The endocannabinoid system controls key epileptogenic circuits in the hippocampus

Abstract

Balanced control of neuronal activity is central in maintaining function and viability of neuronal circuits. The endocannabinoid system tightly controls neuronal excitability. Here, we show that endocannabinoids directly target hippocampal glutamatergic neurons to provide protection against acute epileptiform seizures in mice. Functional CB1 cannabinoid receptors are present on glutamatergic terminals of the hippocampal formation, colocalizing with vesicular glutamate transporter 1 (VGluT1). Conditional deletion of the CB1 gene either in cortical glutamatergic neurons or in forebrain GABAergic neurons, as well as virally induced deletion of the CB1 gene in the hippocampus, demonstrate that the presence of CB1 receptors in glutamatergic hippocampal neurons is both necessary and sufficient to provide substantial endogenous protection against kainic acid (KA)-induced seizures. The direct endocannabinoid-mediated control of hippocampal glutamatergic neurotransmission may constitute a promising therapeutic target for the treatment of disorders associated with excessive excitatory neuronal activity.

Figure 1

Specific Deletion of CB1 in Different Neuronal Subpopulations in Conditional CB1 Mutant Mice as Revealed by In Situ Hybridization with a Specific Riboprobe for CB1 mRNA Micrographs showing CB1 mRNA expression in the brain of CB1^f/f (wt, [A]–[D]), CaMK-CB1^−/− (E–H), Glu-CB1^−/− (I–L), and GABA-CB1^−/− mice (M–P). Sections were stained throughout the brain and micrographs were taken at the level of caudate putamen (A, E, I, and M), dorsal hippocampus (B, F, J, and N) and cerebellum (C, G, K, and O). (D, H, L, and P) Detailed enlargements of the dorsal hippocampus of (B), (F), (J), and (N), respectively. In all three mutant lines, deletion of CB1 is mainly restricted to the forebrain, with little or no alterations in the hindbrain (C, G, K, and O). In CaMK-CB1^−/− mice, CB1 mRNA is absent from all principal neurons and expressed only in GABAergic interneurons (intense scattered dots; E, F, and H). In Glu-CB1^−/− mice, CB1 mRNA is absent in the majority of cortical glutamatergic neurons (I, J, and L). In GABA-CB1^−/− mice, CB1 mRNA is absent in all GABAergic neurons and it is expressed only in non-GABAergic cells (low uniform gray staining; M, N, and P).

Figure 2

Deletion of CB1 in Glutamatergic Cortical Neurons, but Not in GABAergic Neurons, Increases Susceptibility to KA-Induced Seizures (A–C) Micrographs showing double in situ hybridization of CB1 mRNA (red staining) together with GAD65 mRNA (silver grains) in the CA3 region of CB1^f/f (A), Glu-CB1^−/− (B), and GABA-CB1^−/− mice (C). In (A), CB1 is present in both GABAergic interneurons and pyramidal neurons. In (B), CB1 is present only in GABAergic interneurons. In (C), CB1 is present only in pyramidal neurons. Blue staining, toluidine blue nuclear counter-staining. Filled arrows, GABAergic interneurons (GAD65-positive) expressing CB1 mRNA. Open arrows, GABAergic interneurons lacking CB1 expression. Bar, 25 μm. For similar analysis of CB1 expression in CaMK-CB1^−/− mice, see Figure 2 of Marsicano et al. (2003). (D–F) Seizure scoring (30 mg/kg KA) of conditional CB1 mutant mice (filled symbols), as compared to control CB1^f/f littermates (open symbols). Seizures are worsened in CaMK-CB1^−/− (D) and in Glu-CB1^−/− (E), whereas they do not differ from wild-type controls in GABA-CB1^−/− mice (F). In brackets, number of animals for each experimental group. Data are presented as mean ± SEM. **p < 0.005; ***p < 0.001 (by two-way ANOVA, factor genotype).

Figure 3

Functions of CB1 Receptors in the Control of GABAergic Transmission in Conditional CB1 Mutant Mice and in the Development of KA-Induced Seizures (A) Depolarization-induced suppression of inhibitory currents (DSI) in CA1 hippocampal pyramidal neurons of CaMK-CB1^−/− mice (filled circles, n = 11 cells from six mice) and of respective CB1^f/f littermate controls (open circles, n = 13 cells from six mice). The 5 s depolarization step is indicated (see Experimental Procedures). Data are mean ± SEM. (B) Summary of DSI data obtained by calculating the mean of the five eIPSCs before and after the depolarization step. Data are mean ± SEM. Top, representative traces before and after depolarization step for each genotype, respectively. Data are mean ± SEM. ***p < 0.001 versus respective baseline before depolarization (100%, dotted line). (C) DSI in CA1 hippocampal pyramidal neurons of GABA-CB1^−/− (filled triangles, n = 13 cells from six mice) and of respective CB1^f/f littermate controls (open triangles, n = 10 cells from six mice). Data are mean ± SEM. The 5 s depolarization step is indicated. (D) Summary of DSI data obtained by calculating the mean of the five eIPSCs before and after the depolarization step. Top, representative traces before and after depolarization step for each genotype, respectively. Data are mean ± SEM. ***p < 0.001 versus respective baseline before depolarization (100%, dotted line). (E) Effects of vehicle (open symbols) and of the allosteric modulator of GABA-A receptor diazepam (6 mg/kg, filled symbols) on KA-induced seizures in CB1^+/+ (triangles) and on CB1^−/− littermates (squares). Data are mean ± SEM. **p < 0.01 as compared to respective vehicle-treated groups (by two-way ANOVA, factor treatment).

