Kcnj10 encodes the inwardly rectifying K+ channel 4.1 (Kir4.1) and is expressed in the basolateral membrane of late thick ascending limb, distal convoluted tubule (DCT), connecting tubule (CNT), and cortical collecting duct (CCD). In the present study, we perform experiments in postneonatal day 9 Kcnj10−/− or wild-type mice to examine the role of Kir.4.1 in contributing to the basolateral K+ conductance in the CNT and CCD, and to investigate whether the disruption of Kir4.1 upregulates the expression of the epithelial Na+ channel (ENaC). Immunostaining shows that Kir4.1 is expressed in the basolateral membrane of CNT and CCD. Patch-clamp studies detect three types of K+ channels (23, 40, and 60 pS) in the basolateral membrane of late CNT and initial CCD in wild-type (WT) mice. However, only 23- and 60-pS K+ channels but not the 40-pS K+ channel were detected in Kcnj10−/− mice, suggesting that Kir.4.1 is a key component of the 40-pS K+ channel in the CNT/CCD. Moreover, the depletion of Kir.4.1 did not increase the probability of finding the 23- and 60-pS K+ channel in the CNT/CCD. We next used the perforated whole cell recording to measure the K+ reversal voltage in the CNT/CCD as an index of cell membrane potential. Under control conditions, the K+ reversal potential was −69 mV in WT mice and −61 mV in Kcnj10−/− mice, suggesting that Kir4.1 partially participates in generating membrane potential in the CNT/CCD. Western blotting and immunostaining showed that the expression of ENaCβ and ENaCγ subunits from a renal medulla section of Kcnj10−/− mice was significantly increased compared with that of WT mice. Also, the disruption of Kir4.1 increased aquaporin 2 expression. We conclude that Kir4.1 is expressed in the CNT and CCD and partially participates in generating the cell membrane potential. Also, increased ENaC expression in medullary CD of Kcnj10−/− mice is a compensatory action in response to the impaired Na+ transport in the DCT.
- inwardly rectifying K+ channel 4.1
- EAST syndrome
- Giltelman syndrome
- connecting tubule
- cortical collecting duct
- epithelial sodium ion channel
loss-of-function mutations of Kcnj10 [inwardly rectifying K+ channel 4.1 (Kir4.1)] cause EAST/SeSAME syndrome in humans (seizures, sensorineural deafness, ataxia, mental retardation, and electrolyte imbalance) (1). The renal phenotype of the disease is reminiscent to Gitelman's syndrome, including hypomagnesemia, hypokalemia, and metabolic alkalosis, suggesting that the disruption of Kir4.1 mainly impairs the transport in the distal convoluted tubule (DCT) (15, 16). This is also indicated by our previous finding that the disruption of Kir4.1 significantly decreased NCC expression (21, 27). However, Kir4.1 has been shown to be expressed not only in the DCT but also in the thick ascending limb (TAL), connecting tubule (CNT), and initial cortical collecting duct (CCD) in human kidney (10, 14, 25). Studies performed in the TAL of Kcnj10−/− mice demonstrated that the disruption of Kir4.1 did not significantly decrease the basolateral K+ conductance and had no effect on the membrane potential of the TAL (25). We further demonstrated that a Na+- and Cl−-activated K+ channel in the TAL was upregulated, thereby compensating the function of Kir4.1 (25). This compensation may explain why patients with loss-of-function mutations of Kir4.1 in the kidney do not have phenotypes of Bartter's syndrome although Kir4.1 is expressed in the TAL. Whereas NCC expression was significantly decreased in mouse models in which Kir4.1 activity was inhibited, the phenotype of Na+ wasting was relatively modest (16, 21). We speculate that disruption of Kir4.1 in the CNT and CCD may not significantly affect the electrochemical driving force for Na+ entry through epithelial Na+ channel (ENaC) in the CNT and CCD. Moreover, it is possible that ENaC activity in collecting duct may be upregulated in Kcnj10−/− mice due to either high aldosterone or vasopressin. Therefore, aims of the present study are as follows: 1) to determine whether the depletion of Kir4.1 affects K+ channel expression in the basolateral membrane of the CNT/CCD and membrane potential; and 2) to determine whether ENaC expression is upregulated in the collecting duct of Kcnj10−/− mice.
