Schaffer collateral and perforant path inputs activate different subtypes of NMDA receptors on the same CA1 pyramidal cell

Abstract

The two major inputs to CA1 pyramidal neurons, the perforant pathway (PP) that terminates on distal dendrites and the Schaffer collaterals (SCH) that terminate on proximal dendrites, activate both AMPA and N-methyl-D-aspartate (NMDA) receptors. In an in vitro slice preparation, the pharmacologically isolated NMDA receptor-mediated excitatory postsynaptic currents (EPSCs) (NMDA-EPSCs) of either pathway can be selectively activated onto a single CA1 pyramidal neuron. Analysis of the decay phase of PP and SCH NMDA-EPSCs revealed no significant difference in their time constants, suggesting no apparent different distribution in NR2-subunit composition in the NMDA receptors (NMDAR) activated by the two synaptic inputs. However, application of the NR2B-selective antagonist, ifenprodil, differently affected the NMDA-EPSCs activated by the PP and SCH inputs. The reduction of the PP responses was only 30% compared to 75% for the SCH responses. In addition, for both pathways, the ifenprodil-insensitive component of the NMDA-EPSCs had significantly more rapid decay kinetics than those prior to application of ifenprodil. Our results show a greater NR2B subunit contribution to the NMDA component of the SCH EPSC, compared to the NMDA component of the PP EPSC and that in single CA1 pyramidal neurons NMDA composition is anatomically specific to the afferent input.

Figure 1

Isolation of NMDA receptor-mediated EPSCs. NMDA-EPSCs are recorded in a single CA1 pyramidal neuron in response to stimulations of the PP (left) and SCH (right) inputs. Recordings in normal ACSF (control) show that both PP and SCH activations evoke in the CA1 neuron a fast AMPA receptor-mediated EPSC followed by a GABA_B receptor-mediated IPSC (asterisks). To obtain isolated NMDA-EPSCs, recordings are performed at V_h=−60 mV, in ACSF (low Mg²⁺), bicuculline methiodide (20 μM), CGP55845 (100 nM) and NBQX (12 μM). The NMDA-EPSCs are completely blocked by the application of DL-APV (100 μM).

Figure 2

Ifenprodil effects on NMDA-EPSC evoked charge and decay kinetics. (a) Example of ifenprodil effects on NMDA-EPSCs recorded in a single CA1 pyramidal neuron evoked by PP (left) and SCH (right) stimulations. In this neuron ifenprodil (10 μM) reduces the PP NMDA-EPSC charge by 35% (control 12.15 pA × ms; ifenprodil 7.38 pA × ms) and the SCH NMDA-EPSC charge by 77% (control 11.31 pA × ms; ifenprodil 2.56 pA × ms). (b) Averaged data from ifenprodil-mediated inhibition of the NMDA-EPSC charge. Ifenprodil reduces the PP NMDA-EPSC charge by 29.6 ± 4.4% and SCH NMDA-EPSC charge by 75.1 ± 4.07% (P <0.01, paired t-test; n = 10). Errors bars represent the s.e.m. (c) Ifenprodil effect on the kinetics of NMDA-EPSC decay phase. The ifenprodil-insensitive component of the NMDA-EPSCs activated by either PP and SCH pathways have more rapid decay kinetics compared to the corresponding NMDA-EPSCs measured in control. The traces are the result of averaging peak-aligned, normalized NMDA-EPSCs recorded in five neurons. The decay phase is best fit with a single exponential equation which is shown superimposed on the average NMDA-EPSC. Correlation coefficients (R) for PP were: 0.9805 (control) and 0.9852 (ifenprodil); and for SCH: 0.9771 (control) and 0.9659 (ifenprodil).

Figure 3

Time course of the effect of ifenprodil on NMDA-EPSC charge. Average responses of four CA1 pyramidal neurons to the application of ifenprodil (10 μM) and DL-APV (100 μM) on NMDA-EPSCs evoked by stimulation of both PP (empty dots) and SCH (filled dots). The percentage of NMDA-EPSC charge is calculated by taking the mean NMDA-EPSC charge during the first 5 min (control) as 100%.