Université Bordeaux Segalen
Année 2012 Thèse n° 1928
THÈSE
pour le
DOCTORAT DE L’UNIVERSITÉ BORDEAUX 2 Mention : Sciences, technologie, santé Option : Neurosciences
Présentée et soutenue publiquement le 26 juin 2012, par Mathilde Metna-Laurent Née le 19 mars 1984 à Creil, France
SPECIFICITE DU TYPE CELLULAIRE DANS LE CONTROLE
DES FONCTIONS DE MEMOIRE PAR LES RECEPTEURS
CANNABINOIDES CB 1
Membres du Jury
Dr Christian Giaume Paris, France ...... Président, Rapporteur Pr Manuel Guzmán Madrid, Espagne ...... Rapporteur Dr Aline Desmedt Bordeaux, France ...... Examinatrice Dr Pier Vincenzo Piazza Bordeaux, France...... Examinateur Dr David Robbe Barcelone, Espagne ...... Examinateur Dr Stéphane Oliet Bordeaux, France ...... Examinateur invité Dr Giovanni Marsicano Bordeaux, France ...... Directeur de thèse
University Bordeaux Segalen
2012 Thesis n° 1928
THESIS
For the
DOCTORATE OF BORDEAUX 2 UNIVERSITY Speciality : Sciences, technologie, santé Option : Neurosciences
Presented and defended publicly on June 26 th 2012, by Mathilde Metna-Laurent Born on Marsh 19 th 1984 at Creil, France
CELL TYPE -SPECIFIC CONTROL OF MEMORY FUNCTIONS BY
CB 1 CANNABINOID RECEPTORS
Members of the Jury
Dr Christian Giaume Paris, France ...... Chairman, Reviewer Pr Manuel Guzmán Madrid, Espagne ...... Reviewer Dr Aline Desmedt Bordeaux, France ...... Examiner Dr Pier Vincenzo Piazza Bordeaux, France ...... Examiner Dr David Robbe Barcelone, Espagne ...... Examiner Dr Stéphane Oliet Bordeaux, France ...... Invited examiner Dr Giovanni Marsicano Bordeaux, France ...... Thesis supervisor
REMERCIEMENTS
Tout d’abord merci à Aline Desmedt, Christan Giaume, Manuel Guzmán, Stéphane Oliet et David Robbe pour avoir accepté d’accorder votre temps et votre expertise à l’évaluation de ce travail. Merci à Pier Vincenzo Piazza d’avoir accepté de participer à cette commission, et, en tant que directeur du NeuroCentre Magendie, merci de m’avoir permis de travailler dans des conditions matérielles exceptionnelles.
Cette thèse est le fruit d’un travail collectif dont Giovanni Marsicano est le superviseur. Gio, merci de m’avoir accueilli dans ton équipe et de m’avoir fait confiance pour la réalisation de ce projet. Je mesure l’immense chance que j’ai eu d’être formée par un chercheur dont l’excellence scientifique n’a d’égale que les qualités humaines.
Je tiens à remercier tout particulièrement les docteurs Pauline Lafenêtre, Joana Lourenço et Francis Chaouloff. Votre soutien permanent durant ces années a conditionné le prologue ainsi que le bon déroulement des peripéties de cette aventure. Merci à toute l’équipe Marsicano, qui malgré la croissance fulgurante de son effectif, a toujours gardé sa cohésion et sa bonne humeur. Bien que les listings soient un peu ennuyeux, « un Marsicano » se laisse difficilement citer sans son voisin. Ainsi, merci à Dany, Peggy, Astrid, Sarah, Barbara, Elodie, Tifany, Isabel, Bo, Aya, Theresa, Ilaria, Luigi, Edgar, Little Giovanni, Michelangelo, Etienne, Federico, Arnau… Je n’oublie pas Pierrick, Martina et Laurie: merci d’avoir partagé mon travail pour un temps, je vous souhaite le meilleur!
Mes remerciements s’adressent également à Xia Zhang, ainsi qu’à tous nos collaborateurs cannadiens, de Chine, d’Espagne, de France et de Navarre, avec qui j’ai participé à une expérience collaborative hors du commun, et qui a éveillé ma curiosité pour ces cellules outsiders…
Merci à Carsten Wotjak de m’avoir accueillie dans son équipe durant ces semaines de training à la microdialyse et au breakfast bavarois. Merci pour ton soutien et pour avoir apporté, à de nombreuses reprises, ton regard d’expert sur mes résultats. Merci également à Elmira, Linda, Vincenzo et toute l’équipe du « Max Planck » pour leur accueil chaleureux.
Un clin d’œil à mes neurocopines, Amélie, Aurore et Marion. J’espère que l’on aura toujours l’occasion de partager ces bons moments de science de comptoir…
Merci enfin à toute ma famille et mes amis, les Aquitains, les Picards, les Champenois, les Limousins, la Languedocienne, les Franciliens… Un immense merci à ma sœur et à ma maman pour leur soutien et leur amour inconditionnel. Merci à toi pour qui l’anonymat est un droit non négociable : ce travail est en partie anonyme.
LIST OF PUBLICATIONS
• Articles published or in process of publication in peer reviewed scientific journals:
Boucher AA, Vivier L, Metna-Laurent M , Brayda-Bruno L, Mons N, Arnold JC, Micheau J (2009) Chronic treatment with Delta(9)-tetrahydrocannabinol impairs spatial memory and reduces zif268 expression in the mouse forebrain . Behavioral Pharmacology 20(1):45-55.
Dubreucq S, Kambire S, Conforzi M, Metna-Laurent M , Cannich A, Soria-Gomez E, Richard E; Marsicano G, Chaouloff F (2012). CB1 receptors located on sim1-expressing neurons control emotional behaviors . Neuroscience 204:230-244.
Bénard G 1, Massa F 1, Puente N, Lourenço J, Bellocchio L, Soria-Gómez E, Matias I, Delamarre A, Metna-Laurent M , Cannich A, Hebert-Chatelain E, Mulle C, Ortega-Gutiérrez S, Martín-Fontecha M, Klugmann M, Guggenhuber S, Lutz B, Gertsch J, Chaouloff F, López-Rodríguez ML, Grandes P, Rossignol R, Marsicano G (2012) Mitochondrial CB(1) receptors regulate neuronal energy metabolism . Nature Neuroscience 15(4):558-564.
Han J 1, Kesner P 1, Metna-Laurent M 1, Duan T, Xu L, Georges F, Koehl M, Abrous DN, Mendizabal-Zubiaga J, Grandes P, Liu Q, Bai G, Wang W, Xiong L, Ren W, Marsicano G 2, Zhang X2 (2012) Acute Cannabinoids Impair Working Memory through Astroglial CB1 Receptor Modulation of Hippocampal LTD . Cell 148, 1039-1050.
Metna-Laurent M 1, Soria-Gόmez E 1, Verrier D, Conforzi M, Jégo P, Lafenêtre P, Marsicano G (2012) Bimodal Control of Fear-Coping Strategies by CB1 Cannabinoid Receptors . The Journal of Neuroscience, in press .
• Oral communications
Metna-Laurent M , Jégo P, Conforzi M, Lafenêtre P, Marsicano G The endocannabinoid system bidirectionally determines fear-coping strategies. Meeting of the Université franco- allemande, Deutsch-Französische Hochschule, Munich, Germany, 2011 July 13 th -14 th .
Metna-Laurent M , Han J, Kesner P, Duan T, Xu L, Georges F, Koehl M, Abrous DN, Mendizabal-Zubiaga J, Grandes P, Liu Q, Bai G, Wang W, Xiong L, Ren W, Marsicano G, Zhang X. Astroglial CB1 receptors and memory . 2 nd Symposium of the NeuroCentre Magendie, Bordeaux, France, 2011 December 15 th -16 th ; International Astrocytes Summer school, Bertinoro, Italy, 2012 Marsh 25 th -31 th .
1 Share first authorship
2 Share last authorship
• Posters
Metna-Laurent M , Lafenêtre P, Marsicano G Role of the endocannabinoid system in the associative components of extinction. Annual meeting of the Network of European Neuroscience Institutes. Fodelebeach, Greece, 2009 May 20 th -21 st ; 9th meeting of the French Neurosciences Society. Bordeaux, France, 2009 May 26 th -29 th .
Metna-Laurent M , Lafenêtre P, Marsicano G Role of the endocannabinoid system in fear- induced avoidance behaviors. Gordon Research Conference on Amygdala in Health and Disease. Colby College, Waterville, ME, USA, 2009 July 12 th -17 th .
Metna-Laurent M , Lafenêtre P, Jégo P, Marsicano G Bimodal control of fear-coping strategies by the endocannabinoid system. 40 th Annual meeting of the Society for Neuroscience. San Diego, CA, USA, 2010 November 13 th -17 th .
Metna-Laurent M , Lafenêtre P, Jégo P, Conforzi M, Marsicano G The endocannabinoid system bidirectionally determines fear-coping strategies. Gordon Research Conference on Cannabinoid functions in the CNS. Les Diablerets, Switzerland, 2011, May 22 th -27 th .
Conforzi M, Metna-Laurent M , Wiesner T, Bosier B, Massa F, Marsicano G Implication of astrocytic CB 1 receptors in fear memory. Gordon Research Conference on Cannabinoid functions in the CNS. Les Diablerets, Switzerland, 2011, May 22 th -27 th .
Metna-Laurent M , Soria-Gomez E, Verrier D, Jégo P, Conforzi M, Lafenêtre P, Marsicano G nd Bimodal control of fear-coping strategies by CB 1 cannabinoid receptors. 2 symposium of the NeuroCentre Magendie, Bordeaux, France, 2011 December 15 th -16 th .
• Scientific article for the general public
Metna-Laurent M, Marsicano G (2012) Cannabis et mémoire de travail . La Recherche, in press.
Address of the laboratory: NeuroCentre Magendie INSERM U862 group « Endocannabinoid and NeuroAdaptation » 146, rue Léo Saignat 33077 Bordeaux Cedex
RESUME
Le système endocannabinoïde est un important modulateur des fonctions physiologiques. Dans le cerveau, son contrôle s’exerce essentiellement par les récepteurs cannabinoïdes de type 1 (CB 1).
Les récepteurs CB 1 sont abondamment exprimés sur les neurones excitateurs glutamatergiques et les neurones inhibiteurs GABAergiques et leur stimulation inhibe la libération du glutamate et du GABA.
Récemment, l’activité des récepteurs CB1 sur les astrocytes a été proposée comme facilitant la transmission excitatrice. Par ce contrôle général de la neurotransmission, l’activité des récepteurs CB 1 induit différents phénomènes de plasticité synaptique associés aux processus de mémoire. Les récepteurs CB 1 jouent un rôle complexe dans les fonctions de mémoire. En particulier, la stimulation exogène des récepteurs CB 1 perturbe la mémoire de travail. D’autre part, la signalisation endogène des récepteurs CB 1 est nécessaire à l’adaptation des réponses de peur apprises. Cependant, les mécanismes par lesquels les récepteurs CB 1 régulent ces processus de mémoire n’ont été que peu analysés. L’objectif de ce travail fut de caractériser les mécanismes cellulaires par lesquels les récepteurs CB 1 contrôlent la mémoire de travail et les réponses de peur apprises. Nous avons utilisé les modèles de mutation constitutive et conditionnelle des récepteurs CB 1 chez la souris afin d’analyser les conséquences de la délétion de ces récepteurs sur des types cellulaires particuliers.
Dans une première étude, nous avons montré que les cannabinoïdes exogènes tels que le ∆9- tetrahydrocannabinol (THC, principal composé psychoactif du cannabis) induisent des déficits de mémoire de travail spatiale par la stimulation des récepteurs CB 1 exprimés sur les astrocytes. Les cannabinoïdes induisent une forme de dépression à long-terme dans l’hippocampe dont plusieurs mécanismes cellulaires sont similaires à ceux supportant les déficits de mémoire mis en évidence par l’analyse comportementale. Ces résultats suggèrent que les cannabinoïdes altèrent la mémoire de travail spatiale par une modification de la plasticité synaptique de l’hippocampe induite par la stimulation des récepteurs CB 1 astrogliaux.
Dans une seconde étude, nous avons mis en évidence que les récepteurs CB 1 localisés sur les neurones GABAergiques et glutamatergiques exercent un contrôle opposé sur le type de réponse
élicité par un stimulus conditionné aversif. La ré-expression sélective des récepteurs CB 1 dans l’amygdale des souris mutantes constitutives a permis de préciser l’implication de cette structure dans la régulation des réponses de peur conditionnées par les récepteurs CB 1.
L’ensemble de ces travaux indiquent que le système endocannabinoïde contrôle les fonctions de mémoire par une régulation de l’activité de cellules spécifiques dans le cerveau. L’implication des astrocytes dans les effets des cannabinoïdes sur la mémoire souligne l’importance de ces cellules dans les processus cognitifs et suggère que les récepteurs CB 1 astrogliaux jouent un rôle dans d’autres fonctions cérébrales. Nos résultats révèlent également l’importance de l’évaluation de différents comportements dans le cadre des modèles expérimentaux d’adaptation à la peur.
Mots clés : récepteurs CB 1, astrocytes, mémoire de travail, peur conditionnée, évitement, souris mutantes
ABSTRACT
The endocannabinoid system is an important regulator of physiological functions. In the brain, this control is mainly exerted through the type-1-cannabinoid (CB 1) receptors. CB 1 receptors are abundant at excitatory glutamatergic and inhibitory GABAergic neuron terminals where their stimulation inhibits neurotransmitter release. The activity of CB 1 receptors on astrocytes has been recently proposed as facilitating excitatory transmission. Through this general control on brain neurotransmission, CB 1 receptors mediate distinct forms of synaptic plasticity that are associated with memory processing. Indeed, CB 1 receptors control memory functions. In particular, the exogenous stimulation of CB 1 receptors impairs working memory. Moreover, the endogenous CB 1 receptor signalling ensures the adaptation of learned fear responses. However, the brain mechanisms of this
CB 1-mediated control of memory functions are poorly characterized. The goals of this research work were to dissect the cellular mechanisms by which CB1 receptors control both working memory and learned fear responses. We used constitutive and conditional mutagenesis in mice to address the roles of CB 1 receptors on particular cell types in these functions.
We first showed that exogenous cannabinoids, including ∆9-tetrahydocannabinol (THC, the main psychoactive constituent of cannabis), impairs spatial working memory through the stimulation of astroglial CB 1 receptors. Cannabinoids also induce a form of in vivo long-term depression in the hippocampus that shares several cellular mechanisms with the cannabinoid-induced working memory impairments. These results suggest that cannabinoids disrupt spatial working memory by altering hippocampal synaptic plasticity through astroglial CB 1 receptor stimulation.
We then showed that CB 1 receptors expressed on GABAergic and glutamatergic neurons oppositely control fear coping strategies in the presence of fear conditioned stimuli. The selective and local re-expression of CB 1 receptors in the amygdala of constitutive CB 1 mutant mice allowed to precise the involvement of this brain structure in the regulation of conditioned fear responses by CB 1 receptors.
Altogether, these studies indicate that the endocannabinoid system differentially controls memory functions through its distinct modulation of the activity of specific brain cells. The involvement of astrocytes in the effects of cannabinoids on memory highlights their key roles in cognitive processes and further suggests that astroglial CB 1 receptors might play a role in other high order brain functions. Our results also point the importance of performing thorough behavioral analyses in the experimental models of fear adaptation.
Key words: CB 1 receptors, astrocytes, working memory, conditioned fear, avoidance, mutant mice.
ABBREVIATIONS LIST
AEA N-arachidonoylethanolamide, Anandamide 2-AG 2-Arachidonoylglycerol AMPA Acide alpha-amino-3-hydroxy-5-méthyl-4-isoxazolepropionique AMPAR AMPA receptors ATP Adenosine triphosphate BA Basal amygdala BAPTA 1,2-bis(o-aminophenoxy)ethane- N,N,N',N'-tetraacetic acid BrdU Bromodeoxyuridine BLA Basolateral amygdala cAMP Cyclic adenosine monophosphate
CB 1 receptors Type-1-cannabinoid receptors CB-LTD Cannabinoid-induced long-term depression CCK Cholecystokinin CeA Central nucleus of the amygdala CNS Central nervous system COX Cyclooxygenase CREB cAMP response element-binding CRF Corticotropin-releasing factor CS Conditioned stimulus DAB 3,3'-Diaminobenzidine DAG Diacylglycerol DAGL / DGL DAG lipase DMTS Delayed-matching-to-sample DNA Deoxyribonucleic acid DSE Depolarization-induced suppression of excitation DSI Depolarization-induced suppression of inhibition EAAT1/2 Excitatory amino acid transporter 1/2 eCB(s) Endocannabinoid(s) eCB-LTD Endocannabinoid long-term depression ECS Endocannabinoid system EGTA Ethylene glycol tetraacetic acid EMT Endocannabinoid membrane transporter EPSC / IPSC Excitatory / inhibitory post-synaptic currents ERK1/2 Extracellular signal-regulated kinase 1/2 FAAH Fatty acid amide hydrolase FLAT FAAH-like protein complex GABA Gamma-Aminobutyric acid GAD65 Glutamic acid decarboxylase 65 GDP Guanosine diphosphate GFAP Glial fibrillary acidic protein GLAST Glutamate aspartate transporter GLT-1 Glutamate transporter 1 GPCR G protein-coupled receptors
HFS High frequency stimulation HPA axis Hypothalamic-pituitary-adrenal axis HSP70 Heat shock protein 70 IL β Interleukin β IP3 Inositol triphosphate ITI Intertrial intervals JNK c-Jun N-terminal kinases KO Knock-out LA Lateral amygdala LFS Low frequency stimulation LOX Lipoxygenase LTD Long-term depression LTP Long-term potentiation mAchR Metabotropic acetylcholine receptors MAGL Monoacylglycerol lipase MAPK Mitogen-activated protein kinase MDMA 3,4-méthylène-dioxy-N-méthylamphétamine MFS Mild frequency stimulation mGluR Metabotropic glutamate receptors MPEP 2-Methyl-6-(phenylethynyl)pyridine mPFC Medial prefrontal cortex MCT1/4 Monocarboxylate transporter 1/4 NAPE N-Acylphosphatidylethanolamine NAPE-PLD NAPE-phospholipase D NAT N-acyltransferase NMDA N-methyl-D-aspartate NMDAR NMDA receptors NO Nitric oxide PAG Periaqueducal gray PFC Prefrontal cortex PI3K Phosphatidylinositol 3-kinases PKA, B, C Protein kinases A, B, C PLC Phospholipase C PPR Paired pulse inhibition PVN Paraventricular nucleus (hypothalamus) REM Rapid eye movement RHA Roman high avoidance (rats) RLA Roman low avoidance (rats) RNA Ribonucleic acid SNARE Soluble N-ethylmaleimide sensitive factor attachment protein receptor SNR Substantia nigra parse reticulata SWM Spatial working memory THC Delta-9-tetrahydrocannabinol TMEV Theiler's Murine Encephalomyelitis Virus TNF α Tumor necrosis factor α US Unconditioned stimulus
VMH Ventromedial hypothalamus VTA Ventral tegmental area WM Working memory WT Wild type
LIST OF FIGURES
Figure 1. Brain distribution of the mouse CB 1 receptor protein...... 40
Figure 2. Main biosynthesis pathways of AEA and 2-AG...... 43
Figure 3. Cellular pathways for AEA and 2-AG production, transport and degradation ...... 44
Figure 4. Main intracellular CB 1 receptor signalling pathways ...... 47
Figure 5. Mechanisms of endocannabinoid-mediated long-term depression in the hippocampus and the amygdala ...... 52
Figure 6. Mechanisms of endocannabinoid-dependent long-term depression in the striatum ...... 54
Figure 7. Mechanisms of endocannabinoid-dependent time-spiking long-term depression in the neocortex...... 55
Figure 8. Cell type-specific deletion of the CB 1 gene by the Cre/loxP system strategy in mice ...... 58
Figure 9. Photomicrographs of fluorescent in situ hybridization of CB 1 receptor mRNA in the brain of wild-type (WT), Glu-CB 1-KO and GABA-CB 1-KO mice ...... 59
Figure 10. Cajal’s histological preparations and drawings of astrocytes glial cells already underlined the neurovascular and neuronal coupling of astrocytes...... 62
Figure 11. The different glial cells ...... 64
Figure 12. Astrocyte calcium waves propagation...... 65
Figure 13. Long-term memory systems...... 78
Figure 14. Baddeley's multicomponents model of Working Memory ...... 79
Figure 15. The spatial alternation task in the T-maze ...... 82
Figure 16. The delayed-matching-to-place version of the Morris Water Maze paradigm ...... 83
Figure 17. Representation of US and SC neuronal pathways in fear conditioning...... 90
Figure 18. Representation of the genetic strategy employed to induce a specific rescue of CB 1 receptors expression in GFAP-expressing cells of constitutive CB 1-KO mice...... 117
Table 1 : Available neuron type-specific CB 1-KO mouse lines ...... 60
INDEX
GENERAL INTRODUCTION ...... 21
Part I Résumé de l’introduction générale...... 25 I.1 Le système endocannabinoïde dans le cerveau...... 25
I.2 Les récepteurs CB 1 modulent les fonctions astrocytaires ...... 28
I.3 Les récepteurs CB 1 et la mémoire...... 30 I.4 Objectifs de la thèse ...... 33 Part II The endocannabinoid system in the brain...... 35 II.1 History and Characterization ...... 35 II.1.1 From cannabis to its physiological target ...... 35 II.2 The cannabinoid receptors ...... 37 II.2.1 Characterization and pharmacology...... 37 II.2.2 Distribution of CB 1 receptors in the brain ...... 38 II.3 The endocannabinoids and their synthesis, transport and degradation pathways ...... 41
II.4 Central CB 1 receptor signalling and modulation of synaptic functions ...... 45 II.4.1 CB 1-mediated intracellular signalling...... 45 II.4.2 CB 1 receptor-mediated modulation of synaptic transmission and plasticity...... 48 II.4.3 How does CB 1 receptor signalling modulate long-term synaptic plasticity? ...... 51 II.4.4 CB 1 receptors in the control of brain networks activity and related behavioral consequences ..... 56 II.4.5 Using knock-out mouse models to study CB 1 receptor functions ...... 57
Part III CB 1 receptors and astrocytes ...... 61 III.1 Endocannabinoid system machinery is present in astrocytes ...... 63 III.1.1 What are astrocytes and how do they communicate?...... 63 III.1.2 Cannabinoid receptor expression on astrocytes ...... 66 III.1.3 Do astrocytes produce and metabolize endocannabinoids? ...... 67
III.2 CB 1 receptors participate in astroglial neuro-protective functions...... 68 III.2.1 Astrocytic defense mechanisms...... 68 III.2.2 Astrocyte prevention from excitotoxicity: putative role of CB 1 receptors...... 69 III.2.3 CB 1 receptors regulate astrocytic immune responses...... 69 III.2.4 CB 1 receptors control astroglial differentiation and astrogliosis...... 70
III.3 Role of CB 1 receptors in astroglial metabolic support of neurons...... 71 III.3.1 Astrocytes-neurons metabolic coupling...... 71 III.3.2 Role of CB 1 receptors in astrocytic metabolic functions...... 72
III.4 Astroglial control of information processing: CB 1 receptors also matter ...... 73 III.4.1 Astrocytes and neurons are necessary partners for adapted memories ...... 73 III.4.2 Astroglial CB 1 receptors are key elements of the tripartite synapse...... 74
Part IV CB 1 receptors and memory...... 77 IV.1 The memory systems...... 77 IV.1.1 Short and long –lasting memories...... 77 IV.1.2 Long-term memory systems...... 78 IV.2 Cannabinoid-induced impairment of working memory...... 79 IV.2.1 Working memory ...... 79 IV.2.2 Effects of cannabinoids on human working memory ...... 80 IV.2.3 Cannabinoid and animal models of working memory...... 81 IV.3 The endogenous cannabinoid system controls aversive memories...... 83 IV.3.1 Fear and its regulation: theoretical accounts...... 84 IV.3.2 Aversive learning paradigms in animals...... 85 IV.3.3 How CB 1 receptors modulate aversive memories ...... 90 IV.3.4 CB 1 receptors in other forms of memories ...... 93
Part V Research goals ...... 95
RESULTS ...... 99
Part I CB 1 receptors and working memory...... 101 I.1 Résumé de l’article 1: Les cannabinoïdes perturbent la mémoire de travail par une modulation de la LTD hippocampique exercée par les récepteurs CB 1 astrogliaux...... 101
I.2 Article 1: Acute Cannabinoids Impair Working Memory through Astroglial CB 1 Receptor Modulation of Hippocampal LTD ...... 103
Part II CB 1 receptors and fear responses...... 105 II.1 Résumé de l’article 2: contrôle bimodal des stratégies d’adaptation à la peur par les récepteurs cannabinoïdes CB 1 ...... 105
II.2 Article 2: Bimodal control of fear-coping strategies by CB 1 cannabinoid receptors...... 107
GENERAL DISCUSSION...... 109
Part I Résumé de la discussion générale...... 111
Part II Cellular mechanisms involved in the control of working memory by CB 1 receptors ...... 115 II.1 Cannabinoid-induced LTD and working memory impairment: looking for causality ...... 116
II.2 Roles of CB 1 receptors in vitro versus in vivo ...... 118
II.3 Endogenous CB 1 receptor signalling and memory ...... 121
Part III Neuronal subtype-specific control of learned fear responses by CB 1 receptors ...... 123 III.1 Individual variability of the fear responses ...... 124 III.2 The ECS as a determinant of the types of fear coping strategies ...... 126 III.3 Conclusion...... 127
REFERENCES...... 129
GENERAL INTRODUCTION
21
Introducing the endocannabinoid system (ECS) generally imply to follow a tacit convention, according to which the ancestral consumption of cannabis sativa
(cannabis, marijuana) extracts is the starting point. Cannabis is one of the oldest drugs of abuse in the world, but is also the drug with the longest recorded history of medicinal value. From the beginning of the 20 th century, the legal restrictions in the medical experimentation on cannabis and the social consideration about its consumption for hedonic purposes probably impeded the characterization of its physiological effects and mechanisms of action. Nowadays, knowledge about the
ECS, the endogenous target of cannabinoid agents, is far away from the description of the effects of cannabis intoxication and, in particular, those of its well known active component ∆9-tetrahydrocannabinol (THC). Exogenous cannabinoids are currently used for therapeutical treatment of many diseases including epilepsy and neuropathic pain but their use is limited by several adverse effects mainly due to their psychoactivity, including memory impairments. Indeed, the ECS emerged as a powerful modulator of many physiological functions. Therefore, the description of the cellular mechanisms by which the ECS control body functions is necessary for improving the therapeutical strategies based on cannabinoids.
