Cl - channels of the distal nephron

【摘要】  Cl - currents were observed under whole cell clamp conditions in cells of the rat cortical collecting duct (CCD), connecting tubule (CNT), and thick ascending limb of Henle's loop (TALH). These currents were much larger in intercalated cells compared with principal cells of the CCD and were also larger in the TALH and in the CNT compared with the CCD. The conductance had no strong voltage dependence, and steady-state currents were similar in inward and outward directions with similar Cl - concentrations on both sides of the membrane. Current transients were observed, particularly at low Cl - concentrations, which could be explained by solute depletion and concentration in fluid layers next to the membrane. The currents had a remarkable selectivity among anions. Among halides, Br - and F - conductances were only 15% of that of Cl -, and I - conductance was immeasurably small. SCN - and OCN - conductances were 50%, and aspartate, glutamate, and methanesulfonate conductance was 5% that of Cl -. No conductance could be measured for any other anion tested, including NO 3 -, HCO 3 -, formate, acetate, or isethionate; NO 3 - and I - appeared to block the channels weakly. Conductances were diminished by lowering the extracellular pH to 6.4. The properties of the conductance fit best with those of the cloned renal anion channel ClC-K2 and likely reflect the basolateral Cl - conductances of the cells of these nephron segments.

【关键词】  ClCK anion selectivity basolateral membrane intercalated cells

CL - CHANNELS IN THE DISTAL nephron serve at least two important functions. In cells such as those of the thick ascending limb of Henle's loop (TALH) and the distal convoluted tubule (DCT), which reabsorb NaCl from the urine via cotransporters, Cl - channels facilitate transepithelial transport by forming a pathway for the exit of Cl - across the basolateral membrane ( 31, 33 ). In acid-secreting cells of the distal nephron, anion channels help to recycle the Cl - that enters the cells in exchange for HCO 3 - ( 14 ). In both of these cases, however, KCl cotransporters operating in parallel with the channels offer an alternative pathway for Cl - movement. Although basolateral Cl - channels have been identified at the single-channel level ( 15, 24, 29, 31 ), quantitative measurements of cellular Cl - conductance are lacking. It is therefore difficult to assess the precise role of these channels in the transport process.

The molecular identity of these renal channels is also uncertain. Genetic evidence points to the importance of the ClC-Kb/barttin complex in the reabsorption of salt by the TALH. Defects in these proteins lead to Bartter's syndrome (types III and IV, respectively), characterized by excess loss of salt in the urine and extracellular volume contraction ( 18, 19 ). Immunocytochemical data suggest that the acid- and base-transporting intercalated cells (ICs) of the collecting duct and connecting tubule also express ClC-K channels. However, there is no clear phenotype in Bartter's syndrome that is attributable to the lack of such channels in these cells ( 18 ).

In this study, we use whole cell recording techniques to characterize and quantify a Cl - conductance in principal (PCs) and ICs of the rat connecting tubule (CNT) and cortical collecting duct (CCD), and in cells of the TALH. We find that all the cells that we studied express a conductance that varies in magnitude but that consistently exhibits a remarkable selectivity for Cl - over other anions. The ClC-K2 protein may underlie this conductance.

METHODS

Animals. Sprague-Dawley rats of either sex (100-150 g) raised free of viral infections (Charles River Laboratories, Kingston, NY) were fed with either standard rat chow or a high-K diet containing 10% KCl (Harlan-Teklad, Madison, WI). A third group of animals was fed normal chow and were implanted subcutaneously with osmotic minipumps (model 2002 Alza, Palo Alto, CA) to increase levels of circulating aldosterone. Aldosterone was dissolved in polyethylene glycol 300 at a concentration of 2 mg/ml to give a calculated infusion rate of 24 µg/day. Animals were killed using the volatile anesthetic isoflurane, and one kidney was removed for tubule dissection. All animal-handling procedures were approved by the Institutional Animal Care and Use Committee of Weill Medical College.

CCD and CNT segments were identified and isolated as described previously ( 9, 27 ). PCs and ICs were distinguished visually. PCs were flat and polygonal, whereas ICs were more rounded and raised. TALH segments were isolated from medullary rays of the cortex and were readily distinguishable from CCDs and proximal tubules.

Electrophysiology. After dissection, tubules were opened manually with a very fine needle and forceps to expose the luminal surface. The split tubules were attached to a small plastic rectangle coated with Cell-Tak (Collaborative Research, Bedford, MA) and placed in a perfusion chamber mounted on an inverted microscope. The perfusate was prewarmed to 37°C.

Tubules were superfused with solutions containing (in mM) 135 Na +, 5 K +, 2 Ca 2+, 1 Mg 2+, 5 Ba 2+, 2 glucose, and 10 HEPES. The anions in the medium were varied as described in the text, and the solution was adjusted to pH 7.4 with NaOH. In experiments where F - was used in the perfusate, Ca 2+ was omitted from all solutions. When tubules were taken from animals on low-Na + or high-K + diets, or infused with aldosterone, amiloride (10 -5 M) was added to the bath to block currents through apical Na + channels. The patch-clamp pipettes were filled with solutions containing (in mM) 10 or 2 KCl, 113 or 131 aspartic acid, 20 CsOH, 20 TEAOH, 5 EGTA, 10 HEPES, and 3 MgATP, with the pH adjusted to 7.4 with KOH. The final concentration of K + was 120 mM. Most K + currents are blocked by Ba 2+ in the superfusate ( 12 ). Outward K currents at large positive voltages that remain in the presence of Ba 2+ are apparently reduced by internal Cs + and TEA, as indicated by very small outward currents measured in the absence of a permeant anion in the bath solution. In some measurements, a high-Cl - solution was used in which the aspartic acid was replaced with 131 mM HCl. Amiloride-sensitive currents were measured as the difference in current with and without 10 µM amiloride in the bath solution. Electrical contact with the solutions was made with Ag/AgCl wires in the micropipette and in a larger pipette, which was filled with agarose dissolved in 1 M Na acetate and was backfilled with the pipette solution.

Pipettes were made from hematocrit tubing, pulled in a three-step process, coated with Sylgard, and fire polished with a microforge. Pipette resistances ranged from 2 to 5 M. After formation of a G seal on the apical membrane, whole cell voltage clamp was achieved by breaking the membrane patch with suction. Current traces and current-voltage ( I - V ) plots indicate raw data. Capacitative transients measured under cell-attached conditions were compensated electronically. Cell capacitance transients were not compensated. Currents were recorded using a List EPC7 amplifier and digitized through a Digidata 1322A interface (Axon Instruments, Union City, CA). Voltage-clamp pulse protocols and data acquisition were controlled using pCLAMP 8 software (Axon Instruments). Holding potentials varied as described in the text. Voltages were pulsed for 10 ms to values from -100 to +100 mV in 10-mV increments. The voltage was returned to the holding potential for 140 ms between pulses.

