BACKGROUND AND PURPOSE:Cilostazol may be effective in dementia associated with a cerebral ischaemia. In this study, we examined whether it exerts beneficial effects on learning and/or memory impairment induced by Aβ(25-35) in mice, and compared its effects with those of aspirin. EXPERIMENTAL APPROACH:Aβ(25-35) (9 nmol) was administered to mice i.c.v. Learning and memory behaviour were evaluated by measuring spontaneous alternation in a Y-maze and a step-down type passive avoidance test, on the 5th and 8th days after injection respectively. Levels of lipid peroxidation (malondialdehyde) and cytokines in the frontal cortex and hippocampus were measured 2, 3, 5 and 7 days after the Aβ(25-35) injection. The effects of repeated administration of cilostazol and aspirin (both at 30 and 100 mg·kg(-1), p.o.) on any changes induced by Aβ(25-35) were evaluated. KEY RESULTS:Repeated administration of cilostazol significantly attenuated the impairment of spontaneous alternation and the shortened step-down latency induced by Aβ(25-35) . Aspirin did not show any beneficial effect. A significant increase in the levels of malondialdehyde (MDA) and IL-1β (only measured in hippocampus) was observed 2, 3 and 5 days after the Aβ(25-35) injection in the frontal cortex and hippocampus. Repeated administration of cilostazol (100 mg·kg(-1)) completely prevented the increase in MDA levels but failed to antagonize the increase in the expression of IL-1β induced by Aβ(25-35). CONCLUSIONS AND IMPLICATIONS:These results suggest that the protective effect of cilostazol on Aβ(25-35)-induced memory impairment may be related to oxidative stress in the frontal cortex and the hippocampus.