Both sorbitol accumulation-linked osmotic stress and "pseudohypoxia" [increase in NADH/NAD+, similar to that in hypoxic tissues, and attributed to increased sorbitol dehydrogenase (1-iditol:NAD+ 5-oxidoreductase; EC 1.1.1.14; SDH) activity] have been invoked among the mechanisms underlying oxidative injury in target tissues for diabetic complications. We used the specific SDH inhibitor SDI-157 [2-methyl-4(4-N,N-dimethylaminosulfonyl-1-piperazino)pyrimid ine] to evaluate the role of osmotic stress versus "pseudohypoxia" in oxidative stress occurring in diabetic precataractous lens. Control and diabetic rats were treated with or without SDI-157 (100 mg/kg/day for 3 weeks). Lens malondialdehyde (MDA) plus 4-hydroxyalkenals (4-HA), MDA, GSH, and ascorbate levels, as well as the GSSG/GSH ratios, were similar in SDI-treated and untreated control rats, thus indicating that SDI-157 was not a prooxidant. Intralenticular osmotic stress, manifested by sorbitol levels, was more severe in SDI-treated diabetic rats (38.2+/-6.8 vs 21.2+/-3.5 micromol/g in untreated diabetic and 0.758+/-0.222 micromol/g in control rats, P<0.01 for both), while the decrease in the free cytosolic NAD+/NADH ratio was partially prevented (120+/-16 vs 88+/-11 in untreated diabetic rats and 143+/-13 in controls, P<0.01 for both). GSH and ascorbate levels were decreased, while MDA plus 4-HA and MDA levels were increased in diabetic rats versus controls; both antioxidant depletion and lipid aldehyde accumulation were exacerbated by SDI treatment. Superoxide dismutase (superoxide:superoxide oxidoreductase; EC 1.15.1.1), GSSG reductase (NAD[P]H:oxidized-glutathione oxidoreductase; EC 1.6.4.2), GSH transferase (glutathione S-transferase; EC 2.5.1.18), GSH peroxidase (glutathione:hydrogen-peroxide oxidoreductase; EC 1.11.1.9), and cytoplasmic NADH oxidase activities were increased in diabetic rats versus controls, and all the enzymes but GSH peroxidase were up-regulated further by SDI. In conclusion, sorbitol accumulation and osmotic stress generated oxidative stress in diabetic lens, whereas the contribution of "pseudohypoxia" was minor. SDIs provide a valuable tool for exploring mechanisms of oxidative injury in sites of diabetic complications.