This thesis work starts with a presentation of the ECS and its cellular mechanisms in the brain. In a second part, the role of the ECS in the regulation of astroglial functions will be extensively described. The control of the ECS on memory functions will be then presented. Two articles published or accepted for publication represent the results section of this report. Finally, the results of these studies will be further discussed and perspectives will be proposed in a general discussion section.
French summaries precede each part of the manuscript.
23
Part I RÉSUMÉ DE L ’INTRODUCTION GÉNÉRALE
I.1 Le système endocannabinoïde dans le cerveau
Les plus anciennes traces relatant l'usage de la plante Cannabis chez l'homme remontent à 6000 ans. Bien avant notre ère, certaines propriétés médicinales du cannabis, comme ses qualités analgésiques, anti-convulsivantes et stimulantes de l'appétit étaient déjà connues (Russo, 2007). A partir du 19 ème siècle, les effets du cannabis commencent à être étudiés par les scientifiques. Dans les années 1960, la caractérisation pharmacologique des molécules contenues dans le cannabis, les cannabinoïdes, et notamment le ∆9-tetrahydrocannabinol (THC), principal composant actif du cannabis (Gaoni and Mechoulam, 1964), a permis d'évaluer leurs effets comportementaux chez l'animal. Par exemple, les cannabinoïdes ont été montrés comme régulant la réactivité au stress, perturbant la mémoire et induisant des effets sédatifs (Carlim and Kramer, 1965; Kiplinger and Manno, 1971; Mokler et al., 1986; Little et al., 1988).
Contrairement à la nature hydrosoluble des neurotransmetteurs connus à cette époque, les cannabinoïdes sont des lipides. Une interaction directe avec la membrane cellulaire a été le premier mécanisme d'action des cannabinoïdes proposé (Hillard et al., 1985). Ce n'est qu'à partir de 1990 que les cibles physiologiques des cannabinoïdes ont été identifiées (Matsuda et al., 1990; Munro et al., 1993). Ils interagissent avec des récepteurs couplés aux protéines G dont les deux principaux sont appelés récepteurs aux cannabinoïdes de type 1 (CB 1) et récepteurs aux cannabinoïdes de type 2
(CB 2). Les récepteurs CB 2 sont principalement exprimés sur les cellules immunitaires et sur les terminaisons nerveuses du système nerveux périphérique (Calignano et al., 1998; Klein et al., 2003).
Les récepteurs CB 1 sont exprimés dans tout le cerveau (Figure 1). En particulier, ils sont abondants dans le néocortex, la formation hippocampique, le striatum, l'amygdale, l'hypothalamus, le cervelet et la substance grise périaqueducale (PAG). Ils sont majoritairement présents sur les terminaisons de différentes populations de neurones, comme les neurones GABAergiques du prosencéphale exprimant la cholecystokinine (CCK) et les neurones glutamatergiques corticaux (Mailleux and
25 GENERAL INTRODUCTION
Vanderhaeghen, 1992; Marsicano and Lutz, 1999). Cependant, des études récentes ont montré que les récepteurs CB 1 cérébraux sont également présents sur la membrane des astrocytes (voir ci-après).
Les deux principaux ligands endogènes des récepteurs aux cannabinoïdes (endocannabinoïdes; eCBs) ont aussi été identifiés dans les années 1990 (Devane et al., 1992; Mechoulam et al., 1995). Il s'agit de l'anandamide (AEA) et du 2-arachidonoylglycerol (2-AG). Ce sont des lipides synthétisés à partir des phospholipides membranaires (Figure 2). Compte-tenu de leur nature hydrophobe, ils ne sont pas stockés dans le milieu aqueux des vésicules synaptiques. Différents mécanismes de transport membranaire et au sein des milieux intra- et extra-cellulaires ont été proposés (Figure 3). Les eCBs sont dégradés par une machinerie enzymatique intracellulaire. La FAAH ( fatty acid amide hydrolase ) et la MAGL ( mono-acyl glycerol lipase ) sont responsables de la dégradation de la majorité de l'AEA et du 2-AG, respectivement. La synthèse et la libération des eCBs sont déclenchées par l'activité neuronale (Piomelli, 2003). La récupération cellulaire et la dégradation des eCBs s’opèrent rapidement, leur diffusion est donc temporellement et spatialement restreinte. Par conséquent, l'action des eCBs sur les récepteurs CB1 se fait « à la demande ». Les eCBs, les récepteurs aux cannabinoïdes et les complexes enzymatiques de synthèse et de dégradation des eCBs constituent le système endocannabinoïde (SEC).
L'activation des récepteurs CB 1 engendre une cascade d'évènements intracellulaires, qui comprend la modulation de l'ouverture de canaux ioniques et de la synthèse protéique, dont la finalité est l'inhibition de la libération de neurotransmetteurs (Nicholson et al., 2003; André and Gonthier,
2010). Ainsi, la signalisation dépendante des récepteurs CB 1 peut induire une diminution transitoire de la libération des neurotransmetteurs (Chevaleyre et al., 2006; Kano et al., 2009). Cette forme de plasticité synaptique est induite par une brève dépolarisation de la membrane post-synaptique et résulte en la suppression de l’excitation neuronale si elle concerne l’inhibition de la libération de glutamate ( depolarization-induced suppression of excitation , DSE) ou la suppression de l’inhibition de l’activité neuronale dans le cas d’une diminution de la libération de GABA ( depolarization-induced suppression of inhibition , DSI). Les phénomènes de DSE et DSI sont dépendants de l’augmentation de la concentration intracellulaire de Ca 2+ au niveau post-synaptique et sont induits par une diminution des courants d’origine pré-synaptique. Ils ont été observés dans différentes régions du cerveau comme l’hippocampe, l’amygdale, le néocortex, le cervelet ou le striatum. L’activité du SEC peut aussi induire des processus de plasticité synaptique à long-terme. En particulier, la stimulation des récepteurs CB 1 peut entrainer une diminution durable de la libération de neurotransmetteurs ou dépression à long-terme ( endocannabinoid-mediated long-term depression , eCB-LTD ; Chevaleyre et al., 2006). Une fois induit, la maintenance du phénomène d’eCB-LTD ne nécessite pas l’activation continue des récepteurs CB 1. Plusieurs formes d’eCB-LTD ont été décrites dans différentes structures du cerveau; elles diffèrent notamment par leur protocole d’induction, par leur contrôle homo- ou hétéro-synaptique de la libération de neurotransmetteurs, et par les mécanismes cellulaires qui les sous-tendent. Contrairement à d’autres formes très étudiées de plasticité à long-terme, les eCB-LTDs ne sont pas dépendantes de l’activité des récepteurs NMDA (N-methyl-D-aspartate) post-synaptiques, mais elles requièrent la signalisation des récepteurs métabotropiques au glutamate (mGluR) de groupe 1 et/ou une élévation de la concentration de Ca 2+ intracellulaire au niveau post-synaptique
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 26 Part I - Résumé de l’introduction générale
(Figure 5-6). Par exemple, une stimulation à haute fréquence des neurones pyramidaux de l’hippocampe induit une dépression à long-terme de la transmission inhibitrice (I-LTD; Chevaleyre and Castillo, 2003; Heifets et al., 2008; Figure 4). Cette forme d’eCB-LTD est dépendante des récepteurs 2+ CB 1. La I-LTD ne dépend pas de variation de la concentration de Ca intracellulaire au niveau post- synaptique et est caractérisée par une expression pré-synaptique, ce qui indique qu’elle s’exprime par une inhibition de la libération de neurotransmetteurs.
Qu’elles soient à court terme ou persistantes, ces formes d’inhibition de la neurotransmission dépendantes des récepteurs CB 1 faciliteraient l’induction d’autres phénomènes de plasticité synaptique associés à des processus de mémoire observés au niveau comportemental tels que l’adaptation des réponses de peur conditionnées (Marsicano et al., 2002; Azad et al., 2004). De surcroît, la plasticité synaptique a été associée à l’activité synchrone de populations de neurones (Dragoi et al., 2003; Girardeau et al., 2009). L’administration de cannabinoïdes est connue pour ses effets délétères sur les performances de mémoire spatiale dépendants de l’hippocampe. Ces effets ont été associés à une désynchronisation des rythmes oscillatoires des neurones de l’hippocampe par les cannabinoïdes chez le rat réalisant un test de mémoire spatiale (Robbe et al., 2006 ; Robbe and Buszaki, 2009).
Les récepteurs CB 1 sont localisés sur différents types de populations cellulaires. Afin de distinguer le rôle de ces récepteurs sur ces types cellulaires, il est nécessaire d’employer des outils expérimentaux permettant un ciblage des récepteurs CB 1 spécifique à un type cellulaire particulier. L’utilisation du système Cre/LoxP est une stratégie de mutation génétique permettant de supprimer un gène uniquement dans une population cellulaire donnée chez la souris (Sauer and Henderson, 1989; Figure 8). Ce système repose sur le croisement de deux lignées de souris génétiquement modifiées.
Dans le cas d’une mutation conditionnelle des récepteurs CB 1, chez les souris de la première lignée, la séquence génétique codant pour les récepteurs CB1 est encadrée par deux séquences loxP (“ locus of crossover P1 ”). Le gène dit « floxé » s’exprime normalement, ces souris peuvent donc être considérées comme de type sauvage. La deuxième lignée de souris est caractérisée par l’expression de la protéine Cre recombinase sous le contrôle d’une séquence promotrice spécifique d’un type cellulaire donné. La protéine Cre excise la séquence comprise entre les sites loxP, dans ce cas, celle codant pour le récepteur CB 1. Ainsi, lorsque ces deux lignées de souris sont croisées, leur descendance exprimera une délétion des récepteurs CB 1 restreinte au type cellulaire dans lequel la protéine Cre s’exprime, tandis que l’expression de ces récepteurs sera maintenue dans les autres cellules de l’organisme. Plusieurs lignées de souris mutantes conditionnelles pour les récepteurs CB 1 ont été générées par ce système (Tableau 1). En particulier, la génération des souris caractérisées par une délétion des récepteurs CB 1 restreinte aux neurones glutamatergiques corticaux (Glu-CB 1-
KO) et celles caractérisées par une délétion des récepteurs CB 1 spécifique aux neurones
GABAergiques du prosencéphale (GABA-CB 1-KO) a permit de révéler des conséquences distinctes de la régulation des transmissions glutamatergiques et GABAergiques par les récepteurs CB 1 sur plusieurs fonctions comme la neuroprotection ou la prise alimentaire (Figure 9 ; Marsicano et al., 2003; Monory et al., 2006; Bellocchio et al., 2010).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 27 GENERAL INTRODUCTION
I.2 Les récepteurs CB 1 modulent les fonctions astrocytaires
Les cellules gliales sont les cellules les plus abondantes dans le cerveau humain. Elles ont été décrites il y a 150 ans et les premières techniques de marquage cellulaire ont permis d’observer leurs connections avec les vaisseaux sanguins du cerveau et les neurones (Virchow, 1858 ; Cajal, 1913). Les astrocytes ont été distingués des autres types de cellules gliales par leur forme étoilée due à leurs nombreux prolongements autour de leur corps cellulaire (Lenhossek, 1891 ; Del Río-Hortega, 1919; 1921). Au 19 ème siècle, la caractérisation des communications neuronales a conduit à considérer les neurones comme seuls supports du stockage et de la propagation de l’information (Sherrington, 1906). Les astrocytes ne sont pas électriquement excitables, ils ne génèrent pas de potentiels d’action. C’est pourquoi ils n’ont été étudiés que dans le cadre de leurs fonctions métaboliques et protectrices du cerveau jusqu’à très récemment. Les astrocytes se distinguent également des autres populations cellulaires du cerveau par leur composition biochimique. Ils expriment des protéines spécifiques, comme la protéine acide fibrillaire gliale ( glial fibrillary acidic protein, GFAP ; Middeldorp and Hol, 2011) qui est très utilisée pour le marquage anatomique et le ciblage génétique des astrocytes. Les astrocytes communiquent avec les neurones, les vaisseaux sanguins et entre eux. Ils sont organisés en réseaux grâce à un couplage par les jonctions gap qui assure le passage de petites molécules. Les jonctions gap sont cruciales pour le maintien de l’homéostasie du cerveau ainsi que pour les processus de plasticité synaptique (Giaume et al. 1997; Rouach, 2008). L’excitabilité des astrocytes repose en partie sur des variations de concentration de Ca 2+ intracellulaire qui peuvent se propager à d’autres astrocytes, vers les vaisseaux sanguins ou les neurones. Les astrocytes sont également capables de libérer des messagers chimiques appelés gliotransmetteurs (par analogie aux neurotransmetteurs) comme le GABA, la D-serine, l’ATP, l’adénosine, le monoxyde d’azote (NO) et le facteur de nécrose tumorale α ( tumor necrosis factor α, TNF α; Perea et al., 2009).
La présence des récepteurs CB 1 sur les astrocytes a fait l’objet de débats dans la littérature, puisque leur détection était variable selon le type de préparation ( ex vivo ou cultures astrocytaires) et le type d’anticorps utilisés (Stella, 2010). Cependant, des preuves fonctionnelles montrant un contrôle direct des récepteurs CB 1 sur la signalisation astrocytaire de l’hippocampe ont été reportées ex vivo (Navarrete and Araque, 2008 ; 2010). Les astrocytes sont également équipés de la machinerie de synthèse et de dégradation des eCBs, ce qui suggère qu’ils sont capables de libérer des eCBs. Cependant, leurs rôles dans la communication astrocytaire sont actuellement inconnus.
Les astrocytes sont équipés pour détecter et réagir face aux agents toxiques et infectieux qui menacent le système nerveux central. Ils expriment différents transporteurs qui assurent l’évacuation des neurotransmetteurs, comme le GABA et le glutamate, de l’espace inter-synaptique. Les astrocytes maintiennent également l’homéostasie ionique en accumulant l’excès de K + résultant de l’activité neuronale. Ainsi, ils assurent un maintien de l’excitabilité neuronale et la protection des neurones contre l’excitotoxicité (Takana et al., 1997 ; Walz, 2000; Takuma et al., 2004). Les astrocytes sont immunocompétents ; ils sont capables de libérer des facteurs inflammatoires comme les cytokines (par exemple, le TNF α) en réponse à des agents infectieux, des lésions tissulaires où à
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 28 Part I - Résumé de l’introduction générale une accumulation protéique anormale. Ces réponses neuro-inflammatoires sont nécessaires à la restriction des dommages cérébraux mais peuvent également altérer les fonctions neuronales et la viabilité des neurones. La neuro-inflammation a été impliquée dans certaines pathologies neuro- dégénératives comme la maladie d’Alzheimer (Farina et al., 2007). Les récepteurs CB 1 ont été impliqués dans la récupération des acides aminés excitateurs par les astrocytes. Chez des astrocytes en culture, la stimulation des récepteurs CB 1 inhibe la récupération de D-aspartate (Shivachar, 2007). Au contraire, l’AEA prévient les effets neurotoxiques induit par l’AMPA via la stimulation des récepteurs CB 1 in vitro et in vivo , suggérant que le SEC protège contre l’excès de glutamate en facilitant sa récupération par les astrocytes (Loria et al., 2010). Par ailleurs, la stimulation des récepteurs CB 1 inhibe la libération de facteurs inflammatoires comme le NO et le TNF α in vitro par les astrocytes, indiquant le potentiel anti-inflammatoire des cannabinoïdes par leur action sur les récepteurs CB 1 astrogliaux (Molina-Holgado et al., 1997 ; Ortega-Gutiérrez et al., 2005 ; Sheng et al., 2005 ; Froger et al., 2009). Les cellules progénitrices des neurones et des astrocytes expriment la protéine GFAP ainsi que d’autres marqueurs spécifiques des cellules immatures. Les récepteurs CB 1 sont présents sur ces progéniteurs cellulaires et leur stimulation favorise leur prolifération ainsi que leur différentiation en astrocytes (Aguado et al., 2006). A l’inverse, la suppression des récepteurs CB 1 inhibe cette astrogliogénèse chez des souris adultes, indiquant que le SEC joue un rôle dans le développement du cerveau. D’autre part, la stimulation des récepteurs CB 1 empêche la prolifération et la croissance astrocytaire induite dans certaines conditions pathologiques évoquées plus haut (astrogliose ; Esposito et al., 2007).
20 % de l’énergie produite par l’organisme est consommée par le cerveau. De part leur proximité avec les vaisseaux sanguins cérébraux, les astrocytes sont les principaux fournisseurs de substrats métaboliques aux neurones. A partir du glucose provenant de la circulation sanguine ou des stocks de glycogène préformés, les astrocytes synthétisent le lactate qui sera transporté aux neurones comme principale source énergétique (Pellerin and Magistretti, 2005 ; Brown and Ransom, 2007). Les acides gras sont une source énergétique alternative utilisée par l’organisme, par exemple, dans des conditions pauvres en carbohydrates. Les corps cétoniques sont synthétisés à partir des acides gras principalement par le foie mais aussi par les astrocytes et sont utilisés comme carburant pour la respiration cellulaire (Guzman and Blazquez, 2001; 2003). Peu d’études se sont intéressées au rôle du SEC dans les fonctions métaboliques astrocytaires. Sur des astrocytes en culture, la stimulation des récepteurs CB 1 augmente l’oxydation du glucose et du glycogène mais les conséquences de ce mécanisme in vivo sont inconnues (Sanchez et al., 1998). De même, les cannabinoïdes exogènes stimulent la cétogenèse astrocytaire in vitro (Blazquez et al., 1999) . Vu les propriétés anti-convulsivantes des régimes alimentaires cétoniques et des cannabinoïdes, il est possible que le SEC protège contre l’hyper-excitabilité neuronale par une facilitation de la cétogenèse.
Comme précisé plus haut, les astrocytes sont capables de libérer des gliotransmetteurs. Ils expriment également une grande variété de récepteurs. Ainsi, les astrocytes communiquent avec les neurones. Par exemple, le glutamate libéré par les astrocytes contrôle l’excitabilité neuronale dans plusieurs structures cérébrales incluant l’hippocampe. La signalisation astrocytaire est capable de moduler les phénomènes de plasticité synaptique par différents mécanismes. Il a été proposé que la
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 29 GENERAL INTRODUCTION libération exocytotique d’ATP par les astrocytes contribue aux conséquences négatives de la privation de sommeil sur la potentialisation à long-terme induite dans l’hippocampe (long-term potentiation , LTP) et sur la reconnaissance spatiale (Halassa et al., 2009). La D-sérine libérée par les astrocytes est nécessaire à la LTP dépendante des récepteurs NMDA dans l’hippocampe, l’hypothalamus et le cortex préfrontal (Panatier et al., 2006 ; Henneberger et al., 2010 ; Fossat et al., 2012). De manière intéressante, la D-sérine est également connue pour faciliter différents processus de mémoire (Bado et al., 2011). Le rôle direct des fonctions métaboliques des astrocytes, en particulier du transport du lactate vers les neurones, dans la plasticité synaptique de l’hippocampe et dans la consolidation de la mémoire a été récemment établi (Suzuki et al., 2011). Deux études ont montré l’implication des récepteurs CB 1 dans la communication astrocytes-neurones dans l’hippocampe (Navarrete and
Araque, 2008 ; 2010). La stimulation des récepteurs CB 1 localisés sur les astrocytes augmente la concentration de Ca 2+ astrocytaire et est nécessaire au maintien de l’excitabilité des neurones. Leur stimulation favorise également la transmission neuronale excitatrice par un mécanisme dépendant des récepteurs mGluR de groupe 1, suggérant que les récepteurs CB 1 astrogliaux pourraient être impliqués dans certains phénomènes de plasticité synaptique ainsi que dans des fonctions de mémoire associées.
I.3 Les récepteurs CB 1 et la mémoire
La mémoire réfère à la capacité à retenir et rappeler des expériences passées (Squire, 2004). Ces informations peuvent être retenues pendant une courte période de temps (par exemple, quelques secondes ou minutes) ou pendant des périodes plus longues, comme des mois ou des années. La mémoire à long-terme se distingue de la mémoire à court-terme par la synthèse de protéines de novo. Les systèmes de mémoire peuvent également se distinguer selon le type d’information traité. Par exemple, la mémoire représentative relève de la mémoire des faits (« savoir quoi »). La mémoire non- déclarative relève quant à elle de l’ensemble des processus qui permettent d’acquérir des automatismes et habitudes (« savoir comment »), ou des réponses conditionnées. Le concept de mémoire de travail a été proposé afin de rendre compte du stockage et du traitement de l’information pendant une courte période de temps (Miller et al., 1960 ; Atkinson et Shiffrin, 1968 ; Baddeley et Hitch, 1974). La mémoire de travail peut être évaluée chez l’homme au cours de tests qui impliquent de retenir pendant un bref délai une liste de lettres ou de chiffres et de les rappeler dans l’ordre ou le désordre ( digit span tests ). Le modèle de la mémoire de travail de Baddeley (2007) est le plus reconnu actuellement. Il est composé d’un administrateur central permettant le stockage temporaire et la manipulation d’informations afin de réaliser des tâches cognitives. L’administrateur central est en relation avec trois systèmes de stockage indépendants des informations sensorielles, la boucle phonologique (informations auditives) et le calepin visuo-spatial (informations visuelles). Le tampon épisodique est une unité de stockage multi-sensorielle assurant le transfert d’information entre l’administrateur central et la mémoire à long-terme (Figure 14). Chez l’animal, certaines tâches sont associées à des performances requérant la mémoire de travail comme le labyrinthe radial (Olton, 1979), permettant ainsi d’en étudier les mécanismes neurophysiologiques. Des études d’imagerie chez l’homme et les modèles animaux permettent de comprendre les mécanismes cérébraux
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 30 Part I - Résumé de l’introduction générale impliqués dans des dimensions spécifiques de la mémoire de travail. Ainsi, les interactions neuronales entre l’hippocampe et le cortex préfrontal seraient nécessaires au stockage à la manipulation d’informations spatiales (Laroche et al., 2000).
