点击显示 收起
【摘要】 Activity of the epithelial Na + channel (ENaC) is limiting for Na + reabsorption at the distal nephron. Gain-of-function mutations in ENaC cause Liddle's syndrome: a severe form of inheritable hypertension. Several polymorphisms in -hENaC possibly associated with abnormal Na + handling by the kidney and the salt-sensitive hypertension prevalent in black populations have been reported. The functional effects of -hENaC polymorphisms on channel activity, however, remain controversial and have not been directly tested in a mammalian background. We ask here whether polymorphisms at positions 334, 618, and 663 in -hENaC influence channel activity. Activity of wild-type (A334, C618, A663) and polymorphic ENaC expressed in Chinese hamster ovary cells was assessed with patch-clamp electrophysiology. While the A334T polymorphism had little effect on macroscopic ENaC currents, the C618F and A663T polymorphisms 3.3- and 1.6-fold, respectively. Similarly, polymorphic ENaC had greater activity compared with wild-type channels in excised patches with activity of C618F and A663T channels increased 3.8- and 2.6-fold, respectively. Unitary channel conductances and reversal potentials were not different for polymorphic and wild-type ENaC. Increases in activity resulted primarily from increases in the apparent number of active (polymorphic) channels in the plasma membrane. Moreover, addition of a reducing agent to the cytosol significantly increased activity of wild-type ENaC equal to that of C618F polymorphic channels but had no effect on these latter channels. These results are consistent with the C618F and A663T polymorphisms leading to elevated ENaC activity with the possibility that they facilitate altered Na + handling by the kidney.
【关键词】 hypertension Liddle‘s syndrome sequence variations
THE EPITHELIAL Na + channel is a heteromeric channel composed of three similar but distinct subunits,,, and ( 4, 10, 12 ). This channel is an end-effector of the rennin-angiotensin-aldosterone system and resides in the apical plasma membrane of renal distal nephron epithelia where its activity is limiting for Na + reabsorption. Due to its regulation and function, the epithelial Na + channel (ENaC) plays a central role in modulating systemic Na + balance and thus chronic blood pressure. Gain-of-function mutations in ENaC, indeed, cause the rare but severe form of inheritable hypertension Liddle's syndrome ( 2, 6, 10, 12 ). Liddle's syndrome is marked by salt sensitivity, low renin activity, and low ANG II and aldosterone levels. In contrast to this rare form of monogenic hypertension, most hypertension results from compounding gene defects influenced by environmental factors. This more prevalent but less well-understood hypertension is referred to as essential hypertension.
The importance of genetic predisposition in essential hypertension has long been appreciated ( 6, 12, 14, 16 ). Similarly, the role of the kidney or rather its dysfunction in hypertension has long been recognized ( 2, 6, 10, 16 ). Improper Na + retention by the kidney elevates blood pressure and can be causative for salt-sensitive hypertension. Western populations of African ancestry appear to be particularly predisposed to intrinsic kidney defects, for these populations compared with those of European ancestry often have low levels of renin activity and aldosterone resembling a mild form of Liddle's syndrome with hypertension in black individuals often being salt sensitive ( 1, 9, 14, 16 ). Accordingly, hypertension is both more common and severe in African Americans (reviewed in Refs. 14, 16 ).
Several sequence variations in the gene encoding the -subunit of ENaC, including polymorphisms leading to substitutions at residues 334, 618, and 663, have been identified ( 1, 9, 14 - 16 ). In some studies but not others, polymorphisms at these sites have been linked with elevated blood pressure ( 1, 9, 14 - 16 ). It is likely that the effects of ENaC polymorphisms on blood pressure are subtle and population dependent. This may explain why it has been difficult to reach a consensus about the contribution of specific ENaC polymorphisms to abnormal blood pressure. Nevertheless, polymorphisms at these sites do have distinct allele frequencies in white and black populations. For instance, the C618F polymorphism is almost completely absent in the former population (0.002) but more prevalent in the latter at a frequency approaching 0.08 ( 1, 9 ). It has been postulated that subtle changes in ENaC activity resulting from polymorphisms may lead to elevations in blood pressure and a predisposition for blacks toward salt-sensitive hypertension ( 8, 14, 16 ). Consistent with this idea, the A663T polymorphism has been reported to increase hENaC activity when the channel is heterologously expressed in Xenopus laevis oocytes ( 11 ). Increases in activity in this system resulted from increases in the number of active channels in the membrane.
