The Na+-dependent chloride-bicarbonate exchanger SLC4A8 mediates an electroneutral Na+ reabsorption process in the renal cortical collecting ducts of mice

Abstract

Regulation of sodium balance is a critical factor in the maintenance of euvolemia, and dysregulation of renal sodium excretion results in disorders of altered intravascular volume, such as hypertension. The amiloride-sensitive epithelial sodium channel (ENaC) is thought to be the only mechanism for sodium transport in the cortical collecting duct (CCD) of the kidney. However, it has been found that much of the sodium absorption in the CCD is actually amiloride insensitive and sensitive to thiazide diuretics, which also block the Na-Cl cotransporter (NCC) located in the distal convoluted tubule. In this study, we have demonstrated the presence of electroneutral, amiloride-resistant, thiazide-sensitive, transepithelial NaCl absorption in mouse CCDs, which persists even with genetic disruption of ENaC. Furthermore, hydrochlorothiazide (HCTZ) increased excretion of Na+ and Cl- in mice devoid of the thiazide target NCC, suggesting that an additional mechanism might account for this effect. Studies on isolated CCDs suggested that the parallel action of the Na+-driven Cl-/HCO3- exchanger (NDCBE/SLC4A8) and the Na+-independent Cl-/HCO3- exchanger (pendrin/SLC26A4) accounted for the electroneutral thiazide-sensitive sodium transport. Furthermore, genetic ablation of SLC4A8 abolished thiazide-sensitive NaCl transport in the CCD. These studies establish what we believe to be a novel role for NDCBE in mediating substantial Na+ reabsorption in the CCD and suggest a role for this transporter in the regulation of fluid homeostasis in mice.

Figure 1. Pharmacological characterization of transepithelial transport of Na⁺, K⁺, and Cl^– in collecting ducts isolated from wild-type mice.

(A) Effects of amiloride (10^–5 M) and HCTZ (10^–4 M) on Na⁺, Cl^–, and K⁺ transepithelial fluxes and on transepithelial voltage in CCDs isolated from Na⁺-restricted mice. J_Na, rate of Na⁺ absorption; J_Cl, rate of Cl^– absorption; V_te, transepithelial voltage; J_K, rate of K⁺ secretion. Statistical significance was assessed by ANOVA; comparisons between groups were tested by Bonferroni’s post-hoc test. n = 5 in each group, *P < 0.05 versus control, ^#P < 0.05 versus low Na⁺; ^†P < 0.05 versus low Na⁺ amiloride (amil). (B) Effects of HCTZ alone (10^–4 M) on Na⁺, Cl^–, and K⁺ transepithelial fluxes and on transepithelial voltage in CCDs isolated from Na⁺-restricted mice. Statistical significance was assessed by 2-tailed unpaired Student’s t test. n = 5 in each group; *P < 0.05 versus control (low Na).

Figure 2. Analyses of J_Na, J_Cl, and J_K and V_te in CCDs isolated from collecting duct–specific ENaC–KO mice maintained on a Na⁺-depleted diet.

The control group consists of littermate mice floxed for α-ENaC but negative for the HoxB7-Cre transgene, as detailed elsewhere (22). The control group was kept on a normal Na⁺ diet to provide the zero baseline values for each variable. Statistical significance was assessed by 2-tailed unpaired Student’s t test. n = 5 in each group; *P < 0.05.

Figure 3. Effects of amiloride (10^–5 M) and HCTZ (10^–4 M) on Na⁺ and Cl^– transepithelial fluxes in CCDs isolated from Ncc^+/+ and Ncc^–/– mice fed a Na⁺-replete diet.

Statistical significance was assessed by ANOVA; comparisons between groups were tested by Bonferroni’s post-hoc test. n = 5 in each group; *P < 0.05 versus Ncc^+/+; ^#P < 0.05 versus Ncc^–/–; ^†P < 0.05 versus Ncc^–/– amil.