Les conséquences négatives d’une prise aigüe de cannabinoïdes comme le THC sur les performances de mémoire de travail ont été fréquemment reportées chez l’homme (Ranganathan and D’Souza, 2006). Le test d’appariement ou delayed-matching-to-sample (DMTS) est une tâche de mémoire de travail qui consiste à sélectionner parmi plusieurs stimuli (par exemple des images) celui qui correspond à un stimulus de référence présenté précédemment. Chez des sujets consommateurs occasionnels de THC, l’inhalation ou l’administration par voie intraveineuse de THC à une dose de 2,5 mg suffit à augmenter le nombre d’erreurs au cours du test de DMTS (D’Souza et al., 2004). Les conséquences d’une consommation chronique de THC sur la mémoire de travail sont plus controversées. Du reste, il a été fréquemment reporté qu’une administration aigüe de THC chez des consommateurs réguliers engendre une tolérance aux déficits de mémoire de travail (Heisman et al., 1997). En revanche, ces effets délétères ne persisteraient pas chez des consommateurs réguliers lorsqu’ils ne sont pas sous l’influence de THC (Solowij, 1998). Les déficits de mémoire de travail ont été reproduits chez l’animal. Chez le rongeur, le test d’alternance spatiale dans le labyrinthe en T (Brito and Thomas, 1981) et la procédure de delayed-matching-to-position dans le labyrinthe aquatique de Morris (Morris et al., 1982) sont deux modèles très utilisés pour évaluer les performances de mémoire de travail spatiale chez le rongeur. Au cours de la tâche d’alternance spatiale dans le labyrinthe en T, les animaux doivent alterner la visite de deux bras opposés afin d’obtenir une récompense alimentaire (Figure 15). Au cours de la tâche de delayed-matching-to- position dans le labyrinthe aquatique de Morris, les animaux doivent retrouver la position d’une plateforme invisible dans un bassin à partir d’indices spatiaux, afin d’échapper à l’immersion dans l’eau. La mémoire de travail est engagée dans ce test lorsque la position de la plateforme est modifiée à chaque séance (Figure 16). L’administration aigüe de THC (3,0 à 5,0 mg/kg) diminue ces performances par la stimulation des récepteurs CB 1 (Jensch et al., 1997; Ferrari et al., 1999; Nava et al., 2000; 2001; Varvel et al., 2001; 2004). Ces effets ont été associés à des changements d’activité des neurones de l’hippocampe, et l’administration de cannabinoïdes dans l’hippocampe induit des déficits de mémoire de travail dans le test labyrinthe aquatique de Morris (Heyser et al., 1993 ;
Lichtman et al., 1995; Wise et al., 2009). Ainsi, les récepteurs CB 1 de l’hippocampe sont impliqués dans les conséquences de l’administration de cannabinoïdes sur la mémoire de travail. Pourtant, les mécanismes cellulaires impliqués dans ces effets sont inconnus.
Le SEC exerce un contrôle sur la mémoire émotionnelle et en particulier sur l’adaptation des réponses de peur conditionnées. La peur peut être définie par un état émotionnel subjectif déplaisant élicité par la présence d’un danger. Elle s’accompagne de réponses physiologiques dont la finalité est de supprimer la source du danger. La peur est une réaction normale et nécessaire pour la survie des individus. Néanmoins, les réponses de peur doivent s’adapter aux changements de situation afin de maintenir des activités physiques et mentales normales (Gross, 1998 ; 2002). Certains auteurs utilisent le terme de coping pour définir les comportements engagés par les individus dans des situations aversives ayant pour but d’atténuer les effets de stimuli aversifs sur la physiologie de
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 31 GENERAL INTRODUCTION l’organisme (Wechsler, 1995). Il est admit que les émotions, et la peur en particulier, soient préservées au cours de l’évolution des espèces (Darwin, 1872). Il serait donc possible d’étudier les comportements émotionnels des animaux et les réponses physiologiques qui les accompagnent afin de comprendre les mécanismes des émotions humaines (MacLean, 1949 ; Blanchard and Blanchard, 1972). Le paradigme du conditionnement classique de peur est très utilisé pour comprendre les bases neurales de la peur chez l’animal (Pavlov, 1927 ; Blanchard and Blanchard, 1969 ; Fanselow, 1980 ; LeDoux, 2000). Dans ce test, un stimulus initialement neutre (qui ne suscite aucune réponse émotionnelle observable, par exemple un son de faible intensité) est associé à la délivrance d’un stimulus aversif inconditionnel (SI). Suite à cette association, la présentation du son suffit à éliciter une réponse de peur (réponse de peur conditionnée), il devient alors un stimulus conditionnel (SC). Le conditionnement classique de peur est observé chez de nombreuses espèces, des invertébrés aux mammifères incluant l’homme. Chez le rongeur, plusieurs types de réponses de peur conditionnées peuvent être mesurées, qu’elles soient autonome, endocriniennes ou comportementales. En particulier, la quantification de la réponse de freezing , ou absence de tout mouvement excepté ceux dédiés à la respiration, est souvent utilisée pour mesurer l’intensité du conditionnement de peur chez le rongeur (Blanchard and Blanchard, 1969; Maren, 2008). Cependant, d’autres réponses peuvent être induites par le SC comme le comportement de digging (creusement dans la litière) ou de rearing (posture debout sur les membres postérieurs) et rivaliser avec l’occurrence du comportement de freezing (Gozzi et al., 2010). Une fois apprise, les réponses de peur conditionnée peuvent être atténuées par la présentation répétée ou prolongée du SC en l’absence du SI, c’est l’extinction de la mémoire de peur (Pavlov, 1927 ; Rescola, 1968). L’extinction de la mémoire de peur n’est pas considérée comme l’oubli de l’association SC-SI mais comme un nouvel apprentissage puisque la réponse de peur conditionnée peut être restaurée, par exemple, suite au seul passage du temps (Myers and Davis, 2007). L’atténuation de la réponse de peur observée au cours de la procédure d’extinction impliquerait également un processus non associatif (habituation) due à une présentation prolongée du SC (Kamprath and Wotjak, 2004). Les mécanismes neuronaux du conditionnement de peur sont complexes et impliquent l’activité de plusieurs régions du cerveau. L’amygdale est une structure clef pour l’intégration des informations sensorielles relatives aux SC et SI (LeDoux, 2000). Elle est composée de différents noyaux incluant le noyau latéral (LA), le noyau central (CeA) et le noyau basal (BA), formant des circuits neuronaux organisés et dont la plasticité permet l’association entre le SC et SI (Rogan et al. 1997 ; Ehrlich et al., 2009 ; Figure 14). L’intégrité de l’amygdale est également nécessaire à l’extinction de la mémoire de peur (Myers and Davis, 2007). Par ailleurs, les connexions entre l’hippocampe et l’amygdale sont importantes pour l’intégration des stimuli relatifs au contexte de l’apprentissage aversif (LeDoux, 2000). Le cortex préfrontal médian (mPFC) est également nécessaire à l’expression et à l’extinction des réponses de peur conditionnées ; son activité serait importante pour l’encodage de la saillance des informations (Laviolette et al., 2005; Quirk and Mueller, 2008).
Les tests d’évitement conditionnés sont d’autres types d’apprentissages aversifs qui reposent sur l’acquisition d’un comportement dont l’occurrence supprime l’apparition du SI (Estes and Skinner, 1941 ; Sidman et al., 1957, Brady and Harris, 1977). Au cours de l’évitement actif, les animaux doivent
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 32 Part I - Résumé de l’introduction générale engager une réponse de fuite pour échapper au SI tandis que l’évitement passif suppose d’éviter d’approcher l’endroit associé à la délivrance du SI. L’intégrité de l’amygdale est également nécessaire pour l’acquisition de l’évitement conditionné. Des études lésionnelles chez le rat ont proposé que le LA soit nécessaire à l’association entre le SC et le SI. Le CeA est nécessaire à l’acquisition et l’expression de la réponse conditionnée de freezing tandis que et le BA est nécessaire à l’acquisition et l’expression de l’évitement actif (Amorapanth et al., 2000; Choi et al., 2010; Figure 14). De manière intéressante, plusieurs études ont reporté que les performances d’évitement actif sont sujettes à la variabilité individuelle chez le rongeur, suggérant que l’apprentissage des réponses de peur conditionnées pourraient dépendre de tendances individuelles spontanées à adopter des réponses défensives particulières (Bolles and Popp, 1964; Blanchard and Blanchard, 1969; Steimer et al., 1997; Koolhaas et al., 1999, 2010; Lázaro-Muñoz et al., 2010; Vicens-Costa et al., 2010; Díaz-Morán et al., 2011).
Bien que le rôle des récepteurs CB 1 dans l’acquisition de la réponse conditionnée de freezing soit très discuté dans la littérature, leur signalisation endogène est nécessaire à l’extinction de la mémoire de peur (Marsicano et al., 2002 ; Arenos et al., 2006 ; Cannich et al., 2004). L’activité endogène des récepteurs CB 1 faciliterait l’extinction de la mémoire de peur au moins en partie par l’habituation de la réponse de freezing à la présentation prolongée du SC (Kamprath et al., 2006).
D’autre part, la délétion génétique des récepteurs CB 1 facilite l’acquisition de l’évitement actif (Martin et al. 2002). L’activité des récepteurs CB 1 permettrait l’extinction de la réponse conditionnée de freezing en favorisant l’induction de la plasticité des connexions synaptiques excitatrices et inhibitrices de l’amygdale, et la phosphorylation de protéines comme ERK1/2 dans cette structure (Marsicano et al., 2002 ; Cannich et al. 2004 ; Chhatwal et al., 2009 ; Kamprath et al., 2011). Les récepteurs CB 1 localisés dans le mPFC, l’hippocampe et la PAG sont également impliqués de manière spécifique dans l’acquisition ou l’expression de la réponse conditionnée de freezing (Laviolette and Grace, 2006 ;
De Olivera Alvares et al., 2008 ; Resstel et al., 2008). Cependant, la contribution des récepteurs CB 1 localisés sur les neurones GABAergiques et glutamatergiques dans ces apprentissages conditionnés est inconnue.
I.4 Objectifs de la thèse
Les récepteurs CB 1 sont abondants dans le cerveau et sont présents sur différentes populations cellulaires. Sur les neurones, l’activité des récepteurs CB 1 inhibent la libération des neurotransmetteurs excitateurs glutamatergiques et inhibiteurs GABAergiques tandis que sur les astrocytes, leur activation favoriserait la transmission excitatrice. La caractérisation des mécanismes cellulaires par lesquels les récepteurs CB 1 contrôlent les fonctions du cerveau nécessite l’emploi d’outils expérimentaux permettant de discriminer leur rôle sur ces populations cellulaires. L’approche
électrophysiologique a permit de mettre en évidence l’importance des récepteurs CB 1 dans différentes formes de plasticité synaptique dans le cerveau qui sont associées aux fonctions de mémoire.
L’activité des récepteurs CB 1 contrôle les processus de mémoire. En particulier, leur stimulation par les cannabinoïdes exogènes perturbe la mémoire de travail et la signalisation endogène des
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 33 GENERAL INTRODUCTION
récepteurs CB 1 assure l’adaptation des réponses de peur conditionnées. Cependant, les contributions respectives des récepteurs CB 1 exprimés sur distinctes populations cellulaires dans ces processus de mémoire sont inconnues.
L’objectif général de ce travail de thèse est de caractériser les mécanismes cellulaires par lesquels les récepteurs CB 1 contrôlent ces fonctions de mémoires. Nous avons utilisé le modèle de mutation conditionnelle des récepteurs CB 1 chez la souris afin d’analyser les conséquences comportementales de la délétion de ces récepteurs sur des types cellulaires particuliers. Nous nous sommes intéressés dans un premier temps à la description des mécanismes cérébraux par lesquels les récepteurs CB 1 modulent la mémoire de travail. Cette étude constitue la première partie des résultats. La seconde étude est consacrée à la description des mécanismes neuronaux impliqués dans la régulation des réponses de peur apprises par les récepteurs CB 1. Les résultats de ce travail sont présentés dans la seconde partie de cette section.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 34
Part II THE ENDOCANNABINOID SYSTEM IN THE BRAIN
II.1 History and Characterization
II.1.1 From cannabis to its physiological target
The written history of the relationship between cannabinoids and humans starts in China, where the first archaeological traces of cannabis cultures were found and dated at 6000 years ago. Chinese manufactured textiles and papers with the plant’s stem, but also comestible oil. The world’s oldest Chinese medicine used also cannabis against pain, intestinal and reproductive troubles. Medicinal and religious uses of the hemp plant were also noticed in India well before Christian era, where lots of its therapeutic properties were already sensed. Indeed, the dry leaves or flowers were consumed as an analgesic/anesthetic, anticonvulsant, sedative, antibiotic, antiparasite, antispasmodic, pro-digestive, appetite stimulant, diuretic and aphrodisiac. In Europe, few archaeological pieces of evidences attested that Greeks and Romans consumed cannabis seeds before Christian era, mainly for euphoric purposes (Russo, 2007).
In the 16 th century, the medicinal use of cannabis was diffused in China, India, Arabia and reached Africa and South America. It is only from the 19 th century that cannabis medication was introduced into Western Europe. In 1839, O’Shaughnessy, an Irish medical officer of the British Army in Calcutta, described the narcotic and some therapeutic properties of cannabis preparations experimentally assessed in human, such as anti-rheumatisms, -convulsions and -muscular spasms, and evaluated the plant’s toxicity in animals (O’Shaughnessy, 1839ab). The diffusion of his publications on cannabis pharmacology marked the introduction of cannabis medicine in Western Europe. Six years later, in ‘Du Hashish et de l’Aliénation Mentale’, the French psychiatrist Moreau provided a thorough analysis of the psychomimetic effects of cannabis resin ( Hashish ) from his self or students’ consumption reports, arguing that it represents a powerful mean to investigate mental illness (Moreau, 1845). Both these contributions sparked off the multiplication of studies on cannabis medical
35 GENERAL INTRODUCTION values in Europe and North America, facilitated by the selling of plant extracts by private laboratories (e.g. Merck in Germany or Eli Lilly in USA) from the second half of the 19 th century. Indeed, several effects of these preparations, including sedation, analgesia, appetite-promotion and others were experimentally identified (Mikuriya, 1969).
At the beginning of the 20 th century, research on the medical asset of cannabis declined. Since the active component of the plant was not identified yet, studies were based on variable concentration extracts, impairing the reliability of the results (Rödner Sznitman et al., 2008). Moreover, the development of vaccines against infectious diseases, of effective analgesics such as aspirin and morphine, and the use of barbiturates as sedative altogether rivaled with cannabis medical use. On the top of that, in 1937, medicinal cannabis use was restricted to an excise tax payment followed by its removal from the American pharmacopeia in 1941. Paradoxically, the worldwide cannabis consumption for hedonic and recreational purposes exploded until the 1970’s, and remains high nowadays (Mikuriya, 1969; Rödner Sznitman et al., 2008; World Drug Report 2011 1).
The identification of the chemical components of cannabis, named cannabinoids, started already from the 1940’s with the isolation of cannabinol and cannabidiol (Adams, 1940; Jacob and Todd, 1940). In 1964, the structural characterization of the ∆9-tetrahydrocannabinol (THC) by Gaoni and Mechoulam marked the first return of cannabis research (Gaoni and Mechoulam, 1964). A large amount of pharmacological studies in human and animal highlighted that THC mediates the major psychoactive properties of the plant (Kiplinger and Manno, 1971; Carlini et al., 1974). Mechoulam et al. then achieved the complete synthesis of the pure compounds, established their molecular structures, and began to study their structure-activity relationships (Mechoulam et al., 1972; Razdan, 1986). The already known consequences of cannabis consumption were thus rediscovered and precisely characterized by systemic administration of various doses of THC in animals. The characteristic sedative effects of THC (called “tetrad”) consisting in hypolocomotion, hypothermia, analgesia, and catalepsy (impaired ability to initiate movements) were observed at the same dose and time ranges (Little et al., 1988). THC was also shown to modulate the activity of the hypothalamo- pituitary-adrenal axis (HPA) and stress reactivity (Mokler et al., 1986). THC was also observed to have anticonvulsant properties and THC was efficient to promote food intake at low doses (Corcoran et al., 1973; Wada et al., 1973; Brown et al., 1977). Soon after its identification, THC was found to mediate the cannabis-induced memory impairments (Carlim and Kramer, 1965). Despite this regain of interest in the behavioral effects of THC, the difficulty to establish the physiological mode of action of cannabinoids attenuated the renewed enthusiasm for cannabinoid research.
Due to the lipophilic nature of cannabinoids, a direct interaction with the cell membranes was first proposed (Hillard et al., 1985). The first report that THC might exert its effects by interacting with a specific receptor protein in the brain was proposed by Howlett (1984), demonstrating that cannabinoids decrease cyclic adenosine monophosphate (cAMP) levels in neuroblastoma cell cultures, suggesting a G αi/o -coupled receptor-mediated action. Soon after, the synthesis of new THC
1 http://www.unodc.org/unodc/en/data-and-analysis/WDR-2011.html
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 36 Part II - The endocannabinoid system in the brain analogues with higher metabolic stability and potency in typical assays of central cannabimimetic activity, such as CP-55,940 and HU-210 (Mechoulam et al., 1990) led to the identification of a receptor-mediated interaction between cannabinoid compounds and biological tissues (Howlett et al., 1988), suggesting the existence of a still-to-be identified G protein-coupled receptor (GPCR) sensitive to cannabinoid compounds, including THC (Devane et al., 1988). Two years later, Matsuda et al.
(1990) cloned the first cannabinoid receptor (CB 1) and opened the way to the molecular approach of the biology of cannabinoids. CB 1 receptors were found mainly in the brain, indicating a major function in the central effects of exogenous cannabinoids. Later, Munro et al. (1993) cloned a second receptor, named CB 2 receptors, also able to bind cannabimimetic compounds such as THC, but whose expression is mainly limited to peripheral cells and tissues of the immune system. Furthermore, the real raison d’être of the cannabinoid receptors was found with the identification of their endogenous agonists in mammals, named endocannabinoids (eCBs) by analogy with the plant natural products. An amide of arachidonic acid, named anandamide (AEA), was the first eCB discovered (Devane et al., 1992) followed by 2-arachidonoylglycerol (2-AG; Mechoulam et al., 1995). Furthermore, a system for eCBs uptake was uncovered in cells (Beltramo, 1997; Hillard et al., 1997) and specific synthesis and degradation pathways were identified (Di Marzo et al., 1994). Such an impressive efflux of new data and discoveries led to postulate the existence of a new signalling system, made by endogenous compounds, their specific receptors and mechanisms for production, release, uptake and degradation (Piomelli, 2003). The ECS is a key physiological system involved in establishing and maintaining human health. In each tissue, the ECS performs different tasks to guarantee body homeostasis, the maintenance of a stable internal environment despite fluctuations in the external environment (Pagotto, 2005).
II.2 The cannabinoid receptors
II.2.1 Characterization and pharmacology
The first identified receptor, named type-1-cannabinoid (CB 1) receptor, was cloned in rat (Matsuda et al., 1990), in human (Gérard et al., 1991) and in mouse (Chakrabarti et al., 1995). Two splice variants of the human CB 1 receptors, named CB 1A and CB 1B respectively, were also isolated (Shire et al., 1995; Ryberg et al., 2005). The characterization and cloning of the other well known cannabinoid receptor, designated CB 2 receptors, were realized subsequently in human (Munro et al., 1993) and mouse (Shire et al., 1996).
As members of the GPCR super family, the actions of CB 1 and CB 2 receptors are transduced via the activation of G proteins (Piomelli 2003). CB 1 and CB 2 receptors contain seven transmembrane domains, connected by three intracellular and three extracellular loops, an intracellular C-terminus region and an extracellular N-terminus domain (Shim et al., 2011). CB 1 and CB 2 receptors share 44% amino acid identity and are encoded by different genes. In human, CB 1 receptors are found ubiquitously but are highly and preferentially localized in the brain and the spinal cord (Howlett, 2002).
At the peripheral level, CB 1 receptors are expressed on several cell types of the adrenal and thyroid glands. CB 1 receptors are also expressed in several cells related to metabolism, such as fat cells,
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 37 GENERAL INTRODUCTION muscle cells, liver cells and in the gastrointestinal tract. It is also expressed in the lungs and the kidney
(Pagotto et al., 2006). In contrast, CB 2 receptors are mainly expressed in cells of the immune system such as leukocytes. At a lower extent, CB 2 receptors are found on peripheral nerve terminals and current research suggests that these receptors play a role in nociception (Calignano et al., 1998; Klein et al., 2003). In the brain, they are mainly expressed by microglial cells (Walter and Stella, 2004).
Consequently, the original concept that CB 1 receptors play an exclusive role in the brain, and CB 2 receptors in the immune system, has evolved into the idea that both cannabinoid receptors can control both central and peripheral functions. Moreover, functional studies have suggested that the activation of CB 2 receptors by the administration of exogenous CB 2 receptor ligands may open novel therapeutic avenues for a number of brain pathological conditions such as neurodegenerative disorders, stroke, and neuropathic pain (Mackie, 2006).
Numerous pharmacological studies suggest the existence of additional cannabinoid receptors.
The transient receptor potential vanilloid type 1 (TRPV 1) ion channel, which interacts with eCB ligands such as AEA, was shown to be involved in some effects of cannabinoids including nociception (Starowicz et al., 2007) and anxiety (Marsch et al., 2007). Two other GPCRs, the G protein-coupled receptor 55 (GPR55) and the G protein-coupled receptor 119 (GPR119) were affiliated as novel potential cannabinoid receptors (Brown, 2007).
As previously mentioned the use of the marijuana-derived THC and of its synthetic analogues was instrumental for the discovery and characterization of the cannabinoid receptors. Among the synthetic CB 1 receptor agonists, some of them are widely used in experimental models. HU-210 is the most potent synthetic compound (Mechoulam et al., 1990). CP-55,940 (Mechoulam et al., 1990) and
WIN-55,212 (Compton et al., 1992) also exhibit a more potent and long lasting agonistic activity at CB1 site than THC. Among the synthetic ligands showing antagonistic properties at the cannabinoid receptors, SR 141716 (Rinaldi-Carmona et al., 1994), AM251 (Gatley et al., 1996) and AM281 (Lan et al., 1999) are specific for CB 1 receptors. Finally, although not acting as ligands of cannabinoid receptors, two interesting classes of compounds are able to inhibit the cellular uptake of eCBs, such as AM 404 (Beltramo, 1997) or AEA hydrolysis, such as URB532 and URB597 (Kathuria et al., 2003). More recently, the synthesis of two inhibitors of eCB degradation, such as JZL184 and JZL195 allowed new insights regarding the physiological functions of eCBs (Long et al., 2009ab). The advantage of these compounds is to interfere with the function of the ECS through a targeted increase in the concentration of eCBs, possibly avoiding some of the side effects known to be associated with unspecific cannabinoid receptor activation by direct agonists.
II.2.2 Distribution of CB 1 receptors in the brain
CB 1 receptors are among the most abundant GPCR in the rat and human CNS and their density in the brain rivals that of ionotropic receptors for glutamate and GABA, the main excitatory and inhibitory neurotransmitters, respectively (Herkenham et al., 1990).
Soon after its cloning, brain localizations of CB 1 mRNA by in situ hybridization were observed to mismatch with the areas of detection of CB 1 protein, suggesting that a majority of CB 1 receptors are
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 38 Part II - The endocannabinoid system in the brain
transported away from the soma (Matsuda et al., 1993). The pre-synaptic location of CB 1 receptors was confirmed using light microscopy in the rat hippocampus (Katona et al., 1999) and later supported by electrophysiological approaches (Hoffman and Lupica, 2000). However, a putative weak post- synaptic location of CB 1 receptors supported by anatomical studies or indirect evidences is still under debate (Freund et al., 2003; Bacci et al., 2004; Katona et al., 2006; Kawamura et al., 2006; Lafourcade et al., 2007; Marinelli et al., 2009).
