a1–5 Representative recordings showing the effect of different treatments on NE-induced repetitive IPSC bursts in the BLA principal neurons. 1. NE application induced phasic IPSC bursts following an initial increase in IPSCs. 2. Inhibiting PV neuron-mediated transmission with the P/Q calcium channel blocker ω-agatoxin selectively blocked the NE-induced repetitive IPSC bursts, but not the initial increase in IPSCs. 3. Co-application of ω-agatoxin and the CB1 receptor agonist WIN 55,212-2 blocked both the initial increase in IPSCs and the repetitive IPSC bursts. 4. Pretreatment of a slice with the N-type calcium channel blocker ω-conotoxin selectively inhibited the initial increase in IPSCs induced by NE, but not the repetitive IPSC bursts. 5. Application of the CB1 receptor agonist WIN 55,212-2 blocked the NE-induced initial IPSC increase but did not affect the repetitive IPSC bursts. b, c Expanded traces of individual IPSC bursts in a1 (indicated with green and blue shading), showing the fast acceleration in the intra-burst IPSC frequency and resulting depression in the baseline holding current (indicated with dashed lines) induced by NE. d Histogram showing the distribution of phase durations of NE-induced repetitive IPSC bursts. 10-s bins from time −5 s to +5 s. (n = 38 bursts from 16 cells, 5 mice). e Mean (±SEM) instantaneous intra-burst IPSC frequency over the course of the NE-stimulated accelerating IPSC bursts (n = 38 bursts from 16 cells, 5 mice). f Time course of the effect of blocking P/Q-type calcium channels and activating CB1 receptors on sIPSC amplitude. The NE-induced plateau increase in sIPSC amplitude (mean ± SEM) was blocked by ω-agatoxin (P/Q blocker), while the peak increase, corresponding to the NE-induced initial increase in IPSCs shown in a2, was unaffected (n = 10 cells from 4 mice). The ω-agatoxin-insensitive IPSCs were blocked by CB1 receptor activation with WIN 55,212-2 (n = 8 cells from 3 mice), corresponding to the recording in a3. g Time course of the effect of blocking N-type calcium channels and activating CB1 receptors on sIPSC amplitude (mean ± SEM). Both treatments selectively eliminated the NE-induced initial increase in IPSCs, with little effect on NE-induced repetitive IPSC bursts (n = 12 and 10 cells from 4 and 3 mice, respectively), which corresponds to the recordings in a4 and a5. h Representative loose-seal extracellular recording showing the NE-stimulated repetitive bursts of action potentials in a PV interneuron. A burst of action potentials indicated by the blue box was expanded below to show the accelerating intra-burst spike frequency. i Autocorrelation showing the rhythmicity of NE-induced AP bursts in the PV interneuron shown in h. j Mean (±SEM) instantaneous intra-burst action potential frequency over the course of PV neuron action potential bursts induced by NE (n = 7 cells from 3 mice). k, l Mean (±SEM) change in sIPSC amplitude over time in response to NE (n = 16 cells from 5 mice), NE + α1A receptor antagonist WB4101 (n = 7 cells from 4 mice), α1A receptor agonist A61603 (n = 10 cells from 4 mice), NE + TTX (n = 10 cells from 5 mice), NE + Gq antagonist YM-254890 (n = 8 cells from 3 mice); paired t-tests (two-tailed): NE vs. baseline, p = 0.0003, A61603 vs. baseline, p = 0.023; One-Way ANOVA, F(4, 46) = 6.22, p = 0.0004, Dunnett’s multiple comparisons test, NE vs. NE + WB4101, p = 0.0098, NE vs. A61603, p > 0.99, NE vs. NE + TTX, p = 0.0052, NE vs. NE + YM-254890, P = 0.0078; *p < 0.05, **p < 0.01 vs. baseline, ##p < 0.01 vs. NE, ns not significant; values at time = 13 min were used to compare ω-agatoxin-sensitive IPSCs. Source data are provided as a Source Data file.