Roles of the fast-releasing and the slowly releasing vesicles in synaptic transmission at the calyx of Held

Abstract

In the calyx of Held, fast and slow components of neurotransmitter release can be distinguished during a step depolarization. The two components show different sensitivity to molecular/pharmacological manipulations. Here, their roles during a high-frequency train of action potential (AP)-like stimuli were examined by using both deconvolution of EPSCs and presynaptic capacitance measurements. During a 100 Hz train of AP-like stimuli, synchronous release showed a pronounced depression within the 20 stimuli. Asynchronous release persisted during the train, was variable in its amount, and was more prominent during a 300 Hz train. We have shown previously that slowly releasing vesicles were recruited faster than fast-releasing vesicles after depletion. By further slowing recovery of the fast-releasing vesicles by inhibiting calmodulin-dependent processes (Sakaba and Neher, 2001b), the slowly releasing vesicles were isolated during recovery from vesicle depletion. When a high-frequency train was applied, the isolated slowly releasing vesicles were released predominantly asynchronously. In contrast, synchronous release was mediated mainly by the fast-releasing vesicles. The results suggest that fast-releasing vesicles contribute mainly to synchronous release and that depletion of fast-releasing vesicles shape the synaptic depression of the synchronous phase of EPSCs, whereas slowly releasing vesicles are released mainly asynchronously during high-frequency stimulation. The latter is less subject to depression presumably because of a rapid vesicular recruitment process, which is a characteristic of this component.

Figure 1.

Synaptic response to a 100 Hz train of AP-like stimuli at the calyx of Held. A, A single 1 ms depolarizing pulse to +40 mV was applied to the presynaptic terminal (AP-like stimulus) to elicit a Ca²⁺ current (I_pre). An EPSC was evoked in response to Ca²⁺ influx. Quantal release rates (rel. rate) were estimated using the deconvolution method. The amount of release before and after the dotted line was termed as synchronous and asynchronous release, respectively. B, The same cell pair as in A. A 100 Hz train of AP-like stimuli was applied, which was followed by a step-depolarizing pulse (prepulse to +70 mV for 2 ms, followed by repolarization to 0 mV for 50 ms). The dotted line in the EPSC trace indicates the residual current component attributable to delayed clearance of glutamate as modeled by the deconvolution method (residual current) (Neher and Sakaba, 2001a). The cumulative amount of release was obtained by integrating release rates from the deconvolution method. Filled circles in the cumulative release plot indicate cumulative amount of synchronous release during the train. C, The same cell pair as in A, and a single step-depolarizing pulse (prepulse to +70 mV, followed by repolarization to 0 mV for 50 ms) was applied to the presynaptic terminal. In cumulative release traces, the black trace indicates an original cumulative release. The gray trace indicates the cumulative release that is corrected by the recruitment of new vesicles after depletion (Sakaba and Neher, 2001a). D, Cumulative release traces during a step-depolarizing pulse with (B) and without (C) a preceding 100 Hz train (black traces). The difference between the two traces is shown as a gray trace and was used to estimate how many fast-releasing and slowly releasing vesicles had been released during a 100 Hz train by fitting the time course.

Figure 2.

Correlation between the amounts of cumulative release and presynaptic capacitance responses. A, A 100 Hz train of AP-like stimuli was applied, and presynaptic calcium currents and EPSCs are shown. The dotted line in the EPSC trace indicates the current component attributable to delayed clearance of glutamate. B, Presynaptic capacitance (C_m), conductance (G_m), and series conductance (G_s) values before and after a 100 Hz train of AP-like stimuli as shown in A. C, Cumulative amounts of release estimated by the deconvolution method. The number of AP-like stimuli was varied from 1 to 20. The same cell pair as in A and B is shown. D, The same cell pair, but presynaptic capacitance changes in response to the trains are shown. The cross symbol indicates where the amplitude of capacitance increase relative to baseline was measured. E, Correlation between the amount of cumulative release and the capacitance jump elicited by a train, obtained from C and D.

Figure 3.

