点击显示 收起
【摘要】 We studied the effects of two mutations of the extracellular loop of the -subunit of the (ENaC) on amiloride-sensitive current in Xenopus laevis oocytes and the inhibition of this current by 3-morpholinosydnonimine (SIN-1). Injection of oocytes with wild-type (wt) -, -, -rENaC cRNA (8.3 ng/subunit) resulted 48-72 h later in inward Na + currents (-5.5 ± 0.8 µA; means ± SE at -100 mV; n = 21), which were completely inhibited by amiloride. Oocytes injected with either Y279A - or Y283A - and -, -rENaC cRNAs had significantly lower Na + currents. Furthermore, Y279A -, -, -rENaC-injected oocytes had a higher K i for amiloride (0.54 ± 0.97 vs. 0.10 ± 0.04 µM; P < 0.01). Exposure of oocytes to SIN-1 (1 mM) for 5 min decreased both total Na + and amiloride-sensitive currents across wt and Y279A - but not Y283A -, -, -rENaC. Furthermore, exposure to SIN-1 increased the K i for amiloride across wt but not Y279A -, -, -rENaC-injected oocytes. These data indicate that both tyrosines are important for proper ENaC function and their oxidative modifications contribute to altered ENaC function.
【关键词】 peroxynitrite tyrosines morpholinosydnonimine Xenopus laevis oocytes whole cell currents
AMILORIDE - SENSITIVE EPITHELIAL Na + channels (ENaC), located at the apical membrane of Na + -transporting epithelia, play an essential role in the control of salt and fluid homeostasis in a variety of epithelia including kidney, lung, and colon ( 5, 23 ). Three homologous subunits (,, and ) of amiloride-sensitive ENaC were cloned from rat, human, and other species ( 6, 21, 24 ). Mice lacking -ENaC are unable to clear fetal fluid and die shortly after birth from respiratory distress ( 12 ). In contrast, -ENaC (-/-) mice showed normal prenatal development but on low-salt diets developed type 1 pseudohypoaldosteronism ( 25 ). In addition, abnormalities in ENaC channel number and opening have been linked to the pathogenesis of cystic fibrosis ( 31 ) and Liddle's syndrome ( 25 ).
Inflammatory responses lead to the generation of reactive oxygen and nitrogen species (RONS) including superoxide (O 2 - ·), nitric oxide (NO), and their product peroxynitrite (ONOO - ) ( 3 ). In response to various inflammatory stimuli, lung endothelial, alveolar, and airway epithelial cells, as well as activated alveolar macrophages, produce both NO and O 2 - · ( 9, 13 ). Recent studies implicated NO and ONOO - in the pathogenesis of many diseases, such as septic shock, arthritis, acute lung injury, and atherosclerosis ( 19, 28, 30, 33 ). Reactive oxygen-nitrogen intermediates (RONS) have also been demonstrated to have inhibitory effects on Na + transport across epithelia ( 1, 11, 15, 20 ).
The decomposition of 3-morpholinosydnonimine (SIN-1) produces NO, O 2 - ·, ONOO -, and hydrogen peroxide ( 10 ) as well as a number of reactive intermediates formed by the reaction of peroxynitrite with SIN-1c, its stable byproduct ( 32 ). We showed that small concentrations of ONOO -, generated by SIN-1, profoundly inhibit amiloride-sensitive whole cell sodium conductance in Xenopus laevis oocytes expressing wild-type -, -, -rENaC ( 7 ). ONOO - may exert its effects by oxidizing, nitrating, or nitrosylation critical amino acid residues in either ENaC or other structural proteins necessary for its proper function. The specific residues involved in these responses have yet to be identified.
A limited domain within the extracellular loops of the - and -subunits has been shown, based on mutagenesis studies, to play an important role in mediating changes in channel gating that were observed with extracellular Ni + and Na + (i.e., Na + self inhibition) ( 26, 27 ). In channels composed solely of -subunits, mutations in this domain affect both channel gating and channel block by amiloride ( 14, 17 ). This region within the -subunit includes residues 278-283 (WYRFHY). As this region has a role in modulating channel gating, modification of these tyrosine residues by reactive oxygen-nitrogen species might diminish Na + currents and their response to amiloride. We therefore generated -rENaC mutant proteins, where tyrosine residues at positions 279 (mutation Y279A ) and 283 (mutation Y283A ) were replaced with alanine. We then expressed these mutants, either alone or in combination with wild-type - and -rENaC in X. laevis oocytes, and studied their effects on amiloride-sensitive whole cell currents as well as the sensitivity of these currents following exposure of oocytes to SIN-1.
