Distinct mechanisms of CB1 and GABAB receptor presynaptic modulation of striatal indirect pathway projections to mouse globus pallidus

Abstract

Presynaptic modulation is a fundamental process regulating synaptic transmission. Striatal indirect pathway projections originate from A2A-expressing spiny projection neurons (iSPNs), targeting the globus pallidus external segment (GPe) and control the firing of the tonically active GPe neurons via GABA release. It is unclear if and how the presynaptic G-protein-coupled receptors (GPCRs), GABA_B and CB1 receptors modulate iSPN-GPe projections. Here we used an optogenetic platform to study presynaptic Ca²⁺ and GABAergic transmission at iSPN projections, using a genetic strategy to express the calcium sensor GCaMP6f or the excitatory channelrhodopsin (hChR2) on iSPNs. We found that P/Q-type calcium channels are the primary voltage-gated Ca²⁺ channel (VGCC) subtype controlling presynaptic calcium and GABA release at iSPN-GPe projections. N-type and L-type VGCCs also contribute to GABA release at iSPN-GPe synapses. GABA_B receptor activation resulted in a reversible inhibition of presynaptic Ca²⁺ transients (PreCaTs) and an inhibition of GABAergic transmission at iSPN-GPe synapses. CB1 receptor activation did not inhibit PreCaTs but inhibited GABAergic transmission at iSPN-GPe projections. CB1 effects on GABAergic transmission persisted in experiments where Na_V and K_V 1 were blocked, indicating a VGCC- and K_V 1-independent presynaptic mechanism of action of CB1 receptors. Taken together, presynaptic modulation of iSPN-GPe projections by CB1 and GABA_B receptors is mediated by distinct mechanisms. KEY POINTS: P/Q-type are the predominant voltage-gated Ca²⁺ channels controlling presynaptic Ca²⁺ and GABA release on the striatal indirect pathway projections. GABA_B receptors modulate iSPN-GPe projections via a VGCC-dependent mechanism. CB1 receptors modulate iSPN-GPe projections via a VGCC-independent mechanism.

Keywords: basal ganglia; calcium imaging; neurotransmitter release; optogenetics; presynaptic.

Figure 1. An optogenetic strategy to study GABAergic transmission and PreCaTs at iSPN‐GPe projections

A, representative micrographs of EYFP expression in the A2aCre‐AiCOP4 line. From left to right: coronal slices containing the dorsal striatum, GPe and SNr. Dense EYFP fluorescence was observed in the striatum and GPe, but not in the SNr. B, representative confocal micrographs where immunohistochemical staining was performed in A2aCre‐GCaMP6f mice using anti‐GFP (green) and anti‐NeuN (red). From left to right: 20× images of dorsal striatum (scale bar: 100 µm); 63× images of NeuN, GFP and the two channels merged in the dorsal striatum (scale bar: 25 µm); 20× image of GPe (scale bar: 100 µm). C, diagram of optogenetic and whole‐cell patch clamp experiments. D, summary bar graph and example traces of oIPSC inhibition by gabazine. Gabazine fully inhibited oIPSCs recorded at iSPN‐GPe synapses indicating their GABA_A dependency (n = 6 cells from n = 3 mice). E, summary bar graph and example traces of whole‐cell patch clamp experiments where (1) baseline, (2) TTX effect and (3) TTX + 4‐AP effects were compared. TTX fully inhibited oIPSCs, which were restored above baseline levels by TTX + 4‐AP application (n = 5 cells from n = 4 mice). F, onset latency of oIPSCs recorded using high‐ or low‐chloride intracellular solutions; cells from panels D and E. G, schematic of slice photometry experiments. H, input/output (I/O) curve experiment. The amplitude of the PreCaTs scaled with stimulation intensity indicating activity dependence (n = 5 slices, n = 4 mice). I, time course of Ca²⁺‐free aCSF application effects on PreCaTs. J, bar graph summary of effects: baseline, last 2 min of Ca²⁺‐free aCSF, last 2 min of washout (n = 5 slices from n = 4 mice). ^* P < 0.05; ^** P < 0.01;    ^****/#### P < 0.0001. [Colour figure can be viewed at wileyonlinelibrary.com]