Kcnj10−/−, Kcnj10+/−, and Kcnj10+/+ mice were obtained through mating Kcnj10+/− mice, which were kindly provided by Dr. Paulo Kofuji at the University of Minnesota to Dr. R. P. Lifton (9, 12). Primers used for genotyping were the following: Kcnj10, forward 5′-TGG ACG ACC TTC ATT GAC ATG CAG TGG-3′ and reverse 5′-CTT TCA AGG GGC TGG TCT CAT CTA CCA CAT-3′ (9). The Neo forward primer is GAT TCG CAG CGC ATC GCC TTC TAT C. Kcnj10−/− mice were viable and had no obvious abnormality at birth compared with their wild-type (WT) littermates. However, their growth was stopped at postnatal days (P) 6-7, and the mice died within 2 wk (12). Thus, we carried out the experiments using P7-10 homozygous Kcnj10−/−, heterozygous Kcnj10+/−, and Kcnj10+/+ mice. The animal use protocol was approved by the independent Institutional Animal Care and Use Committees at both Yale University and New York Medical College.
Preparation of the CNT and early CCD.
After mice were killed by cervical dislocation, we perfused the left kidney with 2 ml collagenase type 2 (250 U/ml) containing L-15 medium (Life Technology). The collagenase-perfused kidney was removed, and we only cut the renal cortex with a sharp razor. The renal cortex at the top half was further cut into small pieces that were then incubated in collagenase-containing L-15 media for 30–40 min. After the collagenase treatment, the tissue was washed three times with L-15 medium and transferred to an ice-cold chamber for dissection. The method for dissecting CNT and early CCD was similar to those described previously for dissecting DCT (26). The patch-clamp experiments were performed in the late CNT and early CCD near the branch of CNT/CCD as shown in Fig. 1. The tubules were adhered to a cover glass coated with polylysine, and the cover glass was placed on a chamber mounted on an inverted microscope. The tubule was superfused with a bath solution containing (in mM) 140 NaCl, 5 KCl, 1.8 MgCl2, 1.8 CaCl2, and 10 HEPES (pH 7.4).
For the single channel recording, an Axon200B patch-clamp amplifier was used to record the channel current. Borosilicate glass (1.7 mm OD) was used to make the patch-clamp pipettes using a Narishige electrode puller. The pipette solution contained (in mM) 140 KCl, 1.8 MgCl2, and 10 HEPES (pH = 7.4). The currents were low-pass filtered at 1 kHz and digitized by an Axon interface with a sampling rate of 4 kHz. The channel open probability (Po) was calculated from the channel number (N) and NPo (a product of channel number and open probability), which was calculated from data samples of 60 s duration in the steady state as follows:
where ti is the fractional open time spent at each of the observed current levels. The channel conductance was determined by measuring the current amplitudes over several voltages.
For the measurement of K+ reversal potential with the whole cell recording, we used an Axon 200A amplifier. The tip of the pipette was filled with pipette solution containing (in mM) 140 KCl, 2 MgCl2, 1 EGTA, and 5 HEPES (pH 7.4). The pipette was then backfilled with amphotericin B (20 μg/0.1 ml) containing the pipette solution. The bath solution contained 140 mM NaCl + 5 mM KCl. After a high-resistance seal was formed (>2 GΩ), the membrane capacitance was monitored until the whole cell patch configuration was formed. The currents were low-pass filtered at 1 kHz, digitized by an Axon interface with 4-kHz sampling rate (Digidata 1440A). Data were analyzed using the pClamp software system 9.0 (Axon).
