Ghrelin signaling in the cerebellar cortex enhances GABAergic transmission onto Purkinje cells

Abstract

Ghrelin, an orexigenic peptide ligand for growth hormone secretagogue receptor 1a (GHS-R1a), occurs not only in the stomach but also in the brain, and modulates neuronal activity and synaptic efficacy. Previous studies showed that GHS-R1a exists in the cerebellum, and ghrelin facilitates spontaneous firing of Purkinje cells (PCs). However, the effects of ghrelin on cerebellar GABAergic transmission have yet to be elucidated. We found that ghrelin enhanced GABAergic transmission between molecular layer interneurons (MLIs) and PCs using electrophysiological recordings in mouse cerebellar slices. This finding was consistent with the possibility that blocking synaptic transmission enhanced the ghrelin-induced facilitation of PC firing. Ghrelin profoundly increased the frequency of spontaneous inhibitory postsynaptic currents (IPSCs) in PCs without affecting miniature or stimulation-evoked IPSCs, whereas it significantly facilitated spontaneous firing of MLIs. This facilitation of MLI spiking disappeared during treatments with blockers of GHS-R1a, type 1 transient receptor potential canonical (TRPC1) channels and KCNQ channels. These results suggest that both activating TRPC1 channels and inhibiting KCNQ channels occur downstream the ghrelin-GHS-R1a signaling pathway probably in somatodendritic sites of MLIs. Thus, ghrelin can control PC firing directly and indirectly via its modulation of GABAergic transmission, thereby impacting activity in cerebellar circuitry.

Figure 1

Effects of ghrelin on spontaneous firing of cerebellar PCs. (A) Representative traces of spontaneous firing recorded by the cell-attached mode from PCs before treatment (control), and in the presence of ghrelin (0.1 μM). (B) left: Time course of the firing rate of PCs with perfusion of ghrelin (0.1 μM: gray circles, 0.3 μM: black circles). The firing rate of PCs was expressed as a percentage of the baseline, which was determined for 4 min before application of ghrelin. Each point represents the mean values obtained from 13 cells for 0.1 μM and 12 cells for 0.3 μM. right: The magnitude of ghrelin-induced firing facilitation of PCs at 0.1 (n = 13, gray circles) and 0.3 μM (n = 12, black circles). (C) The peptide increased the firing rate of PCs significantly (left) without changing the coefficient of variation (CV) (middle) or CV2 (right) of the inter-spike interval (n = 13). (D) Effects of blockers for synaptic transmission on the ghrelin-induced firing facilitation of PCs. Time course of the firing rate of PCs in the presence of synaptic blockers (SBs), NBQX (5 μM) and PTX (100 μM). The firing rate of PCs was expressed as a percentage of the baseline, which was determined for 4 min before application of ghrelin. Each point represents the mean values obtained from 9 cells. (E) Ghrelin strongly increased the firing rate of PCs without changing the CV (middle) or CV2 (right) of the inter-spike interval (n = 9). (F) The magnitude of ghrelin-induced firing facilitation of PCs in the presence of blockers for synaptic transmission (SBs, n = 9) is larger than that in the absence of the blockers (Ctrl, n = 13).

Figure 2

Effects of ghrelin on stimulation-evoked PF-EPSCs and IPSCs in PCs. (A) Representative averaged traces of four consecutive PF-EPSCs recorded from PCs held at –60 mV before treatment (control), and in the presence of ghrelin (0.1 μM). (B) Time course of the peak amplitude of the first PF-EPSCs recorded from PCs. The first PF-EPSC amplitude was expressed as a percentage of the baseline, which was determined for 4 min before application of ghrelin. Each point represents the mean values obtained from 10 cells. (C, D) The peptide affected neither the amplitude (C) nor the paired-pulse ratio (PPR) (D) of PF-EPSCs (n = 10). (E) Representative averaged traces of four consecutive eIPSCs recorded from PCs held at 10 mV before treatment (control), and in the presence of ghrelin (0.1 μM). (F) Time course of the peak amplitude of the first eIPSCs recorded from PCs. The first eIPSC amplitude was expressed as a percentage of the baseline, which was determined for 4 min before application of ghrelin. Each point represents the mean values obtained from 7 cells. (G) (H) The peptide affected neither the amplitude (G) nor the paired-pulse ratio (PPR) (H) of eIPSCs (n = 7).

