Mechanistic Target of Rapamycin Complex 1 in Distal Convoluted Tubule and Renal Potassium Handling

Visual Abstract

Abstract

Key Points

• Rapamycin inhibited Kir4.1/Kir5.1, decreased Na-Cl-cotransporter (NCC), and increased renal potassium (K⁺) excretion.

• Inhibition of mechanistic target of rapamycin complex 1 in the distal convoluted tubule suppressed Kir4.1/Kir5.1, inhibited NCC, and increased renal K⁺ excretion.

• The mechanistic target of rapamycin complex 1 in the distal convoluted tubule plays a role in maintaining K⁺ homeostasis by controlling baseline activity of Kir4.1/Kir5.1 and NCC.

Background

Mechanistic target of rapamycin complex 1 (mTORc1) plays a role in maintaining potassium (K⁺) homeostasis. We now examine whether mTORc1 of distal convoluted tubule (DCT) regulates Kir4.1/Kir5.1 channels and thiazide-sensitive Na-Cl cotransporter (NCC), which plays a role in regulating renal K⁺ excretion.

Methods

We used patch clamp technique to examine basolateral Kir4.1/Kir5.1 in early DCT, immunoblotting to examine NCC expression, and in vivo measurement of urinary K⁺ excretion to determine baseline renal K⁺ excretion (E_K) in the mice treated with rapamycin and in DCT-specific regulatory-associated protein of mechanistic target of rapamycin knockout mice (DCT-RAPTOR-KO).

Results

Application of rapamycin decreased Kir4.1/Kir5.1-mediated K⁺ currents and depolarized DCT membrane potential in Fkbp1a^flox/flox mice. However, the effect of rapamycin on Kir4.1/Kir5.1 was absent in kidney-specific 12kDa FK506-binding protein knockout mice. Rapamycin decreased basolateral 40-pS K⁺ channel activity (Kir4.1/Kir5.1 heterotetramer) of the DCT. This effect was absent in the DCT treated with hydrogen peroxide, which stimulated the 40-pS K⁺ channel activity, suggesting the role of reactive oxygen species in mediating the effect of mTORc1 on Kir4.1/Kir5.1. Rapamycin treatment significantly decreased the abundance of both phosphorylated NCC and total NCC in Fkbp1a^flox/flox mice but not in kidney-specific 12kDa FK506-binding protein knockout mice. Moreover, in vivo measurement of urinary Na⁺ excretion and urinary K⁺ excretion demonstrated that rapamycin treatment decreased hydrochlorothiazide-induced natriuresis but increased renal K⁺ excretion in Fkbp1a^flox/flox mice. Moreover, Kir4.1/Kir5.1-mediated K⁺ currents of the DCT were lower, and DCT membrane potential was less negative in DCT-RAPTOR-KO than those of Ncc-Cre-Raptor^flox/flox mice. In addition, the abundance of phosphorylated NCC was lower in DCT-RAPTOR-KO mice than Ncc-Cre-Raptor^flox/flox mice. By contrast, the abundance of phosphorylated type II Na-Cl-K cotransporter was the same between two genotypes, while cleaved alpha subunit of epithelial sodium channel abundance was higher in DCT-RAPTOR-KO mice than Ncc-Cre-Raptor^flox/flox mice. Consequently, DCT-RAPTOR-KO mice had a higher urinary K⁺ excretion and lower plasma K⁺ concentrations than Ncc-Cre-Raptor^flox/flox.

Conclusions

mTORc1 in the DCT plays a significant role in maintaining K⁺ homeostasis by controlling the basolateral Kir4.1/Kir5.1 of the DCT and NCC.

Introduction

The mechanistic target of rapamycin complex 1 (mTORc1) is a serine/threonine protein kinase and is a multiprotein complex.^1,2 The core components of mTORc1 are composed of three major subunits including the kinase mammalian target of rapamycin, regulatory associated protein of mammalian target of rapamycin (RAPTOR), and mLST8 (mammalian lethal with Sec 13 protein 8).³ The inhibition of mTORc1 has been shown to cause hypokalemia in rats, whereas the stimulation of mTORc1 of the cortical collecting duct has been shown to cause hyperkalemia in mice due to the inhibition of alpha subunit of epithelial sodium channel (αENaC) and renal outer medullary potassium (K⁺) channel (Kir1.1) (ROMK) expression.^4,5 Maintaining K⁺ homeostasis is essential for the function of the heart, skeletal muscles, and neurons because hypokalemia or hyperkalemia could cause life-threatening consequences such as cardiac arrhythmia. In this regard, it is now well established that thiazide-sensitive Na-Cl-cotransporter (NCC) not only plays a role in reabsorbing filter Na^+6–8 but also plays a key role in the regulation of K⁺ homeostasis and renal K⁺ excretion.^9,10 The role of NCC in regulating renal K⁺ excretion is achieved by controlling Na⁺ delivery to the late distal nephron. The coordinated action among NCC, epithelial Na⁺ channel (ENaC), Ca²⁺-activated big-conductance K⁺ channel, and ROMK is essential for maximally enhancing renal K⁺ excretion during high dietary K⁺ intake and for effectively preventing K⁺ wasting during low dietary K⁺ intake.^9,11–15 Furthermore, we and others have demonstrated that the basolateral K⁺ channels in the distal convoluted tubule (DCT) play an important role in controlling NCC abundance and activity, thereby playing a role in regulating renal K⁺ excretion.^14,16–20 The basolateral K⁺ channels of the DCT are composed of Kir4.1 and Kir5.1, which interact with each other to form a 40-pS inwardly rectifying K⁺ channel.^16,21 Kir4.1 provides the K⁺ conductance for the heterotetramer since Kir4.1 deletion almost completely abolishes the basolateral K⁺ conductance in the DCT.^14,20 On the other hand, Kir5.1 is an important regulatory subunit for the heterotetramer. We have demonstrated that Kir5.1 interacts with the E3 ubiquitin ligase Nedd4-2, which then regulates the ubiquitination of Kir4.1.²² Our previous studies have demonstrated that high-K⁺ intake–induced inhibition of Kir4.1/Kir5.1 is an essential step for high-K⁺ intake–induced inhibition of NCC.^14,23 Conversely, low-K⁺ intake–induced stimulation of Kir4.1/Kir5.1 of the DCT is an essential step for low-K⁺ intake–induced stimulation of NCC.^14,23 The role of Kir4.1 and Kir5.1 in the regulation of renal K⁺ excretion has been demonstrated in the mouse models: The deletion of Kir4.1 inhibits NCC activity and causes renal K⁺ wasting,^17,20 whereas the deletion of Kir5.1 increases NCC activity and reduces renal K⁺ excretion ability during high dietary K⁺ intake.²³ Thus, Kir4.1/Kir5.1 heterotetramer in the DCT is an important component of “K⁺ sensor” mechanism and plays a role in regulating K⁺ excretion in the kidney.¹⁶ Because mTORc1 is known to have an effect on renal K⁺ excretion and it is also expressed in the DCT,²⁴ it raises the question of whether mTORc1 plays a role in regulating Kir4.1/Kir5.1 and NCC in the DCT. Thus, the aim of this study was to examine the role of the mTORc1 in regulating Kir4.1/Kir5.1 of the DCT using pharmacologic and transgenic mouse models.

