Careful analysis revealed a clear principle underlying the fine-scale organization of these inputs: synapses that are located near each other on the same dendritic branch exhibit a higher degree of temporal correlation than synaptic pairs on different dendrites. By blocking action potential firing or N-methyl-D-aspartate (NMDA) receptor activation in slices for several days we showed that this clustering of synaptic inputs is activity-dependent. Thus, by quantifying and comparing a large population of functional synaptic inputs across the dendritic arborization

(the “synaptome”; DeFelipe, 2010) of developing pyramidal neurons, we revealed that developing synapses are Lumacaftor chemical structure functionally clustered on developing dendrites and that clustering requires spontaneous activity. To monitor the spatiotemporal patterns of spontaneous synaptic activation in developing neurons we performed simultaneous patch-clamp recordings and calcium imaging of hippocampal CA3 pyramidal neurons in organotypic slices from neonatal rats (postnatal [P] 0–2, days in vitro [DIV] 2–4). Patch-clamp recordings in voltage-clamp mode

revealed spontaneously occurring synaptic currents, most likely representing unitary synaptic events (1.8 ± 0.62 Hz; mean ± standard deviation [SD] per cell), as well as bursts of synaptic inputs, previously described as giant depolarizing potentials (GDPs; Ben-Ari et al., 1989 and Bonifazi et al., 2009). We determined selleck inhibitor the occurrence of bursts using an adapted version of the Rank Surprise method (Gourévitch and Eggermont, 2007; for details see Experimental Procedures). Bursts of synaptic inputs occurred at a rate of 15.02 ± 2.06 min−1, which is in the range measured in previous in vitro and in vivo recordings (Ben-Ari et al., 1989 and Leinekugel et al., 2002). In fact, the distribution of burst interevent intervals (Figure S1 available online) was virtually identical to that previously described in the hippocampus of developing rats in vivo (P4–6; Leinekugel et al., these 2002), demonstrating that not only the general connectivity

(Frotscher et al., 1990 and Stoppini et al., 1991), but also fundamental functional parameters are maintained in the hippocampal slice culture preparation. Calcium imaging in apical dendrites within stratum radiatum and stratum pyramidale (<200 μm from the soma) revealed spontaneous local calcium transients that occurred at an average rate of 68 ± 43.8 min−1 mm−1 dendrite ( Figure 1A). The majority of local calcium transients were observed in dendritic shafts and not in spines, because there are only very few spines present on dendrites of CA3 pyramidal neurons during this developmental period. Global calcium transients, which can also occur spontaneously in developing CA3 pyramidal neurons and depend on action potential firing, were not observed, since the membrane potential was clamped at −55 mV, the resting membrane potential of neonatal CA3 pyramidal neurons ( Safiulina et al., 2006 and Sipilä et al., 2006).