Climbing fiber-evoked endocannabinoid signaling heterosynaptically suppresses presynaptic cerebellar long-term potentiation

Abstract

Endocannabinoid signaling has been demonstrated to mediate depolarization-induced suppression of excitation at climbing fiber (CF) and parallel fiber (PF) synapses onto cerebellar Purkinje cells. Here, we show that CF-evoked release of cannabinoids (CBs) additionally suppresses a presynaptic form of long-term potentiation (LTP) at PF synapses. PF-LTP can be induced by 8 Hz PF tetanization but is blocked when the PF tetanization is paired with 4 or 1 Hz CF coactivation. CF activity can be substituted for by bath application of the CB receptor agonist WIN55,212-2 [R(+)-[2,3-dihydro-5-methyl-3-[(morpholinyl)methyl]pyrrolo[1,2,3-de]-1,4-benzoxazinyl]-(1-naphthalenyl) methanone]. In the presence of the CB1 receptor antagonist AM251 [N-1-(2,4-dichlorophenyl)-5-(4-iodophenyl)-4-methyl-N-1-piperidinyl-1H-pyrazole-3-carboxamide], CF activity no longer suppresses PF-LTP. Presynaptic potentiation can also be obtained by the adenylyl cyclase activator forskolin. WIN55,212-2 blocked this forskolin-mediated enhancement, showing that CB1 receptor activation interferes with the adenylyl cyclase-protein kinase A cascade, which participates in LTP induction. CF activity has been described to promote the induction of postsynaptic PF-long-term depression (LTD) and to impair postsynaptic PF-LTP. Our observation that CF activity blocks the induction of presynaptic LTP suggests that the CF input controls all forms of presynaptic and postsynaptic PF plasticity and that CF activity provides a "safety lock" to prevent an enhancement of transmitter release while postsynaptic AMPA receptor function is downregulated during LTD.

Figure 1.

CF activity suppresses the induction of presynaptic PF-LTP. A, PF-LTP can be induced by PF stimulation at 8 Hz for 15 s (n = 12). Each data point represents the average of three successive test responses evoked at 0.05 Hz. The traces on top show EPSCs before and after LTP induction. The arrow indicates the onset of tetanization. B, PF-LTP can be induced when BAPTA (30 mm) is added to the internal saline (n = 5). C, PPF ratio (EPSC 2/EPSC 1) from the LTP group shown in A. D, PPF ratio from the BAPTA group shown in B. E, PF-LTP is abolished when 8 Hz PF stimulation is paired with 4 Hz CF stimulation (n = 7). LTP suppression using paired CF stimulation can also be observed when PKC [19-36] (100 μm) is added to the internal saline (n = 6). F, PF-LTP is blocked when the 8 Hz PF stimulation is paired with CF stimulation at 1 Hz (n = 7). Error bars indicate SEM.

Figure 2.

The CB1 receptor antagonist AM251 rescues PF-LTP. A, PF-LTP is observed when PF and CF synapses are coactivated in the presence of AM251 (5 μm; n = 12). B, Bath application of AM251 does not alter EPSC amplitudes (n = 6). The horizontal bar indicates the period in which AM251 was bath applied. Error bars indicate SEM.

Figure 3.

The CB receptor agonist WIN55,212-2 blocks PF-LTP. A, Bath application of WIN55,212-2 (2 μm) depresses PF-EPSCs (n = 7). B, PPF ratio (EPSC 2/EPSC 1) from the group shown in A. The horizontal bar indicates the presence of WIN55,212-2 in the bath. C, PF-LTP induced by 8 Hz PF stimulation for 15 s is abolished in the presence of WIN55,212-2 (n = 17). Error bars indicate SEM.

Figure 4.

WIN55,212-2 prevents the activation of adenylyl cyclases by forskolin. A, Bath application of the adenylyl cyclase activator forskolin (50 μm) enhances PF-EPSCs (n = 6). B, PPF ratio (EPSC 2/EPSC 1) from the group shown in A. C, WIN55,212-2 (2 μm) abolishes the forskolin-mediated enhancement of PF-EPSCs (n = 6). The horizontal bars indicate the presence of WIN55,212-2 and forskolin, respectively, in the bath. Error bars indicate SEM.