Figure 2. P/Q‐type are the predominant VGCCs controlling presynaptic Ca²⁺ and GABAergic transmission at the indirect pathway projections to the GPe

A, time course of VGCC blocker application effects on PreCaTs and raw photometry traces. B, summary of drug effects: baseline, last 2 min of drug, last 2 min of washout. Blockade of P/Q‐type VGGCs (agatoxin, n = 6 slices from n = 6 mice) significantly decreased PreCaTs from baseline, whereas no significant effect on PreCaTs was observed by blocking N‐type VGCCs (conotoxin, n = 8 slices from n = 8 mice) or L‐type VGCCs (nifedipine, n = 5 slices from n = 5 mice). C, time course of VGCC blocker application effects on oIPSCs and raw electrophysiological traces. D, comparison of drug effects: baseline, last 2 min of drug, last 2 min of washout. Blockade of P/Q‐type VGGCs (agatoxin, n = 5 cells from n = 4 mice) significantly decreased PreCaTs from baseline, and smaller but significant effects were induced by blocking N‐type VGCCs (conotoxin, n = 5 cells from n = 4 mice) or L‐type VGCCs (nifedipine, n = 5 cells from n = 4 mice). ^*/# P < 0.05; ^**/## P < 0.01; ^****/#### P < 0.0001. [Colour figure can be viewed at wileyonlinelibrary.com]

Figure 3. VGCC‐dependent modulation of iSPN‐GPe projections by GABA_B receptors

A, time course of Baclofen effects on PreCaTs and raw photometry traces. B, summary of drug effects: baseline, last 2 min of drug, last 2 min of washout. GABA_B activation (baclofen, n = 7 slices from n = 5 mice) significantly decreased PreCaTs amplitude. C, time course experiment of baclofen followed by CBP 55840 effects on PreCaTs. D, baclofen significantly decreased PreCaTs amplitude and CGP55840 application reversed PreCaTs to baseline levels (n = 5 slices from n = 2 mice). E, time course of baclofen effects on oIPSCs at iSPN‐GPe synapses and raw electrophysiological traces. F, baclofen significantly reduced oIPSCs at iSPN‐GPe terminals (n = 7 cells from n = 5 mice). G, effects of baclofen application on sIPSCs. Example traces and summary bar graphs. Baclofen application significantly decreased the frequency and amplitude of sIPSCs recorded in the GPe, indicating a presynaptic site of action (n = 12 cells from n = 10 mice). ^* P < 0.05; ^** P < 0.01; ^***/### P < 0.001; ^####/*** P < 0.0001. [Colour figure can be viewed at wileyonlinelibrary.com]

Figure 4. VGCC‐ and KV1‐independent mechanisms of presynaptic modulation of iSPN‐GPe projections by CB1 receptors

A, time course of WIN effects on PreCaTs and raw photometry traces. B, summary of drug effects: baseline, last 2 min of drug, last 2 min of washout. CB1 receptor activation (WIN 55212‐2, n = 5 slices from n = 4 mice) did not change PreCaT amplitude. C, time course of WIN effects on oIPSCs at iSPN‐GPe synapses and raw electrophysiological traces. D, WIN 55212‐2 application significantly decreased oIPSC amplitude (n = 5 cells from n = 5 mice). E, effects of WIN 55212‐2 application on mIPSCs. Example traces and summary bar graphs. WIN 55212‐2 application significantly decreased the frequency and amplitude of mIPSCs (n = 7 cells from n = 4 mice). F, time course of WIN 55212‐2 effects on oIPSCs recorded in the presence of TTX + 4AP. G, WIN 55212‐2 significantly reduced oIPSCs at iSPN‐GPe terminals (n = 4 cells from n = 3 mice). H and I, no effect of AM‐251 application on oIPSCs (n = 5 cells from n = 3 mice). J, no effect of AM‐251 application on sIPSC frequency or amplitude (n = 5 cells from n = 3 mice). K, Cd²⁺ decreased the amplitude but not frequency of mIPSCs (n = 7 cells from n = 3 mice). L, baclofen effects were preserved in slices pre‐incubated with AM‐251 (n = 4 cells from n = 2 mice). ^* P < 0.05; ^** P < 0.01. [Colour figure can be viewed at wileyonlinelibrary.com]