Glucose/Fructose Delivery to the Distal Nephron Activates the Sodium-Chloride Cotransporter via the Calcium-Sensing Receptor

Abstract

Background: The calcium-sensing receptor (CaSR) in the distal convoluted tubule (DCT) activates the NaCl cotransporter (NCC). Glucose acts as a positive allosteric modulator of the CaSR. Under physiologic conditions, no glucose is delivered to the DCT, and fructose delivery depends on consumption. We hypothesized that glucose/fructose delivery to the DCT modulates the CaSR in a positive allosteric way, activating the WNK4-SPAK-NCC pathway and thus increasing salt retention.

Methods: We evaluated the effect of glucose/fructose arrival to the distal nephron on the CaSR-WNK4-SPAK-NCC pathway using HEK-293 cells, C57BL/6 and WNK4-knockout mice, ex vivo perfused kidneys, and healthy humans.

Results: HEK-293 cells exposed to glucose/fructose increased SPAK phosphorylation in a WNK4- and CaSR-dependent manner. C57BL/6 mice exposed to fructose or a single dose of dapagliflozin to induce transient glycosuria showed increased activity of the WNK4-SPAK-NCC pathway. The calcilytic NPS2143 ameliorated this effect, which was not observed in WNK4-KO mice. C57BL/6 mice treated with fructose or dapagliflozin showed markedly increased natriuresis after thiazide challenge. Ex vivo rat kidney perfused with glucose above the physiologic threshold levels for proximal reabsorption showed increased NCC and SPAK phosphorylation. NPS2143 prevented this effect. In healthy volunteers, cinacalcet administration, fructose intake, or a single dose of dapagliflozin increased SPAK and NCC phosphorylation in urinary extracellular vesicles.

Conclusions: Glycosuria or fructosuria was associated with increased NCC, SPAK, and WNK4 phosphorylation in a CaSR-dependent manner.

Graphical abstract

Figure 1.

Extracellular glucose promotes SPAK phosphorylation in a CaSR-WNK4–dependent manner in HEK-293 cells. (A) Representative immunoblots of cells transfected with hSPAK-GFP-HA, mWNK4-HA, and hCaSR and exposed to different concentrations of extracellular glucose or to 200 nM R-568 for 30 minutes. (B) Densitometric analysis of experiments as follows: 0 mM glucose, n=10; 5.5 mM glucose, n=9; 25 mM glucose, n=10; 0 mM glucose plus R-568, n=9. (C) Representative immunoblots of cells transfected with hSPAK-GFP-HA and hCaSR with or without mWNK4-HA and exposed to different concentrations of extracellular glucose or to 200 nM R-568 for 30 minutes. (D) Densitometric analysis of experiments as follows: cells expressing hSPAK-GFP-HA, hCaSR, and mWNK4-HA with 0 mM glucose, n=5; 5.5 mM glucose, n=5; 25 mM glucose, n=5 or 0 mM glucose plus R-568, n=5; and cells expressing only hSPAK-GFP-HA and hCaSR with 0 mM glucose, n=6; 5.5 mM glucose, n=5; 25 mM glucose, n=6; 0 mM glucose plus R-568, n=5. (E) Representative immunoblots of cells transfected with hSPAK-GFP-HA and mWNK4-HA with or without hCaSR and exposed to different concentrations of extracellular glucose or to 200 nMR-568 for 30 minutes. (F) Densitometric analysis of experiments as follows: cells expressing hSPAK-GFP-HA, mWNK4-HA, and hCaSR with 0 mM glucose, n=5; 5.5 mM glucose, n=5; 25 mM glucose, n=4; or 0 mM glucose plus R-568, n=4; and cells expressing only hSPAK-GFP-HA and mWNK4-HA with 0 mM glucose, n=5; 5.5 mM glucose, n=5; 25 mM glucose, n=6; 0 mM glucose plus R-568, n=3. (G) Representative immunoblots of cells transfected with hSPAK-GFP-HA, mWNK4-HA, and hCaSR or hCaSR-R185Q and exposed to different concentrations of extracellular glucose for 30 minutes, in the absence or presence of the calcilytic NPS2143 (300 nM). (H) Densitometric analysis of experiments as follows: cells expressing hSPAK-GFP-HA, mWNK4-HA, and hCaSR with 0 mM glucose, n=7; 5.5 mM glucose, n=5; 25 mM glucose, n=6; or cells expressing hSPAK-GFP-HA, mWNK4-HA, and hCaSR R185Q with 0 mM glucose, n=7; 5.5 mM glucose, n=6; 25 mM glucose, n=7; or cells expressing hSPAK-GFP-HA, mWNK4-HA, and hCaSR with 0 mM glucose plus NPS2143, n=5; 5.5 mM glucose plus NPS2143, n=4; 25 mM glucose plus NPS2143, n=5. For the quantitative analysis, in each experiment, groups with hSPAK-GFP-HA, mWNK4-HA, and hCaSR exposed to 0 mM of extracellular glucose were arbitrarily set to one and the corresponding groups were normalized accordingly (fold change). Data shown in bars represent mean±SEM. Data were analyzed using ANOVA with Bonferroni multiple comparison. *P<0.05, **P<0.01, ****P<0.0001. E, empty lane; GFP, green fluorescent protein; V, empty vector (pCDNA 3.1 as transfection control).

