Background: We have previously reported that 5-Aza-2-deoxycytidine (5-Aza-cdR) can repress protein serine/threonine phosphatase-1 gamma (PP1 gamma) expression and activity in the mouse hippocampus and affect the behaviour of mice in a water maze. It is well known that the phosphorylation of N-methyl-D-aspartate receptor 2B subunit (NR2B) plays a role in behaviour. In this study, we examined whether 5-Aza-cdR affects NR2B phosphorylation at tyrosine 1472 (p-Y1472 NR2B) and whether it affected the responses of the mice in a passive avoidance test. Methods: 5-Aza-cdR (10 mu M) was administered to mice via intracerebroventricular injection (i.c.v). The learning and memory behaviour of the mice were evaluated by measuring their response in a step-down type passive avoidance test 24 h after the injection. The mRNA level of NR2B was measured by real-time PCR. NR2B and p-Y1472 NR2B protein expression in the mouse hippocampus was detected by western blot and immunofluorescence. CDK5 activity was detected by the ADP-Glo(TM) + CDK5/p35 Kinase Enzyme System. To further clarify whether the 5-Aza-cdR effects on behaviour were dependent on cellular proliferation or not, the effect of 5-Aza-cdR on the expression level of NR2B, the phosphorylation level of p-Y1472 NR2B, cell viability and the cell cycle were analysed using the immortalized mouse hippocampal neuronal cells neural cell line HT22 treated with 10 mu M 5-Aza-cdR compared with an untreated control group. Results: After injection with 5-Aza-cdR, the behaviour of the mice in the step-down test was improved, while their phosphorylation level of p-Y1472 NR2B was increased and their CDK5 activity was decreased in the hippocampus. In vitro experiments showed 10 mu M 5-Aza-cdR increased the p-Y1472 NR2B phosphorylation level with inhibition of cell viability and cell cycle arrest. Conclusions: Our results suggested that the effect of 5-Aza-cdR on behaviour may be related to the increase in phosphorylation of p-Y1472 NR2B in the hippocampus.