Figure 4

CB1 Protein Is Present in Glutamatergic Hippocampal Neurons Micrographs showing the immunohistochemical analysis of CB1 expression in wild-type (A–C), CaMK-CB1^−/− (D–F), GABA-CB1^−/− (G–I) and complete CB1 knock-out mice (J–L). (B, E, H, and K) Higher magnification micrographs of the areas enclosed in the square in (A), (D), (G), and (J), respectively. (C, F, I, and L) Detail of the CA3 hippocampal region. GC, granule cell layer of dentate gyrus; Hil, hilar region of dentate gyrus; LMol, stratum lacunosum-molecularis; Mol, stratum molecularis; Or, stratum oriens; Pyr, CA1/CA3 pyramidal cell layer of hippocampus; Rad, stratum radiatum. Asterisks indicate the inner third of the molecular layer. Bar, 100 μm (A, C, D, F, G, I, J, and L); 25 μm (B, E, H, and K).

Figure 5

CB1 mRNA Is Expressed in Glutamatergic Cortical Neurons and in Glutamatergic Mossy Cells of the Dentate Gyrus, as Revealed by Double In Situ Hybridization Experiments in Wild-Type C57BL/6N Mice and Conditional Mutants (A–D) Micrographs showing coexpression of CB1 mRNA (red staining) with VGluT1 mRNA (silver grains) in the CA3 region of hippocampus (A), basolateral amygdala (B), entorhinal cortex (C), and hilus of dentate gyrus (D) of wild-type C57BL/6N adult mice. Filled arrows, GABAergic interneurons, expressing CB1, but not VGluT1. Open arrows, glutamatergic neurons, coexpressing CB1 and VGluT1. (E–H) Micrographs showing coexpression of CB1 mRNA (red staining) with GAD65 mRNA (silver grains), in the hilus of dentate gyrus in CB1^f/f (E), GABA-CB1^−/− (F), CaMK-CB1^−/− (G), and Glu-CB1^−/− mice (H). Filled arrowheads, GABAergic interneurons, coexpressing CB1 and GAD65 mRNAs. Open arrowheads, presumable glutamatergic neurons, not coexpressing CB1 and GAD65 mRNAs. Note that high CB1-expressing cells do not coexpress VGluT1 mRNA (A–D), but do coexpress GAD65 mRNA (E–H), showing that they are GABAergic interneurons. Conversely, low CB1-expressing cells coexpress VGluT1 mRNA (A–D), but do not contain GAD65 mRNA (E–H), showing that they are glutamatergic neurons. In particular, glutamatergic hilar mossy cells contain CB1 mRNA in wild-type (D), CB1^f/f mice (E), and GABA-CB1^−/− (F), but not in CaMK-CB1^−/− (G) and Glu-CB1^−/− mice (H). Blue staining, toluidine blue nuclear counterstaining. GC, granule cells of dentate gyrus. Pyr, CA3 pyramidal neurons. Bar, 25 μm.

Figure 6

Glutamatergic Terminals of Mossy Cells Contain Functional CB1 Receptors (A) Low-magnification micrograph showing expression of VGluT1 protein in the hippocampal formation. (B–D) High-magnification confocal micrographs showing detailed expression of CB1 and VGluT1 protein in the inner third of the molecular layer (approximately corresponding to the dotted area in [A]). Expression of VGluT1 (green, [B]), CB1 (red, [C]), and merged image (D) are shown. Yellow areas indicate colocalization of the two proteins. (E–G) Activation of CB1 receptors reduces glutamatergic inputs onto dentate gyrus granule cells. (E) Time-course analysis of the effects of 5 μM of the CB1 agonist WIN55,212-2 on eEPSCs in wild-type (CB1^f/f, open circles, n = 6 slices from three mice) and CaMK-CB1^−/− littermates (filled circles, n = 6 cells from three mice). (F) Summary of eEPSCs data obtained by calculating the average of the measurements obtained during the last 5 min of experiment (35–40 min after drug application). (G) Representative electrophysiological traces before (baseline) and 40 min after application of WIN 55,212-2 (WIN). Bars, 75 μm (A), 3.5 μm (B–D); **p < 0.01 as compared to baseline (average of the measurements during the last 5 min before drug application, dotted line).

Figure 7

Virally Induced Deletion of CB1 Gene in the Hippocampal Formation Increases the Sensitivity to KA-Induced Seizures (A and B) Dark-field micrographs showing CB1 mRNA expression in CB1^f/f mice 9 weeks after intrahilar injections of AAV-GFP (A) and of AAV-Cre (B), showing the Cre-mediated deletion of CB1 in the hilus of the dentate gyrus and in part of the CA1 and CA3 hippocampal regions. Cx, neocortex; Lh, lateral habenula; bar, 500 μm. (C) Schematic diagrams showing the approximate extension of Cre-mediated recombination of CB1 (gray shading) in AAV-Cre-injected CB1^f/f mice (see Experimental Procedures). Numbers, distance from bregma (Paxinos and Franklin, 2001). (D) Seizure scoring (30 mg/kg of KA) of AAV-GFP-injected (open circles) and AAV-Cre-injected CB1^f/f mice (filled circles). In brackets, number of mice in each experimental group. Data are presented as mean ± SEM. *p < 0.05, by two-way ANOVA.