Loss of the Secretin Receptor Impairs Renal Bicarbonate Excretion and Aggravates Metabolic Alkalosis in Mice during Acute Base-Loading

Abstract

Significance statement: During acute base excess, the renal collecting duct β -intercalated cells ( β -ICs) become activated to increase urine base excretion. This process is dependent on pendrin and cystic fibrosis transmembrane regulator (CFTR) expressed in the apical membrane of β -ICs. The signal that leads to activation of this process was unknown. Plasma secretin levels increase during acute alkalosis, and the secretin receptor (SCTR) is functionally expressed in β -ICs. We find that mice with global knockout for the SCTR lose their ability to acutely increase renal base excretion. This forces the mice to lower their ventilation to cope with this challenge. Our findings suggest that secretin is a systemic bicarbonate-regulating hormone, likely being released from the small intestine during alkalosis.

Background: The secretin receptor (SCTR) is functionally expressed in the basolateral membrane of the β -intercalated cells of the kidney cortical collecting duct and stimulates urine alkalization by activating the β -intercalated cells. Interestingly, the plasma secretin level increases during acute metabolic alkalosis, but its role in systemic acid-base homeostasis was unclear. We hypothesized that the SCTR system is essential for renal base excretion during acute metabolic alkalosis.

Methods: We conducted bladder catheterization experiments, metabolic cage studies, blood gas analysis, barometric respirometry, perfusion of isolated cortical collecting ducts, immunoblotting, and immunohistochemistry in SCTR wild-type and knockout (KO) mice. We also perfused isolated rat small intestines to study secretin release.

Results: In wild-type mice, secretin acutely increased urine pH and pendrin function in isolated perfused cortical collecting ducts. These effects were absent in KO mice, which also did not sufficiently increase renal base excretion during acute base loading. In line with these findings, KO mice developed prolonged metabolic alkalosis when exposed to acute oral or intraperitoneal base loading. Furthermore, KO mice exhibited transient but marked hypoventilation after acute base loading. In rats, increased blood alkalinity of the perfused upper small intestine increased venous secretin release.

Conclusions: Our results suggest that loss of SCTR impairs the appropriate increase of renal base excretion during acute base loading and that SCTR is necessary for acute correction of metabolic alkalosis. In addition, our findings suggest that blood alkalinity increases secretin release from the small intestine and that secretin action is critical for bicarbonate homeostasis.

Graphical abstract

Figure 1

SCTR, CFTR, and pendrin KO mice display delayed correction of an orally imposed metabolic alkalosis. (A) Venous pH at baseline and1 hour and 2 hours after 4 mmol NaHCO₃/kg. (B) Venous standard bicarbonate at baseline and 1 hour and 2 hours after a gavage load of 4 mmol NaHCO₃/kg. Baseline: n=15–16. 4 mmol NaHCO₃/kg: n=9–11. (C and D) SCTR, CFTR, and pendrin KO mice show a similar inability to compensate an oral base load. (C) Change in venous standard bicarbonate from baseline to 1 hour and 2 hours and (D) 1–2 hours after a gavage load of 4 mmol NaHCO₃/kg in WT mice (n=28) and pendrin (n=7), CFTR (n=6), and SCTR KO mice (n=11). Statistical differences were assessed by the use of a mixed-effects analysis using genotype, treatment (0, 2.24, and 4 mmol NaHCO₃/kg), and time (as a categorical variable) with an interaction between each variable as fixed effects (A and B) and one-way ANOVA, followed by multiple comparisons (D). Note that the baseline values from Figure 2C are used for comparisons in (A and B). KO, knockout; SCTR, secretin receptor; WT, wild-type. Figure 1 can be viewed in color online at www.jasn.org.

Figure 2

SCTR KO mice display no acid–base disturbances and normal β-IC function during control conditions. (A) Urine pH, HCO₃⁻, TA, NH₄⁺, and net acid (NA=[NH₄⁺+TA]−HCO₃⁻) concentration in spot urine samples (n=10–12). (B) Twenty-four–hour urine NAE in metabolic cages (mean of two consecutive days, n=6–9). (C) Venous pH and standard HCO₃⁻ concentration (n=15–16). (D) Representative immunohistochemistry images stained for pendrin (brown); scale bars represent 50 µm. (E) Pendrin protein abundance determined by immunoblotting (n=10–12). See Supplemental Figure 1 for full-length gel with sex indicated for each lane. (F) Pendrin function as determined by the intracellular alkalization rate in ICs from isolated perfused cortical collecting ducts after a luminal Cl⁻ removal (n=14–41 cells). Statistical differences were assessed by Student' t tests. IC, intercalated cell; KO, knockout; NAE, net acid excretion; SCTR, secretin receptor; TA, titratable acid. Figure 2 can be viewed in color online at www.jasn.org.

