Metformin is the most frequently used drug for the treatment of type 2 diabetes in humans. However, only little is known about effects of metformin on brain metabolism. To investigate potential metabolic consequences of an exposure of brain cells to metformin, we incubated rat astrocyte-rich primary cultures with this compound. Metformin in concentrations of up to 30 mM did not acutely compromise the viability of astrocytes, but caused a time- and concentration-dependent increase in cellular glucose consumption and lactate production. For acute incubations in the hour range, the presence of 10 mM metformin doubled the glycolytic flux, while already 1 mM metformin doubled glycolytic flux during incubation for 24 h. In addition to metformin, also other guanidino compounds increased astrocytic lactate production. After 4 h of incubation, half-maximal stimulation of glycolysis was observed for metformin, guanidine and phenformin at concentrations of around 3 mM, 3 mM and 30 µM, respectively. The acute stimulation of glycolytic lactate production by metformin was persistent after removal of extracellular metformin and was also observed, if glucose was absent from the incubation medium or replaced by other hexoses. The metformin-induced stimulation of glycolytic flux was not prevented by compound C, an inhibitor of AMP-dependent protein kinase, nor was it additive to the stimulation of glycolytic flux caused by respiratory chain inhibitors. These data demonstrate that the antidiabetic drug metformin has the potential to strongly activate glycolytic lactate production in brain astrocytes.

To clarify discrepancies in the literature on the adverse effects of hydrogen peroxide on neurons, this study investigated the application of this peroxide to cultured cerebellar granule neurons with six assays frequently used to test for viability. Cultured neurons efficiently cleared exogenous H2O2. Although viability was not affected by exposure to 10 µM hydrogen peroxide, an exposure to the peroxide in higher concentrations rapidly lowered, within 15 min, the cellular 3-(4,5-dimethylthiazol-2-yl)-2,5-diphenyltertrazolium bromide (MTT) reduction capacity to 53% ± 1% (100 µM) and 31% ± 1% (1,000 µM) and the 3-amino-7-dimethylamino-2-methyl-phenazine hydrochloride (neutral red; NR) uptake to 84% ± 6% (100 µM) and 33% ± 1% (1,000 µM) of control cells. The release of glycolytically generated lactate was stopped within 30 min in neurons treated with 1,000 µM peroxide. In contrast, even hours after peroxide application, the cell morphology, the number of propidium iodide-positive cells, and the extracellular activity of the cytosolic enzyme lactate dehydrogenase (LDH) were not significantly altered. The rapid loss in MTT reduction and NR uptake after exposure of neurons to H2O2 for 5 or 15 min correlated well with a strongly compromised MTT reduction and a very high extracellular LDH activity observed after further incubation in peroxide-free medium for a total incubation period of 24 hr. These data demonstrate that cultured neurons do not recover from damage that is inflicted by a short exposure to H2O2 and that the rapid losses in the capacities of neurons for MTT reduction and NR uptake are good predictors of delayed cell damage.