Action potentials in excitatory L2/3 barrel cortex neurons of awake mice are driven by large and rapid depolarization of ∼10 mV in the 20 ms preceding spike initiation (Poulet and Petersen, 2008; Gentet et al., 2010) (Figure 6B). Although membrane potential fluctuations are in general highly correlated in nearby excitatory neurons, the postsynaptic potentials that drive AP firing are entirely specific for the spiking neuron and no correlated signal is seen in neighboring excitatory neurons during ongoing spontaneous activity in awake L2/3 mouse barrel cortex

(Poulet and Petersen, 2008; Gentet et al., 2010). These large and rapid depolarizations that drive spiking might result from the postsynaptic integration of one, or more, of these rare large-amplitude Cabozantinib molecular weight synaptic inputs specifically innervating the spiking neuron (Figure 6C). In future experiments, it might therefore be of key importance to better characterize selleckchem these large-amplitude synaptic connections examining their functional relevance in vivo and whether they preferentially occur within specific subnetworks. Of specific functional significance, excitatory L2/3 neurons in mouse primary visual cortex preferentially make synaptic connections with other excitatory neurons

sharing the same orientation tuning (Figures 6D and 6E) (Ko et al., 2011; Hofer et al., 2011). However, PV neurons receive uEPSPs from excitatory Oxalosuccinic acid neurons without orientation-specific connectivity (Hofer et al., 2011), consistent with the broad tuning properties of PV cells (Sohya et al., 2007) and the extremely high connectivity between excitatory neurons and PV neurons, which in itself precludes specificity (Figures 6D and 6E). Strongly connected subnetworks of L2/3 excitatory neurons with the same orientation preference may thus help drive these neurons to respond to specific visual stimuli escaping from strong, but weakly tuned, inhibition. Both the in vitro and the in vivo membrane potential measurements that we have discussed until now were recorded at the soma. It is interesting to record from the soma because it is electrotonically close to the axon

initial segment, where APs are typically initiated (Stuart and Sakmann, 1994). The somatic membrane potential of excitatory L2/3 neurons is therefore a good predictor of AP firing, which occurs at a relatively constant threshold potential (Azouz and Gray, 2000; Poulet and Petersen, 2008; Mensi et al., 2012). However, the synaptic conductances that drive membrane potential changes are distributed across the neuronal arborizations, often at large electrotonic distances. Most excitatory synapses are located on dendritic spines and many GABAergic synapses are also on dendrites (although not primarily on spines). The passive membrane properties of dendrites follow from the properties of the electrical cables (Rall, 1969; Jack et al., 1975; Spruston et al., 1994).