The spark and its ember: separately gated local components of Ca(2+) release in skeletal muscle.

Amplitude, spatial width, and rise time of Ca(2+) sparks were compared in frog fast-twitch muscle, in three conditions that alter activation of release channels by [Ca(2+)]. A total of approximately 17,000 sparks from 30 cells were evaluated. In cells under voltage clamp, caffeine (0.5 or 1 mM) increased average spark width by 28%, rise time by 18%, and amplitude by 7%. Increases in width were significant even among events of the same rise time. Spontaneous events recorded in permeabilized fibers with low internal [Mg(2+)] (0.4 mM), had width and rise times greater than in reference, and not significantly different than those in caffeine. The spark average in reference rides on a continuous fluorescence "ridge" and is continued by an "ember," a prolongation of width approximately 1 microm and amplitude <0.2, vanishing in approximately 100 ms. Ridge and ember were absent in caffeine and in permeabilized cells. Exposure of voltage-clamped cells to high internal [Mg(2+)] (7 mM) had effects opposite to caffeine, reducing spark width by 26% and amplitude by 27%. In high [Mg(2+)], the ember was visible in individual sparks as a prolongation of variable duration and amplitude up to 1.2. Based on simulations and calculation of Ca(2+) release flux from averaged sparks, the increase in spark width caused by caffeine was interpreted as evidence of an increase in radius of the release source-presumably by recruitment of additional channels. Conversely, spark narrowing suggests loss of contributing channels in high Mg(2+). Therefore, these changes in spark width at constant rise times are evidence of a multichannel origin of sparks. Because ridge and ember were reduced by promoters of Ca(2+)-dependent activation (caffeine, low [Mg(2+)]) and became more visible in the presence of its inhibitors, they are probably manifestations of Ca(2+) release directly operated by voltage sensors.