Methods

All methods, including animal preparations, electrophysiology, Western blot, fluorescence immunostaining, and in vivo measurement of hydrochlorothiazide (HCTZ)-induced natriuresis, are available within the article and its online Supplemental Material.

Generating Kidney-Specific 12kDa FK506-Binding Protein Knockout Mice and DCT-Specific RAPTOR Knockout Mice

Mice expressing Pax8-rtTA and tet-on Li-Cor Biotech Odessey-1 transgene were crossed with FKBP1a-floxed mice for generating inducible kidney tubule–specific FK506-binding protein-12 knockout mice (Ks-12kDa FK506-binding protein [FKBP12]-KO), as previously described.²⁵ FKBP12 is a binding protein for rapamycin, and rapamycin/FKBP12 complex is required for the inhibition of mTORc1.²⁶ A previous study demonstrated that deletion of FKBP12 in the renal tubule had no obvious phenotypes regarding the baseline kidney function.²⁵ Moreover, we previously showed that deletion of FKBP12 in the kidney had no significant effect on baseline activity of Kir4.1/Kir5.1 of the DCT.²⁷ Fkbp1a deletion was conducted in 8-week-old male and female mice homozygous for floxed Fkbp1a gene and heterozygous for Pax8-rtTA/Li-Cor Biotech Odessey-1 transgene by providing doxycycline (5 mg/ml containing 5% sucrose) in the drinking water for 2 weeks. This was followed by at least 2 additional weeks without doxycycline treatment before performing experiments. Littermate mice of the same age and genetic background drinking 5% sucrose were used as controls (Fkbp1a^flox/flox). The deletion of FKBP12 was confirmed by Western blot showing that FKBP12 protein was absent in doxycycline-treated mice (Supplemental Figure 1). Moreover, the expression of Kir4.1 seems also similar between two genotypes. To generate DCT-specific regulatory-associated protein of mechanistic target of rapamycin knockout mice, we used NCC-Cre system, as described previously.²⁸ The mice expressing Slc12a3-Cre-ERT2 were crossed with Raptor^flox/flox mice (purchased from Jackson Lab) to generate Slc12a3-Cre-ERT2-Cre-Raptor^flox/flox (NCC-Cre-Raptor^flox/flox) mice. We then treated NCC-Cre-Raptor^fl/fl mice with tamoxifen (daily peritoneal injection with tamoxifen at 1 mg in 100 μl corn oil, for 5 days) to induce Cre entering the nucleus for recombination at floxed alleles, leading to DCT-specific conditional deletion of RAPTOR. Tail DNA was PCR amplified, and the primers for genotyping are presented in Supplemental Table 1. The animal use protocol has been approved by New York Medical College Institutional Animal Care and Use Committee.

Materials

Chemicals including Inactin, rapamycin, hydroperoxide, and hydrochlorothiazide were purchased from Sigma-Aldrich (St. Louis, MO). Supplemental Table 2 summarizes the information regarding antibodies used in this study.

Statistical Analysis

We used software (Sigma plot 14) for the statistical analysis. For analyzing the values between two groups, we used the t test, and for comparisons of the values within the same group, we used the paired t test. We used one-way or two-way ANOVA for analyzing results of more than two groups, and the Holm–Sidak test was used as post hoc analysis. P values < 0.05 were considered statistically significant. The sample size of animals was based on our previous publication in the similar research topic. Data are presented as the mean± SEM.

