Functional polymorphism in the carboxyl terminus of the alpha-subunit of the human epithelial sodium channel.

A common human epithelial sodium channel (ENaC) polymorphism, alphaT663A, is present in the cytoplasmic C terminus of the alpha-subunit, although it is unclear whether this polymorphism segregates with blood pressure. We examined whether this polymorphism was associated with differences in functional Na(+) channel expression. Whole cell amiloride-sensitive currents in Xenopus oocytes expressing wild type channels (alphaT663betagamma) were significantly approximately 1.3-2.0-fold higher than currents measured in oocytes expressing channels with an Ala, Gly or Leu, or Lys at position alpha663. In contrast, differences in functional human ENaC expression were not observed with oocytes expressing channels having Thr (wild type), Ser, or Asp at this position. The surface expression of channels, measured using an epitope-tagged beta-subunit, was significantly reduced in oocytes expressing alphaT663Abetagamma when compared with oocytes expressing alphaT663betagamma. The corresponding polymorphism was generated in the mouse alpha-subunit (malphaA692T) and was not associated with differences in functional alphabetagamma-mouse ENaC expression. The polymorphism is present in a region that is not well conserved between human and mouse. We generated a mouse/human chimera by replacement of the distal C terminus of the mouse alpha-subunit with the distal C terminus of the human alpha-subunit. Co-expression of this m(1-678)/h(650-669)T663A chimera with mouse betagamma led to a significant reduction in whole cell Na(+) currents and surface expression when compared with m(1-678)/h(650-669)T663-mbetagamma. Our results suggest that halphaT663A is a functional polymorphism that affects human ENaC surface expression.

Epithelial sodium channels (ENaC) 1 are expressed in principal cells in the late distal convoluted tubule and collecting tubule, where they serve as a final site for reabsorption of Na ϩ from the glomerular ultrafiltrate. Volume regulatory hormones, such as aldosterone, have a key role in modifying rates of renal tubular Na ϩ reabsorption through regulation of functional ENaC expression at the apical plasma membrane (1). Epithelial Na ϩ channels are composed of three structurally related subunits, termed ␣-, ␤-, and ␥-ENaC that likely assemble as a ␣2,␤1,␥1 tetramer (2,3), although an alternative subunit stoichiometry has been proposed (4). The three subunits share limited (ϳ30 -40%) sequence identity but share a common topology of two membrane-spanning domains and intracellular N and C termini (5)(6)(7).
Changes in ENaC functional expression are associated with alterations in blood pressure (8,9). Na ϩ channel gain-of-function mutations have been identified in patients with Liddle's syndrome, a disorder characterized by volume expansion, hypokalemia, and hypertension (10,11). ENaC loss-of-function mutations have been identified in patients with type I pseudohypoaldosteronism, a disorder characterized by volume depletion, hypotension, and hyperkalemia (12,13). Some common human ENaC polymorphisms may segregate with blood pressure (i.e. ␤T594M) (14), suggesting that ENaC polymorphisms that alter functional channel expression may contribute to the development of hypertension in the general population.
A common polymorphism has been described in C terminus of the ␣-subunit of human ENaC (hENaC), ␣T663A (15,16). The question of whether this polymorphism segregates with blood pressure has been addressed by several groups, and conflicting results have been reported (15,16). We examined whether this polymorphism was associated with altered Na ϩ channel activity when expressed in Xenopus oocytes. We observed that oocytes expressing wild type ␣T663␤␥ channels had significantly higher currents than oocytes expressing ␣T663A␤␥. Furthermore, the higher currents observed with ␣T663␤␥ channels were associated with higher levels of cell surface expression of channels, suggesting that this polymorphism altered channel trafficking. This polymorphism is present in the distal C terminus of the ␣-subunit, a region that is not well conserved between the human and mouse ␣-subunit. Although the ␣A692T polymorphism in mouse ENaC (mENaC), corresponding to human ␣T663A, was not associated with differences in functional ␣␤␥-mENaC expression, replacement of the distal C terminus of the mouse ␣-subunit with the distal C terminus of the human ␣-subunit restored the functional differences that were observed with the human ␣T663A polymorphism.