Two types of corrections were made to the raw data. Conductance measurements (e.g., see Fig. 10 ) were corrected a posteriori for series resistance, assuming values equal to the pipette resistance. Reversal potential measurements (e.g., see Fig. 11 B ) were corrected for liquid junction potentials at the solution-salt bridge interface using a 3 M KCl solution as a reference. These were 6 mV.

Fig. 1. Cl - currents in a principal cell (PC) of the cortical collecting duct (CCD). A : currents obtained under whole cell clamp conditions. Pipette Cl - was 2 mM. Bath solution contained 20 mM Cl - +10 µM amiloride ( top ), 0 mM Cl - ( middle ), or 0 mM Cl - +10 µM amiloride ( bottom ). Methanesulfonate was the major extracellular anion and was used to replace Cl -. Cell voltages were held at 0 mV and pulsed to ±100 mV in 10-mV increments for 10 ms. B : steady-state current-voltage ( I - V ) relationships for the traces in A. Currents were measured between 9 and 10 ms after the start of each voltage pulse.

Fig. 2. Cl - currents in an intercalated cell (IC) of the CCD. A : currents obtained under conditions identical to those of Fig. 1. Currents after the addition of amiloride are not shown. B : steady-state I - V relationships for the traces in A. Currents in the presence of amiloride overlay those in the absence of the blocker.

Fig. 3. Replacement of Na + and Cl - with impermeant ions. I - V relationships were determined for a PC of the CCD in the presence of 150 mM NaCl in the external bath with 2 mM Cl solution in the pipette. The holding potential was 0 mV. Next, Na + in the bath was replaced entirely by N -methyl- D -glucamine (NMDG + ), a presumably impermeant cation, which did not alter the I - V relationship significantly. Cl - in the bath was subsequently replaced entirely by gluconate (Gluc), a presumably impermeant anion that reduced both inward and outward conductances.

Fig. 4. Effect of holding potential ( V H ) on reversal potentials and currents. A : bath solution contained 150 mM Cl -, and pipette solution contained 30 mM Cl -. Recording is from a PC of the CCD. Initially, the V H was -40 mV, close to the equilibrium potential (mV). Under these conditions, the reversal potential was -38 mV, and the outward current at V = 0 was 1,330 pA. The V H was then switched to 0 mV. B : time course of the current response over a 7-s interval. After the capacitative transient decayed, the current declined to a new steady state that could be described by a double-exponential relaxation process: I ( t ) = A 1 exp(- t / 1 ) + A 2 exp(- t / 2 ) + B. The best fit is shown as the white line through the data points and had parameters A 1 = 308 pA; 1 = 0.35 s; A 2 = 487 pA; 2 = 1.60 s; B = 447 pA. A new I - V relationship was then generated using 0 mV as the V H. The reversal potential was -11 mV, and the inward current at 0 mV was reduced to 450 pA.

Fig. 5. Cl - dependence of currents. A : currents obtained from a PC of the CCD. Pipette Cl - was 2 mM. Bath Cl - (Cl o ) was 2, 6.7, 20, and 67 mM, varied by substituting gluconate for Cl -. Holding potentials were 0, -30, -60, and -90 mV, respectively. B : steady-state I - V relationships for the traces in A. C : conductance as a function of Cl -. Slope conductance was measured for outward currents between 0 and 40 mV. For each cell, the conductance was normalized to that measured with 20 mM external Cl -, which was used in each experiment. The solid line represents a hyperbola fitted to the data points with a base ( y -intercept) of 0.18, a maximum of 2.24, and a Cl - concentration ([Cl - ]) for half-maximal activation of 20 mM.

Fig. 6. Halide selectivity of anion conductance. A : currents obtained from a PC of the CCD. Pipette Cl - was 2 mM. The bath contained 140 mM gluconate or 120 mM gluconate+20 mM Cl -, Br -, F - (not shown), or I -. The V H was -60 mV. B : steady-state I - V relationships for the traces in A.

Fig. 7. Selectivity of anion conductance. Steady-state I - V relationships for a PC of the CCD. Pipette contained 2 mM Cl -. The bath contained 140 mM gluconate or 40 mM gluconate+100 mM Cl -, SCN -, or isethionate -. V H was -60 mV.

Fig. 8. Selectivity of anion conductance. Steady-state I - V relationships for a PC of the CCD. Pipette contained 2 mM Cl -. The bath contained 140 mM Cl, 140 mM aspartate, 140 mM glutamate, or 140 mM gluconate.

Fig. 9. Selectivity of anion conductance. A : steady-state I - V relationships for a PC of the CCD. Pipette contained 2 mM Cl -. The bath contained 140 mM MeSO, 120 mM MeSO +20 mM Cl -, 140 mM gluconate, or 120 mM gluconate+20 mM NO 3 -. V H was -60 mV. B : steady-state I - V relationships for a PC of the CNT. Pipette contained 2 mM Cl -. The bath contained 140 mM MeSO, 120 mM MeSO +20 mM NO 3 -, 120 mM MeSO +20 mM Cl -, or 100 MeSO +20 mM Cl - +20 mM NO 3 -. V H was -60 mV.

Fig. 10. Selectivity of anion conductance. Outward conductances were measured in the presence of various anions in the bath as shown in Figs. 4 - 6. For each measurement, the conductance in the presence of gluconate was subtracted, and the anion-specific conductance expressed as a ratio of that measured with Cl - in the bath. Values are means ± SE for 4 or more cells. ISE, isethionate.

Fig. 11. Permeability vs. conductance ratios. Pipette solutions contained 161 mM Cl -. Bath solutions contained 161 mM Cl - or solutions in which 140 mM Cl - was replaced with another anion. The steady-state 0 current potential was determined and then used to set the V H. A : representative I - V relationships with Cl -, Br -, I -, and gluconate. B : measured changes in 0 current potential relative to that with 161 mM Cl - in the bath are plotted vs. the conductance ratio as shown in Fig. 10. Values are means ± SE for 4 or more experiments. The solid line represents the theoretical relationship derived from the GHK equation assuming that the permeability ratio equals the conductance ratio (see text).