Despite its wide expression in the CNS, the cellular distribution of CB 1 receptors seems to be restricted to particular cell types. CB 1 receptor expression is particularly dense in cortical regions such as the hippocampus, the cerebral cortex and the amygdala (Mailleux and Vanderhaeghen, 1992;
Marsicano and Lutz, 1999; Figure 1). Within these regions, CB 1 mRNA is detected at high level in subsets of local GABAergic interneurons co-expressing the glutamic acid decarboxylase (GAD65) and cholecystokinine (CCK). In contrast, both CB 1 mRNA and CB 1 protein are rarely observed in inhibitory interneurons expressing parvalbumin (Marsicano and Lutz, 1999). In spite of early anatomical studies pointing-out the presence of the CB 1 gene sequence in excitatory pyramidal neurons (Mailleux and Vanderhaeghen, 1992; Marsicano and Lutz, 1999) several years were then needed to recognize the presence of pre-synaptic CB 1 receptors on glutamatergic neuron terminals (Katona et al., 2000;
Freund et al., 2003), mostly due to the much lower CB 1 expression on glutamatergic cells as compared to GABAergic interneurons (Marsicano and Lutz, 1999; Marsicano et al., 2003). Recently, the genetic deletion of CB 1 receptor gene restricted to cortical glutamatergic neurons in mouse (see Part II.4.5), associated with anatomical and functional approaches allowed to clearly demonstrate the presence of CB 1 receptor protein on excitatory pyramidal neurons in the hippocampus, neocortex and amygdala (Domenici et al., 2006; Monory et al., 2006; Bellocchio et al., 2010).
Moreover, serotonin-releasing neurons originated from the mouse raphe nuclei also express
CB 1 mRNA whose protein is detected on the terminals projecting to the basolateral amygdala (BLA) and the hippocampal CA3 region (Häring et al., 2007). Oropeza et al. (2007) also provided immunostaining of CB 1 receptors on the noradrenergic terminals of the coeruleo-frontal fibers. The striatum exhibits a dense expression of CB 1 receptors, mainly in the medium spiny neurons (MSNs) of the dorsal striatum, co-expressing GAD65 mRNA but also the types I- and II-dopamine receptors (D1 and D 2 receptors, respectively; Marsicano and Lutz, 1999; Hermann et al., 2002; Monory et al., 2007).
CB 1 protein expressed in the MSNs is observed in their projection regions such as the substantia nigra pars reticulata (SNr; Tsou et al., 1998). CB 1 mRNA is detected at low levels in the thalamus. In contrast, its expression in the septum, the subthalamic nucleus and in the habenula is abundant (Marsicano and Lutz, 1999; Soria-Gómez et al., unpublished data).
The ECS has a major functional importance in the hypothalamic nuclei, including the regulation of energy metabolism, adaptation to stress and sexual behaviors (Pagotto, 2005; Cota et al., 2006; Hill et al., 2010). Accordingly, CB 1 receptors are enriched in the hypothalamus, at both genomic and protein levels (Marsicano and Lutz, 1999; Wittmann et al., 2007) Notably, they are expressed on glutamatergic neurons in the ventromedial nucleus of the hypothalamus (VMH), and co-localize with
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 39 GENERAL INTRODUCTION hormone-producing neurons, such as the corticotropin releasing factor (CRF) –positive neurons in the paraventricular nucleus (PVN; Cota et al., 2003).
CB 1 receptors are also present in several midbrain nuclei. In the ventral tegmental area (VTA),
CB 1-immunostaining has been detected but contradictory results exist concerning the nature of the neurons involved (i.e. glutamatergic, GABAergic and/or dopaminergic; Wenger et al., 2003; Mátyás et al., 2006). Importantly, CB 1 protein is abundant in the periaqueducal gray (PAG) where it has been implicated in the adaptation to aversive and noxious stimuli, likely by controlling both excitatory and inhibitory transmissions (Maione et al., 2006; Drew et al., 2008; Moreira et al., 2009).
Figure 1. Brain distribution of the mouse CB 1 receptor protein. Immunolabelling of the CB 1 receptor protein in parasagittal brain slices of wild-type (A) and CB 1-KO (D) mice, and in coronal sections of wild type mice (B, C and E). Note the high levels of CB 1 expression in the anterior olfactory nucleus (AON, A), neocortex (A-C), caudate putamen (CPu, A-C), hippocampus (Hi, A,C), thalamus (Th, A,C) basolateral (BLA) and central (Ce) amygdaloid nuclei (C), cerebellum (Cb, A) and spinal cord (E). CB 1 mRNA is absent in the CB 1-KO mouse brain (D). M1, primary motor cortex; S1, primary somatosensory cortex; V1, primary visual cortex; Cg, cingulate cortex; Ent, entorhinal cortex; DG, dentate gyrus; NAc, nucleus accumbens, GP, globus pallidus; VP, ventral pallidum; Mid, midbrain; SNR, substantia nigra pars reticulata; PO, pons; MO, medulla oblongata; EP, entopedoncular nucleus; VMH, ventromedial hypothalamus; DH, dorsal horn; DLF, dorsolateral funiculus. Scale bars: 1 mm (A-C, E), 200 µm (D) (from Kano et al., 2009).
In the cerebellum, CB 1 mRNA is fund in the granular glutamatergic cells and the protein labels the parallel fibers to the Purkinje cells as well as inhibitory interneurons controlling the activity of the
Purkinje cells (Marsicano and Lutz, 1999; Kawamura et al., 2006). Although CB 1 receptors are likely absent from the Purkinje cell surface, the endocannabinoid signalling tightly regulates their electrical activity (Takahashi and Linden, 2000).
In the light of the above studies, CB 1 receptors are expressed on terminals of specific neuronal subpopulations, suggesting the importance of CB 1 physiological activation on their activity within the brain. Although the presence of CB 1 receptors on other neuronal cell types, such as dopaminergic or serotoninergic neurons is still not fully accepted, recently-generated conditional mutant mice lacking
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 40 Part II - The endocannabinoid system in the brain
CB 1 receptors specifically on these cell types will provide new insight on the expression and functions of the “under-represented” CB 1-expressing neurons (Terzian et al., 2011; Bellocchio et al., unpublished data; see section II.4.5). In summary, CB 1 receptors are abundant in the brain, mainly localized at neuronal terminal membranes, and expressed on several types of neurons. This wide distribution explains the complexity of the mechanisms by which CB 1 receptors modulate brain functions. However, such a picture is still too simplistic.
Due to their high liposolubility, cannabinoids can easily cross the plasma membrane and thereby act intracellularly. Mitochondria are intracellular organelles regulating the energy metabolism of living cells. In presence of oxygen, mitochondria metabolize glucose to produce adenosine triphosphate (ATP), an energetic source directly used by cells. Mitochondria also participate in the apoptotic cascade, and mitochondria dysfunctions have been associated to many neurodegenerative diseases (Mattson et al., 2008). In the 1970’s, well before the CB 1 receptor cloning, some pharmacological assays of THC on subcellular functions pointed out that the drug alters the enzymatic activity of mitochondria (Chari-Bitron and Bino, 1971; Mahoney and Harris, 1972) and more recent work suggested a direct effect of THC on mitochondria (Whyte et al., 2010). Using electron microscopy approaches, Bénard and colleagues in our laboratory recently demonstrated the presence of CB 1 receptors on the mitochondrial membrane of hippocampal neurons (Bénard et al., 2012). That is, THC affects mitochondrial functions, such as energy production, by acting at mitochondrial CB 1
(mtCB 1) receptors. Strikingly, this study further indicates that mtCB 1 receptors participate in the cannabinoid-modulation of synaptic transmission, extending the potential mechanisms by which the ECS regulates neuronal functions (Bénard et al., 2012). This study suggests that GPCR can be expressed on mitochondrial membranes, thus profoundly challenging the classical conception of GPCR functioning.
Besides the neuronal expression of CB 1 receptors, several in vitro and ex vivo studies have indicated the expression of CB 1 receptors on brain glial cells. Although resting microglia shows few to any CB 1 and CB 2 receptor expression in brain tissue, the neuronal damage-induced expression of CB 2 receptors is well characterized in microglial cells (Walter and Stella, 2004; Stella, 2010).The presence of CB 1 receptors on astrocytes is better documented; I will provide an extended discussion of the known roles of the ECS in astroglial functions in the Part III.
II.3 The endocannabinoids and their synthesis, transport and degradation pathways
The endocannabinoids (eCBs) are defined as the endogenous ligands for cannabinoid receptors. They were discovered shortly after the CB 1 and CB 2 receptors cloning and are so-named because they were first identified as activating the same receptors as exogenous cannabinoids. This was an important breakthrough as eCBs gave a physiological significance to the cannabinoid receptors (Di Marzo et al., 1998). Structurally, most eCBs are lipids deriving from amides or esters of long chain fatty acids. The first eCB identified was arachidonoyl-ethanolamide, named anandamide (AEA) from the Sanskrit word for “bliss”, ananda (Devane et al., 1992). AEA was rapidly shown to
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 41 GENERAL INTRODUCTION mimic the cannabinoid “tetrad” effects (i.e. hypolocomotion, analgesia, catalepsy, hypothermia) in rodents (Crawley et al., 1993; Fride and Mechoulam, 1993). AEA acts as a partial agonist of both CB 1 and CB 2 receptors. However, AEA has also an agonistic activity at TRPV1 receptors, which led to propose it as an "endovanilloid" molecule (Di Marzo and Petrocellis, 2006). The second discovered eCB, 2-arachidonoyl-glycerol (2-AG) was also isolated from the brain (Mechoulam et al., 1995; Sugiura et al., 1995; Stella et al., 1997). The brain 2-AG levels are about 200-fold greater than AEA and evidence suggests that 2-AG is the main effectors of the CB 1 receptors-mediated regulation of synaptic transmission (Kim and Alger, 2004; Hashimotodani et al., 2007; see section II.4.2.1). Other chemically similar eCBs were identified during the last 10 years. Interestingly, recent data suggested that other endogenous molecules not directly related to "classical" eCBs can activate or inhibit cannabinoid receptors, such as hemopressin (Chen et al., 2008; Pamplona et al., 2010). Nevertheless, AEA and 2-AG have remained the only ones for which the pharmacological activity and metabolism have been most thoroughly investigated (Mackie, 2006; Alger and Kim, 2011). Therefore, these two compounds are still referred to as the major eCBs.
The synthesis, cellular transport and degradation of eCBs are tightly regulated processes. A feature that distinguishes eCBs from many other neuromodulators is that they are lipids and, therefore, they are not stored in the aqueous medium of synaptic vesicles. Rather, their precursors exist in cell membranes and are cleaved by specific enzymes. This form of synthesis / release is often referred to as “on demand” (i.e. spatially and temporally-restricted). Following physiological or pathological stimuli, several mechanisms underlie the production and/or release of eCBs. For instance, membrane depolarization of neurons can induce de novo formation and/or release of eCBs (Di Marzo et al., 1994; Stella and Piomelli, 2001). ECBs production and release is potentiated by pharmacological activation of the hippocampal post-synaptic group I metabotropic glutamate receptors (mGluR) and the muscarinic acetylcholine receptors (mAChR; Varma et al., 2001; Kim et al., 2002; Figure 3). Importantly, eCBs production and release can be reversed by chelating intracellular Ca 2+ concentration, thus characterizing the activity-dependent action of the ECS (Cadas et al., 1996; Piomelli, 2003; Figure 3).
The biosynthesis of AEA initiates by the formation of N-arachidonoyl- phosphatidylethanolamines (NAPE) from the catalysis of phosphatidylethanolamines by the N- acyltransferase (NAT) enzyme. NAPE are then hydrolyzed by a phospholipase D-like enzyme named NAPE-PLD (Di Marzo et al., 1994; 1998) resulting in the production of AEA (Figure 2;.Di Marzo et al., 1994, 1998) resulting in the production of AEA (Figure 2). Biochemical studies have revealed several pathways for 2-AG synthesis (Piomelli, 2003). 2-AG can be formed from arachidonic acid-containing membrane phospholipids such as phosphatidylinositol through the action of phospholipase C (PLC) leading to the formation of diacylglycerol (DAG), which in turn is catalyzed by a selective DAG lipase (DAGL or DGL). More recently, several enzymes for eCBs synthesis such as NAPE-PLD and DGL have been knocked-down in mouse, resulting in a deeper characterization of eCBs synthesis pathways and kinetics (Chanda et al., 2010; Alger and Kim, 2011).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 42 Part II - The endocannabinoid system in the brain
Phosphatidylethanolamines Phosphatidylinositol
NAT PLC N-Acetyltransferase Phospholipase C NAPE DAG N-arachidonoyl-phosphatidylethanolamines Diacylglycerol
NAPE-PLD DAGL NAPE-Phospholipase D DAG lipase AEA 2-AG 2-arachidonoylglycerol anandamide
Figure 2. Main biosynthesis pathways of AEA and 2-AG. Adapted from Piomelli, 2003
Classical neurotransmitters are usually inactivated by facilitated re-uptake from neurons and / or astrocytes and subsequently degraded. ECBs also need mechanisms for a rapid removal from their molecular targets and subsequent degradation. A major question mark in the activity cycle of eCBs is how they are transported away from the internal cell space to reach the trans-synaptic CB 1 receptors and then re-uptaken for breakdown processes. In other words, how these hydrophobic molecules diffuse through the intracellular or extracellular aqueous media?
No clear answer is provided to this question yet (Figure 3). A temperature-sensitive and Na +- independent "endocannabinoid membrane transporter" (EMT) protein was described in primary neuron cultures (Di Marzo et al., 1994; Ligresti et al., 2004), but the reason why lipophilic compounds such as eCBs need carriers to diffuse across the cell membrane is not clear. Inconsistent with this hypothesis, it was proposed that AEA can rapidly and efficiently cross the cell membrane by a concentration gradient-driving process (Glaser et al., 2003), or by a facilitated-diffusion process (Hillard et al., 1997) that may be induced by cholesterol binding (Di Pasquale et al., 2009). An alternative stipulated mechanism would be that AEA undergoes endocytosis by membrane invagination (McFarland and Barker, 2004). Several protein partners have been involved in AEA transport within the cytosol. The folding heat shock protein 70 (Hsp70) as well as albumin are able to bind AEA in vitro (Oddi et al., 2009). Particular isomers of the fatty acid binding proteins have been shown to potentiate the uptake and degradation rate of AEA in vitro (Kaczocha et al., 2009). More recently, a cytosolic truncated fatty acid amide hydrolase (FAAH, see below)-like protein complex, the FLAT, has been shown to participate in AEA transport (Fu et al., 2012). The mechanisms of 2-AG carrying are even less understood. This is an important issue because 2-AG is considered as the major eCB mediating the regulation of synaptic transmission by CB 1 receptors (see part II.4.2.1).
Once inside the cell, eCBs are hydrolyzed by two enzymes: the FAAH and the monoacyl- glycerol lipase (MAGL; Figure 3). FAAH catabolizes AEA into arachidonic acid and ethanolamine (Cravatt et al., 1996). FAAH would also participates in the degradation of 2-AG (Goparaju et al., 1998),
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 43 GENERAL INTRODUCTION but it is now clear that MAGL is the main enzyme degrading ~ 85 % of 2-AG in vivo (Dinh et al., 2004; Blankman et al., 2007; Chanda et al., 2010). The degradation products of 2-AG are arachidonic acid and glycerol (Piomelli, 2003). These metabolites are recycled into the membrane phospholipids in order to be used, at least in part, for de novo biosynthesis of the two eCBs (Bisogno et al., 2005). Recent studies pointed out the importance of the α-β hydrolase 6 and 12 (ABHD6 and ABHD12, respectively), members of the serine hydrolases family, in 2-AG degradation (Blankman et al., 2007). ABHD6 is located at pre-synaptic sites of neurons and its inhibition induces the accumulation of 2-AG at cannabinoid receptors and favors the induction of CB 1-dependent long-term depression (LTD; Marrs et al., 2010). Another pathway of eCBs degradation is their oxidation by the cyclooxygenase (COX) and lipooxygenase (LOX) enzymes by targeting their arachidonic acid group (P ăunescu et al., 2011).
Due to the on demand property of eCBs actions, an interesting strategy to potentiate CB 1 receptors signalling is to inhibit eCBs degradation. In that sense, the comprehension of eCBs breakdown mechanisms is crucial for developing tools facilitating ECS activity when beneficial.
Overall, the on demand post-synaptic release of eCB, together with the presence of CB 1 receptors on several cell types (and neuronal mitochondria) suggest that the ECS play an important role in modulating synaptic transmission in the nervous system.
Membrane phospholipids
MAGL ABHD6 Transporter- Facilitated mediated Passive diffusion diffusion diffusion 2-AG
FLAT Chol CB 1
2-AG AEA
R u R l h G c DAGL m A m
FAAH NAPE Ca 2+ -PLD PLC β Ca 2+
Membrane phospholipids
Figure 3. Cellular pathways for AEA and 2-AG production, transport and degradation 2-AG and AEA are synthesized from membrane phospholipids in an activity-dependent way (i.e. Postsynaptic intracellular Ca 2+ elevation). Their production can be increased by mGluR and mAchR stimulation. The black square represents the 3 putative membrane diffusion mechanisms for eCBs release into the synaptic cleft (Chol, cholesterol). AEA and 2-AG can be degraded in the post-synaptic element by the FAAH. However, 2-AG is mainly degraded by the MAGL but also by ABHD6 at the pre-synapse.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 44 Part II - The endocannabinoid system in the brain
II.4 Central CB 1 receptor signalling and modulation of synaptic functions
It is now clear that plastic events outlast the early stages of nervous system development and occur throughout adult life, providing, for instance, the basis of learning and memory (Milner et al., 1998; Kandel, 2001). Synapses are able to change their strength in response to specific synaptic or extra-synaptic events. Several underlying mechanisms cooperate to achieve synaptic plasticity, including changes in the quantity of neurotransmitters released at synapses and changes in how effectively cells respond to neurotransmitters. Plasticity can be categorized as short-term, lasting few seconds to few minutes, or long-term, which lasts from minutes to hours. Pre-synaptic events, such as action potentials, can increase the probability that synaptic terminals release transmitters (Kandel, 2001). Neural activity can also generate persistent forms of synaptic plasticity, such as long-term potentiation (LTP) and long-term depression (LTD). LTP is characterized by a long-lasting increase of synaptic efficacy whereas LTD corresponds to an enduring decrease of synaptic response (Citri and Malenka, 2008). Long-term plasticity involves intracellular cascades of events, such as proteins phosphorylation, that lead to modifications of genes expression. As a result, new proteins are synthesized, inducing neuronal plasticity phenomena including synaptic morphological changes, modification of receptors conformation, surface expression or trafficking. There is now plenty of evidence indicating that LTP and LTD are used to retain and update information in activated networks of neurons (Citri and Malenka, 2008).
As mentioned above, CB 1 receptors are considered to be likely the most abundant GPCR in the mammalian brain, and its presence throughout the brain accounts for most of the behavioral actions of cannabinoids. ECB signalling is a principal regulator of synaptic communication; its molecular and anatomical organization is a common feature of most synapses. Moreover, the great heterogeneity of
CB 1 receptor intracellular signalling pathways may also account for the various physiological consequences of CB 1 receptors stimulation.
II.4.1 CB 1-mediated intracellular signalling
The signal transduction cascades activated by CB 1 receptors are primarily mediated by their G protein coupling. This coupling was assessed by the ability of CB 1 receptor agonists to stimulate
[35S]GTP γS binding in presence of GDP excess. Potent CB 1 receptor agonists such as WIN 55212-2 were reported to increase this binding activity in various brain areas (Selley et al., 1996; Pertwee, 1997). Moreover, many cannabinoid effects are reversed by pertussis toxin treatments, revealing a
Gαi/o protein coupling (Howlett et al., 1986). G αi/o activation mainly leads to an inhibition of adenylyl cyclase enzymatic activity and subsequent diminution of cyclic adenosine monophosphate (cAMP) production (Figure 4) in brain regions showing the highest concentration of CB 1 receptors including the hippocampus, cerebral cortex, striatum and cerebellum (Bidaut-Russell et al., 1990; Matsuda et al., 1990).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 45 GENERAL INTRODUCTION
It should be noticed that an increase of cAMP accumulation has been reported in cultured striatal neurons as well as peripheral cells, this effect being attributed to a G αs protein coupling of CB 1 receptors and / or to distinct signalling outcomes of the adenylyl cyclase isoforms (Hampson et al.,
2000; Demuth and Molleman, 2006). A CB 1 receptors-Gαq coupling has been reported in response to WIN55,212-2, that was able to induce intracellular increase of Ca 2+ level through activation of phospholipase C (PLC) – inositol triphosphate (IP3) in particular cells (Lograno and Romano, 2004; Navarrete and Araque, 2008, see also section III.4.2).
Phosphorylation of the protein kinase A (PKA) is a major consequence of cAMP increases
(Brandon et al., 1997). Thus, CB 1 receptor-dependent PKA dephosphorylation reverses the effect of PKA activity, such as the inhibition of ion channels, including A-type K + channels (Figure 4). This results in a hyperpolarization of the neuronal membrane (Childers and Deadwyler, 1996; Mu et al.,
2000). Indeed, several neuronal responses induced by CB 1 receptors activation, including inhibition of synaptic transmission and modulation of neuronal plasticity, depends on the PKA phosphorylation pathways (Chevaleyre et al., 2007; Azad et al., 2008).
Activation of CB 1 receptors is also able to up-regulate the mitogen-activated protein kinases (MAPK) pathway, c-Jun N-terminal kinases (JNK) and phosphatidylinositol 3-kinases (PI3K) both in vitro and in vivo (Bouaboula et al., 1995b; Valjent et al., 2001; Rubino et al., 2004; André and Gonthier, 2010). MAPK, including the extracellular signal-regulated kinases 1 and 2 (ERK1/2), are essential for cell development, differentiation and apoptosis, and are known to cross the nuclear membrane to regulate gene transcription. In neurons, they also play key roles in shaping synaptic terminals, promoting spine formation and their signalling has been strongly associated with synaptic plasticity and memory formation (Wiegert and Bading, 2011). The exact mechanisms by which CB 1 receptor stimulation activates these kinases are not clear and futures studies will have to address this question, since the ERK1/2 pathway is considered as a major intracellular process by which CB 1 receptors might modulate synaptic plasticity (Miyamoto, 2006).
A faster way for cannabinoids to hyperpolarize cell membranes through CB 1 receptors is the modulation of voltage-dependent ion channels directly through the G αi/o and G β/γ protein subunits 2+ (Figure 4). Indeed, CB 1 receptor activation leads to an inhibition of the N- L- or P/Q-type Ca + channels, inhibition of Na channels (Nicholson et al., 2003) and activation of inwardly rectifying K ir channels (Mackie et al., 1993). Crucially, the modulation of voltage-dependent ion channels by CB 1 receptors results in a decrease in intracellular Ca2+ concentration and subsequent reduction of excitatory and inhibitory synaptic transmission (Demuth and Molleman, 2006). Interestingly, 2+ cannabinoid-induced inhibition of Ca and activation of K ir channels, respectively, is reversed by PKC activity (Garcia et al., 1998), suggesting the existence of an auto-regulatory control of CB 1 receptors signalling.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 46 Part II - The endocannabinoid system in the brain
Ca 2+ Extracellular
CB 1
Intracellular Gαi/o + AC K+ K ir Gγ Gβ A
ATP cAMP PKA MAPK JNK PKB
Gene expression
Nucleus
Figure 4. Main intracellular CB 1 receptor signalling pathways Stimulation of CB 1 receptors lead to 3 main cascades of events. A direct modulation of ion channels conductance 2+ + + + including an inhibition of Ca channel and activation of K ir channels. Activation of A-type K channels (K A) can be induced through the inhibition of adenylyl cyclase (AC). Phosphorylation of several protein kinases including MAPK, JNK and PKB leads to de novo gene expression. Adapted from Pagotto et al., 2006.
Other mechanisms ensure the adaptation of CB 1 receptor signalling. Sustained or persistent agonist occupancy can lead to an attenuation of CB 1 receptors efficacy to subsequent activations, a process referred as desensitization. Agonist-induced CB 1 receptors desensitization has been observed in the brain of both human chronic cannabis consumers and in rodent chronically treated with potent CB 1 receptors agonists (Sim-Selley, 2003; Villares, 2007). These effects are particularly robust within the hippocampus and are associated with G-protein activity as assessed with 35 [ S]GTP γS binding in brain tissue (Sim-Selley, 2003). CB 1 receptors desensitization has been associated with agonist-stimulated CB 1 internalization processes (endocytosis) and through plasma membrane trafficking (Hsieh et al., 1999; Bari et al., 2005). Several GPCR-interacting proteins are involved in CB 1 receptor desensitization, including kinases and β-arrestins (Pitcher et al., 1998).