EXPERIMENTAL PROCEDURES
Preparation of hENaC Mutants-␣-, ␤-, and ␥hENaC cDNAs were from M. J. Welsh (University of Iowa). Mouse ENaC cDNAs have been described and used by our group previously (17). All mutants and mouse/human chimeras were generated using a PCR-based mutagenesis technique as described previously (18). All sequences were confirmed by automated DNA sequence analyses performed at sequencing facilities at the University of Pennsylvania, the Children's Hospital of Philadelphia, or the University of Pittsburgh.
Expression of ENaC in Xenopus Oocytes-cRNAs for wild type and mutant ␣hENaC and ␣mENaC, wild type ␤hENaC and ␤mENaC, and wild type ␥hENaC and ␥mENaC were synthesized from linearized plasmids containing the appropriate cDNAs using appropriate RNA polymerases (T 3 , T 7 , or SP6, mMessage mMachine, Ambion Inc., Austin, TX) and stored at Ϫ80°C. cRNA concentration was determined spectroscopically. Stage V-VI oocytes were surgically harvested from female Xenopus laevis (Nasco, Fort Atkinson, WI) and pretreated with 2 mg/ml collagenase (type IV) as described previously (18). Oocytes were injected with 2 ng/subunit of hENaC cRNAs or 0.33 ng/subunit of mENaC cRNAs in 50 nl of H 2 O. After injection, oocytes were incubated at 18°C in modified Barth's saline (MBS) (88 mM NaCl, 1 mM KCl, 2.4 mM NaHCO 3 , 15 mM HEPES, 0.3 mM Ca(NO 3 ) 2 , 0.41 mM CaCl 2 , 0.82 mM MgSO 4 , pH 7.2) supplemented with 10 g/ml sodium penicillin, 10 g/ml streptomycin sulfate, and 100 g/ml gentamycin sulfate. Animal protocols were approved by the University of Pennsylvania, the University of Pittsburgh, and the Children's Hospital of Philadelphia Institutional Animal Care and Use Committees.
Two-electrode Voltage Clamp-A two-electrode voltage clamp (TEV) was performed 24 -48 h after cRNA injection at room temperature using a DigiData 1320 interface and Axon Geneclamp 500B Amplifier (Axon Instruments, Foster City, CA). Data were acquired at 200 Hz, and analyses were performed using pClamp 8.0 or 8.1 software (Axon Instruments) on 833-MHz Pentium III personal computers (Dell Computer, Austin, TX). Pipettes were pulled from borosilicate glass capillaries (World Precision Instruments, Inc., Sarasota, FL) with a micropipette fuller (Sutter Instrument Co., Novato, CA) and had resistances of 0.5-5 megohms when filled with 3 M KCl and inserted into the bath solution. Oocytes were maintained in a recording chamber with 1 ml of bath solution and continuously perfused with bath solution at a flow rate of 4 -5 ml/min. The bath solution contained 100 mM sodium gluconate, 2 mM KCl, 1.8 mM CaCl 2 , 3 mM BaCl 2 , 10 mM tetraethylammonium chloride, and 10 mM HEPES, pH 7.4. A series of voltage steps (1 s) from Ϫ140 to ϩ60 mV (adjusted for resting membrane potential) in 20-mV increments were performed, and whole cell currents were recorded 750 ms after the initiation of the Ϫ100-mV voltage step for data analysis. ENaC-mediated current was defined as the difference in whole oocyte current at a Ϫ100-mV holding potential (adjusted for resting membrane potential) before and after the addition of 10 M amiloride-HCl (Sigma) to the bath solution.