Whereas some studies have linked sequence variations in -hENaC and elevated blood pressure, it has been difficult to determine the possible causal role played by these polymorphisms for their effects on channel activity have not been extensively studied. Variants of mammalian ENaC, in particular, have never been directly investigated in a mammalian background. The current study asked whether the A334T, C618F, and A663T polymorphisms in -hENaC increase channel activity in a mammalian background. This is an important question for if these polymorphisms possibly contribute to increases in blood pressure, they must do so by increasing ENaC activity and/or disrupting normal regulation of this channel.
MATERIALS AND METHODS
Materials. All chemicals were reagent grade and purchased from Sigma and Fischer Scientific unless noted otherwise. The plasmids encoding human -, -, and -ENaC subunits have been described previously ( 7, 17 ). Site-directed mutagenesis of -hENaC was performed using a QuikChange Site-Directed Mutagenesis Kit (Stratagene) or outsourced to Bio S & T (Montreal Quebec, Canada). Our initial -hENaC construct had the sequence A334, C618, and A663T (e.g., the A663T polymorphism). However, in the current work, we reference wild-type -hENaC as A334, C618, A663 (see also Refs. 1, 9, 14 - 16 ). From the initial A663T polymorphic construct, we created wild-type -hENaC and the double (with respect to wild-type) polymorphisms A334T + A663T and C618F + A663T. From wild-type, we next created the single polymorphic variants A334T and C618F, and from A334T + A663T we created the triple polymorphism A334T + C618F + A663T. All constructs were sequenced to confirm orientation, reading frame, and sequence and to identify and to ensure proper incorporation of sequence variations. Chinese hamster ovary (CHO) cells were cultured and prepared for experiments using standard methods as described previously ( 13, 17 ).
Patch-clamp recording and single-channel analysis. For both whole cell and excised outside-out patch-clamp recordings, polymorphic and wild-type channels were always assessed on the same day, alternating between each group. For each group, at least three different transfections were assayed.
Whole cell macroscopic current recordings of recombinant hENaC expressed in CHO cells were made under voltage-clamp conditions using standard methods ( 13, 17, 18 ). In brief, current through ENaC (at room temperature) was the inward, amiloride-sensitive Na + current with an extracellular bath solution of (in mM) 160 NaCl, 1 CaCl 2, 2 MgCl 2, and 10 HEPES (pH 7.4), and an intracellular pipette solution of (in mM) 120 CsCl, 5 NaCl, 2 MgCl 2, 5 EGTA, 10 HEPES (pH 7.4), 2.0 ATP, and 0.1 GTP. In some experiments, 10 mM DTT was also included in the pipette solution. Current recordings were acquired with a PC-505B patch-clamp amplifier (Warner Instruments) interfaced via a Digidata 1320A (Axon Instruments) to a PC running the pClamp 9.2 suite of software (Axon Instruments). Current was acquired under constant bath perfusion with a holding potential of 40 mV. Voltage ramps (500 ms) from 40 to -100 mV were used to measure ENaC activity. Whole cell capacitance was routinely compensated and was 10 pF. Series resistances, on average 2-5 M, were also compensated.
Single-channel current recordings were performed as described previously ( 17, 18 ). In brief, all experiments were performed at room temperature with constant perfusion using fire-polished pipettes of borosilicate glass (World Precision Instruments) with tip resistances of 7 M. All recordings were made in excised, outside-out patches with pipette and bath solutions of (in mM) 140 CsCl, 5 NaCl, 2 MgCl 2, 3 ATP, 0.1 GTP, 5 EGTA, 10 HEPES (pH 7.2), and 160 NaCl, 1 CaCl 2, 2 MgCl 2, 10 HEPES (pH 7.4), respectively. Inward currents are shown as downward deflections. All experiments were acquired using pClamp9.2 software with time and current amplitude data analyzed with this software in conjunction with Igor Pro 4.0 (Wavemetrics). Data were filtered at 100 Hz and collected at 500 Hz. For presentation, some current data were subsequently software filtered at 20 Hz.
