Paired-pulse modulation at individual GABAergic synapses in rat hippocampus.

1. Unitary inhibitory postsynaptic currents (uIPSCs) were recorded in synaptically coupled pairs of CA1 hippocampal interneurons and pyramidal neurons in rat brain slices by using dual patch-clamp techniques. Paired-pulse modulation of uIPSCs at individual GABAergic synapses was tested. 2. GABAergic synapses could be divided into two subgroups, high and low failure, depending on their failure rate. 3. The external Ca2+ levels modulate the failure rate of uIPSCs. In 0.51 mM Ca2+, low-failure pairs had a high-failure characteristic, whereas high-failure pairs had a low-failure characteristic in 8 mM Ca2+. The results suggest that uIPSC failures result from the Ca2+-dependent release mechanism rather than axon propagation failures. 4. Paired-pulse facilitation (PPF) occurred in high-failure pairs when the interspike interval was 20 ms. Paired-pulse depression (PPD) was not predominant in high-failure pairs. 5. Potency of uIPSCs, the average amplitude of non-failure events, was enhanced by PPF, suggesting that multiple synapses connect each pair. Differing numbers of activated synapses contributed to the variable amplitude of uIPSCs from a given pair. 6. PPD occurred in low-failure pairs at the tested range of interspike intervals (20-200 ms). The uIPSC2 after a large uIPSC1 was smaller than the uIPSC2 after a small uIPSC1, suggesting that PPD is use dependent and due to a decrease in the quantal content (m) after the first release. 7. In 8 mM Ca2+, PPD occurred in high-failure pairs and was larger in low-failure pairs, suggesting that the occurrence of PPF or PPD depends on the baseline release probability. 8. The GABAB receptor antagonist CGP 55845A (5 microM) decreased the baseline release probability of inhibitory synapses and attenuated PPD indirectly, rather than by blocking presynaptic GABAB autoreceptors.