IGF-1 and insulin exert opposite actions on ClC-K2 activity in the cortical collecting ducts

Abstract

Despite similar stimulatory actions on the epithelial sodium channel (ENaC)-mediated sodium reabsorption in the distal tubule, insulin promotes kaliuresis, whereas insulin-like growth factor-1 (IGF-1) causes a reduction in urinary potassium levels. The factors contributing to this phenomenon remain elusive. Electrogenic distal nephron ENaC-mediated Na⁺ transport establishes driving force for Cl⁻ reabsorption and K⁺ secretion. Using patch-clamp electrophysiology, we document that a Cl⁻ channel is highly abundant on the basolateral plasma membrane of intercalated cells in freshly isolated mouse cortical collecting duct (CCD) cells. The channel has characteristics attributable to the ClC-K2: slow gating kinetics, conductance ∼10 pS, voltage independence, Cl⁻>NO₃⁻ anion selectivity, and inhibition/activation by low/high pH, respectively. IGF-1 (100 and 500 nM) acutely stimulates ClC-K2 activity in a reversible manner. Inhibition of PI3-kinase (PI3-K) with LY294002 (20 μM) abrogates activation of ClC-K2 by IGF-1. Interestingly, insulin (100 nM) reversibly decreases ClC-K2 activity in CCD cells. This inhibitory action is independent of PI3-K and is mediated by stimulation of a mitogen-activated protein kinase-dependent cascade. We propose that IGF-1, by stimulating ClC-K2 channels, promotes net Na⁺ and Cl⁻ reabsorption, thus reducing driving force for potassium secretion by the CCD. In contrast, inhibition of ClC-K2 by insulin favors coupling of Na⁺ reabsorption with K⁺ secretion at the apical membrane contributing to kaliuresis.

it is generally recognized that transport in the distal part of renal tubule, including the connecting tubule and the collecting duct, plays an important role in maintenance of whole body fluid homeostasis by shaping excretion rates of water and electrolytes with urine (32, 40). This part contains two morphologically and functionally distinct cell populations, where principal cells (PC) mediate sodium and water reabsorption and secretion of potassium, and intercalated cells (IC) control acid-base balance and also participate in chloride reabsorption (40, 44). Activity of the epithelial sodium channels (ENaC) located to the apical membrane of PC underlies electrogenic Na⁺ transport in PC, which, in turn, provides driving force for luminal K⁺ exit via the renal outer medulla potassium channels (40). Cl⁻ reabsorption occurs via both paracellular (through tight junctions) and transcellular (across IC) routes (44).

The initial step of the transcellular Cl⁻ reabsorption is largely mediated by the electroneutral Cl⁻/HCO₃⁻ transporter Slc26a4 (pendrin) in base secreting B type of IC and Slc4a11 transporter acting as an electrogenic Cl⁻/HCO₃⁻ exchanger or a Cl⁻ channel in acid secreting A type of IC on the apical side (46). Recent experimental evidence indicates that basolateral Cl⁻ export to the interstitium in the cortical collecting duct (CCD) is mediated by the activity of the chloride ClC-K2 channel present in both A and B types of IC (18, 28, 42). ClC-K2 (ClC-Kb in humans) is predominantly expressed in the kidney and shares substantial homology with other members of ClC channels (13, 42, 43). In addition to the CCD, ClC-K2 is also present in other parts of the distal renal tubule, specifically in the thick ascending limb (TAL) and the distal convoluted tubule (DCT) (18). Interestingly, gain-of-function polymorphism ClC-Kb^T481S in humans is associated with elevated blood pressure due to augmented renal salt retention (12). This suggests that the channel possibly plays an essential role in mediating Cl⁻ reabsorption in the distal parts of renal tubule. Consistently, ClC-K2 loss-of-function mutations lead to salt-wasting phenotype of Bartter's syndrome type III associated with hypotension (39). However, little is known about mechanisms controlling function of the channel in different segments of the renal tubule, and specifically in the CCD, expressing ClC-K2.

