|
|
|
|
|
|
|
||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
ARTICLE |
Correspondence to Marc Paulais: Marc.Paulais{at}bhdc.jussieu.fr
 Top ABSTRACT INTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
20), its conductance ranged from 140 to 180 pS with subconductance levels, and its current/voltage relationship displayed intermediate, Na+-dependent, inward rectification. Internal Na+ and Cl– activated the channel with 50% effective concentrations (EC50) and Hill coefficients (nH) of 30 ± 1 mM and 3.9 ± 0.5 for internal Na+, and 35 ± 10 mM and 1.3 ± 0.25 for internal Cl–. Channel activity was unaltered by internal ATP (2 mM) and by internal pH, but clearly decreased when internal free Ca2+ concentration increased. This is the first demonstration of the presence in the epithelial cell membrane of a functional, Na+-activated, large-conductance K+ channel that closely resembles native KNa channels of excitable cells. This Slo2.2 type, Na+- and Cl–-activated K+ channel is primarily located in the thick ascending limb, a major renal site of transcellular NaCl reabsorption.
 |
INTRODUCTION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
), and neurons from various species (for review see Dryer, 1994). It has been suggested that KNa channels may either serve to control the resting membrane potential or be involved in the Na+-dependent slow afterhyperpolarization that follows a neuronal burst of action potentials (Dryer, 1994). They are also thought to have a protective role on guinea pig cardiomyocytes under pathological conditions, such as is chemia or hypoxia (Kameyama et al., 1984). The only report of functional KNa channels in a nonexcitable tissue of which we are aware is a very brief description of Na+-activated K+ channel activity in the plasma membrane of Xenopus oocytes (Egan et al., 1992a).
The recently cloned Slo2.1 (Bhattacharjee et al., 2003) and Slo2.2 (Joiner et al., 1998; Yuan et al., 2000, 2003) channels, two members of the Slo channels family, share 74% identity in the rat (Bhattacharjee et al., 2003) and display functional properties reminiscent of native KNa channels in excitable cells (Joiner et al., 1998; Bhattacharjee et al., 2003; Yuan et al., 2003): channel conductance in the 140–180 pS range, the existence of several conductance substates, and the fact of being stimulated by internal Na+. As well as being stimulated by internal Na+, Slo2.1 and Slo2.2 channels are both coactivated by internal Cl– (Bhattacharjee et al., 2003; Yuan et al., 2003), a yet unadressed property of native KNa channels. The molecular distributions of Slo2.1 and Slo2.2 channels correlate closely with the functional observation of KNa channels in excitable cells. Thus, mRNA and transcripts of both channels have been detected in the brain and, although to a lesser extent, in the heart of the mouse (Yuan et al., 2003) and rat (Bhattacharjee et al., 2003; Joiner et al., 1998). These studies also indicated a very limited distribution of channels among nonexcitable cells, as no Slo2.1 or Slo2.2 signal was detected in the spleen, lung, and liver of either rodent species. But, although Northern blot analysis also failed to demonstrate the presence of Slo2.1 mRNA in rat kidney (Bhattacharjee et al., 2003), significant levels of Slo2.2 mRNA and transcripts were unexpectedly detected in whole kidneys of the rat (Joiner et al., 1998) and mouse (Yuan et al., 2003). To date, no such large-conductance, Na+-sensitive K+ channels have been functionally identified in renal cells.
We therefore used real-time PCR and patch-clamp techniques on microdissected tubules of mouse nephron for molecular and functional characterization of renal KNa channels. We report here that, in addition to excitable cells and Xenopus oocytes, a Na+-activated K+ channel is present in the membranes of epithelial renal cells. Moreover, this Slo2.2-type K+ channel may have physiological significance, since it is primarily expressed in the thick ascending limb of Henle's loop, a major site of transcellular NaCl absorption.
|
MATERIALS AND METHODS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
Animal and Tissue Preparation
The experiments were conducted according to the rules of the French Ministry of Agriculture (permit no. 75–096). Male, 15–20 g, CD1 mice (Charles River Laboratories France) were fed ad libitum with a control diet (SAFE). On the day of the experiment, one mouse was killed by cervical dislocation and small pieces of renal cortex and medulla were treated with collagenase (Paulais et al., 2002).
