Review

FXYD Proteins: New Tissue-Specific Regulators of the Ubiquitous Na,K-ATPase

See allHide authors and affiliations

Science's STKE  21 Jan 2003:
Vol. 2003, Issue 166, pp. re1
DOI: 10.1126/stke.2003.166.re1

Abstract

Maintenance of the Na+ and K+ gradients between the intracellular and extracellular milieus of animal cells is a prerequisite for basic cellular homeostasis and for functions of specialized tissues. The Na,K-ATPase, an oligomeric P-type adenosine triphosphatase (ATPase), is composed of a catalytic α subunit and a regulatory β subunit and is the main player that fulfils these tasks. A variety of regulatory mechanisms are necessary to guarantee appropriate Na,K-ATPase expression and activity adapted to changing physiological demands. Recently, a regulatory mechanism was defined that is mediated by interaction of Na,K-ATPase with small proteins of the FXYD family, which possess a single transmembrane domain and so far have been considered as channels or regulators of ion channels. The mammalian FXYD proteins FXYD1 through FXYD7 exhibit tissue-specific distribution. Phospholemman (FXYD1) in heart and skeletal muscle, the γ subunit of Na,K-ATPase (FXYD2) and corticosteroid hormone-induced factor (FXYD4, also known as CHIF) in the kidney, and FXYD7 in the brain associate preferentially with the widely expressed Na,K-ATPase α1-β1 isozyme and modulate its transport activity in a way that conforms to tissue-specific requirements. Thus, tissue- and isozyme-specific interaction of Na,K-ATPase with FXYD proteins contributes to proper handling of Na+ and K+ by the Na,K-ATPase, and ensures correct function in such processes as renal Na+-reabsorption, muscle contraction, and neuronal excitability.

Introduction

The Na+,K+-pump, or Na+,K+-adenosine triphosphatase (Na,K-ATPase), is a plasma membrane transporter expressed in all tissues examined so far whose minimal functional unit consists of two subunits, a catalytic α subunit with 10 transmembrane segments, and a type II β subunit (Fig. 1). By using the energy of hydrolysis of adenosine triphosphate (ATP), the Na,K-ATPase exchanges three Na+ and two K+ ions against their respective electrochemical gradients across the cell membrane and thus fulfils its fundamental task in maintaining the characteristic Na+ and K+ gradients of animal cells. These Na+ and K+ gradients are essential to preservation of cell volume and the membrane potential; in addition, the Na+ gradient provides the energy for the activity of secondary transporters of vital importance (Fig. 1). The Na,K-ATPase is indispensable, not only for basic cellular homeostasis but also for specialized functions of various tissues. For instance, in renal epithelial cells, the Na,K-ATPase is exclusively expressed in the basolateral membrane and thus becomes the driving force for net transepithelial Na+ reabsorption, which is necessary for the maintenance of the extracellular volume and blood pressure. Moreover, in skeletal and heart muscle, the Na,K-ATPase activity is tightly coupled to the activity of a Na+/Ca2+exchanger that controls muscle contractility. Finally, in neurons and glial cells, the presence of the Na,K-ATPase assures neuronal excitability by reestablishing extracellular and intracellular K+ and Na+ homeostasis during neuronal activity.

Fig. 1.

Function, structure, and regulation of Na,K-ATPase. For description, see text.

Considering the central physiological role of the Na,K-ATPase, it is obvious that its expression and activity must be finely regulated to achieve appropriate Na,K-ATPase activity adapted to changing physiological demands.

Numerous mechanisms are involved in the regulation of Na,K-ATPase expression and function (Fig. 1). A basic requirement for cellular Na,K-ATPase expression is the stoichiometric synthesis of its α and β subunits. The β subunit acts as a specific molecular chaperone. Its association with the catalytic α subunit is necessary for the correct and stable membrane insertion of that subunit, which is required for the functional maturation of the α subunit (1). Tissue-specific differences of the Na,K-ATPase function are achieved by the expression of four α and three β isoforms, which potentially permit formation of 12 different Na,K-ATPase isozymes with distinct functional properties (2, 3). Acute modulation of Na,K-ATPase activity occurs in response to changes in the intracellular concentration of Na+, and under resting cellular conditions intracellular Na+ is limiting for the Na,K-ATPase transport activity. Moreover, certain peptide hormones and neurotransmitters that stimulate protein kinase A (PKA) or protein kinase C (PKC) provoke phosphorylation of the α subunit, which affects the transport properties of the Na,K-ATPase or the distribution of Na,K-ATPase between the plasma membrane and intracellular stores, or both (4, 5). Finally, long-term modulation of Na,K-ATPase is mediated by aldosterone and thyroid hormones, which alter α and β subunit gene transcription and ultimately produce an increased number of Na,K-ATPase units at the cell surface (5) (Fig. 1).

