Atypical delayed paired-pulse depression at an identified synapse.

At chemical synapses, the interplay between the stimulation pattern, the dynamics of presynaptic calcium concentration and the use and replenishment of the vesicle pool causes plasticity phenomena such as synaptic facilitation and depression. These phenomena may coexist, with their relative contribution depending mostly on the initial release probability. Synaptic facilitation is caused by an increased probability of release as a result of presynaptic calcium accumulation, whereas synaptic depression is attributed to depletion of the releasable vesicle pool. Here, we show an unusual form of paired-pulse delayed depression at the synapse between pressure-sensitive and anterior pagoda neurons of the leech. Two presynaptic impulses with short (700 ms) interstimulus intervals produced delayed depression, that increased with the interval up to 1000 ms, but was reduced by pre-incubation with EGTA-AM. Depression emerged with an abrupt deviation from the facilitation decrease curve, and not with the continuous transition expected from a gradual decrease in release probability and a slow recovery of the vesicle pool. In addition, strong facilitation could be elicited after delayed depression, even when depression was increased by increasing previous release. Classical models of plasticity, solely considering vesicular depletion-replenishment and calcium-dependent changes in release probability, are unable to explain our data. We propose a new model that includes a slow calcium-dependent mechanism removing vesicles from the readily releasable pool with a delay after each presynaptic impulse. KEY POINTS: Synaptic plasticity produces dynamic changes in the efficiency of transmission at chemical synapses. Synaptic depression is usually attributed to depletion of the presynaptic pool of releasable vesicles. Here, we show an unusual form of synaptic depression, which occurs with a delay after previous release, depends on calcium and is not caused by depletion of the vesicle pool by its previous use. We propose a mechanism where vesicles are removed from the releasable pool after activation of a second messenger cascade.