Patch-clamp experiments were performed on the basolateral membrane of tubules microdissected as previously described (Paulais et al., 2002). For real-time PCR experiments, tubular fragments and glomeruli were microdissected at 4°C in a medium supplemented with RNase-free BSA (1 mg/ml), rinsed, and directly treated for RNA extraction (see below).
Single Channel Analysis
Current Recordings.
Single-channel currents were recorded with a patch-clamp amplifier (LM-EPC 7, List Electronic; RK-400, Bio-Logic) and stored on digital audio tapes (DTR-1205, Bio-Logic). The bath reference was 0.5 M KCl in a 4% agar bridge connected to an Ag/AgCl pellet. In the cell-attached configuration, the clamp potential (Vc = Vbath – Vpipette) is superimposed on the cell membrane potential (Vm). In excised inside-out membrane patches Vc = Vm. All Vc values were corrected for liquid junction potentials, as calculated by a routine of the AxoScope software (Axon Instruments Inc.). All experiments were conducted at room temperature.
Data Analysis.
Signals were typically low-pass filtered at 1 kHz by an 8-pole Bessel filter (LPBF-48DG; npi electronic GmbH) and digitized at 3 kHz by a Digidata 1200 A/D converter (Axon Instruments Inc.). The mean current (I) passing through N channels was used to calculate the normalized current (NPo) according to the equation NPo = I/i, where i is the unitary current amplitude.
For the analysis of channel kinetics, recordings were low-pass filtered at 3 kHz and digitized at 10 kHz. TAC and TACfit softwares (Bruxton Corp.) were used for semi-automatic event detection and to fit open and closed time distribution histograms, respectively. Open and closed time intervals covering a wide range (see RESULTS) fits were generated on plots of the square root of the number of observations against the logarithm of the time intervals (Colquhoun and Sigworth, 1995).
Unless otherwise stated, the maximal inward conductance (gin) was determined by linear regression analysis of the i/Vc relationship between Vc = –40 mV and more hyperpolarizing voltages. When appropriate, the outward chord conductance (gout) was defined as the slope of the i/Vc curve when measured between the reversal potential (Erev) and Vc = +60 mV.
Channel Selectivity.
PK/PNa was determined in inside-out membrane patches from Erev values obtained from i–Vc relationships established with opposite Na+ and K+ ion gradients across the membrane patch, using the Goldman, Hodgkin and Katz voltage equation (Hille, 1992):
 |
Solutions.
Pipettes were typically filled with a high-K+ solution containing (in mM) 145 KCl, 1 MgCl2, 10 glucose, 10 HEPES, pH 7.4/KOH. In cell-attached patches, bath medium contained 140 NaCl, 5 KCl, 1 CaCl2, 1 MgCl2, 10 glucose, 10 HEPES, pH 7.4/NaOH.
The intracellular side of inside-out patches was typically bathed by a high-Na+ medium, containing 140 NaCl, 5 KCl, 10 HEPES, 10 glucose, 2 K2-EDTA, pH 7.4/NaOH. When appropriate (see RESULTS), various free Ca2+ concentrations (under constant low free Mg2+ concentration) were obtained by replacing EDTA with appropriate amounts of CaCl2 and EGTA, as previously described (Teulon et al., 1987). In experiments where channel sensitivities to internal Na+ and Cl– concentrations were investigated (see RESULTS), LiCl, Li-acetate, or Na-acetate were substituted for equimolar amounts of NaCl. Na+-induced rectification was investigated by adding appropriate amounts of NaCl to a high-K+ solution (145 KCl, 10 HEPES, 10 glucose, 2 K2-EDTA, pH 7.4/KOH) bathing the cytoplasmic side of inside-out patches.
Real-time PCR
mRNA was extracted from pools of 20–50 microdissected tubules or glomeruli, according to the technique previously described (Elalouf et al., 1993). RT was performed using the first strand cDNA synthesis kit for RT-PCR (Roche Diagnostics).