Results

Rapamycin Inhibited Kir4.1/Kir5.1 in DCT

We first examined the effect of rapamycin on the basolateral K⁺ channel activity in the early part of the DCT (DCT1) using whole-cell patch clamp recording technique in the control mice (Fkbp1a^flox/flox mice). The reason for conducting experiments in DCT1 was due to that Kir4.1/Kir5.1 heterotetramer was the predominant form of K⁺ channel in the basolateral membrane.^18,21,29 Moreover, since ROMK (Kir1.1) activity was absent in the apical membrane,³⁰ Ba²⁺-sensitive whole-cell K⁺ current of DCT1 represented only Kir4.1/Kir5.1 activity. Figure 1A shows a whole-cell recording demonstrating the Ba²⁺-sensitive K⁺ currents measured with step protocol from −100 to 60 mV in DCT1 of control mice treated with vehicle or rapamycin at 1.5 μg per gram body weight (body wt) (by peritoneal injection 30 minutes before the experiment). Figure 1B is a bar graph with a scatter plot summarizing the results in three male and three female mice demonstrating that rapamycin treatment decreases Ba²⁺-sensitive whole-cell K⁺ currents from 57.3±2.9 to 29±5 pA/pF at −60 mV (male) or from 57.6±2.9 pA to 28.6±4 pA/pF at −60 mV (female), respectively. To test whether the effect of rapamycin on Kir4.1/Kir5.1 was mediated by mTORc1, we also examined the effect of rapamycin on Kir4.1/Kir5.1 in the Ks-FKBP12-KO mice. Supplemental Figure 2A shows a whole-cell recording demonstrating the Ba²⁺-sensitive K⁺ currents measured with step protocol from −100 to 60 mV in DCT1 of Ks-FKBP12-KO mice, and the results are summarized in Figure 1B. It is apparent that rapamycin treatment failed to inhibit Kir4.1/Kir5.1 activity in the DCT of Ks-FKBP12-KO mice.

Figure 1.

Inhibition of mTORc1 decreased Kir4.1/Kir5.1 activity in the DCT. (A) A whole-cell recording shows Kir4.1/Kir5.1-mediated K⁺ currents of the DCT with step protocol in the control mice (Fkbp1a^flox/flox). A symmetric 140 mM KCl solution was used for the bath and the pipette. (B) A bar graph with scatter plot summarizes values measured at −60 mV in male/female control mice or Ks-FKBP12-KO (KO) mice (each point represents one tubule and three mice for each group). Significant difference as determined by two-way ANOVA. (C) A voltage clamp recording shows inward-to-outward currents reversal potential (I-reversal potential) in the DCT treated with rapamycin in Fkbp1a^flox/flox mice. (D) The values measured in above experiments of male/female mice are summarized in a bar graph with scatter data (each point is obtained from one tubule and three mice for each group). The bath solution contains 140 mM Na⁺, and the pipette solution contains 140 mM K⁺. Significance was determined by two-way ANOVA. Rapamycin was resolved in ethanol and diluted with saline (1:500) to make a stock solution (1 ml saline containing 1.5 mg rapamycin). Rapamycin was applied by peritoneal injection at 15 μg/per 10 gram body wt). Body wt, body weight; DCT, distal convoluted tubule; FKBP12, 12kDa FK506-binding protein; K⁺, potassium; KCl, potassium chloride; KO, knockout; Ks-FKBP12-KO, kidney-specific FKBP12 knockout mice; mTORc1, mechanistic target of rapamycin complex 1.

Because Kir4.1/Kir5.1 determines the basolateral membrane potential of the DCT,¹⁸ it is conceivable that the inhibition of Kir4.1/Kir5.1 by rapamycin should decrease the negativity of DCT membrane potential. Thus, we next examined the effect of rapamycin on the reversal potential of inward-to-outward current (I-reversal-potential), which was an index of the membrane potential, in Fkbp1a^flox/flox mice (Figure 1C) and in Ks-FKBP12-KO mice (Supplemental Figure 2B). Figure 1D is a bar graph with a scatter plot summarizing the results of experiments performed in both male/female mice. It is apparent that I-reversal-potential of the DCT in rapamycin-treated mice was significantly less negative (from −61±1 to −52±1 mV in male; from 62±1 to −51±1 mV in female mice) than treated with vehicle.

Hydroperoxide Abolished the Effect of Rapamycin on Kir4.1/Kir5.1

To test whether rapamycin inhibited Kir4.1/Kir5.1 by directly inhibiting mTORc1 in the DCT, we used the single-channel recording to examine the effect of rapamycin on 40-pS K⁺ channel activity (Kir4.1/Kir5.1 heterotetramer) in isolated DCT. Figure 2A is a single-channel recording (cell attached mode) showing the effect of rapamycin (500 nM) on the basolateral 40-pS K⁺ channel, and Figure 2B summarizes the results of the experiments. Acutely addition of rapamycin decreased the 40-pS K⁺ channel activity in the DCT and decreased the NP_o from 1.52±0.1 to 0.4±0.06 (3m/3f mice), and the effect of rapamycin on the K⁺ channel was reversible. This suggests that mTORc1 in the DCT regulated the baseline activity of Kir4.1/Kir5.1. Stimulation of mTORc1 has been reported to inhibit SOD, thereby increasing the reactive oxygen species (ROS),³¹ which is known to stimulate Kir4.1/Kir5.1.³² Thus, we next examined the effect of rapamycin on the K⁺ channel in the DCT treated with 100 μM hydrogen peroxide. Figure 2C is a single-channel recording (cell attached patches) demonstrating the effect of hydroperoxide and rapamycin on the 40-pS K⁺ channel, and Figure 2D summarizes the results. Adding 100 μM hydroperoxide not only increased the 40-pS K⁺ channel NP_o (from 1.55±0.10 to 2.44±0.11) but also abolished the effect of rapamycin on the Kir4.1/Kir5.1 activity in the isolated DCT. This suggests that the mTORc1 inhibits Kir4.1/Kir5.1, at least in part, by suppressing ROS-dependent mechanism.

Figure 2.

Rapamycin decreases the 40-pS K⁺ channel activity. (A) A single-channel recording (cell attached) shows the effect of rapamycin (500 nM; by directly adding to the bath) on the basolateral 40-pS K⁺ channel (Kir4.1/Kir5.1) of the DCT. The top trace shows the time course of the experiments. Three parts of the recording indicated by numbers are extended to show the fast time resolution. The experiments were performed in the isolated DCT of three male or female mice. (B) A scatter plot summarizes the above result. (C) A single-channel recording (cell attached patch) shows the effect of H₂O₂ (100 μM) and rapamycin (500 nM) on the basolateral 40-pS K⁺ channel of the DCT. The top trace shows the time course of the experiments. Three parts of the recording indicated by numbers are extended to show the fast time resolution. The channel closed line is indicated by a dotted line plus “C” and the patch clamp voltage was held at 0 mV. (D) A scatter plot summarizes the above result. The experiments were performed in the isolated DCT of three male and three female mice and each point represents one tubule. H₂O₂, hydrogen peroxide.