MATERIALS AND METHODS
RNA synthesis. The pSport plasmid (GIBCO BRL, Gaithersburg, MD), containing either -, -, or -rENaC (a generous gift from Dr. B. Rossier, University of Lausanne, Switzerland), mutation Y279A, or mutation Y283A, was linearized by overnight incubation with Not I (Promega, Madison, WI). Sense RNA was in vitro transcribed from purified plasmid DNA using T7 polymerase according to the manufacturer's instructions (Ambion, Austin, TX). The integrity of the cRNA was verified by denaturing gel electrophoresis through 1% agarose-formaldehyde gel ( 7 ).
Injection into oocytes. Oocytes were obtained from appropriately anesthetized adult female X. laevis toads ( Xenopus Express, Burley Hill, FL) by standard techniques ( 7 ). Ovarian tissue was digested with 2 mg/ml collagenase (type 1A, Sigma) in Ca 2+ -free OR-2 medium, containing (in mM) 82.5 NaCl, 2.4 KCl, 1.0 MgCl 2, and 5 HEPES, pH 7.4, under rotation at 24°C for 2 h. Defolliculated oocytes were washed three times in OR-2 medium. Stage VI oocytes were selected to recover overnight in half-strength Leibovitz medium (GIBCO BRL), containing 15 mM HEPES, penicillin (100 U/ml), streptomycin (100 µg/ml), and 5% horse serum, pH 7.6, at 18°C. A Nanoject (Drummond, Broomall, PA) microinjector was used to inject 12.5 or 25 ng of total cRNA in a 50-nl volume into each oocyte. Control oocytes were injected with 50 nl of RNAase-free H 2 O. Injected cRNA contained -, -, and -rENaC in equal amounts (i.e., 4.17 or 8.3 ng of each subunit). Oocytes received either wild-type -rENaC, Y279A -rENaC or Y283A -rENaC, and - and -rENaC. In other experiments, cells were injected with either 50 ng wild-type -rENaC or Y279A -rENaC alone. Oocytes were incubated in oocyte culture medium at 18°C until use. The culture medium was changed daily.
Whole cell current measurements. The membrane current was evaluated 48-72 h postinjection using a two-electrode voltage clamp ( 7 ). The oocytes were held in a small groove in an experimental chamber of 0.5-ml volume at room temperature (21°C). The chamber was perfused continuously at a rate of 2 ml/min with normal Ringer solution, containing (in mM) 100 NaCl, 2 KCl, 1 MgCl 2, 1.8 CaCl 2, and 15 HEPES at pH 7.6 (osmolarity 200-220 mosM). Oocytes were impaled with two 3 M KCl-filled electrodes, with resistances of 0.5-1.5 M. The electrodes were connected to a GeneClamp 500 (Axon Instruments, Foster City, CA) current-voltage ( I-V ) clamp amplifier via Ag-AgCl pellet electrodes and connected to the bath via a 3 M KCl-Agar Bridge. A MacIntosh IIci computer running Axodata (Axon Instruments) acquisition software controlled the voltage clamp.
Oocytes were clamped at a holding potential of 0 mV. Currents were recorded every 20 s by stepping from the holding potential to -100 mV for 400 ms, back to the holding potential for 50 ms and then to +100 mV for 400 ms. I-V relationships were determined by stepping from the holding potential to -100 mV through +100 mV in 20-mV increments for 600 ms. In a number of cases, once control I-V curves were obtained, oocytes were perfused with increasing concentrations of external solutions containing amiloride (1 x 10 -5 to 100 µM). I-V relationships were then repeated when currents reached plateau values for each amiloride concentration (usually with 2-4 min).