A BLAST search for the base sequence that encodes for the highly conserved region of the chloride bowl (FMFRLPFA) of human, rat, and Caenorhabditis elegans Slo2 channels (Yuan et al., 2003) was performed in the mouse genome. This yielded two sequences. A 3396-bp sequence was identified (GenBank/EMBL/DDBJ accession no. XM_136252), which, when translated, showed 93% homology with the amino acid sequence of rat Slo2.1 (NM_198762) (Bhattacharjee et al., 2003). A 3759-bp mouse cDNA sequence was also identified (AK048406), which, when translated, showed 99% homology with the 703–1238 amino acid sequence of rat Slo2.2 (NM_021853) (Yuan et al., 2003). The primers for the mouse Slo2.1 sequence were as follows: sense 5'-TATGATCTGCGTTGCCC-3' (bases 747–762) and antisense 5'-CCGAACTTGCACATCC-3' (bases 1296–1311). The primers for the mouse Slo2.2 sequence were as follows: sense 5'-AAACACAACAGCTACGAG-3' (bases 188–205) and antisense 5'-GGGAAAAGCCGGAACA-3' (bases 555–570). The primers for the mouse cyclophilin sequence were as follows: sense 5'-ATGGCAAATGCTGGACCAAA-3' (bases 339–358) and antisense 5'-GCCTTCTTTCACCTTCCCAAA-3' (bases 426–446).
Real-time PCR was performed on a LightCycler System (Roche Diagnostics) using the FastStart DNA Master SYBR Green 1 kit (Roche Diagnostics) on an 8-µl final reaction volume as previously described (Nissant et al., 2004). In brief, measurements were performed on cDNA corresponding to either 0.1 mm of tubule or 0.2 glomerulus in the presence of either 4 pmol Slo2.2 primers/8 µl or 2 pmol Slo2.1 primers/8 µl. Samples were submitted to 50 cycles of three temperatures (Slo2.2: 95°C for 10 s, 60°C for 10 s, 72°C for 20 s; Slo2.1: 95°C for 10 s, 62°C for 10 s, 72°C for 20 s). A melting curve was then established in order to check for nonspecific PCR products. After 50 cycles, no DNA was detectable in blank runs.
In each experiment, a standard curve was plotted using serial dilutions (1 to 1/500) of a cDNA stock solution obtained from mouse whole kidney RNA, dilution 1 corresponding to the cDNA produced by 10 ng RNA. For all samples, the amount of PCR product was calculated as a percentage of the RNA standard (arbitrary unit).
Data Presentation and Statistics
Results are given as means ± SEM for n experiments. P values <0.05 (paired or unpaired t test, when appropriate; SigmaPlot, SPSS) were taken to represent statistically significant differences. Nonlinear regression analyses were performed using either the SigmaPlot or the Origin (Microcal Software, Inc.) software.
|
RESULTS |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
) was observed in 27% (23 out of 85) of the patches. The activity of a larger conductance K+ channel was also observed in 12% (10 out of 85) of the patches. Both channel types were observed on the same patch in one case. Intermediate- and large-conductance K+ channels were also observed on the basolateral membrane of MTAL tubules, both in 19% of the cell-attached patches (6 out of 32 patches, for each channel type). In contrast, no activity of a large conductance K+ channel was detected in basolateral membrane of DCT (Lourdel et al., 2002). All the data presented hereafter were obtained on patches of the basolateral membrane of CTAL tubules.
Representative current tracings of large-conductance K+ channel activity in situ and the corresponding i/Vc relationship are shown in Fig. 2. Channel conductance averaged 82 ± 6 pS at Vc = 0 mV (n = 8) and maximal inward conductance (gin; see MATERIALS AND METHODS) reached 138 ± 7 pS (n = 8). No outward current was detected in these conditions and Erev could be only estimated using linear regression analysis of three i/Vc relationships for inward currents. A value of 62 ± 8 mV was obtained, in good agreement with the activity of a K+-selective channel. This was confirmed by two measurements of ionic currents with 75 mM K+ in the patch pipette (70 mM Na+ subsituted for K+) (Fig. 2 B). Under these conditions, gin was reduced to 83 pS and Erev was shifted toward Vc 42 mV. This relative shift in Erev yielded a PK/PNa value >20 (for rationale see Paulais et al., 2002).
|
Channel Properties in the Inside-out Configuration
Excising patches into a high-NaCl and low-KCl medium led to greater channel activity. Under theses conditions, the large conductance K+ channel was observed on 31 out of 85 patches (36.5%). Intermediate- and large-conductance K+ channels were observed on the same patch in seven cases (8.2%).
