KCC3a, a Strong Candidate Pathway for K+ Loss in Alkalemia

Abstract

Loss-of-function mutations in the human potassium chloride cotransporter-3 (KCC3) cause a hereditary motor sensory neuropathy associated with agenesis of the corpus callosum. While recapitulating the neuropathy, KCC3-knockout mice also exhibit high blood pressure. This phenotype is believed to have neurogenic and/or vascular origins. The role of KCC3 in the kidney is poorly understood. KCC3 is encoded by two major isoforms originating from alternative promoters: KCC3a and KCC3b, with KCC3b being the predominant transcript in the kidney. Although the transporter has previously been localized to the proximal tubule, we show here the unique expression of the KCC3a isoform in the connecting tubule. Using a KCC3a-specific polyclonal antibody validated for both immunofluorescence and immunoblotting, we showed an intense KCC3a signal restricted to cortical intercalated cells. No overlap is detected between KCC3a and sodium chloride cotransporter (NCC), a distal convoluted tubule (DCT) marker; or between KCC3a and ENaC or calbindin, which are both principal cell markers. KCC3a signal was observed in cells expressing the apical V-ATPase and pendrin, establishing a unique expression pattern characteristic of intercalated cells of type-B or type-nonA/nonB. We further show that treatment of wild-type mice with hydrochlorothiazide, amiloride, or fed a K⁺-deficient diet up-regulates KCC3a level, suggesting that volume depletion increases KCC3a abundance. This hypothesis was confirmed by showing a higher abundance of KCC3a protein after 23-h water restriction or after placing the mice on a low-salt diet. More importantly, abundance of the Cl^-/HCO₃ ^- exchanger, pendrin, which is known to secrete bicarbonate in alkalotic conditions, was significantly diminished in KCC3-knockout mice. In addition, KCC3a abundance increased significantly alongside pendrin abundance in bicarbonate-treated alkalotic mice, providing a credible mechanism for K⁺ loss in metabolic alkalosis.

Keywords: K+ loss; K–Cl cotransport; bicarbonate; intercalated cells; metabolic alkalosis.

FIGURE 1

Generation of a potassium chloride cotransporter 3a (KCC3a)–specific antibody. (A) Structure of the mouse Slc12a6 (KCC3) gene showing alternative promoters and exons 1a and 1b, as well as exons 2, 3, and 25. (B) Glutathione-S-transferases (GST)–exon 1a fusion protein used to purify the rabbit immune serum and fusion protein exon 1a–enhanced green fluorescent protein (EGFP) purified from transfected HEK293 cells used to assess antibody specificity. (C) Immunoblot showing EGFP expression in HEK293 cells transfected with 1a-EGFP or 2-3-EGFP but not untransfected cells. KCC3 signal in cells transfected with 1a-EGFP, but not 2-3-EGFP or untransfected cells. (D) Immunoblot shows the band at ∼130 kDa with kidney (K) and brain (B) samples from wild-type mice, but not from KCC3-knockout (KO) mice. Notably, there is an abundance of KCC3a in brain tissue, compared to the kidney.

FIGURE 2

Potassium chloride cotransporter 3a (KCC3a) expression in pendrin-expressing cells. KCC3a signal (purple) in the kidney cortex of wild-type mice (A), but not KCC3-knockout mice (B). KCC3a expression [green, (C)] and pendrin expression [red, (D)] with an overlap of signal [yellow, (E)] in the kidney cortex at 40X. Notably, there is a presence of cells with faint basolateral KCC3a staining (white stars). Similarly, KCC3a expression [green, (F)] and pendrin expression [red, (G)] with overlap [yellow, (H)] at 63X. The absence of KCC3 expression in most tubule segments surrounding the positive cells is also noted. Scale bars = 50 µm for panels (A–E) and 10 µm for panels (F–H).

