MARTIN BOOTMAN INTRACELLULAR CALCIUM Questions about quantal Ca 2+ release Several suggestions have been put forward to explain how the release of Ca 2 + from intracellular stores can be graded, but the mechanism underlying this phenomenon remains elusive. Our understanding of the way in which hormones gen- erate complex intracellular Ca 2+ signals has been greatly enhanced by recent studies investigating the regulation of inositol 1,4,5-trisphosphate (InsP 3 )-induced Ca 2+ release from intracellular stores. The receptors for InsP 3 , which mediate the Ca 2+ release, are controlled by several factors including covalent modifications result- ing from phosphorylation or thiol-group oxidation, and also by allosteric interactions with cytoplasmic Ca 2+ and adenine nucleotides [1]. In many cell types, the amplitude or spatial location of hormone-evoked Ca 2+ signals can be smoothly graded, reflecting the partial release of the InsP 3 -sensitive Ca 2+ stores. This phenomenon, known as 'quantal Ca 2+ release', was first described by Muallem and col- leagues [2], five years ago, yet the underlying mecha- nism and its physiological significance still eludes explanation. How? Quantal Ca 2+ release from InsP 3 -sensitive stores has been demonstrated in several cell types, using both intact and permeabilized cells. The same mechanism may also control Ca 2+ release from ryanodine receptors, channels that release Ca 2 + in muscle and neuronal cells [3]. In its simplest form, quantal Ca 2 + release implies that low concentrations of either hormone or InsP 3 , applied to intact and permeabilized cells respectively, are unable to evoke the complete release of the entire intracellular Ca 2+ pool. One can imagine several trivial explanations for this observation, such as hormone/ InsP 3 receptor desensitization, or the rapid attainment of a steady state where Ca 2+ release and re-uptake are balanced. However, these obvious explanations have been tested and largely rejected, and we are left with two recurring models. The steady-state model, proposed by Irvine [4], suggests that the quantal response results from the rapid attenua- tion of Ca 2+ release from intracellular stores that are homogeneously sensitive to InsP 3 . In this scheme, the opening of Insl' 3 -sensitive Ca 2+ channels is allosterically regulated by both cytoplasmic InsP 3 and the Ca 2+ con- centration within the lumen of the stores. The effect of a submaximal InsP 3 concentration is to reduce the luminal Ca 2+ concentration to a point where the sensi- tivity of the channel to InsI 3 is decreased, and Ca2+ release stops. Addition of a higher InsP 3 concentration will reopen the channel, promoting more Ca 2+ release, until the luminal Ca 2+ concentration is further reduced and the channel closes again. The second scheme suggests that quantal Ca 2+ release results from the all-or-none emptying of discrete intra- cellular stores that vary in their sensitivity to InsP 3 . Low hormone concentrations release the most sensitive Ca 2+ stores, and higher hormone concentrations, which gen- erate larger increases in cytosolic InsP 3 levels, are required to recruit the less sensitive stores. These models are the cause of considerable debate, as both are supported by data from various experimental systems and it is unclear at present which scheme is correct. For example, a major component of the steady- state model is that changes in the luminal Ca 2+ content control the sensitivity of InsP 3 -induced Ca 2+ release. This effect, however, has not been consistently observed with the cell types so far tested [3]. The all-or-none model is not without problems either, as for this scheme to work, several conditions must be met. Firstly, InsP 3 receptors require a variable sensitivity to InsP 3 . Secondly, Ca 2+ release would need to be highly co-operative to allow individual stores to unload over a narrow range of InsP 3 concentrations. Finally, the Ca 2+ stores would have to be functionally or physi- cally discrete. The first two conditions have some experimental support, but the requirement for function- ally or physically discrete intracellular Ca 2+ stores is an interesting issue that has been brought sharply into focus by two recent papers investigating InsP 3 -mediated Ca 2+ release in permeabilized cells. Where? InsP 3 -sensitive Ca2+ stores are thought to be located within the endoplasmic reticulum (ER). This organelle usually appears as a continuous structure, suggesting that Ca 2+ stores could be homogeneously distributed within cells. However, observations that some Ca 2+ stores are InslP3-insensitive, and that these stores some- times transfer Ca 2+ to others that are InsP3-sensitive, argue that the Ca 2+ release might be localized to functionally discrete ER regions [5]. In many studies investigating the regulation of Ca 2+ release, cell permeabilization has been used to allow the introduction of water-soluble inositol phosphates into cells. Typically, emphasis was given to obtaining a high degree of permeabilization, rather than preserving the integrity of the intracellular milieu. However, Renard-Rooney and colleagues [6] have shown recently that the permeabilization procedure used for many studies of Ca 2 + release might dramatically alter the structural organization of the ER Ca 2+ pool, as the © Current Biology 1994, Vol 4 No 2 169 MARTIN BOOTMAN INTRACELLULAR CALCIUM