Representative tracings and the corresponding i/Vc curve for the large-conductance K+ channel under these conditions are shown in Fig. 3 A. Channel conductance averaged 86 ± 2 pS at Vc = 0 mV (n = 13), and fitting the i/Vc data with the Goldman-Hodgkin-Katz current equation (Hille, 1992) yielded a gin value of 154 ± 4 pS (n = 13) and an Erev of 62 ± 2 mV (n = 13), confirming the K+ selectivity of the channel with a PK/PNa ratio of 20 ± 3 (n = 13).
|
; Wang et al., 1991, Dryer 1994) and rat Slo2.1 (Bhattacharjee et al., 2003) and Slo2.2 (Yuan et al., 2003) channels is the presence of numerous subconductance levels. The examples shown in Fig. 3 B illustrate the similar behavior of the large conductance K+ channel in the TAL. Intermediate current levels more frequently resulted from partial channel closures from the fully open state during a burst of activity, although partial openings also arose from the closed state. Within a burst, the time spent at subconductance levels accounted for 9.8 ± 1% of the total open time (n = 7).
A continuous recording on a patch containing one active channel at Vc = 0 mV is shown in Fig. 4. Typically, channel activity occurred in bursts of openings separated by closures of various duration, with no evident sign of rundown over several minutes (unpublished data). Kinetic analysis revealed that, as observed for the KNa channels of excitable cells (Kameyama et al., 1984; Veldkamp et al., 1994; Mistry et al., 1997), the time distribution of the main open state was best fitted by the sum of two exponentials, longer openings being the most frequent (Fig. 4; Table I). Fitting the closed times distribution required at least four exponentials (Table I). Closures with the longest time constant were the least frequent and displayed some variability from patch to patch. Indeed, the very long and infrequent closures seen on some recordings fell outside the fitted relationship (unpublished data).
|
|
Na+-dependent Inward Rectification
As for KNa channels in guinea pig ventricular myocytes (Wang et al., 1991) and rat olfactory bulb neurons (Egan et al., 1992b), the blockade by internal Na+ of outward currents flowing through the TAL large-conductance K+ channel induces an inward rectification. Fig. 5 A compares single channel current amplitudes observed at Vc = +60 mV and –60 mV under symmetrical 145 mM K+ conditions, in the presence and absence of Na+ on the inner face of the membrane patch. In the absence of Na+, ion currents were of similar amplitudes at both potentials. Adding 140 mM internal Na+ had little influence on the amplitude of the inward current (Vc = –60 mV), but reduced the amplitude of outward current (Vc = +60 mV) by nearly 50%. Accordingly, the i/Vc relationship in the absence of Na+ was linear over the –70 to +80 mV range, with a conductance of 177 ± 19 pS (n = 3) (Fig. 5 B). Adding 140 mM internal Na+ reduced gout to 93 ± 3 pS (n = 3; P = 0.001 vs. Na+-free), but did not affect gin (159 ± 8 pS; n = 4; P = 0.4), yielding a rectification coefficient gin/gout of 1.8.
|
). Fig. 6 shows a continuous recording of an inside-out patch initially bathed by 140 mM internal NaCl then exposed to a Na+-free medium. With high NaCl, the activity of three large conductance K+ channels was detected, yielding an NPo of 0.61. When all internal Na+ was replaced by Li+ (the internal Cl– being kept concentration to 145 mM), the K+ channel currents were virtually abolished. The NPo fell to 0.05, and there was no change in single-channel current amplitude (Fig. 6, insets). K+ channel activity was fully and rapidly restored when internal Na+ was reintroduced (NPo = 0.69). The effects of internal Na+ were concentration dependent, altering NPo in a sigmoidal fashion (Fig. 7 A). Fitting data points using a modified Hill equation (see legend) yielded an EC50 of 30 ± 1 mM Na+. A Hill coefficient nH of 3.95 ± 0.5 was found, suggesting that the cooperative action of four Na+ ions is required to bring about channel activation.
