The slower recovery after EPSC depression in Munc13-1W464R calyces was more pronounced after 300 Hz trains (WT, τ1 = 0.23 s [71%], τ2 = 10.8 s [29%], n = 9; Munc13-1W464R, τ1 = 0.2 s [19%], τ2 = 3.6 s [81%], n = 7; Figure 5B). In mature calyces (P14–P17), we observed a significant reduction of the recovery rate of EPSCs in Munc13-1W464R calyces after AP trains of 100 and 300 Hz, as compared to WT calyces (300 Hz train; WT, τ1 = 0.12 s [34%], τ2 = 3.3 s [66%], n = 16; Munc13-1W464R, τ1 = 0.13 s [25%], τ2 = 4.55 s [75%], n = 26; Figures 5C and 3-MA molecular weight 5D). Therefore, a stimulatory

effect of Ca2+-CaM on Munc13-1 is important for the recovery of EPSCs after high-frequency stimulation in calyces before and after hearing onset. Depletion of the

readily releasable SV pool is thought to be a major cause of STD in the calyx of Held, (von Gersdorff et al., 1997; Weis et al., 1999; Wu and Borst, 1999), and presynaptic introduction of CaM inhibitors leads to stronger SSD (Hosoi et al., 2007; Lee et al., 2012). In light of the substantially slower recovery rate after depletion of the fast releasing SV pool in the this website Munc13-1W464R mice, we tested next whether the Ca2+-CaM interaction with Munc13-1 is critical for frequency dependent STD. We triggered presynaptic AP trains of different frequencies by stimulating the afferent fibers and measured EPSCs in P9–P11 mice. Cyclothiazide was not included in the bath solution in these experiments to prevent alterations of the presynaptic AP (Ishikawa and Takahashi, 2001) and release time course (Taschenberger et al., 2005). No differences between WT GBA3 and Munc13-1W464R synapses were detectable with regard to the SSD levels during trains of 25 APs at frequencies of 2–300 Hz (Figures 6A–6C). Likewise, no significant differences in the average amplitude of the first EPSC in a given train (100 Hz train; WT, 10.02 ± 0.72 nA, n = 8; Munc13-1W464R, 11.78 ± 1.51 nA, n = 8; p > 0.05) or in the time course of EPSC depression were detectable (Figures S3A and S3B), indicating that the total

RRP size as well as release probability (pr) under both resting and activated conditions are similar in WT and Munc13-1W464R calyces. Accordingly, paired-pulse ratios (PPR) were comparable between WT and Munc13-1W464R calyces (Figure 6E). To account for the possibility that postsynaptic receptor saturation and/or desensitization may have masked differences in glutamate release, we repeated above experiments in the presence of kynurenic acid (1 mM) in the extracellular solution to minimize such effects, but again failed to detect significant differences between WT and Munc13-1W464R calyces (Figures 6D, S3C, and S3D). In juvenile calyces of Held, inactivation of presynaptic Ca2+ currents contributes strongly to STD elicited by low-frequency AP trains (Xu and Wu, 2005).