Exposure to SIN-1. Cells were incubated in half-strength Leibowitz medium with or without 1 mM SIN-1 for 2 h at 21°C before electrophysiological recordings were made. Measurements were repeated following perfusion of oocytes with amiloride (10 µM). Because whole cell currents in oocytes vary, control measurements were obtained in a separate group of oocytes before or after the SIN-1 measurements. In a number of additional experiments, oocytes, injected with 4.17 ng each of -, -, -rENaC, were perfused with a solution containing 1 mM SIN-1 in normal Ringer at 21°C for up to 60 min. Currents were recorded every 20 s by stepping from the holding potential of 0 mV to -100 mV for 400 ms, and back to the holding potential for 50 ms and then to +100 mV for 400 ms. In other cases, oocytes were perfused with a solution containing 1 mM SIN-1 and 0.2 mM urate, a well-known scavenger of ONOO - ( 8 ).
Quantitation of ONOO - production. ONOO - generation by 1 mM SIN-1 (Calbiochem, La Jolla, CA) in 10 mM phosphate buffer (pH 5.5) was quantified by measuring the rate of rhodamine formation from the oxidation of dihydrorhodamine 123, as previously described ( 7 ).
Statistical analysis. All results are reported as means ± SE. Data were analyzed by paired or unpaired t -tests or one-way ANOVA and the Bonferroni multiple comparison tests using INSTAT Software (GraphPad Software). P values of <0.05 were considered significant.
RESULTS
Y279 and Y283 are important in ENaC function and in the inhibition of ENaC by amiloride. Injection of oocytes with 8.3 ng each of -, -, -rENaC cRNAs resulted in the expression of significant amounts of inward Na + current at -100 mV (-5.5 ± 0.8 µA; means ± SE; n = 21). Substitution of Y279A or Y283A for wild-type -rENaC resulted in Na + currents at -100 mV that were significantly lower than wild-type controls (-1.92 ± 0.5; n = 24 and -1.91 ± 0.4; n = 14, respectively; see Fig. 1 ). In a number of cases, once control I-V relationships were obtained, oocytes were perfused with increasing concentrations of external solutions containing amiloride (1 x 10 -5 to 100 µM). As can be seen ( Fig. 2 and Table 1 ), in all cases, 10 µM amiloride inhibited the inward (Na + ) currents. However, substitution of Y279A, but not Y283A, for wild-type -rENaC significantly increased the K i for amiloride from 0.10 ± 0.04 to 0.54 ± 0.97 µM ( P < 0.01; Fig. 2 ).
Fig. 1. Whole cell current-voltage ( I-V ) relationships across oocytes injected with wild-type (wt) -, -, -r epithelia Na + channels (ENaC), Y279A -, -, -rENaC, or Y283A -, -, -rENaC: oocytes were injected with 25 ng total cRNA (8.3 ng/subunit) and held at a potential of 0 mV. Currents were elicited by applying voltage steps from -100 to +100 mV in 20-mV increments, with each step lasting 600 ms. The pipettes were filled with the standard internal solution (3 M KCl). Currents were recorded after perfusing the oocytes with standard external solution (normal Ringer) for 10 min. Values are means ± SE; n = number of oocytes. See Table 1 for statistical analysis of these data.
Fig. 2. Substitution of alanine for tyrosine 279 affects the sensitivity of whole cell Na + current to amiloride: whole cell current recordings were obtained from oocytes injected with wt -, -, -rENaC, Y279A -, -, -rENaC, or Y283A -, -, -rENaC (8.3 ng/subunit, 25 ng total cRNA). Inward Na + currents were recorded at -100 mV. Cells were perfused with normal Ringer solution containing the indicated concentrations of amiloride until a steady-state current was obtained, usually within 3-5 min. The percentage (%) of total amiloride-sensitive (amil.sensit.) current ( y -axis) obtained at each amiloride concentration was calculated as follows: % = [( I x - I 100 )/( I 0 - I 100 )] x 100, where I 0 is the current obtained without amiloride, I x is the steady-state current during perfusion with a given concentation of amiloride (in µM), and I 100 is the current measured during perfusion with 100 µM amiloride. Values are means ± SE; n = number of oocytes.
