The complex phenotype of Dent disease 1 is probably explained by the predominant expression of ClC-5 in the PT segments, with more discrete expression in the thick ascending limb (TAL) of Henle's loop and the α-type intercalated cells (IC) of the collecting ducts of the kidney [
19]. In PT cells, ClC-5 co-distributes with the vacuolar H
+-ATPase (V-ATPase) in early endosomes [
19,
20], which are responsible for the reabsorption and processing of albumin and LMW proteins that are filtered by the glomerulus (Figure ). These vesicles belong to the receptor-mediated endocytic pathway, which involves the multiligand receptors, megalin and cubilin, located at the apical brush border of PT cells [
21]. Progression along the endocytic apparatus depends on endosomal acidification, driven by the V-ATPase and requiring a parallel Cl
- conductance to maintain electroneutrality. It has long been assumed that ClC-5 could provide such an electrical shunt to neutralize the H
+ gradient. Accordingly, the loss of the endosomal Cl
- conductance mediated by ClC-5 would impair vesicular acidification, causing dysfunction of PT cells. Two independent strains of ClC-5 knock-out (KO) mice have been generated, which both recapitulate the major features of Dent's disease including LMW proteinuria and other manifestations of PT dysfunction [
22,
23]. Furthermore,
in vitro experiments have shown a decreased acidification of early endosomes in ClC-5-deficient mice [
24,
25].
However, ClC-5 is a 2Cl
-/H
+ exchanger rather than a Cl
- channel [
14], and the relevance of this exchange activity for Dent's disease was unknown. To address that important issue, Jentsch and colleagues engineered a knock-in (KI) mouse harbouring a point mutation in a critical glutamate residue which converts the exchanger into an uncoupled Cl
- channel that should facilitate endosomal acidification. They then compared these KI mice with the conventional ClC-5 KO mouse [
26]. As expected, acidification of the renal endosomes from wild-type and KI mice was normal, but severely impaired in KO mice. However, despite normal endosomal acidification, KI mice showed the same renal phenotype than KO mice and patients with Dent's disease, including LMW proteinuria, hyperphosphaturia and hypercalciuria. Furthermore, both the KI and KO mouse showed impaired PT endocytosis, indicating that PT dysfunction in Dent's disease may occur despite normal acidification of the endosomes. These findings suggest a role for a reduced endosomal Cl
- accumulation in Dent's disease and, by extension, point to the importance of Cl
- concentration for organelle physiology [
26].
Studies in mice have demonstrated that inactivation of ClC-5 is associated with a severe trafficking defect in PT cells, with loss of megalin and cubilin at the brush border, subsequent loss of their ligands in the urine, and impaired lysosomal processing [
22,
23,
27]. Since the megalin/cubilin complex mediates the reabsorption of the vitamin D-binding protein, the 25(OH)-vitamin D3 and parathyroid hormone (PTH) that are ultrafiltrated by the glomerulus, the urinary loss of these mediators could potentially lead to opposite effects in PT cells, resulting in variable levels of active 1,25(OH)
2-vitamin D
3 levels in the serum [
28]. Such variability could explain why renal hypercalciuria and kidney stones are present in one strain of ClC-5 KO mouse [
23] but not in the other [
22], potentially reflecting the phenotype variability observed in patients harbouring ClC-5 mutations [
7]. Recently, Gailly et al. showed that the deletion of ClC-5 in mouse and human PT cells is associated with increased cell proliferation, oxidative stress and the specific induction of type III carbonic anhydrase [
29]. Furthermore, ClC-5 inactivation is associated with impaired lysosome biogenesis, which also contributes to defective endocytosis and urinary loss of LMW ligands and lysosomal enzymes [
30]. It must be emphasized that other inherited disorders targeting the PT cells, such as lysosomal storage disorders (cystinosis) or mitochondrial cytopathies, may result in PT dysfunction similar to that observed in Dent's disease [
31].