Release of the fast-releasing and the slowly releasing vesicles during a 100 Hz train of AP-like stimuli. A, Similar to Figure 1, but a 100 Hz train (for 200 ms) was applied twice with an interval of 500 ms in the presence of the calmodulin inhibitor peptide (calmodulin binding domain, 20 μm). The duration of the AP-like stimulus was 1 and 1.5 ms, in the first (left) and the second (right) stimulation protocol, respectively. After a train, a step-depolarizing pulse (prepulse to +70 mV for 2 ms, followed by repolarization to 0 mV for 50 ms) was applied to deplete the remaining pool of releasable vesicles. Dotted lines in the EPSC traces indicate the current component mediated by delayed clearance of glutamate. The left and right panels use the same scale bars (right), except in the insets. The inset in the left panel shows release rates evoked by the first AP-like stimulus (gray) and the average release rates of the last five AP-like stimuli (black) during the first train. Bi, Superimposed traces of presynaptic calcium currents during the first (gray) and the second (black) trains. Only the responses to the first AP-like stimulus are shown. Bii, Average amount of presynaptic Ca²⁺ influx in response to a single AP-like stimulus during the first and the second trains. Biii, Total amount of presynaptic Ca²⁺ influx during the first and the second trains. C, Cumulative amounts of release during the first and the second trains (solid traces). By subtracting the cumulative release during the second train (slowly releasing vesicles only) from that during the first train, the release time course of the fast-releasing vesicles was estimated (open circles). The amount of synchronous release (synch. rel.) during the first train is shown as filled circles. The amount of synchronous release during the second train is shown as filled triangles.

Figure 4.

The amount of fast-releasing vesicles during a 100 Hz train of AP-like stimuli. From the left, the amount of synchronous release, the amount contributed by fast-releasing vesicles, the total amount of release during the first 100 Hz train, and the total amounts of release including those during a depleting pulse are shown. Data were obtained by using the stimulation protocol shown in Figure 3. The average data are from n = 6 cell pairs. Error bars indicate SEM.

Figure 5.

Synaptic response to a 300 Hz train of AP-like stimuli. AP-like stimuli were applied 30 times to the presynaptic terminal at a frequency of 300 Hz. After the train, a single step-depolarizing pulse (prepulse to +70 mV for 2 ms, followed by repolarization to 0 mV for 20 ms) was applied to release remaining releasable vesicles.

Figure 6.

Release of the fast-releasing and the slow releasing vesicles during a 300 Hz train of AP-like stimuli. A, The same as Figure 3, but a 300 Hz train (for 100 ms) was applied twice with an interval of 500 ms in the presence of the calmodulin inhibitor peptide. From the top, presynaptic calcium currents (I_Ca), EPSCs, and release rates are shown. The left and right panels use the same scale bars. The expanded EPSC trace (left) and release rates (right) later during the train (time period indicated by a bar in the EPSC trace) are shown as an inset. The dotted line in the expanded release trace indicates a release rate of 0. Bi, Superimposed traces of presynaptic calcium currents during the first and the second trains. Only the responses to the first AP-like stimulus were shown. Bii, Average amount of presynaptic Ca²⁺ influx in response to a single AP-like stimulus during the first and the second trains. Biii, Total amount of presynaptic Ca²⁺ influx during the first and the second trains. C, Cumulative amounts of release during the first and the second trains (solid traces). By subtracting the cumulative release during the second train from that during the first train, the release time course of the fast-releasing vesicles was estimated (open circles), assuming that only the slowly releasing vesicles were isolated during the second train. The amount of synchronous release during the first train is shown as filled circles. The expanded cumulative release time course during the first train is shown as an inset.

Figure 7.

The amount of fast-releasing vesicles during a 300 Hz train of AP-like stimuli. From the left, the amount of synchronous release, the amount contributed by the fast-releasing vesicles, and the total amount of release during a train are shown. Data were obtained by using the stimulation protocol shown in Figure 6. The average data are from n = 9 cell pairs. Error bars indicate SEM.