Insulin and structurally related insulin growth factor-1 (IGF-1) have been long recognized to play a role in controlling renal function (2, 8). Both hormones trigger activation of PI3-kinase (PI3-K)- and mitogen-activated protein kinase (MAPK)-dependent pathways via predominant binding to respective insulin and IGF-1 receptors expressed at both apical and basolateral sides along the renal tubule (2, 7, 8). In the CCD, insulin and IGF-1 stimulate ENaC-mediated sodium reabsorption through a PI3-K-dependent mechanism (41). Acute intravenous IGF-1 injection results in a significant reduction in the fractional Na⁺ excretion in humans (6). In patients with acromegaly, augmented circulating IGF-1 levels result in antinatriuresis and hypertension, which can be corrected with ENaC inhibitor amiloride (14, 15). In contrast, the effects of insulin are very dependent on plasma glucose and K⁺ levels and often do not lead to salt retention and elevation in blood pressure (5, 20). Furthermore, IGF-1 reduces renal potassium excretion (6), whereas insulin can promote kaliuresis (5), particularly when plasma K⁺ levels are exogenously clamped (9, 36). The molecular details of the distinct effects of insulin and IGF-1 on urinary excretion patterns are not clear.

Despite the fact that transcellular reabsorption of Na⁺ and Cl⁻ is carried out in different cell types in the CCD, accumulating experimental evidence suggests that PC and IC often cooperate to promote reabsorption of NaCl with little change in net pH balance (reviewed in Ref. 44). Thus, genetic ablation of pendrin causes respective decreases in ENaC activity and protein abundance, contributing to a lower blood pressure and decreased renal salt retention (17, 33). On the other side, pendrin overexpression in IC leads to activation of sodium reabsorption without significant changes in trans-epithelial voltage and potassium excretion (11). In contrast, augmentation of Na⁺ transport in the PC only will result in stimulation of both potassium secretion and, to a lesser extent, paracellular Cl⁻ reabsorption (44). It is not clear whether ClC-K2-mediated basolateral chloride conductance contributes to the separation of Cl⁻ and K⁺ fluxes in the CCD.

In this study, we tested the effects of insulin and IGF-1 on the basolateral Cl⁻ conductance in mouse CCD. Using patch-clamp electrophysiology, we report that highly abundant 10-pS anion-selective channel at the basolateral plasma membrane demonstrates properties attributable to ClC-K2. Nanomolar concentrations of IGF-1 reversibly stimulated channel activity in a PI3-K-dependent manner. In contrast, insulin inhibited ClC-K2 activity in CCD cells via a mechanism dependent on MAPK activation. We propose that the opposite effects of insulin and IGF-1 on Cl⁻ transport in the CCD may, at least partially, underlie distinct patterns of urinary electrolyte excretion in response to the hormones.

MATERIALS AND METHODS

Reagents and animals.

All chemicals and materials were from Sigma (St. Louis, MO), VWR (Radnor, PA), and Tocris (Ellisville, MO) unless noted otherwise and were at least of reagent grade. Animal use and welfare adhered to the National Institutes of Health Guide for the Care and Use of Laboratory Animals following protocols reviewed and approved by the Animal Care and Use Committees of the University of Texas Health Science Center at Houston. For experiments, male C57BL/6J mice (Charles River Laboratories, Wilmington, MA) 6–10 wk old were used. Animals were maintained on standard rodent regimen (Purina, #5001) and had free access to tap water.

Tissue isolation.

The procedure for isolation of the CCDs suitable for electrophysiology is a modification from the protocols described previously (23, 24, 26, 47). Mice were killed by CO₂ administration followed by cervical dislocation and the kidneys were removed immediately. Kidneys were cut into thin slices (<1 mm) with slices placed into ice-cold physiologic saline solution [PSS; in mM: 150 NaCl, 5 KCl, 1 CaCl₂, 2 MgCl₂, 5 glucose, and 10 HEPES (pH 7.35)]. Straight cortical-medullary sectors, containing ∼30–50 renal tubules, were isolated by microdissection using watchmaker forceps under a stereomicroscope. Isolated sectors were further incubated in PSS containing 0.8 mg/ml collagenase type I (Alfa Aesar, Ward Hill, MA) and 5 mg/ml of dispase II (Roche Diagnostics, Mannheim, Germany) for 20 min at 37°C followed by extensive washout with an enzyme-free saline solution. Individual CCDs were visually identified by their morphological features (pale color, coarse surface and, in some cases, bifurcations) and were mechanically isolated from the sectors by microdissection. Isolated CCDs were attached to a 5 × 5-mm cover glass coated with poly-l-lysine. A cover glass containing a CCD was placed in a perfusion chamber mounted on an inverted Nikon Eclipse Ti microscope and perfused with PSS at room temperature. The tubules were used within 1–2 h after isolation.