Recent experimental evidence suggests that changes in Na,K-ATPase activity are important not only to keep the intracellular ionic milieu constant but also as part of a signaling pathway that ultimately leads to changes in cell growth and differentiation (6). This concept of Na,K-ATPase as a signal transducer is originally based on the observation that the Na,K-ATPase acts as the pharmacological receptor for cardiac glycosides, drugs that have been used for more than 200 years to treat congestive heart failure. Recent findings suggest that certain well-known cardiotonic glycosides such as ouabain are not only toxins from plants, but may be synthesized in the mammalian adrenal gland and act as hormones (7). In cardiac myocytes, ouabain binding and partial inhibition of Na,K-ATPase lead to changes in intracellular Na+ and Ca2+ concentrations and the expected inotropic effect. However, ouabain binding also produces a signal that is transmitted by protein-protein interactions from the Na,K-ATPase in the plasma membrane to the nucleus through intracellular signaling cascades that are at least partly independent of ouabain-induced changes in intracellular ions (6).

An exciting new field in Na,K-ATPase research is concerned with the identification of the nature and function of cytoplasmic or membrane proteins that interact with Na,K-ATPase and influence cell function either by direct signal transduction or through changes in intracellular concentrations of K+, Na+, and Ca2+. Changes in ion concentrations appear to result from tissue- and isozyme-specific association of the Na,K-ATPase with small membrane proteins of the FXYD family, and from regulatory effects of the FXYD proteins on the Na,K-ATPase transport function.

The FXYD Protein Family

On the basis of sequence similarity, Sweadner and Rael (8) defined an FXYD protein family. In mammals, this family contains at least seven members, including phospholemman (PLM, or FXYD1) (9), the γ-subunit of the Na,K-ATPase (FXYD2) (10, 11), mammary tumor marker 8 (MAT-8, or FXYD3) (12), corticosteroid hormone-induced factor (CHIF, or FXYD4) (13), related to ion channel (RIC, or FXYD5) (14), phosphohippolin (FXYD6) (15), and FXYD7 (Fig. 2). Two other family members that are not obvious orthologs of any identified mammalian FXYD protein exist in zebrafish (8). All these proteins share a signature sequence of six conserved amino acids comprising the FXYD motif in the NH2-terminus, and two glycines and one serine residue in the transmembrane domain (Fig. 2). The NH2- and COOH-termini are variable among FXYD proteins. FXYD1 (9), FXYD2 (16), FXYD4 (17), and FXYD7 (18) were shown to be type I proteins with a single membrane span and the COOH-terminus exposed to the cytoplasm. FXYD1 and FXYD4, but not FXYD2 or FXYD7, adopt this membrane orientation after cleavage of a signal sequence.

Fig. 2.

