We analyzed determinants of paired pulse facilitation (PPF) at Purkinje neuron (PN) to PN synapses by quantifying presynaptic Ca²⁺ signals and quantal release parameters in wild-type (WT), calbindin (CB^-/-) and parvalbumin (PV^-/-) deficient mice. Incorporating the data into a reaction-diffusion model of Ca²⁺ -dependent vesicular release allowed us to estimate the relative contributions of residual free Ca²⁺ ("residual Ca^2+"), of CB and PV, and of Ca²⁺ bound to the release sensor ("active Ca^2+") to PPF. Surprisingly (1–3), none of the endogenous buffers significantly influenced the paired pulse ratio (PPR, 1.46 ± 0.25 in WT at 200 Hz), indicating that neither buffer saturation nor a residue of free Ca²⁺ was the main determinant of PPF. Rather, our analysis revealed that at the estimated influx-release coupling distance of 35 nm free Ca²⁺ in response to the first action potential (AP) was elevated for only 2 ms without any substantial residue of Ca²⁺ or saturation of buffers afterwards. In consequence, at the interstimulus interval of 5 ms, residual Ca²⁺ was at rest prior to the second AP. On the other hand, the active Ca²⁺, i.e. the Ca²⁺ occupancy of the release sensor, outlasted free Ca²⁺ by 6–8 ms, leading to residual sensor occupancy of 15% for the second AP and an increase in release probability that is in accord with the experimental results. We conclude that facilitation is not determined by residual free Ca²⁺ or by buffer saturation but by 'residue of the active Ca^2+'(4), most likely at the release sensor.

1. O. Caillardet al., PNAS 97, 13372 (2000).

2. M. Blatow, A. Caputi, N. Burnashev, H. Monyer, A. Rozov, Neuron 38, 79 (2003).

3. E. Eggermann, P. Jonas, Nat Neurosci 15, 20 (2011).

4. B. Katz, R. Miledi,J Physiol 195, 481 (1968).

Supported by DFG grants to JE and HS (El 342/4-1) and SH (HA 6386/2-1).