Single channel recordings.

Single channel activity of ClC-K2 on the basolateral membrane in CCD cells was determined under voltage-clamp conditions in cell-attached and inside-out configurations. Recording pipettes had resistances of 8–10 MΩ. Bath and pipette solutions were (in mM) 150 NaCl, 5 KCl, 1 CaCl₂, 2 MgCl₂, 5 glucose, and 10 HEPES (pH 7.35); and 150 mM KCl, 2 mM MgCl₂, 10 mM HEPES (pH 7.35). For experiments testing pH sensitivity, pH of bath solution was adjusted with either HCl or NaOH to 5.0 and 8.0, respectively. The current-voltage (I–V) relationships were obtained by monitoring channel activity at applied pipette voltages (−V_p) from −100 to +40 mV with a step of 20 mV for at least 180 s.

For paired patch-clamp experiments, cell-attached configuration was used with pipette voltage being −V_p = −60 or −40 mV. Current recordings were made in a permanently perfused bath (1.5 ml/min). For each experimental condition, CCDs from at least three different mice were assayed. Gap-free single channel current data from GΩ seals were acquired with an Axopatch 200B (Molecular Devices, Sunnyvale, CA) patch-clamp amplifier interfaced via a Digidata 1440 (Molecular Devices) to a PC running the pClamp 10.4 suite of software (Molecular Devices). Currents were low-pass filtered at 1 kHz with an eight-pole Bessel filter (Warner Instruments, Hamden, CT). Events were inspected visually before acceptance. In paired experiments, ClC-K2 activity was analyzed over a span of 60–120 s for each experimental condition after a new steady-state was reached in response to a treatment. Channel activity (NP_o) and open probability (P_o) were assessed using Clampfit 10.4 software (Molecular Devices). For calculating P_o in paired experiments, N was fixed as the greatest number of active channels observed in control or experimental conditions. For representation, current traces were filtered at 200 Hz and corrected for slow baseline drifts, as necessary.

Ion selectivity.

The permeability ratio for Cl⁻ and NO₃⁻ was calculated using the equation of Goldman, Hodgkin, and Katz. For this, the reversal potential (E_rev) was determined in inside-out patch-clamp experiments, where 150 mM KCl in the recording pipette was substituted with 150 mM KNO₃.

Basolateral membrane voltage measurements.

To monitor real-time changes in membrane voltage, CCD cells were studied under current-clamp mode using the perforated-patch technique. Freshly made Amphotericin-B, 400 μM (Enzo Life Sciences, Farmingdale, NY) was dissolved in the pipette solution containing 150 mM KCl, 2 mM MgCl₂, 10 mM HEPES (pH 7.35) by ultrasonication. Electrical recordings were made once the access resistance from the pipette to the cell interior fell to <15 MΩ, usually 5–10 min after achieving a pipette-to-membrane seal resistance of 5–10 GΩ.

Data analysis.

All summarized data are reported as means ± SE. In paired experiments, data from before and after treatment were compared using the paired t-test. Data from unpaired experiments were compared with a Student's (2-tailed) t-test or a one-way ANOVA as appropriate. P < 0.05 was considered significant.

RESULTS

Determination of molecular identity of the basolateral Cl⁻ channel in CCD cells.

Previous studies suggest the presence of anion conductance on the basolateral membrane of IC (28, 30). We first employed patch-clamp electrophysiology in a cell-attached configuration in freshly isolated enzymatically treated mouse CCD to perform functional characterization of this conductance. Using patch pipette containing 150 mM KCl, we observed a highly abundant Cl⁻ channel in ∼40% of patches. Typical patch-clamp recordings at different pipette potentials and the I–V relationship of the channel with slow gating kinetics in CCD cells are shown in Fig. 1, A and B, respectively. The E_rev was ∼0 mV and the estimated channel conductance was 11.5 ± 0.3 pS. Equimolar substitution of K⁺ to Na⁺ and Li⁺ in the pipette solution, had no measurable effect on single channel conductance and E_rev (data not shown). In contrast, replacement of Cl⁻ with acetate induced rightward shift of the I–V curve by ∼20 mV (Fig. 1B, gray trace). This indicates an anion-selective nature of the channel. The majority of patches have more than five channels that make evaluation of the P_o not always accurate. Thus, we assessed P_o of the channel only in patches containing fewer than five channels. Figure 1C demonstrates no apparent voltage dependence of channel P_o at the tested pipette potentials from −100 to +40 mV.