The FXYD protein family. (A) Alignment (CLUSTAL X software) (64) of human FXYD1-3, rat FXYD4, and human FXYD 5-7. The extended NH2-terminus of FXYD5 has been omitted (indicated by X). Amino acids in red are identical in all FXYD proteins, with the exception of the proline (P) residue preceding the FXYD motif, which is a lysine in Xenopus FXYD2. Amino acids in white are similar in at least 80% of mammalian FXYD proteins. Sequence similarity in and adjacent to the transmembrane (TM) domain is indicated by a bar. (B) Phylogenetic tree of mammalian FXYD proteins obtained by TreeView software (65) from an alignment of all available mammalian sequences. Only FXYD3 and FXYD4 cosegregate in a common branch, whereas all other FXYD proteins diverge considerably. h, human; r, rat; m, mouse; p, pig; sh, sheep; d, dog; and b, bovine. (C) Structural and functional features of FXYD proteins. In the linear models of FXYD proteins, the positions of conserved residues are highlighted in red. The type I membrane topology of FXYD3, 5, and 6 are still hypothetical. The signal sequences in the NH2-terminus that were shown to be cleaved in FXYD1 and FXYD4 are shown in blue. The hatched bar in the NH2-terminus of FXYD5 represents a region highly enriched in proline-serine or proline-threonine doublets. The hexagonal structures in the NH2-terminus of FXYD7 represent O-glycosylated chains. Clusters of positively charged residues in the COOH-terminus of FXYD proteins are indicated by +. In the COOH-terminus of FXYD1, serine residues are indicated that are phosphorylated by protein kinases. In the TM domain of FXYD2, the glycine residue that is replaced by an arginine in cases of human primary hypomagnesemia is indicated. At the bottom, the state of knowledge on the functional effects of FXYD proteins on the affinity of different ligands of the Na,K-ATPase is summarized. The results presented for Na+ and K+ affinities have been obtained at a membrane potential of –50 mV. ↓, decrease in the apparent affinity; ↑, increase in the apparent affinity; –, no change in the apparent affinity; n.d. = not determined, ATP, adenosine triphosphate. For more details see text.

FXYD proteins are widely distributed in mammalian tissues with prominent expression in tissues that perform fluid and solute transport or that are electrically excitable (8). FXYD3 was originally identified in murine breast tumors initiated by Ras or Neu (19), and FXYD5 was identified in various cell lines transformed with the E2a-Pbx1 oncoprotein (14).

Initial functional characterization showed that FXYD1 (20), FXYD3 (12), and FXYD4 (13) induce ion-specific conductances when overexpressed in Xenopus oocytes. These results suggest that FXYD proteins act as channels or as modulators of ion channels, but the physiological relevance of these putative functions remains unclear. During the last two years, studies have revealed that most FXYD proteins have another specific function and act as tissue-specific modulators of the Na,K-ATPase. This line of investigations was guided by the observation, made more than 20 years ago, that purified renal Na,K-ATPase preparations are associated with FXYD2 (10), and by the more recent demonstration of a regulatory effect of the mammalian γ subunit, (16) and of a shark phospholemman-like protein (21), on Na,K-ATPase activity.

Regulation of Na,K-ATPase by FXYD2 (γ Subunit) and FXYD4 (CHIF) in Na+-Reabsorbing Tissues

Structural and Functional Aspects of FXYD2, the γ Subunit of Na,K-ATPase

FXYD2 was originally characterized as a small acidic proteolipid that copurified with renal Na,K-ATPase and which, similar to the α subunit, covalently bound to photoaffinity derivatives of ouabain (10). After FXYD2 was cloned (11), studies confirmed that FXYD2 indeed associates specifically with Na,K-ATPase, and showed that FXYD2 can modify the transport activity of Na,K-ATPase after expression of the proteins in Xenopus oocytes (16).

The human FXYD2 gene is located on chromosome 11q23; it spans 9.2 kilobases (kb) and includes six exons (22, 23) that encode two splice variants, FXYD2a and FXYD2b (γa and γb), which differ only in their most NH2-terminal amino acids (22, 24). In adult animals, FXYD2 is confined to the kidney (25, 26), but during development it is also expressed in other tissues (26). Interestingly, the onset of expression of FXYD2 in the amphibian kidney occurs in the pronephric stage and coincides with the expression of Na,K-ATPase α1 and β1 subunits (26). This result indicates that a common regulatory mechanism may initiate the transcription of all three of these genes during development, and highlights the potential functional importance of FXYD2 in renal Na,K-ATPase function.