Table 1. Whole cell oocyte currents at -100 mV during perfusion with vehicle or amiloride
In a second series of experiments, we injected oocytes with 50 ng each of either wild-type -rENaC or Y279A -rENaC cRNA alone. Wild-type injected oocytes exhibited significant values of total Na + and amiloride-sensitive currents, which as previously reported ( 6 ), were considerably lower than those injected with -, -, -rENaC ( Table 1 ). On the other hand, oocytes injected with Y279A -rENaC had significantly lower Na + current values, which were not different from those injected with water ( Table 1 ). However, a small component of this current ( 20%) was still inhibited by amiloride. These data indicate that while both Y279 and Y283 are important in determining ENaC current, Y279 also plays a role in determining the sensitivity of sodium current to amiloride.
Effects of reactive oxygen-nitrogen intermediates on Na + currents. Exposure to SIN-1 reduced whole cell and amiloride-sensitive currents of oocytes injected with 8.3 ng each of -, -, -rENaC cRNA and Y279A -, -, -rENaC. On the other hand, SIN-1 did not alter either total or the amiloride-sensitive currents of oocytes injected with Y283A -, -, -rENaC ( Figs. 3, 4, and Table 2 ). In addition, SIN-1 decreased amiloride-sensitive currents of oocytes injected with 50 ng cRNA of wild-type -rENaC from -0.15 ± 0.04 to -0.04 ± 0.02 µA, n = 12, P < 0.01 but had no effects on Y279A -rENaC currents, perhaps due to their very small magnitude ( Table 2 ).
Fig. 3. Effects of 3-morpholinosydnonimine (SIN-1) on whole cell Na + currents in oocytes injected with -, -, -rENaC, Y279A -, -, -rENaC, and Y283A -, -, -rENaC. Whole cell currents were recorded in oocytes injected with 8.3 ng/subunit of wt -, -, -rENaC ( A ), Y279A -, -, -rENaC ( B ), and Y283A -, -, -rENaC ( C ). Oocytes were incubated in half-strength Leibowitz medium with or without 1 mM SIN-1 for 2 h at 21°C before I-V relationships were recorded. For each group (control and SIN-1), I-V relationships were repeated following perfusion of oocytes with amiloride (10 µM). Values are means ± SE; n = number of oocytes. See Table 2 for statistical analysis of these data.
Fig. 4. Effects of SIN-1 on whole cell amiloride-sensitive Na + currents in oocytes injected with -, -, -rENaC ( A ), Y279A -, -, -rENaC ( B ), and Y283A,, -rENaC ( C ). Amiloride-sensitive currents were calculated by subtracting the currents at each voltage following pefusion with amiloride (10 µM) from the corresponding control value for the given condition (vehicle or SIN-1). All other conditions were as described in Fig. 3. Values are means ± SE; n = number of oocytes; n 8; see Table 2 for exact n values and statistical analysis.
Table 2. Whole cell oocyte currents at -100 mV during perfusion with vehicle or amiloride before and after incubation with 1 mM SIN-1 for 2 h
To examine the time course of the effect of SIN-1 on Na + currents, oocytes injected with wild-type -, -, -rENaC (4.17 ng/subunit) were held at -100 mV and inward Na + current was recorded while the oocytes were perfused with SIN-1 (1 mM). Inward current in oocytes decreased 5 min from the start of perfusion with SIN-1 and remained stable for the rest of the perfusion period ( Fig. 5 ). As previously shown, SIN-1 generates ONOO - in a time-dependent fashion, with very small levels being generated within the first 5 min ( 7 ). Addition of urate, an ONOO - scavenger, inhibited production of ONOO - (data not shown) and the SIN-1-induced reduction of both total and amiloride-sensitive Na + currents ( Fig. 6 ).
Fig. 5. Reduction of inward Na + current by SIN-1 is rapid and inhibited by urate. The time course of total inward current recorded from oocytes injected with wt -, -, -rENaC (4.17 ng/subunit) at -100 mV was recorded. The holding potential was 0 mV. Cells were perfused with normal Ringer solution containing either 1 mM SIN-1, 0.2 mM urate, 0.2 mM urate plus 1 mM SIN-1, or the same volume of vehicle (Na + /Na + -phosphate buffer) for 80 min at 21°C. Increasing concentrations of amiloride (100 nM to 100 µM) were added to the bath solution after 60 min. One-millimolar SIN-1 decreased both total inward current and amiloride-sensitive current and these effects were inhibited by 0.2 mM urate (which totally inhibits peroxynitrite formation). Values are means ± SE.
