Montelukast treatment suppresses retinal and leukocyte markers of oxidative stress and inflammation. A: Mice were harvested at 3 months’ diabetes duration. Fresh retinas from each experimental group were isolated and evaluated for superoxide generation as measured by a lucigenin-based chemiluminescent technique. Retinas from mice with streptozotocin-induced diabetes (SD) demonstrated a more than twofold rise in superoxide levels compared with retinas from nondiabetic mice (ND). Montelukast treatment administered in the drinking water (SD + montelukast) prevented this diabetes-induced rise in superoxide. Retinas were assayed from five mice per experimental group. *P < 0.05, ND vs. SD and SD vs. SD + montelukast. B: Superoxide generation by isolated peripheral blood leukocytes was determined by a lucigenin-based chemiluminescent technique. Montelukast administration via drinking water to diabetic mice dampened the typical diabetes-induced rise in leukocyte superoxide production. Leukocytes were isolated from five to seven mice per experimental group. *P < 0.05, ND vs. SD and SD vs. SD + montelukast. C: The number of adherent leukocytes attached to the retinal microvasculature was determined for each experimental group as described in A. The diabetes-induced adherence of leukocytes to the microvasculature (representative image, yellow arrow) was reduced significantly in the presence of montelukast. Retinas were assayed from 4 to 5 mice per group. *P < 0.007, ND vs. SD and SD vs. SD + montelukast. D: Coculture of isolated mouse leukocytes with mouse mRECs was evaluated after 24 h for endothelial cell viability as described in Research Design and Methods. Death of mRECs increased when cocultured in the presence of leukocytes from diabetic mice compared with leukocytes from nondiabetic mice or montelukast-treated mice. Leukocytes from six mice per group were studied. HG, high glucose (25 mmol/L). *P < 0.001.