Loss of Rut or Epac, but not PKA or the I _h channel, results in defective PTP. (A) Representative recordings of WT, rut¹/Y, Epac^Δ1/Epac^Δ3, I_h^e01599/I_h^f03355, OK6-GAL4/+, and OK6-GAL4/UAS-PKAinh¹ (OK6>PKAinh¹) before, during, and after the 10 Hz, 1 min PTP induction protocol (0.3 mM Ca²⁺; Fig. 5 A). Mean EJP amplitudes normalized to the initial EJP amplitude at 0.5 Hz (basal). Each point indicates the mean normalized amplitude of consecutive EJPs recorded every 10 s. (B–D) Bar graphs of EJP amplitudes in larvae of indicated genotypes at the end (B) and 60 s after (C) tetanic stimulation, and PTP decay time constants (D). n = 12 NMJs. (E) NMJs with FM1-43 (pink) after ECP-RP loading (E1), ECP unloading (E2), and RP unloading (E3; as in Fig. 5 F) shown for WT, rut¹/Y, Epac^Δ1/Epac^Δ3, OK6-GAL4/+, and OK6>PKAinh¹. (F) FM1-43 fluorescence intensity after ECP-RP loading (whole columns), ECP unloading (gray columns), and RP unloading (black columns). n = 12 boutons. Data represent mean ± SEM. Comparisons are with WT or OK6-GAL4/+ (**, P < 0.01; ***, P < 0.001 by one-way ANOVA with multiple comparisons). Scale bar: 5 μm.