Figure 2.

Extracellular fructose promotes SPAK phosphorylation in a CaSR-WNK4–dependent manner in HEK-293 cells. (A) Representative immunoblots of cells transfected with hSPAK-GFP-HA, mWNK4-HA, and hCaSR and exposed to different concentrations of extracellular fructose or to 200 nM R-568 for 30 minutes. (B) Densitometric analysis of experiments as follows: 0 mM fructose, n=6; 5.5 mM fructose, n=4; 25 mM fructose, n=6; 0 mM fructose plus R-568, n=5. (C) Representative immunoblots of cells transfected with hSPAK-GFP-HA and hCaSR, with or without mWNK4-HA, and exposed to different concentrations of extracellular fructose or to 200 nM R-568 for 30 minutes. (D) Densitometric analysis of experiments as follows: cells expressing hSPAK-GFP-HA, hCaSR, and mWNK4-HA with 0 mM fructose, n=3; 5.5 mM fructose, n=3; 25 mM fructose, n=3; or 0 mM fructose plus R-568, n=3; and cells expressing only hSPAK-GFP-HA and hCaSR with 0 mM fructose, n=3; 5.5 mM fructose, n=5; 25 mM fructose, n=6; or 0 mM fructose plus R-568, n=3. (E) Representative immunoblots of cells transfected with hSPAK-GFP-HA and mWNK4-HA, with or without hCaSR, and exposed to different concentrations of extracellular fructose or to 200 nM R-568 for 30 minutes. (F) Densitometric analysis of experiments as follows: cells expressing hSPAK-GFP-HA, mWNK4-HA, and hCaSR with 0 mM fructose, n=6; 5.5 mM fructose, n=5; 25 mM fructose, n=5; or 0 mM fructose plus R-568, n=4; and cells expressing only hSPAK-GFP-HA and mWNK4-HA with 0 mM fructose, n=6; 5.5 mM fructose, n=7; 25 mM fructose, n=7; or 0 mM fructose plus R-568, n=5. (G) Representative immunoblots of cells transfected with hSPAK-GFP-HA, mWNK4-HA, and hCaSR or hCaSR-R185Q and exposed to different concentrations of extracellular fructose for 30 minutes, in the absence or presence of the calcilytic NPS2143 (300 nM). (H) Densitometric analysis of experiments as follows: cells expressing hSPAK-GFP-HA, mWNK4-HA, and hCaSR with 0 mM fructose, n=8; 5.5 mM fructose, n=7; 25 mM fructose, n=7; or cells expressing hSPAK-GFP-HA, mWNK4-HA, and hCaSR R185Q with 0 mM fructose, n=10; 5.5 mM fructose, n=9; 25 mM fructose, n=11; or cells expressing hSPAK-GFP-HA, mWNK4-HA, and hCaSR with 0mM fructose plus NPS2143, n=7; 5.5 mM fructose plus NPS2143, n=7; 25 mM fructose plus NPS2143, n=7. For the quantitative analysis, in each experiment, groups with hSPAK-GFP-HA, mWNK4-HA, and hCaSR exposed to 0 mM of extracellular fructose were arbitrarily set to one and the corresponding groups were normalized accordingly (fold change). Data shown in bars represent mean±SEM. Data were analyzed by using ANOVA with Bonferroni multiple comparison. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001. E, empty lane; GFP, green fluorescent protein; V, empty vector (pCDNA 3.1 as transfection control).