Rapamycin Treatment Inhibited NCC and Increased Renal K⁺ Wasting

Since basolateral Kir4.1/Kir5.1 of the DCT is closely associated with NCC expression/activity,^19,33 we speculate that rapamycin-induced inhibition of Kir4.1/Kir5.1 in the DCT may inhibit NCC. Thus, we next examined the effect of rapamycin treatment (by peritoneal injection 30 minutes before experiments) on the abundance of phosphorylated Na-Cl cotransporter (pNCC) and total NCC (tNCC) in the control and in Ks-FKBP12-KO mice. Figure 3A and Supplemental Figure 3A show two Western blots demonstrating the abundance of pNCC and tNCC in male/female (three mice for each group) control and Ks-FKBP12-KO mice treated with rapamycin or vehicle, respectively. Figure 3B and Supplemental Figure 3B show two scatter plots summarizing the normalized band density of pNCC and tNCC for male and female samples, respectively. From the inspection of Figure 3, it is apparent that rapamycin treatment decreased the abundance of pNCC (by 39%±5%) and tNCC (by 33%±4%). Moreover, we observed that the effect of rapamycin treatment on pNCC expression was largely abolished in Ks-FKBP12-KO mice. Similar results were also observed in female mice with rapamycin treatment. The finding that acute rapamycin treatment decreased NCC and pNCC abundance in the control mice led us to test NCC activity in in vivo by measuring HCTZ (30 mg/kg body wt)–induced urinary Na⁺ excretion (E_Na). Figure 3C is a scatter plot summarizing the delta value of HCTZ-induced E_Na in male control mice treated with vehicle or rapamycin. HCTZ-induced natriuresis in rapamycin-treated control mice (2.10±0.18 μEq/min per 100 g body wt, n=7 male mice) was significantly smaller than in vehicle-treated control mice (2.88±0.26 μEq/min per 100 g body wt, n=7 male mice). By contrast, HCTZ-induced E_Na was similar in Ks-FKBP12-KO mice treated with vehicle or rapamycin. We next examined the effect of rapamycin treatment on renal K⁺ excretion using in vivo measurement of renal K⁺ excretion rate. The results are summarized in Figure 3C, showing that rapamycin treatment increased renal K⁺ excretion from 0.33±0.04 to 0.65±0.04 μEq/min per 100 g body wt (n=7). By contrast, baseline E_K was similar in Ks-FKBP12-KO mice treated with vehicle or rapamycin. Similar results were observed in female mice showing that rapamycin treatment decreased HCTZ-induced natriuresis (Supplemental Figure 4A) and increased renal K⁺ excretion (Supplemental Figure 4B).

Figure 3.

Rapamycin treatment decreases NCC abundance/activity. (A) A Western blot shows the effect of rapamycin treatment on the abundance of pNCC and tNCC in male control (Fkbp1a^flox/flox) and Ks-FKBP12-KO mice. (B) A scatter plot summarizes the normalized band density of pNCC and tNCC in mice treated with vehicle or rapamycin, respectively (two experiments were performed for each mouse with two sides of kidney sample and three male mice for each group). (C) Two scatter plots summarize the results of experiments in which the effect of rapamycin on HCTZ-induced net renal Na⁺ excretion (E_Na) and baseline renal K⁺ excretion (E_K) were measured in male control (Fkbp1a ^flox/flox) mice and in Ks-FKBP12-KO mice (seven mice for each group). Rapamycin was given by peritoneal injection at 15 μg/per 10 g body wt 30 minutes before the experiment. For measurement of urinary E_K, urine collections started 1 hour after infusion of 0.3 ml saline and a total of six collections (every 30 minutes) were performed to calculate E_K. For measurement of HCTZ-induced urinary E_Na, urine collections were performed before HCTZ application (twice) and after HCTZ (30 mg/kg body wt; four times). GAPDH, glyceraldehyde-3-phosphate dehydrogenase; HCTZ, hydrochlorothiazide; IB, immunoblotting; NCC, Na-Cl cotransporter; pNCC, phosphorylated NCC; tNCC, total NCC.

Inhibition of mTORc1 in DCT Decreased Kir4.1/Kir5.1 Activity

To further examine the role of mTORc1 in the regulation of Kir4.1/Kir5.1 and renal K⁺ excretion, we generated inducible DCT-specific RAPTOR KO mice (DCT-RAPTOR-KO) by crossing the mice expressing Slc12a3-Cre-ERT2 with Raptor^flox/flox mice. The deletion of Raptor was confirmed by quantitative RT-PCR experiments in which the Raptor mRNA expression in renal tubules was examined. Supplemental Figure 5A is a bar graph with a scatter plot summarizing quantitative analysis of Raptor mRNA to Gapdh mRNA expression ratio. The deletion of RAPTOR in the DCT was also confirmed by immunofluorescence staining showing that RAPTOR was expressed in parvalbumin-positive tubule (DCT) in Ncc-Cre-Raptor^flox/flox mice (Supplemental Figure 5B) but was almost absent in DCT-RAPTOR-KO mice (Supplemental Figure 5C). Moreover, immunohistochemistry examination shows that the expression of Kir4.1 (Supplemental Figure 6) and Kir5.1 (Supplemental Figure 7) was similar in both genotypes. After confirming the deletion of RAPTOR in DCT, we then used metabolic cage to examine the body weight, food/water intakes, urinary Na⁺ (U_Na)/K⁺ excretion rate (U_K), and plasma Na⁺ and K⁺ concentrations in male/female DCT-RAPTOR KO mice and Ncc-Cre-Raptor^flox/flox mice (Table 1). Although two genotypes did not have significant difference regarding body weight, food/water intakes, U_Na, and plasma Na⁺ concentrations, DCT-RAPTOR-KO mice had a significantly higher U_K and lower plasma K⁺ concentrations than their corresponding control mice. Thus, the deletion of RAPTOR in the DCT increased the renal K⁺ excretion and decreased plasma K⁺ concentrations.