|
|
Only one patch with a single active channel was seen in this experimental series. This recording, obtained under low Cl– conditions, is shown in Fig. 7 B. It is reassuring to observe that, as previously observed for KNa channels in excitable cells (Kameyama et al., 1984; Mistry et al., 1997), a predominant effect of reducing internal Na+ was an increase in the number and duration of the intervals between bursts of activity.
The concentration dependence of the effects of internal Cl– on channel activity was then examined (Fig. 8). Under 140 mM internal Na+ concentration, varying internal Cl– concentration over the 0–145 mM range dose dependently affected channel activity, with EC50 = 35 ± 9.8 mM and nH = 1.3 ± 0.25. These results constitute the first demonstration of a native Na+-sensitive K+ channel being coactivated by internal Cl–.
|
1 nM free Ca2+) and was reduced by increasing the free Ca2+ concentration. At 1 µM Ca2+, NPo was 30% of that in Ca2+-free medium (P = 0.0007; n = 4), and channel activity was further decreased by increasing free Ca2+ concentration to 1 mM (NPo = 14 ± 3% of that in the Ca2+-free medium; n = 6). The dose–response curve for the inhibitory effect of internal Ca2+ (Fig. 9 B) displayed EC50 300 nM.
|
|
DISCUSSION |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
). First, the conductance of KNa channels ranges from 100 to 200 pS, and the occurrence of multiple substates has been frequently reported (Wang et al., 1991; Mistry et al., 1996). Second, internal Na+ induces an inward rectification of KNa channels. The intermediate rectification coefficient (gin/gout) of 2 that we observed here is close to the ratio of 3 which can be deduced from published data on KNa channels in guinea pig cardiomyocytes (Wang et al., 1991) and rat olfactory bulb neurons (Egan et al., 1992b). Third, KNa channel open probability in excitable cells is usually weakly (Haimann et al., 1990; Egan et al., 1992b; Koh et al., 1994) or not (Kameyama et al., 1984; Safronov and Vogel, 1996; Mistry et al., 1997) potential dependent, and very moderately pH sensitive within the 6.5–8 pH range in the presence of an elevated internal Na+ concentration (Veldkamp et al., 1994). Finally, KNa channel activity in guinea pig cardiomyocytes (Kameyama et al., 1984) and in rat bulb olfactory neurons (Egan et al., 1992b) is not affected by internal ATP.
The activity of the 150 pS channel in the TAL increases with internal Na+. In the presence of 145 mM internal Cl–, the half-maximum activating Na+ concentration (EC50) and Hill coefficient (nH) were 30 mM and 4, respectively. These values closely match the EC50 (30–75 mM) and nH (2.4–4.6) values reported for KNa channels of excitable membrane under similar conditions (Dryer, 1994). These properties provide functional identification of the 150 pS K+ channel of TAL basolateral membrane as a KNa channel.
Comparison with Slo2 Channels
To date, two paralogues, Slo2.1 and Slo2.2, have been shown to encode for Na+- and Cl–-activated, large-conductance K+ channel (Bhattacharjee et al, 2003; Joiner et al, 1998; Yuan et al, 2000, 2003). Using real-time PCR, Slo2.2 mRNA, but not that of Slo2.1, was detected in the TAL. This strongly suggests that Slo2.2 underlies the TAL KNa channel.
