Depletion of the endoplasmic reticulum Ca²⁺ stores in electrically non-excitable cells activates a Ca²⁺ entry mechanism known as ‘capacitative’ Ca²⁺ entry (CCE). Recent evidence has pointed to the widespread expression of Ca²⁺ entry channels that are not dependent on store depletion, but are activated by lipid messengers such as arachidonic acid (AA) (Shuttleworth & Thompson, 1999). We have investigated the properties of such a non-CCE influx pathway in HeLa cells.

The imidazole compound calmidazolium (10 µM) was used to stimulate phospholipase A₂ (PLA₂) and consequently activate the non-CCE channel. Intracellular Ca²⁺ ([Ca²⁺]_i) and strontium (Sr²⁺) concentrations were measured using a video imaging system. Cells were loaded with the ratiometric fluorescent dye fura-2 and fluorescent images were obtained by excitation at 340/380 nm. Mn²⁺ entry experiments were recorded as fura-2 quenches Mn²⁺ following excitation at the Ca²⁺-independent wavelength 360 nm. The calmidazolium-induced Ca²⁺ influx reached a peak amplitude ([Ca²⁺]_i) of 256 ± 31.77 nM (n = 30). Data are means ± S.E.M. This response was completely blocked by the PLA₂ inhibitor N-(p-amylcinnamoyl) anthranilic acid (n = 30) and mimicked by exogenous application of 1-10 µM AA. Both calmidazolium and AA gated non-CCE channels that permitted Ca²⁺, Mn²⁺ and Sr²⁺ entry. In contrast, CCE channels only permitted Ca²⁺ and Mn²⁺ entry. CCE was completely inhibited by 1 µM Gd³⁺ (IC₅₀ = 0.19 µM, n = 30) while non-CCE was uninhibited by Gd³⁺ ≤ 100 µM (n = 20). Both calmidazolium and AA caused mobilisation of Ca²⁺ from inositol 1, 4, 5-trisphosphate (InsP₃)-sensitive intracellular stores. This was not required for Ca²⁺ influx since co-application of calmidazolium with the InsP₃ receptor antagonists 2-aminoethoxydiphenyl borate (100 µM, n = 60) or caffeine (40 µM, n = 20), completely blocked Ca²⁺ release without affecting Ca²⁺ entry. Although CCE is autoinhibited during prolonged activation using the Ca²⁺ ATPase inhibitor thapsigargin (2 µM), non-CCE can still be activated after CCE run-down (n = 20), suggesting this entry is distinctively different from CCE. The Ca²⁺ influx mediated by CCE and AA were found to be additive in all cells (n = 20). Application of both cyclo-oxygenase and lipo-oxygenase inhibitors which prevent AA metabolite production had no inhibitory effect on calmidazolium-evoked non-CCE (n = 50 and 20, respectively). These data indicate that HeLa cells express distinct Ca²⁺ influx channels activated by store depletion (CCE) and AA (non-CCE), and that these two Ca²⁺ entry systems can operate independently of each other.

This work was funded by the BBSRC.