In the adult kidney, FXYD2a and FXYD2b proteins show a distinct expression pattern (27, 28). Both of these proteins are abundantly expressed in the inner stripe of the outer medulla in the thick ascending limb (TAL), whereas smaller amounts of FXYD2 proteins are expressed in the cortical portion of the TAL, in proximal tubules, and in macula densa. Moreover, large amounts of FXYD2b are found in the distal convoluted tubule and the connecting tubule (28), whereas little or no FXYD2 is expressed in medullary and cortical portions of the collecting duct (27, 28). Little is known about the regulation of expression of FXYD2 under various physiological conditions. In mice induced to consume large amounts of water, urinary osmolarity is reduced to about one sixth that in control animals, and expression of FXYD2b and, to a lesser extent, that of FXYD2a are decreased in the inner medulla. Exposure of inner medullary collecting cells to hypertonic medium induces expression of FXYD2b and, to a lesser extent, that of FXYD2a (29).

FXYD2a and FXYD2b colocalize with the Na,K-ATPase in basolateral membranes of renal epithelial cells (27). Both variants can be co-immunoprecipitated from renal microsomes with antibodies to the Na,K-ATPase α subunit (17, 28, 30), but any individual Na,K-ATPase α subunit associates exclusively with either FXYD2a or FXYD2b, as suggested by co-immunoprecipitation experiments using FXYD2a-specific antibodies (30).

The functional effect of FXYD2 interaction on the Na,K-ATPase has been studied in partially purified Na,K-ATPase from cells with or without FXYD2, or after expression of the proteins in cultured cells or in Xenopus oocytes (Fig. 2C). In contrast to the β subunit, FXYD2 is not essential for the cellular expression of Na,K-ATPase (16). However, FXYD2 itself needs association with Na,K-ATPase to become stably expressed and to be efficiently transported to the cell surface (16). Electrophysiological measurements of Na,K-ATPase transport in Xenopus oocytes revealed that FXYD2 decreases the apparent affinity of the Na,K-ATPase for intracellular Na+, and modulates its apparent affinity for external K+, depending on the membrane potential (17). Moreover, it has been shown that FXYD2 mediates a parallel decrease in the Na+ and K+ activation of the ATPase activity (31) and that it increases the ATP sensitivity (25, 32) and the K+ antagonism of intracellular Na+ binding of the Na,K-ATPase (27). Although FXYD2 binds photoaffinity-labeled ouabain analogs (10), the intrinsic affinity of Na,K-ATPase for ouabain is similar in the presence or absence of FXYD2 (17).

Comparison of the functional effects of FXYD2a and FXYD2b on Na,K-ATPase activity showed that both FXYD2 variants similarly affect Na,K-ATPase function (17, 27). However, recent evidence suggests that so far uncharacterized posttranslational modifications, which may be cell-specific or may depend on the physiological state of the animal, differentially influence the functional effect of FXYD2 variants on Na,K-ATPase (33). These results provide a first rationale for the cell-specific expression of FXYD2a and FXYD2b in the kidney.

Little is known about the relationship between structure and function in FXYD2. Both the extracytoplasmic FXYD motif and the cytoplasmic, positively charged residues appear to be important for the efficient association of FXYD2 with Na,K-ATPase (17). However, short COOH-terminal deletions in FXYD2a, or removal of the NH2-terminal amino acids that differ between FXYD2a and FXYD2b, do not abrogate the cytoplasmic antagonism between Na+ and K+ but do abolish the decrease in the ATP affinity of Na,K-ATPase observed with wild-type FXYD2a and FXYD2b (34). Moreover, the tendency in FXYD2 peptides that compose the transmembrane domain to oligomerize is abolished after mutation of one of the conserved glycine residues that is part of a LAFVVGLLILLSK motif (35) that resembles the LAXXVGXXIGXXI motif (where X represents any amino acid) involved in membrane helix packing (36). All together, these data indicate that different regions are responsible for the possible self-association, the efficient interaction of FXYD2 with Na,K-ATPase, and the various functional effects of FXYD2. The interaction site of FXYD2 with the 10 transmembrane segments of the α subunit of the Na,K-ATPase has not yet been determined. Thermal denaturation experiments on purified Na,K-ATPase preparations showed that the FXYD2 is lost from the membrane together with COOH-terminal membrane spans of the α subunit, suggesting that FXYD2 may be associated with transmembrane domains M8 through M10 of the α subunit (37). Density maps obtained by electron crystallography at 9.5 Å indeed suggest that FXYD2 may be positioned close to a cavity formed by M2, M6, and M9 of the α subunit (38).

