Pore directions for the expression of a Ca2+‐activated chloride channel

A major step has been made in determining how cytosolic Ca2+ might contribute to the regulation of membrane potential, cell excitability, ion transport and sensation. In this issue of The Journal of Physiology, Adomaviciene et al. (2013) implicate a portion of a putative pore in Ca2+-activated chloride channels (CaCCs) as a non-canonical Ca2+-sensitive region that enhances channel residency in the cell membrane. Many questions remain regarding the time-dependent interplay between Ca2+ concentration and membrane potential in cells expressing CaCCs. While these observations raise questions regarding the positioning of a putative extracellular loop and the structure of the anion permeation pathway, they provide a new direction for the field in the characterization of CaCCs. Much effort has been expended over more than three decades to define the molecular basis of CaCCs. A first goal is to define the molecular identity and structure that accounts for the physiological functions that have been attributed to CaCCs, including epithelial ion transport, smooth muscle contraction, olfaction, gustation, photoreception, somatic sensation and fertilization. Once defined, CaCCs would constitute potential drugable targets to treat or circumvent a variety of diseases, including cancer and cystic fibrosis. Perhaps unwittingly, a major step in defining CaCCs was made in 2003 with the identification of a chromosome 11 gene that coded for a putative transmembrane protein that was expressed in oesophageal, bladder and breast tumours, which was ultimately named Tmem16a (Katoh, 2003). This original identification included the observation that two splice variants (with/without exon 15) were expressed. In silico analysis revealed that Tmem16a was the founder for a 10-member gene family that is characterized by eight putative transmembrane sequences and cytosolic N and C termini. Five years later, three laboratories employing distinctly different approaches independently reported that Tmem16a was (or contributed to) a CaCC (Caputo et al. 2008; Schroeder et al. 2008; Yang et al. 2008). Importantly, Tmem16a exhibited characteristics of classical CaCCs, including Ca2+ and voltage sensitivity, outward rectification at low Ca2+ concentrations with linear conductance at high concentrations, a selectivity sequence of I− > Cl−, permeability to polyatomic anions including SCN− and NO3−, high anion to cation selectivity, and inhibition by niflumic acid. The second family member to be identified, Tmem16b, was also shown to exhibit CaCC characteristics. Based upon the channel function and the structure, it was proposed (Yang et al. 2008) to rename the gene family ‘anoctamins’[anion channels with eight (oct) membrane-spanning segments], which provides the basis for the accepted HUGO nomenclature, ANO1–10. Ultimately, protein identification is only the starting point to define the physiology. Calcium sensitivity is a hallmark of both ANO1 and ANO2, although the Ca2+ concentration dependency differs. ANO1 is reported to have a 10-fold higher sensitivity to Ca2+ (apparent KD 100–500 nm) when compared with ANO2, although direct comparisons in a single expression system have not been reported. The Ca2+ binding site(s) have not been identified; neither canonical helix-loop-helix domains such as the ‘F-hands’ first described for carp parvalbumin, nor classical calmodulin binding domains (i.e., IQ domains) have been identified. Likewise, the voltage sensor has not been defined. Unlike many voltage-sensitive channels, the anoctamins do not harbour a putative transmembrane segment with multiple charged residues. The first putative intracellular loop of ANO1 contains four consecutive glutamate residues, which have been speculated to confer voltage sensitivity. Adjacent to these is a four-residue segment (coded by exon 13) that is absent in some variants that exhibit reduced Ca2+ sensitivity, which suggests that the first intracellular loop may contain critical components for both voltage and Ca2+ sensitivity. Homology modelling suggested that the third extracellular loop includes a component that partly penetrates the membrane and forms a critical portion of the ion channel. Indeed, mutation and cysteine-modifying studies have supported this notion. It is logical to evaluate this putative pore-loop further to define the channel, which Adomaviciene et al. (2013) set out to do. The authors defined a number of parameters that differentiate the whole-cell currents associated with ANO1 and ANO2, including activation and inactivation rates along with voltage and Ca2+ dependency. Permselectivity, however, was similar for a series of anions. Additionally, the outcomes revealed differences in current magnitude and channel density that suggested differences in membrane residency. A chimeric approach was used, and the data set was extended to show that a small segment of the third putative extracellular loop is a determinant of channel density. These 38 residues do not contain a recognized targeting sequence, yet the simplest interpretation of these data is that this segment from ANO1 either increases the trafficking to or decreases the removal from the cell membrane. Additionally, the putative extracellular loop carried with it effects on both voltage and Ca2+ sensitivity. These results suggest that substantial effort is warranted to dissect the many facets of channel behaviour that are imparted by this putative pore-loop sequence. There are many questions regarding the anoctamins and specifically ANO1 and ANO2 that remain to be addressed. Adomaviciene et al. (2013) show that the scaffold around the putative pore-loop also affects channel behaviour. Voltage dependency seems to associate with the pore sequence, but the Boltzmann constant does not, which suggests that additional portions of the protein contribute to the voltage-sensing apparatus. Likewise, the putative pore-loop contributes to Ca2+ sensitivity, but is not the sole determinant. The concentration dependency is not consistent with a simple bimolecular interaction. The Hill coefficient approaches 3, which is greater than the value reported by other researchers. Nonetheless, the simplest interpretation is that multiple Ca2+ binding sites contribute to channel activation, which could reflect an interaction with calmodulin, as has been postulated. Alternatively, the channel may be a homo- or heterodimer, as has been postulated by others, with a requisite Ca2+ binding site on each monomer. The absence of consensus binding sites makes this a difficult but intriguing question to address. Numerous splice variants of ANO1 have been reported, with alternative forms of exon 1 or 6 and the absence or presence of exons 13 and 15. Likewise, splice variants of ANO2 are expressed. There are likely to be differences in developmental expression or tissue distribution, but these differences remain to be defined. Different forms may be targeted to specific cell regions, such as epithelial apical or basolateral membranes, that can contribute to anion secretion or to K+ secretion, as has been suggested. Calcium sensitivity has been associated with residues in the first intracellular loop and now with the putative pore-loop region. Thus, these distinct and distant segments of the protein may work in concert to direct the protein to the appropriate cellular destination. The paper by Adomaviciene et al. (2013) builds substantially on previous work in the field and clearly changes the frontier for ongoing investigations. These authors provide a new starting point for studies that are designed to determine how this putative pore directs CaCC distribution, regulation and gating.