Remacemide hydrochloride ((±)-2-amino- N-(1-methyl-1,2-diphenylethyl)-acetamide hydrochloride or FPL 12924AA) is a novel compound undergoing clinical trials for patients with generalized tonic/clonic and complex partial epilepsy. Remacemide exhibits efficacy against maximal electroconvulsive shock (MES) in rodents and seizures elicited by N- methyl- d , l-aspartate (NMDLA) in mice. Using rat synaptic membrane fractions, remacemide was shown to possess relatively weak noncompetitive binding to the ionic channel site of the NMDA ( N- methyl- d-aspartic acid ) receptor complex. With the hypothesis that activity against NMDLA-elicited seizures might be reflected by transformation to a more active metabolic species, the aim of the present study was to evaluate potential pharmacological effects of the 9 identified metabolites of remacemide which were all found in human and dog urine. Moreover, specific entities were recognized in plasma (including the rat's), as well as dog and rat cerebrospinal fluid. Five putative metabolites were also examined. A major route of metabolic transformation of remacemide in rats yields the formation of a pharmacologically active more potent desglycine derivative, namely FPL 12495 (±). Potency over the parent compound is revealed in the MES test in mice and rats, the NMDA-induced convulsions/ mortality test in mice, and especially involving in vitro displacement of MK801 binding to the channel subsite of the NMDA receptor. The S isomer (FPL 12859) of this desglycinate is even more potent, while the R isomer is less potent than the corresponding racemate. Unlike the non-competitive NMDA antagonist, MK801, these desglycinates did not prevent kindled seizures. Three other identified metabolites show efficacy in the mouse and rat in vivo tests, namely the N-hydroxy-desglycinate (FPL 15053) and the p-hydroxy-desglycinates (FPL 14331 and FPL 14465). FPL 15053 exhibited modest activity in all tests. The only in vivo activity exhibited by the 2 p-hydroxy-desglycinates was evidenced in the MES test following i.p. and i.v. dosing. However, FPL 14331 was active in the MK801 binding assay. An oxoacetate metabolite, FPL 15455, failed to demonstrate any biological activity. Of potential metabolites tested 2 β-Miydroxy-desglycinates (FPL 14991 and FPL 14981) displayed modest activity in the MES test, however, only FPL 14981 prevented NMDLA-induced convulsions/mortality in mice and was 2-fold more active regarding MK801 binding. The hydroxy-methyl derivative of remacemide (FPL 13592) and its desglycinate (FPL 15112) prevented MES-induced convulsions only after i.v. administration; only the desglycine derivative displaced MK801 binding. FPL 14467 ( p-dihydroxy-desglycine) was inactive in vivo and weak in the MK801 binding assay. The discrepancy between in vivo/in vitro activities for many of the compounds is presumably due to pharmacokinetic factors.