Desensitization of CB 1 receptors efficacy has important physiological consequences. For instance, prolonged treatment of hippocampal neurons with WIN55212-2 attenuated the inhibition of glutamatergic transmission induced by acute CB 1 receptors activation in vitro (Kouznetsova et al., 2002).
Recently, the cannabinoid receptor-interacting protein 1 (CRIP 1), specific to CB 1 receptors, have 2+ been cloned and shown to attenuate the Ca channels inhibition induced by CB 1 receptor activity
(Niehaus et al., 2007). CRIP 1 binds to the C-terminus domain of CB 1 proteins, which also contains a PDZ ligand sequence, possibly interacting with others PDZ-containing proteins in the synaptic density.
Although little is known about the functional role of CRIP 1, some studies revealed that its expression is
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 47 GENERAL INTRODUCTION modulated in hippocampus of epileptic patients as well as after kainic acid-induced seizures in rat (Ludányi et al., 2008; Bojnik et al., 2012).
Lastly, another mean to potentially modulate the responsiveness of CB 1 receptors is the formation of multimers (Vischer et al., 2011). Virtually all the GPCR can heterodimerize, both between receptors from the same class and between members of different classes. GPCR heterodimerization is thought to increase the binding possibilities with ligands that are not able to interact with the individual receptor, and also to regulate receptor signalling cascades. Indeed, GPCR heterodimerization likely expands the physiological repertoire of receptor responses to single molecules (Mackie, 2005). Several studies reported that CB 1 receptors can form both homodimers and heteromultimers in the brain (Katona et al., 2001; Mackie, 2005). Notably, CB 1 receptors are able to dimerize with dopamine D 2 receptors, opioid receptors, adenosine A 2 receptors and orexin-1 receptors (Smith et al., 2010). Again, the functional significance of these complexes formation is poorly understood, even though CB 1 receptor heterodimerization seems to decrease cannabinoid effects (Smith et al., 2010; Rozenfeld et al., 2012).
II.4.2 CB 1 receptor-mediated modulation of synaptic transmission and plasticity
As described above, eCBs are released on demand, after cellular depolarization or receptor stimulation in a calcium-dependent manner. Because of the activation of K + currents and the inhibition 2+ of Ca entry into cells, the net effect of CB 1 receptor stimulation is a local membrane hyperpolarization that leads to the general inhibition of neurotransmitter release including glutamate, GABA, glycine, acetylcholine, norepinephrine, dopamine, serotonin and CCK (Kano et al., 2009). Due to the widespread distribution of CB 1 receptors and their expression on many brain cell subtypes, the impact of CB 1-mediated pre-synaptic inhibition of transmitter release on cell communication is of major importance.
II.4.2.1 Depolarization-induced suppression of neurotransmitter release
The most studied form of endocannabinoid-dependent synaptic plasticity is the depolarization- induced suppression of neurotransmitter release. A brief depolarization (up to few seconds) of the post-synaptic neurons, for instance the pyramidal neurons of the hippocampus, induces a transient suppression (< 90 s) of either GABAergic (depolarization-induced suppression of inhibition, DSI) or glutamatergic synaptic inputs (depolarization-induced suppression of excitation, DSE), as measured by the amplitudes of evoked inhibitory post-synaptic currents (eIPSC) and evoked excitatory post- synaptic currents (eEPSC), respectively (Pitler and Alger, 1992; Kreitzer and Regehr, 2001). From the physiological point of view, DSI and DSE are characterized by 2 important properties:
(i) An increase of Ca 2+ influx in the post-synaptic neuron. Both DSI and DSE are prevented by Ca 2+ chelators, such as BAPTA and EGTA, when applied post-synaptically (Pitler and Alger 1992; Kreitzer and Regehr, 2001).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 48 Part II - The endocannabinoid system in the brain
(ii) The decrease of post-synaptic currents is a consequence of pre-synaptic transmission changes. The post-synaptic responses of GABA A receptors induced by locally-applied GABA are not altered by DSI induction (Pitler and Alger 1992). DSI is also observed when the amplitude of miniature inhibitory post-synaptic currents (mIPSC) is constant, that is interpreted as a constant response to single vesicle release, another indicator of unchanged post-synaptic responses. However, the frequency of mIPSC decreases during DSI and DSE, that reflects a decrease in the number of vesicles released during a synaptic event (Pitler and Alger 1992). Induction of both DSE and DSI is concurrent with modifications of the paired-pulse ratio (PPR) that corresponds to the increase of amplitude of a PSC evoked by a second impulse as compared to the first PSC (Pitler and Alger 1992; Kreitzer and Regehr, 2001). This change is associated to a decrease of vesicular release probability (McNaughton, 1982).
Therefore, the post-synaptic elevation of Ca 2+ associated with pre-synaptic changes is typical of a retrograde messenger signalling (i.e. released post-synaptically and acting at the pre-synaptic terminal). Wilson and Nicolls in 2001 first reported that DSI is mediated by eCBs retrogradely acting at pre-synaptic CB 1 receptors in the hippocampus. Indeed, DSI was blocked in presence of CB 1 receptor antagonists and in CB 1 knock-out ( CB 1-KO) mice (Wilson and Nicoll, 2001; Wilson et al., 2001). Furthermore, application of both exogenous and endogenous cannabinoids (2-AG) suppressed eIPSC and was able to occlude DSI. The CB 1-triggering DSE was indentified in the cerebellum within the same year (Kreitzer and Regehr, 2001). Since then, these two forms of modulation of synaptic transmission by CB 1 receptors have been described in many brain structures including the hippocampus, cerebellum, neocortex, striatum, in both central and basolateral amygdala, ventral tegmental area (VTA), hypothalamus and the brain stem nuclei (Chevaleyre et al., 2007; Kano et al., 2009; Kamprath et al., 2011).
CB 1 receptors are thought to be mainly expressed in CCK-positive inhibitory interneurons, but not in parvalbumin expressing ones (Marsicano and Lutz, 1999). In the hippocampus, DSI can be induced only in cannabinoid-sensitive neuronal subpopulations (Martin et al., 2001). On the other hand, DSI is observed at CCK-positive terminals in the CA1 (Ali, 2007), suggesting that endocannabinoid-driven DSI is restricted to neurons expressing CCK. Beside this anatomical specificity, CCK has been proposed to play a role in DSI, as its bath application in hippocampal slices inhibits IPSC in the CA1 in a CB 1 receptor-dependent manner, suggesting that eCBs could induce DSI by favoring CCK release (Földy et al., 2006).
Another crucial question is which eCB(s) induce(s) DSI / DSE through CB 1 receptors? In their seminal study, Wilson and Nicolls in 2001 reported that DSI can be occluded by exogenous applications of 2-AG. In contrast, preventing AEA degradation by selective inhibition of FAAH, either by URB597 or in FAAH knock-out mice, does affect neither DSE nor DSI, suggesting that AEA is not involved in DSI / DSE (Kim and Alger, 2004; Pan et al., 2009). Increasing 2-AG availability by suppressing its degradation in MAGL knock-out mice prolonged DSI in hippocampal CA1 pyramidal neurons (Pan et al., 2011). Similarly, the selective pharmacological blockade of MAGL using JZL184 prolongs DSE in Purkinje neurons in cerebellar slices and DSI in CA1 pyramidal neurons in
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 49 GENERAL INTRODUCTION hippocampal slices (Pan et al., 2009), suggesting an important role of 2-AG in DSI / DSE. However, more controversial results have been observed after suppression of 2-AG synthesis by DAGL inhibition. While pharmacological inhibition of DAGL do not block neither DSI nor DSE (Chevaleyre and Castillo, 2003; Safo and Regehr, 2005; Min et al., 2010), genetic suppression of the DAGL α isoform prevented both DSI and DSE (Chanda et al., 2010; Tanimura et al., 2010). Several hypotheses have been proposed to explain this discrepancy (Min et al., 2010). In particular, the existence of intracellular pre-existing pools of 2-AG could account for the rapid retrograde suppression of transmission observed in DSI/DSE. Indeed, already synthesized 2-AG would be released following Ca 2+ elevation, and this pathway could not be prevented by DAGL inhibition. However, only chronic DAGL suppression, as in genetically modified mice, might suppress the formation of these pools, and thus impedes DSE / DSI induction (Min et al., 2010). This is an important postulate that needs to be experimentally assessed because it challenges the “on demand” nature of eCB synthesis and / or release.
Beside the common general features of DSI and DSE, distinctions exist, mainly in regard with their expression. Indeed, DSE is more difficult to induce than DSI, needs stronger depolarization parameters to be observed and results in weaker depression of glutamatergic transmission (Ohno-
Shosaku et al., 2002b). One possible reason is that CB 1 receptors are much less expressed at glutamatergic terminals as compared to GABAergic terminals. Alternatively, the sensitivity of CB 1 receptors to exogenous and/or endogenous cannabinoids might differs in respect to their neuronal expression site, so that a sustained eCB release is required to shut-down glutamatergic transmission.
II.4.2.2 Others forms of CB 1-dependent regulation of synaptic transmission
Pre-synaptic inhibition of both excitatory and inhibitory neurotransmitter release can be mediated by activation of post-synaptic group I metabotropic glutamate receptors (mGluR1 and mGluR5) and has been observed in the hippocampus, the cerebellum, the dorsal striatum and the VTA (Maejima et al., 2001; Varma et al., 2001; Brown et al., 2003; Melis et al., 2004; Galante and Diana, 2004; Uchigashima et al., 2007). Importantly, group I mGluR-dependent modulation of synaptic transmission does not require post-synaptic Ca 2+ elevation because it is not prevented by BAPTA applied at post-synaptic neuron, contrary to DSI and DSE (Maejima et al., 2001). However, it is blocked by CB 1 receptor antagonists and it is absent in CB 1-KO mice, indicating the involvement of endocannabinoid signalling. In the hippocampus, the group I mGluR-dependent modulation of excitation and inhibition has been proposed to depend on mGluR5 as it was blocked by MPEP, a specific mGluR5 antagonist (Ohno-Shosaku et al., 2002a; Straiker and Mackie, 2007). Importantly, group I mGluR activation triggers depression of IPSC and eCB production in the hippocampus that is prevented by suppressing PLC activity in PLC β1 knock-out mice. However, DSI was intact in these animals, emphasizing that mGluR-mediated suppression of inhibition through CB 1 receptors and DSI are mediated by distinct mechanisms in the hippocampus (Hashimotodani et al., 2005).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 50 Part II - The endocannabinoid system in the brain
Similarly, pharmacological activation of the post-synaptic metabotropic muscarinic acetylcholine receptors (mAchR) is able to transiently inhibit IPSC in hippocampal slices in a CB 1-dependent manner as this effect was blocked in CB 1-KO mice (Kim et al., 2002). In particular, the M 1 and M 3 mAchR subtypes mediate this regulatory process that is not dependent of post-synaptic Ca 2+ elevation (Kim et al., 2002; Fukudome et al., 2004). The mGluR and mAChR-mediated effects are proposed to be initiated independently, because antagonists of one receptor do not affect the responses mediated by the other. Even though the mAchR-mediated suppression of inhibition is G-protein dependent (Kim et al., 2002), the intracellular mechanisms by which mAchR activation might increase eCB release are not known.
CB 1 receptors can control synaptic transmission by many other pathways that can vary among different brain nuclei. For instance, in the hippocampus, glutamate release induced by stimulation of the Schaffer collaterals can hetero-synaptically depress GABA release through the activation of pre- synaptic kainate receptors on inhibitory interneurons. Lourenço et al. (2010) in our laboratory showed that this inhibition of GABA release by kainate receptors is conditioned by the presence of CB 1 receptors on inhibitory interneurons, and requires a retrograde release of 2-AG. This new property of the ECS might help to sense and adapt to changes in the overall network activity (Lourenço et al., 2010).
In the supra-optic nucleus of the hypothalamus, oxytocin can reduce EPSC in magnocellular neurons and IPSC from oxytocin-producing cells, both effects being proposed to depend on pre- synaptic CB 1 receptor activity (Hirasawa et al., 2004; Oliet et al., 2007). In addition, CB 1 receptors were shown to mediate glucocorticoid-induced suppression of excitatory transmission both in the paraventricular and the supra-optic nuclei of the hypothalamus (Di et al., 2005, 2003).
II.4.3 How does CB 1 receptor signalling modulate long-term synaptic plasticity?
Beside the transient suppression of neurotransmission by CB 1 stimulation, ECS activity has also a long-term impact on synaptic plasticity. Noteworthy, the stimulation of CB 1 receptors can induce particular forms of LTD known as eCB-LTDs (Chevaleyre et al., 2006). Unlike eCB-induced short-term depression, eCB-LTDs are thought to be induced by a sustained activation of pre-synaptic CB 1 receptors and result in a long-lasting decrease of neurotransmitter release. eCB-LTDs maintenance does not require continuous CB 1 receptor stimulation (Chevaleyre et al., 2006). Unlike the most studied forms of long-term plasticity, none of the previously described eCB-LTDs depends on post- synaptic NMDA receptor signalling but instead requires post-synaptic activation of the group I mGluR and / or post-synaptic Ca 2+ elevation. eCB-LTDs are inducible in several brain regions including the hippocampus, amygdala, striatum, and the neocortex but differ in their establishment protocols and underlying mechanisms (Chevaleyre et al., 2006).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 51 GENERAL INTRODUCTION
II.4.3.1 Heterosynaptic eCB-LTD in the hippocampus and amygdala
eCB-LTD of IPSC was first described in the amygdala (termed as LTDi; Marsicano et al., 2002) and then in the hippocampus (termed as I-LTD; Chevaleyre and Castillo, 2003; Figure 5). In the presence of ionotropic glutamate receptor antagonists, low frequency stimulation (1 Hz; LFS) of lateral amygdala afferent fibers and moderate to high frequency stimulation (10-100 Hz; HFS) of excitatory pyramidal neurons in the hippocampus induce a long-lasting (at least 30-40 min) depression of inhibitory inputs in amygdala and hippocampus, respectively (Figure 5). Both phenomena are blocked by pre-incubation of CB 1 receptor antagonists and absent in CB 1-KO mice (Marsicano et al., 2002; Chevaleyre and Castillo, 2003). Both in the amygdala and the hippocampus, LTDi / I-LTD appear to be expressed pre-synaptically as they are associated with modifications of paired-pulse ratios, and are independent of post-synaptic Ca 2+ elevation (Marsicano et al., 2002; Chevaleyre and Castillo, 2003; Azad et al., 2004). I-LTD in the hippocampus depends on post-synaptic activation of group I mGluR, which is followed by activation of PLC and DAGL (Figure 5). Thus, 2-AG would mediate I-LTD in the CA1 portion of the hippocampus (Chevaleyre and Castillo, 2003). I-LTD is also conditioned by the initial spiking activity of the CB 1-containing GABAergic interneurons, as preventing their firing inhibits I- LTD induction (Heifets et al., 2008). LTDi in the basolateral amygdala depends on post-synaptic release of eCBs, induced by the post-synaptic activation of mGluR1, and of the adenylyl cyclase-PKA pathway, but does not require PLC-DAGL activation (Azad et al., 2004). Moreover, LTDi is facilitated in FAAH-KO mice, suggesting that LTDi is mediated by AEA (Azad et al., 2004; Figure 5).
Hippocampus Amygdala I-LTD LTDi Glutamate HFS LFS
GABA GABA
-I m GluR Glu m R-I C B 1 B C 1 AC 2-AG PLC AEA
PKA DGL
Figure 5. Mechanisms of endocannabinoid-mediated long-term depression in the hippocampus and the amygdala Adapted from (Chevaleyre et al., 2006)
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 52 Part II - The endocannabinoid system in the brain
II.4.3.2 Homosynaptic eCB-LTD in the striatum
In 2002, two major studies first uncovered the importance of CB 1 receptors signalling in long- term depression of excitatory prelimbic cortical inputs to the nucleus accumbens (Nac) (Robbe et al., 2002) and excitatory cortical afferents to medium spiny neurons of the dorsal striatum (Gerdeman et al., 2002; Figure 6).
In the Nac, mild frequency (13 Hz; MFS), long lasting (10 min) stimulation of cortical afferences evoked LTD of EPSC. LTD induction was blocked both in CB 1-KO mice and following administration of
CB 1 receptor antagonists. However, pharmacological blockade of CB 1 receptors after LTD induction did not prevent the maintenance of EPSC inhibition, indicating that continuous CB 1 receptor activation is not necessary for the late phase of eCB-LTD (Robbe et al., 2002). The authors then showed that this eCB-LTD was independent of both NMDA and dopamine D 1 and D 2 receptors, but instead required post-synaptic Ca 2+ elevation and mGluR5 receptors intact signalling (Figure 6). Moreover, similarly to LTDi in the amygdala, eCB-LTD at these synapses was shown to be dependent on the cAMP-PKA pathway (Mato et al., 2008). Interestingly, eCB-LTD at cortico-Nac excitatory synapses is restricted to medium spiny neurons expressing D 2 receptors and also requires activation of TRPV1 receptors (Grueter et al., 2010). Because eCB-LTD in the Nac was enhanced in FAAH-KO mice, it was suggested that AEA primarily mediated LTD at these synapses (Grueter et al., 2010).
At cortical excitatory projections to medium spiny neurons of the dorsal striatum, eCB-LTD can be induced by a stronger stimulation protocol (100 Hz; HFS; Gerdeman et al., 2002). This eCB-LTD is 2+ also dependent on post-synaptic Ca rise, of group-I mGluR and D 2 receptors endogenous activity (Gerdeman et al., 2002; Kreitzer and Malenka, 2005; Figure 6). Importantly, cortico-striatal eCB-LTD requires pre-synaptic activity coincident with CB 1 receptor activation (Singla et al., 2007) but it is transiently impaired after a single systemic administration of THC (3 mg/kg) in mice, suggesting that eCB plasticity is a tightly regulated process (Mato et al., 2004).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 53 GENERAL INTRODUCTION
Dorsal striatum Nac Glutamate HFS MFS
C B 1 B C 1
mG l R u5 T -I D2 R R luR P mG V 1
AEA eCB Ca 2+ ? AC
PKA
Figure 6. Mechanisms of endocannabinoid-dependent long-term depression in the striatum Adapted from (Chevaleyre et al., 2006)
II.4.3.3 eCB-LTD in the neocortex
Sjöström et al. (2003) showed that retrograde eCB signalling is also at the basis of spike timing- dependent LTD (tLTD; Figure 7) of glutamatergic currents in pyramidal neurons of the neocortex (Sjöström et al., 2003). tLTD can be induced by mild frequency paired stimulations of the pre- and post-synaptic elements, and observed when the post-synaptic firing precedes the pre-synaptic firing (Chevaleyre et al., 2006). Strong evidence indicates that tLTD is mediated by a reduction of pre- synaptic release of glutamate (Sjöström et al., 2003) and requires post-synaptic activity (Sjöström et al., 2001). tLTD induction needs CB 1 receptor endogenous signalling and, under certain conditions, application of CB 1 agonists can substitute post-synaptic firing provoking tLTD (Sjöström et al., 2003), indicating that eCBs release might rhythm this form of synaptic plasticity in the cerebral cortex. tLTD is prevented by post-synaptic application of BAPTA, indicative of a post-synaptic Ca 2+ -dependent mechanism (Figure 7). tLTD also requires NMDAR activity at the pre-synaptic site (Sjöström et al.,
2003). Moreover, CB 1 receptor-dependent tLTD has been recently described at cortico-striatal synapses (Fino et al., 2010).
Recently, Lafourcade et al. (2007) described a similar homosynaptic eCB-LTD at pyramidal neurons of the prelimbic cortex as previously observed in the Nac. Finally, heterosynaptic I-LTD was observed within the prefrontal cortex and mediated by CB 1 receptors and pre-synaptic D 2 receptors that co-localize within this region (Chiu et al., 2010).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 54 Part II - The endocannabinoid system in the brain
tLTD Glutamate MFS 2
? NM CB 1 DA
MFS 1
eCB Ca 2+ ?
Figure 7. Mechanisms of endocannabinoid-dependent time-spiking long-term depression in the neocortex Adapted from (Chevaleyre et al., 2006)
II.4.3.4 Autocrine modulation of neuronal activity by CB 1 receptors
GABAergic interneurons are a heterogeneous cell population that forms complex functional networks. They differ according to their anatomy, electrophysiological properties and biochemical signature (Bacci et al., 2005). GABAergic interneurons activity modulate the activity of other cells (extrinsic modulation), but they can also regulate their own excitability (intrinsic modulation) through activity-dependent mechanisms known as autaptic transmission (Van der Loos and Glaser, 1972). For instance, in the neocortex, two distinct populations of self-inhibitory GABAergic interneurons were described. The major class of GABAergic interneurons called multipolar basket cells and chandelier cells are parvalbumin-positive and CCK-negative cells. They are characterized by autaptic synaptic contacts and generate fast and powerful self-inhibition when depolarized (Van der Loos and Glaser, 1972; Bacci et al., 2003). Another class of CCK-positive GABAergic interneurons has been characterized in the neocortex as generating low frequency firing rate but not showing GABAergic autaptic innervations (Bacci et al., 2003). However, following multiple trains of evoked action potentials, these interneurons develop a strong and long-lasting hyperpolarization. This self-induced inhibition is mediated by Ca 2+ elevation, blocked by AM251 and can be induced by exogenous application of 2-AG, indicating that eCBs acting at CB 1 receptors produce this self-induced inhibition (Bacci et al., 2004). Moreover, the pharmacological blockade of PLC and DAGL activity prevented this self-induced inhibition, indicating that this phenomenon is mediated by endogenous release of 2-AG
(Marinelli et al., 2008). Remarkably, a subset of neocortical pyramidal neurons shows similar CB 1 receptor-dependent self-inhibition properties (Marinelli et al., 2009). These neurons could evoke DSE more commonly than non self-inhibitory pyramidal neurons and showed long-lasting depression of
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 55 GENERAL INTRODUCTION inhibition after repeated depolarization (d-LTDi). These functional properties would provide a fine tuning of neocortical network activity (Bacci et al., 2004; Marinelli et al., 2008; 2009).
In conclusion, eCB-mediated modulation of synaptic plasticity is one of the most described phenomena of brain plasticity (Malenka and Bear, 2004; Citri and Malenka, 2008). Due to the high expression level of CB 1 receptors within the CNS, eCB-dependent changes in neuronal communication and excitability are considered as the most widespread forms of pre-synaptic plasticity
(Chevaleyre et al., 2006). Great efforts are being made to describe the mechanisms of CB 1 receptor- mediated regulation of neuronal transmission and plasticity at local, synaptic level. However, the neural bases of any function do not rely on plastic changes of one single cell (Neves et al., 2008). Therefore, analyzing network activities is important to understand how local cellular mechanisms regulate integrated physiological functions and behaviors.
II.4.4 CB 1 receptors in the control of brain networks activity and related behavioral consequences
A common function of both short and persistent CB 1-dependent depression of neuronal transmission would be to facilitate the induction of synaptic plasticity. In this sense, Chevaleyre and Castillo (2004) proposed that the ECS regulates brain metaplasticity. For instance, the stimulation protocol inducing LTDi at BLA GABAergic neurons is also able to enhance subsequently-induced LTP in a CB 1 receptor-dependent manner (Marsicano et al., 2002; Azad et al., 2004). Because CB 1 receptor blockade delays extinction of conditioned freezing (see section IV.3.3.1), this disinhibitory effect of CB 1 receptor signalling on excitatory transmission has been proposed to allow adaptation of conditioned fear responses (Marsicano et al., 2002).
