Exposure of rat hippocampal slices to 12-min hypoxia produced only mild neuronal damage, as 72% of all slices recovered their CA1-evoked population spike following a 30-min recovery period. However, when this hypoxic insult was administered in the presence of 2.5 microM kainate or AMPA, only 6 and 15% of the slices, respectively, recovered their neuronal function. This enhancement of hypoxic damage by kainate could be attenuated in a dose-dependent fashion by the kainate/AMPA antagonist GYKI 52466 but not by the competitive NMDA antagonist APV. Unexpectedly, the noncompetitive NMDA antagonist MK-801 also attenuated the kainate- and AMPA-enhanced hypoxic neuronal damage and was more efficacious than GYKI 52466. Considering (1) the ability of MK-801 to antagonize hypoxic neuronal damage in the absence or the presence of NMDA, kainate or AMPA; (2) the antihypoxic effect of MK-801 in the presence of APV + 7-chlorokynurenate, a pairing that supposedly blocks MK-801 binding to the NMDA receptor; (3) the ability of MK-801 to protect hippocampal slices against brain damage induced by depolarization + excitotoxin (50 mM KCl + mM glutamate for 60 min); and (4) the ability of diltiazem, an L-type calcium channel blocker, to protect hippocampal slices against hypoxic neuronal damage, we conclude that the mode of action of MK-801 cannot be explained by its NMDA receptor antagonistic properties alone. A possible blockade of Ca2+ channels, most likely of the L-type, by MK-801 should be considered along with other mechanisms.