Life-threatening agranulocytosis and hepatotoxicity during prophylactic administration of amodiaquine have led to its withdrawal. Agranulocytosis is thought to involve bioactivation to a protein-reactive quinoneimine metabolite. The toxicity of amodiaquine and the lack of cheap drugs have prompted a search for alternative antimalarial agents. The aim of this study was to determine the metabolism and neutrophil toxicity of amodiaquine, pyronaridine, and other related antimalarial agents. Horseradish peroxidase and hydrogen peroxide were used to activate drugs to their respective quinoneimine metabolites. Metabolites were trapped as stable glutathione conjugates, prior to analysis by LC/MS. Amodiaquine was metabolized to a polar metabolite (m/z 661), identified as a glutathione adduct. Tebuquine was converted to two polar metabolites. The principal metabolite (m/z 686) was derived from glutathione conjugation and side chain elimination, while the minor metabolite gave a protonated molecule (m/z 496). Only parent ions were identified when chloroquine, cycloquine, or pyronaridine was incubated with the activating system and glutathione. Calculation of the heat of formation of the drugs, however, demonstrated that amodiaquine, tebuquine, cycloquine, and pyronaridine readily undergo oxidation to their quinoneimine. None of the antimalarial compounds depleted the level of intracellular glutathione (1-300 microM) when incubated with neutrophils alone. Additionally, with the exception of tebuquine, no cytotoxicity below 100 microM was observed. In the presence of the full activating system, however, all compounds except chloroquine resulted in depletion of the level of glutathione and were cytotoxic. Pretreating the cells with glutathione and other antioxidants inhibited metabolism-dependent cytotoxicity. In summary, our data show that amodiaquine and related antimalarials containing a p-aminophenol moiety undergo bioactivation in vitro to chemically reactive and cytotoxic intermediates. In particular, pyronaridine, which is currently being investigated in humans, was metabolized to a compound which was toxic to neutrophils. Thus, the possibility that it will cause agranulocytosis in clinical practice cannot be excluded, and will require careful monitoring.