In the hippocampus, both DSI and I-LTD of inhibitory transmission facilitate the induction of LTP of excitatory inputs (Chevaleyre and Castillo, 2004). However, a good illustration of the different functional impacts of the endogenous CB 1 receptors signalling as compared to its exogenous stimulation is that over-activation of CB 1 receptors by exogenous cannabinoids reduces both LTP and LTD efficacy in the hippocampus by reducing pre-synaptic transmission (Collins et al., 1995; Misner and Sullivan, 1999). In the same way, cannabinoid agonists have been shown to favor LTD at the expense of LTP after a tetanic stimulation of excitatory pyramidal neurons of the rodent’s prefrontal cortex (Auclair et al., 2000). The effects of CB 1 receptor agonists on neuronal plasticity have been associated with the well-known memory impairments induced by cannabinoid administration in human and animals (Misner and Sullivan, 1999; Abush and Akirav, 2010; see section IV.2). Memory processes are strongly associated with experience-dependent modifications in the temporal coordination of specific neural ensembles, which would in turn favor plasticity at synaptic levels (Dragoi et al., 2003; Girardeau et al., 2009). Few studies aimed at uncovering how cannabinoids impact on brain network activities during memory processing. The hippocampal neuronal network displays oscillatory rhythms characterized by different frequencies. Robbe et al., (2006) recorded these hippocampal oscillations in rats performing a hippocampus-dependent spatial memory task. Both systemic and unilateral intra-hippocampus administration of exogenous cannabinoids impaired
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 56 Part II - The endocannabinoid system in the brain memory abilities. Interestingly, this impairment was associated with a de-synchronization of oscillatory rhythms but not with changes in the intrinsic firing rate of individual neurons, suggesting that cannabinoid agonists impair spatial memory by altering the coordination of neuronal networks (Robbe et al., 2006; Robbe and Buzsáki, 2009).
II.4.5 Using knock-out mouse models to study CB 1 receptor functions
Until recently, the available tools to study CB 1 receptor functions where mostly based on (i) natural or synthetic pharmacological compounds, agonists and antagonists of different specificity / potency for CB 1 receptors (Yao and Mackie, 2009) and (ii) null CB 1 mutant lines, characterized by an ubiquitous deletion of CB 1 receptors (Ledent et al., 1999; Zimmer et al., 1999; Marsicano et al., 2002).
As previously mentioned, CB 1 receptors are expressed on many types of cells. Therefore, the consequences of systemic or even local manipulations of CB 1 receptor activity reflect a “net” effect of
CB 1 signalling onto these various target cells. A good example is the inhibitory control exerted by CB 1 receptor stimulation on both GABA and glutamate release, the main inhibitory and excitatory neurotransmission systems in the brain, respectively; indeed, what is the functional outcome of dampening both inhibitory and excitatory tone? Because CB 1 receptors are expressed in great majority at GABAergic terminals, the phenotypes reported following their experimental manipulation were often assigned to a direct regulation of GABAergic neurotransmission (Katona et al., 2000, 2001; Freund et al., 2003). A gene alteration may exert its effects in multiple different cell and tissue types, creating a complex phenotype in which it is difficult to distinguish its direct functions in particular cells. Therefore, methods have been developed to control conditions such as the timing and / or the cell- type specificity of gene activation or repression.
The Cre/LoxP system is an approach for generating tissue-specific knockout mice (Sauer and Henderson, 1989; Figure 8). Two different genetically engineered mouse lines are needed to achieve a tissue-specific gene deletion. In most cases, Cre- and loxP-containing strains of mice are developed independently and crossed to generate offspring bearing the tissue-specific gene knock-out. The first mouse strain contains the targeted gene sequence flanked by two loxP (“locus of crossover P1”) sites (“floxed” gene). The loxP sites are 34-base pair DNA sequences placed on each side of a target gene sequence by homologous recombination in mice (Sauer and Henderson, 1989). The small dimensions of the loxP sites guarantee that the mutation does not alter the expression of the floxed gene. Thus, floxed mouse lines can be considered as phenotypically wild-type. The Cre recombinase is a member of the integrase family of recombinases which catalyzes recombination between two loxP sites properly oriented. Indeed, the Cre excises the DNA segment between two loxP sites, resulting in a single remaining loxP site (Sauer and Henderson, 1989). Once the normal expression of the gene is verified, the floxed mouse line can thus be crossed to any transgenic mouse line expressing the Cre recombinase under the control of a promoter that is specific for a particular cell type (referred to as Cre line). Offspring of this crossing will express both the floxed gene and the Cre-expressing transgene. In the cells where the Cre recombinase is expressed, the DNA segment flanked by the
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 57 GENERAL INTRODUCTION loxP sites will be excised (Figure 8), and consequently inactivated. The floxed gene remains active in the cells that do not express the Cre (Sauer and Henderson, 1989).
By this genetic system, several cell-type specific CB 1-KO mouse lines were generated (Table
1). It is important to note that constitutive as well as the available conditional CB 1-KO mice are viable, fertile, develop normally and do not have severe deficits that can distinguish them from the control wild-type littermates.
The Cre/loxP system applied on CB 1 receptors
CB 1 pr Cre CB 1 X
Target cell type pr Specific promotor Other cell types Cre Cre recombinase
CB 1 CB 1 gene CB 1 loxP sites
Cell type-specific CB 1-KO
Figure 8. Cell type-specific deletion of the CB 1 gene by the Cre/loxP system strategy in mice The generation of cell-type specific CB 1-KO mouse lines reposes on the crossing of two different lines. The CB 1- flox line (upper left) is characterized by a flanked CB 1 sequence by two loxP sites in all the cells where CB 1 is normally expressed. The cre line (upper right) is characterized by the cre expression under the control of a promoter specific to the targeted cell type. In the offspring of this crossing (lower), the cre will excise CB 1 gene in the targeted cells, leaving the normal expression of the receptor in the remaining cells of the body.
For instance, the CB 1-floxed mice (Marsicano et al., 2002) were crossed with mice that express Cre recombinase under the control of the regulatory sequences of the Calcium/calmodulin-dependent kinase II α gene ( CaMKII α-Cre mice) that is specifically expressed in forebrain principal neurons
(Casanova et al., 2001), to obtained the CaMKII α-CB 1-KO mice in which CB 1 receptors are deleted in all principal neurons of the forebrain (Marsicano et al., 2003). These mice were used to show that the protective function of the ECS against excitotoxic seizures induced by kainic acid is due to the CB 1- dependent control of principal forebrain neurons activity (Marsicano et al., 2003). These results were important also because they showed for the first time the functional presence of CB 1 receptors on other cells than GABAergic interneurons, as it was believed until then (Freund et al., 2003). The
CaMKII α-CB 1-KO mice were also shown to be resistant to diet-induced obesity, suggesting that central
/ sympathetic but not non-neuronal peripheral CB 1 receptor signalling is determinant for the ECS control of energy balance (Quarta et al., 2010).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 58 Part II - The endocannabinoid system in the brain
WT GLU-CB 1-KO GABA-CB 1-KO
Figure 9. Photomicrographs of fluorescent in situ hybridization of CB 1 receptor mRNA in the brain of wild-type (WT), Glu-CB 1-KO and GABA-CB 1-KO mice Red squares focus on the hippocampus. Note the remaining CB 1 receptor expression in interneurons of the Glu- CB 1-KO and in the pyramidal cell layer of the GABA-CB 1-KO, Bar, 1 mm. Adapted from Bellocchio et al., 2010.
In 2006, Monory and colleagues generated two key conditional CB 1-KO lines: the Nex -CB 1-KO mice that are characterized by a selective deletion of CB 1 on cortical glutamatergic neurons (referred to as Glu-CB 1-KO; Monory et al., 2006) and the Dlx5/6 -CB 1-KO, showing a suppression of CB 1 restricted to forebrain GABAergic neurons (referred to as GABA-CB 1-KO; Monory et al., 2006). The
Glu-CB 1-KO mice were generated from the Nex -Cre mouse line (Goebbels et al., 2006). The helix- loop-helix transcription factor Nex participates in the development of mature cortical glutamatergic neurons originating from embryonic neuronal progenitors expressed only in the dorsal telencephalon proliferative zone (Ross et al., 2003). Indeed, the Cre-mediated recombination of Nex is expressed in mature glutamatergic cortical neurons, including neocortical and hippocampal glutamatergic neurons (Figure 9), but not in cortical GABAergic interneurons and to a much lesser extent in subcortical regions in the adult brain (Akagi et al., 1997). The GABA-CB 1-KO mice were generated from the Dlx5/6 -Cre mouse line (Monory et al., 2006). The homeobox Dlx5/Dlx6 genes are expressed in subpallial embryonic precursor cells that will later give rise to forebrain GABAergic neurons (Stühmer et al., 2002). Thus, the expression of the cre recombinase under the control of the regulatory sequences of Dlx5/Dlx6 genes drives recombination of the loxP sites in forebrain GABAergic neurons (Monory et al., 2006; Figure 9).
The generation of GABA-CB 1-KO and Glu-CB 1-KO lines provided crucial information on the anatomical distribution of CB 1 receptors by allowing detection and quantification of CB 1 receptor expression on both cell populations (Monory et al., 2006; Bellocchio et al., 2010), thereby closing the controversy regarding the actual expression of CB 1 receptors on glutamatergic neurons (Katona et al., 1999; Marsicano and Lutz, 1999; Freund et al., 2003; Domenici et al., 2006).
At the functional level, the use of GABA-CB 1-KO and Glu-CB 1-KO mice allowed first to show the importance of CB 1 receptors expressed on hippocampal glutamatergic neurons and not on GABAergic interneurons for the neuro-protective action of eCBs against excitotoxicity (Monory et al., 2006). Another key study aimed at defining the neuronal populations involved in the typical “tetrad” effects of
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 59 GENERAL INTRODUCTION
THC (Monory et al., 2007), which are commonly used to measure the cannabimimetic activity of drugs (Howlett, 2002). The tetrad model consists of hypolocomotion, hypothermia, antinociception, and catalepsy induced by cannabinoids (Little et al., 1988). Surprisingly, the deletion of the majority of CB 1 receptors in the brain in the GABA-CB 1-KO mice did not alter the THC-induced tetrad. Conversely,
THC-induced hypolocomotion and hypothermia were partially mediated by CB 1 receptors expressed on glutamatergic neurons (Monory et al., 2006). Bellocchio et al., (2010) reported that inhibition of food consumption by blocking CB 1 receptor signalling was mostly due to the activation of cortical glutamatergic CB 1 receptors. Unexpectedly, the same study revealed that deleting CB 1 receptors from forebrain GABAergic neurons increases stimulated food intake, demonstrating a so far unsuspected bimodal control of CB 1 receptors on food intake. Overall, the specific deletion of CB 1 receptors in excitatory and inhibitory neurotransmissions in the Glu-CB 1-KO and GABA-CB 1-KO, respectively, is a powerful tool to characterize the neuronal mechanisms of CB 1 receptor functions. Important functions are supported by CB 1 receptors located on glutamatergic terminals that represent the minority of the brain CB 1 expression (Marsicano et Lutz, 1999; Monory et al., 2006; 2007; Lafenêtre et al., 2009;
Bellocchio et al., 2010). This suggests that the CB1 receptor expression level on a given cell type is not indicative of its functional importance. Recently, other conditional CB 1-KO lines were developed to study the role of CB 1 receptors on particular neuronal populations, their pattern of deletions are listed in the Table 1.
CB 1-KO mouse line Cre mouse line Target cell type Main CB 1 recombination pattern
CB -KO 1 Ubiquitous (Marsicano et al., 2002)
Neocortex, hippocampus, amygdala, CamKII α-CB -KO CamKII α-BAC -iCre Principal forebrain 1 striatum, thalamus, hypothalamus, (Marsicano et al., 2003) (Casanova et al., 2001) neurons sympathetic ganglia
GABA-CB 1-KO Dlx5/6 -Cre Forebrain GABAergic CCK-expressing forebrain GABAergic (Monory et al., 2006) (Monory et al., 2006) interneurons interneurons
glutamatergic neurons form the Glu-CB -KO Nex -Cre Cortical glutamatergic 1 neocortex, archicortex, hippocampus (Monory et al., 2006) (Goebbels et al., 2003) neurons and amygdala D1R-YAC -Cre D -CB -KO Dopamine D receptors- 1 1 (Lemberger et al., 1 MSN of both ventral and dorsal striatum (Monory et al., 2007) expressing neurons 2007) TPH-CB -KO 1 TPH2 -CreERT2 Serotoninergic neurons Serotoninergic neurons of the raphe (Dubreucq et al., in (Weber et al., 2009) of the raphe nuclei nuclei press)
Sim1-CB -KO Sim1 -BAC-Cre Sim1 expressing 1 hypothalamus, mediobasal amygdala (Dubreucq et al., 2012) (Balthasar et al., 2005) neurons
Table 1. Available neuron type-specific CB 1-KO mouse lines.
In summary, genetic tools allow to address the specific roles of CB 1 receptors in distinct cell populations, thereby helping the dissection of the ECS functions. Moreover, given the general modulatory role of CB 1 receptor signalling and its importance in neuronal and cellular processes, these tools also provide interesting observations on the general mechanisms of brain activity.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 60
Part III
CB 1 RECEPTORS AND ASTROCYTES
Glial cells are the most abundant cell type in the human brain. Rudolf Virchow originally described them in 1858 as connective tissue of neurons and termed them neuroglia, suggesting that they could act as the “glue” that physically structures the brain (Virchow, 1858). Some years later in Italy, Camillo Golgi discovered one of the most important pigments staining method in the history of science, the silver nitrate staining now referred as Golgi’s staining. Using this approach, Golgi could observe individual cells in the brain, and drew glial cells surrounding the large body of neurons and axons. Because the glial processes appeared to connect blood vessels and neurons, Golgi hypothesized that the function of these cells is to feed neurons (Koob, 2009). Santiago Ramon y Cajal was contemporaneous of Golgi. In Madrid, he used an enhanced Golgi’s staining method with gold chloride and represented glial cells in the brain. We now know that this stain targeted intermediate filaments consisting mainly of glial fibrillary acidic protein (GFAP), a protein used today as an astrocytic marker (Middeldorp and Hol, 2011). The use of Golgi staining led to further characterization of glial cells. In 1891, the term “astrocyte” was introduced by Michael von Lenhossek because of the stellate form of these cells (von Lenhossek, 1891). Subsequently, other principal classes of glial cells, the oligodendrocytes and the microglia, were distinguished from astrocytes as a group of “adendritic” cells (Del Río Hortega, 1918). At the same time, Golgi’s staining method also revealed the anatomy of neurons, composed by the cell body, the axon and the dendritic extensions. Because axonal length can project over brain regions, and neurons appeared to be organized in networks, they were thought to support information storage (Koob, 2009).
61 GENERAL INTRODUCTION
Figure 10. Cajal’s histological preparations and drawings of astrocytes glial cells already underlined the neurovascular and neuronal coupling of astrocytes A. Drawing by Cajal showing protoplasmic astrocytes (containing relatively few fibrils) in the gray matter of the cerebral cortex labelled with the gold chloride method. A, large type of neuroglia cell; B, small type of neuroglia cell; C, end-foot inserted in a capillary; D, pyramidal cell; a, b capillaries. B 3. Drawing by Cajal of fibrous astrocyte in the white matter of adult brain (B 1 and B 2, formalin-uranium nitrate and gold-sublimated chloride). C 1 and C 2. Confocal microscope images of double-labelled immunohistochemical sections to visualize NeuN-immunoreactive neurons (in red) and GFAP-labelled astrocytes (green) in the human temporal neocortex. C1, processes from a GFAP-labelled astrocyte in apposition to a NeuN-labelled pyramidal neuron (arrow). C 2, a GFAP-labelled astrocyte emitting a process that contacts (arrow) a blood vessel (bv) which can be recognized by the red blood cells. Scale bar = 15 m. Note the similarity of the vascular end-feet on the blood vessel and processes surrounding neurons between tissue microscope images and the drawing of Cajal. Modified from (Cajal, 1913 in Garcia-Lopez et al., 2010).
By then the neuronal doctrine became absolutely dominating, and the synapse, identified by Charles Scott Sherrington (Sherrington, 1906), was considered as the principal place of integration in the nervous system. Rapidly developing electrophysiological experimental techniques allowed monitoring neuronal activity. After almost a century of study of neuronal functions, we have now a universally accepted vision of the general principles mediating exchanges of information in the CNS. Action potentials convey the information though neuronal membranes and activate synaptic release of neurotransmitters at synaptic terminals. In turn, neurotransmitters trigger changes in post-synaptic potentials, thereby transferring the electrochemical information though neuronal networks. Unlike neurons, astrocytes are not electrically excitable and do not generate action potentials. For this reason, astrocytes were mainly studied for their energy supply to neurons and neuro-protective properties throughout the nineteenth century. Only recently, the discovery that astrocytes are chemically excitable and can influence neuronal signalling led to an on going reconsideration of brain functions including astrocytes as active components of information transfer and processing.
In this section I will provide information about the expression of the ECS elements in astrocytes.
I will then describe the involvement of CB 1 receptors neuro-protective and metabolic functions of astrocytes. Finally, I will present evidence implicating CB 1 receptor signalling in the control of synaptic transmission and plasticity by astrocytes.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 62 Part III - CB1 receptors and astrocytes
III.1 Endocannabinoid system machinery is present in astrocytes
III.1.1 What are astrocytes and how do they communicate?
III.1.1.1 Astrocyte genesis and anatomical considerations
Neurons and neuroglia both originate from the neuroepithelial cells of the embryo (Rowitch and Kriegstein, 2010). The neuroepithelial cells then transform into the radial glia, which act as a universal neural precursor. Through several transitional forms, radial glia produces neuronal precursors and also originates astrocytes and oligodendrocytes. Some of the astrocytes retain these stem cell properties throughout life and underlie the adult neuro- and glio-genesis from specific brain areas (Rowitch and Kriegsten, 2010). In addition, neuroglial cells are instrumental in promoting neuronal survival at different developmental stages through the release of numerous neurotrophic factors.
The GFAP (glial fibrillary acidic protein) is the main intermediate filament of astrocytes and is a typical marker of these cell types, distinguishing them from other cell types in the CNS (Kimelberg, 2004; Middeldorp and Hol, 2011). As a key element of their cytoskeleton, GFAP warrants cell structural integrity and is associated with the transduction of molecular signals within astrocytes (Middeldorp and Hol, 2011). However, it is not clear whether (i) all astrocytes do express GFAP and (ii) all GFAP-expressing cells can be considered as astrocytes (Kimelberg et al., 2004). For instance, during adulthood, GFAP is expressed on stellate stem cells that give rise to neurons in the subventricular zone and the subgranular zone of the dentate gyrus (Type B cells; Kimelberg, 2004; Bordey, 2006). Furthermore, it has been proposed that both Golgi staining and GFAP immunolabelling reveal only 15% of the total astrocytic surface when compared to an astroglial membrane labelling (Bushong et al., 2002), indicating that astrocytes are morphologically more complex than initially appreciated. Depending on brain regions, a single mature rodent astrocyte can contact about 100,000 individual synapses (Oberheim et al., 2006). Other proteins are selectively expressed in astrocytes and are also used as specific astrocyte markers, such as the intermediate filament vimentin, the S100 calcium binding protein beta (S100 β) involved in the control of Ca 2+ influx and astrocyte proliferation, or the astroglial glutamate transporters EAAT1/GLAST and EAAT2/GTL-1. However, these proteins are expressed in region-specific astrocytes (Middeldorp and Hol, 2011). Therefore, GFAP represents the most used target to visually and genetically distinguish astrocytes from neurons (Middeldorp and Hol, 2011; Hirrlinger et al., 2006).
Two classes of central astrocytes were first described: the fibrous astrocytes of the white matter, and the protoplasmic astrocytes from the grey matter (Figure 11). Numerous fine processes characterize protoplasmic astrocytes whereas fibrous astrocytes show poor branching arborisation. Region-specific astrocytes or macroglial cells can be also distinguished. For instance, the Bergman glia and the Müller glia are found in the cerebellum and in the retina, respectively (Kettenmann and Verkhratsky, 2008). Because their molecular and phenotypic specificities are not yet fully described, the classification of astrocytes and neuroglial cells in general is still evolving nowadays. An intriguing
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 63 GENERAL INTRODUCTION particularity of glial cells is that their growing number and complexity is directly correlated to species evolution. For instance, the rat cerebral cortex contains twice more neurons than glial cells (Bass et al., 1971) while the human brain bears 1.5 times more glial cells than neurons (Nedergaard et al., 2003). Oberheim et al. (2009) compared astrocyte shapes between rodent and human neocortical astrocytes. The size of human protoplasmic astrocytes is about 2.5 – 3 times larger than in rodents. The human protoplasmic astrocytes possess 10 times more primary processes, and display more complex arborisations of processes than rodent astroglia. As a result, a single human protoplasmic astrocyte occupies a volume in the brain that is almost 30 times the volume in rodents and can contact about 2 millions of synapses, indicating that glial cell functions may be important for the increased complexity of brain functions that has emerged during evolution (Oberheim et al., 2009).
Figure 11. The different glial cells Pictures representing the morphology of protoplasmic and fibrous astrocytes, microglial cells and oligodendrocytes (from Barrett and Ganong, 2010)
The physical strategic position and connections of astrocytes with the other brain elements already led pioneer anatomists 100 years ago to propose that astrocytes would have key roles in metabolic support of neurons, modulation of brain information processing and recycling brain wastes (Golgi, 1885; Lugaro, 1907; Kettenmann and Ransom, 1995; Verkhratsky et al., 2011).
III.1.1.2 Network organization of astrocytes and ways of communication
Astrocytes are physically connected to both neurons and blood vessels and they are also organized in networks, thus communicating each other. Astrocytes can communicate with their cell partners by many ways. In particular, astrocytes represent the largest gap junction-coupled cell
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 64 Part III - CB1 receptors and astrocytes network within the CNS, where this mode of direct intercellular connections plays a role in the homeostatic regulation of ion levels (including Ca 2+ , Na + and K +) and of many molecules including glucose, glutamate and ATP (Giaume et al., 1997; Anderson and Swanson, 2000; Ransom et al., 2003). Astrocytes also influence CNS vascular tone and neuronal synaptic activity, which are facilitated, in part, via gap junction coupling (Bezzi and Volterra, 2001; Mulligan and MacVicar, 2004; Tabernero et al., 2006; Rouach et al., 2008).
Figure 12. Astrocyte calcium waves propagation. a. Photographs (~1 s time interval) showing the temporal sequence of Ca 2+ propagation across astrocytes after local Ca 2+ concentration elevation. b. Proposed mechanisms for waves propagation across astrocytes. IP3, generated by the active phopholipase C pathway, would diffuse to nearby glia through gap junctions and release Ca 2+ from their internal stores. The release of ATP, which can cause further generation of IP3 and release of ATP of nearby cells, would be involved in long range astrocyte signalling. From Haydon, 2001.
Gap junctions represent intercellular physical contact points allowing the direct trafficking of small molecules between adjacent cells without contacting the extracellular milieu and form a molecular association for the long-distance propagation of signals across astrocytic networks (Figure 12). The intensity of astrocyte gap junctional coupling led to the concept that astrocytes form syncytia , or networks, in which signals can be coordinated. Analysis of dye diffusion within the mouse cortex in situ revealed restricted astrocyte coupling (Houades et al., 2008). However, it is possible that the delay of dye diffusion would not be sufficient to label the whole astrocyte network. Moreover, metabolic processes could also limit protein diffusion, so that the dimensions of such astrocytes networks in the brain might be wider than currently assumed. Gap junctional coupling by astrocytes is mainly ensured by connexins (Cx), which are present in two main isoforms (Cx43 and Cx30; Dermietzel et al., 2000; Altevogt and Paul, 2004; Houades et al., 2008; Giaume et al., 2010). Recently, Cx43 and Cx30 down-regulation has been implicated in synaptic plasticity impairments in several brain regions as well as in the pathophysiology of neuro-degenerative diseases (Orellana et al., 2011; Pannasch et al., 2011).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 65 GENERAL INTRODUCTION
As previously noted, astrocytes do not generate action potentials. Instead, astrocyte excitability reposes partly on variations of intracellular Ca 2+ concentrations. The cellular Ca 2+ signal is manifested as elevations of cytosolic Ca 2+ from Ca 2+ stored in the endoplasmic reticulum (Perea and Araque, 2005). Ca 2+ mobilization from the internal stores depends on activation of receptors including ryanodine receptors and inositol tri-phosphate (IP3) receptors that are sensitive to IP 3 produced by activation of the phospholipase C (PLC). An increase of Ca 2+ signal in astrocytes is viewed as waves that can propagate to other astrocytes, neuronal vasculature or neurons, thus serving as an intracellular and intercellular signal (Figure 12; Haydon, 2001; Perea and Araque, 2005). Thanks to important technological advances, recent studies provided impressive insights into the mechanisms of intra-astrocyte Ca 2+ signalling at a nano-scale. Within astrocytes processes, Ca 2+ is accumulated in discrete segregated stores or micro-domains that can individually respond to single synaptic information, indicating that astrocytes discriminate multiple levels of communication and may serve as a functional integrator of synaptic events (Di Castro et al., 2011; Panatier et al., 2011). Ca 2+ variations have relevant functional consequences in the CNS as they can be observed either spontaneously or in response to different neurotransmitters both in vitro and in vivo . For instance, astrocyte Ca 2+ elevation can be triggered by whiskers sensory stimulation in living mouse cortical astrocytes (Wang et al., 2006). In the same region, combined stimulation of mouse whiskers and cholinergic cortical afferences enhanced whisker-evoked neuronal responses. This plasticity phenomenon depends on intracellular astrocyte Ca 2+ elevation (Takata et al., 2011). Other forms of synaptic plasticity controlled by astrocytes depend on their Ca 2+ signalling, demonstrating that astrocyte Ca 2+ signalling is a key determinant of neuronal processes and may have an important impact on animal behavior (Perea et al., 2009).