A second approach to examine surface expression used cell surface biotinylation to identify the surface pool of channels as described previously (22). ␤hENaC with a C-terminal V5 epitope tag (␤-V5) was generated in pcDNA6 as described previously (23). cRNAs for ␣␤-V5␥ hENaC were co-injected into Xenopus oocytes. As a control, oocytes were co-injected only with cRNAs for ␤-V5 and ␥. After 48 h, oocytes were stripped mechanically of their vitelline membranes in hypertonic FIG. 2. The ␣T663A polymorphism reduces hENaC functional expression in oocytes. ␣T663␤␥ or ␣T663A␤␥ hENaC was expressed in oocytes, and TEV was performed. Amiloride-sensitive whole cell currents were determined at a Ϫ100-mV holding potential (adjusted for resting membrane potential) and are expressed relative to the mean whole oocyte current for ␣T663␤␥hENaC-expressing oocytes. Data are presented as mean Ϯ S.E. with the p value determined by a two-tailed t test as described under "Experimental Procedures." medium (300 mM sucrose in MBS without penicillin, streptomycin, and gentamycin), and surface proteins were labeled with N-hydroxysulfosuccinimidobiotin (Pierce) as described previously (22). Oocytes (10/ group) were subsequently lysed in 0.15 M NaCl, 0.01 M Tris-Cl, pH 8.0, 0.01 M EDTA, 1.0% Nonidet P-40, 0.5% sodium deoxycholate, 1.0 mM phenylmethanesulfonyl fluoride, 0.1 mM N-␣-p-tosyl-L-lysine chloromethyl ketone, 0.1 mM L-1-tosylamide-2-phenylethyl-chloromethyl ketone, and 2 g/ml aprotinin for 1 h at 4°C and centrifuged at 13,000 ϫ g for 15 min at 4°C. Biotinylated proteins were precipitated with streptavidin-agarose (Pierce) and subjected to SDS-PAGE.
Statistical Analyses-Whole cell amiloride-sensitive current data were expressed relative to that of wild type ENaC. To decrease the influence of batch-to-batch variability in ENaC expression, data were normalized by the mean amiloride-sensitive current for the control condition (usually ␣T663␤␥ hENaC) within a batch of oocytes prior to combining data of multiple independent batches for statistical analysis. These data are presented as mean Ϯ S.E. Because a Poisson distribu- , and ␣T663L␤␥ (E) were expressed in oocytes, and TEV was performed. The functional expression of these mutant hENaCs was compared with ␣T663␤␥ by determination of amiloride-sensitive whole cell currents at a Ϫ100-mV holding potential (adjusted for resting membrane potential). These data (mean Ϯ S.E.) are expressed relative to the mean whole cell amiloride-sensitive current measured in oocytes expressing ␣T663␤␥hENaC. p values were determined by a two-tailed t test as described under "Experimental Procedures." tion, rather than a Gaussian distribution, best described these combined data, p values were determined by a two-tailed t test after a square root transformation to better approximate a Gaussian distribution (24). A p value of Յ0.05 was considered significant. We also independently analyzed the statistical significance of these data without normalization and transformation using the Wilcoxon rank sum/Mann-Whitney U test for non-parametric results and obtained similar p values to the method outlined above (data not shown). Other data that were normally distributed are expressed as mean Ϯ S.E. with p values determined by a two-tailed t test. All data analyses were performed with SigmaStat version 2.03.

␣T663A Polymorphism Is Associated with Altered Functional
Channel Expression in Xenopus Oocytes-Functional hENaC expression was assessed by TEV in oocytes expressing either ␣T663␤␥ hENaC (n ϭ 56) or the ␣T663A polymorphism (␣T663A␤␥) (n ϭ 57). Fig. 1 illustrates current/voltage relationships for oocytes expressing ␣T663␤␥ and ␣T663A␤␥ hENaC before and after the addition of 10 M amiloride. The linear current/voltage relationships are characteristic of ␣␤␥ ENaC expressed in oocytes, and the reversal potential of ϳ0 mV is consistent with intracellular Na ϩ loading observed in oocytes expressing ENaC and incubated in MBS (17,18). Oocytes expressing ␣T663A␤␥ channels exhibited mean Ϯ S.E. whole cell amiloride-inhibited currents of Ϫ2.15 Ϯ 0.29 A at a Ϫ100-mV holding potential that were significantly lower than mean Na ϩ currents measured in oocytes expressing wild type channels (Ϫ4.65 Ϯ 0.67 A, p Ͻ 0.001 (Fig. 2 for normalized data)). Given the small difference in whole cell Na ϩ currents we observed with oocytes expressing ␣T663␤␥ and ␣T663A␤␥, experiments were performed with two different cRNA preparations and four different batches of oocytes to confirm our findings.