Figure 3.

Glucosuria is associated with WNK4-SPAK-NCC phosphorylation via CaSR in vivo. (A) C57BL/6 mice were treated with vehicle, dapagliflozin (Dapa; IP, 1 mg/kg), or dapagliflozin (IP, 1 mg/kg) along with the calcilytic NPS2143 (oral gavage, 30 mg/kg). Three hours later, kidneys were extracted and processed for immunoblot. Shown are representative immunoblots of renal protein samples performed with the indicated antibodies. Densitometric analysis of (B) pNCC, (C) phosphorylated WNK4 (pWNK4), and (D) pSPAK as follows: vehicle, n=8; Dapa, n=9; Dapa plus NPS2143, n=10. (E) Urinary glucose measurement after 3 hours of dapagliflozin injection. Vehicle, n=7; Dapa, n=9; and Dapa plus NPS2143, n=10. (F) WNK4-KO mice or their WT littermates were subjected to vehicle or dapagliflozin. Shown are representative immunoblots performed with kidney protein samples. (G) Densitometric analysis of pSPAK as follows: WT vehicle, n=5; WT Dapa, n=7; WNK4-KO vehicle, n=5; WNK4-KO Dapa, n=6. (H) Urinary glucose measurement after 3 hours of dapagliflozin injection. WT vehicle, n=5; WT Dapa, n=7; WNK4-KO vehicle, n=5; WNK4-KO Dapa, n=6. For the quantitative analysis, WT mice exposed to vehicle were arbitrarily set to one and the corresponding groups were normalized accordingly (fold change). Data shown in bars are mean±SEM. Data were analyzed by using ANOVA with Bonferroni multiple comparison. (I) Effect of dapagliflozin administration on HCTZ-induced changes in sodium ion (Na⁺) excretion in WT mice. HCTZ was given by IP injection (25 mg/kg) after 3 hours of dapagliflozin injection. Urinary Na⁺ excretion was monitored at baseline; after 3 hours of dapagliflozin injection; and at 3, 6, and 24 hours after HCTZ. Vehicle, n=7; Dapa, n=8; Dapa plus HCTZ, n=8; HCTZ, n=7. Data are presented as mean±SEM and were analyzed by using two-way ANOVA with Bonferroni multiple comparison. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Figure 4.

Fructose intake is associated with NCC phosphorylation via CaSR-WNK4-SPAK in vivo. (A) C57BL/6 mice were administered vehicle, 20% fructose dissolved in the drinking water, or 20% fructose in the drinking water along with the calcilytic NPS2143 (30 mg/kg) through oral gavage. Three hours later, kidneys were extracted and processed for immunoblot. Representative immunoblots showing the effect of fructosuria-inducing maneuver on NCC, SPAK, and WNK4 phosphorylation. Densitometric analysis of (B) pNCC, (C) phosphorylated WNK4 (pWNK4), (D) and pSPAK as follows: vehicle, n=7; fructose, n=9; fructose plus NPS2143, n=9. (E) Urinary fructose measurement after 3 hours of fructose intake. Vehicle, n=7; 20% oral fructose ad libitum, n=9; 20% oral fructose ad libitum plus NPS2143, n=9. (F) WNK4-KO mice or their WT littermates were subjected to vehicle or 20% oral fructose ad libitum. Shown are representative immunoblots performed with kidney protein samples. (G) Densitometric analysis of pSPAK as follows: WT vehicle, n=4; WT fructose, n=4; WNK4-KO vehicle, n=7; WNK4-KO fructose, n=8. (H) Urinary fructose measurement after 3 hours of fructose intake in WT and WNK4-KO mice. WT vehicle, n=4; WT 20% oral fructose ad libitum, n=4; WNK4-KO vehicle, n=7; 20% WNK4-KO oral fructose ad libitum, n=8. For the quantitative analysis, WT mice exposed to vehicle were arbitrarily set to one and the corresponding groups were normalized accordingly (fold change). Data shown in bars are mean±SEM. Data were analyzed by using ANOVA with Bonferroni multiple comparison. (I) Effect of fructose intake on HCTZ-induced changes in sodium ion (Na⁺) excretion in WT mice. Urinary Na⁺ excretion was monitored at baseline; after 3 hours of fructose intake; and at 3, 6, and 24 hours after the HCTZ. HCTZ was given by IP injection (25 mg/kg) after 3 hours of fructose intake. Vehicle, n=10; 20% oral fructose ad libitum, n=9; 20% oral fructose ad libitum plus HCTZ, n=10; HCTZ, n=8. Data are presented as mean±SEM and were analyzed by using two-way ANOVA with Bonferroni multiple comparison. *P<0.05, **P<0.01, ***P<0.001.

