Chloride transport across vesicle and cell membranes by steroid-based receptors.

It is well-established that molecules which transport cations across cell membranes (cationophores) can have potent biological effects. Anion flux is also important to the cell and, correspondingly, anion carriers may be capable of biological activity. Indeed, chloride transporters have direct medical potential as treatments for cystic fibrosis and other diseases caused by defective channel proteins. Despite this motivation, there have been relatively few reports of anionophore natural products or of anion transport by synthetic systems. Most, moreover, have involved cationic centers, which can assist anion passage through the formation of an ion pair. Anion transport by purely electroneutral systems is still quite rare. The recently described “cholapod” anion receptors 1 are intrinsically lipophilic, potentially membranesoluble, and tuneable to very high affinities. Certain cholapods have proved capable of “flippase” activity, that is, they can convey polar head groups across phospholipid membranes. We therefore supposed that electroneutral cholapods might transport inorganic anions by a “shuttle” mechanism, in the manner of cationophores such as valinomycin. We now report that cholapods 2 are indeed capable of transporting chloride ions across liposomal membranes and also across live cells grown as polarized epithelia. Receptors 2 were prepared from cholic acid 3 by conversion into diamine 4 and treatment with appropriate isocyanates. Initially, their anion-transport properties were studied by following chloride efflux from unilamellar vesicles using a chloride-selective electrode. More specifically, 30 mm unilamellar vesicles (200 nm mean diameter) were prepared by extruding a 7:3 mixture of 1-palmitoyl-2-oleoyl-sn-glycero3-phosphocholine (POPC) and cholesterol in aqueous NaCl (500 mm). This stock vesicle dispersion was dialyzed against aqueous NaNO3 (500 mm) to replace the external chloride ions with nitrate and then diluted with 500 mm NaNO3 to give a 1 mm total lipid concentration. As shown in Figure 1, addition of a solution of 2a–e in THF (4.0 mm final cholapod concentration) caused chloride efflux from the inner phase of the vesicles. The initial rates of chloride efflux and the association constants of 2a–e for Et4N Cl and Et4N NO3 in