ensured that the ouabain-sensitive NKA isoform is completely blocked (98%) and thus prevented contamination of the I_Pump generated by the ouabain-resistant NKA with the ouabain-sensitive current. NKA-α₁ generated a maximum I_Pump of 1.2 ± 0.2 pA/pF, which is ∼65% of the total pump current, while the maximum current produced by NKA-α₂ was 0.41 ± 0.05 pA/pF, ∼23% of total I_Pump (Fig. 2B). The combined maximum I_Pump generated by NKA-α₁ and -α₂ is less than the total current measured in WT or SWAP mice (∼88%), because 20 μM ouabain also blocks a small fraction (∼16%) of the ouabain-resistant NKA. This does not affect in any way the K⁺ dose-response curves or the estimation of the external K⁺ affinity of NKA-α₁ and -α₂. It only reduces the apparent maximum I_Pump of each isoform similarly.

Fig. 2.

Extracellular K⁺ affinity of NKA-α₁ and -α₂. The ouabain-sensitive NKA isoform was blocked with 20 μM ouabain, and we measured the K⁺ dependence of the ouabain-resistant I_Pump [NKA-α₁ in wild-type (WT) and NKA-α₂ in SWAP mice]. A: I_Pump dependence on external K⁺ in myocytes from WT mice with no ouabain, i.e., I_Pump is generated by both NKA α₁- and α₂-isoforms (n = 13, ·); WT mice in the presence of ouabain, i.e., I_Pump is generated by NKA-α₁ (n = 8, ○); and SWAP mice in the presence of ouabain, i.e., I_Pump is due to NKA-α₂ (n = 12, □). All points are means ± SE. For SWAP mice, error bars are smaller than symbol. [K⁺]_o, extracellular K⁺ concentration. B: maximum I_Pump in these three groups. I_max, maximum current. C: half-maximal pump activation (K_0.5) for external K⁺ in these three groups. Values are means ± SE. **P < 0.01. ***P < 0.001.

The K_0.5 for external K⁺ was 1.3 ± 0.1 mM in WT mice (NKA α₁-isoform) and 2.9 ± 0.3 mM in SWAP mice (NKA-α₂). Thus, NKA-α₁ has a higher affinity for external K⁺ than NKA-α₂, and would be more fully activated at any given extracellular [K⁺].

Regulation of external K⁺ affinity by PLM in cardiac myocytes.

To investigate the effect of PLM on the K⁺ affinity of NKA, we performed similar I_Pump measurements in myocytes isolated from WT and PLM-KO mice (Fig. 3). I_max is smaller in myocytes from PLM-KO mice compared with WT (1.51 ± 0.03 vs. 1.82 ± 0.07 pA/pF). However, there is a 25% decrease in the NKA expression level in the PLM-KO mice, as previously shown (10, 25). When normalized to the same NKA level, I_max is actually higher (2.01 pA/pF) in PLM-KO vs. WT mice. The data shown in Fig. 3A are normalized to the maximum I_Pump to emphasize the effect of PLM on the K_0.5. Knockout of PLM significantly increased the affinity of NKA for external K⁺ as shown by a decrease in K_0.5 (2.0 ± 0.2 vs. 1.05 ± 0.08 mM in myocytes from WT and PLM-KO mice, respectively; Fig. 3B). Thus, PLM decreases the apparent NKA affinity for external [K⁺] in cardiac myocytes.

Fig. 3.

K⁺ dependence of I_Pump in cardiac myocytes from WT and phospholemman-knockout (PLM-KO) mice. A: I_Pump was normalized to the maximum current to emphasize the effect of PLM on the pump affinity for external K⁺. Shown are means ± SE from 13 WT and 8 PLM-KO myocytes. For some points, error bars are smaller than symbol. B: K_0.5 and I_max for external K⁺ in cardiac myocytes from WT and PLM-KO mice. **P < 0.01.

We also assessed the K⁺ affinity of NKA using an ATPase activity assay in sarcolemmal vesicles. Fig. 4A shows the normalized curve for the K⁺ dependence of the ATPase activity in WT and PLM-KO mice. Figure 4B shows the mean K_0.5 and V_max of NKA for external K⁺ from WT and PLM-KO mice. In WT mice, NKA had a K_0.5 of 2.7 ± 0.3 mM. In PLM-KO mice, NKA had a significantly higher affinity for external K⁺ (K_0.5 = 1.5 ± 0.2 mM). The ATPase V_max in the PLM-KO group is significantly lower than that in the WT group (0.67 ± 0.03 vs. 0.86 ± 0.01); however, this is again consistent with the decrease in the NKA expression level in PLM-KO mice. The data are consistent with those obtained from I_Pump measurements, suggesting that PLM decreases the NKA affinity for external K⁺.