Figure 3

Effects of ghrelin on spontaneous IPSCs in PCs. (A) Representative traces of sIPSCs recorded from PCs held at − 35 mV before treatment, and in the presence of ghrelin (0.3 μM). Gray dotted lines indicate the basal current level before application of the peptide. The lower traces represent sIPSCs displayed on an expanded time scale before and during the application of ghrelin. (B) Ghrelin caused an outward shift of baseline currents (n = 8). (C) Time course of the frequency of sIPSCs in PCs. The sIPSC frequency was expressed as a percentage of the baseline, which was determined for 4 min before application of ghrelin. Each point represents the mean values obtained from 8 cells. (D, E) The peptide increased both the frequency (D) and the amplitude (E) of sIPSCs significantly (n = 8).

Figure 4

Effects of ghrelin on miniature IPSCs in PCs. (A) Representative traces of mIPSCs recorded from a PC before treatment (control) and during the application of ghrelin (0.3 μM). (B) Cumulative probability fractions of the mIPSC amplitude were obtained from the same cell as in (A). Amplitude distribution was not affected by ghrelin (P = 0.220 by Kolmogorov–Smirnov test). (C) The time course of the mIPSC frequency in PCs. The mIPSC frequency was expressed as a percentage of the baseline, which was determined for 4 min before application of ghrelin. Each point represents the mean values obtained from seven cells. (D–G) Ghrelin had no effect on the frequency (D), amplitude (E), rise time (F) or decay time (G) of mIPSCs in PCs (n = 7).

Figure 5

Effects of ghrelin on spontaneous firing of MLIs. (A) Representative traces of spontaneous action potentials of MLIs recorded by a cell-attached recording before treatment (control) and during the application of ghrelin (0.3 μM). (B) left: Time course of the firing rate of MLIs with perfusion of ghrelin (gray circles: 0.1 μM, black circles: 0.3 μM). The firing rate was expressed as a percentage of the baseline, which was determined for 4 min before application of ghrelin. Each point represents the mean values obtained from 12 cells for 0.1 μM and 8 cells for 0.3 μM. right: The magnitude of ghrelin-induced firing facilitation of MLIs at 0.1 (n = 12, gray circles) and 0.3 μM (n = 8, black circles). (C) Ghrelin increased the firing rate of MLIs (left) without changing the CV (middle) or CV2 (right) of the inter-spike interval (n = 8). (D) Time course of the firing rate of MLIs. The GHS-R1a antagonist JMV3002 (1 μM) was applied during the periods indicated by a horizontal gray bar. The firing rate of MLIs was expressed as a percentage of the baseline, which was determined for 4 min before application of the antagonist. Each point represents the mean values obtained from 7 cells. (E, F) The GHS-R1a antagonist did not affect the baseline firing rate (E), while it inhibited the ghrelin-induced firing facilitation of MLIs (F) (n = 7).

Figure 6

Effects of inhibitors for channels possibly involved in ghrelin signaling downstream on ghrelin-induced firing facilitation of MLIs. (A) Time course of the firing rate of MLIs in the presence of blockers for synaptic transmission (SBs), NBQX (5 μM), APV (15 μM), and PTX (100 μM). The firing rate of MLIs was expressed as a percentage of the baseline, which was determined for 4 min before application of ghrelin. Each point represents the mean values obtained from 13 cells. (B) Ghrelin strongly increased the firing rate of MLIs (left) changing significantly both the CV (middle) and CV2 (right) of the inter-spike interval (n = 13). (C) Effects of a non-selective TRPC channel inhibitor 2-APB (30 μM) on the ghrelin-induced firing facilitation of MLIs. Time course of the firing rate of MLIs. Each point represents the mean values obtained from 9 cells. (D) Effects of a blocker for KCNQ channels XE991 (10 μM) on the ghrelin-induced firing facilitation of MLIs. Time course of the firing rate of MLIs. Each point represents the mean values obtained from 7 cells. (E) The magnitudes of ghrelin-induced firing facilitation of MLIs in the presence of 2-APB and XE991 were lower than that in the absence of blocker.