FIGURE 3

Potassium chloride cotransporter 3a (KCC3a) expression along the connecting tubule. KCC3a expression [green, (A)] and vacuolar-type ATPase (V-ATPase) expression [red, (B)] with overlap of signal [yellow, (C)] in kidney cortex. KCC3a expression [purple, (D)], phosphorylated sodium chloride cotransporter (pNCC) [green, (E)], epithelial sodium channels (ENaC) [red, (F)], and merged signal (G) in the kidney cortex. Expression of KCC3a [green, (H)], connecting tubule-specific principal cell marker—Calbindin-28 [red, (I)] and merged image (J) in the renal cortex. The signal for KCC3a is interspersed among the calbindin-positive cells. Scale bar = 50 µm.

FIGURE 4

Change in potassium chloride cotransporter 3a (KCC3a) abundance in the kidney in mice on K⁺-deficient diet. (A) KCC3a abundance increased on K⁺-free diet (KFD) than in the control (Ctrl) diet (two-tailed unpaired t-test). (B) The phosphorylated sodium chloride cotransporter (pNCC) abundance was higher on the KFD than on the control diet (two-tailed unpaired t-test), indicating the diet’s efficiency. For blot quantification, densitometric values were normalized to β-actin. Values are means ± SEM; values in parentheses indicate n values. *p < 0.05; ****p < 0.0001.

FIGURE 5

Potassium chloride cotransporter 3a (KCC3a) abundance changes in response to hydrochlorothiazide (HCTZ) treatment. (A) KCC3a abundance increases after HCTZ (37.5 mg/kg body weight) treatment compared with the vehicle (Veh, glycol:water = 300:100)-treated group (two-tailed unpaired t-test). (B) After HCTZ treatment, compared with the Veh-treated mice, the abundance of phosphorylated sodium chloride cotransporter (pNCC) increases (two-tailed unpaired t-test), demonstrating the drug’s effectiveness. (C) The abundance of KCC3a in amiloride-treated mice was higher than in vehicle-treated mice (two-tailed unpaired t-test). (D) In mice treated with vehicle, the abundance of pNCC was lower than in mice treated with amiloride (two-tailed unpaired t-test). For blot quantification, densitometric values were normalized to β-actin. Values are means ± SEM; values in parentheses indicate n values. *p < 0.05; ***p < 0.001.

FIGURE 6

Potassium chloride cotransporter 3a (KCC3a) abundance changes in response to water deprivation and high-salt versus low-salt diets. (A) Mice were water restricted for 23 h, (Wat. Dep.) and KCC3a abundance was compared to mice maintained on water during the same period Ctrl, (two-tailed unpaired t-test). (B) Mice were fed for 4 days on a high-salt or low-salt diet and analyzed for KCC3a expression (two-tailed unpaired t-test). For blot quantification, densitometric values were normalized to β-actin. Values are means ± SEM; values in parentheses indicate n values. *p < 0.05; ***p < 0.001.

FIGURE 7

Link between pendrin and KCC3a. (A,B) NaHCO₃ treatment for 24 h increased KCC3a and pendrin abundances compared to control (Ctrl), (two-tailed unpaired t-test). (C) Urinary K⁺ excretion was higher in NaHCO₃-induced alkalotic condition (two-tailed paired t-test). For blot quantification, densitometric values were normalized to β-actin. Values are means ± SEM; values in parentheses indicate n values. *p < 0.05; ***p < 0.001.

FIGURE 8

KCC3 deletion in KCC3-knockout (KCC3-KO) mice downregulated pendrin abundance compared to control (Ctrl), (two-tailed unpaired t-test). For blot quantification, densitometric values were normalized to β-actin. Values are means ± SEM; values in parentheses indicate n values. ***p < 0.001.

FIGURE 9

KCC3a participates in K⁺ loss in isolated alkalemia caused by metabolic alkalosis. Increased HCO₃ ⁻ with a consequent increase in blood pH is known as alkalemia. NonA/nonB and/or type-B intercalated cells express pendrin and KCC3a at the apical membrane. Pendrin and KCC3a are upregulated during base loading or metabolic alkalosis. As KCC3a secretes K⁺ and Cl⁻ and pendrin recycles Cl⁻ in exchange for HCO₃ ⁻, this leads to a net loss of KHCO₃.