Analyses of ␣T663 Mutants-A conservative substitution of Ser for the Thr residue at position ␣663 did not alter levels of amiloride-sensitive Na ϩ current (Fig. 3A, p ϭ n. s.). Thr and Ser residues are potential targets of Ser/Thr kinases, and phosphorylation of this residue would alter the charge at this site. We examined whether an acidic residue (Asp) at position ␣663 affected functional expression of amiloride-sensitive Na ϩ current. Levels of amiloride-sensitive Na ϩ current expression in oocytes injected with either wild type (␣T663␤␥) hENaC or ␣T663D␤␥ hENaC were similar (Fig. 3B, p ϭ n. s.). In contrast, amiloride-sensitive currents in oocytes expressing Na ϩ channels with a basic residue at position ␣663 (␣T663K␤␥) were significantly lower than currents observed with wild type hENaC (Fig. 3C, p ϭ 0.025). Non-polar amino acids were also introduced at position ␣663, and mutant or wild type channels were expressed in oocytes. Mean whole cell amiloride-sensitive currents in oocytes expressing ␣T663G␤␥ (Fig. 3D, p ϭ 0.011) or ␣T663L␤␥ (Fig. 3E, p Ͻ 0.001) were significantly lower than the currents measured in oocytes expressing wild type ENaC. These data are summarized in Table I.
Functional Changes Associated with the T663A Polymorphism Are Species-specific-The alignments of the C termini of mouse and human ␣-subunits are illustrated in Fig. 4A start-ing with the Tyr in the PY motif that is present within all three ENaC subunits and is conserved across species. T663 is present within a region of the ␣-subunit C terminus that is not well conserved across species. An Ala residue is present at the equivalent site in the mouse ␣-subunit (m␣A692). We mutated mouse ␣A692 to Thr. Oocytes expressing mouse ␣A692␤␥ (i.e. wild type, n ϭ 38) or mouse ␣A692T␤␥ (n ϭ 36) showed similar levels of amiloride-sensitive currents (Fig. 4B, p ϭ n. s.). In contrast, oocytes expressing human ␣T663 with mouse ␤␥ (n ϭ 43) had significantly higher amiloride-sensitive currents than oocytes expressing human ␣T663A with mouse ␤␥ (Fig. 4C, n ϭ 44, p ϭ 0.017).
␣ENaC Polymorphism Affects Surface Expression of ENaC-The reduced functional expression of ␣T663A␤␥ channels in oocytes likely reflects changes in the numbers of channels at the plasma membrane or changes in channel open probability. Because these residues are not within the pore of the channel (1,18), it is unlikely that this polymorphism affects singlechannel conductance. Previous studies (25) have also suggested that it is difficult to assess differences in ENaC open probability because the channel has inherently large variability in open probability.
We therefore examined whether differences in functional expression of ␣T663␤␥ and ␣T663A␤␥ hENaC were associated with altered levels of cell surface expression of channels. In these experiments, the ␤-subunit had a C-terminal V5 epitope tag to identify the protein on immunoblots. Cell surface proteins of oocytes expressing ␣T663␤-V5␥ or ␣T663A␤-V5␥ were labeled by biotinylation and recovered with streptavidin beads. The biotinylated ␤-subunits were subsequently detected by immunoblots. As shown in Fig. 5A, expression of ␣T663A␤-V5␥ hENaC resulted in a relative recovery of biotinylated ␤-V5 of 0.64 Ϯ 0.08 compared with that observed when ␣T663␤-V5␥ hENaC was expressed (n ϭ 5, p ϭ 0.003). Whole oocyte ␤-V5 expression was similar in the two groups ( Fig 5B). These data suggest that the increases in functional expression observed with ␣T663␤␥, when compared with ␣T663A␤␥, are because of differences in surface expression.
Previous studies (25,26) suggested that the ␣-subunit has a key role in facilitating the trafficking of ␤and ␥-subunits to the plasma membrane in oocytes. In the absence of ␣-subunit expression, levels of surface expression of ␤and ␥-subunits in oocytes are reduced markedly (25,26). We observed that the recovery of biotinylated ␤-V5 in oocytes expressing only ␤␥-hENaC was 0.18 Ϯ 0.06 (n ϭ 3) of that observed with oocytes expressing ␣T663␤␥ hENaC (data not shown). These data suggest that oocytes that are stripped of their vitelline membrane have only a limited labeling of intracellular ␤-subunits with N-hydroxysulfosuccinimidobiotin.
