Figure 7. Premotor circuit dysfunction causes altered NMJ bouton architecture via ROS-dependent activation of the JNK signaling pathway upon loss of eaat1.

(A–J) Representative confocal images of NMJs co-stained with α-HRP (red) and α-Dlg (green) obtained from third instar larvae of controls (w¹¹¹⁸) and the indicated genotypes. Scale bar: 10 µm. (K) Quantification data for the number of NMJ boutons per muscle area normalized to the value of controls (n ≥ 7 NMJs of A2 muscles 6 and 7 derived from n ≥ 7 animals for each genotype). (L–M) Loss of eaat1 increases mitochondrial ROS in motor neurons. (L) Representative confocal images of motor neurons of third instar larvae expressing mitotimer using D42-GAL4 obtained from controls (w¹¹¹⁸) and eaat1^hypo/hypo mutants. Scale bar: 5 µm. (M) The RFP/GFP ratio of mitotimer was quantified and normalized to the value of controls (n ≥ 10 animals for each genotype). (N,O) Expression of fos^DN in motor neurons rescues the NMJ bouton phenotype based on normalized quantal content, but does not affect premotor circuit dysregulation in eaat1^hypo mutants. (N) Quantification data for quantal content recorded from A3 muscle 6 of third instar larvae of controls (w¹¹¹⁸) and the indicated genotypes with 0.2 Hz electric stimulation in 0.5 mM Ca²⁺-containing HL3 solution (n ≥ 6 animals for each genotype). (O) Quantification data for overall firing time (for bursts of >15 s) per recording minute (n ≥ 8 animals for each genotype). (P) Quantification data for larval locomotion of eaat1^hypo mutants and eaat1^hypo mutants who express fos^DN in motor neurons (n ≥ 20 animals for each genotype). P values: ns, not significant; *, p<0.05; **, p<0.01; ***, p<0.001; ****, p<0.0001. n: replicate number. Error bars indicate SEM. Statistics: Student's t-test or one-way ANOVA with Tukey’s post hoc test.