The heterologous expression of Slo2.1 and Slo2.2 yields K+ channels with very similar conductive properties (Joiner et al., 1998; Bhattacharjee et al., 2003; Yuan et al., 2000, 2003). Both genes encode for 140– 180 pS K+ channels with subconductance levels that are coactivated by internal levels of both Na+ and Cl–. Several functional observations confirm that Slo2.2 underlies the KNa channel in the TAL. First, the Na+-dependent activation of TAL KNa channel also depends on the internal Cl–, and reducing the internal Cl– from 140 to 40 mM increased the EC50 for Na+ from 30 to 74 mM (Fig. 7). These values are of the same order of magnitude as those reported for Slo2.2 channels under similar conditions (Yuan et al., 2003). Furthermore, Slo2.1 displays a significant activity in a 35 mM Cl– and Na+-free medium (Bhattacharjee et al., 2003), whereas Slo2.2 is inactive under these conditions (Bhattacharjee et al., 2003; Yuan et al., 2003). Similar ionic conditions (0 Na+ and 40 mM Cl–, Fig. 7) abolished TAL KNa channel activity. Also, the EC50 for Cl– (35 mM, Fig. 8) we observed in the presence of an elevated internal Na+ concentration is of the same order of magnitude as that reported for Slo2.2 channel (20 mM Cl–) under similar conditions (Yuan et al., 2003).
Second, Slo2.1 is inhibited by internal ATP (Bhattacharjee et al., 2003), possibly via the presence of a consensus ATP binding motif in the COOH-terminal tail of its sequence. There is no such ATP binding motif in the Slo2.2 sequences of the rat (Bhattacharjee et al., 2003) and mouse (present study). Although the possibility that ATP could inhibit rat Slo2.2 by ATP was not tested for (Bhattacharjee et al., 2003), the fact that ATP had no effect on the TAL KNa channel clearly shows that this channel is not of a Slo2.1 type.
Finally, an elevation in the cytosolic free Ca2+ concentration inhibits rat Slo2.2 (Joiner et al., 1998). Our observation that KNa channel activity in the TAL also decreased in response to an increase in the cytosolic free Ca2+ is also consistent with the presence of a Slo2.2 channel.
Physiological Role of a KNa Channel in the TAL Basolateral Membrane
The TAL transcellularly reabsorbs 15–20% of the filtered NaCl load. Basically, NaCl enters TAL cells via an apical Na+-K+-2Cl– cotransporter driven by the basolateral Na+/K+ ATPase and, on the basolateral side, Cl– exits through chloride channels (Fig. 10). Basolateral K+ channels first maintain the membrane potential difference and thus the driving force for basolateral Cl– diffusion and, second, provide a major pathway for the efflux of K+ ions entered into the cell by Na+/K+ ATPase activity. We (Paulais et al., 2002) and others (Hurst et al., 1992; Gu and Wang, 2002) have previously demonstrated the presence of an 45 pS K+ channel in this membrane. This channel is observed in 30% of cell-attached patches, with an average NPo of 1.9 and a conductance of 10 pS under physiological conditions (Paulais et al., 2002). This makes it possible to calculate a mean conductance of 6 pS per membrane patch. The KNa channel was observed here in 12% of the patches, with an average NPo of 0.6. Assuming a conductance of 45 pS under physiological conditions (Safronov and Vogel, 1996), one may calculate a conductance of 3 pS per membrane patch. Thus, the 150 pS K+ channel would account for 33% of the basolateral membrane K+ conductance. But it should be pointed out that a significant fraction of KNa channels may be inactive under resting conditions, since the channel frequency of observations increased from 12% to 35% after patch excision into a Na+-rich medium. This may indicate that this channel is in a position to more significantly contribute to the K+ conductance of the TAL basolateral membrane.
|
0.6. This observation is somewhat surprising since the Na+ and Cl– sensitivities we measured in cell-free patches indicate that the channel should be nearly inactive in situ. Thus, resting intracellular Na+ and Cl– concentrations in the TAL are 15 mM (Rajerison et al., 1986; Lu et al., 1996) and 35 mM (Greger et al., 1983), respectively. According to the Na+ dependence relationship in the presence of 40 mM Cl– in cell-free patches, channel activity is very low under these conditions. Furthermore, these later experiments were performed under Ca2+-free conditions, whereas we observed that the activity of the channel was inhibited when the cytosolic free Ca2+ concentration was increased to a value near resting free Ca2+ concentration in the mouse TAL (Paulais et al., 1996). This further indicates that the TAL KNa channel sensitivities to internal Na+ and Cl– we measured in cell-excised membrane patches are indeed underestimated, as reported for the Na+ sensitivity of KNa channel in excitable cells (Haimann et al., 1992; Rodrigo, 1993), and suggests that they may be subject to modulation. Further studies are obviously needed to identify these modulators in order to provide a more precise picture of the conditions under which it is activated, and hence its role in the reabsorption of NaCl in the TAL.