We also examined whether differences in the functional expression of mouse/human ␣-chimeras (m(1-678)/h(650 -669)T663-m␤␥ and m(1-678)/h(650 -669)T663A-m␤␥) were associated with altered levels of cell surface expression of channels. Cell surface expression was determined using a ␤-subunit with an extracellular FLAG epitope tag and a chemi- FIG. 4. Residues in the vicinity of h␣T663A have a role in conferring differences in function ENaC expression. A, alignments of the distal C termini of human and mouse ␣-subunits. Spaces were added to separate a conserved six-residue N-terminal tract (YATLGP) from the remainder of the C terminus and to highlight h␣T663 and m␣A692 (in bold). The first and last amino acid residue numbers for each sequence are listed. Sequence alignments were performed with MacVector 7.2 (Accelrys) for the mouse and human ␣-subunits (GenBank TM Accession Numbers AF112185 and L29007). B, TEV was performed with oocytes expressing either mouse ␣A692␤␥ or ␣A692T␤␥. Amiloride-sensitive whole cell currents were determined at the Ϫ100-mV holding potential (adjusted for resting membrane potential) and are expressed relative to the mean whole cell amiloride-sensitive current for ␣A692␤␥. C, TEV was performed with oocytes expressing either human ␣T663/mouse ␤␥ or human ␣T663A/mouse ␤␥. Amiloride-sensitive whole cell currents were determined at the Ϫ100-mV holding potential and expressed relative to the mean whole cell amiloride-sensitive current for human ␣T663/mouse ␤␥. D, TEV was performed with oocytes expressing either m(1-678)/h(650 -669)T663/mouse ␤␥ or m(1-678)/h(650 -669)T663A/mouse ␤␥. Amiloride-sensitive whole cell currents were determined at the Ϫ100-mV holding potential and expressed relative to the mean whole cell amiloride-sensitive current for m(1-678)/h(650 -669)T663/mouse ␤␥. p values were determined with a two-tailed t test as described under "Experimental Procedures." luminescence-based assay that does not require stripping of the oocyte vitelline membrane. The mean relative cell surface expression of m(1-678)/h(650 -669)T663A-m␤␥ was 0.59 Ϯ 0.10 (n ϭ 46) compared with that of m(1-678)/h(650 -669)T663-m␤␥ (Fig. 6, 1.00 Ϯ 0.18, n ϭ 47, p ϭ 0.016). Nonspecific labeling was examined in oocytes expressing m(1-678)/ h(650 -669)T663-m␤␥ where the ␤-subunit did not contain an epitope tag. Mean nonspecific labeling was 0.18 Ϯ 0.03 (n ϭ 35) of the labeling observed in oocytes expressing epitopetagged m(1-678)/h(650 -669)T663-m␤␥. We also performed separate control experiments to assess the labeling of intracellular ␤-subunits in this chemiluminescence-based assay. The labeling of oocytes expressing only human ␤-FLAG, which, as discussed above, has markedly reduced surface expression (25,26), was indistinguishable from that of uninjected oocytes and was significantly less than that of oocytes expressing m(1-678)/h(650 -669)T663-m␤-FLAG␥ (25, 26) (data not shown). DISCUSSION A common T663A polymorphism is present within the C terminus of the ␣-subunit of human ENaC (15,16). We observed a significant increase in functional hENaC expression in oocytes expressing ␣T663␤␥, when compared with oocytes expressing ␣T663A␤␥ (Fig. 1). One study (15) suggested previously that the human ␣T663A polymorphism was not associated with the changes in functional ENaC expression in Xenopus oocytes; however the number of oocytes analyzed was not reported. Given the inherent variability of the levels of ENaC expression in oocytes, it may be difficult to detect small changes in the level of ENaC expression.
Additional studies were performed to assess whether substitution of other amino acid residues at this site influenced functional ENaC expression. Whole cell amiloride-sensitive currents in oocytes expressing channels with a Thr, Ser, or Asp at position ␣663 were ϳ1.3-2.0-fold higher than currents measured in oocytes expressing channels with an Ala, Gly or Leu, or Lys at this position (Table I and Figs. 2 and 3). Our observations that channels with Thr, Ser, or Asp at position ␣663 showed similar levels of functional expression in oocytes suggest that ␣663 may be a site of phosphorylation by Ser/Thr kinases. At present, we have no direct evidence that this site is phosphorylated in vivo nor whether phosphorylation of this site affects ENaC activity.
