CLC Cl^– channels form an evolutionarily old gene family that has nine members in mammals. While the first branch of this family encodes plasma membrane channels, it is now clear that channels belonging to the other two branches reside primarily in intracellular organelles.

ClC-Ka and ClC-Kb (both belonging to the first branch) are expressed in different nephron segments. Mutations in ClC-Kb underlie Bartter syndrome III, while the disruption of ClC-Ka in mice caused diabetes insipidus. While this indicated a role in transepithelial transport, both proteins did not yield currents upon heterologous expression. It is now clear that they need barttin, a protein with two transmembrane spans, for functional expression (Estévez et al. 2001). Mutations in barttin underlie Bartter syndrome type IV, which combines severe renal salt loss with deafness. Indeed, ClC-K/barttin heteromers are additionally involved in K⁺ secretion across the stria vascularis of the inner ear by recycling Cl^– taken up by a Na⁺-K⁺-2Cl^– cotransporter (Estévez et al. 2001).

It has long been known that Cl^– conductances provide an electrical shunt that is needed for the efficient operation of the H⁺-ATPase that acidifies vesicles of the endocytotic and secretory pathways. However, the molecular identities of the underlying Cl^– channels have remained obscure until such a role was recognised for ClC-5. This channel is mutated in Dent’s disease, a disorder characterised by low molecular weight proteinuria and kidney stones (Lloyd et al. 1996). ClC-5 resides in endosomes of the proximal tubule (PT), where it co-localises with the H⁺-ATPase and endocytosed proteins (Günther et al. 1998). This suggested a role in the acidification of the endocytotic pathway. Disrupting ClC-5 in mice affects both fluid-phase and receptor-mediated endocytosis, as well as the endocytotic retrieval of certain plasma membrane proteins in the PT (Piwon et al. 2000). As the PT endocytoses hormones such as PTH and 25(OH)VitD3, this leads to changes in calcio-tropic hormone levels and to secondary changes in the renal handling of phosphate and calcium. Thus the vesicular ClC-5 Cl^– channel is crucial for endocytosis.

We have also disrupted the highly homologous ClC-3 Cl^– channel that is expressed in brain and several other organs (Stobrawa et al. 2001). This led to a nearly complete degeneration of the hippocampus and photoreceptors. ClC-3 was localised to late endosomes and synaptic vesicles, to whose acidification it contributes. The degeneration of the hippocampus may be due to an altered filling of synaptic vesicles (which depends on the electrochemical H⁺ gradient), or to altered intracellular trafficking.

Finally, we have disrupted ClC-7, a broadly expressed member of the third branch of the CLC family (Kornak et al. 2001). This led to severe osteopetrosis, which is due to a failure of osteoclasts to acidify the resorption lacuna. ClC-7 is normally present in late endosomal to lysosomal compartments, but is inserted together with the H⁺-ATPase into the osteoclast ruffled border upon attachment to bone. Stimulated by this finding, we also demonstrated that human patients with severe osteopetrosis have mutations in either the ClC-7 Cl^– channel (Kornak et al. 2001), or in a subunit of the H⁺-ATPase (Kornak et al. 2000).

Thus the interplay between the H⁺-ATPase and various CLC Cl^– channels has diverse and important roles for the cell and the organism.