Fig. 6. Urate, a peroxynitrite scavenger, prevents the effects of SIN-1 on Na + currents. Oocytes were injected with wt -, -, -rENaC (4.17 ng/subunit). Forty-eight hours later, they were perfused with SIN-1 (1 mM) in the presence or absence of urate (200 µM). Membrane potentials were varied from the holding value of 0 to + or -100 mV. After 60 min of perfusion with vehicle, SIN-1, urate, or SIN-1 plus urate, oocytes were perfused with a solution containing 10 µM amiloride. Difference ( I ) currents were calculated by subtracting the current at -100 mV from the corresponding value before the application of amiloride or the indicated agent. Values are means ± SE; n = number of oocytes. * P < 0.01 compared with the vehicle value (unpaired t -test). # P < 0.01 compared with the corresponding SIN-1 value (analysis of variance followed by the Bonferroni multiple comparison test).
In additional experiments, we examined the effects of the amiloride sensitivity of Na + currents from SIN-1-exposed oocytes. Exposure to SIN-1 (1 mM for 2 h) significantly reduced the amiloride sensitivity of inward Na + current in oocytes injected with wild-type -, -, -rENaC [ K i 0.41 ± 0.1 vs. 1.53 ± 0.23 µM ( P < 0.01; Fig. 7 A ) but not Y279A -, -, -rENaC ( Fig. 7 B )]. These experiments were not conducted in Y283A -, -, -rENaC-injected oocytes because in this case similar values of amiloride-sensitive currents were obtained before and after treatment of oocytes with SIN-1 (see Fig. 3 ).
Fig. 7. SIN-1 reduces the sensitivity of whole cell Na + current to amiloride in oocytes injected with -, -, -rENaC or Y279A -, -, -rENaC but not Y283A -, -, -rENaC. Whole cell currents (mean values) were recorded in oocytes injected with wt -, -, -rENaC ( A ) or Y279A -, -, -rENaC (8.3 ng/subunit; B ). Oocytes were incubated in half-strength Leibowitz medium with or without 1 mM SIN-1 for 2 h at 21°C before I-V relationsips were made. Inward Na + currents were recorded at -100 mV. The percentage (%) of total amiloride-sensitive current ( y -axis) obtained at each amiloride concentration was calculated as follows: % = [( I x - I 100 )/( I 0 - I 100 )] x 100, where I 0 is the current obtained without amiloride, I x is the steady-state current obtained during perfusion with x M amiloride, and I 100 is the current obtained during perfusion with 100 µM amiloride. Values are means ± SE.
DISCUSSION
Amiloride-sensitive Na + transport in epithelia is mediated by the ENaC, which has at least three homologous subunits, termed -, -, and -rENaC ( 6, 24 ). Our data, in agreement with that of others, indicate that the -subunit alone is sufficient to mediate amiloride-sensitive Na + current, but this current is greatly potentiated by the addition of the - and -subunits ( 6 ).
The -rENaC subunit has two membrane-spanning domains and an extracellular loop that is rich in tyrosine residues ( 23 ). Recent work examining the properties of channels composed solely of -subunits identified a domain within the extracellular loop of -rENaC (residues 278-283) where mutations or deletions affect amiloride block as well as the open probability of the channels ( 14, 17 ). Furthermore, selected mutations within this region of -ENaC or within the corresponding region of -ENaC altered the channel's response to external Ni 2+ ( 27 ) and to changes in extracellular Na + ( 26 ). Our results indicate that residues Y279 and Y283 in -rENaC are important in -rENaC-mediated Na + transport as mutations of either markedly inhibit whole cell Na + current in X. laevis oocytes ( Fig. 1 ).