␣T663 is present within a region of the ␣-subunit that is not well conserved between the human and mouse (Fig. 4A). An Ala residue is present at the homologous site in the mouse ␣-subunit (m␣A692), and we observed that oocytes expressing mouse ␣A692␤␥ and ␣A692T␤␥ had similar levels of amiloride-sensitive currents (Fig. 4B). However, when the distal C terminus of the mouse ␣-subunit was replaced with the distal C terminus of the human ␣-subunit, creating a mouse/human chimera, oocytes co-expressing m(1-678)/h(650 -669)T663-m␤␥ had significantly higher whole cell Na ϩ currents than oocytes co-expressing m(1-678)/h(650 -669)T663A-m␤␥ (Fig. 4D). These results suggest that residues in the vicinity of h␣T663 have a role in conferring the observed differences in function expression between h␣T663␤␥ and h␣T663A␤␥. Furthermore, the differences in ENaC surface expression that we observed with the h␣T663A␤␥ and the m(1-678)/h(650 -669)T663A-m␤␥ polymorphisms (Figs. 5 and 6) were correlated with differences in function expression (Figs. 2 and 4D) and likely reflect differences in the delivery of channels to the plasma membrane and/or differences in the rate of retrieval from the plasma membrane. Our data suggest that the distal C terminus of the human ␣-subunit has a novel site that affects the intracellular trafficking of ENaC that is distinct from the PY motif in the ␤and ␥-subunits (25,27,28).
Increases in ENaC activity in renal collecting ducts may be associated with extracellular fluid volume expansion and hypertension, as has been reported in studies of patients with Liddle's syndrome (8,10,29,30). Whole cell Na ϩ currents measured in oocytes expressing mutations that are found in patients with Liddle's syndrome were significantly higher than currents measured in oocytes expressing wild type ENaC (31,32). However, correlations between increases in ENaC activity measured in oocytes and increases in blood pressure in humans with specific ENaC mutations or polymorphisms are not a consistent finding. For example, the ␤T594M polymorphism has been found to segregate with blood pressure in selected populations, suggesting that this polymorphism is associated with alterations in ENaC activity (14,33). Lymphocytes obtained from patients homozygous for ␤T594 exhibited amiloride-sensitive whole cell Na ϩ currents that were enhanced by cAMP (34). Phorbol 12-myristate 13-acetate blocked the activation of Na ϩ currents by cAMP. In contrast, lymphocytes obtained from ␤T594M homozygotes exhibited cAMP-activated whole cell currents that were not blocked by phorbol 12-myristate 13-acetate, suggesting that this polymorphism was associated with altered ENaC activity (34). However, no differences in ENaC activity have been reported with the ␤T594M polymorphism in oocyte expression studies (35).
Although we have observed differences in hENaC activity with the ␣T663A polymorphism with oocyte expression, it is unclear whether the h␣T663A polymorphism is associated with differences in blood pressure in humans. Ambrosius et al. (15,36) reported that ␣A663 was more prevalent in African-American subjects than in Caucasian subjects, and its presence (␣A663) was associated with a modest increase in blood pressure. However, other investigators have not observed differences in blood pressure with the ␣T663A polymorphism (16). Our functional studies in oocytes predict that ␣A663 would be associated with a reduction in blood pressure. Certainly environmental and other genetic factors have an important role in the regulation of blood pressure in humans, and it may be difficult to demonstrate that specific ENaC polymorphisms contribute to essential hypertension in humans despite modest differences in functional activity when assayed in heterologous expression systems (37).
In summary, our results suggest that h␣T663␤␥ has increased activity in oocytes when compared with h␣T663A␤␥. The increase in channel activity observed with h␣T663␤␥ reflects an increase in surface expression, suggesting that this polymorphism modifies the intracellular trafficking of ENaC. Furthermore, our studies using mouse/human ␣-subunit chimeras suggest that residues in the vicinity of h␣T663 have an important role in conferring differences in functional expression between h␣T663␤␥ and h␣T663A␤␥.