Single-channel unitary current ( i ) was determined from the best-fit Gaussian distribution of all-point amplitude histograms. Channel activity ( NP o ) was calculated as: NP o = I / i, where I is the mean total current in a patch and i is the unitary current for ENaC. By definition, current at the closed state then is 0. Amiloride, applied to the extracellular face of outside-out patches containing ENaC, was used to establish the 0 currrent level. With this method, NP o is calculated independent from first establishing N and P o. Open probability ( P o ) was subsequently calculated by normalizing NP o for the total number of observed/estimated channels ( N ) in the patch. The precision in calculating P o with this method (or any other biophysical approach) ( 5 ) decreases as a function of the number of channels in a patch.
Statistics. All patch-clamp data are presented as means ± SE. Unpaired data were compared using a t -test. P < 0.05 was considered significant.
RESULTS
Figure 1 aligns the established sequences for -ENaC from several species showing the areas surrounding residues 334, 618, and 663 in -hENaC. The consensus amino acids at 334, 618, and 663 are threonine, cysteine, and alanine, respectively. Sequence variations at these sites in -hENaC, as discussed earlier, have previously been linked to increases in blood pressure and have distinct frequencies in people of European and African ancestry. The sequence around and including residue 334 is extremely well conserved in all species from amphibians to humans. This region of -ENaC resides in the extracellular domain of the channel. The most common residue at position 334 in humans, which differs from the consensus sequence, is alanine ( 1, 9 ). Thus the A334T polymorphism returns -hENaC to the consensus sequence. The cysteine at 618 and its variant C618F are in the cytosolic COOH terminus of -ENaC. This residue and its surrounding sequence are also strongly conserved in all mammals. However, an equivalent sequence is not present in amphibian and avian -ENaC. Comparatively, the region including residue 663 is less well conserved and also not found in the amphibian and avian sequences. Although first reported as a threonine, the most common residue at 663 in humans is alanine with A663T being the polymorphism ( 1, 9 ). Thus wild-type -hENaC contains A334, C618, and A663.
Fig. 1. Residues at 334, 618, and 663 in -hENaC are well conserved. Shown is the sequence alignment around residues 334, 618, and 663 for -ENaC from mammals, birds, and amphibians. Residue numbers are referenced to human -ENaC. Light gray represents sequence identity and conserved residues with dark gray shading residues that are weakly similar. Dissimilar residues are not shaded. Bottom : consensus sequence with residues containing sequence variations in -hENaC boxed. Human, cow, feral pig, mouse, rat, guinea pig, rabbit, bull frog, chicken, Xenopus laevis, and X. tropicalis -ENaC sequences are from NCBI locus P37088 , NP_777023 , NP_998923 , AAH49956 , P37089 , CAB64910 , AAS00455 , AAM53957 , AAB04954 , I51682 , and AAH64718 , respectively.
To determine whether sequence variations at positions 334, 618, and 663 affect channel activity, we overexpressed wild-type and polymorphic ENaC in CHO cells and quantified channel activity under voltage-clamp conditions. Figure 2 A shows typical macroscopic ENaC currents before (arrow) and after application of amiloride to CHO cells expressing wild-type ENaC and channels containing an -subunit bearing a single polymorphism (A334T, C618F, or A663T), two polymorphisms (A334T + A663T and C618F + A663T), and three polymorphisms (A334T + C618F + A663T). For these experiments, wild-type and polymorphic -hENaC were coexpressed with wild-type - and -hENaC. Currents were elicited by voltage ramps. Figure 2 B displays the macroscopic current-voltage ( I - V ) relationships for wild-type and polymorphic ENaC. Summarized in Fig. 2 C are the effects of -hENaC polymorphisms on channel activity. Compared with wild-type channels, those containing the single A663T and C618F but not A334T polymorphism had significantly more activity. Similarly, those containing two or more polymorphisms also had significantly more activity compared with wild-type channels. In addition, the double polymorphisms A334T + A663T and C618F + A663T both had significantly more activity compared with the single A334T polymorphic channel. In addition, C618F + A663T had significantly greater activity compared with A663T but not C618F channels. These results support the conclusion that polymorphisms at position 618 and 663 elevate activity with the C618F polymorphism having the single largest effect.
