Evidence suggests that N-methyl-D-aspartate receptors (NMDARs) have a relatively high affinity for agonist compared with non-NMDA receptors. Dose-response curves constructed with sustained agonist application suggest that the 50% effective concentration (EC(50)) for peak glutamate-evoked current at NMDARs is 1 to 10 microM, whereas that of alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate (AMPA) receptors is approximately 500 microM. Given estimates of synaptic cleft glutamate concentration in the millimolar range, it would be expected that NMDARs would be saturated with agonist. However, studies of synaptic NMDAR responses indicate that these receptors may not be saturated during single release events at many synapses. To address this apparent contradiction, we have compared the glutamate dose-response curve for the peak NMDAR current generated by sustained glutamate application with that obtained during brief synaptic-like pulses of agonist. Our results using both recombinant and native NMDARs indicate a marked reduction in glutamate potency with reduced agonist application duration (EC(50) = 100 to 200 microM with 1 ms application). A kinetic model suggested that the reduction in potency with shorter agonist application duration could be attributed to the relatively slow activation and deactivation rates of the NMDARs. Comparison of room temperature to 37 degrees C indicated that NMDAR activation and deactivation were strongly accelerated by increased temperature. However, at 37 degrees C, we still observed a significant increase in potency with longer agonist application duration. We propose that glutamate has a relatively lower potency at NMDARs than previously thought from agonist application under equilibrium conditions. This lower potency would account for data that shows nonsaturation of NMDARs during synaptic transmission.