Fig. 1.

Functional properties of ClC-K2-like channels abundantly expressed on the basolateral membrane of cortical collecting duct (CCD) cells. A: representative current traces of single channel activity recorded from the same patch at different pipette potentials, as indicated. A c denotes closed nonconducting state. B: average current-voltage (I–V) relationship of the unitary current amplitude for channels similar to that shown in A when patch pipette contains 150 mM KCl (black trace) and 150 mM KAcetate (gray trace). Number of experiments is also indicated. C: dependence of open probability (P_o) of the channel from voltage applied to the recording pipette. For each condition, P_o was estimated over a time span of at least 180 s.

We next quantitatively determined anion selectivity of the recorded channel upon excision of the patch and controlling solute composition at both sides. Figure 2A demonstrates respective I–V relationship in cell-attached (black trace) and inside-out (gray trace) configurations when the majority of Cl⁻ in the pipette was replaced with NO₃⁻. The calculated permeability ratio P_Cl/P_NO3 = 1.46 (see materials and methods). Furthermore, the leftward shift of ∼15 mV in the I–V upon excitation suggests that the channel predominates in the IC, which are known to have much more depolarized basolateral membrane potential than PC (27).

Fig. 2.

Anion selectivity and contribution to the basolateral voltage of ClC-K2-like channel. A: average I–V relationship of the unitary current amplitude for the basolateral Cl⁻ channels when patch pipette contains 150 mM KNO₃ in cell-attached (black trace) and excised inside-out (gray trace) configurations. Number of experiments is also indicated. B: representative continuous voltage trace monitoring basolateral membrane potential in an intercalated cell (IC) in the control, upon application of a broad spectrum Cl⁻ channel inhibitor 100 μM NPPB and 150 mM NaCitrate. Application times are shown with respective bars on top.

We next tested whether basolateral Cl⁻ channels participate in setting resting membrane potential in IC. For this, we directly monitored changes in the basolateral membrane voltage using current-clamp configuration upon application of broad spectrum Cl⁻ channel inhibitors NPPB (100 μM) or DIDS (100 μM). Previous reports showed that similar concentrations of the antagonists drastically reduced ClC-K2 activity in renal tubule cells (1, 28). We focused on IC having values of basolateral membrane voltage of ∼−10 mV, in contrast to PC having basolateral membrane voltage of −70 mV (47). NPPB treatment has no measurable effect on the basolateral membrane voltage (n = 5), whereas replacement of bath NaCl for NaCitrate elicits prominent depolarization, as expected (Fig. 2B). Application of DIDS also has no effect on membrane potential (n = 4, data not shown). These results suggest that either ClC-K2 does not directly participate in controlling basolateral membrane potential or ClC-K2 inhibition can be compensated by other membrane transporter proteins.

We next probed regulation of Cl⁻ channel activity by pH. An acute increase in extracellular pH from 7.35 to 8.0 induces pronounced augmentation of channel activity as demonstrated on the representative current trace in Fig. 3A. Washout with control saline (pH 7.35) returns activity of the cannel to control values. Furthermore, application of acidic extracellular media (pH 5.0) drastically reduces channel P_o in a reversible manner (Fig. 3B). Figure 3C summarizes the effect of extracellular pH on the activity of Cl⁻ channels in paired experiments similar to that demonstrated in Fig. 3, A and B. The mean values of P_o are 0.05 ± 0.01, 0.30 ± 0.04, and 0.64 ± 0.02 for pH 5.0, 7.35, and 8.0, respectively. We concluded that the channel is highly pH sensitive. Alkalic pH potentiates Cl⁻ channel, whereas acidic pH decreases channel activity.

Fig. 3.