Fig. 2. Polymorphic ENaC has increased activity. A : overlays of macroscopic ENaC currents before (arrow) and after amiloride elicited by voltage ramps applied to Chinese hamster ovary (CHO) cells expressing wild-type ( top trace) and polymorphic ENaC. B : current-voltage relationship for the amiloride-sensitive ENaC current in voltage-clamped CHO cells expressing wild-type and polymorphic ENaC. C : summary of the amiloride-sensitive current density for ENaC at -80 mV in voltage-clamped CHO cells expressing wild-type and polymorphic ENaC. No. of observations appear in bars. *Vs. wild-type (wt); **vs. wt and A334T; vs. wt, A334T, and A663T; and vs. wt, A334T, A663T, and A334T + A663T.
Introducing additional polymorphisms on the background of C618F, as exemplified by the triple variant A334T, C618F, A663T, did not significantly increase activity. Thus the A663T and C618F polymorphisms did not have an additive effect. We interpret these results as the C618F and A663T polymorphisms increasing activity through a similar general mechanism, such as increasing the number of channels in the membrane, with the C618F polymorphism having a greater single effect on a possibly common step in a shared mechanism.
Macroscopic current through a population of channels is equal to the mean current through the individual channels multiplied by the number of these channels in the membrane and the mean open probability these channels have. Because the C618F and A663T polymorphisms increased ENaC activity at the macroscopic level, we asked whether these sequence variations did so by increasing the number of channels in the membrane, the open probability of these channels, and/or the unitary conductance of ENaC.
Figure 3 A shows results from excised, outside-out patches containing three to four wild-type ( top ) and A334T + C618F + A663T polymorphic channels ( bottom ) before and after addition of 10 µM amiloride to the extracellular face of the patch. As expected, amiloride blocked both wild-type and polymorphic channels. Figure 3 B shows typical current traces for wild-type ENaC in an excised, outside-out patch stepped from a holding potential of 0 mV down to -100 mV by -20-mV steps. Similar I - V relationships were established for polymorphic ENaC and are summarized in Fig. 3 C. The unitary current at 0 mV and conductance for wild-type and polymorphic ENaC were not different (see also Table 1 ). Thus increases in macroscopic ENaC activity must arise from changes in channel activity ( NP o ).
Fig. 3. Polymorphic ENaC have similar conductances but increased activity in excised, outside-out patches. A : representative current traces of ENaC in excised patches created from CHO cells expressing wild-type ( top ) and polymorphic ( bottom ) channels before and after addition of amiloride (10 µM; arrow) to the extracellular face of the channel. In all cases, channel blockade was quickly reversed on washout of amiloride (not shown). Holding potential is 0 mV with inward current down. B : representative current trace of wild-type ENaC in an excised patch held at 0 mV and stepped down to -100 mV by -20-mV voltage steps. C : single-channel current-voltage ( I - V ) relationships for wild-type and polymorphic ENaC in excised outside-out patches. D : summary graph of activity for wild-type and polymorphic ENaC in excised patches. *Vs. wt.