Cell capacitance was measured from the rapid decay of current at the start of the voltage pulse. To maximize the amplitude of the transient relative to the steady-state current and to minimize effects of the slower current decays often observed in the presence of Cl -, we chose outward current traces measured in the absence of a conducted anion. In most cells studied, currents between 0.5 and 2 ms could usually be described by a single exponential function and were used to estimate capacitance as described previously ( 11 ). Capacitance values for PCs of the CCD ( Table 1 ) were similar to those reported previously ( 28 ) under similar conditions. Those of the other cell types were also in the same range. We presume that the cells are not electrically coupled to neighboring cells ( 11 ), or if they are, that the coupling is weak and similar in different segments.

Table 1. Capacitance in distal nephron cells

RESULTS

Figures 1 and 2 illustrate currents from a PC and an IC, respectively, of the CCD from an aldosterone-treated rat. We showed previously that the cells of the CCD were not coupled by gap junctions ( 11 ), so that these measurements are from individual epithelial cells. In both cases, whole cell clamps were formed with pipettes containing 2 mM Cl - and 131 mM aspartate. The conductances measured under these conditions were observed immediately on breaking of the membrane patch with suction. The potential was held at 0 mV and changed from -100 to +100 mV in 10-mV increments for 10-ms intervals. In the PC ( Fig. 1 ), the bath initially contained 20 mM Cl - and 120 mM methanesulfonate. Both inward and outward conductances were large. The bath solution was then switched to one in which Cl - was completely replaced by methanesulfonate. Removal of Cl - reduced both outward and inward currents. The reduction in outward current is attributable directly to the removal of permeant anions from the outside of the cell. Although we cannot rule out more complex mechanisms, the large inward currents in the presence of 20 Cl - may simply reflect accumulation of the anions in an unstirred layer inside the cell membrane during the holding potential intervals. This could also explain the finding that the reversal potential did not change as much as expected for removal of all permeant anions. Finally, 10 µM amiloride was added to the methanesulfonate bath solution. Amiloride reduced inward currents consistent with the presence of apical Na + channels expected in these cells.

Figure 2 shows a similar protocol applied to an IC, identified visually from its raised appearance. The currents in the presence of Cl - were substantially larger than those in the PCs. They were also evident immediately on formation of the whole cell clamp and were markedly reduced by replacement of Cl - with methanesulfonate. In ICs, the current transients observed after the change in the voltage were larger in magnitude and longer lived; steady-state currents were not achieved within the 10-ms pulses and could not be accounted for by cell capacitance. Indeed, when 50-ms pulses were used the currents continued to decline (not shown). This behavior is also consistent with the presence of unstirred layers in which ions may accumulate or be depleted during the pulse, altering the transmembrane concentration gradients. These transients were not confined to ICs but were also observed in PCs, particularly in the CNT, that expressed large Cl - currents. Finally, amiloride had no measurable effect on the currents even after replacement of Cl - with methanesulfonate. This was expected, since ICs are thought not to express apical Na + channels.

Further evidence that these currents represent Cl - conductances is presented in Fig. 3. Currents were measured as described above in a PC that had a relatively high conductance under these conditions. Replacement of all the Na + in the bath with the large, presumably impermeant cation N -methyl- D -glucamine had no detectable effect on the conductance. However, replacement of all Cl - in the bath with a presumably impermeant anion, in this case gluconate, virtually abolished both inward and outward conductances, as in Figs. 1 and 2. Similar results were obtained in ICs (not shown). This strongly suggests that the reduction of inward conductance under the latter condition reflects a decrease in the outward movement of anions. The most likely anion is Cl - that enters the cells from the bath and is trapped in an unstirred layer.

Figure 4 further illustrates this phenomenon under more physiological conditions in a PC. Here, the pipette solution contained 30 mM Cl -, whereas the bath contained 150 mM Cl -. When the holding potential was maintained at -40 mV, approximately equal to the Cl - equilibrium potential ( E Cl ), such that no net Cl - movement would be expected, the holding current was close to 0 and the reversal potential was therefore close to -40 mV. However, when the holding potential was depolarized to 0 mV the reversal potential was much less negative, about -10 mV, and the outward current at 0 mV was reduced. The time course of this reduction is shown in Fig. 4 B. The decay to the new steady-state value is quasi-exponential but requires two time constants for a precise fit. The simplest explanation is that when the membrane potential was zero, net movement of Cl - into the cell results in the accumulation of Cl - in a restricted space beneath the cell membrane and/or a depletion of Cl - from a restricted space just outside the membrane. The elaborate system of infoldings in the basolateral membrane of these cells ( 5 ) might provide sites for ion accumulation and depletion. A more quantitative analysis is given in the APPENDIX. These effects are reminiscent of gradual transepithelial voltage changes across the isolated perfused rabbit CCD under current-clamp conditions ( 16 ). These changes, termed "polarization potentials," were also thought to represent redistribution of ions within the tissue, although neither the ions nor the physical compartments involved were identified.

The effects of changing external Cl - concentrations are illustrated more quantitatively in Fig. 5. In these experiments, gluconate was used instead of methanesulfonate to replace Cl - since methanesulfonate turned out to have a measurable conductance. Again, nominal internal Cl - was 2 mM, and external Cl - was increased from 2 to 67 mM. In this case, the holding potential in between pulses was changed to the Cl - equilibrium potential at each external Cl - concentration. The reversal potentials of the resulting I - V plots were close to E Cl. As in Fig. 4, this is due, in part, to setting the holding potential near E Cl to prevent the net movement of the anion across the membrane.

Outward conductance was a monotonically increasing function of external Cl -. At lower Cl - concentrations, the currents tended to saturate at large positive potentials. It is not clear to what extent this represents a true rectification of the conduction process at low concentrations or ion depletion processes. The results of these and other experiments in which Cl - concentration was as high as 140 mM are shown in Fig. 5 C. The data are described reasonably well by a hyperbola with a half-maximal conductance value at 20 mM Cl -. We do not attribute any theoretical significance to the systematic deviations of the data at high and low concentrations.

This conductance displayed an unusually strong selectivity for Cl - over other anions. Figure 6 indicates the selectivity among halides with respect to conductance. For these experiments, we used a low pipette Cl - concentration and a negative holding potential (-60 mV) to minimize the Cl - concentration on the cytoplasmic side of the membrane. This reduced currents due to the outward flow of Cl -. Replacement of Cl - with Br - decreased the conductance substantially. Similar currents were observed in the presence of F - ( Fig. 6 B ). Replacement with I - reduced the conductance further to levels indistinguishable from those measured with gluconate as the major anion in the bath. Our interpretation is that neither gluconate nor I - is conducted through these anion channels. In these measurements, Ca 2+ was removed from the bathing solution to prevent precipitation with F -. In similar experiments with Ca 2+ present, a similar selectivity for Cl - over Br - and I - was observed.