The physiological relevance of the modulatory effects of FXYD2 on Na,K-ATPase function in the control of Na+-reabsorption in the kidney remains to be established. However, among the various functional effects of FXYD2 reported, the effect of FXYD2 on the Na+ affinity of Na,K-ATPase is the most prominent and generally observed, and thus may be physiologically significant. Nephron segment-specific differences in the Na+ affinity of Na,K-ATPase have been reported (39) that correlate with the presence or absence of FXYD2. In the cortical collecting duct, which lacks FXYD2, the constant for half-maximal activation of Na,K-ATPase by Na+ (K1/2Na+) is lower than that in the TAL, where FXYD2 is abundant. The K1/2Na+ value of 10 mM reported for the Na,K-ATPase in the TAL correlates well with that determined for Na,K-ATPase expressed in Xenopus oocytes in the presence of FXYD2. Another argument for the physiological relevance of the effect of FXYD2 on the affinity of Na,K-ATPase for Na+ is provided by the observation that cells transfected with FXYD2 variants exhibit reduced proliferation rates, but only if the FXYD2 variant reduces the Na+ affinity of Na,K-ATPase (33).

In view of the crucial importance of Na,K-ATPase in the reabsorptive capacity of the kidney, small changes in its Na+ affinity may be necessary and sufficient for fine control of the transepithelial ion transport. In the TAL, which reabsorbs up to 25 to 40% of the Na+ filtered in the glomerulus, association of basolateral Na,K-ATPase with FXYD2, which reduces its apparent Na+ affinity, may permit efficient extrusion of cellular Na+ at high intracellular Na+ concentrations (Fig. 3).

Fig. 3.

Functional role of FXYD2 and FXYD4 in the kidney. In a renal nephron, the thick ascending limb (TAL, in green), which exhibits prominent expression of FXYD2 (γ subunit), and the collecting duct (CD, in blue) which exhibits prominent expression of FXYD4 (CHIF), are highlighted. The depiction of an epithelial cell in the TAL (left side) and in the CD (right side) shows the respective intra- and extracellular Na+ concentrations, the contribution of each nephron segment to Na+ reabsorption (as percentage of total Na+ filtered in the glomerulus), and the relative number of Na,K-ATPase (red ovals) associated with FXYD2 (green rectangles) or FXYD4 (blue rectangles) in the basolateral membrane. Na,K-ATPase associated with FXYD2 or FXYD4 acquires a low or a high Na+ affinity, respectively. Grey ovals represent Na+ transporters in the apical membrane facing the lumen of the nephron. Under Na+-depleted conditions, aldosterone acts on the CD, increases apical Na+ entry, Na,K-ATPase, and FXYD4 expression, and thus promotes increased Na+ reabsorption (indicated as -fold increase) and K+ secretion.

Mutation of a conserved glycine residue in the transmembrane domain of FXYD2 (Fig. 2) is associated with cases of human primary hypomagnesemia (23, 40). Studies investigating the effect of the G41R mutant of FXYD2 revealed that the mutation abolishes its interaction with Na,K-ATPase, which results in failure of FXYD2 to traffic to the cell surface and to modulate Na,K-ATPase (34). Because expression of the Na,K-ATPase at the cell surface is not affected in cells expressing the G41R mutant, the hypomagnesemia appears to be an indirect consequence of the loss of Na,K-ATPase modulation by FXYD2. Nothing is known, however, about the cellular mechanisms that link abrogation of FXYD2 modulation of Na,K-ATPase to the loss of Mg2+, increased Ca2+ absorption, and hypocalciuria observed in these patients.

Though the functional role of FXYD2 in Na,K-ATPase modulation is well established, some experimental evidence indicates that, under certain conditions, FXYD2 has a different function. When expressed in Xenopus oocytes, FXYD2 can induce an ion conductance that was attributed to the activation of a Ca2+- and voltage-gated, nonspecific large diameter pore present in the oocyte (41). Whether this observation made in Xenopus oocytes is an artifact due to overexpression of FXYD2, or whether, under physiological conditions, FXYD2 can indeed associate with proteins other than the Na,K-ATPase, remains to be determined. During embryonic development, FXYD2 is not only present in the basolateral membrane of blastocyst cells, which expresses Na,K-ATPase α1-β1 isozymes, but also in the apical membrane, which is devoid of this Na,K-ATPase isozyme (42). Because antisense disruption of FXYD2 expression delays blastocoel formation, the authors concluded that FXYD2 has a role in transepithelial Na+ reabsorption, and hence in blastocoel formation, that is independent of its association with the Na,K-ATPase (42). However, a recent study (43) reports that apical FXYD2 localizes with an apically expressed α3 isoform (43), suggesting that FXYD2 may function in the modulation of Na,K-ATPase activity in both the apical and basolateral membranes of the trophectoderm.

