, 1979, Schonewille et al., 2006 and Thach, 1968); hence, in previous work in vitro we selected for analysis Purkinje cells that generate stable and continuous action potential firing (Nolan et al., 2003). In this study, we find that 37%–50% of Purkinje cells from control Nfascfl/− (TAM) and TCE/Nfascfl/− (Oil) mice maintained spontaneous

action potential firing for > 10 min ( Figures 6A, 6B, and 6D). The properties of these spontaneous action potentials did not differ between the two control groups and were similar to spontaneous action potentials recorded previously ( Häusser www.selleckchem.com/products/abt-199.html and Clark, 1997 and Nolan et al., 2003), indicating that tamoxifen and the expression of Cre do not on their own affect spike firing ( Figure S3). In contrast to the control groups, none of the Purkinje cells in which the initial segment was disassembled were able to maintain spontaneous action potential firing for > 10 min (χ2(2, n = 82) = 14.98, p < 0.005) ( Figures 6C and 6D). Instead, these neurons maintained a modal resting potential of −43.0 ± 1.8 mV (n = 19) ( Figure 6). Using less restrictive criteria, in which we simply compared the number of cells that could generate spontaneous action potentials at any point during a recording, we found that a majority of Purkinje cells from INCB018424 chemical structure Nfascfl/− (TAM) (73.08%, 19/26), and TCE/Nfascfl/− (Oil) groups (68.75%, 22/32) fired

action potentials spontaneously, whereas spontaneous action potential firing was rarely observed in Purkinje cells obtained from TCE/Nfascfl/− (TAM) mice (8.33%, 2/24). Nevertheless, Purkinje cells from all TCE/Nfascfl/−

(TAM) mice were able to generate action potentials in response to positive current steps ( Figure 7C), indicating that an intact initial segment is not required for evoked action potential firing. We next asked if disassembly of the initial segment alters the heptaminol properties of evoked action potentials. For these experiments, the membrane potential was adjusted to −60 mV by injection of a negative holding current and action potentials were evoked by superimposed positive current steps (Figures 7A–7E). The current required to drive spike firing at a frequency of 52–58 Hz was significantly greater in TCE/Nfascfl/− (TAM) mice compared with either control group ( Figure 7F). Comparison of the waveform of action potentials evoked at this frequency revealed that deletion of Nfasc186 leads to a reduced maximum rate of rise ( Figure 7G) and peak amplitude ( Figure 7H) and increased half-width ( Figure 7I), but no difference in the peak membrane potential of the afterhyperpolarization (AHP) ( Figure 7J) although there was an increase in the delay until the peak of the AHP ( Figure 7K). There was no significant difference between groups in the frequency of spikes chosen for analysis (F2,18 = 0.94, p = 0.