Figure 4. Effects of HCTZ on urinary excretion of Na⁺ and Cl^– in Ncc^+/+ and Ncc^–/– mice.

One single dose of HCTZ (50 mg/kg body weight) or vehicle was administered intraperitoneally to Ncc^+/+ and Ncc^–/– mice. Urine samples were collected from 0 to 6 and from 6 to 12 hours after injection to measure urinary Na⁺ (top panels) or Cl^– (bottom panels) excretion. Results are expressed as the ratio to urinary creatinine. Statistical significance was assessed by 2-tailed Student’s unpaired t test. n = 5 in WT groups and n = 8 in KO groups; *P < 0.05, **P < 0.01 versus vehicle.

Figure 5. Effects of bicarbonate on transepithelial fluxes of Na⁺ and Cl^– in CCDs isolated from Ncc^–/– mice.

CCDs from Ncc^–/– mice were either perfused in CO₂/HCO₃^–-containing solutions or in CO₂/HCO₃^–-free (no bicarbonate [no bicarb]) buffer. Statistical significance was assessed by 2-tailed Student’s unpaired t test. n = 5 in each group; *P < 0.001 versus control.

Figure 6. Detection and functional characterization of NDCBE in collecting ducts isolated from wild-type mice fed a Na⁺-depleted diet.

(A) Na⁺ dependence of pH_i changes in intercalated cells of CCDs isolated from mice fed a Na⁺-depleted diet. Traces are the average of pH_i changes recorded when luminal Na⁺ was removed from, and then readded to the perfusate. Upper-left panel: Both Cl^– and HCO₃^–/CO₂ were present in the extracellular fluid; mean starting pH_i (immediately before luminal Na⁺ removal) was 7.03 ± 0.04. Upper right: extracellular Cl^– was absent; mean starting pH_i was 7.03 ± 0.05. Lower left: extracellular HCO₃^–/CO₂ was absent; mean starting pH_i was 7.09 ± 0.11. Lower right: initial rates of pH_i changes during exposure to different solutions. Values are mean ± SEM, with number of tubules in parentheses. Statistical significance was tested by ANOVA followed by Bonferroni’s post-hoc test. **P < 0.01 versus all other groups. o, outside. (B) Western blot analyses of NDCBE in the mouse kidney and CCD. Western blot analyses of 50 μg membrane fraction proteins from renal cortex isolated from Ndcbe^–/– mice or wild-type mice or of proteins from 100 CCDs (corresponding to ~25 mm of tubule) isolated from wild-type mice. NDCBE was detected in the renal cortex or isolated tubules of wild-type but not of Ndcbe ^–/– mice. Note that the apparent difference in abundance of NDCBE between the lane loaded with total cortex and the one loaded with isolated CCD does not reflect an enrichment of the cortex versus the CCD, but actually only reflects the difference in the quantity of protein loaded. (C) Analyses of amiloride-resistant Na⁺ and Cl^– transepithelial fluxes in CCDs isolated from NDCBE-KO mice maintained on a Na⁺-depleted diet. Fluxes were measured in the presence of 10^–5 M amiloride in the perfusate to reflect only the amiloride-resistant, HCTZ-sensitive component of NaCl reabsorption. Values are mean ± SEM; statistical significance was assessed by unpaired Student’s t test with Welch’s correction, when appropriate. n = 5 in each group.

Figure 7. Effects of HCTZ (10^–4 M) on NDCBE or PDS activity in isolated collecting ducts or on recombinant NDCBE or PDS expressed in Xenopus oocytes.

