Abstract
1. The rate of calcium release (Rrel) from the sarcoplasmic reticulum (SR) in voltage clamped segments of frog skeletal muscle fibres was calculated from myoplasmic free calcium transients (.DELTA.[Ca2+]) measured with the calcium indicator dye Antipyrylazo III. 2. During a 100-200 ms depolarizing pulse Rrel reached an early peak and then declined markedly. The time course and extent of decline of Rrel were independent of membrane potential over a range of potentials where release activation varied severalfold. 3. For test pulses applied shortly after relatively large or long conditioning pulses Rrel completely lacked the early peak. The peak gradually recovered as the interval between the conditioning and the test pulses was increased to 1 s. 4. A latency was often observed before the start of recovery of the peak in Rrel. The latency appeared to be correlated with the time for .DELTA.[Ca2+] to fall below a certain level, indicating that recovery of the peak might represent reversal of a calcium-dependent process. It was therefore proposed that the rapid decline in Rrel during a pulse was due to calcium-dependent inactivation of the SR calcium release channels. 5. Inactivation continued to develop during the interval between a relatively large 20 ms conditioning pulse and a test pulse applied 20 ms later. This was as expected for calcium-dependent inactivation of SR calcium release becaused of the elevated [Ca2+] between the conditioning and test pulses. It was not as expected for external membrane potential-dependent inactivation. 6. Small steady elevations in [Ca2+] due to relatively small 200 ms conditioning pulses produced marked inactivation of Rrel, indicating an apparent dissociation constant for calcium-dependent inactivation only slightly above resting [Ca2+]. 7. All observations could be well simulated by a two-step model for inactivation in which myoplasmic free calcium equilibrates rapidly with a high-affinity calcium receptor on the release channel and then the calcium-receptor complex undergoes a slower conformational change to the inactivated state of the channel. 8. An alternative model in which calcium binds to a soluble receptor (e.g. free calmodulin) and then the calcium-receptor complex binds to and directly inactivates the channel was shown to be formally identical to the preceding model. Either model could closely simulate all observations.