Calcium is a crucial element modulating the probability of neurotransmitter release, because its interaction with key components of the active site in the presynaptic terminal.

The increase of extracellular calcium will increase the equilibrium potential for this ion (ECa), so when the voltage gated calcium channels in the presynaptic terminal open when an action potential arrives, more calcium will enter the presynaptic active zones, evoking the release of a higher number of neurotransmitter vesicles. So increasing extracellular calcium increases neurotransmitter release. To preserve the ratio of divalent ions (a variable itself) constant, during experiments with changes in extracellular concentration of calcium, magnesium is accordingly changed. But the main character affecting PPR is calcium because its action onto the release machinery.

Classical explanation for paired-pulse facilitation is the residual calcium hypothesis. According to this hypothesis, during the first action potential there is an entrance of calcium in the presynaptic terminal. Most of these calcium ions will activate neurotransmitter release by its binding to the sensor for release (synaptotagmin), in a well-known cooperative manner. However, a small fraction of calcium ions will remain in the active zone as "passive ions", namely residual calcium. If there is a second action potential after some milliseconds (and assuming the same entrance of calcium), this new calcium income will sum up with the residual calcium (which needs more time to be removed as the inter-stimulus interval), thus activating an elevated number of vesicles to release. According with this hypothesis, if we reduce levels of extracellular calcium and so ECa, less calcium ions will enter the terminal after an action potential, but the same residual calcium is expected to remain. So, when we reduce extracellular calcium, residual calcium would become a more significant fraction of the local calcium signal after a second action potential. That means that the second action potential will be "more facilitated" with respect to the first one (>PPR) in this second “low calcium” scenario, than with higher extracellular calcium concentrations. This model predicts that PPR will decrease with higher extracellular calcium concentrations. However, it seems that for most synapses this hypothesis fails to explain paired-pulse facilitation, because residual calcium and PPF decay at different rates.

There are other models proposed to explain PPF. A "facilitator sensor" different than synaptotagmin has been suggested. This sensor would have slow kinetics of action, unable to work after a first action potential, but ready when a second action potential arrives, and somehow facilitating a greater amount of neurotransmitter release.

It has been also suggested that paired-pulse facilitation is dependent on endogenous calcium buffers. These buffers would have high affinity, so after a first action they would become saturated, but when a second action potential arrives after a short period, the kinetics of the buffer are slower than the inter-stimulus interval, so assuming that the same amount of calcium is entering the presynaptic terminal, less calcium will be "caught" by the buffers, so a greater number of ions will be available to active neurotransmitter release, and in this way facilitate the second response of the pair. An important prediction of this model is that PPR will increase with higher extracellular calcium concentrations.

In general, paired-pulse facilitation is a synapse-specific property. Mossy fiber to CA3 connections behave more accordingly to the "buffer saturation" hypothesis, because in this synapse, PPR does increase with increases in extracellular calcium concentration. However, this doesn’t happen in the Schaffer collateral to CA1 synapse, where PPR decreases with increases in extracellular calcium.

Hope this helps you to understand better the complexities of PPF, which even today, still remain very mysterious!

I would recommend you to consult the review from Regehr about short-term plasticity (Regehr Cold Spring Harb Perspect Biol. 2012).

Good luck with your PPF experiments!

Hello,

Diego's answer is quite correct but I 'd like to add a precision for people not very familiar with synaptic transmission. Synaptic transmission is thought to be mono-vesicular at most synapses. It means that when an action potential reaches a release site (a synapse) one vesicle would be released at most. Vesicle release is a stochastic phenomenon meaning that the release will obey to a probability law. When Diego speaks about an increase of the number of released vesicles during PPF, it does not mean that multi-vesicular release  has occurred. It means that the probability of release has increased at several sites (several synapse) and therefore the mean number of release vesicles has increased.

The difference may sound subtle but that an important point often ignored by people.

Regards

Hello,

just wanted to mention a nice review where the monovesicular release vs multivesicular release hypothesis is discussed. In this review, it is concluded that much evidence support a multivesicular release mechanism. In any case, the issue seems not to be definitely solved towards the monovesicular release manner.

See Xu friedman and regehr 2004 Physiol Rev.

Thank you so much Diego and Fabien. Very comprehensive and useful information!!