ClC-K2-like channel is pH sensitive. A: representative continuous current trace from a cell-attached patch monitoring activity of basolateral ClC-K2-like channels in the control (pH 7.35), after application of basic (pH 8.0) media, and following washout. The patch was clamped to −V_p = −40 mV. A c denotes closed nonconducting state. Application times are shown with respective bars on top. B: representative continuous current trace from a cell-attached patch monitoring activity of basolateral ClC-K2-like channels in the control (pH 7.35), after application of acidic (pH 5.0) media, and following washout. All other conditions are as in A. C: summary graph of changes in channel P_o from pH in paired experiments similar to that shown in A and B. Number of experiments is indicated. *Significant change vs. pH 7.35.

All defined properties: slow channel kinetics (Fig. 1A), single channel conductance (Fig. 1B), independence from applied voltage (Fig. 1C), expression in IC, anion permeability pattern (Fig. 2A), and pH sensitivity (Fig. 3) of the basolateral Cl⁻ channel in CCD cells are consistent with those reported for the ClC-K2 channel, which was previously shown to be functionally expressed in distal tubular segments, including the CCD (22, 28, 45).

IGF-1 stimulates ClC-K2 channel activity in a PI3-K-dependent manner.

Previous reports document that IGF-1 receptors are expressed at the basolateral membrane of CCD (2, 35). ClC-K2, in tandem with HCO₃⁻/Cl⁻ exchangers Slc26a4 (pendrin) and Slc4a11, is thought to be involved in the net Cl⁻ reabsorption in IC of the CCD (44). IGF-1 stimulates Na⁺ reabsorption by increasing ENaC activity in PC (41). Thus, we next determined whether IGF-1 also targets ClC-K2 to promote Cl⁻ transport in the CCD. Basolateral application of 100 nM IGF-1 acutely stimulates ClC-K2 activity in a reversible manner, as demonstrated by the representative continuous current trace in Fig. 4A. The summary graph of changes in ClC-K2 P_o in paired experiments upon treatment with IGF-1 is shown in Fig. 4B. The mean values of P_o are 0.31 ± 0.04, 0.55 ± 0.06, and 0.41 ± 0.06 in the control, during IGF-1 (100 nM) application, and washout, respectively (n = 7). Application of a higher IGF-1 concentration (500 nM) yields virtually identical results (Fig. 4C). The mean values of P_o are 0.35 ± 0.05, 0.59 ± 0.06, and 0.42 ± 0.07 in the control, during IGF-1 (500 nM) application, and washout, respectively (n = 5).

Fig. 4.

Insulin-like growth factor (IGF)-1 acutely increases open probability of basolateral ClC-K2 channel. A: representative continuous current trace from a cell-attached patch monitoring activity of basolateral ClC-K2 channels in CCD cells in the control, after application of 100 nM IGF-1, and following washout with control media. The patch was clamped to −V_p = −60 mV. IGF-1 application time is shown with the bar on top. Areas (1) and (2) are shown below with an expanded time scale. B: summary graph of changes in ClC-K2 open probability upon treatment with 100 nM IGF-1 from paired patch-clamp experiments similar to that shown in A. C: summary graph of changes in ClC-K2 open probability upon treatment with 500 nM IGF-1. *Significant increase vs. control.

Stimulation of IGF-1 receptors generally leads to activation of a PI3-K-dependent cascade (8). Therefore, we next tested whether inhibition of PI3-K affects regulation of ClC-K2 activity by IGF-1 in CCD cells. Application of a PI3-K inhibitor, LY294002 (20 μM, 3 min), has no measurable effect on ClC-K2, as summarized in Fig. 5. However, this treatment abolishes stimulation of ClC-K2 activity by IGF-1. The mean values of P_o are 0.31 ± 0.06, 0.33 ± 0.07, 0.31 ± 0.07, and 0.30 ± 0.06 in the control, after LY294002 (20 μM), during application of IGF-1 (100 nM) in the continued presence of the PI3-kinase antagonist, and washout, respectively (n = 5). Overall, we concluded that IGF-1 stimulates ClC-K2 in a PI3-K-dependent manner.

Fig. 5.

IGF-1 stimulates ClC-K2 activity in a PI3-K-dependent manner. Summary graph of changes in ClC-K2 P_o in paired cell-attached patch-clamp experiments in the control, upon treatment with PI3-K inhibitor LY294002 (20 μM), IGF-1 (100 nM) in the continued presence of the inhibitor, and following washout.

