1. Intracellular recording techniques were used to investigate the physiological and pharmacological properties of stimulus-induced excitatory postsynaptic potentials (EPSPs) recorded in regular-spiking cells located in layers II/III of rat sensorimotor cortical slices maintained in vitro. 2. Depending on the strength of the extracellular stimuli, a pure EPSP or an EPSP-inhibitory postsynaptic potential sequence was observed under perfusion with normal medium. The EPSPs displayed short latency of onset [2.4 +/- 0.7 (SD) ms] and were able to follow repetitive stimulation (tested less-than-or-equal-to 5 Hz). Variation of the membrane potential (V(m)) revealed two types of voltage behavior for the short-latency EPSP. The first type decreased in amplitude with depolarization and increased in amplitude with hyperpolarization. In contrast, the second type behaved anomalously by increasing and decreasing in size after depolarization and hyperpolarization, respectively. 3. Several experimental procedures were carried out to investigate the mechanism underlying the anomalous voltage behavior of the EPSP. Results indicated that this type of V(m) dependency could be mimicked by an intrinsic response evoked by a brief pulse of depolarizing current and could be abolished by N-(2,6-dimethylphenylcarbamoylmethyl)triethylammonium bromide (50 mM). Furthermore, the EPSP was not sensitive to the N-methyl-D-aspartate (NMDA) receptor antagonist 3-((+/-)-2-carboxypiperazin-4-yl)-propyl-1-phosphonate (CPP, 10-mu-M). Thus the anomalous voltage behavior was attributable to the nonlinear current-voltage relationship of the neuronal membrane. 4. The involvement of non-NMDA receptors in excitatory synaptic transmission was investigated with their selective antagonist 6-cyano-7-nitroquinoxaline-2,3-dione (CNQX, 1-10-mu-M). This drug greatly reduced or completely blocked the EPSP in a dose-dependent manner (1-10-mu-M). The IC50 for the CNQX effect was approximately 2-mu-M. In the presence of CNQX (10-mu-M) and glycine (10-mu-M), synaptic stimulation failed to elicit firing of action potential. However, a CPP-sensitive EPSP was observed. 5. When synaptic inhibition was reduced by low concentration of bicuculline methiodide (BMI, 1-2-mu-M), extracellular stimulation revealed late EPSPs (latency to onset: 10-30 ms) that were not discernible in normal medium. Similar to the short-latency EPSP, the V(m) dependency displayed by this late EPSP could be modified by inward membrane rectifications. The late EPSP appeared to be polysynaptic in origin because 1) its latency of onset was long and variable and 2) it failed to follow repetitive stimuli delivered at a frequency that did not depress the short-latency EPSP. 6. The role of NMDA receptors in mediation of the short-latency and late EPSPs evoked in the presence of BMI was examined. Bath perfusion of CPP (3-10-mu-M) progressively increased the latency to onset of the late EPSP while reducing its amplitude, until the response was completely abolished. In contrast, CPP attenuated the decay phase of the short-latency EPSP, whereas its peak amplitude remained insensitive to the drug. 7. Our data indicate that excitatory synaptic transmission in the sensorimotor cortex is predominantely mediated by the CNQX-sensitive non-NMDA receptors. Although a CPP-sensitive EPSP was observed in the presence of CNQX and glycine, the appearance of this NMDA-mediated potential might be explained by the effects of CNQX, which reduces the IPSP and glycine that augment NMDA-mediated responses. In addition, our results demonstrate that a slight reduction of inhibition by BMI is sufficient to reveal latent polysynaptic pathways that are mediated by NMDA receptors. Under such conditions, NMDA receptors can play a crucial role in neuronal synchronization.
