Abstract: Objective To investigate the effects of metformin (Met) on the phosphatidylinositol-3-kinase (PI3K)/protein kinase B (Akt) pathway and cognitive dysfunction in Alzheimer's disease (AD) rats. Methods Fifty Sprague Dawley rats were selected and divided into the following groups: normal, AD model (AD group), Met low (50 mg/kg), Met medium (100 mg/kg) and Met high (200 mg/kg) according to the random number table method. Each group included 10 rats. With the exception of the normal group, AD models were established by injecting 10 μL streptozotocin (STZ, 3 mg/kg) into the bilateral ventricles. After the model was successfully established, Met at the corresponding dose was administered by gavage. The normal and AD groups were administered the same amount of normal saline by gavage. The medicine was administered once a dayfor 14 days. After the last administration, the Morris water maze test was used to detect the spatial cognitive function of rats. The rats were sacrificed, and the hippocampal tissues were collected. The contents of β-amyloid protein 42 (Aβ42) and phosphorylated tau protein (p-tau) in hippocampus were detected by enzyme-linked immunosorbent assay (ELISA). The pathological changes of neurons in hippocampus were observed by HE staining. The detection of PI3K positive expression was performed by immunohistochemistry. Western blots were used to detect the relative expression levels of PI3K, phosphorylated Akt (p-Akt), insulin receptor substrate-1 (IRS-1), and glycogen synthase kinase-3 (GSK-3) in hippocampus. Results Compared with the normal group, there were pathological damages, such as disordered arrangement and decrease of neurons in hippocampus of rats in the AD group. The number of times the platform was crossed, the ratio of swimming distance to the total distance in the original platform quadrant, the ratio of swimming time to the total time, and the protein expressions of PI3K, p-Akt and IRS-1 in hippocampus of the AD group were significantly low (P < 0.05). The average escape latency, the contents of Aβ42 and p-tau, and the protein expression of GSK-3 in hippocampus were high (P < 0.05). Compared with the AD group, the pathological damages of hippocampal tissue in low, medium, and high dose Met rats were alleviated. The number of times the platform was crossed, the ratio of swimming distance to the total distance in the original platform quadrant, the ratio of swimming time to the total time, and the protein expressions of PI3K, p-Akt and IRS-1 in hippocampus of the AD group were significantly high (P < 0.05). The average escape latency, the contents of Aβ42 and p-tau, and the protein expression of GSK-3 in hippocampus were low (P < 0.05). Moreover, the above indexes were dose-dependent in the Met groups. Conclusion Met can activate the PI3K/Akt pathway, decrease the levels of p-tau and Aβ42 in hippocampus, and improve cognitive impairment in AD rats.

Key words: Metformin, Alzheimer's disease, Phosphatidylinositol-3-kinase, Protein kinase B, Cognitive dysfunction, Rats