Mice were anesthetized with ketamine (100 mg/kg) and xylazine (10 mg/kg), and the abdomens were cut open for perfusion of kidneys with 2 ml PBS containing heparin (40 U/ml) followed by 20 ml of 4% paraformaldehyde. After perfusion, the kidneys were removed and subjected to postfixation with 4% paraformaldehyde for 12 h. The kidneys were dehydrated and cut in 8- to 10-μM slices with a Leica1900 cryostat (Leica). The tissue slices were dried at 42°C for 1 h. The slides were washed with 1× PBS for 15 min and permeablized with 0.4% Triton dissolved in 1× PBS buffer containing 1% BSA and 0.1% lysine (pH = 7.4) for 15 min. Kidney slices were blocked with 2% horse serum for 30 min at room temperature and then incubated with primary antibodies [Kir.4.1, aquaporin 2 (AQP2), and ENaCβ] for 12 h at room temperature. Slides were thoroughly washed with 1× PBS followed by addition of a secondary antibody mixture in 0.4% Triton dissolved in 1× PBS for 2 h at room temperature. Kir.4.1, AQP2, and ENaCβ antibody was diluted at 1:500, 1:1,000, and 1:500, respectively. The secondary antibody (Alexa Fluor 488 or 594) was diluted 1:1,000.
Preparation of protein samples and Western blot.
The tissue of renal cortex was homogenized in an ice-cold solution containing 250 mM sucrose, 50 mM Tris·HCl, 1 mM EDTA, 1 mM EGTA, 1 mM DTT, and 1% protease and phosphatase inhibitor cocktails (Sigma) titrated to pH 7.6. After homogenization, the sample was subjected to centrifugation at 2,000 revolutions/min for 15 min at 4°C, and the protein concentration was measured using the DC Protein Assay Kit (Bio-Rad, Hercules, CA). The proteins were separated by electrophoresis on 4–15% SDS-polyacrylamide gels and transferred to nitrocellulose membrane. The membranes were blocked with LI-COR blocking buffer (PBS). An Odyssey infrared imaging system (LI-COR, Lincoln, NE) was used to scan the membrane at a wavelength of 680 or 800 nM. For Western blot, ENaCα, ENaCβ, and ENaCγ antibody was diluted at 1:1,000.
Experimental materials and statistics.
We obtained polyclonal Kir4.1 antibody from Millipore (Temecula, CA) and ENaCβ antibody from Sigma-Aldrich (St. Louis, MO). ENaCα and ENaCγ antibodies were purchased from StressMarg (Victoria, Canada). AQP2 antibody was purchased from Santa Cruz (Dallas, TX). Data are presented as means ± SE. The Chi squared test, paired Student's t-test, or one-way ANOVA (Kruskal-Wallis H-test) was used determine statistical significance. Results were considered to be significant at P < 0.05.
We first carried out immunostaining experiments to examine whether Kir4.1 is expressed in the AQP2-positive tubules in P9 WT mice. Figure 2A is an immunostaining image with a low magnification demonstrating the expression of Kir4.1, and Fig. 2B is an image showing AQP2 expression in the kidney of P9 Kcnj10+/+ mice (WT). Figure 2C is the merged image showing Kir4.1 and AQP2 expression in the renal cortex. We confirm the previous report that Kir4.1 is expressed in some AQP2-positive tubules, presumably in the CNT and the CCD (10). However, Kir4.1 is not expressed in all AQP2-positive CNT/CCDs. From the inspection of Fig. 2, C and D, it is apparent that the expression of Kir4.1 is clearly detected in AQP2-positive CNTs or CCDs located in the top half of the renal cortex but almost absent in AQP2-positive tubules located in the lower portion of the renal cortex. Because Kir4.1 is mainly expressed in the CNT/CCDs located in the top part of the renal cortex, we next performed the patch-clamp experiments also in the tubule dissected from the top half of the renal cortex.