Insulin inhibits ClC-K2 via stimulation of a MAPK-dependent cascade.

We next aimed to investigate regulation of ClC-K2 by insulin, which also gives a stimulatory effect on ENaC-mediated sodium reabsorption in the CCD (10, 31). In contrast to IGF-1, application of insulin (100 mM) greatly decreased ClC-K2 activity and the effect was reversed upon washout of the hormone, as demonstrated in the representative continuous current trace in Fig. 6A. The summary graph of changes in ClC-K2 P_o in paired experiments upon treatment with insulin is shown in Fig. 6B. The mean values of P_o are 0.34 ± 0.04, 0.11 ± 0.02, and 0.24 ± 0.04 in the control, during insulin (100 nM) application, and washout, respectively (n = 7).

Fig. 6.

Insulin reversibly inhibits ClC-K2 activity in CCD cells. A: representative continuous current trace from a cell-attached patch monitoring activity of basolateral ClC-K2 channels in CCD cells in the control, after application of 100 nM insulin, and following washout with control media. The patch was clamped to −V_p = −60 mV. Insulin application time is shown with the bar on top. Areas (1) and (2) are shown below with an expanded time scale. B: summary graph of changes in ClC-K2 open probability upon treatment with 100 nM insulin from paired patch-clamp experiments similar to that shown in A. *Significant decrease vs. control.

Activation of insulin receptors may stimulate PI3-K and MAPK intracellular cascades (8). However, we demonstrate that PI3-K-dependent pathway increases ClC-K2 activity in response to IGF-1. Indeed, inhibition of PI3-K with LY294002 (20 μM, 3 min) failed to preclude inhibitory action of insulin to ClC-K2 (Fig. 7A). The mean values of P_o are 0.38 ± 0.05, 0.37 ± 0.05, 0.24 ± 0.05, and 0.33 ± 0.05 in the control, after LY294002 (20 μM), during application of insulin (100 nM) in the continued presence of the PI3-kinase antagonist, and washout, respectively (n = 6). In contrast, application of MEK inhibitor U0126 (20 μM, 3 min) to disrupt MAPK signaling abolished inhibition of ClC-K2 activity by insulin (Fig. 7B). The mean values of P_o are 0.38 ± 0.07, 0.38 ± 0.06, 0.37 ± 0.07, and 0.38 ± 0.06 in the control, after U0126 (20 μM), during application of insulin (100 nM) in the continued presence of the MEK antagonist, and washout, respectively (n = 5). Thus, we concluded that insulin inhibits ClC-K2 activity in the CCD cells via MAPK-dependent signaling cascade.

Fig. 7.

Insulin decreases ClC-K2 activity via a MAPK-dependent pathway. A: summary graph of changes in ClC-K2 P_o in paired cell-attached patch-clamp experiments in the control, upon treatment with PI3-K inhibitor LY294002 (20 μM), insulin (100 nM) in the continued presence of the inhibitor, and following washout. *Significant decrease vs. control. B: summary graph of changes in ClC-K2 P_o in paired cell-attached patch-clamp experiments in the control, upon treatment with MEK inhibitor U0126 (5 μM), insulin (100 nM) in the continued presence of the inhibitor, and following washout.

DISCUSSION

In this study, we provide the first functional evidence that basolateral Cl⁻ conductance in CCD cells is under direct hormonal control by IGF-1 and insulin. We further put forward the hypothesis that the opposing effects of the aforementioned hormones on the basolateral ClC-K2 chloride channel participate in separation of Cl⁻ and K⁺ fluxes stimulated by ENaC-mediated Na⁺ reabsorption depending on physiological needs. This, in turn, suggests a possible mechanism that contributes to different patterns of urinary excretion of electrolytes caused by IGF-1 and insulin (Fig. 8). Specifically, we propose that activation of ClC-K2 channels by IGF-1 and subsequent stimulation of a PI3-K-dependent pathway facilitate net NaCl retention (Fig. 8A), whereas inhibition of the channel by insulin via a MAPK-dependent mechanism creates favorable conditions for coupling of Na⁺ and K⁺ transport in the CCD (Fig. 8B), therefore, elevating excretion of K⁺ with urine.

Fig. 8.