Table 1. Sequence variations in -ENaC affect activity
As summarized in Fig. 3 D and expanded on in Table 1, C618F and A663T polymorphisms significantly increased ENaC activity in excised, outside-out patches. Indeed, the fold- increase in activity as measured in excised patches was comparable to the increase in activity observed for ENaC in whole cell voltage-clamp experiments ( Fig. 2 C ). Similar to the C618F and A663T single variants, the double variants A334T + A663T and C618F + A663T also had greater activity compared with wild-type channels in excised patches. Also paralleling findings in whole cell voltage-clamp experiments, the activity of the triple variant A334T + C618F + A663T in outside-out patches trended greater then that of the A334T + A663T double variant but similar to that for the C618F + A663T double variant and the C618F single polymorphism. As shown in Table 1, the resistances and by extension the size of the pipettes used to investigate wild-type and polymorphic ENaC in excised patches were uniform.
In an attempt to define the mechanism by which polymorphisms increase channel activity, we estimated N and P o for wild-type and polymorphic ENaC in excised, outside-out patches (see Table 1 ) and made a direct comparison of representative patches containing three wild-type and polymorphic channels ( Fig. 4 ). Perusal of the representative current traces ( left ) and corresponding all point histograms ( right ) for wild-type and polymorphic ENaC shown in Fig. 4 supports the idea that these sequence variations have little effect on channel P o. The P o for wild-type, C618F, A334T + A663T, and A334T + C618F + A663T channels in these representative patches were 0.42, 0.50, 0.41, and 0.59, respectively. For the entire data set (see Table 1 ), mean P o ranged from extremes of 0.44 ± 0.04 to 0.57 ± 0.03 for wild-type and A334T + C618F + A663T channels, respectively. This agrees well for ENaC in epithelia stimulated with aldosterone ( 5 ). The modest 1.3-fold increase in P o for channels containing C618F clearly does not explain the approximately 3.8-fold increase in activity for these channels observed in whole cell and excised patches. Comparison of N for patches containing wild-type and polymorphic channels reveals that polymorphisms rather most likely increase the number of active channels in the membrane. Patches containing channels with either the C618F or A663T polymorphism alone or in combination had a significantly greater N compared with those containing wild-type channels. Indeed, N for C618F channels was 3.4-fold greater than wild-type. This fold increase in N approximates the increase in activity. Similarly, the significant 2.2-fold increase in N for patches containing A663T channels closely approximates the corresponding 2.6-fold increase in activity for these channels compared with wild-type channels. It is likely that underestimation of N for patches containing the C618F and A663T variants resulted in artificially high P o values for these patches often contained multiple channels, which complicated data analysis. Nevertheless, the fact that N is significantly increased for C618F and A663T channels, even though it may be underestimated, emphasizes the likelihood that these polymorphisms increase activity by increasing the number of channels in the membrane.
Fig. 4. Polymorphisms at 618 and 663 increase ENaC number not open probability. Representative current traces ( left ) and corresponding all-point amplitude histograms ( right ) for ENaC in excised patches created from CHO cells expressing wild-type ( top ) and polymorphic channels. Holding potential is 0 mV with inward current down. Only patches containing 3 observable channels were chosen for comparison.
Because cysteine at 618 is well conserved and cysteine residues, in some cases, are able to be modified and sensitive to redox reactions, as well as involved in formation of disulfide bridges, we tested the effects of the reducing agent DTT on wild-type and C618F polymorphic channels. Figure 5 A shows the steady-state activity of wild-type and polymorphic ENaC after dialyzing the cytosol with (10 mM) DTT. As reported earlier, C618F channels, in the absence of DTT, have greater activity compared with wild-type channels (see Fig. 2 C ). However, in the presence of DTT, activity of wild-type and C618F polymorphic channels was not different. As shown in Fig. 5 B, this reflects a significant increase in activity in response to DTT for only wild-type channels. Thus DTT had a markedly greater effect on wild-type channels compared with C618F variants.
Fig. 5. Reducing agents have a greater effect on wild-type compared with C618F channels. A : summary graph of wild-type and polymorphic ENaC activity in the presence of 10 mM cytosolic DTT. B : summary graph of the relative increase in activity of wild-type and polymorphic ENaC in response to DTT. *Vs. C618F.