Both SCN - and OCN - were conducted reasonably well. I - V relationships with Cl - and SCN - in the bath are shown in Fig. 7. The conductance of SCN - had a stronger voltage dependence than that of Cl -, increasing as the intracellular voltage became more positive. OCN - conductance was similar (not shown). However, isethionate, like I - in Fig. 6, was similar to gluconate and presumed to be nonconducted.

The anionic amino acid aspartate also showed a small conductance under these conditions. As shown in Fig. 8, replacement of bath Cl - with aspartate reduced outward and inward conductance, but not to the same extent as did replacement with gluconate. Similar results were observed with glutamate as the major anion in the bath. Because aspartate was the major intracellular anion in most of the experiments described here, there will be a small basal conductance attributable to aspartate under most conditions.

The only other ion to have a measurable conductance was methanesulfonate. As shown in Figs. 1 and 2, replacement of Cl - with this anion strongly reduced currents, but as illustrated in Fig. 9 A, replacement with gluconate lowered them further, indicating that methanesulfonate is conducted better than gluconate. Nitrate ions were not conducted and even appeared to block these channels. NO 3 - had no effect on currents when gluconate was the major anion. NO 3 - reduced Cl - currents to a slight extent ( Fig. 9 B ). However, NO 3 - blocked currents in the presence of methanesulfonate to a proportionately greater degree to levels indistinguishable from those seen with gluconate replacement. Although inhibition constants were not directly measured, NO 3 - seems to be a stronger blocker of methanesulfonate currents, perhaps because these ions interact competitively with Cl -.

HCO 3 -, formate, and acetate also failed to increase conductance beyond that observed with gluconate. The simplest interpretation is that these ions are not conducted to a measurable extent and that the currents seen in their presence represent nonspecific pathways, including that of the membrane seal with the pipette. The selectivity assessed by conductance ratios is summarized in Fig. 10. Conductances were measured at voltages between +60 and +100 mV and corrected for leak conductance using currents measured in the presence of gluconate, as well as for series resistance. The overall selectivity sequence with respect to conductance is Cl - SCN - OCN - Br - F - aspartate glutamate methanesulfonate isethionate, nitrate, formate, acetate, HCO 3 -, gluconate.

Although we did not measure conductances to all of the above ions in ICs, those that were tested had selectivity ratios indistinguishable from those of PCs. In particular, the sequence and conductance ratios for halide ions were Cl - ( 1 Br - (0.16) I - (0.003). The conductance of I - was not distinguishable from 0. Thus the characteristic pattern of high selectivity for Cl - over Br -, and almost absolute selectivity over I -, is shared by PCs and ICs. In addition, the conductance ratios for SCN - /Cl - and NO 3 - /Cl - were similar ( Table 2 ). Thus the predominant Cl - channels in these two cell types may be identical, although this cannot be shown conclusively with these macroscopic recordings. Selectivity among halides was also assessed in cells of the TALH. We found this segment to be more difficult to study using whole cell clamp techniques, due to a lower rate of G seal formation. Nevertheless, in five satisfactory recordings the conductance ratios of Br - and I - over Cl - were indistinguishable from those of the CNT and CCD; G Br / G Cl was slightly but not significantly higher than that in the CCD, and I - conductance was undetectably small. This suggests that the major basolateral Cl - channel type may be the same in all of these cells.

Table 2. Conductance relative to Cl - in distal nephron cells

Selectivity can also be assessed as permeability ratios by measuring reversal potentials under bi-ionic conditions. As described above, reversal potentials appear to be strongly affected by accumulation and depletion of ions in the fluid layers next to the membrane, especially when Cl - concentrations are low. To partially overcome this difficulty, we increased intracellular and extracellular Cl - to 161 mM. We then measured the change in reversal potential after reducing Cl - to 21 mM by replacement with various anions. The high Cl - concentrations meant that Cl - fluxes across the membrane produced smaller percent changes in concentrations. Theoretically, if the replacement anion was completely impermeable the shift in reversal potential would be ( RT / F )ln(161/21) = 54 mV. I - V relationships for Cl -, Br -, I - and gluconate are shown in Fig. 11 A. The reversal potential was shifted by 30 mV in the Br - solution and by 45 mV in the I - and gluconate solutions. In addition, the outward conductance was decreased more in the Br - and I - solutions than in the gluconate solutions. Assuming that gluconate is essentially an inert anion, i.e., it does not permeate or block the channels very well, this finding suggests that both Br - and I - can block the Cl - conductance. A summary of results for these and similar experiments is shown in Fig. 11 B. The three nonconducted ions tested (I -, gluconate, and NO 3 - ) shifted the reversal potential by 40 mV, indicating permeabilities relative to Cl - of 10% or less. Smaller shifts were observed with F - and Br -, and the smallest shift was seen with SCN - replacement. Quantitation of permeability ratios measured in this way is probably not very meaningful due to unstirred layer effects. However, the permeability sequence is at least qualitatively the same as the conductance sequence.

Some Cl - channels are sensitive to the extracellular pH. To test this point, we measured conductance in the presence of Cl - while either acidifying the bath to pH 6.4 or alkalizing to pH 8.4. Non-Cl - conductances were measured in the presence of gluconate and subtracted to give the Cl - -specific conductance ( G Cl ). As shown in Fig. 12, acidification reduced G Cl by 90%. Alkalinization to pH 8.4 had no measurable effect.

Fig. 12. Effect of pH on Cl - conductance. Outward conductance was measured in PCs from the CCD with 2 mM Cl - in the pipette and 100 mM Cl - in the bath at pH 6.4, 7.4, and 8.4. Values were corrected for nonspecific conductance measured with gluconate replacing Cl - and normalized to values obtained at pH 7.4. V H potential was -60 mV. Values are means ± SE for 5 cells.

Figure 13 summarizes the magnitude of G Cl in different cells of the kidney and under different physiological conditions. Because of significant variability in this magnitude, we compared G Cl in PCs of the CCD and the CNT of the same animals. As seen on the left side of the figure, G Cl was about fivefold higher in the CNT cells. This confirms our previous report, where we first identified the high Cl - conductance in this segment ( 9 ). In a separate set of experiments, we measured Cl - conductance in cells of the TALH. These were also quite large, although the number of cells studied was smaller for technical reasons as described above. The middle part of the figure compares PCs and ICs from the same CCDs. G Cl was considerably larger in the ICs, a finding that may reflect the physiological role of these channels (see DISCUSSION ). Finally, we compared G Cl in CCD PCs taken from animals in different physiological states. Increasing dietary KCl upregulates apical Na + and K + channels ( 10, 26, 49 ) but had no clear effect on G Cl. Chronic infusion of aldosterone stimulates Na + channels and also enhanced G Cl by about twofold.

