We tested the abilities of two potent non-N-methyl-D-aspartate (non-NMDA) glutamate antagonists [2,3-dihydroxy-6-nitro-7-sulfamoylbenzo(F)quinoxaline (NBQX)] and [1-(4-aminophenyl)-4-methyl-7,8-methylene-dioxy-5H-2,3-benzodiazep ine hydrochloride (GYKI 52466)] to reduce neocortical infarction following 2 h of transient middle cerebral artery occlusion in a hypertensive stroke model in the rat and compared these effects against, and in combination with, a potent NMDA antagonist [(+)-5-methyl-10,11-dihydro-5H-dibenzo[a,d]cyclohepten-5,10-amine maleate (MK-801)]. In Expt. 1, an already established cytoprotective dose of Na(+)-NBQX (30 mg/kg i.p. x 3) was compared with saline (1 ml), the NMDA antagonist MK-801 (1 mg/kg i.p. x 3), and a combination of the same doses of both NBQX and MK-801. Initial doses were delayed to 90 min following occlusion with subsequent injections at the time of reperfusion and 30 min following reperfusion. Saline-treated rats sustained 181 +/- 32 mm3 (n = 15) of neocortical infarction (mean +/- SD). This was significantly reduced by NBQX to 137 +/- 25 mm3 (n = 15, p < 0.05) of damage. Neither MK-801 (170 +/- 33 mm3; n = 11) nor the combination of MK-801 and NBQX (169 +/- 20 mm3; n = 6) proved to be cytoprotective when given with a 90-min delay. In Expt. 2, NBQX (30 mg/kg) was dissolved (6 mg/ml) in 5% dextrose and compared with both saline and dextrose (1.2 ml) i.v. infusions given over a 4-h period starting 1 h after occlusion. Saline-treated rats had a mean infarct of 183 +/- 27 mm3 (n = 6), dextrose-treated had 200 +/- 30 mm3 (n = 9), while for NBQX-treated rats it was reduced to 129 +/- 60 mm3 (n = 10, p < 0.05). Intravenous NBQX precipitated into the renal tubules, causing nephrotoxicity. In Expt. 3, rats were given either saline (1 ml i.p.) or GYKI 52466 (10 mg/kg i.p.) at 30 and 90 min following occlusion and at 30, 90, and 150 min following reperfusion. Saline-treated rats sustained 187 +/- 27 mm3 of neocortical infarction (n = 7), while those treated with GYKI 52466 were protected, with 139 +/- 38 mm3 of infarction (n = 7, p < 0.05). A clinically useful role for alpha-amino-3-hydroxy-5-methyl-4-isoxazole propionate antagonists in embolic stroke is envisaged if nontoxic drugs can be developed, since cerebroprotection was achieved with delayed treatment with both of these lead compounds.