Abstract
The natural product MGN-3 (Biobran) is a defatted, partially hydrolysed rice bran-derived hemicellulose enzymatically modified with an extract of Lentinus edodes. It has a high proportion of arabinoxylan. It has a protective action against intracerebroventricular streptozotocin-induced murine sporadic Alzheimer's disease and reverses spatial memory deficit in this disease model. The aim was to test the hypothesis that MGN-3 increases glutamatergic burst activity in human neuronal and glial cells by conducting an in vitro multielectrode array-based micro-electrophysiological study in a cultured mixture of human glutamatergic neurones, GABAergic neurones and astrocytes. The effects of MGN-3 at two concentrations, 0.750 g L(-1) and 0.375 g L(-1), and vehicle (control), on glutamatergic burst activity in a triculture of human-induced pluripotent stem cell (hiPSC)-derived GABAergic neurones, glutamatergic neurones and astrocytes were studied. The change in the number of glutamatergic bursts normalised to the vehicle control was analysed using a normal or Gaussian generalised linear model. This statistical model was highly significant (p = 1.468 × 10(-17)). Both MGN-3 concentrations were associated with highly significant main effects. These results provide strong evidence to reject the null hypothesis that MGN-3 does not affect glutamatergic burst activity in human neuronal and astrocytic cells. The study's strengths include the novel use of hiPSC-derived neurones and astrocytes and the robust statistical significance of the findings. Limitations include in vitro conditions that may not fully replicate in vivo conditions, potential variability in hiPSC-derived cell preparations, and the need to test other neuronal subtypes or additional doses to assess dose-dependent effects. These should be addressed in future studies.