We then carried out the patch-clamp experiments to examine the basolateral K+ channel activity in the late CNT and early CCD as shown in Fig. 1. The most abundant K+ channel type in the basolateral membrane of the late CNT and initial CCD is an intermediate-conductance (40 ± 2 pS) K+ channel (n = 9). Figure 3A is a typical recording demonstrating the 40-pS K+ channel activity in a cell-attached patch, and Fig. 3B is a current (I)-voltage (V) relationship curve. We detected the 40-pS K+ channel activity in 9 patches from a total of 35 experiments (26%) (Fig. 3C). Moreover, this 40-pS K+ channel was completely absent in all 148 patches from Kcnj10−/− mice, suggesting that Kir4.1 is a key component of the 40-pS K+ channel. The second abundant K+ channel type in the late CNT and early CCD is a small-conductance K+ channel (Fig. 4A). Figure 4B is an I–V curve yielding a slope conductance of 23 ± 1 pS (n = 6) at −60 mV. From the inspection of Fig. 4A, it is apparent that the 23-pS K+ channel has a high and voltage-independent channel open probability (0.9 ± 0.1) from 0 to −60 mV. We detected this 23-pS K+ channel in 6 out of 35 patches (17%) in P9 WT mice. The disruption of Kir4.1 did not significantly affect the probability of finding the 23-pS K+ channel since we detected the 23-pS K+ channel in 21 out of 148 patches (14%) in Kcnj10−/− mice. Moreover, mean NPo (1.36 ± 0.2) of the 23-pS K+ channel in the CNT and CCD was similar in Kcnj10−/− and WT mice (n = 6, data not shown). Because the depletion of Kir4.1 did not affect 23-pS K+ channel activity, it is unlikely that the 23-pS K+ channel is a Kir4.1 homotetrammer, which has slope conductance of 20–23 pS (14, 26). The patch-clamp experiments have further detected the third type of K+ channel, a large-conductance channel, in the CNT/CCD (Fig. 5A). Figure 5B is an I–V curve yielding the channel slope conductance of 60 ± 2 pS (n = 4) at −60 mV. Figure 5A is a recording showing channel activity in a cell-attached patch, and it demonstrates that the channel activity defined by NPo is increased at a more negative membrane potential (20 mV, NPo = 0.4 ± 0.1; −60 mV, NPo = 1.56 ± 0.3). Thus, this 60-pS K+ channel is most likely a hyperpolarization-activated K+ channel detected previously in rat CCD (22). We detected this 60-pS K+ channel in 4 patches from a total of 35 experiments (11%) in WT mice and in 15 patches from total of 148 experiments (10%) in Kcnj10−/− mice, respectively (Fig. 5C). Moreover, the mean NPo of the 60-pS K+ channel at −60 mV was similar in Kcnj10−/− (NPo = 1.45 ± 0.2, n = 4) and WT (NPo = 1.40 ± 0.2, n = 4) mice. Thus, the disruption of Kir4.1 also had no effect on the probability of finding the 60-pS K+ channel in the CNT/CCD and the channel activity.
Because the disruption of Kir4.1 fails to increase the activity of the 23- and 60-pS K+ channel activity in the CNT and CCD, we speculate that the membrane should be depolarized in Kcnj10−/− mice compared with that of WT. Thus, we used the perforated whole cell recording to measure K+ reversal potential with 5 mM K+/140 mM Na+ in the bath and 145 mM K+ in the pipette. Figure 6 is a recording showing that the disruption of Kir4.1 shifts the K+ reversal potential slightly to the right (depolarization). Figure 6, right, summarizes the results of five experiments in which the K+ reversal potential was measured in the CNT/CCD segments in WT and Kcnj10−/− mice, respectively (WT, −69 ± 3 mV; Kcnj10−/−, 59.5 ± 2 mV). This suggests that Kir4.1 is partially responsible in generating the membrane potential in the CNT/CCD. Because K+ channels other than Kir4.1 are expressed in the CNT and CCD and Kir4.1 is also not expressed in the whole length of these nephron segments, the disruption of Kir4.1 should have a modest effect on the electrochemical gradient of Na+ entry only in part of the CNT and CCD.