Structural and Functional Aspects of FXYD4 (CHIF)

FXYD4 was initially identified as a protein that induces conductance of K+, but not of other ions, when expressed in Xenopus oocytes (13). FXYD4 expression is confined to basolateral membranes of cells of the distal colon and the medullary portion of the renal collecting duct (44) and shows no overlap with FXYD2 expression in various nephron segments (30).

Like FXYD2, when FXYD4 is expressed in Xenopus oocytes, it associates specifically and stably with the Na,K-ATPase but not with the closely related nongastric H,K-ATPase. FXYD4 also forms complexes with the Na,K-ATPase in native renal tissue (17). Functional analysis of FXYD4 coexpressed with Na,K-ATPase in Xenopus oocytes revealed that it reduces the K+ activation of Na,K-ATPase transport over a wide range of membrane potentials through an increased competition by Na+ at external K+ binding sites (17). Moreover, FXYD4 increases the apparent Na+ affinity of Na,K-ATPase, an effect which is opposite to that of FXYD2 (17, 30) (Fig. 2).

In contrast to FXYD2, which is mainly expressed in tubular segments with high Na+-reabsorptive capacities, FXYD4 is present in the medullary collecting duct and the colon, which are ultimate sites for electrolyte and fluid conservation. Thus, FXYD4 and FXYD2 may allow adaptation to physiological demands of particular nephron segments. In the collecting duct, association of FXYD4 with Na,K-ATPase, which increases its Na+-affinity, is favorable because it permits efficient Na+ reabsorption even at low intracellular Na+ concentrations (Fig. 3). From the change in the apparent Na+ affinity of Na,K-ATPase associated with FXYD4, it can be estimated that, at physiologically low intracellular Na+ concentrations, the transport rate of FXYD4-associated Na,K-ATPase is about four times higher than that of Na,K-ATPase lacking FXYD4 (17). Because K+ secretion is tightly coupled to Na+ reabsorption and Na,K-ATPase activity (45), association of FXYD4 with Na,K-ATPase may also be favorable for K+ secretion. This is reflected in studies showing that a decrease in renal FXYD4 mRNA, observed during acute renal failure, correlates with hyperkalemia. The observed increase in colonic FXYD4 mRNA may be part of a compensatory mechanism that maintains K+ balance through increased colonic K+ secretion (46, 47).

The collecting duct is a target site for aldosterone, which can increase Na+ reabsorption by two- to four-fold when Na+ conservation is required (5) (Fig. 3). During the early phase of aldosterone action, which mediates an increased apical Na+ entry and produces small changes in intracellular Na+ concentrations, Na+ reabsorption should improve in the presence of Na,K-ATPase associated with FXYD4. Moreover, in conditions of Na+ depletion, expression of both Na,K-ATPase and FXYD4 expression is increased (44), which assures that FXYD4 does not become limiting for the formation of Na,K-ATPase with high Na+ affinity.

Results from FXYD4 knockout mice at least partially confirm the putative role of FXYD4 as a Na,K-ATPase modulator in the medullary collecting duct (48). The major phenotype of these knockout mice, a larger volume of urine excretion, is quite mild and is observed only after K+ loading or Na+ deprivation, suggesting that efficient compensatory mechanisms exist in these mice that prevent severe perturbations in electrolyte homeostasis.

