Loss of chloride channel ClC-5 impairs endocytosis by defective trafficking of megalin and cubilin in kidney proximal tubules

Abstract

Loss of the renal endosome-associated chloride channel, ClC-5, in Dent's disease and knockout (KO) mice strongly inhibits endocytosis of filtered proteins by kidney proximal tubular cells (PTC). The underlying mechanism remains unknown. We therefore tested whether this endocytic failure could primarily reflect a loss of reabsorption by the multiligand receptors, megalin, and cubilin, caused by a trafficking defect. Impaired protein endocytosis in PTC of ClC-5 KO mice was demonstrated by (i) a major decreased uptake of injected 125I-beta 2-microglobulin, but not of the fluid-phase tracer, FITC-dextran, (ii) reduced labeling of endosomes by injected peroxidase and for the endogenous megalin/cubilin ligands, vitamin D- and retinol-binding proteins, and (iii) urinary appearance of low-molecular-weight proteins and the selective cubilin ligand, transferrin. Contrasting with preserved mRNA levels, megalin and cubilin abundance was significantly decreased in kidney extracts of KO mice. Percoll gradients resolving early and late endosomes (Rab5a, Rab7), brush border (villin, aminopeptidase M), and a dense peak comprising lysosomes (acid hydrolases) showed a disappearance of the brush border component for megalin and cubilin in KO mice. Quantitative ultrastructural immunogold labeling confirmed the overall decrease of megalin and cubilin in PTC and their selective loss at the brush border. In contrast, total contents of the rate-limiting endocytic catalysts, Rab5a and Rab7, were unaffected. Thus, impaired protein endocytosis caused by invalidation of ClC-5 primarily reflects a trafficking defect of megalin and cubilin in PTC.

Fig. 1.

Selective impairment of protein endocytosis in PTC from ClC-5 KO mice. (A) Quantitation of the uptake deficit. At 7 min or 1 h after i.v. injection of¹²⁵I-β₂M to control (WT) and ClC-5 KO mice (KO), radioactivity was measured in bleached kidneys and liver (n = 4). (B and C) Visualization of the apical endocytic apparatus by light (B) and ultrastructural (C) cytochemistry at 7 or 60 min after HRP injection. (B Upper) At 7 min, the reaction product stains the brush border (gray signal next to the collapsed lumen) in WT and KO, whereas multiple apical dots immediately thereunder are seen only in WT. (B Lower and C) At 1 h after injection to WT mice, HRP mostly labels small endocytic vesicles and dense apical recycling tubules and includes larger structures, whereas labeling is greatly reduced in KO mice. (D) Urine analysis by Western blotting. Urinary excretion of LMWPs, transferrin, and cathepsin B in KO mice is obvious, contrasting with decreased urinary loss of megalin and normal excretion of cubilin.

Fig. 2.

Immunofluorescence. The strong apical signal, seen for megalin and cubilin in WT proximal tubules, is markedly decreased in KO mice. Likewise, the granular staining for reabsorbed endogenous DBP and RBP is strongly reduced in ClC-5 KO kidneys.

Fig. 3.

Subcellular fractionation in Percoll gradients. Average density in seven gradients is indicated by a broken line. Dotted lines indicate initial concentration. Distributions after centrifugation are presented as C/Ci indices, where values >1 reflect organelle enrichment and values <1 reflect organelle depletion. Typical distributions representative of at least three fractionation analyses for each constituent are shown for control (WT, ○) and corresponding KO littermates (•). Distributions of¹²⁵I-β₂-M were comparable after 7 min and 1 h.

Fig. 4.

Immunogold labeling of megalin and cubilin on ultrathin frozen sections of renal cortices. Arrows indicate labeling at brush border, which is much reduced in KO as compared with WT mice.