Moreover, we hypothesize that the disruption of Kir4.1 may stimulate ENaC expression because the inhibition of NCC in the DCT in Kcnj10−/− mice should increase aldosterone or vasopressin induced by volume depletion. Therefore, we next examined the expression of ENaC expression in the WT, Kcnj10+/−, and Kcnj10−/− mice. Figure 7 is a Western blot showing the expression of ENaCα, -β, and -γ subunits in the renal cortex (Fig. 7A) and outer medulla (OM) (Fig. 7B). Although the disruption of Kir4.1 had no significant effect on ENaCα subunit expression, it significantly increased the expression of ENaCβ (cortex, 140 ± 20% of the control; OM, 170 ± 20% of the control) and -γ (160 ± 20% of the control for both cortex and OM) subunits (Fig. 7D). Moreover, although full-length ENaCα was not significantly changed in Kcnj10−/− mice compared with WT control, the cleaved form of ENaCα was significantly increased in Kcnj10−/− mice by 70 ± 20% (Fig. 7C). The notion that the depletion of Kir4.1 stimulates ENaCβ expression in medullary collecting duct is further indicated by immunostaining of ENaCβ subunit (Fig. 8). It is apparent that ENaCβ subunit expression is higher in the medulla of Kcnj10−/− mice than that of WT mice. Because it is not possible to directly measure either aldosterone or vasopressin in P9 neonatal mice, we could only speculate that an increase in vasopressin level may be mainly responsible for stimulation of ENaCβ and ENaCγ subunits. This view is supported by the report that aldosterone increased ENaCα expression without having an effect on ENaCβ and ENaCγ expression (11). On the other hand, it has been reported that vasopressin stimulated ENaCβ subunit expression (3). Our previous experiments supported the notion that the disruption of Kir4.1 increased vasopressin level as evidenced by the fact that AQP2 expression was significantly increased in P9 Kcnj10−/− mice (4). Figure 9 is an image showing that the intensity of AQP2 expression is higher in Kcnj10−/− mice than that of WT, suggesting a possible high vasopressin level in Kcnj10−/− mice. Thus, it is possible that a high vasopressin level may be mainly responsible for the stimulation of ENaCβ and ENaCγ expression.
The basolateral K+ channels play an important role in the regulation of NaCl transport in renal epithelial cells (2, 6). First, it has been suggested that changes in transepithelial NaCl transport must be accompanied by proportional change in basolateral K+ exit, thereby sustaining Na+-K+-ATPase activity (17). Second, we and other investigators have recently suggested that the basolateral K+ channels may play a role in controlling the Cl−-sensitive with-no-lysine kinase (WNK) pathway by changing the intracellular Cl− level (19, 20, 27). Because WNK activity is enhanced by lowering intracellular Cl− (19), it is conceivable that a high basolateral K+ channel activity should stimulate while the inhibition of the basolateral K+ channel should suppress WNK activity. The role of the basolateral K+ channels in regulating NaCl transport by a Cl−-sensitive pathway is supported by our previous experiments in which the downregulation of Kir4.1 decreased the ste20 proline-alanine rich kinase activity, a downstream kinase of WNK, thereby inhibiting NCC in the DCT (27). Terker et al. have also demonstrated that an increase in intracellular Cl− induced by depolarization is responsible for the inhibition of NCC activity in the cells expressing loss-function-of-Kir4.1 mutants (20). Thus, the basolateral K+ channel activity is important not only in sustaining Na+-K+-ATPase but also in regulating the Cl−-sensitive WNK pathway.
Molecular approaches and immunostaining have identified several types of inwardly rectifying K+ channels expressed in the basolateral membrane of the CNT and the CCD, and they are Kir4.1, Kir5.1, Kir2.3, and Kir7.1 (10, 13, 24). Our present study has also confirmed that Kir.4.1 is expressed in the basolateral membrane of the CNT and CCD. In addition to the inwardly rectifying K+ channels, immunostaining shows that the α-subunit of Ca2+-activated big-conductance K+ channels (BK) is also expressed in the basolateral membrane of the CCD in Na+-restricted animals, suggesting a possible role of BK in stimulating Na+ absorption during volume depletion (5). In the present experiments, we have identified three types of K+ channels [a small-conductance (23-pS), intermediate-conductance (40-pS), and large-conductance (60-pS) channel] in the basolateral membrane of the CNT and CCD. Because the 40-pS K+ channel is absent in the basolateral membrane of the CNT/CCD in Kcnj10−/− mice, this suggests that the 40-pS K+ channel is a Kir4.1/Kir5.1 heterotetramer (10). However, the disruption of Kir4.1 did not affect the probability of finding the 23- and 60-pS K+ channels in the basolateral membrane of the CNT/CCD, suggesting that these two types of K+ channels are not related to Kir4.1. Thus, unlike in the DCT in which Kir4.1 plays a predominant role in determining the basolateral K+ conductance (27), multiple types of K+ channels participate in generating the membrane potential in the late CNT and initial CCD (7). This notion is also supported by the observation that the depletion of Kir4.1 causes only a modest depolarization in the late CNT and initial CCD, suggesting that Kir4.1 participates only partially in generating the membrane potential in these nephron segments. A similar finding that multiple types of K+ channels are expressed in the basolateral membrane of principal cells was previously reported from experiments performed in rat CCD (8, 23). Although the finding that three types of K+ channels are expressed in the basolateral membrane is confirmatory, our study establishes firmly that Kir4.1 is a pore-containing component of the 40-pS K+ channel in the basolateral membrane. Moreover, we demonstrated that the function of Kir4.1 in the CNT/CCD is not completely compensated by two other types of K+ channels. However, because the present experiments were performed in P7-9 neonatal mice, we could not exclude the possibility that different types of K+ channels will be expressed in the CNT/CCD of adult mice. Thus, new experiments are required to examine the effect of depletion of Kir4.1 on K+ channel expression in the basolateral membrane of CNT/CCD of adult mice.
