Chloride conductances that are sensitive to calcium (CaCC) are found in most cells and tissues. However, assigning a molecular identity to a particular conductance has proved challenging, not least because of the variety of observed current phenotypes. The CLCAs are a recently identified mammalian gene family that when heterologously expressed behave as chloride channels when exposed to calcium. They are widely distributed in both excitable and non-excitable tissues including epithelia, endothelia, smooth muscle, neurons and brain. The prototypical CLCA family member was isolated and subsequently cloned from bovine tracheal epithelium (Ran et al. 1992). When incorporated into planar lipid bilayers, the purified protein formed channels of approximately 25 pS, had a linear I-V relationship in symmetrical solutions and was inhibited by DIDS. Similar properties were found for the cloned protein when transiently expressed in Xenopus oocytes or COS-7 cells (Cunningham et al. 1995). It is now recognized that four CLCA family members are expressed in mammals, although this family does not seem to be represented in C. elegans, Drosophila, or birds (Gruber et al. 2000). When compared to calcium-sensitive chloride currents recorded from native cells, the heterologously expressed currents have many properties in common, including ion selectivity profile (I– > Cl–), sensitivity to several agents widely used as chloride channel blockers (e.g. DIDS, NPPB), and an outwardly rectified current-voltage relationship under whole-cell recording conditions. However, other features of heterologously expressed and/or purified CLCAs, such as the lack of a time dependent component to the whole-cell current, sensitivity to the reducing agent DTT, and relatively large single channel conductance suggest that by themselves they do not fully recapitulate the native channel/current phenotype. This may reflect a role for the CLCAs as subunits of a more complex multimer including other as yet unidentified or unrecognized polypeptides. Surprising evidence for such an interaction has recently been presented when current more typical of that found in vascular smooth muscle was reconstructed in HEK 293 cells by co-expression of mCLCA1 and the non-pore forming β subunit of the large conductance BK channel (Greenwood et al. 2002). However, the potential functions of members of the CLCA family are not confined to their channel properties; intriguingly, certain members of the CLCA family have also been demonstrated to function both as cell adhesion molecules, and as tumour suppressors. When expressed in an underlying substrate of endothelial cells, one bovine member of this family (LuECAM1, bCLCA2) mediates adherence of metastatic melanoma cells, an interaction that can be blocked by a monoclonal CLCA antibody (Goetz et al. 1996). Furthermore, a murine family member, mCLCA1 binds to β4-integrin, activating downstream FAK/ERK signalling pathways (Abdel-Ghany et al. 2002). Additionally both hCLCA1 and hCLCA2 are downregulated in colon and breast cancer (Gruber & Pauli 1999; Bustin et al. 2001), while heterologous expression of the murine homologues in tumour cells reduce colony formation in inoculated animals and promote apoptosis (Elble & Pauli, 2001). A role for the CLCAs is also emerging in the field of asthma research, where the asthma-associated cytokine IL-9 both promotes expression of hCLCA1 and mucus secretion in airway goblet cells (Toda et al. 2002). In this context, the localization of the murine homologue of hCLCA1 (mgob5, mCLCA3) on the membrane of mucus granules is highly consistent (Leverkoehne & Gruber, 2002). Given the diverse nature of potential functions of the CLCA family, considerable further studies will be required to determine whether the CLCAs form independent ionic channels, in addition to elucidating the interdependence of conductance, adhesion and tumour suppressor roles.
Toda et al. (2002). J Allergy Clin Immunol 109, 246-250.
Work in the authors’ laboratory is supported by NIH Grant DK 53090.