Fig. 4.

K⁺ dependence of NKA ATPase activity in sarcolemmal vesicles from WT (n = 5) and PLM-KO (n = 9) mice. A: normalized curve for the K⁺ dependence of ATPase activity in WT and PLM-KO mice. Data are means ± SE of 5 myocytes from WT mice and 9 from PLM-KO mice. For some points, error bars are smaller than symbol. B: K_0.5 and V_max for external K⁺ in sarcolemmal vesicles from WT and PLM-KO mice. **P < 0.01.

Effect of isoproterenol on the external K⁺ affinity of NKA.

We have previously shown that PKA activation by isoproterenol (Iso) phosphorylates PLM and thus relieves the PLM-dependent inhibition of intracellular Na⁺ affinity of NKA (10). Here we tested whether PLM phosphorylation by PKA also increases the external K⁺ affinity of NKA (relieves PLM-induced reduction in K⁺ affinity). We measured I_Pump in myocytes from WT mice at decreasing external K⁺, in the absence and in the presence of 1 μM Iso (Fig. 5). In mouse myocytes, Iso phosphorylates PLM at Ser68, with a maximal effect occurring in 10–15 min (10). Thus, with Iso, I_Pump measurements were taken 10 min after Iso application to allow for PKA activation and PLM phosphorylation. The result indicated that I_Pump was similar with and without Iso, i.e., no significant alteration in either I_max, K_0.5, or n_Hill by Iso application. Thus, phosphorylation of PLM does not relieve the PLM-induced inhibition of extracellular K⁺ dependence of NKA activity. We conclude that PLM phosphorylation does not affect the K⁺ affinity of either NKA-α₁ or -α₂. However, our results do not rule out the possibility that NKA-α₁ and -α₂ are affected in opposite directions by PLM phosphorylation and that the effects cancel each other in WT mice. That seems unlikely, because in Xenopus oocytes, Bibert et al. (4) could not detect any effect of PLM phosphorylation by PKA on K⁺ affinity of either NKA-α₁ or α₂.

Fig. 5.

A: effect of isoproterenol (Iso) on the external K⁺ dependence of I_Pump in myocytes from WT mice (n = 13). B: effect of Iso on I_max. C: effect of Iso on K_0.5 for external K⁺. Data in B and C were obtained by fitting the curve in A with a Hill expression.

DISCUSSION

The present study shows that, in adult ventricular myocytes, 1) NKA-α₁ has a higher affinity for extracellular K⁺ than NKA-α₂ (based on measurements of [K⁺]_o dependence of I_Pump in myocytes from WT and SWAP mice in the presence of 20 μM ouabain), 2) PLM reduces the apparent NKA K⁺ affinity (I_Pump and ATPase activity measurements in myocytes from WT vs. PLM-KO mice), and 3) PLM phosphorylation by PKA (upon Iso stimulation) does not affect the K⁺ dependence of NKA activity.

Different external K⁺ affinity of NKA-α₁ and -α₂.

There are multiple NKA isoforms in cardiac myocytes. NKA-α₁ is the dominant, ubiquitous isoform, whereas NKA-α₂ and NKA-α₃ are present in relatively small amounts and in a species-dependent manner (32). It has been proposed that NKA-α₂ and NKA-α₃ are located mainly in the T tubules, at the junctions with the sarcoplasmic reticulum, where they could function differently than NKA-α₁, that is, by regulating local Na⁺/Ca²⁺ exchange and thus cardiac myocyte Ca²⁺ signaling. There is rather convincing evidence supporting such a model in smooth muscle (28). However, there is less compelling evidence in cardiac myocytes. Notably, the T tubules can also experience local [K⁺] gradients, and differential K⁺ dependence of NKA isoforms could be functionally relevant. In this context, it is important to determine whether different NKA isoforms have different affinities for external K⁺ in the ventricular myocyte environment.

Previous studies of NKA isoform function have mainly used either heterologous expression systems or myocyte measurements of the total and ouabain-resistant NKA activity (with the activity of the ouabain-sensitive component derived only by subtraction). Heterologous expression of NKA isoforms is useful; however, it lacks the physiological myocyte environment for the NKA. The subtraction of ouabain-resistant from total NKA activity in myocytes is practical and useful for studying the ouabain-resistant NKA-α₁ isoform in myocytes. However, since α₁ is the predominant isoform, the subtraction of two large numbers introduces large errors with respect to inferences regarding NKA-α₂ activity (2). Thus, the SWAP mice, in which NKA-α₁ is ouabain sensitive and NKA-α₂ is ouabain resistant (13), are an excellent tool for the study of functional properties of NKA-α₂. Swapping the ouabain affinity only involves minor changes in the coding sequence (two amino acids), and these mutations do not affect other NKA properties (13). Combining measurements in WT and SWAP mice with the ouabain-sensitive isoform blocked, we could thus measure separately and directly the external K⁺-dependence of NKA-α₁ (in WT mice) and NKA-α₂ (in SWAP mice). We do not see evidence for appreciable NKA-α₃ in mouse ventricular myocytes, and the fact that we get a similar estimate of the fraction of NKA function attributed to NKA-α₂ in both SWAP and WT mice is consistent with this idea.