Figure 5.

Luminal glucose is associated with SPAK and NCC phosphorylation ex vivo. (A) Rat kidneys were extracted and perfused with solutions containing 0, 4, 11.1, or 25 mM glucose at a rate of 0.60 ml/min. Shown are representative immunoblots performed with kidney protein samples. Each column represents one kidney. (B and C) Results from densitometric analysis of blots represented in (A). (D and E) Kidneys were perfused with physiologic solutions containing different concentrations of glucose. The calcilytic NPS2143 was added to the solution in the indicated groups. Representative immunoblots are shown. (F and G) Densitometric analysis of immunoblots represented in (D and E). 0 mM glucose, n=4; 0 mM glucose plus NPS2143, n=3; 4 mM glucose, n=3; 4 mM glucose plus NPS2143, n=4; 11.1 mM glucose, n=4; 11.1 mM glucose plus NPS2143, n=3; 25 mM glucose, n=3; 25 mM glucose plus NPS2143, n=4. For the quantitative analysis, rat kidneys perfused with 0 mM glucose to vehicle were arbitrarily set to one and the corresponding groups were normalized accordingly (fold change). Data shown in bars are mean±SEM. Data were analyzed using ANOVA with Bonferroni multiple comparison. ***P<0.001, ****P<0.0001.

Figure 6.

Glucosuria or fructosuria-inducing interventions are associated with increased NCC-SPAK phosphorylation in UEVs of healthy volunteers. (A) Schematic representation of the prospective intervention protocol designed to study UEVs in healthy volunteers. Three randomly ordered interventions were performed on four healthy male volunteers. In all interventions (Int.), subjects were asked to drink 250 ml of water hourly to ensure enough urine output. Urine samples were collected every hour for 4 hours. After the first hour, which was considered time zero, the treatment was started, which could consist of (1) a single dose of 30 mg oral cinacalcet, (2) a single dose of 10 mg oral dapagliflozin, or (3) hourly 5% oral fructose. A baseline evaluation was also preformed where only water was provided to evaluate the steady-state behavior. All subjects were exposed to the four maneuvers in random order (three treatments and one baseline evaluation), with at least 15 days between each one. All subjects started the interventions with the baseline evaluation (only water intake). Immunoblots were performed with the UEVs extracted from each collection. (B) Representative results of blots performed with samples of volunteer 1, volunteer 2, volunteer 3, and volunteer 4 are presented. Results from densitometric analysis of pNCC/NCC and pSPAK/SPAK are shown in panels C and D for cinacalcet treatment, panels E and F for dapagliflozin treatment, and panels G and H for fructose treatment. For the quantitative analysis, pNCC/NCC or pSPAK/SPAK at 0 hour was arbitrarily set to one, and the corresponding groups were normalized accordingly (fold change). Data shown in bars are mean±SEM. Data were analyzed by using two-way ANOVA with Bonferroni multiple comparison. *P<0.05, **P<0.01, ***P<0.001, ****P<0.0001.

Figure 7.

Proposed model for the activation of CaSR-WNK4-SPAK-NCC in the presence of glucose or fructose. The arrival of glucose or fructose to the distal convoluted tubule exerts a positive allosteric effect type II, making the CaSR more sensitive to Ca²⁺, causing activation of WNK4-SPAK-NCC, with a consequent increase in salt retention. Cl⁻, chloride ion; Na⁺, sodium ion; P, phosphorylation.