The potential roles of ClC-5 in the TAL (involved in the urinary concentration mechanism and the regulated reabsorption of divalent cations) [
32] and in the α-type IC (responsible for distal urinary acidification) remain to be defined [
33]. Of interest,
CLCN5 mutations have not been detected in patients with idiopathic hypercalciuria and in the hypercalciuric stone-forming (GHS) rat strain [
34]. The hypercalciuria observed in patients with Dent's disease and some ClC-5-deficient mice may be secondary to the PT dysfunction (urinary loss of vitamin D binding protein and reduced phosphate absorption, leading to increased 1,25(OH)
2-vitamin D
3 synthesis) or, at least in part, caused by the functional loss of ClC-5 in the TAL. A small fraction of patients with Dent's disease may have nephrocalcinosis without hypercalciuria [
10], which could indeed reflect the fact that ClC-5 is distributed in several nephron segments that can contribute to the genesis of kidney stones through different mechanisms. For instance, it has been suggested that collecting duct cells lacking ClC-5 may show an impaired ability of internalization of calcium crystals adhering to apical cell surface [
35]. In summary, we can hypothesize that the functional loss of ClC-5 is essentially reflected by manifestations of PT dysfunction and may contribute to the genesis of kidney stones in different ways, reflecting its involvement in specific tubular functions. The issue is further complicated by the existence of a significant inter- and intra-familial variability in the manifestations of nephrocalcinosis and kidney stones.
Although ClC-5 mRNA and protein are detected in rodent intestine [
36] and thyroid [
37], no clear phenotype related to these tissues has been reported in patients. Of note, ClC-5 is highly expressed in the mouse thyroid, located in various endosomes at the apical pole of the thyrocytes. Mice lacking ClC-5 develop a euthyroid goiter, which results from impaired apical iodide efflux (secondary to down-regulated pendrin) rather than defective apical endocytosis [
37].
The phenotype of Dent disease 2 due to
OCRL1 mutations may in part be attributed to the role of
OCRL1 in lysosomal trafficking and endosomal sorting.
OCRL1 encodes a member of the type II family if inositol polyphosphate 5-phosphatases [
38]. These enzymes hydrolyze the 5-phosphate of inositol 1, 4, 5-trisphosphate and of inositol 1,3,4,5-tetrakisphosphate, phosphatidylinositol 4,5-bisphosphate, and phosphatidylinositol 3,4,5-trisphosphate, thereby presumably inactivating them as second messengers in the phosphatidylinositol signalling pathway [
39]. The preferred substrate of OCRL1 is phosphatidylinositol 4,5-bisphosphate (PIP
2), and this lipid accumulates in the renal PT cells of patients with Lowe syndrome [
39]. OCRL1 is localised to lysosomes in renal PT cells and to the trans-Golgi network in fibroblasts. This localisation is consistent with the role of OCRL1 in lysosomal enzyme trafficking from the trans-Golgi network to lysosomes, and the activities of several lysosomal hydrolases are found to be elevated in the plasma of affected patients [
40]. OCRL1 has also been shown to interact with clathrin and indeed co-localises with clathrin on endosomal membranes that contain tranferrin and mannose 6-phosphate receptors [
41]. Mannose 6-phosphate receptor-bound lysosomal enzymes are recruited by appendage (AP) subunits and Golgi-localised binding proteins into clathrin-coated vesicles that transport them from the trans-Golgi network to endosomes [
41]. More recently, Erdmann et al. showed that OCRL1 plays a role in the early endocytic pathway, by interacting with the Rab5 effector APPL1 [
42]. Thus, it seems likely that the
OCRL1 mutations in Lowe syndrome patients result in OCRL1 protein deficiency, which leads to disruptions in the endosomal and/or lysosomal trafficking. This abnormality is similar to that observed in Dent disease 1, and it seems that Dent's disease therefore may be due to abnormalities in either endosomal acidification and sorting, or lysosomal trafficking. It must be noted that the targeted disruption of the murine ortholog for OCRL1 does not cause Lowe syndrome, because
Ocrl1 deficiency is complemented in mice by inositol polyphosphate 5-phosphatase (
Inpp5b) [
43]. Thus, no mouse model recapitulating Lowe syndrome caused by the deficiency in OCRL1 is available.