Fig. 13. Cl - conductances in different cells under different conditions. Data are grouped together from the same animals (PCs of the CCD and CNT; PCs and ICs from the CCD) or from the same batches of animals (control, high-KCl diet, aldosterone-treated). Values are means ± SE for 14-38 cells.

As a further test for the relationship between Na + and Cl - channels, we looked for correlations between amiloride-sensitive Na + currents and Cl - -specific conductance in PCs of the CCD from rats treated with a high-K + diet and aldosterone. The results are shown in Fig. 14. Although there was considerable variability in both parameters, there was no evident correlation between the two.

Fig. 14. Cl - conductance vs. Na + currents in PCs from the CCD. Pipette solution contained 2 mM Cl -. For each cell, the inward Na + current was measured as the change in current with 10 µM amiloride at -100 mV, measured with gluconate as the major extracellular anion. The Cl - conductance was then measured as the increase in outward conductance after replacement of either 20 or 140 mM gluconate with Cl -. Conductances with 20 mM Cl - were scaled up by a factor of 2.2 to match the expected currents with 140 mM Cl - (see Fig. 5 C ). Each point represents a single cell from a rat on a high-K diet or infused with aldosterone.

DISCUSSION

Epithelial cell Cl - conductances. Cl - conductances were large, 100-300 nS/cell, in several of the cell types examined, including TALH cells, PCs of the CNT, and ICs of the CCD. Most of this is presumably across the basolateral membrane, since no Cl - channel activities in the apical membranes are expected, nor have they been observed in our laboratory. These conductances should have significant impacts on the electrical properties of the cells; in the case of the ICs, the Cl - conductance may be dominant and is likely to be critical in determining cell electrical potentials, as no other major conductances have been reported.

The high Cl - conductance of the basolateral membrane of TAL cells presumably permits the rapid reabsorption of NaCl in this segment, as genetic studies have clearly implicated both ClC-Kb and its regulatory protein Barttin in salt transport ( 18 ). Net Cl - fluxes in rat TALH are 6 peq·mm -1 ·s -1 ( 43, 50 ). For an epithelium with 500 cells/mm, this corresponds to 1,000 pA/cell of Cl - current. For the measured Cl - conductance of 300 nS/cell, this current would require a driving force of only 3 mV. Thus these channels could carry most of the reabsorptive flux of Cl - in this segment.

The most obvious physiological role for these channels in the CCD and CNT is in the ICs, which are believed to mediate acid/base transport in the distal nephron. When these cells secrete protons, CO 2 is hydrated in the cytoplasm to form H + and HCO 3 -. The H + is actively transported into the urine by an ATP-driven pump, while the HCO 3 - leaves the cell across the basolateral membrane in exchange for Cl - ( 14, 18, 38 ). The Cl - conductance would then serve to recycle the Cl - that entered the cell, although a KCl cotransporter is also thought to contribute to this efflux ( 18 ). When the cells secrete HCO 3 -, the positions of the H + pump and HCO 3 /Cl exchanger are thought to be reversed ( 14 ). In this case, assuming that the channels remain in the basolateral membrane, they would serve to reabsorb Cl -.

HCO 3 - reabsorption in the isolated, perfused rat CCD under acid-loaded conditions was 1.2 peq·cm -1 ·s -1 ( 3 ). Again, assuming 500 cells/mm, one-third of which are ICs, this gives 7.2 x 10 -4 peq·s -1 ·IC -1. According to the model above, Cl - will need to recycle out of the cell at a rate equal to that of H + secretion or HCO 3 - reabsorption, giving a net outward Cl - current of 70 pA/cell. For a conductance of 200 nS/cell, this would require a driving force of only 0.3 mV. Thus this conductance would be sufficient to permit Cl - recycling even if Cl - ions were close to electrochemical equilibrium across the membrane.

The existence of Cl - conductance in the PCs of the CCD was somewhat surprising based on electrophysiological studies of the isolated, perfused rat CCD. Measuring basolateral membrane potentials, Schlatter and Schafer ( 37 ) found that the voltage was insensitive to changes in extracellular Cl - concentration, implying a transference number of the basolateral membrane for Cl - of near 0. We do not have a complete explanation for this discrepancy. However, many of the PCs that we have studied do indeed have a very low whole cell Cl - conductance. In addition, the mean conductance for PCs (20-30 nS) was considerably lower than the K + conductance (70-140 nS) ( 12 ). It is also worth noting that in the rabbit CCD the PCs have a high basolateral Cl - conductance as measured by transference number using intracellular microelectrodes ( 36 ).

The role of the channels in the PCs is more obscure. One study of isolated, perfused CCDs indicated that bradykinin inhibited a significant fraction of net Na + reabsorption without affecting transepithelial voltage or K + secretion ( 47 ). This could be explained by electroneutral NaCl uptake across the apical membrane, in which case a basolateral Cl - conductance could contribute to salt reabsorption in this segment, analogous to the situation in the TALH. In rabbit CCD, however, blocking Na + channels with amiloride completely inhibited net Na + reabsorption ( 42 ). Thus there is no compelling evidence thus far linking the Cl - conductance with either Na + or K + transport. It is possible that it contributes to cell volume regulation by controlling the Cl - content of the cells. However, we do not have any evidence that the conductance is modulated in response to changes in the volume of the cell. The importance of this Cl - channel in PCs therefore remains a mystery.

Selectivity. The Cl - channels in the distal nephron have an unusual selectivity pattern. The most striking aspect is the very strong selectivity for Cl - even with respect to other halide ions, particularly I -, which does not appear to be conducted at all. The conductance in I - solutions was indistinguishable from that in gluconate, and clearly even smaller than in methanesulfonate, whose conductance was 5% of that of Cl -. In many other Cl - channels, selectivity among halide anions is modest. These include GABA- and glycine-activated channels ( 4 ), CFTR ( 40, 45 ), bestrophin 1 ( 30 ), and the outwardly rectifying Cl - channel found in some secretory epithelia ( 13 ).

