Objective: To study the effect and mechanism of ginsenosides Rb1 on memory impairment of type 2 diabetes. Method: Sixteen 10-week C57BL mice were as normal group (N group) and 64 Kkay mice 10-week with high blood sugar (≥13.9 mmol·L-1) as model group (M group)

positive group (P group) and composite group with high

middle and low doses (C1

C2

C3 group). Fed the positive group was given rosiglitazone maleate tablets(3.0 mg·kg-1)

and composite group was given rosiglitazone maleate tablets(3.0 mg·kg-1) and ginsenosides Rb1 (60

30

15 mg·kg-1). Normal and model groups were given equal volume of saline. After two Morris water mazes

samples were collected and enzyme-linked immunosorbant assay (ELISA) method wasused to test plasma insulin and reverse transcription polymerase chain reaction (RT-PCR) was used to test 1-phosphatidylinosital 3-kinase(PI-3K )/p85. Result: Compared with normal group

model group used much longer time to find the plat(P<0.05)

had more active plasma insulin and PI-3K/p85(P<0.05). Compared with control group

there was no significant differences in positive group

while plasma insulin and PI-3K/p85 were contrary in composite group(P<0.05). Conclusion: Ginsenosides Rb1 can improve the study and memory level of type 2 diabetes mice with cognitive impairment based on the reducing of plasma insulin

and the possible mechanism may reduce the level of PI-3K/p85.