Two potential mechanisms may account for these effects. First, the introduced mutations in the -subunit may interfere with proper trafficking or accelerate the removal of ENaC subunits to and from the plasma membranes. ENaC is a heteromultimeric channel consisting of at least three subunits. Presently, it is unclear which subunit is the rate-limiting one in the trafficking of this complex to the level of the plasma membrane. Although some studies suggest that both the - and -subunits are essential for this process ( 18 ), others feel that -ENaC or some of its recently identified partners which interact with the ubiquitin-ligase Nedd4-2 play an important role ( 22 ). Second, channels formed by mutant - and wild-type -, -subunits may have different biophysical properties, including decreased open times and open probabilities. Indeed, recent findings indicate that expression of Y279A and Y283A -rENaC into Chinese hamster ovary cells results in channels with altered biophysical properties and considerably higher K i for amiloride than channels formed by wild-type -rENaC ( 17 ). Interestingly, oocytes injected with either Y279A - rENaC or Y279A -, -, -rENaC expressed Na + currents that were less sensitive to amiloride than oocytes injected with the corresponding wild-type controls. Because we did not measure surface levels of ENaC expression and single-channel activity following wild-type and mutant ENaC injections into oocytes, we cannot distinguish between these two possibilities.
RONS may cause injury to host tissues in inflammatory diseases ( 19 ). In particular, ONOO - or reactive intermediates generated by the myeloperoxidase-catalyzed reaction of reactive species released from activated neutrophils may nitrate tyrosine residues and oxidize and/or nitrosate cysteines in host proteins, thus impairing their function ( 4 ). Tyrosine nitration is a covalent posttranslational protein modification derived from the reaction of proteins with nitrating agents. Protein nitration appears to be a selective process because not all tyrosine residues in proteins or all proteins are nitrated in vivo ( 29 ). In previous studies, we demonstrated that ONOO -, formed by the decomposition of SIN-1, inhibits amiloride-sensitive Na + conductance in X. laevis oocytes injected with -, -, -rENaC ( 7 ). Bolus addition of peroxynitrite also inhibited amiloride-sensitive 22 Na transport across suspensions of freshly isolated ATII cells ( 11 ) and membrane vesicles of colonic cells of dexamethasone-treated rats known to contain Na + channels ( 1 ). More recently, we showed that exposure of airway cells to NO donors decreases cystic fibrosis transmembrane regulator expression in both heterologously transfected LLC-PK 1 cells ( 16 ) and airway cells ( 2 ). In the latter, reactive oxygen-nitrogen intermediates also decreased cAMP-activated Cl - currents and this effect was correlated with CFTR nitration. It should be noted that while most investigators attribute the oxidizing and nitrating effects of SIN-1 to ONOO -, recent evidence indicates that complex intermediates, formed by the reaction of ONOO - with SIN-1c, its stable byproduct, are capable of activating Cl - currents ( 32 ), presumably by the oxidative modification of target proteins.
Herein we demonstrated that incubation with SIN-1 of X. laevis oocytes injected with wild-type -, -, -rENaC reduced both total amiloride-sensitive current and caused a rightward shift in the dose response for amiloride. Because of the shift in the dose-response curve, we investigated if either of two tyrosine residues in a 6-amino acid stretch of -rENaC previously associated with amiloride binding were targets for SIN-1 nitration. When oocytes expressing Y279A -, -, -rENaC were exposed to SIN-1, there was a significant reduction in whole cell current. Conversely, the same maneuver carried out with the companion construct, Y283A -, -, -rENaC had nearly no effect on total amiloride-sensitive current. Because the substitution of alanine for tyrosine at residue position 279 did not prevent current downregulation by SIN-1, whereas SIN-1 had no effect on current associated with mutation of Y283, these results suggest that Y283 may be a potential target for SIN-1-mediated nitration. However, SIN-1 had no effect on the dose-response curve for amiloride for the Y279A construct, which was not significantly different from that for the wild-type channel in the absence of SIN-1, although as noted above, it caused a significant shift in the amiloride dose-response curve for the wild-type channel. These results are consistent with a role for Y279 in the binding of amiloride and suggest that nitration at this residue may hinder amiloride access to the binding domain.
GRANTS
This study was supported by grants to Dr. S. Matalon (HL-51173, HL-31197, P30 DK-54781), Dr. Kleyman (DK-54354), and Dr. Fuller (DK-37206).
ACKNOWLEDGMENTS
The authors thank Dr. M. Duvall, C. Myles, and E. C. Walthall for advice and excellent technical assistance.
Address for reprint requests and other correspondence: S. Matalon, Dept. of Anesthesiology, Univ. of Alabama at Birmingham, 901 19th St. S, BMR II, Rm. 224, Birmingham, AL 35205-3703 (E-mail: sadis{at}uab.edu