revail under physiological conditions producing a net disinhibition of cortical PNs. To purchase Nigericin (sodium salt) determine the effects of CB1 activation on glutamatergic neurotransmission, NMDA EPSCs were recorded from PNs in layer II/III and layer V/VI before and after the application of WIN. There was no interaction between cortical layer and WIN treatment = 0.132, p = 0.723). However, there was a main effect of WIN to reduce EPSC area = 44.71, p,0.0001), and EPSCs were significantly reduced by WIN application in recordings from both layer II/III = 5.16, p = 0.0004) and layer V/VI neurons = 4.33, p = 0.0016). In a complementary set of experiments, the effects of CB1 activation on evoked and spontaneous IPSCs on layer II/III and layer V/VI neurons were compared. There was no interaction between cortical layer and WIN treatment = 4.29, p = 0.056), but there was a main effect of WIN to reduce IPSC area = 14.68, Endocannabinoid Modulation of Up-States p = 0.0016). A within-group planned comparisons found that IPSC area was reduced in recordings from layer II/III neurons = 4.60, p,0.001) but not layer V/VI = 1.15, p = 0.538). Data from recordings of sIPSCs confirmed the layer specific effects of CB1 activation. Overall, sIPSCs recorded from layer V/VI neurons were of higher amplitude than those recorded from layer II/III cells = 11.94, p = 0.0026). There was also a main effect of CB1 activation = 15.29, p,0.001), but this effect was layer specific as WIN treatment only reduced sIPSC amplitude in layer PubMed ID:http://www.ncbi.nlm.nih.gov/pubmed/19645691 II/III neurons = 3.47, p = 0.0051). The sIPSC interevent interval was significantly greater in recordings from layer V/VI neurons = 21.54, p,0.001), but there was no effect of WIN = 0.23, p = 0.636) on sIPSC IEI. In order to compare the effects of CB1 activation on glutamatergic and GABAergic synapses across cortical layers, the percent of inhibition was calculated for evoked currents,. WIN-induced inhibition of PSCs differed significantly between PSC type = 11.2, p = 0.002) and cortical layer = 10.47, p = 0.0027). Specifically, activation of CB1 produced less inhibition of IPSCs in layer V/VI than EPSCs within the same layer = 2.80, p = 0.017), and CB1-mediated inhibition of IPSCs in layer II/III was greater than in deep-layer cells = 2.90, p = 
0.013). These data support the hypothesis that GABAergic synapses onto layer II/III PNs are more sensitive to the effects of CB1 activation than GABA synapses on layer V/VI PNs. Likewise our results indicate that glutamategic synapses in different layers show similar sensitivity to activation of CB1. Antagonists of CB1 block the manifestation of neocortical up-states As discussed above, inhibitors of FAAH and MAGL augmented up-state amplitude and spiking suggesting a role for ECs in regulating this form of cortical network activity. To investigate the necessity of EC/CB1 signaling in generating up-states, time course experiments were performed where up-states were evoked every 30 sec for 55 min in the presence of three structurally similar, wellcharacterized, selective antagonists of the CB1 receptor. Specifically, the effects of 1 mM AM281, 1 mM AM251, and 0.1 mM NESS0327 were compared. The doses used for AM281 and AM251 were chosen based on commonly used concentrations in slice physiology experiments. Because NESS0327 is much more potent that the other two CB1 antagonists, an order of magnitude lower concentration was used. Application of any one of the CB1 antagonists significantly reduced up-state duration relative to ups