The ClC family of Cl - channels is one group that can display strong selectivity among halides. ( Table 3; see also Ref. 7 ). They generally show the basic Cl - Br - I - sequence that we observed in the renal cells. However, in most of these (ClC-2, -3, -4, -5, and -K1), Br - is conducted nearly as well as Cl - and I - is conducted to at least some extent ( 6, 8, 20, 46 ). In ClC-1 channels, Br - conductance is low, 15% that of Cl -, and I - conductance is undetectable ( 34 ), but the permeability of I - is relatively large, 20-30% that of Cl -, and NO 3 - has a small but measurable conductance ( 34 ). In ClC-Ka, I - conductance is low but both Br - and NO 3 - conductances are high ( 6 ). The best match for the characteristics of the conductances we observed in the distal nephron is ClC-Kb ( 6 ), although for this clone the currents measured in exogenous expression systems were small.

Table 3. Conductance and permeability relative to Cl - in ClC family members

Molecular identity. ClC-K1 and -K2 are members of the ClC family that are expressed rather specifically in the kidney and the inner ear ( 22 ). Loss-of-function mutations in ClC-Kb, the human counterpart of ClC-K2, lead to a form of Bartter's syndrome, presumably due to decreased salt reabsorption in the TALH ( 39 ). Thus ClC-K2 is likely to form a major anion conductance in the basolateral membrane of these cells. Immunocytochemical evidence supports this conclusion. Studies using antibodies that recognize both ClC-K1 and ClC-K2 revealed a broad distribution of these channels in basolateral membranes of cells in the loop of Henle and the distal nephron ( 6, 48 ). Using mice in which the ClC-K1 gene had been disrupted, Kobayashi et al. ( 23 ) localized ClC-K2 to the basolateral membranes of the TALH, DCT, CNT, and the -type ICs of the collecting duct. This distribution fits that of the conductances that we observed; ICs always expressed high Cl - currents. Expression in PCs of the CCD was much lower and in many cells was undetectable.

The biophysical properties of ClC-Kb agree reasonably well with those of the conductances we observed in renal cells, although the cloned channels have been difficult to study in detail due to their low expression levels in heterologous systems. When expressed in Xenopus laevis oocytes along with barttin, ClC-Kb has a conductance sequence Cl - Br - I -, at least qualitatively similar to that in the native cells ( 6 ). Furthermore, the T481S variant of ClC-Kb, which expresses better in oocytes, has an I - conductance that is very low compared with that of Cl -, consistent with our measurements ( 17 ). These data support the idea that ClC-K2 underlies the conductances that we have studied. The ClC-Kb/barttin channels were also inhibited by lowering the extracellular pH, although unlike the native kidney conductance the cloned channels were sensitive to alkalinization as well as to acidification ( 6 ).

Some members of the ClC family function not as simple ion channels but as Cl/H exchangers ( 1, 2 ). The exchanger activity of these proteins seems to correlate perfectly with the presence of a glutamate residue at a critical position (E203 of the bacterial ClC-ec1 transporter) in the anion transport pathway ( 2 ). Because ClC-Kb has a valine at this position, as do the known channels ClC-0, -1, and -2, it is presumed but not proven to function as a pore rather than as an exchanger.

Comparison with other renal Cl - channels. Other basolateral Cl - channels in renal tubules have been characterized by patch-clamp recordings at the single-channel level. Sansom et al. ( 35 ) found channels in the basolateral membrane of the rabbit CCD that showed double-barrel behavior, appearing two at a time with a unitary conductance of 23 pS. Selectivity among anions was not assessed. Teulon and colleagues ( 24, 25 ) measured currents through single Cl - channels in the basolateral membrane of the DCT. These channels had a linear conductance of 11 pS in symmetrical Cl - solutions and a permeability sequence Cl - I - Br - NO 3 - F -; all the ions tested had measurable conductances. The channels were also proposed to be composed of ClC-K1 and/or ClC-K2 proteins, although the selectivity sequence is very different from the one we observed. This group also reported Cl - channels with a 45-pS conductance in the basolateral membrane of the TALH ( 29 ). A similar channel was observed in both basolateral membrane patches and in bilayers after fusion of basolateral membrane vesicles derived from TALH cells ( 32 ) and has been characterized in detail ( 31 ). Reeves and colleagues ( 51 ) also showed that the abundance of Cl - channels was decreased when cells were pretreated with antisense oligonucleotides against the (rabbit) ClC-Ka sequence, suggesting that this gene product is involved in formation of these channels. Paulais and Teulon ( 29 ) assessed the selectivity of this channel, which had the sequence Cl - Br - NO 3 - F -. The significant NO 3 - permeability seems to distinguish it from the conductance reported here. Finally, Hanrahan et al. ( 15 ) observed 64-pS channels in rabbit urocytes. These channels did not distinguish strongly among Cl -, Br -, and I -. One caveat in these comparisons is that all the channels observed in the studies described above were studied in tissues treated with collagenase, a procedure that may have altered the properties of the channels ( 35 ).

In conclusion, we have measured Cl - conductances in a variety of cells of the distal nephron using the whole cell patch clamp. Although it is likely that several Cl - channels contribute to conductances in the distal nephron, our results indicate that either one particular channel type or multiple types with the same highly characteristic selectivity pattern dominate this conductance at least under the conditions of our measurements. The most likely candidate for this channel is ClC-K2.

APPENDIX

Estimation of Unstirred Layer Effects

We analyzed the possible contribution of unstirred layer effects to the currents and reversal potentials observed in this study using a simplified three-compartment model as shown in Fig. A1. Compartment 1 on the left represents the stirred bath solution. Compartment 2 represents a layer beneath the cell membrane. Compartment 3 represents the bulk solution of the cell and patch-clamp pipette. We assume that Cl - is transferred between compartments 2 and 3 through restricted diffusion. For simplification, we have assumed that all of the restricted diffusion compartment is intracellular, with no restriction in the diffusion of solute up to the cell membrane from the bath solution. This is probably not realistic. We further assume that there are no electrical potential differences between compartments 2 and 3.

Fig. A1.

We use this model to analyze the experiment shown in Fig. 4. When the cell voltage is held at -40 mV, there is no driving force for Cl - to move into or out of the cell, and the concentration of Cl - in compartment 2 will be 30 mM, the same as in compartment 3. When the holding potential is depolarized to 0 mV, Cl - will enter compartment 2 and begin to accumulate there. A new steady-state will be obtained when the rate of Cl - movement from compartment 1 to compartment 2 equals the rate of movement from compartment 2 to compartment 3. These rates will be given by

We estimate k 12 = 7.0 pA/mM from the initial current after the holding potential is changed to 0 mV assuming [Cl - ] 2 = 30 mM. We then use the steady-state current at V = 0 mV to estimate [Cl - ] 2 assuming that k 12 does not change. In the case of the experiment illustrated, the steady-state value of [Cl - ] 2 is 99 mM, much higher than that of compartment 3 (30 mM). A value of k 23 = 4.2 pA/mM can then be estimated from the second equation above. The ratio of k 23 / k 12 is 0.6. Note that if the Cl - concentration gradient is diminished in part because of depletion of Cl - outside the cell, the values of both [Cl - ] 2 and k 23 / k 12 will be overestimated.