Principal scheme of regulation of ClC-K2 activity in CCD cells by IGF-1 (A) and insulin (B). Solid bold and dashed thin arrows represent dominant and minor ion transporting pathways, respectively.

Using patch-clamp technique, we routinely monitor the activity of the anion channel at the basolateral membrane of CCD cells, which demonstrates all the molecular signature characteristics of ClC-K2 (28). This includes ∼10-pS conductance, slow gating, independence of P_o from applied voltage, Cl⁻>NO₃⁻ selectivity, and pH sensitivity. Our observations resonate with previous work by Teulon's group (28) that recorded a channel with virtually identical characteristics in the CNT/CCD region of the mouse renal tubule. It also appears that the channel is expressed in cells having greatly depolarized basolateral membrane potential (Fig. 2A), which are IC. Indeed, previous studies in perfused tubules suggest low-K⁺ conductance in the basolateral membrane of IC (as opposed to PC having predominantly K⁺ exiting mechanisms) and major Cl⁻ conductance (19, 27, 38). An anion channel with similar properties was reported to be functionally expressed in the DCT and the TAL (22, 28, 30, 45). The expression pattern of the channel is consistent with the immunofluorescent detection of ClC-K2 in mouse kidney sections (18). Overall, the existing evidence strongly supports the view that ClC-K2 activity mediates basolateral Cl⁻ conductance in IC of CCD. It should be noted, although, that the final conclusion can be made only after ablation of ClC-K2 in this segment using genetic tools.

One of the most striking features of ClC-K2 is its functional expression levels. Thus far, this is the most abundant channel recorded with patch clamping in the CCD cells. The majority of cell-attached patches have more than five simultaneously active ClC-K2 channels, which makes it difficult to reliably estimate channel P_o in all cases. While not exactly precise, patch-clamp studies indicate that functional expression levels of ClC-K2 in CCD region are no less or even possibly higher than those in the DCT (22) and TAL (45). It is proposed that ClC-K2 participates in Cl⁻ reabsorption in the distal renal tubule, starting from TAL (18). Inhibition of ClC-K2 in vivo promotes diuresis (21) and loss-of-function mutations in ClC-K2 result in salt-wasting phenotype of Bartter's syndrome type III (39). However, the levels of ClC-K2 in CCD are far more than necessary to perform equimolar reabsorption of NaCl at this site considering functional expression levels of ENaC (23, 25, 34), the major route for electrogenic Na⁺ transport, and similar conductance of both channels. This argues that ClC-K2 is likely contributing to processes other than Cl⁻ reabsorption, such as regulatory volume increase/decrease and possibly establishing basolateral membrane voltage. While our data (Fig. 2B) fail to demonstrate the direct effect of ClC-K2 activity on membrane voltage, this possibility cannot be ruled out because compensatory mechanisms likely exist. The absence of currently available selective ClC-K2 inhibitors does not also exclude the scenario that nonspecific adverse effects of NPPB and DIDS may contribute to the observed negative results. In addition, remarkable pH sensitivity of ClC-K2 indicates that the channel may participate in regulation of acid-base balance in IC. Further studies are necessary to carefully examine these possibilities.