Figure 3

SCTR KO mice do not respond to secretin with urinary alkalization or increased pendrin function. The effect of i.p. application of secretin (5 µg/25 g, red) or saline (black) in (A) SCTR WT and (B) KO mice. (C) Summary of peak urine pH difference between secretin- and saline-injected SCTR WT and KO mice (n=6). (D) Representative image of a 2′,7′-Bis-(2-Carboxyethyl)-5-(and-6)-Carboxyfluorescein, Acetoxymethyl Ester–loaded isolated perfused cortical collecting duct. Note the preferential BCECF dye loading in ICs. (E) Representative pH_i traces after a luminal Cl⁻ removal in SCTR WT and KO pendrin positive cells with or without preincubation with secretin (10 nM for 10 minutes). (F) Summary of pendrin function in WT and KO ICs with and without secretin preincubation (n=16–27 cells). Statistical differences were assessed by two-way ANOVA, followed by multiple comparisons. IC, intercalated cell; KO, knockout; SCTR, secretin receptor; WT, wild-type. Figure 3 can be viewed in color online at www.jasn.org.

Figure 4

Impaired bicarbonate secretion in SCTR KO mice. (A) Three-hour cumulated net base (HCO₃⁻−[NH₄⁺+TA]), HCO₃⁻, TA, and NH₄⁺ excretion in SCTR WT mice after a gavage load of approximately 0, 2.24, or 4 mmol NaHCO₃/kg. (B–D) Cumulated. (B) HCO₃⁻, (C) TA, and (D) NH₄⁺ excretion as a function of hours after gavage loading in SCTR WT mice. (E–H) Three-hour cumulated. (E) Net base, (F) HCO₃⁻, (G) TA, and (H) NH₄⁺ excretion in SCTR WT and KO mice after a gavage load of approximately 0, 2.24, or 4 mmol NaHCO₃/kg (n=4–12). Note that one SCTR KO mouse was excluded from net base and NH₄⁺ comparisons because of extremely large NH₄⁺ excretions. Statistical differences were assessed by one-way ANOVA, followed by multiple comparisons (A), and by the use of a mixed-effects analysis using genotype, treatment (0, 2.24, and 4 mmol NaHCO₃/kg), and time (as a categorical variable) with an interaction between each variable as fixed effects (E–H). KO, knockout; SCTR, secretin receptor; TA, titratable acid; WT, wild-type. Figure 4 can be viewed in color online at www.jasn.org.

Figure 5

Acute base loading causes hypoventilation in SCTR KO mice. (A and B) Difference in air convection requirement for CO₂ (ΔACR_CO2) and minute ventilation (ΔV_E) in SCTR WT and KO mice after a gavage load of approximately 0 or 4 mmol NaHCO₃/kg as a function of time (6 hours before and 6 hours after gavage). (C and D) Summary of the differential response in SCTR WT and KO mice during the first 2 hours after gavage loading (n=11–12, one SCTR KO mouse was excluded because of an extremely large decrease of ΔACR_CO2). Statistical differences were assessed by the use of a mixed-effects analysis using genotype and period (baseline versus 2 hours) with an interaction between each variable as fixed effects. KO, knockout; SCTR, secretin receptor; WT, wild-type. Figure 5 can be viewed in color online at www.jasn.org.

Figure 6

SCTR KO mice display aggravated and delayed correction of a metabolic alkalosis imposed by i.p. NaHCO₃ application. Venous (A) pH, (B) standard bicarbonate, (C) pCO₂, and (D) Cl⁻ at baseline and 1 hour and 2 hours after an i.p. injection of approximately 0 or 2.24 mmol NaHCO₃/kg. Baseline: n=15–16; approximately 0 mmol NaHCO₃/kg: n=6–7; 2.24 mmol NaHCO₃/kg: n=6–7. Note that the baseline values from Figure 2C are used for comparisons. Statistical differences were assessed by the use of a mixed-effects analysis using genotype, treatment (0 and 2.24 mmol NaHCO₃/kg), and time (as a categorical variable) with an interaction between each variable as a fixed effect. KO, knockout; SCTR, secretin receptor. Figure 6 can be viewed in color online at www.jasn.org.

Figure 7

Perfusing the superior mesenteric artery with alkaline perfusion buffer (pH: 7.6, HCO₃⁻: 38 mM, pCO₂: 40 mm Hg) increases secretin release from the rat small intestine. (A) Schematic illustration of the experimental setup. The rat upper small intestine is isolated and perfused through a catheter inserted into the superior mesenteric artery, and the venous effluent is collected from a catheter in the portal vein for secretin concentration measurements. (B) Secretin concentrations in the venous effluent from isolated perfused rat small intestines as a function of alkalinity of the arterial perfusate and after luminal HCl (0.1 mM) stimulation. (C) Mean baseline (first 10 minutes) secretin concentration in the venous effluent is approximately two-fold higher with alkaline perfusion buffer (pH: 7.6, HCO₃⁻: 38 mM) as compared with normal (pH: 7.4, HCO₃⁻: 24 mM) and acidic perfusion buffer (pH 7.2, HCO₃⁻: 15 mM). Statistical differences were assessed by one-way ANOVA, followed by multiple comparisons.