Fig. 4.
Slow frequency-dependent depression of inhibitory synapses in culture. A, Representative responses from three different pairs stimulated 1000 times at 2 (left), 10 (middle), or 20 (right) Hz. IPSCs were normalized to the baseline amplitude obtained from 10 responses during preceding stimulation at 0.2 Hz. Synaptic depression was characterized by a rapid phase of decay within ∼10 stimuli, followed by a slower decay. The amplitude of the slow decay was increasingly prominent at higher stimulation frequencies. All of the 1000 responses of the train are plotted as single points. Recovery from depression was assessed by 30 pulses at 1 Hz, starting 1 sec after the end of the train (action potentials 1001–1030). Action potentials and IPSCs from the 1st, 100th, 1000th, and 1030th action potential are displayed in theinset at all three stimulation frequencies. Top abscissa shows number of action potentials in the train and during recovery; bottom abscissa shows time during the train. Axis titles hold for all three graphs. B, Averaged waveforms of synaptic depression induced by 1000 stimuli applied at 2 (n = 8), 10 (n = 16), or 20 (n = 3) Hz. Averages of five consecutive (average starting with the second response) responses are displayed as single points. Steady-state amplitude decreased with increasing stimulation frequency. C, Plot of the first 25 responses in the train. Inset shows an overlay of the normalized mono-exponential curves fitted to the three datasets. The time course of fast synaptic depression was not dependent on frequency (see Results). D, Recovery from synaptic depression assessed with single stimuli delivered at 1 Hz, starting 1 sec after the end of the train. For the train at 2 Hz, recovery was complete with the first response after the train (n = 4). For 10 Hz, recovery showed a slowly relaxing component, with a time constant of 4.4 ± 0.7 sec (n = 9;arrow).