Frequency jumps from 100 to 300 Hz. (A) Analogous to Figure 5C. Displayed are results from three types of trials: one where the frequency of stimulation was increased to 300 Hz after 500 ms at 100 Hz (green); one where the increase to 300 Hz occurred after 750 ms of 100 Hz; and one where stimulation was maintained at 100 Hz throughout, all as diagrammed at top. Mean responses are quantified as the quantal content of sequential 10 ms segments, allowing direct comparison of the rate of release when stimulation was 100 to when 300 Hz; single segments contain quanta released by a single action potential for times when the stimulating frequency was 100 Hz, and three consecutive action potentials when the frequency was 300 Hz. Boxes demarcate responses used to calculate the differential release in (B). (B) Differential release calculated as for Figure 5D; ‘M’ and ‘S’ signify the same as in Figure 5D. Note that the amount of differential release after 500 ms of 100 Hz stimulation was equivalent to after 750 ms, confirming that 100 Hz stimulation drives the standing fullness of the RRP to a steady state. (C) Average of traces during first 50 ms after jump to 300 Hz after 500 ms of 100 Hz stimulation. Blue is WT, magenta is QKO. Dashed light blue lines are baseline.
Alternate scenario: We use the results in Figure 5 and here to conclude that the probability of release of low vesicles is elevated at QKO synapses. We have additionally considered the alternate scenario where the low vesicles are not immediately releasable (Miki et al., 2016; Gustafsson et al., 2019). In this case, the ongoing synchronous transmitter release occurring after eliminating high vesicles could be explained by a fast recruitment mechanism that would transfer a tiny fraction of the vesicles from the un-releasable state to a releasable state in the time between action potentials. The fraction would be the value of for the low vesicles. However, this alternate scenario is not compatible with the results of the frequency jump experiments because the acceleration mechanism could not influence the rate of transfer from the un-releasable to releasable state until after the first 3.33 ms interstimulus interval during 300 Hz stimulation. In contrast, the rate of transfer during the first 3.33 ms interstimulus interval would have to be equivalent to the rate during the first 3.33 ms of the 20 or 10 ms interstimulus intervals during the preceding 50 or 100 Hz stimulation. Therefore, one would expect 6-fold or 3-fold fewer new releasable vesicles at the end of the first 3.33 ms interstimulus interval compared to before the preceding action potential. If so, the second pulse during 300 Hz stimulation would elicit paired-pulse depression rather than the robust facilitation seen in Figure 5D and (B). And, indeed, one would expect twice as much paired-pulse depression in Figure 5D compared to in (B) (i.e. because the transfer rate during 50 Hz stimulation would be half the rate when stimulation is 100 Hz).