Our observation that insulin and IGF-1 exert opposing effects on the basolateral ClC-K2 channels in the CCD cells is somehow surprising from the first glance. Insulin and IGF-1 have substantial structural similarity and IGF-1 receptor is also very similar to insulin receptor (16, 37). Both receptors possess tyrosine kinase activity and involve activation of PI3-K and MAPK pathways (8). Despite this, insulin and IGF have clearly distinct physiological functions. Activation of IGF-1 receptors is primarily important for growth and development, whereas insulin receptors are viewed as metabolic mediators (16). Furthermore, insulin and IGF-1 have much lower affinity (up to 500-fold) for the respective counterpart receptor (7). We found that activation of distinct intracellular signaling pathways, namely PI3-K-dependent (Fig. 5) and MAPK-dependent (Fig. 7), mediates regulation of ClC-K2 activity by IGF-1 and insulin, respectively. This indicates that different receptors are involved in conferring opposite effects on ClC-K2 activity in response to the hormones. A recent study demonstrates that similar concentrations of IGF-1 (up to 300 nM) cause inhibition of basolateral 10-pS Cl⁻ channel (presumably ClC-K2) in TAL cells via activation of PI3K-AKT-mTOR pathway (45). It was proposed that IGF-1-induced inhibition of trans-epithelial Cl⁻ absorption in the TAL would decrease oxygen consumption in the medullary TAL, thus contributing to a beneficiary role of IGF-1 during ischemia-reperfusion injury (29). However, intrarenal infusions of similar IGF-1 concentrations induce antidiuresis and antinatriuresis (4, 6), which is consistent with augmented net NaCl reabsorption by renal tubule. Indeed, stimulatory effects of IGF-1 on sodium-transporting proteins, NKCC2 in the TAL and ENaC in CCD, are reported (3, 41). We did observe that 100 and 500 nM IGF-1 elicit virtually identical reversible increases in the activity of basolateral ClC-K2 channel in CCD cells (Fig. 4, B and C), which likely facilitates salt retention at this site. It is possible that different sets of membrane receptors for IGF-1 and insulin are present in TAL and CCD, to convey segment-specific actions of the hormones. In support of this view, insulin in concentrations for up to 500 nM had no effect on ClC-K2 activity in the TAL (45). It should be noted that micromolar insulin concentration did inhibit ClC-K2, pointing that the effect is mediated by IGF-1R.

According to the classical view, the distal nephron, and specifically CCD, is the site where potassium secretion is coupled with ENaC-mediated sodium reabsorption (40). ENaC activity creates a favorable driving force for K⁺ to exit from principal cells which, in turn, is a major contributor for urinary potassium levels. Mineralocorticosteroid aldosterone is the principal ENaC activator (40). However, the substantial portion of patients with elevated circulating aldosterone levels does not develop renal potassium wasting and hypokalemia (48). This suggests that Na⁺ and K⁺ fluxes in the CCD can occur independently based on physiological needs. Modulation of trans-cellular Cl⁻ reabsorption might be one of the possible mechanisms contributing to this separation. Thus, overexpression of the apical Cl⁻ transporter pendrin in IC of the CCD leads to respective activation of sodium reabsorption without significant changes in trans-epithelial voltage and potassium excretion (11). In addition to aldosterone, both IGF-1 and insulin activate ENaC in the CCD (10, 31, 41). However, only insulin can increase urinary K⁺ excretion (9, 36), whereas IGF-1 primarily causes salt retention and presumably ENaC-dependent hypertension, at least in patients with acromegaly (14, 15). Our finding that IFG-1 and insulin have opposite effects on the basolateral ClC-K2 in the CCD may provide an explanation of different patterns of urinary K⁺ excretion produced by these hormones. IGF-1, by stimulating ENaC in PC and ClC-K2 channels in IC, elicits cooperative NaCl reabsorption and, therefore, a reduction of potassium secretion by the CCD. In contrast, insulin by stimulating ENaC but inhibiting ClC-K2 speeds up ion fluxes in PC cells only (i.e., coupling of Na⁺ reabsorption with K⁺ secretion).

In summary, this manuscript provides the first observation of direct effects of insulin and IGF-1 on the basolateral Cl⁻ conductance, and specifically on the basolateral ClC-K2 channel, in the IC of CCD via distinct intracellular signaling pathways. It is possible that modulation of trans-cellular Cl⁻ reabsorption by controlling ClC-K2 activity may provide a physiologically relevant mechanism enabling dissociation of sodium reabsorption from potassium secretion at this site.

GRANTS

This research was supported by National Institutes of Health-National Institute of Diabetes and Digestive and Kidney Diseases Grant DK095029 (to O. Pochynyuk), American Heart Association (AHA)-GIA-13GRNT16220002 (to O. Pochynyuk), and AHA-14POST20380979 (to M. Mamenko).

DISCLOSURES

No conflicts of interest, financial or otherwise, are declared by the author(s).

AUTHOR CONTRIBUTIONS

Author contributions: O.Z., M.M., and N.B. performed experiments; O.Z. and M.M. analyzed data; O.Z., M.M., and O.P. prepared figures; O.Z., M.M., N.B., and O.P. edited and revised manuscript; O.Z., M.M., N.B., and O.P. approved final version of manuscript; N.B. and O.P. interpreted results of experiments; O.P. conception and design of research; O.P. drafted manuscript.