Fig. 5: A homeostatic increase in AZ numbers compensates for the drop in release probability.
From: Homeostatic synaptic plasticity rescues neural coding reliability
A Left, experimental scheme. Postsynaptic responses to spontaneously occurring single-vesicle fusions (minis) were recorded in PNs in 2–4-day-old flies. Right, averaged minis in PNs of wt (black) and upon cac knockdown in ORNs (green). B Neither mini frequency (left) nor amplitude (right) differ significantly between genotypes (wt, n = 6; cacRNAi, n = 9 flies). C Left, experimental scheme. Nicotine (100 µM) was applied to the AL and currents were measured in PNs. Right, Driving cacRNAi with Orco-GAL4 in ORNs increases the PN current amplitude in response to a nicotinic puff. Together with the unaltered minis, this indicates an increase in postsynaptic receptor fields. GH146-QF drove QUAS-GFP (wt, n = 23; cacRNAi, n = 22 flies). Two-sided permutation test, p = 0.002. D Confocal microscopy shows a significant increase in the number and fluorescence intensity of individual Brp punctae at ORN-PN connections following cac knockdown. The average size of Brp punctae remained unaltered (wt, n = 7; cacRNAi, n = 10 flies). Two-sided two sample t-test, p = 0.02 (# of Brp punctae), p = 0.009 (Mean intensity). E Example confocal images of an individual plane through the antennal lobe in control (upper panel; orco-Gal4/+; GH-146-QF,QUAS-GFP/+) and cacRNAi flies (lower panel; orco-Gal4/ UAS-cacRNAi; GH146-QF,QUAS-GFP/+) stained against Brp (fire). To restrict the analysis to ORN AZs with an excitatory PN as postsynaptic partner, the Brp signal was overlaid with masks generated by imaging GFP under control of GH146 (gray). For all panels *p < 0.05, **p < 0.01, for detailed statistical analysis, see Table S1.