The molecular nature of 23- and 60-pS K+ channels is not known. It has been reported that Kir2.3 formed a 15-pS K+ channel when it was expressed in Xenopus oocytes (24). Moreover, immunostaining shows that Kir2.3 is expressed in the basolateral membrane of the CCD. In addition, a strong Kir.7.1 immunostaining was detected in the basolateral membrane of the CNT and CCD, whereas a weak Kir.7.1 staining was also detected in outer medullary collecting duct (OMCD) and inner medullary collecting duct of the rat kidney (13). Furthermore, the expression of Kir.7.1 was only detected in principal cells of the CCD (18). Further study is required to determine the role of Kir7.1 and Kir2.3 in forming the 23- and 60-pS K+ channels in the CNT and CCD.
Another finding of the present study is that the disruption of Kir4.1 increases the expression of ENaCβ and -γ subunits and the cleaved form of ENaCα in the CCD and OMCD. The upregulation of ENaC expression may explain the fact why patients with EAST/SeSAME syndrome have a modest phenotype of salt wasting despite of the inhibition of Na+ transport in the DCT (16). We suspect that increased expression of ENaCβ and -γ subunits and cleaved ENaCα should be a compensation action as a consequence of the downregulation of NCC. The inhibition of NCC is expected to cause a volume depletion that should increase the aldosterone and vasopressin level, thereby stimulating ENaC expression. Because the present experiments were performed in P7-9 neonatal mice, it is not possible to use pharmacological approaches to dissect the mechanism by which the depletion of Kir4.1 stimulates ENaC expression. Thus, further experiments are required to test how the disruption of Kir4.1 increases ENaC expression in adult Kir4.1 knockout mice. However, we reasoned that a high vasopressin level may be mainly responsible for the stimulation of ENaCβ and -γ expression in P9 neonatal Kcnj10−/− mice because vasopressin has been shown to stimulate the expression of ENaCβ subunit (3). Moreover, previous and present studies showed that the disruption of Kir4.1 increased AQP2 expression, suggesting Kcnj10−/− mice have a high vasopressin level (4). We conclude that Kir4.1 is expressed in the basolateral membrane of the CNT and initial CCD and plays a role in generating the negative membrane potential. Moreover, the depletion of Kir4.1 stimulates the expression of ENaCβ and -γ subunits in the medullary CCD. An increase in ENaC expression should be responsible for preventing excessive salt wasting despite inhibition of NCC in patients with EAST/SeSAME syndrome.
This work was supported by National Institutes of Health Grants DK-54983 and HL34100.
X.-T.S. performed the experiments, analyzed data, and wrote the paper; C.Z. performed the experiments, analyzed data; L.W. performed the experiments, analyzed data; R.G. designed experiments and analyzed data; D.-H.L. designed the experiments and performed experiments; W.-H.W. designed the experiments, analyzed data, interpreted the results, wrote and edited the paper.
We thank Drs. Junhui Zhang and Richard Lifton for providing Kcnj10−/−, Kcnj10+/−, and Kcnj10+/+ mice.
- Copyright © 2016 the American Physiological Society