A major way for astrocytes to directly influence the activity of neurons and also regulate vascular flow is to release different chemical messengers referred as gliotransmitters. These include glutamate, GABA, D-serine, ATP, adenosine, nitric oxide (NO) but also several cytokines such as the tumor necrosis factor α (TNF α), arachidonic acid derivatives and as seen below, very likely eCBs (Walter, 2002; Walter and Stella, 2003; Volterra and Meldolesi, 2005; Perea et al., 2009). To date, the mechanisms underlying gliotransmitters release from astrocytes are not fully understood. Some of them are released at least in part through Ca 2+ -dependent exocytosis such as glutamate (Volterra and Meldolesi, 2005). Indeed, anatomical evidence demonstrates the presence of vesicular glutamate transporters (VGLUTs) and SNARE proteins in astrocytes, and inactivation of exocytotic proteins prevents gliotransmitter release from astrocytes (Volterra and Meldolesi, 2005). The recent generation of knock-out mouse lines lacking astrocytic-restricted exocytotic proteins (such as SNARE) and connexins should provide further information on the release mechanism of gliotransmitters (Halassa et al., 2009).
III.1.2 Cannabinoid receptor expression on astrocytes
The actual expression of cannabinoid receptors on astrocytes has been long debated in the endocannabinoid research field. Only few years ago, the group of Araque provided functional evidence of a direct control of CB 1 receptors on astroglial-neurons interactions (Navarrete and Araque,
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 66 Part III - CB1 receptors and astrocytes
2008, 2010; see section III.4.2). The presence of CB 1 receptor mRNA in astrocytes was first reported in human glioma cell lines and in primary cultures of rat astrocytes (Bouaboula et al., 1995a).
However, the same study did not detected CB 2 receptor gene expression in the same preparation. The expression of both CB 1 and CB 2 receptors was later observed both in primary cultures of human astrocytes and in glioma cells (Sánchez et al., 2001; Sheng et al., 2005). Other in vitro studies revealed the expression of CB 1 receptors in cultured astrocytes from rat and mice from CD1 genetic background (Sánchez et al., 1998; Molina-Holgado et al., 2002). The fact that CB 1 were not detected in other mouse strains such as C57BL6 and Swiss-Webster (Sagan et al., 1999; Walter and Stella, 2003; Lou et al., 2012) suggested that their astroglial expression may depend on animal strains or species, on the ability of different anti-CB 1 antibodies to recognize its glial counterpart, but also on the conditions of cell cultures. Furthermore, cultured astrocytes are much less complex in term of process ramification than astroglial networks in vivo (Verkhatsky et al., 2011), suggesting that their cytoarchitecture and molecular signature can be different. In rat brain slices, immunodetection coupled with electron microscopy pointed-out both peri-vascular and peri-synaptic expressions of astrocytic
CB 1 receptors in the neocortex, hippocampus, amygdala and the Nac (Moldrich and Wenger, 2000;
Rodriguez et al., 2001; Navarrete and Araque, 2008), which suggest important functions of CB 1 receptors in the control of both astrocyte-neurovascular coupling and regulation of neuronal function by astrocytes. Again, due the existence of contradictory results concerning the presence of astroglial
CB 1 receptors in the hippocampus as assessed by immnunohistochemistry in rat brain (Katona et al., 1999), the specific mutagenesis approach appears to be a landmark step to determine the anatomical expression and functions of astroglial CB 1 receptors.
III.1.3 Do astrocytes produce and metabolize endocannabinoids?
Although it is now recognized that astrocytes are equipped for eCB synthesis and breakdown, astroglial detection of eCBs raised technical challenges to set methods to quantify low eCB levels. The Stella’s laboratory was able to detect and quantify femtomole levels of eCBs using a chemical ionization gas chromatography / mass spectrometry and found that mouse astrocytes in culture produce AEA and 2-AG at similar levels as neurons in culture through a Ca 2+ -dependent mechanism (Walter, 2002; Walter and Stella, 2003). Interestingly, ATP was shown to potently increase eCB levels in astrocytes (Stella, 2010). Furthermore, primary astrocytes inactivate eCBs by uptake and / or hydrolysis (Di Marzo et al., 1994; Beltramo, 1997; Bisogno et al., 2001). However, the majority of FAAH expression found in the brain, as assessed by immunohistochemistry, is localized in neurons (Romero et al., 2002; Benito et al., 2003, 2007). Suarez et al. (2010) proposed an extended description of the protein expression of the ECS elements in in situ rat astrocytes located near the brain ventricular areas. They reported that the AEA and 2-AG synthesis enzymes, NAPE-PLD and DAGL α/β, as well as their degradation enzymes, FAAH and MAGL, co-localize with GFAP- and / or Vimentin–expressing cells (Suárez et al., 2010). Importantly, a recent detailed gene and protein expression of MAGL within the dentate gyrus uncovered that MAGL consistently co-expresses with both GFAP and GLAST in this region (Uchigashima et al., 2011).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 67 GENERAL INTRODUCTION
Overall, these data suggest that eCBs are likely released and degraded by astrocytes in vivo, but their putative role in astrocyte signalling is presently unknown. As noticed below, CB 1 receptors can directly modulate several astrocyte functions, but the origin (i.e. neuronal or astroglial) of the eCB signal has never been assessed.
III.2 CB 1 receptors participate in astroglial neuro- protective functions
III.2.1 Astrocytic defense mechanisms
One of the best characterized functions of glial cells is to protect the CNS against injuries. Astrocytes are strategically placed, and remarkably well-equipped to prevent, detect and struggle against toxic agents as well as infections that might compromise CNS functions.
Astrocytes maintain a constant balance in the extracellular medium by uptaking metabolic wastes and extracellular signalling molecules. For instance, astrocytic processes express transporters for a variety of neurotransmitters including glutamate and GABA. These transporters participate in the rapid removal of neurotransmitters released into the synaptic cleft, which is essential for both the termination of synaptic transmission and maintenance of neuronal excitability (Takuma et al., 2004). Glutamate uptake by astrocytes is also crucial for protecting neurons against excitotoxicity (Tanaka et al., 1997). Overstimulation of glutamate receptors is highly toxic for neurons and, in the absence of efficient mechanisms for its removal; it can cause extensive neuronal injuries. This is mainly achieved by the astrocyte-specific glutamate transporters GLT-1 and GLAST (Tanaka et al., 1997; Gadea and López-Colomé, 2001). In vivo studies demonstrated the primary importance of astrocytic glutamate uptake in preventing glutamate-induced excitotoxicity. For instance, the GLT-1 knock-out mice show strong spontaneous seizures and over-expression of GLT-1 prevents neuronal loss in models of excitotoxicity (Tanaka et al., 1997). This suggests that modulation of the glutamate uptake capacity of astrocytes may represent a promising therapeutic target for pathologies involving excitotoxicity.
Neuronal activity also results in local increases of extracellular K + in the extracellular space. Without tight regulatory mechanisms, this could dramatically alter the neuronal membrane potential, leading to neuronal hyperexcitability and pathological consequences (Walz, 2000). Astrocytes also ensure ion homeostasis by buffering extracellular K+. Indeed, astrocytes have a strongly negative resting potential (-85 to -90 mV) and express a number of K + channels, resulting in a high membrane permeability to K +. These features enable astrocytes to accumulate the excess extracellular K +, which can diffuse in the astrocytic syncitium through gap junctions so lowering its concentration gradient. This regulatory mechanism allows the maintenance of physiological extracellular K + concentration (Seifert and Steinhäuser, 2011).
Although microglial cells are generally considered as the main resident immune cells of the brain, it is important to note that astrocytes are immunocompetent as well, and that they act as important regulators of brain inflammation induced by various insults such as infections, injuries or abnormal protein accumulation (Farina et al., 2007). Astroglial pro-inflammatory actions are
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 68 Part III - CB1 receptors and astrocytes physiological processes aiming at protecting the brain from potentially deleterious threats. However, if the insult persists, the inflammatory process may negatively affect neuronal functions and viability, thus contributing to disease progression. Neuro-inflammation has indeed been implicated in several neuropathologies including Alzheimer's and Parkinson's diseases, multiple sclerosis, and strokes. Activated astrocytes can release a wide array of immune mediators such as cytokines (for instance interleukins and TNF α), chemokines and growth factors that may exert either neuroprotective or neurotoxic effects (Farina et al., 2007).
III.2.2 Astrocyte prevention from excitotoxicity: putative role of CB 1 receptors
The ECS is known to protect against excitotoxicity (Marsicano et al., 2003). Even though this process involve a direct regulation of neuronal excitability by CB 1 receptors located on glutamatergic neurons (Monory et al., 2006), recent studies reported that CB 1 receptors activity might also control glutamate uptake pools by astrocytes. In rat astrocyte cultures from the cortex, Shivachar (2007) observed that both CB 1 receptor agonists CP55,940 and WIN55,212-2 inhibited D-aspartate uptake from these cells. This effect was partially reversed by SR141716A, suggesting that over-activation of astroglial CB 1 receptors would interfere with excitatory amino acid in resting conditions. In contrast, AEA was shown to reverse AMPA-induced neurotoxicity and down-regulation of GLT-1 and GLAST mRNA in neocortical CD1 mouse astrocyte cultures and in spinal cord of a mouse model of multiple sclerosis. These effects were reversed by SR141716A, suggesting a CB 1 receptor-dependent action (Loría et al., 2010). In this study, eCBs would prevent glutamate overload by facilitating its uptake by astrocytes through an uncharacterized CB 1 receptor signalling pathway. Thus, contrasting results are present in the literature regarding the role of astroglial CB 1 receptors in the regulation of excitatory amino acids in the context of excitotoxic processes.
III.2.3 CB 1 receptors regulate astrocytic immune responses
Early in vitro evidence showed that CB 1 receptor agonists inhibit the release of pro-inflammatory cytokines from reactive astrocytes. In this sense, activation of CB 1 receptors would have anti- inflammatory properties. The Theiler's Murine Encephalomyelitis Virus (TMEV) is known to induce encephalomyelitis comparable to multiple sclerosis (Stavrou et al., 2010). Primary cultures of neonatal mouse cortical astrocytes stimulated with TMEV release NO and TNF α and this effect is prevented by
AEA (Molina-Holgado et al., 1997) in a CB 1 receptor-dependent manner (Molina-Holgado et al., 2002). In comparable conditions, UCM707, a potent and selective anandamide uptake inhibitor, is able to reduce NO release, TNF α and interleukin-1beta (IL-1β) and these effects are blocked by administration of both CB 1 and CB 2 receptors antagonists, further supporting the involvement of the ECS in the anti-inflammatory responses of astrocytes (Ortega-Gutiérrez et al., 2005). Interestingly, IL- 1β stimulation has been shown to mediate both NO and cytokines release from astrocytes, thereby activating microglial cells and favoring astrocyte apoptosis (Ehrlich et al., 1999). Altogether, these data suggest that CB 1 receptor activation could primarily prevent IL-1β production. Moreover, WIN55,212-2
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 69 GENERAL INTRODUCTION blocked the release of both NO and TNF α induced by IL-1β administration in human cultured astrocytes, this effect being partially reversed by CB 1 receptors antagonists (Sheng et al., 2005). Froger et al. (2009) reported that exogenous cannabinoids prevent astrocytic Cx43 deregulation induced by IL-1β and TNF α in mice cultured astrocytes, an effect mediated by CB 1 receptor activation.
Indeed, the anti-inflammatory effect of CB 1 receptor agonism is likely mediated by gap junctional maintenance at astrocytes. Robust evidence for the immunomodulatory and neuroprotective properties of cannabinoids exists in animal models of multiple sclerosis (Zajicek and Apostu, 2011; Sánchez and García-Merino, 2012). Indeed, these anti-inflammatory properties of cannabinoids could be mediated, at least partly, by a direct control of CB 1 receptors on astrocytes reactivity.
III.2.4 CB 1 receptors control astroglial differentiation and astrogliosis
A particular class of astrocytes acts as stem or progenitor cells for both astro- and neuro- genesis during embryonic development as well as in adulthood. The progenitor cells do express several astrocyte-specific markers including GFAP, but exhibit specific expression of immature cell markers such as nestin, an intermediate filament implicated in cell growth characterizing pre-mitotic cells (Alvarez-Buylla and Lim, 2004). Importantly, both CB 1 receptors and FAAH proteins are present in these progenitors (Aguado et al., 2006). As assessed both in vitro and in vivo in 2-days post-natal rat and mouse brains as well as in 3-months old mice, exogenous cannabinoid administration increases the proliferation of GFAP- and nestin-co-expressing cell progenitors. Both exogenous and endogenous cannabinoids favour the differentiation of such progenitors into astrocytes in a CB 1 receptor-dependent manner, and this developmental feature is associated with a decrease of neuronal specification of the progenitors (Aguado et al., 2006). This progliogenic effect of cannabinoids does not influence progenitor apoptosis. In the dentate gyrus of adult CB 1-KO mice, the same authors reported a decrease of bromodeoxyuridine- (BrdU, commonly used in the detection of proliferating cells in living tissues) and S100 β-positive cells, while the opposite consequence was observed in the
FAAH-KO mice, indicating that endogenous CB 1 receptor signalling is necessary for normal adult astrogliogenesis (Aguado et al., 2006). The mechanisms underlying this promoting effect of CB 1 receptors on astrocytic proliferation are unknown, even though the presence of CB 1 receptors on these cells suggests a direct control of these processes, which may have an important role in both normal brain development and hippocampal plasticity in adults (Massa et al., 2011). On the other hand, neuronal CB 1 receptor signalling might also indirectly modulate astrogliogenesis and neurogenesis.
The behavior of glial cells and their progenitors may change after brain injury ranging from neurodegenerative or inflammatory conditions to acute invasive brain injury, such as trauma or stroke. In these conditions, astrocytes react by multiple and complex changes in morphology, gene expression and function, a process referred to as “astrogliosis”. Astrocytes become hypertrophic and up-regulate most prominently GFAP and vimentin that allow, for instance, scare formation but also prevent neuronal regeneration (Pekny and Pekna, 2004). Several in vivo studies in animals pointed- out particular effects of CB 1 receptor signalling manipulations on reactive astrocytes. In rat, an intra-
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 70 Part III - CB1 receptors and astrocytes cortical administration of the amyloid β peptide leads to an increase of GFAP expression. This effect was prevented by systemic administration of a CB 1 receptor agonist and reversed by SR141716A (Esposito et al., 2007). Interestingly, WIN55,212-2 also reverses amyloid β–induced spatial reference memory decline in rat (Ramírez et al., 2005), suggesting that stimulation of CB 1 receptors could have beneficial effects on memory losses observed within Alzheimer disease possibly through an attenuation of gliosis. Recently, two studies from the Viveros laboratory also suggested that stimulation of CB 1 receptors could alleviate brain injury–induced astrogliosis via an enhancement of brain estradiol availability (López Rodríguez et al., 2011), but also decrease reactive astrogliosis in the hippocampus following maternal deprivation (López-Gallardo et al., 2012). In the same way, THC is able to protect against MDMA neurotoxicity (i.e. dopaminergic neurons loss) by totally preventing the huge astrogliosis induced by the drug in the mouse striatum in a CB 1 receptor–dependent manner (Touriño et al., 2010). Within the mouse hypothalamus, a 2-weeks high fat diet regime doubles GFAP expression as compared to standard diet. Blocking CB 1 receptors abolishes both the preference for high fat diet and astrocyte proliferation in the same region (Higuchi et al., 2010), suggesting that reducing astrogliosis could be a mechanism by which CB 1 receptor antagonists protect against obesity.
III.3 Role of CB 1 receptors in astroglial metabolic support of neurons
III.3.1 Astrocytes-neurons metabolic coupling
About 20 % of the body energy production at rest is consumed by the brain whereas this organ represents only 2 % of the body mass (Magistretti and Pellerin, 1999). The close proximity of astrocytes to blood vessels places them in the optimal location to monitor circulating metabolic information. Astrocytes provide energy to neurons trough several pathways. In the brain, neurons require a higher rate of energy as compared to glial cells. Astrocytes take up glucose from blood vessels through the glucose transporter GLUT1 expressed by both capillary endothelial cells and astrocytes. A large portion of the glucose is then metabolized into lactate in astrocytes. Neurons primarily use lactate as an energy substrate (Pellerin and Magistretti, 1994; Bouzier-Sore et al., 2003). Lactate is transported across the astrocytic plasma membrane by the mono-carboxylate transporters MCT1 (Pellerin, 2005) and enters into neuronal cytosol through the MCT4 (Tekkök et al., 2005). Once in neurons, lactate serves as substrate for ATP production in mitochondria.
Brain energy production can also be mediated by the metabolism of glycogen (Magistretti et al., 1993). As a large energy reserve of the brain, glycogen, unlike glucose, can be rapidly metabolized without requirement of ATP supplying. It is also proposed that glycogen represents an alternative energetic source for neurons in case of limited energy availability, such as in hypoglycaemic state. Glycogen is almost exclusively localized in astrocytes in the adult brain and glycogenolysis also results in lactate production and release in the extracellular space (Brown and Ransom, 2007). Interestingly, glycogen content is controlled by neurotransmission including noradrenergic and serotoninergic ones and decreased neuronal activity observed during anaesthesia is associated with
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 71 GENERAL INTRODUCTION increased levels of brain glycogen storage (Brown, 2004). Conversely, sensory stimulation is associated with a decrease in glycogen levels in the activated brain areas, demonstrating a tight coupling between neuronal activity and glycogen mobilization (Dienel et al., 2007). Glycogen metabolism is also implicated in particular cognitive functions and associated synaptic plasticity (Gibbs et al., 2006; Suzuki et al., 2011). I will provide further information regarding this interesting point in a following section.
A third metabolic pathway used by the brain derives from fatty acids, for instance when lipids are the unique energetic source (e.g. during lactation, starvation or high fat diet; Yi et al., 2011). Fatty acids can be metabolized in ketone bodies (i.e. acetoacetate, 3-hydroxybutyrate, acetone) mainly in mitochondria of the liver but also in astrocytes (Guzmán and Blázquez, 2001). Some in vitro and in vivo evidence indicates that ketogenesis may play important roles in maintaining synaptic activity and in the protection against brain injuries, cell death and epilepsy. However, the contribution of this alternative energetic source to brain functions is largely under-studied as compared to carbohydrate– derived fuelling (Guzmán and Blázquez, 2001).
III.3.2 Role of CB 1 receptors in astrocytic metabolic functions
Energy metabolism is one of the most important functions of astrocytes. In parallel, the ECS is a well-characterized regulator of body energy metabolism (Pagotto et al., 2006). However, only very few studies addressed the role of the ECS in astroglial energy supplying. In 1998, Sánchez et al. analyzed the effect of CB 1 receptor agonists on glucose metabolism in cultured astrocytes of new-born rats. In this preparation, both THC and HU210 increased glucose oxidation and glycogen synthesis in cortical and hippocampal astrocytes in a CB 1 receptor- and MAPK activity-dependent manner (Sánchez et al., 1998). A similar effect was observed on glycogen synthesis. The THC–induced potentiation of glucose oxidation was prevented by pertussis toxin, indicating the involvement of G i/o proteins. However, inhibition of cAMP was ineffective in this phenomenon, suggesting a cAMP–independent metabolic effect of CB 1 receptor stimulation (Sánchez et al., 1998). The fact that THC increases brain glucose metabolism in humans (Volkow et al., 1996), together with the known major brain glucose consumption by astrocytes, suggest that a prominent role of astroglial CB 1 receptors in brain energy metabolism exists in vivo .
THC and HU210 were also shown to stimulate ketone bodies production from palmitate in primary rat astrocytes (Blázquez et al., 1999). This increase of astrocyte ketogenesis by THC was dependent on CB 1 receptor activation but was not reversed by inhibition of cAMP and MAPK or by buffering Ca 2+ concentrations. Instead, this effect of THC was proposed to be mediated by a specific enzymatic pathway leading to lipid transport and metabolism in mitochondria (Blázquez et al., 1999). Interestingly, both THC and ketogenic diet (rich in lipids and poor in carbohydrates) are known for their anticonvulsant properties both in humans and animals (Carlini, 2004; Kossoff and Rho, 2009).
Therefore, it would be interesting to assess whether CB 1 receptor activation on astrocytes participates in the protection of cannabinoid against neuronal hyper-excitability possibly by a facilitation of ketogenesis.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 72 Part III - CB1 receptors and astrocytes
III.4 Astroglial control of information processing: CB 1 receptors also matter
III.4.1 Astrocytes and neurons are necessary partners for adapted memories
One of the main responses of astrocytes to external stimuli is an increase of cytoplasmic Ca 2+ concentration (Perea and Araque, 2005). Indeed, astrocytes respond to neurotransmitters including glutamate, GABA, acetylcholine, ATP, noradrenalin and eCBs (Cornell-Bell et al., 1990; Kang et al., 1998; Araque et al., 2002; Bowser and Khakh, 2004; Bekar et al., 2008; Navarrete and Araque, 2008). Theoretically, astrocytes could be sensitive to any molecule already known to affect neurons. Accordingly, astrocytes do express a wide array of receptors including ionotropic and metabotropic glutamate receptors, GABA receptors, dopamine receptors, P2X and P2Y receptors. Therefore, astrocytes are well equipped to listen to synaptic dialogues. Astrocytes also directly participate to synaptic activity by releasing molecules logically called gliotransmitters, so that the concept of the Tripartite Synapse emerged (Araque et al., 1999). Early studies showed that, in hippocampal brain slices; activation of mGluR induces a Ca 2+ -dependent release of glutamate from astrocytes that subsequently activates neuronal AMPA / NMDA receptors, eventually leading to an intra-neuronal Ca 2+ elevation (Pasti et al., 1997). Since then, astrocyte-derived glutamate has been shown to promote neuronal excitability at many synapses including within the hippocampus, the neocortex, the retina, the Nac and the olfactory bulb, and to synchronize the activity of pyramidal neurons (Perea et al., 2009).
One of the most fascinating powers of astrocytes is the control synaptic plasticity. The first pieces of evidence of a direct control of gliotransmission on plasticity are relatively recent, and the consequences of modulating astroglial signalling on different forms of plasticity will be surely much better documented in the future. Following tetanic stimulation of CA1 afferents, astrocytes release
ATP that can be subsequently metabolized into adenosine and acts at pre-synaptic adenosine A 1 receptors to hetero-synaptically depress excitatory transmission (Pascual et al., 2005). Using the tetracycline-controlled transcriptional activation (tet on/off system) in mice, the same group genetically inhibited SNARE-dependent release of transmitters from astrocytes in dnSNARE mice (Halassa et al., 2009). As a consequence, dnSNARE mice were resistant to both recognition memory deficits and impairment of hippocampal LTP induced by sleep deprivation. By coupling this genetic approach with pharmacological targeting of A 1 receptors, they proposed that astrocytic ATP and adenosine A 1 receptor activity contribute to the effects of sleep deprivation on memory (Florian et al., 2011).