Table 1.

Phenotypes of the distal convoluted tubule–specific regulatory-associated protein of mechanistic target of rapamycin knockout mice

Genotype and Sex (n=8)	Ncc-Cre-Raptorflox/flox (Male)	DCT-RAPTOR-KO (Male)	Ncc-Cre-Raptorflox/flox (Female)	DCT-RAPTOR-KO (Female)
Body wt, g	24.3±0.4	24.2±0.5	20.1±0.1	20.6±0.5
Food intake, g/per g body wt/24 h	0.23±0.01	0.22±0.01	0.22±0.01	0.20±0.02
Water intake, ml/per g body wt/24 h	0.30±0.03	0.28±0.03	0.33±0.02	0.30±0.03
Urine volume, ml/per g body wt/24 h	0.10±0.01	0.10±0.01	0.10±0.01	0.10±0.01
UK+ excretion, mmol/per g body wt/24 h	30.5±1.6	39.8±2.7a (P = 0.01)	28.9±2.2	35.7±1.4a (P = 0.03)
UNa+ excretion, mmol/per g body wt/24 h	10.6±0.6	10.1±0.5 (P = 0.50)	11.9±1.0	10.4±0.5 (P = 0.22)
Plasma K+, mM	4.6±0.13	3.9±0.20a (P = 0.046)	4.6±0.09	3.9±0.07a (P = 0.007)
Plasma Na+, mM	143±3	144±2	147±2	146±3

Body wt, body weight.

^aIndicates the difference is significant comparing to the corresponding control value.

We next examined Kir4.1/Kir5.1 activity in DCT1 of control and DCT-RAPTOR-KO mice on normal-K⁺ diet. In Figure 4, A and B, there are two sets of recordings showing whole-cell Kir4.1/Kir5.1-mediated K⁺ currents of DCT1 measured with step protocol (from −100 to 60 mV) and with ramp protocol (from −100 to 100 mV) in Ncc-Cre-Raptor ^flox/flox (wild-type [WT]) mice on normal-K⁺ diet and in DCT-RAPTOR KO mice, respectively. The results of the experiments in which Kir4.1/Kir5.1-mediated whole-cell K⁺ currents were measured at −60 mV are summarized in a bar graph with scatter plots (Figure 4C). It is apparent that Kir4.1/Kir5.1-mediated K⁺ currents of DCT1 in DCT-RAPTOR-KO mice (male, 39±2.3 pA/pF; female, 38.6±2.3 pA/pF, n=5 tubules) were smaller than WT mice (male, 66.5±2.1 pA/pF; female 66.8±3.8 pA/pF). We have also measured I-reversal potential of DCT1 in male/female DCT-RAPTOR-KO mice and WT mice. Figure 4D is a trace showing I-reversal potential of DCT1, and Figure 4E is a bar graph with scatter plots summarizing the results showing the mean value and each data point. We observed that I-reversal potential of DCT1 was significantly more positive in DCT-RAPTOR KO mice (male, −54±1 mV, female, −53±1 mV, n=5 tubules) than WT mice (male, −64±1 mV, female, −64±1 mV). Thus, the inhibition of DCT mTORc1 inhibited Kir4.1/Kir5.1 activity.

Figure 4.

Deletion of RAPTOR in the DCT decreased Kir4.1/Kir5.1 activity. A set of whole-cell recording shows Kir4.1/Kir5.1-mediated K⁺ currents of the DCT measured with step protocol (A) or ramp protocol (B) in Ncc-Cre-Raptor^flox/flox mice and in DCT-RAPTOR-KO mice on normal K⁺. (C) A bar graph with scatter plot summarizes the results showing the mean value and each data point measured at −60 mV in four male and four female Ncc-Cre-Raptor ^flox/flox and DCT-RAPTOR KO mice (each point represent one tubule). (D) A current/voltage curve shows the I-reversal potential of the DCT in Ncc-Cre-Raptor ^flox/flox mice and in DCT-RAPTOR KO mice. (E) A bar graph with scatter plot summarizes the mean value and each data point of I-reversal potential in four male and four female Ncc-Cre-Raptor^flox/flox and DCT-RAPTOR KO mice (each point represents one tubule). RAPTOR, regulatory-associated protein of mammalian target of rapamycin.

Inhibition of mTORc1 in DCT Decreased NCC Abundance

We next examined the abundance of NCC and type II Na-Cl-K cotransporter (NKCC2) in Ncc-Cre-Raptor^flox/flox mice and DCT-RAPTOR-KO mice on normal-K⁺ diet. In Figure 5, A and B, there are two Western blots showing the pNCC and tNCC abundance, respectively. In Figure 5C, there are two bar graphs summarizing the normalized band density in male and female mice. It is apparent that pNCC abundance was significantly lower (70%±5% of WT for male, 63%±5% of WT for female) in DCT-RAPTOR KO mice than WT mice. By contrast, the abundance of total NKCC2 and phosphorylated NKCC2 (pNKCC2) (with pT⁹⁶NKCC2 antibody,³⁴ which has been shown to recognize only pNKCC2 but not pNCC) was similar between two genotypes (Supplemental Figure 8). The notion that the deletion of RAPTOR of the DCT inhibits NCC was also confirmed by experiments in which we used in vivo measurement of HCTZ-induced natriuresis to assess NCC activity in male/female control and DCT-RAPTOR-KO mice. Figure 5D is a scatter plot summarizing the results of experiments showing each data point and mean value (including value of both male and female mice). It is apparent that HCTZ-induced renal Na⁺ excretion in nine Ncc-Cre-Raptor^flox/flox mice (four male/five female) was larger (2.5±0.23 μEq/min per 100 g body wt) than in DCT-rapamycin-insensitive companion of mTOR-KO mice (four male/five female; 1.29±0.16 μEq/min per 100 g body wt).

