Role of redox cycling and lipid peroxidation in bipyridyl herbicide cytotoxicity. Studies with a compromised isolated hepatocyte model system.

The role of active oxygen species and lipid peroxidation in the toxic effects of diquat, paraquat and other bipyridyl herbicides remains controversial. In vitro studies have shown that these compounds are potent generators of active oxygen species by redox cycling and that they stimulate lipid peroxidation. In vivo studies have failed, however, to show clear evidence of lipid peroxidation resulting from toxic exposures to these compounds. We have directly compared the abilities of three bipyridyl herbicides, diquat (DQ), paraquat (PQ) and benzyl viologen (BV), to generate superoxide anion radical (O2-.) in rat liver microsomes and H2O2 in hepatocytes and correlated this with their relative toxicities to a compromised isolated hepatocyte system. DQ was the most potent generator of O2-. and H2O2, being slightly more potent than BV and much better than PQ. This ability of the bipyridyls to generate active oxygen was positively correlated with the ability to induce toxicity in hepatocytes pretreated with 1,3-bis-(2-chloroethyl)-1-nitrosourea (BCNU) to inhibit their glutathione reductase activity, i.e. DQ greater than BV greater than PQ. DQ caused a rapid depletion of cellular GSH and a concomitant increase in GSSG in this system. Toxicity, measured as loss of plasma membrane integrity, was pronounced after only 30-60 min of incubation and was accompanied by a significant increase in lipid peroxidation. The onset of lipid peroxidation could not be separated temporally from the expression of toxicity. However, the total inhibition of lipid peroxidation by the antioxidants Trolox C, promethazine and N,N'-diphenyl-p-phenylenediamine only delayed toxicity, indicating that, even though lipid peroxidation may play some role in enhancing bipyridyl herbicide toxicity, it is not essential for the toxicity to manifest itself.