The amplitude of unitary IPSCs (uIPSCs) at PC-PC recurrent synapses has been shown to be less than 100 pA in mice, being ∼90 pA at P4–P14

( Watt et al., 2009) and ∼60 pA at P15–P19 ( Orduz and Llano, 2007). Considering that a uIPSC is composed of multiple mIPSCs, these results indicate that mIPSCs originating from PC-PC recurrent collaterals must be smaller than 100 pA. Since the amplitude of large mIPSCs is ∼250 pA in control mice ( Figure 7C), it is unlikely that the large mIPSCs are caused by PC-PC recurrent collaterals. Thus, large mIPSCs are considered to arise from synapses from molecular layer interneurons (presumably BCs) to PCs. This argument is consistent with the previous reports that multiple synaptic vesicles are synchronously Obeticholic Acid datasheet released from BC terminals facing PC somata, which results in large-amplitude mIPSCs (sometimes ranging to 1 nA) in developing rat cerebellum ( Conti et al., 2004 and Llano et al., 2000). To directly examine whether GABAergic transmission is attenuated at putative BC to PC synapses in GAD67+/GFP mice, we made paired whole-cell recordings from a putative BC and a PC in control and GAD67+/GFP mice at P10–P13. We found that the amplitude of uIPSCs in GAD67+/GFP mice was significantly smaller than that of control mice (control: 1.05 ± 0.14

nA, n = 14; GAD67+/GFP: 0.58 ± 0.12 nA, n = 15; p = 0.019) (Figures 7E and 7F). The rise time (control: 1.3 ± 0.14 ms, n = 14; GAD67+/GFP: 1.4 ± 0.11 ms, n = 15; p = 0.526), decay time constant (control: MAPK Inhibitor Library solubility dmso 5.0 ± 0.33 ms, n = 14; GAD67+/GFP: 5.7 ± 0.29 ms, n = 15; p = 0.140), and paired-pulse ratio (control: 0.80 ± 0.02, n = 14; GAD67+/GFP: 0.77 ± 0.03, n = 15, p =

0.461) of uIPSCs were not different. These results strongly suggest that GABAergic transmission at putative BC to PC synapses is attenuated in GAD67+/GFP mice. Finally, we investigated many how diminished GABAergic transmission can affect CF synapse elimination in GAD67+/GFP mice. In the standard ACSF without GABAergic blockers, we recorded CF-induced EPSPs and Ca2+ transients simultaneously from the soma of PC that was multiply innervated by a single “strong” CF (CF-multi-S) and one or two “weak” CFs (CF-multi-W) (Figures 8A and 8B). EPSPs elicited by activation of CF-multi-W were often too small to generate action potentials and to induce detectable Ca2+ transients in PCs. Therefore, we focused on CF-multi-W that could induce EPSPs large enough to elicit action potentials and CF-multi-S that inevitably elicits typical complex spikes. Stimulation of such CF-multi-W induced Ca2+ transients in the PC soma but did not elicit measurable Ca2+ elevation in PC dendrites (Figure 8B), which is consistent with the fact that CF-multi-W does not undergo dendritic translocation but stays on the PC soma (Hashimoto et al., 2009a and Hashimoto and Kano, 2003). Integration of Ca2+ transients (for 1.