The interior of TAL cells has to cope with large fluxes of Na+ and Cl–. According to the TAL transport scheme, stimulating apical Na+ and Cl– influx is expected to produce a rapid increase in the Na+/K+ ATPase-mediated Na+ efflux, and a parallel increase in basolateral K+ conductance. Accordingly, apical and basolateral membranes must be functionally linked in order to maintain ionic reabsorption while preventing major perturbations in cellular ionic content and volume (Schultz and Lapointe, 2000). The nature of this functional linkage (or cross-talk) has been debated (Schultz and Lapointe, 2000). Numerous studies in frog skin and in amphibian and mammalian nephrons have provided evidence for the involvement of several factors, including membrane potential, cell volume, and intracellular Ca2+, pH, and ATP (Schultz and Lapointe, 2000). We therefore propose that the basolateral KNa channel may be involved in membrane cross-talk in the TAL by means of its sensitivity to intracellular Na+ and Cl– concentrations. In this case, an elevation in intracellular Na+ and Cl– concentrations would be expected to activate the KNa channel, leading to a coupling between increased Na+/K+ ATPase activity and basolateral recycling of K+.
In conclusion, the present study provides the first evidence of the presence of a KNa channel in an epithelial cell membrane. So far, KNa channels were thought to be present primarily in the membranes of excitable cells (Dryer, 1994). Our findings, together with the observation of KNa channels in Xenopus oocytes (Egan et al., 1992a), may indicate that these channels have hitherto unsuspected functional roles.
|
ACKNOWLEDGMENTS |
|---|
S. Lachheb holds a Ph.D. fellowship from the French Ministère de la Recherche.
Lawrence G. Palmer served as editor.
Submitted: 6 July 2005
Accepted: 13 January 2006
|
REFERENCES |
|---|
|
TopABSTRACTINTRODUCTIONMATERIALS AND METHODSRESULTSDISCUSSIONREFERENCES |
|---|
1-adrenergic receptor in the thick ascending limb of rat kidney. Absolute mRNA quantification by reverse transcription and polymerase chain reaction. J. Clin. Invest. 91:264–272.[Medline]
CiteULike 
Complore 
Connotea 
Del.icio.us 
Digg 
Facebook 
Reddit 
Technorati 
Twitter What's this?
This article has been cited by other articles:
|
|
 |
|
  H. Chen, J. Kronengold, Y. Yan, V.-R. Gazula, M. R. Brown, L. Ma, G. Ferreira, Y. Yang, A. Bhattacharjee, F. J. Sigworth, et al. The N-Terminal Domain of Slack Determines the Formation and Trafficking of Slick/Slack Heteromeric Sodium-Activated Potassium Channels J. Neurosci., April 29, 2009; 29(17): 5654 - 5665. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 D. Heitzmann and R. Warth Physiology and Pathophysiology of Potassium Channels in Gastrointestinal Epithelia Physiol Rev, July 1, 2008; 88(3): 1119 - 1182. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Lachheb, F. Cluzeaud, M. Bens, M. Genete, H. Hibino, S. Lourdel, Y. Kurachi, A. Vandewalle, J. Teulon, and M. Paulais Kir4.1/Kir5.1 channel forms the major K+ channel in the basolateral membrane of mouse renal collecting duct principal cells Am J Physiol Renal Physiol, June 1, 2008; 294(6): F1398 - F1407. [Abstract] [Full Text] [PDF]
|
|
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
; n = 5–8) or in the presence of 75 mM K+ in the pipette (
; n = 2). Inset, voltage dependence of channel activity in the conditions given in A. Channel activity is given as NPo values normalized to that at Vc = 0 mV on the same patch. Each point is the mean of five to eight measurements and the SEM is shown as error bars when larger than symbol.
) conditions. Each point is the average of either 2–3 (