Figure 5.

Deletion of RAPTOR in DCT decreased pNCC abundance and NCC activity. Two sets of Western blots show the abundance of pNCC and tNCC in six male (A) and six female (B) Ncc-Cre-Raptor ^flox/flox and DCT-RAPTOR KO mice on normal-K⁺ diet, respectively. Two bar graphs with scatter plot summarize the normalized band density of pNCC/tNCC in male and female Ncc-Cre-Raptor ^flox/flox (WT) and DCT-RAPTOR-KO mice (KO; C). A scatter plot shows the HCTZ (30 mg/kg)–induced natriuresis in Ncc-Cre-Raptor^flox/flox mice and DCT-RAPTOR-KO mice (four male, circle and five female, triangle). The HCTZ-induced natriuresis was performed with in vivo urinary Na⁺ excretion method. The mean value (black square) includes values from male/female mice (D). For measurement of HCTZ-induced natriuresis, urine collections were performed before HCTZ application (twice) and after HCTZ (30 mg/kg body wt; four times). Na⁺, sodium; WT, wild-type.

Inhibition of mTORc1 in DCT Increased ENaC Activity

We also examined the expression of ENaC in male/female Ncc-Cre-Raptor^flox/flox mice and DCT-RAPTOR KO mice. Figure 6A shows αENaC abundance in male and female mice, and Figure 6B is a bar graph summarizing the normalized band density of Western blot. We observed that the abundance of cleaved αENaC was significantly increased in DCT-RAPTOR-KO mice in comparison with the control mice. However, the expression of βENaC and γENaC was similar in both genotypes (Supplemental Figure 9). We next measured the amiloride-sensitive Na⁺ currents in late-DCT/early connecting tubule and in cortical collecting duct of male control and DCT-RAPTOR-KO mice with the whole-cell recording technique. The results are summarized in Figure 6C, showing that amiloride-sensitive Na⁺ currents in late-DCT/early connecting tubule (12.8±1.3 pA/pF, n=4 tubules) was higher in DCT-RAPTOR-KO mice than the control mice (7.1±0.3 pA/pF). In addition, amiloride-sensitive Na⁺ currents of the cortical collecting duct tends to be higher (4.3±0.4 pA/pF, n=4 tubules) in DCT-RAPTOR-KO mice than the control mice (3.3±0.3 pA/pF), although the effect was NS (P = 0.07). Thus, patching clamp data were consistent with immunoblot results, supporting that ENaC activity in DCT-RAPTOR-KO mice was increased. In summary, the inhibition of mTORc1 in the DCT decreased Kir4.1/Kir5.1 activity, reduced pNCC abundance, increased ENaC activity/expression, and decreased plasma K⁺ concentration due to K⁺ wasting in the kidney.

Figure 6.

The expression of cleaved αENaC is increased in DCT-RAPTOR-KO mice. (A) A Western blot shows the expression of αENaC in male/female Ncc-Cre-Raptor^flox/flox (WT) and DCT-RAPTOR-KO (KO) mice on normal-K⁺ diet. (B) A bar graph with scatter plot summarizes the normalized band density of full-length αENaC and cleaved αENaC abundance in male and female mice on normal K⁺ (three mice for each group and each point represents one kidney sample). (C) A bar graph with scatter plot summarizes the results of experiments in which the amiloride (10 μm)–sensitive Na⁺ currents were measured in DCT2/CNT and CCD at −60 mV from four male WT and four KO mice with the whole-cell voltage clamp. (D) A cell scheme illustrates the role of ROS in mediating the effect of mTORc1 on regulation Kir4.1/Kir5.1 and NCC of the DCT. (E) A cell model illustrates renal K⁺ excretion in DCT of WT mice or DCT-RAPTOR-KO mice, respectively. Solid and dotted lines indicate an enhanced and an inhibited signaling pathway. αENaC, alpha subunit of epithelial sodium channel; CCD, cortical collecting duct; CNT, connecting tubule; DCT2, late distal convoluted tubule; ENaC, epithelial Na⁺ channels; ROMK, renal outer medullary K⁺ channel (Kir1.1); ROS, reactive oxygen species; SPAK, ste20 proline alanine-rich kinase; V, cell voltage; WNK, with no lysine kinase.

Discussion

This study demonstrates the role of mTORc1 in the regulation of the membrane transport of the DCT since the deletion of RAPTOR in the DCT inhibits the basolateral K⁺ channel activity and NCC activity. Both mTORc1 and mechanistic target of rapamycin complex 2 play a role in regulating renal K⁺ excretion. However, the abundance of rapamycin-insensitive companion of mTOR was increased during increased dietary K⁺ intake,^35,36 whereas we observed that K⁺ restriction increased the RAPTOR abundance (Lin's unpublished observation). This suggests that mechanistic target of rapamycin complex 2 plays a role in stimulating renal K⁺ excretion, whereas mTORc1 plays a role in preserving K⁺ and inhibiting renal K⁺ excretion. The role of mTORc1 in the regulation of renal membrane transport from thick ascending limb to DCT and collecting duct has been previously explored by Grahammer et al.²⁴ They have used KspCre mice to constitutively delete RAPTOR from thick ascending limb to whole distal nephrons (Raptor^fl/fl-Ksp-Cre). The main phenotypes of Raptor^fl/fl-Ksp-Cre mice were polyuria due to defective urinary concentrating mechanism and slowly progressive kidney fibrosis. However, it was not reported whether the function of K⁺ channel and NCC was compromised in the DCT of Raptor^fl/fl-Ksp-Cre mice. Since the deletion of RAPTOR in Raptor^fl/fl-Ksp-Cre mice was constitutively induced, it is possible that Raptor^fl/fl-Ksp-Cre mice could have a significant tubule remodeling process, thereby compensating the membrane transport of the DCT. Indeed, transcriptional profiles demonstrated that genes related to cell division, DNA repair, and cell cycles were upregulated in Raptor^fl/fl-Ksp-Cre mice.²⁴ Thus, the remodeling process in Raptor^fl/fl-Ksp-Cre mice could mask the deletion of RAPTOR-induced phenotypes in the DCT. By contrast, the deletion of RAPTOR in Slc12a3-Cre-ERT2-Raptor^flox/flox mice was inducible in the adult mice. Thus, we could determine the role of mTORc1 of the DCT in regulation of Kir4.1/Kir5.1 and NCC.