A Putative Role for FXYD1 (Phospholemman) in Na,K-ATPase Regulation in Heart and Skeletal Muscle

FXYD1 [phospholemman (PLM)] is expressed mainly in heart, skeletal muscle, and liver (9, 49). Multiple functions have been attributed to FXYD1 (Fig. 4). Expressed in Xenopus oocytes, FXYD1 forms or regulates Cl- channels (20). Moreover, in bilayers, FXYD1 forms ion channels that are selective for the zwitterionic taurine, and can switch between conformations with different selectivities for cations and anions (50, 51). FXYD1 also increases taurine efflux and produces a regulatory volume decrease in response to cell swelling when transfected in human embryonic kidney (HEK) cells (52). Accordingly, expression of antisense oligonucleotides to FXYD1 inhibits taurine efflux (53). On the basis of these activities, it was suggested that FXYD1 plays a specific role in muscle contraction and cell volume regulation.

Fig. 4.

Putative physiological roles of FXYD1 in heart and skeletal muscle. One putative role of FXYD1 (PLM) is to form taurine channels in response to cell swelling; thus, FXYD1 may contribute to cell volume regulation. Activation of PKA and phosphorylation (P) of FXYD1 leads to an increase in the expression of FXYD1 at the cell surface and enhanced taurine efflux. FXYD1 can also interact with the Na,K-ATPase and may act as a heart- and muscle-specific regulator of the Na,K-ATPase, which modulates muscle contractility. For more details, see text.

FXYD1 associates specifically with six different Na,K-ATPase isozymes when expressed in Xenopus oocytes, and with α1-β and to a lesser extent with α2-β isozymes in native heart and muscle tissue (54). When expressed in oocytes, FXYD1 produces a slight decrease in the apparent K+ affinity and a larger decrease in the apparent Na+ affinity of Na,K-ATPase (Fig. 2). Interestingly, the K1/2Na+ value of about 20 mM, determined for Na,K-ATPase associated with FXYD1, is close to that predicted for cardiac Na,K-ATPase under physiological conditions (55). This provides an argument for the association of FXYD1 with cardiac Na,K-ATPase and its physiological relevance, although it remains unknown whether FXYD1 has a role in the modulation of the contractile force of cardiac and skeletal muscle. Association of FXYD1 with the Na,K-ATPase, which decreases its apparent Na+ affinity, could result in an increased intracellular Na+ concentration, an increase in intracellular Ca2+ concentrations due to inhibition of the Na+/Ca2+-exchanger, and ultimately an increased contractility of heart and skeletal muscles (Fig. 4). This hypothesis is at least partly supported by the observation that, at low extracellular Ca2+ concentrations, contractility of cardiomyocytes overexpressing FXYD1 increases compared to that of nontransfected cells (56).

Phosphorylation of FXYD1 may be a factor that regulates its putative multiple functions. FXYD1 is considered a major target for protein kinases in the heart (9), and PKA, PKC, and the cell cycle regulated protein kinase NIMA (never in mitosis) phosphorylate specific cytoplasmic serine residues of FXYD1 (57-59). Phosphorylation of FXYD1 by PKA, but not by PKC, increases the amplitude of phospholemman-mediated currents and the amount of FXYD1 at the cell surface, whereas similar effects mediated by NIMA kinases are independent of FXYD1 phosphorylation (59). In contrast, phosphorylation by myotonic dystrophy kinase induces a rapid degradation of FXYD1 (60). Thus, phosphorylation of FXYD1 by different protein kinases seems to influence, on one hand, its targeting and/or its oligomerization into a channel and, on the other hand, its turnover. It remains to be shown whether FXDY1's dissociation from or association with the Na,K-ATPase is regulated by phosphorylation, as suggested for a shark phospholemman-like protein (21).

FXYD7, a Brain-Specific Regulator of the Na,K-ATPase

FXYD7 is exclusively expressed in brain, in both neurons and glial cells (18). FXYD7 is subjected to O-glycosylation, which appears to be important for the protein's cellular stability (18). Expressed in Xenopus oocytes, FXYD7 interacts with α1-β1, α2-β1, and α3-β1 isozymes, but not with α-β2 isozymes. In contrast, in brain, FXYD7 antibodies co-immunoprecipitate only α1-β isozymes, and FXYD7 is probably associated with the α1-β1 isozymes (18). When expressed in Xenopus oocytes, FXYD7 modulates Na,K-ATPase transport properties in a distinctive way (Fig. 2). FXYD7 has no effect on the apparent Na+ affinity of Na,K-ATPase but decreases the apparent K+ affinity over a wide range of membrane potentials, although it affects only α1-β1 and α2-β1 isozymes, not α3-β1 isozymes (18).

