To quark or to spark, that is the question

A rise in the intracellular Ca2+ concentration ([Ca2+]i;) is the first and last signal that most animal cells generate during the span of their life. The question of how these whole-cell Ca2+ signals (e.g. for contraction, secretion, information processing) are generated, is still a matter of great debate. Advancements in optical techniques are constantly revealing new layers of complexity in the mechanisms responsible for cellular Ca2+ signals. For excitable cells, the question was at least partially answered a few years ago by the description of spatially confined Ca2+ transients in resting rat cardiac muscle cells, the so-called ‘Ca2+ sparks’ (Cheng et al. 1993). This initial observation in cardiac muscle cells was thereafter confirmed throughout the entire diversity of excitable cells, from cardiac and smooth to skeletal muscle and recently spatially restricted Ca2+ release was also observed in neuronal cells (for a review see Lipp & Niggli, 1996b). The idea of these Ca2+ sparks being the elementary building blocks of Ca2+ signalling has been established during recent years. The original concept was that the gating of a single sarcoplasmic reticulum (SR) Ca2+ release channel (ryanodine receptor or RyR) gives rise to the occurrence of Ca2+ sparks. Spatio-temporal recruitment of these Ca2+ sparks then leads to a global or whole-cell Ca2+ transient. An accumulating body of evidence, including recent papers on cardiac muscle (Lipp & Niggli, 1996a; Parker et al. 1996) and the report in this issue on skeletal muscle fibres (Shirokova & Ríos, 1997), suggests that Ca2+ sparks can be devolved into even smaller entities. Monitoring Ca2+ sparks by scanning along the transverse axis of cardiac myocytes revealed ‘sub-spark’ Ca2+ release with small, but still discrete amplitudes (Parker et al. 1996). Based on the lack of spatial inhomogeneities (i.e. occurrence of Ca2+ sparks) in Ca2+ signals triggered following photolytic liberation of caged Ca2+ in cardiac muscle cells we suggested the existence of fundamental Ca2+ signalling events (‘Ca2+quarks’, Lipp & Niggli, 1996a). These serve as the fundamental building blocks of Ca2+ signalling in excitable cells. The unresolved nanoscopic Ca2+ quark is proposed to reflect the gating of a single RyR, whereby recruitment of several quarks gives rise to the microscopic Ca2+ sparks. In skeletal muscle, Ca2+ sparks have also been described as elementary Ca2+ release events (Tsugorka et al. 1995) despite the fact of different excitation-contraction coupling (E-C coupling) mechanisms present in cardiac and skeletal muscle. In cardiac muscle, Ca2+-induced Ca2+ release (CICR) is responsible for both the initial and further Ca2+ release from the SR. Conversely, in skeletal muscle the primary step in E-C coupling is a purely voltage-dependent release mechanism originally suggested to recruit Ca2+ spark-like Ca2+ release from the SR. Further SR Ca2+ release, also manifested as Ca2+ sparks, is then triggered by CICR (Klein et al. 1996). This initial scheme is now challenged by Shirokova & Ríos (1997) in a paper in this issue on subcellular aspects of SR Ca2+ release in skeletal muscle fibres. Using confocal line scanning of volt clamped muscle fibres from the frog, they were able to resolve two different types of SR Ca2+ release. Threshold stimulation triggers the sparse occurrence of Ca2+ sparks (1 spark per 6s per triad) on top of a pedestal of Ca2+ increase which was of a very low amplitude (less than 10% of a typical spark) suggesting a second type, i.e. non-spark type, of SR Ca2+ release in skeletal muscle. The latter release was characterized by low amplitude fluctuations of [Ca2+]i at membrane voltages insufficient to trigger spark-type Ca2+ release. Increased depolarizations were indeed able to induce spark-type Ca2+ release as indicated by the occurrence of identifiable Ca2+ sparks and the associated increased fluctuations in [Ca2+]i over time. Besides voltage discrimination, the authors isolated non-spark-type Ca2+ release by applying tetracaine and D600. Tetracaine is an inhibitor of the fast Ca2+ release component in skeletal muscle which is attributed to CICR, but it has little effect on the steady-state Ca2+ release, which is under the control of the membrane voltage. A similar effect is described by Shirokova & Ríos. Application of tetracaine abolished Ca2+ sparks but did not suppress the non-spark release component. This inhibition could not be overcome by stronger depolarizations. In contrast, application of D600, a well-known inhibitor of the membrane voltage sensor, suppressed spark-type Ca2+ release (without inhibiting non-spark release) at low voltages, but this inhibition could be overcome by stronger membrane depolarizations. In summary, the work described above supports the concept of a twofold mechanism for skeletal muscle E-C coupling, but challenges the idea of the Ca2+ spark being the sole elementary event. The primary SR Ca2+release, i.e. the voltage-dependent step, is characterized by a lack of measurable discrete events, while the subsequent Ca2+-induced release occurs as distinct Ca2+ sparks. This points to a striking similarity between cardiac and skeletal muscle Ca2+ signalling, in that both use single-channel events (Ca2+ quarks) as the primary event. Subsequent regeneration occurs via clusters of RyRs (Ca2+ sparks). However, due to the nature of E-C coupling in skeletal muscle, Ca2+ signals will be solely triggered by voltage-operated Ca2+ quarks, whereas in cardiac muscle a CICR mechanism predominates. This hierarchy of events (i.e. quarks sparks global signals) is not unique to excitable cells. It has also been shown for Ca2+ signalling in non-excitable cells, which primarily relies on the gating of inositol 1,4,5,trisphosphate-operated Ca2+ release channels (Bootman et al. 1997; Berridge, 1997). It is tempting to speculate that future studies will confirm this hierarchical concept among all cell types, regardless of the precise coupling mechanism. This unification of ideas across excitable and non-excitable cells is striking given the different physiological consequences of [Ca2+]i signals in various cell types, but will certainly enhance our understanding of cellular Ca2+ signalling.