IntroductionGnetum africanum leaf is consumed as a vegetable in many parts of Africa. Many ethnobotanical surveys indicate its usage in the management of inflammation-related diseases (such as diabetes). However, mechanisms underlying its antidiabetic actions are poorly understood. MethodsThis study conducted phytochemical analysis and investigated mechanisms underlying anti-diabetic actions of aqueous extracts of G. africanum in cellular and high fat-fed mice models of diabetes. Reversed phase high performance liquid chromatography (HPLC) analysis, qualitative phytochemical screening and the estimation of the total flavonoid and total phenolic content were conducted. Mechanisms underlying insulinotropic actions of the extracts as well as its effects on cytotoxicity, cell viability, intracellular calcium and membrane potential were assessed. In vivo actions were evaluated in mice with diet-induced obesity-diabetes. ResultsG. africanum has total phenolic content of 1.24±0.23GE/g and total flavonoid content of 1.54±0.06QE/g. HPLC analysis revealed the presence of the presence of atropine, vicenin 3, vicenin 2, alpha-chaconine, isoswertisin and solanidine. The extract stimulated non-toxic, concentration-dependent insulin-release from BRIN-BD11 cells (1.3-fold to 3.3-fold, P < 0.05 to P < 0.001). As glucose concentration increased from 1.1 mM to 5.6 mM and 5.6 mM to 16.7 mM, these effects increased by 2.0-fold (P < 0.001) and 1.2-fold (P < 0.05) respectively. Insulinotropic effects increased in cells depolarised with KCl (30 mM, 1.3-fold, P < 0.05). These effects reduced in the presence of diazoxide (300 µM, 67 %, P < 0.001), verapamil (50 nM, 50 %, P < 0.001), and absence of extracellular calcium (55 %, P < 0.001). Membrane potential (48 %, P < 0.05) and intracellular calcium (34 %, P < 0.05) increased in cells incubated with the extract. The extract improved glucose tolerance (22.2 %, P < 0.01 at 150 mg kg-1 bw; 35.1 %, P < 0.001 at 300 mg kg-1 bw) and plasma insulin levels (1.4-fold, P < 0.05 at 150 mg kg-1 bw; 2.9-fold, P < 0.001 at 300 mg kg-1 bw) in high fat fed mice in a dose-dependent manner. ConclusionThese results suggest a role for the KATP-dependent pathway in G. africanum insulinotropic actions and encourage further development of the therapeutic potential of the extract.