The peak calcium flux was calculated as the maximum slope of the CFCT, defined as the ratio of the amplitude to four times the fitted logistic exponential steepness (i.e., the derivative of the

logistic function at midpoint). The peak calcium flux of unitary transients (0.16 ± 0.01 ΔG/R·ms−1, n = 17) (Figure 5D) was not correlated with the somatic distance (r = −0.29, p = 0.29, n = 15) (Figure 5E), confirming that dendritic calcium spikes propagate without decrement in spiny dendrites. The peak calcium flux of control CFCTs was smaller (0.04 ± 0.01 ΔG/R·ms−1, n = 45, p < 0.001) and its amplitude distribution only slightly overlapped CAL-101 concentration with that of unitary spikes

(Figure 5D). A calcium flux larger than 0.12 ΔG/R ms−1 can thus be considered as a hallmark of calcium spikes. Control CFCTs with a fast rise time occurred mostly at proximal sites (gray circles, Figure 5E). The duration of calcium influx at these proximal sites (Figure 5B) is shorter than the inactivation of Apoptosis Compound Library manufacturer Cav3.1 channels (Hildebrand et al., 2009), which appear to carry most of the calcium flux (Figure 3F), and much shorter than the inactivation of P/Q channels. Hence, fast closure of T-type channels has to occur, most likely after regenerative repolarization of the proximal dendrites by a K+ conductance. A similar kinetic analysis cannot be performed in smooth dendrites, as intracellular diffusion of calcium will slow the fluorescence

transient rise. However, the amplitude of control CFCTs (<90 μm from soma) was found to be similar to that of the first unitary spikes in DHPG (control: 0.10 ± 0.02 ΔG/R versus DHPG: 0.12 ± 0.007 ΔG/R, n = 4, p = 0.53; paired t test). These results indicate that a dampened regenerative depolarization, similar to a spikelet, may occur in Sodium butyrate the smooth dendrites and proximal spiny dendrites before mGluR1 unlocking, as observed in dendritic electrophysiological recordings (Davie et al., 2008 and Kitamura and Häusser, 2011), but fails to propagate further. To better understand how dendritic spike unlocking can be controlled by the somatic holding potential, we determined the site of spike initiation by monitoring simultaneously the CFCTs in two spiny branchlets. In these paired optical recordings (Figure 5F), unitary transients (the first of the CFCT) always occurred earlier at proximal sites (latency from the first sodium spike 1.52 ± 0.12 ms; n = 4) than at distal sites (1.79 ± 0.19 ms, additional distance 28.2 ± 9.0 μm). This timing difference was not accounted by a change in the rise kinetics of the unitary transients (Figure 5F).