The concentration in compartment 2 will change with an exponential time course having a time constant given by where

where V 2 is the volume of compartment 2. As shown in Fig. 4, the actual time course is more complex but is well described by a double exponential decay process. Rather than constructing a more complicated four-compartment model, we estimated = 1.11 s as the weighted average of the two time constants used to fit the curve. The amplitudes of the exponential processes were used as weighting factors. This gives an estimate of V 2 = 0.19 pl. Average values of V 2 were 0.27 ± 0.08 for five PCs and 0.28 ± 0.06 for four ICs. The best available data for the volume of CCD cells were obtained for rabbit tubules using optical slicing techniques ( 44 ). They estimated volumes of 0.9 pl for PCs and 0.6 pl for ICs. In both cases, the layer of restricted Cl - diffusion that we calculate has an apparent volume equal to a substantial portion of the total cell volume. In fact, at least for ICs it is conceivable that the volume of restricted diffusion could comprise the entire cell, since the volume that we estimate will exclude organelles in which Cl - is not rapidly exchanged with the cytoplasm. This conclusion is plausible given that the conductance of the cell membrane to Cl - is 170 nS. Typically, the pipette resistance is 3-4 M giving a total conductance of 300 nS. The Cl - conductance will be at most half of this conductance, so that the pipette tip could constitute the restriction of Cl - movement assumed in our simple model. In the case of the PC, the situation may be more complicated, since the apparent restricted volume is smaller, the cell volume is larger, and the membrane Cl - conductance is much less. In this case, it is possible that large infoldings of the basolateral membrane ( 5 ) may define the unstirred compartment. A more rigorous test of this idea would require much more elaborate anatomic and numerical models of these cells and their transport properties.

GRANTS

This work was supported by National Institute of Diabetes and Digestive and Kidney Diseases Grant DK-59659.

【参考文献】
  Accardi A and Miller C. Secondary active transport mediated by a prokaryotic homologue of ClC Cl - channels. Nature 427: 803-807, 2004.

Accardi A, Walden M, Nguitragool W, Jayaram H, Williams C, and Miller C. Separate ion pathways in a Cl - /H + exchanger. J Gen Physiol 126: 563-570, 2005.

Atkins JL and Burg MB. Bicarbonate transport by isolated perfused rat collecting ducts. Am J Physiol Renal Fluid Electrolyte Physiol 249: F485-F489, 1985.

Bormann J, Hamill OP, and Sakmann B. Mechanism of anion permeation through channels gated by glycine and gamma-aminobutyric acid in mouse cultured spinal neurones. J Physiol 385: 243-286, 1987.

Brown D, Sabolic I, and Breton S. Membrane macro- and microdomains in electrolyte transporting epithelia: structure-function correlations. In: The Kidney: Physiology and Pathophysiology (3rd ed.), edited by Seldin DW and Giebisch G. Philadelphia, PA: Lippincott Williams and Wilkins, 2000, p. 655-684.

Estevez R, Boettger T, Stein V, Birkenhager R, Otto E, Hildebrandt F, and Jentsch TJ. Barttin is a Cl - channel beta-subunit crucial for renal Cl - reabsorption and inner ear K + secretion. Nature 414: 558-561, 2001.

Fahlke C. Ion permeation and selectivity in ClC-type chloride channels. Am J Physiol Renal Physiol 280: F748-F757, 2001.

Friedrich T, Breiderhoff T, and Jentsch TJ. Mutational analysis demonstrates that ClC-4 and ClC-5 directly mediate plasma membrane currents. J Biol Chem 274: 896-902, 1999.

Frindt G and Palmer LG. Na channels in the rat connecting tubule. Am J Physiol Renal Physiol 286: F669-F674, 2004.

Frindt G and Palmer LG. Short-term regulation of luminal K channels in the rat CCT by K intake. J Am Soc Nephrol 9: 34A, 1998.

Frindt G, Sackin H, and Palmer LG. Whole-cell currents in rat cortical collecting tubule: low-Na diet increases amiloride-sensitive conductance. Am J Physiol Renal Fluid Electrolyte Physiol 258: F562-F567, 1990.

Gray DA, Frindt G, Zhang YY, and Palmer LG. Basolateral K + conductance in principal cells of rat CCD. Am J Physiol Renal Physiol 288: F493-F504, 2005.

Halm DR and Frizzell RA. Anion permeation in an apical membrane chloride channel of a secretory epithelial cell. J Gen Physiol 99: 339-366, 1992.

Hamm LL and Alpern RJ. Cellular mechanisms of renal tubular acidification. In: The Kidney: Physiology and Pathophysiology, edited by Seldin DW and Giebisch G. Philadelphia, PA: Lippincott, Williams & Wilkins, 2000, p. 1935-1979.

Hanrahan JW, Alles WP, and Lewis SA. Single anion-selective channels in basolateral membrane of a mammalian tight epithelium. Proc Natl Acad Sci USA 82: 7791-7795, 1984.

Helman SI, Grantham JJ, and Burg MB. Effect of vasopressin on electrical resistance of renal cortical collecting tubules. Am J Physiol 220: 1825-1832, 1971.

Jeck N, Waldegger P, Doroszewicz J, Seyberth H, and Waldegger S. A common sequence variation of the CLCNKB gene strongly activates ClC-Kb chloride channel activity. Kidney Int 65: 190-197, 2004.

Jentsch TJ. Chloride transport in the kidney: lessons from human disease and knockout mice. J Am Soc Nephrol 16: 1549-1561, 2005.

Jentsch TJ, Neagoe I, and Scheel O. CLC chloride channels and transporters. Curr Opin Neurobiol 15: 319-325, 2005.

Kawasaki M, Suzuki M, Uchida S, Sasaki S, and Marumo F. Stable and functional expression of the CIC-3 chloride channel in somatic cell lines. Neuron 14: 1285-1291, 1995.

Kawasaki M, Uchida S, Monkawa T, Miyawaki A, Mikoshiba K, Marumo F, and Sasaki S. Cloning and expression of a protein kinase C-regulated chloride channel abundantly expressed in rat brain neuronal cells. Neuron 12: 597-604, 1994.

Kieferle S, Fong P, Bens M, Vandewalle A, and Jentsch TJ. Two highly homologous members of the ClC chloride channel family in both rat and human kidney. Proc Natl Acad Sci USA 91: 6943-6947, 1994.