During neuronal activity, the extracellular K+ concentration increases; to prevent K+-mediated perturbations of neuronal excitability, K+ must be rapidly cleared from the extracellular space. Both the K+ recovery during neuronal firing, and the characteristic extracellular K+ undershoot observed after neuronal firing, seem to be mediated by the glial Na,K-ATPase α2-β1 isozymes, which have an intrinsically low affinity for K+, as well as by neuronal α1-β isozymes (61). Neuronal α1-β1 isozymes acquire a low K+ affinity by the associating with FXYD7, which favors efficient removal of K+ from the extracellular space during neuronal firing and may prevent an excessive K+ undershoot after neuronal firing.

Other Members of the FXYD Family

FXYD3 (12), FXYD5 (14), and FXYD6 (15) are poorly characterized, and their ability to interact with and modulate the Na,K-ATPase have not yet been reported.

FXYD3 is expressed mainly in stomach, colon, and uterus. Its expression is increased in certain breast tumors initiated by Ras and Neu oncogenes, but not in those initiated by Myc oncogenes (12). When expressed in Xenopus oocytes, FXYD3 induces a Cl- conductance similar to that associated with expression of FXYD1 (12).

FXYD5 is structurally the most divergent of the FXYD proteins because it includes an unusually long NH2-terminal tail (8) (Fig. 2). A putative isoform called IWU-1 has been described that has a shorter NH2-terminus exhibiting 61% identity with a cytoplasmic COOH-terminal region of the angiotensin II type 1 receptor (62). The RIC gene is activated in cells transformed by the oncogene E2a-Pbxl (14). RIC is also expressed abundantly in several cancer tissues but is expressed in only a few normal cell types, such as lymphocytes, endothelial cells, and basal cells of stratified squamous epithelium (63). Transfection of RIC into liver cancer cells results in decreased abundance of E-cadherin and suggest that RIC, called dysadherin, may have a role in tumor progression and metastasis (63).

Perspectives

Characterization of four out of seven mammalian FXYD proteins indicates that they act as tissue-specific regulatory subunits of Na,K-ATPase. Each of these auxiliary subunits produces a distinct functional effect on the transport characteristics of the Na,K-ATPase that is adjusted to the specific functional demands of the tissue in which the FXYD protein is expressed. FXYD proteins appear to preferentially associate with Na,K-ATPase α1-β isozymes, and affect their function in a way that render them operationally complementary or supplementary to coexisting isozymes. These results emphasize the complex interplay of various regulatory mechanisms that is necessary to ensure appropriate Na+ and K+ handling by the Na,K-ATPase, and allows for the correct function of such mechanisms as renal Na+ reabsorption, muscle contraction, and neuronal excitability under changing physiological conditions. Although much has been learned about FXYD proteins, many questions remain concerning the structure and function relationship of FXYD proteins, and the regulation of their activity and expression. Further studies are also needed to explore the apparent functional versatility of FXYD proteins and their physiological relevance. Moreover, in the light of the recent finding of a link between a mutation in the γ gene and a genetic renal disorder, it is tempting to consider that other genetic diseases may be linked to other alterations in FXYD proteins. Finally, from a pathophysiologic point of view, the functional characterization of MAT-8 and of RIC deserves particular attention, because they may contribute to progression of the cancerous state.

References

  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 10.
  11. 11.
  12. 12.
  13. 13.
  14. 14.
  15. 15.
  16. 16.
  17. 17.
  18. 18.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
  42. 42.
  43. 43.
  44. 44.
  45. 45.
  46. 46.
  47. 47.
  48. 48.
  49. 49.
  50. 50.
  51. 51.
  52. 52.
  53. 53.
  54. 54.
  55. 55.
  56. 56.
  57. 57.
  58. 58.
  59. 59.
  60. 60.
  61. 61.
  62. 62.
  63. 63.
  64. 64.
  65. 65.
  66. 66.
View Abstract

Navigate This Article