Furthermore, when we analyzed the responsive SC boutons, we found that a resting pool of ∼35% was induced (Figure 7A). In response to our 200 AP test stimulus, chronically depolarized boutons released a significantly smaller fraction of their vesicles (Figure 7C). The total number of vesicles did not change after chronic depolarization (Figure S6), and the time constant of endocytosis was not affected (p = 0.37) (Figures 7A and 7C). Together, these experiments suggest that mature SC synapses respond
to prolonged depolarization by removing vesicles from the recycling pool, thus gradually decreasing synaptic output. In a subset of synapses, this process leads to complete presynaptic silencing. PD173074 mouse Protein phosphorylation by cyclin-dependent kinase 5 (CDK5) and dephosphorylation by the calcium-dependent protein phosphatase calcineurin have been shown to regulate the balance between kinetically different modes of endocytosis (Evans and Cousin, 2007; Clayton et al., 2007; Sun et al., 2010) and to affect resting pool size (Kim and Ryan, 2010). At the calyx of Held, however, calcineurin loses its regulatory effect on endocytosis during maturation (Yamashita et al., 2010). We therefore tested whether
pharmacological inhibition of calcineurin by the specific blocker Fluorouracil order FK506 would still affect vesicle cycling at mature SC boutons. Calcineurin block did not significantly decrease resting pool size after 1.5–2 hr of incubation (Figure 7B). The time constant of endocytosis was slightly slowed in FK506, but this effect was also not significant (Figure 7C). Vesicle retrieval rates were not significantly affected by FK506 (RF/τ = 1.5%/s ± 0.48%/s, p = 0.07, n = 8 cells). Because of cell-to-cell variability in average resting pool size (Fernandez-Alfonso and Ryan, 2008), we did not pool boutons from different axons but treated 3-mercaptopyruvate sulfurtransferase every axon as an independent experiment
(n = #cells). Similar sized effects of FK506 with n = #boutons were found to be statistically significant (Marra et al., 2012). Interestingly, across all preparations and pharmacological treatments, the amount of vesicles released during high-frequency AP trains scaled linearly with recycling pool size: in response to 200 APs, synapses on average released ∼30% of their respective recycling pools (slope of the linear fit in Figure 7D). Thus, resting/recycling pool partitioning is likely to affect synaptic output, emphasizing the physiological relevance of the developmental regulation of recycling pool size. At mature synapses, however, drastic measures, such as chronic depolarization, have to be taken to induce a resting pool and thus decrease synaptic strength.