Fig. 3

EV release in immune-activated microglia is dependent on glutamine metabolism. When BV2 medium was changed into serum-free glutamine-free DMEM, additional glutamine was added to the medium at concentrations of 1, 2, and 5 mM 6 h prior to LPS treatment. a, b Supernatants were collected and centrifuged at 1500 rpm for 5 min to remove cells. Samples were prepared for RP-HPLC, and the levels of glutamate and glutamine were determined. c, d Protein lysates were prepared from the whole cell lysates and EVs pellets. The levels of EVs markers Alix and flotillin-2 in EVs, as well as the levels of GAC and β-actin in whole cells, were determined by Western blot. EVs protein loading was normalized with protein concentrations in the whole cell lysates. e, f Densitometric quantifications of the protein levels in EV markers were presented as fold changes relative to that in mock-infected control EV lysates. Western blot results shown are representative of the three independent experiments. Quantification results were normalized to the glutamine 5 mM group; ANOVA and post-test were performed on the remaining groups. *, **, ***, and **** denote p < 0.05, 0.01, 0.001, and 0.0001, respectively, compared with that of control Gln-free-treated microglia cells, n = 3 per group (a–f). g EVs were isolated from culture supernatants and visualized through NanoSight. h Quantifications of NanoSight NTA of vesicle concentration for samples from LPS-treated BV2 cell with different glutamine concentrations. ANOVA and Bonferroni post-test; *, **, ***, and **** denote p < 0.05, 0.01, 0.001, and 0.0001, respectively, compared with that of control Gln-free-treated microglia cells, n = 5 per group