The binding of the excitatory amino acid antagonist dl-2-amino-4-phosphonobutanoic acid (dl-APB) to rat brain synaptic plasma membranes was characterized. As determined by Scatchard analysis, the binding was saturable and homogeneous with &Kd = 6.0 μM and = 380 pmol/mg of protein. The binding was dependent on the presence of Ca2+ and Cl- ions and was diminished upon freezing. The association rate constant was 6.8 × 10-3 μM-1 min-1, and the dissociation rate constant was 2.0 × 10-2 min-1. The l isomers of APB, glutamate, and aspartate were more potent as displacers of APB binding than the d isomers. Previously determined inhibition data obtained for APB-sensitive inputs to hippocampal granule cells are compared to the present displacement data in an attempt to identify this binding protein as the recognition site of the receptor mediating the APB-induced inhibition of synaptic transmission. With the exception of kynurenic acid, all compounds examined in both systems were more potent as displacers of APB binding than as inhibitors of synaptic transmission. This difference in potency was most pronounced for agonists at dentate granule cells. l-Glutamate, d-glutamate, and l-glutamate tetrazole were between 140- and 7500-fold more potent as displacers of dl-APB binding than as inhibitors of synaptic transmission. d-2-Amino-5-phosphonopentanoic acid and a-methyl-APB were between 10- and 20-fold more potent as displacers of binding. Three monomethyl-substituted APB analogues (with methyl groups incorporated at the phosphonate, β, and γ positions) were between 25- and 30-fold more potent as displacers of APB binding than as antagonists of evoked responses in the lateral perforant path, while l-APB is equipotent in both assays. The parallel increases in potency of the methyl-substituted derivatives are the only observation that suggests this binding protein may be a modified form of the receptor mediating the inhibition of lateral perforant path responses by l-APB. On the other hand, the lack of a similar increase for l-APB itself, the significantly altered kinetics of association and dissociation, and the lack of correlation between the ligand specificity of this binding site and the currently accepted pharmacology of "glutamate" receptors do not support this hypothesis.