The first main finding of this study is to demonstrate the role of mTORc1 in regulating the basolateral Kir4.1/Kir5.1 of the DCT. This notion is supported by several lines of evidence: First, rapamycin treatment decreased Kir4.1-Kir5.1-mediated K⁺ currents and depolarized DCT membrane potential in the Fkbp1a^flox/flox mice. Since this effect was absent in Ks-FKBP12-KO mice, the effect of rapamycin on Kir4.1/Kir5.1 was due to the inhibition of mTORc1.²⁶ Second, Kir4.1/Kir5.1-mediated K⁺ currents were lower, and the membrane potentials of the DCT were less negative in Ks-RAPTOR-KO mice than the corresponding WT mice. Third, rapamycin inhibited the basolateral 40-pS K⁺ channel activity in the isolated DCT. This finding also suggests strongly that mTORc1 plays a direct role in regulating Kir4.1/Kir5.1 in the DCT. The mechanism by which mTORc1 regulates Kir4.1/Kir5.1 may be mediated by stimulating ROS production since the inhibition of mTORc1 with rapamycin did not inhibit the 40-pS K⁺ channel in the presence of hydrogen peroxide, which robustly stimulated K⁺ channel activity. In this regard, it has been reported that stimulation of mTORc1 increased ROS production by inhibiting type 1 SOD.³¹

The second main finding of the study is to demonstrate that the inhibition of mTORc1 decreased the NCC expression and activity. First, rapamycin treatment inhibited the abundance of both pNCC and tNCC, an effect that was absent in Ks-FKBP12-KO mice. Second, the expression of pNCC in DCT-RAPTOR-KO mice was significantly lower than the corresponding WT mice. Third, HCTZ-induced natriuresis was smaller in DCT-RAPTOR-KO mice or in the mice treated with rapamycin than the corresponding control mice. The effect of mTORc1 on pNCC should be the result of inhibiting mTORc1 in the DCT because the expression of pNKCC2 was the same between two genotypes. Although we could not completely exclude the possibility that mTORc1 may be able to directly regulate NCC expression/activity, it is likely that the effect of mTORc1 on NCC is, at least in part, achieved by stimulating the basolateral K⁺ channels. The mechanism by which the basolateral Kir4.1/Kir5.1 in the DCT regulates NCC activity depends on Cl⁻-sensitive with no lysine kinase (WNK).^19,37,38 A large body of evidence has convincingly demonstrated that WNK plays a key role in regulating NCC expression/activity. Gain-of-function mutations of WNK1 and WNK4 are responsible for causing familial hyperkalemic hypertension.³⁹ WNK1 and WNK4 are Cl⁻-sensitive kinases with high intracellular Cl⁻ concentration [Cl⁻_i] inhibiting and low [Cl⁻_i] activating them by regulating autoinhibition.³⁷ Because WNK activity regulates ste20-proline-alanine-rich kinase (SPAK) and oxidative stress responsive kinase-1, two kinases responsible for the stimulation of NCC activity,^40–46 high [Cl⁻_i] levels should also inhibit SPAK/oxidative stress responsive kinase-1, decreasing NCC phosphorylation.

Figure 6D is a cell scheme illustrating the role of mTORc1 regulates NCC activity by Kir4.1/Kir5.1-dependent mechanism. Since Kir4.1/Kir5.1 is the only type of K⁺ channel in the DCT for determining the cell membrane potential,^16,21 changes in Kir4.1/Kir5.1 activity should affect DCT membrane potential, which provides the driving force for electrogenic Cl⁻ exit across the basolateral membrane.¹⁹ Indeed, previous study confirmed that the inhibition of Kir4.1/Kir5.1 channels increased the intracellular Cl⁻ concentrations.⁴⁷ Thus, the stimulation of the basolateral K⁺ channels in the DCT facilitates Cl⁻ exit across the basolateral membrane, thereby stimulating WNK/SPAK and NCC. The notion that Kir4.1 activity in the DCT determines NCC activity and expression by regulating [Cl⁻_i] was also supported by the finding that changes in the intracellular Cl⁻ concentration induced by membrane voltage are responsible for the inhibition of NCC activity in DCT cells expressing loss function of Kir4.1 mutants.¹⁹ Thus, our study has demonstrated that mTORc1 plays an essential role in maintaining baseline activity of Kir4.1/Kir5.1 in the DCT and NCC. Since Kir4.1/Kir5.1 activity of the DCT and NCC plays an important role in regulating K⁺ excretion in the kidney, the inhibition of mTORc1 in the DCT is expected to have a significant effect on the renal K⁺ excretion and K⁺ homeostasis. This speculation is confirmed by the finding that DCT-RAPTOR-KO mice had a higher K⁺ excretion rate and lower plasma K⁺ concentrations than the corresponding WT mice.