DISCUSSION
The current results demonstrate that the C618F and A663T polymorphisms but not the A334T polymorphism in -hENaC enhance channel activity when expressed in mammalian cells. This finding is consistent with the former two polymorphisms possibly contributing to elevated blood pressure. Our data suggest that the primary mechanism by which the C618F and A663T polymorphisms increase ENaC activity is by increasing the apparent number of active channels in the plasma membrane, reflecting either an increase in insertion/retention of channels in the membrane or an increase in activity of normally quiescent channels found within the membrane.
It is interesting that the A334T polymorphism is within an extremely well-conserved region of -ENaC. In general, this implies that sequence variations at this site might be expected to affect activity. However, the A334T substitution merely returns -hENaC to the consesnsus sequence for -ENaC (see Fig. 1 ). Thus A334 is actually the sequence variation away from the consensus sequence and might be expected to decrease or increase ENaC activity compared with channels having A334T. We find that the A334T substitution does not influence recombinant hENaC activity in CHO cells. We conclude that having threonine compared with alanine at position 334 is silent with respect to changes in channel structure and regulation.
Similar to A334T, the C618F polymorphism is found in a well-conserved region, at least in mammals, of -ENaC. The current results demonstrate that the C618F polymorphism has a profound effect on ENaC activity. This is the first report, we are aware of, directly demonstrating that this polymorphism has functional consequences with respect to augmenting ENaC activity. Indeed, this is the first report that even directly tested the effects of this polymorphism on ENaC activity. It is interesting that C618 is absolutely conserved in every mammalian -ENaC sequence as yet reported. Cysteines are unique for they are sensitive to cellular redox states and can be modified to sulfinic and sulfenic acids and participate in the formation of disulfide bridges. Thus cysteine residues often directly impact protein structure and regulation, which potentially can affect channel gating, assembly, and/or trafficking. It is provocative to speculate that the conserved cysteine at 618 is sensitive to cellular redox states or involved in formation of a critical disulfide bridge within ENaC. Results in Fig. 5 are consistent with either possibility. Another reasonable interpretation of results in Fig. 5 is that a regulator of ENaC is sensitive to DTT and that wild-type but not C618F channels respond to this regulator. In this scenario, then, the actions of DTT on ENaC would be indirect. A recent finding by Kellenberger and colleagues ( 3 ) demonstrated that SH-reactive reagents, which can modify cysteines, changed ENaC activity by increasing the open probability of channels in the membrane. Thus, while the current results suggest that C618F primarily increases the number of active ENaC in the membrane, it remains unclear whether this polymorphism does so by increasing channel insertion/retention or leading to increased activity of normally quiescent channels found within the membrane.
One final point of consideration with respect to the C618F polymorphism must be that amphibian and avian -ENaC polypeptides do not contain a cysteine comparable to that at 618. In fact, the region surrounding C618 is extremely divergent between amphibian and mammalian ENaC. Thus the requirement for this conserved cysteine in mammalian -ENaC for normal channel activity must be rather new in an evolutionary sense.
Our finding that channels containing the A663T polymorphism have increased activity is consistent with that published recently by Samaha and colleagues ( 11 ). These investigators demonstrated that the A663T polymorphism increased activity by increasing the number of ENaC in the membrane. Although our studies were performed on hENaC in mammalian cells, this earlier work was performed on hENaC expressed in X. laevis oocytes. In apparent conflict with these findings, Ambrosius and colleagues ( 1, 9 ) find that threonine at 663 is protective against elevations in blood pressure. This predicts that channels having the A663T polymorphism would have less and not greater activity. Resolution of this apparent controversy should help clarify the importance of the A663T polymorphism to control of blood pressure in humans.
The current finding that the C618F polymorphism augments hENaC activity in a mammalian background is important for it demonstrates that this sequence variation results in activity changes capable of elevating blood pressure. This supports results linking this polymorphism to elevated blood pressure in people of African ancestry with both being consistent with the idea that ENaC gain of function possibly contributes to some types of polygenic essential hypertension.
GRANTS
This research was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant RO1-DK-59594 and R21-DK-073071 and American Heart Association Texas Affiliate Grant 0355012Y (to J. D. Stockand).
