Extended Data Fig. 8: Behavioral controls of glug KD in cortex glia and pyruvate imaging in glug and nebu KD (related to Fig. 5).
From: Glycolysis-derived alanine from glia fuels neuronal mitochondria for memory in Drosophila
a. Without induction of glug RNAi in cortex glia, memory after single-cycle training (n = 12, F2,33 = 0.40, P = 0.67) and spaced training (n = 10, F2,27 = 0.23, P = 0.79) was normal. b. The behavioral experiments were performed with a second RNAi against glug. Glug knockdown (KD) in adult cortex glia impaired memory after single-cycle training (n = 15, F2,42 = 4.88, P = 0.01) and spaced training (n = 14, F2,39 = 7.10, P = 0.002), but did not affect memory after single-cycle training followed by cold shock (n = 14, F2,42 = 0.32, P = 0.73) or massed training (n = 12, F2,33 = 1.06, P = 0.36). Without induction of glug RNAi in cortex glia, memory after single-cycle training (n = 10, F2,27 = 0.45, P = 0.64) and spaced training (n = 14, F2,39 = 1.11, P = 0.90) was normal. c. Single-cycle training elicited a faster pyruvate accumulation in MB neuron somata following sodium azide application (5 mM) as compared to non-associative unpaired training (n = 12, t22 = 3.89, P = 0.0008). Glug KD in adult cortex glia impaired the single-cycle induced increase in pyruvate accumulation in MB neuron somata following sodium azide application (n = 12, t22 = 0.66, P = 0.62). d. Nebu knockdown in adult cortex glia did not impair the single-cycle-induced increase in pyruvate accumulation ta following sodium azide application in MB neuron soma (n = 12, t22 = 3.21, P = 0.004) or in MB neuron axons (n = 12, t22 = 2.73, P = 0.012). All data are presented as mean ± SEM. Asterisks (**P < 0.01; ***P < 0.001; ns: not significant, P > 0.05) illustrate the significance level of two-sided t-test, or of the least significant pairwise comparison following 1-way ANOVA.