Figure 6E is a cell model illustrating the regulation of renal K⁺ excretion with intact mTORc1 or during the inhibition of mTORc1. We believe that mTORc1 plays a key role in maintaining the baseline activity of basolateral Kir4.1/Kir5.1 of the DCT, which determines the baseline activity of NCC. Because NCC plays an important role in the regulation of renal K⁺ excretion by determining the Na⁺ delivery to the distal nephrons,^9,10,48 mTORc1 should play a significant role in the regulation of baseline K⁺ excretion in the kidney. Thus, decreased NCC activity in DCT-RAPTOR-KO mice should increase Na⁺ delivery, thereby enhancing ENaC-dependent or flow-dependent K⁺ excretion in the aldosterone-sensitive distal nephron.^13,49–51 Moreover, the inhibition of mTORc1 in the DCT is expected to increase ENaC activity. Indeed, the amiloride-sensitive Na⁺ currents in the late DCT were larger in DCT-RAPTOR-KO mice than the corresponding WT mice. Moreover, the finding that cleaved αENaC expression was higher in DCT-RAPTOR-KO mice than the corresponding WT mice suggests the possibility that ENaC activity in the collecting duct should also be increased as a compensative action of decreased NCC1 activity. This notion was also suggested by the finding that ENaC activity in the collecting duct tends to increase in DCT-RAPTOR-KO mice, although the effect was NS. Because DCT-RAPTOR-KO mice had a higher ENaC activity and a lower NCC expression than the WT mice, renal K⁺ excretion was larger in the DCT-RAPTOR-KO mice than WT mice. This could explain why DCT-RAPTOR-KO mice have a lower plasma K⁺ level than the control mice. Although NCC expression in DCT-RAPTOR-KO mice is lower, the net urinary Na⁺ excretion was not significantly higher than the control mice. In addition to high ENaC activity in DCT-RAPTOR-KO mice, low NCC activity may also stimulate electroneutral Na⁺ absorption in the intercalated cells, thereby compensating the effect of low NCC activity on renal Na⁺ absorption.⁵² This study has translational significance because hypokalemia is one of the side effects of using rapamycin, which is approved for treatment of cancer, and also for use in preventing rejection after organ transplantation. Thus, understanding the mechanism by which inhibition of mTORc1 induces hypokalemia is significant for preventing hypokalemia. Our study demonstrates the novel role of mTORc1 of the DCT in regulating baseline renal K⁺ excretion by determining the baseline activity of basolateral Kir4.1/Kir5.1. Thus, rapamycin-induced inhibition of mTORc1 stimulates renal K⁺ excretion and induces hypokalemia due to inhibiting Kir4.1/Kir5.1 and NCC. In conclusion, mTORc1 plays a significant role in determining the baseline activity of Kir4.1/Kir5.1 and NCC, and mTORc1 is indispensable for regulating K⁺ excretion in the kidney.

Disclosures

Disclosure forms, as provided by each author, are available with the online version of the article at http://links.lww.com/JSN/F242.

Funding

W.-H. Wang: National Institute of Diabetes and Digestive and Kidney Diseases (DKR01 133220). D.-H. Lin: National Institute of Diabetes and Digestive and Kidney Diseases (DKR01 136491). Y. Xiao: Heilongjiang Province Natural Science Foundation (LH2022c105) and Research Foundation of Qiqihar Medical College (2022-ZDPY-009)

Author Contributions

Conceptualization: Dao-Hong Lin, Wen-Hui Wang.

Data curation: Xin-Peng Duan, Dao-Hong Lin, Wen-Hui Wang, Yu Xiao, Cheng-Biao Zhang.

Formal analysis: Xin-Peng Duan, Dao-Hong Lin, Wen-Hui Wang, Yu Xiao.

Funding acquisition: Dao-Hong Lin.

Investigation: Xin-Peng Duan, Dao-Hong Lin, Yu Xiao, Cheng-Biao Zhang.

Methodology: Xin-Peng Duan, Dao-Hong Lin, Yu Xiao.

Project administration: Dao-Hong Lin, Wen-Hui Wang.

Resources: Dao-Hong Lin, Wen-Hui Wang.

Supervision: Dao-Hong Lin, Wen-Hui Wang.

Validation: Xin-Peng Duan, Dao-Hong Lin, Wen-Hui Wang, Yu Xiao, Cheng-Biao Zhang.

Visualization: Xin-Peng Duan, Dao-Hong Lin, Wen-Hui Wang, Yu Xiao.

Writing – original draft: Dao-Hong Lin, Wen-Hui Wang.

Writing – review & editing: Xin-Peng Duan, Dao-Hong Lin, Wen-Hui Wang, Yu Xiao, Cheng-Biao Zhang.

Data Sharing Statement

All data are available in the main text or the Supplemental Material. Derived data supporting the findings of this study are available from the corresponding author on request.

Supplemental Material

This article contains the following supplemental material online at http://links.lww.com/JSN/F241.

Supplemental Methods

Supplemental Table 1. Primers for genotyping.

Supplemental Table 2. Information of antibodies used in experiments.

Supplemental Figure 1. FKBP12 was absent in doxycycline-treated mice.

Supplemental Figure 2. The effect of rapamycin on Kir4.1/Kir5.1 in Ks-FKBP12-KO mice.

Supplemental Figure 3. Rapamycin treatment decreases NCC abundance.

Supplemental Figure 4. Rapamycin treatment increased renal K⁺ excretion.

Supplemental Figure 5. RAPTOR is expressed in the DCT.

Supplemental Figure 6. Immunohistochemistry staining of Kir4.1.

Supplemental Figure 7. Immunohistochemistry staining of Kir5.1.

Supplemental Figure 8. Expression of NKCC2 in DCT-RAPTO-KO mice.

Supplemental Figure 9. Expression of βENaC and γENaC in DCT-RAPTOR-KO mice.