Presynaptic action potential waveform determines cortical synaptic latency

Non-technical summary Synaptic delay at cortical synapses is determined by the presynaptic release probability. We show here that the duration and amplitude of the presynaptic action potential also determine synaptic latency at neocortical and hippocampal excitatory synapses. Broadening the presynaptic spike with blockers of potassium channels increased latency by 1-2 ms. Decreasing the amplitude of the presynaptic action potential by partly blocking sodium channels reduced synaptic latency by similar to 0.5 ms. These changes may contribute to stabilization of synaptic timing during repetitive stimulation. The regulation of synaptic timing by these pharmacological agents could not be attributed to modulation of axonal conduction. Rather, the effects are compatible with modifications of the kinetics of the presynaptic calcium current. We conclude that synaptic latency at cortical neurons is not constant but dynamically regulated by presynaptic action potential waveform.Synaptic latency at cortical synapses is determined by the presynaptic release probability (P-r). Short- and long-term presynaptic plasticity is associated with modulation of synaptic delay. We show here that the duration and amplitude of the presynaptic action potential also determine synaptic latency at neocortical and hippocampal excitatory synapses. Blockade of voltage-gated potassium (Kv) channels with 4-aminopyridine or dendrotoxin-I, but not tetraethylammonium, induced a 1-2 ms shift in latency at excitatory synaptic connections formed by pairs of neocortical pyramidal neurons. 4-Aminopyridine or dendrotoxin-I, but not tetraethylammonium, increased the duration of the action potential recorded in the axon, suggesting that presynaptic spike duration is controlled by axonal Kv1 potassium channels. Spike width-dependent changes in latency have been identified at the mossy fibre-CA3 cell synapses and contribute to stabilization of synaptic timing during repetitive stimulation. The effects of presynaptic spike amplitude on synaptic latency were also examined. Decreasing the amplitude of the presynaptic action potential with 15-30 nm TTX reduced synaptic latency by similar to 0.5 ms. The regulation of synaptic timing by potassium and sodium channel blockers could not be attributed to modulation of axonal conduction. Rather, these effects are compatible with modifications of the kinetics of the presynaptic calcium current. We conclude that synaptic latency at cortical neurons is not constant but dynamically regulated by presynaptic action potential waveform.