(A) Effects of HCTZ and Ndcbe disruption on Na⁺-dependent pH_i changes measured in intercalated cells of CCD isolated from Na-depleted Ndcbe^–/– or Ndcbe^+/+ mice fed a low-Na⁺ diet. Traces are the average of pH_i changes recorded when luminal Na⁺ was removed and then readded, in the presence of extracellular Cl^– (122 mM) and HCO₃^– (25 mM). Intracellular Na⁺-dependent acidification was detected in Ndcbe^+/+ mice but absent in Ndcbe^–/– mice or when HCTZ 10^–4 M was present in the perfusate. In these 3 different experimental conditions, mean starting pH_i values were 7.10 ± 0.02, 6.93 ± 0.11, and 7.01 ± 0.04, respectively. (B) Effects of HCTZ on apical Cl^–/HCO₃^– exchange activity in intercalated cells of CCDs isolated from Na-depleted animals. Traces are the average of pH_i changes recorded when luminal Cl^– was removed and then readded, in the presence of extracellular HCO₃^– (25 mM) and in Na⁺-free solutions. Intracellular Cl^–-dependent alkalinization, reflecting apical Cl^–/HCO₃^– exchange, was completely abolished when 10^–4 M HCTZ was present in the perfusate. Mean starting pH_i values (immediately before Cl^– removal) were 6.91 ± 0.03 and 6.89 ± 0.08, in the absence and presence of HCTZ, respectively. (C) Effects of HCTZ on mNdcbe-mediated HCO₃^– influx. Oocytes had been injected with mNdcbe cRNA or H₂O and incubated with HCTZ (0.25 mM). As a control, NDCBE-expressing and H₂O-injected oocytes were incubated with vehicle (methanol). Values are mean ± SEM with 6–9 oocytes per group. **P < 0.01, ***P < 0.001 versus H₂O-injected oocyte. HCO₃^– flux was unaffected by the application of HCTZ (0.25 mM) compared with vehicle alone (P = 0.279). (D) Effects of HCTZ on Pds-mediated ³⁶Cl^– uptake. Pendrin-expressing oocytes (mPds) were incubated in ND96 containing 0.1 or 1 mM HCTZ during the uptake period (16 minutes). As a control, pendrin-expressing and H₂O-injected oocytes were incubated with vehicle (methanol). Values are mean ± SEM, with 6–16 oocytes per group. *P < 0.001 versus H₂O; ^†P < 0.001 versus mPds, HCTZ 0.1 mM.

Figure 8. Effects of ACZ (10^–4 M) on NDCBE- or PDS-dependent transport in isolated collecting ducts.

(A) Effects of 10^–4 M ACZ on Cl^– transepithelial transport in CCDs isolated from Ncc^–/– mice. CCDs were isolated from Ncc^–/– mice and bathed and perfused with CO₂/HCO₃^–-containing solutions. Statistical significance was assessed by 2-tailed Student’s unpaired t test. n = 5 in each group; *P < 0.001 versus control. (B) Effects of luminal 10^–4 M ACZ on pendrin activity in isolated CCDs. Tubules were isolated from wild-type mice. Pendrin activity was assessed by measuring changes in pH_i when Cl^– was removed and then readded from the perfusate. Both bath and perfusate solutions contained 25 mM HCO₃^– and were sodium-free. Traces represent the average of recordings from independent tubules. n = 4–5 independent tubules by group. Mean starting pH_i values were 6.91 ± 0.03 and 6.72 ± 0.05, in the absence and presence of ACZ, respectively. (C) Effects of luminal 10^–4 M ACZ on Na⁺ transepithelial transport in CCDs isolated from Ncc^–/– mice. Statistical significance was assessed by 2-tailed Student’s unpaired t test. n = 5 in each group. (D) Effects of ACZ 10^–4 M on NDCBE activity in isolated CCDs. Tubules were isolated from wild-type mice. NDCBE activity was assessed by measuring changes in pH_i of intercalated cells when Na⁺ was removed and then readded from the perfusate. Both bath and perfusate solutions contained 25 mM HCO₃^– and 122 mM Cl^–. The bath solution was sodium-free to silence basolateral Na⁺/H⁺ exchanger activity. Traces represent the average of recordings from independent tubules. n = 4–5 independent tubules by group. Mean starting pH_i values were 7.03 ± 0.04 and 6.99 ± 0.05, in the absence and presence of ACZ, respectively.