Fig. 7: Modelling of synchronous and asynchronous release – generation of release probability maps for different active zone geometries.

a Schematics illustrating Ca2+ triggering of different release modes. Synchronous release occurs within several milliseconds after an action potential and is triggered by transient local Ca2+-nano/microdomains (~10–100 µM range) in the close vicinity of activated VGCCs (2–100 nm range). In contrast, asynchronous release occurs on a time scale of tens to hundreds of milliseconds and is triggered by longer-lasting global changes in presynaptic Ca2+ concentration (~1-5 µM range)1, 30. b Schematics of the modelling framework. Action potential-evoked Ca2+ dynamics in a presynaptic bouton (left) were simulated using the VCell computational environment on a 10 x 10 x 10 nm mesh (Methods). For each voxel located in the active zone, the obtained Ca2+ transients (middle) were used as an input into the allosteric model of Ca2+ activation of vesicular fusion42 (right) to compute the probability of synchronous release for individual RRP vesicles ps. As discussed in the text, the probability of asynchronous release pa was assumed not to depend on the vesicle location in the active zone and its value was constrained based on the experimental data (Methods). c Nine cases of considered active zone geometries (i to ix) and corresponding model-predicted maps of the overall release probability \({p}_{v}={p}_{s}+(1-{p}_{s}){p}_{a}\), and of the relative fractions of synchronous (blue) and asynchronous (red) release modes for the model with endogenous buffering (see also Supplementary Fig. 13 for the model with the intracellular pipette buffer).