We found that NKA-α₁ has a higher affinity for extracellular K⁺ than NKA-α₂. This is consistent with studies in heterologous expression systems (4–6). NKA-α₂ is concentrated in the transverse tubules (2, 8, 37), where K⁺ could accumulate during ischemia or high heart rate. Thus, the lower K⁺ affinity of NKA-α₂ may provide pump reserve to ensure that NKA-α₂ is able to respond to an increased local [K⁺]. This could be important in both limiting membrane depolarization and keeping [Na⁺]_i low (thereby allowing Na⁺/Ca²⁺ exchange to effectively extrude Ca²⁺ and prevent Ca²⁺ overload). The lower extracellular K⁺ affinity of NKA-α₂ would also mean that, at a given [K⁺]_o, these NKA molecules will be working at a lower level (N.B. that the intracellular Na⁺ dependence is not different between NKA-α₁ and -α₂ in mouse myocytes; 11). If total Na⁺-K⁺-ATPase molecules are evenly distributed in the transverse tubule (9), and if NKA-α₂ are mainly at the junctional cleft (still an open question), then the lower K⁺ affinity of NKA-α₂ could in principle contribute to locally elevated [Na⁺]_i, and thus [Ca²⁺]_i, in the junctional cleft. This idea requires further study.

Na⁺ and K⁺-binding sites on NKA: relationship to PLM.

We found that PLM reduces the external K⁺ affinity of NKA in mouse cardiac myocytes, consistent with former work in heterologous systems (4–6). We previously showed that PLM phosphorylation by PKA relieved the inhibitory effect of PLM on NKA (10). Remarkably, this was not the case for the effect of PLM on the extracellular K⁺ dependence. This is consistent with previous results in heterologous expression systems (4). Thus, while NKA serves to regulate both [Na⁺]_i and [K⁺]_o, the PLM-PKA-dependent NKA regulation seems only to be directly involved in regulating [Na⁺]_i (although K⁺ transport would follow). This lack of symmetry between PLM regulation of Na⁺ and K⁺ may also place important constraints on the detailed molecular mechanism by which PLM exerts its effects on NKA function. The failure to reverse PLM effects on K⁺ dependence (in contrast to Na⁺ dependence) is consistent with a model where phosphorylation alters but does not abolish the PLM-NKA interaction.

Mutagenesis studies (14, 23, 27, 39), homology modeling data (34), and the recently unveiled NKA crystal structure (33) support a model where the first and second NKA cation-binding sites are homologous to the two Ca²⁺-binding sites (sites I and II) of sarco(endo)plasmic reticulum Ca₂₊-ATPase. The two sites are alternately occupied by Na⁺ and K⁺ during the NKA pump cycle. The third Na⁺-binding site (site III) is reported to be located between TM domains 5, 6, and 9 (30). Mutations at each of the residues contributing to this site altered the Na⁺ affinity of NKA. It was also suggested that the TM segment of FXYD proteins could fit in a groove between TM2, TM6, and TM9 (15, 31), although X-ray crystal structure of kidney NKA indicated that the γ-subunit (FKYD2) is close to TM9, but not in the above mentioned groove (33).

We speculate that PLM association may influence binding of intracellular Na⁺ and extracellular K⁺ to one or both of the common sites, as well as Na⁺ binding to the third site (decreasing NKA apparent affinity for both Na⁺ and K⁺, as observed experimentally). Then PLM phosphorylation may only affect the binding of Na⁺ to the third site, with no effect on the two sites common to Na⁺ and K⁺. In this way, PLM phosphorylation would enhance affinity for intracellular Na⁺ without altering the K⁺ affinity. However, this is clearly a speculative model and likely an oversimplification. Indeed, mutational studies (31) have revealed amino acids in the TM9 of NKA α-subunit that are involved in the effect of FXYD2 and -4 on the apparent affinity for K⁺ but not Na⁺. Further structural and functional studies are required to fully understand the mechanism of NKA regulation by PLM.

GRANTS

This work was supported by grants from the National Institutes of Health grants HL-30077, HL-64724, and HL-81562 (to D. M. Bers), the Cardiovascular Research Center at University of Virginia (to A. L. Tucker), and American Heart Association Fellowship (to F. Han) and Scientist Development Grant (0735084 to S. Despa).