ACKNOWLEDGMENTS
J. Medina is recognized for excellent technical assistance.
【参考文献】
Ambrosius WT, Bloem LJ, Zhou L, Rebhun JF, Snyder PM, Wagner MA, Guo C, and Pratt JH. Genetic variants in the epithelial sodium channel in relation to aldosterone and potassium excretion and risk for hypertension. Hypertension 34: 631-637, 1999.
Hummler E and Horisberger JD. Genetic disorders of membrane transport. V. The epithelial sodium channel and its implication in human diseases. Am J Physiol Gastrointest Liver Physiol 276: G567-G571, 1999.
Kellenberger S, Gautschi I, Pfister Y, and Schild L. Intracellular thiol-mediated modulation of ENaC activity. J Biol Chem 280: 7739-7747, 2005.
Kellenberger S and Schild L. Epithelial sodium channel/degenerin family of ion channels: a variety of functions for a shared structure. Physiol Rev 82: 735-767, 2002.
Kemendy AE, Kleyman TR, and Eaton DC. Aldosterone alters the open probability of amiloride-blockable sodium channels in A6 epithelia. Am J Physiol Cell Physiol 263: C825-C837, 1992.
Lifton RP, Gharavi AG, and Geller DS. Molecular mechanisms of human hypertension. Cell 104: 545-556, 2001.
McDonald FJ, Price MP, Snyder PM, and Welsh MJ. Cloning and expression of the - and -subunits of the human epithelial sodium channel. Am J Physiol Cell Physiol 268: C1157-C1163, 1995.
Pratt JH. Low-renin hypertension: more common than we think? Cardiol Rev 8: 202-206, 2000.
Pratt JH. Erratum to genetic variants in the epithelial sodium channel in relation to aldosterone and potassium excretion and risk for hypertension. Hypertension 41: e1, 2003.
Rossier BC, Pradervand S, Schild L, and Hummler E. Epithelial sodium channel and the control of sodium balance: interaction between genetic and environmental factors. Annu Rev Physiol 64: 877-897, 2002.
Samaha FF, Rubenstein RC, Yan W, Ramkumar M, Levy DI, Ahn YJ, Sheng S, and Kleyman TR. Functional polymorphism in the carboxyl terminus of the -subunit of the human epithelial sodium channel. J Biol Chem 279: 23900-23907, 2004.
Snyder PM. The epithelial Na + channel: cell surface insertion and retrieval in Na + homeostasis and hypertension. Endocr Rev 23: 258-275, 2002.
Staruschenko A, Patel P, Tong Q, Medina JL, and Stockand JD. Ras activates the epithelial Na + channel through phosphoinositide 3-OH kinase signaling. J Biol Chem 279: 37771-37778, 2004.
Su YR and Menon AG. Epithelial sodium channels and hypertension. Drug Metab Dispos 29: 553-556, 2001.
Sugiyama T, Kato N, Ishinaga Y, Yamori Y, and Yazaki Y. Evaluation of selected polymorphisms of the Mendelian hypertensive disease genes in the Japanese population. Hypertens Res 24: 515-521, 2001.
Swift PA and MacGregor GA. Genetic variation in the epithelial sodium channel: a risk factor for hypertension in people of African origin. Adv Ren Replace Ther 11: 76-86, 2004.
Tong Q, Gamper N, Medina JL, Shapiro MS, and Stockand JD. Direct activation of the epithelial Na + channel by phosphatidylinositol 3,4,5-trisphosphate and phosphatidylinositol 3,4-bisphosphate produced by phosphoinositide 3-OH kinase. J Biol Chem 279: 22654-22663, 2004.
Tong Q and Stockand JD. Receptor tyrosine kinases mediate epithelial Na + channel inhibition by epidermal growth factor. Am J Physiol Renal Physiol 288: F150-F161, 2005.
作者单位：1 Department of Physiology, University of Texas Health Science Center at San Antonio, San Antonio, Texas; and 2 Department of Molecular Genetics, Biochemistry, and Microbiology, University of Cincinnati Medical Center, Cincinnati, Ohio