Evidence for participation of lipid peroxidation and iron in diquat-induced hepatic necrosis in vivo.

The hepatic necrosis produced in Fischer-344 rats by diquat appears to be mediated by redox cycling of diquat with generation of reactive oxygen species. We have now tested the hypothesis that chelates of iron are important in the cytotoxicity of reactive oxygen species, possibly through initiating the cleavage of peroxyl bonds. Pretreatment with the iron chelator desferrioxamine, 0.24 mmol/kg intraperitoneally, attenuated the hepatic damage produced by diquat. No additional protection was provided by a second dose of desferrioxamine 2 hr after diquat or by administration of the iron chelator by a different route of administration (subcutaneously). Ferrous sulfate (0.36 mmol/kg, intraperitoneally) alone produced no hepatic injury, but when given 15 min before diquat, it potentiated hepatic injury and animal mortality. In contrast, biliary excretion of glutathione disulfide in response to administration of diquat was neither potentiated by pretreatment with FeSO4 nor diminished by pretreatment with desferrioxamine. The marked changes in hepatic injury produced by these pretreatments, without changes in glutathione disulfide production, indicate that shifts in thiol/disulfide equilibria are not likely to be initiating events in the pathogenesis of diquat-induced hepatic necrosis. Administration of a hepatotoxic dose of diquat quickly produced 5-fold stimulation of ethane and pentane expiration rates with return to control rates by 3 hr. Desferrioxamine markedly inhibited, and iron potentiated, hydrocarbon expiration in response to diquat. The parallel changes in diquat hepatic injury and ethane and pentane expiration rates in response to manipulation of iron availability suggest a possible causal role for Fenton chemistry and lipid peroxidation in diquat-generated, reactive oxygen-mediated hepatic injury in vivo.