Figure 3

DFsn is required to restrain synaptic terminal growth at the NMJ. (a) Representative confocal images of muscle 4 synapses co-stained with DVGLUT (green) and FasII (red), in wild-type (WT), and DFsn^f06595/Df-Res [presynaptic rescue: P(f06595)/Df(2R)7872; elav Gal4/UAS-GFP-DFsn] third instar larvae. (b-d) Quantification of bouton number (b), number of branch points (c) and synaptic span (d) in WT, DFsn^f06595[P(f06595)/P(f06595)], DFsn^f06595/Df and DFsn^f06595/Df-Res third instar larvae; n = 36, 21, 23 and 29 cells, respectively. Both DFsn^f06595and DFsn^f06595/Df are significantly different from WT for all parameters measured (p < 0.001), while DFsn^f06595/Df shows a more severe phenotype in number of boutons and number of branching points than DFsn^f06595(p < 0.05). The morphological defects in DFsn^f06595/Df are rescued by presynaptic expression of GFP-DFsn in all parameters measured (p < 0.001 for DFsn^f06595/Df-Res versus DFsn^f06595/Df; p > 0.5 for DFsn^f06595/Df-Res versus WT).