Kobayashi K, Uchida S, Mizutani S, Sasaki S, and Marumo F. Intrarenal and cellular localization of CLC-K2 protein in the mouse kidney. J Am Soc Nephrol 12: 1327-1334, 2001.

Lourdel S, Paulais M, Marvao P, Nissant A, and Teulon J. A chloride channel at the basolateral membrane of the distal-convoluted tubule: a candidate ClC-K channel. J Gen Physiol 121: 287-300, 2003.

Nissant A, Lourdel S, Baillet S, Paulais M, Marvao P, Teulon J, and Imbert-Teboul M. Heterogeneous distribution of chloride channels along the distal convoluted tubule probed by single-cell RT-PCR and patch clamp. Am J Physiol Renal Physiol 287: F1233-F1243, 2004.

Palmer LG, Antonian L, and Frindt G. Regulation of apical K and Na channels and Na/K pumps in rat cortical collecting tubule by dietary K. J Gen Physiol 104: 693-710, 1994.

Palmer LG and Frindt G. Amiloride-sensitive Na channels from the apical membrane of the rat cortical collecting tubule. Proc Natl Acad Sci USA 83: 2767-2770, 1986.

Palmer LG and Frindt G. Regulation of apical membrane Na and K channels in rat renal collecting tubules by aldosterone. Semin Nephrol 12: 37-43, 1992.

Paulais M and Teulon J. cAMP-activated chloride channel in the basolateral membrane of the thick ascending limb of the mouse kidney. J Membr Biol 113: 253-260, 1990.

Qu Z and Hartzell C. Determinants of anion permeation in the second transmembrane domain of the mouse bestrophin-2 chloride channel. J Gen Physiol 124: 371-382, 2004.

Reeves WB, Winters CJ, and Andreoli TE. Chloride channels in the loop of Henle. Annu Rev Physiol 63: 631-645, 2001.

Reeves WB, Winters CJ, Filipovic DM, and Andreoli TE. Cl - channels in basolateral renal medullary vesicles. IX. Channels from mouse MTAL cell patches and medullary vesicles. Am J Physiol Renal Fluid Electrolyte Physiol 269: F621-F627, 1995.

Reilly RF and Ellison DH. Mammalian distal tubule: physiology, pathophysiology, and molecular anatomy. Physiol Rev 80: 277-313, 2000.

Rychkov GY, Pusch M, Roberts ML, Jentsch TJ, and Bretag AH. Permeation and block of the skeletal muscle chloride channel, ClC-1, by foreign anions. J Gen Physiol 111: 653-665, 1998.

Sansom SC, La BQ, and Carosi SL. Double-barrelled chloride channels of collecting duct basolateral membrane. Am J Physiol Renal Fluid Electrolyte Physiol 259: F46-F52, 1990.

Sansom SC, Weinman EJ, and O'Neil RG. Microelectrode assessment of chloride-conductive properties of cortical collecting duct. Am J Physiol Renal Fluid Electrolyte Physiol 247: F291-F302, 1984.

Schlatter E and Schafer JA. Electrophysiological studies in principal cells of rat cortical collecting tubules. Pflügers Arch 409: 81-92, 1987.

Schuster VL. Function and regulation of collecting duct intercalated cells. Annu Rev Physiol 55: 267-288, 1993.

Simon DB, Bindra RS, Mansfield TA, Nelson-Williams C, Mendonca E, Stone R, Schurman S, Nayir A, Alpay H, Bakkaloglu A, Rodriguez-Soriano J, Morales JM, Sanjad SA, Taylor CM, Pilz D, Brem A, Trachtman H, Griswold W, Richard GA, John E, and Lifton RP. Mutations in the chloride channel gene, CLCNKB, cause Bartter's syndrome type III. Nat Genet 17: 171-178, 1997.

Smith SS, Steinle ED, Meyerhoff ME, and Dawson DC. Cystic fibrosis transmembrane conductance regulator. Physical basis for lyotropic anion selectivity patterns. J Gen Physiol 114: 799-818, 1999.

Steinmeyer K, Lorenz C, Pusch M, Koch MC, and Jentsch TJ. Multimeric structure of ClC-1 chloride channel revealed by mutations in dominant myotonia congenita (Thomsen). EMBO J 13: 737-743, 1994.

Stoner LC, Burg MB, and Orloff J. Ion transport in cortical collecting tubule: effect of amiloride. Am J Physiol 227: 453-459, 1974.

Stoner LC and Trimble ME. Effects of MK-196 and furosemide on rat medullary thick ascending limbs of Henle in vitro. J Pharmacol Exp Ther 221: 715-720, 1982.

Strange K and Spring KR. Cell membrane water permeability of rabbit cortical collecting duct. J Membr Biol 96: 27-43, 1987.

Tabcharani JA, Linsdell P, and Hanrahan JW. Halide permeation in wild-type and mutant cystic fibrosis transmembrane conductance regulator chloride channels. J Gen Physiol 110: 341-354, 1997.

Thiemann A, Grunder S, Pusch M, and Jentsch TJ. A chloride channel widely expressed in epithelial and non-epithelial cells. Nature 356: 57-60, 1992.

Tomita K, Pisano JJ, and Knepper MA. Control of sodium and potassium transport in the cortical collecting tubule of the rat. Effects of bradykinin, vasopressin, and deoxycorticosterone. J Clin Invest 76: 132-136, 1985.

Vandewalle A, Cluzeaud F, Bens M, Kieferle S, Steinmeyer K, and Jentsch TJ. Localization and induction by dehydration of ClC-K chloride channels in the rat kidney. Am J Physiol Renal Physiol 272: F678-F688, 1997.

Wang W, Schwab A, and Giebisch G. Regulation of small-conductance K + channel in apical membrane of rat cortical collecting tubule. Am J Physiol Renal Fluid Electrolyte Physiol 259: F494-F502, 1990.

Work J, Galla JH, Booker BB, Schafer JA, and Luke RG. Effect of ADH on chloride reabsorption in the loop of Henle of the Brattleboro rat. Am J Physiol Renal Fluid Electrolyte Physiol 249: F698-F703, 1985.

Zimniak L, Winters CJ, Reeves WB, and Andreoli TE. Cl - channels in basolateral renal medullary vesicles. XI. rbClC-Ka cDNA encodes basolateral MTAL Cl - channels. Am J Physiol Renal Fluid Electrolyte Physiol 270: F1066-F1072, 1996.

作者单位：Department of Physiology and Biophysics, Weill Medical College of Cornell University, New York, New York