Importantly, astrocytes directly participate to NMDA receptor-mediated long-term plasticity by releasing D-serine, a co-agonist of these receptors acting at their allosteric glycine site. This was first demonstrated in the supraoptic nucleus of the hypothalamus by Panatier et al. (2006). In rat, lactation induces glial loss that is associated with an impairment of NMDA receptor-mediated LTP in this region (Theodosis and Poulain, 1993). In lactating animals, exogenous D-serine re-established an NMDA receptors-mediated LTP comparable to the one observed in virgin animals (Panatier et al., 2006). The
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 73 GENERAL INTRODUCTION necessity of D-serine release from astrocytes in NMDA receptor-mediated LTP was also demonstrated in hippocampal as well as in mPFC slices (Henneberger et al., 2010; Fossat et al., 2012). Consistently, inhibition of D-serine synthesis by knocking-out the serine racemase in mice leads to recognition memory impairments and exogenous systemic administration of D-serine in mice facilitates such memory performances (Labrie et al., 2009; Bado et al., 2011).
A demonstration of the implication of glycogen metabolism in higher brain functions was also obtained by Gibbs et al. (2006), who demonstrated that pharmacological inhibition of glycogenolysis in astrocytes by the inhibitor of glycogen phosphorylase DAB (Diaminobenzidine Tetrahydrochloride) abolishes memory consolidation in young chickens in a bead discrimination learning task. Very recently, the involvement of astrocytic glycogen-derived lactate in long-term memory consolidation and in in vivo maintenance of NMDA receptor–dependent LTP in the hippocampus was demonstrated (Suzuki et al., 2011). This study showed that intra-hippocampal injection of DAB specifically impairs long-term consolidation of inhibitory avoidance memory in rats. This effect can be rescued by exogenous lactate administration, but not when the expression of the main neuronal lactate transporter is disrupted, demonstrating the importance of glycogen-derived lactate transport into neurons for memory consolidation. Glycogenolysis and astrocytic lactate transporters were also shown to be critical for the induction of molecular changes required for memory formation, including phospho-cAMP response element-binding (CREB) and activity-regulated cytoskeletal-associated protein ( Arc ), suggesting that lactate may also act as a signalling molecule for adapted memory functions (Suzuki et al., 2011).
Understanding how astrocytes control synaptic plasticity in vivo and what could be the behavioral consequences of such a control is an important challenge in the field. Namely, it implies to generate experimental models allowing the distinction of the neuronal versus astroglial contribution in the control of a given function.
III.4.2 Astroglial CB 1 receptors are key elements of the tripartite synapse
The role of astroglial CB 1 receptors in the modulation of excitatory transmission in the hippocampus was reported in two key studies from the Araque laboratory (Navarrete and Araque,
2008; 2010). First of all, the authors reported that astrocytes located in CA1 express functional CB 1 2+ receptors. Local application of different CB 1 receptor agonists increases intracellular Ca levels in astrocytes. This effect is prevented by CB 1 receptor antagonists and absent in the constitutive CB 1-KO mice, but it is insensitive to pertussis toxin, indicative of a G i/o protein–independent signalling pathway. 2+ CB 1-mediated increase of intracellular Ca levels in astrocytes requires activation of PLC, suggesting that the interaction of CB 1 receptors with the G q protein subunit may be favoured in astrocytes as compared to neurons (Navarrete and Araque, 2008). Astroglial CB 1 receptors activation, through a retrograde signalling from neurons, is necessary for NMDA receptor-mediated neuronal excitability evoked by astrocyte Ca 2+ elevations, suggesting their importance for astrocyte-neuron excitatory signalling (Perea and Araque 2005; Navarrete and Araque, 2008). However, these results do not
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 74 Part III - CB1 receptors and astrocytes characterize neither the endogenous NMDA receptor agonist implicated in this process, nor the cellular origin (astroglial versus neuronal) and identity of the released eCBs that act at astroglial CB 1 receptors in these conditions. Importantly, astroglial CB 1 receptors mediate synaptic potentiation (i.e. facilitation of neurotransmitter release) at excitatory CA3-CA1 synapses in a group I mGluR- dependent fashion, a phenomenon that co-exists with DSE mediated by neuronal CB 1 receptors (Navarrete and Araque, 2010). Therefore, these studies widen the view of the mechanisms by which the ECS can modulate synaptic plasticity and also suggest potential roles of a direct control of CB 1 receptors on astroglial functions in synaptic plasticity.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 75
Part IV
CB 1 RECEPTORS AND MEMORY
Given the widespread expression of CB 1 receptors within the brain and their multiple ways of controlling brain physiology, stating that the involvement of the ECS in memory functions is very complex will not surprise the reader. First, memory is multimodal and each memory process is mastered by specific, albeit sometimes overlapping, brain networks. Second, the ECS signalling is involved in all the memory subtypes in a time-, situation- and brain region-dependent manner. Third, because CB 1 receptors are also involved in many other non-mnemonic functions, it is very difficult to find the best experimental conditions to track their role on “pure” memory processing. In the following section, I will briefly introduce the distinct forms of memory as conceived within the most documented current models. Regarding the ECS, I will first present studies assessing the effect of cannabinoid agonists on working memory. Then I will focus on the role of CB 1 receptors on aversive learning and its adaptation. Lastly, I will summarize the known role of the ECS on other forms of memory.
IV.1 The memory systems
IV.1.1 Short and long –lasting memories
Memory refers to the faculty of retaining and recalling past experiences based on the processes of learning, retention and retrieval (Milner et al., 1998). Information can be retained for a short period of time (e.g. seconds to minute range), for a longer time period (e.g. days, months) or remote period (e.g. years). The functional dissociation of time-dependent memory systems comes from studies of memory performances in brain-lesioned patients and of the cellular mechanisms associated to learning-induced changes in synaptic properties in invertebrates (Milner et al., 1998). For instance, long-term memories were shown to depend on de novo protein synthesis in aplysia (Milner et al., 1998; Kandel, 2001). The observation that long-term changes in synaptic plasticity require new protein synthesis represents one of the most important clues leading to the hypothesis that memory processes are indeed mediated by changes in synaptic plasticity (Milner et al., 1998; Kandel, 2001).
77 GENERAL INTRODUCTION
Moreover, experiments in rodents led to the theory that a shift from hippocampus-dependent to neocortical-dependent memory might delineate ancient memory storage (Bontempi et al., 1999; Frankland et al., 2004).
IV.1.2 Long-term memory systems
The existence of multiple memory systems was already proposed at the early beginning of the 19th century, notably based on the distinction of memory for facts, or representative memory, and memory for skills, or habits (Milner et al., 1998; Squire, 2004). Again, the specific memory impairments described in amnesic patients across tasks led to neuropsychological dissociations between memory types, which were further identified in animal lesion studies (Figure 13). Indeed, human memory can be declarative, as defined by the capacity of explicit recollection about fact and events, or “ knowing that ” (Squire, 2004). Human declarative memory can be further dissociated in semantic memory, defined by general knowledge unrelated to specific experience (“ Paris is the capital of France ”), and episodic memory, which is related to specific, contextualized learning episodes or experiences (“ I visited Paris last year ”). Declarative memory allows referring to a remembered event so as to infer its relationships, similarities and contrasts with other events or items (Squire, 2004). In animals, this form of memory is often called reference or relational memory, in which the medial temporal lobes are critically involved (Slangen et al., 1990; Jaffard et al., 2000; Milner et al., 1998; Squire, 2004). Memory can be also non declarative, implicit, referring to several additional memory subtypes, requiring specialized performances such as skills and habits (“ knowing how ”), sensory learning, priming (influence of a presented stimulus on later responses), classical or Pavlovian conditioning and non associative learning (sensitization / habituation; Squire and Zola-Morgan, 1988; Squire, 2004). It is important to note that these distinctions are not strict. Adapted behaviors often require several memory processes that interact in a given situation (Phelps, 2004; Squire, 2004; Hartley and Phelps, 2010).
MEMORY DECLARATIVE NONDECLARATIVE
SEMANTIC EPISODIC PRIMING PROCEDURAL SENSORY MEMORY CLASSICAL NONASSOCIATIVE NONASSOCIATIVE CONDITIONING LEARNING
Figure 13. Long-term memory systems Taxonomy of the mammalian long-term memory systems. Adapted from Squire, 2004.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 78 Part IV - CB1 receptors and memory
IV.2 Cannabinoid-induced impairment of working memory
IV.2.1 Working memory
The term “working memory” (WM) was first proposed by Miller et al., in 1960 in the context of the computer-based digit span memory test in human. In this test, subjects are submitted to a list of random numbers or letters (without meaningful presentation patterns) presented at a certain time rate (for instance, one per second). At the end of the sequence, subjects are asked to recall the items in a certain order (for instance, in backward order). This task not only imply to store items for a short period of time, but also to manipulate the stored information (Miller, 1960). Indeed, the concept of short-term memory, which refers to the simple temporary storage of few amount of information, evolved into WM, which implies a combination of storage and processing.
A serial model of WM was first proposed by Atkinson and Shiffrin (1968). In this model, all external information passes through a unitary sensory register before being stored and processed on WM and finally flowing in or out of long-term memory. However, in a serial model of memory storage, information could not be stored if one register is impaired or overloaded. However, patients with impaired short-term memory are actually able of long-term learning. In the same way, in healthy subjects, saturating WM, by for instance asking to recall a sequence of learned digits, did not strongly impede performances on a concurrent cognitive task, requiring for instance reading comprehension (Baddeley and Hitch, 1974). Based on these remarks and new experimental data, Baddeley and Hitch in 1974 proposed the first version of the most currently acknowledged WM model.
CENTRAL EXECUTIVE
Visuo -spatial Episodic Phonological Sketch -pad buffer loop
LONG -TERM MEMORY
Figure 14. Baddeley's multicomponents model of Working Memory In this model, the central executive is responsible of information integration from the slave systems, the visuo- spatial sketch-pad that hold visual information and the phonological loop that hold auditory information. The episodic buffer can temporally retain multimodal sensory experience. These slave systems hold information from the external environment or information already encoded in long-term memory. Adapted from (Baddeley, 2010).
It is based on three components that work in parallel and interact each other. The central executive is an attentional control system allowing the temporary storage and manipulation of information necessary for performing a wide range of cognitive activities, as well as interfacing with
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 79 GENERAL INTRODUCTION long-term memory. The central executive is supported by two short-term slave storage systems, one for visual information, the visuo-spatial sketchpad, and one for auditory information, the phonological loop. More recently, Baddeley and Baddeley (2007) added a forth component, the episodic buffer at the interface of long-term memory as capable of holding multidimensional episodes combining, for instance, visual and auditory information. It provides a temporary and very limited store in which the various components of WM can interact with long-term memory.
In animal psychology, the concept of WM was also used to qualify within session acquisition of radial maze performances in rats (Olton, 1979). This test is performed in a multi-arm radial maze and a food pellet is deposed in each arm. The animals are submitted to several trials per day and need to remember which arm had already been visited on that day in order to maximise reward. As stated by Baddeley, WM represents a theoretical framework for investigating a wide range of cognitive activities including “pure” short-term memory, decision making and action planification. Indeed, it is difficult to assume that anatomical localization studies will provide a sharp understanding of the neurobiology of a system as complex as WM. However, human brain injury and functional imaging as well as animal models provided important insights on specific dimensions of WM. Indeed, impairments in WM tasks for objects or spatial locations are seen following lesions or transient inactivation of both the hippocampal system and the prelimbic area of the PFC (Kesner et al., 1996; Hampson et al., 1999; Lee and Kesner, 2003; Yoon et al., 2008). Prefrontal and hippocampal regions have been proposed to cooperate to control behavior via a direct and / or indirect pathway. In this context, inputs from the hippocampus to the PFC may organize the cortical representation of learned events (Laroche et al., 2000).
IV.2.2 Effects of cannabinoids on human working memory
The negative effect of cannabis and THC on WM performances in human is very well documented (Jones, 1978; Hall and Solowij, 1998; Ranganathan and D’Souza, 2006). From the early 1970’s, several clinical studies reported the consequences of THC administration on short-term recall of items from several sensory modalities in moderate to high adult cannabis consumers. For instance, acute THC administered orally or inhaled was shown to decrease accuracy and to increase the time needed to recall a word list or a story (Abel, 1971; Darley et al., 1973; Ranganathan and D’Souza, 2006). These effects were observed at various doses ranging from 6 to 60 mg per individual. Tinklenberg et al (1970) observed an increased number of errors in both forward and backward digit span recall 1.5 hour after a single oral administration of THC (20 to 60 mg) in infrequent THC users. These deleterious effects were still observed 3.5 hours following treatment. In another WM task, in which subjects are asked to associate letters by pressing a key following a previously learned code, THC also increased the error rate in occasional users (Kelly et al., 1990; Wilson et al., 1994). The delayed-matching-to-sample-task (DMTS) is frequently used to assess WM both in human and animals (Walker, 2010). In this test, two or more comparison stimuli are presented after a sample stimulus. Subjects are asked to select the stimulus that matches with the sample stimulus. Acute inhaled or intra-venous THC (2.5 to 11 mg) in infrequent users impaired DMTS performances without changing the response time (D’Souza et al., 2004). Importantly, several studies reported a lack of
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 80 Part IV - CB1 receptors and memory effect of THC on WM either in heavy consumers or when THC was administered chronically in the experimental procedure, suggesting a tolerance effect of THC on WM (Greenwald and Stitzer, 2000). Furthermore, the WM deficits do not persist in heavy THC consumers (for several years) when subjects are not under the influence of the drug (Pope et al., 2001; Hanson et al., 2010), although the question of whether long -term cannabis use can cause irreversible deficits in high brain functions is highly controversial (Fletcher and Honey, 2006; Vadhan et al., 2009).
IV.2.3 Cannabinoid and animal models of working memory
Animal studies, which enable a more controlled drug regime and more constant behavioral testing have confirmed human results and suggest that exogenous cannabinoid treatment selectively affects WM processes. WM can be assessed in animals through several behavioral paradigms. In nonhuman primates, DMTS performances, in which monkeys have to match geometric symbols to receive a food reward, are highly sensitive to acute systemic THC administration (0.3 to 4.0 mg/kg; Schulze et al., 1988). In rats, DMTS can be carried out by training the animals to match colour light signals by lever pressing or hole nose poking. Acute systemic THC administration (0.7 to 2.0 mg/kg) prior training impairs this task in rats and these effects are completely abolished 24 h following drug administration (Heyser et al., 1993; Mallet and Beninger, 1998; Hampson and Deadwyler, 1999).
These effects were also reproduced using the potent CB 1 receptor agonist WIN55,212-2 (0.1 to 0.5 mg/kg) and blocked by SR141716A, indicating that cannabinoids impair WM through activation of CB 1 receptors (Mallet and Beninger, 1998; Hampson and Deadwyler; 2000).
Cannabinoid-induced WM memory deficits in rodents have been also reported in spatial WM (SWM) paradigms highly sensitive to hippocampal dysfunctions, including the spatial alternation task in the T-maze and the delayed-matching-to-place version of the Morris Water Maze test (Brito and Thomas, 1981; Morris, 1984; Deacon and Rawlins, 2006). In the T-maze, rats or mice learn to alternate between the right and left arm of the maze to obtain a reward (Figure 15).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 81 GENERAL INTRODUCTION
Spatial alternation task in the T-Maze
TRAINING Reward
ABAB OR
Start Start
30 s TEST
ABAB Correct choice Error
Start Start
Figure 15. The spatial alternation task in the T-maze During training, mice or rats are presented to a reward (e.g. a food pellet) in one arm of the T-maze. After a short delay (for instance 30 s), the animals are submitted to a choice test in which they have to go into the opposite arm than training to obtain the reward (correct choice).
In the water maze procedure, animals learn to escape form water by reaching a platform hidden in a circular pool (Figure 16). The maze is surrounded by several visual cues, allowing animals to construct a spatial representation of the experimental room. After several trials, mice learn to locate the platform from these spatial cues so that both the latency and the distance to find the platform decrease across the training sessions. In the WM procedure (spatial-matching-to-sample or delayed- matching-to-place, Morris, 1984), the platform location is changed at each daily session, requiring a modification of directionality in relation to extra-maze cues to escape from water. At the first trial of a session, animals always reach the platform randomly. After a short delay (typically 30 s to 60 s), a second trial is performed. One consider that the task is learnt when both the latency and the distance to find the platform decrease between the first trial and the following trials within the training sessions.
Acute systemic administration of THC (3.0 to 5.0 mg/kg) impairs both T-maze alternation and water maze WM performances in trained rats and mice in a CB 1 receptor–dependent manner (Ferrari et al., 1999; Nava et al., 2000, 2001; Varvel et al., 2001, 2005a). Importantly, THC did not alter locomotion as assessed within the tasks, suggesting that these reported effects were not due to the well known hypo-locomotion induced by high doses of THC (Yao and Mackie, 2009).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 82 Part IV - CB1 receptors and memory
Delayed-matching-to-place in the Morris Water Maze
Day 1 Day 2 Day X
Start 3 Start 2 Start 1/4 Start 2
Start 1 Start 4 Start 3 Start 2/4 Start 1/3
Figure 16. The delayed-matching-to-place version of the Morris Water Maze paradigm An invisible platform (dashed circle) is fixed in a circular pool. Its position is changed at each training day. The pool is surrounded by visual cues (orange, green, purple and yellow symbols). Animals are released in the pool at different start points at each trial (excepted at the proximity of the platform location). The latency and distance to reach the platform are measured at each trial.
A key question in the pharmacological effects of cannabinoids on WM is to describe the brain mechanisms leading to this impairment. In rats, the DMTS deficit induced by THC was associated with a decrease in hippocampal population cell discharge as measured during the task (Hampson and Deadwyler, 2000). Accordingly, cannabinoids impair WM when injected intra-hippocampally; indicating that this effect is associated to CB 1 receptors stimulation within the hippocampus (Lichtman et al., 1995; Wise et al., 2009). In addition, THC-induced WM deficits in the delayed alternation T-maze task are associated with altered dopamine and noradrenalin turnover in the rat PFC (Jentsch et al., 1997).
Recently, the potent CB 1 receptor agonist CP-55940 was shown to disrupt theta-frequency coordination of CA1-mPFC activity during WM testing in rats, suggesting that disruption of CA1-mPFC network oscillations mediate the deleterious effects of cannabinoids in this task (Kucewicz et al., 2011). However, the cellular mechanisms implicated in the cannabinoid-induced WM impairments still need to be identified.
IV.3 The endogenous cannabinoid system controls aversive memories
In 2002, a landmark study showed that the endogenous activity of CB 1 receptors is necessary for extinction of conditioned freezing in mice (Marsicano et al., 2002). Since extinction of conditioned freezing is an important indicator of fear adaptation in animals and because our ability to control emotional responses is important to ensure adapted behaviors, this function of the ECS has generated a large interest. In the following sections, I will provide some pieces of information about the way of studying fear in laboratory, before reviewing the current knowledge on the modulation of fear by the ECS.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 83 GENERAL INTRODUCTION
IV.3.1 Fear and its regulation: theoretical accounts
IV.3.1.1 Definitions
Fear can be defined as a subjective unpleasant emotional state elicited by the presence of a threat (Ohman, 2000). Fear is a primary survival mechanism because it prepares the organism to effectively avoid potential dangers. Fear is accompanied by a number of physiological responses leading to the adoption of behavioral responses aiming at removing the threat.
It is important to note that fear is close to the concept of anxiety. According to the last version of the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV-TR, fourth edition, text revision), anxiety denotes an “ apprehensive anticipation of future danger or misfortune accompanied by a feeling of dysphoria or somatic symptoms ”. Indeed, both fear and anxiety can be regarded as emotional states. However, fear is proposed to differ from anxiety in having an identifiable eliciting stimulus (Ohman, 2000). Moreover, anxiety mostly denotes a pathological state, whereas fear is a natural reaction to threatening stimuli (Ohman, 2000).
Fear is a physiological emotion that helps avoiding present or future threats. However, if fear and fear responses do not adapt to the changes in the situations, the individuals’ ability to exert normal life activities may be compromised. Indeed, whereas the ability to produce fear responses is fundamental for individuals' survival, the control of fear and fear responses is also necessary to maintain normal mental and physical activities (Gross, 1998). Many authors used the terms of coping behaviors to define the behavioral repertoire engaged by individuals in aversive situations (Weiss, 1968; Lazarus and Folkman, 1984). Even though the precise definitions of coping behaviors differ among authors, Wechsler (1995) proposed that “ coping is a behavioral response aiming at reducing the effect of aversive stimuli on fitness or physiological measures related to fitness ”. Indeed, effective coping implies that the engaged responses successfully remove the aversive stimuli or decrease the physiological consequences of aversive stimuli that cannot be removed (Wechsler, 1995). As we will see below, coping behaviors are strongly subjected to individual differences (Wechsler, 1995; Koolhaas et al., 1999; De Boer and Koolhaas, 2003; Koolhaas et al., 2010).
IV.3.1.2 Studying fear in animals
Emotions in general and fear in particular are highly preserved across species evolution (Darwin, 1872; Ekman, 2003). Certain emotions are expressed similarly in people around the world, mostly independently of possible cultural transmissions. Moreover, certain emotions are expressed similarly across closely related species further suggesting that they are phylogenetically conserved. Thus, it is likely that emotion circuits are conserved across mammalian species (LeDoux, 2012). Hence, it should be possible to understand human emotions by exploring emotional mechanisms in the nonhuman mammalian brain. Following the definition proposed above, fear would be characterized by 3 main components: (i) physiological changes, (ii) behavioral responses, and (iii) affective experiences (feelings). For obvious reasons, it is not possible to scientifically measure feelings in animals other than humans. However, it is possible to assess animals’ emotional behaviors
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 84 Part IV - CB1 receptors and memory and their related physiological responses that allow them to deal with challenges in their environments. For instance, rodents faced with predator odours or stimuli that predict potential injury will adopt defensive responses (Bolles, 1970). In animal models, these threats can be innately recognized or learned (Blanchard and Blanchard, 1969; Bolles, 1970; McAllister et al., 1971; De Boer and Koolhaas, 2003). In particular, the fear conditioning paradigms, especially in rodent, have received large interests for studying the neuronal basis of fear.
IV.3.2 Aversive learning paradigms in animals
IV.3.2.1 Classical fear conditioning
a. Behavioral procedure
In classical fear conditioning, an initially neutral stimulus, that does not elicit observable emotional response by itself (for instance an auditory cue of mild intensity), is temporally associated to an aversive stimulus or unconditioned stimulus (US; such as the delivery of an electric footshock). Following this pairing, the presentation of the tone (conditioned stimulus, CS) in the absence of US becomes able to induce the fear responses (conditioned fear responses). In this task, subjects learn an association that allows a novel stimulus to become a "warning of danger", eliciting the defensive responses in anticipation of the inescapable danger (Pavlov, 1927; Blanchard and Blanchard, 1969; Fanselow, 1980; LeDoux, 2000). Classical fear conditioning can be induced in many species from invertebrate to mammals, including humans (Blanchard and Blanchard, 1969; LeDoux et al., 1984; LaBar et al., 1995). In rodents, several types of conditioned fear responses were described. These include autonomous arousal (e.g. an increase of heart rate and blood pressure), endocrine responses (hormone release), reflex potentiation and hypoalgesia. Conditioned fear is also expressed by several defensive behavioral responses (Bolles, 1970; Blanchard et al., 2001, 2003). In particular, animals readily respond to CS with the absence of any movements except those dedicated to breathing. This particular behavior is common to several species, but it is particularly strong in rodents and has been defined as "freezing" (Blanchard and Blanchard, 1969; Blanchard et al., 1976). Fear conditioning procedures produce rapid and robust learning as a single footshock can produce high levels of freezing that can be retained for months. Indeed, research from many laboratories used the classical fear conditioning procedure to study the neuronal mechanisms of fear and memory (Maren, 1996; LeDoux, 2000; Ehrlich et al., 2009; Hartley and Phelps, 2010). Importantly, in the early 1990’s, critical observation was made about the different roles of the hippocampus and amygdala in fear conditioning. Whereas the amygdala was found to be necessary for learning about both contextual (i.e. learning about where shocks were delivered) and discrete (i.e. cues) stimuli (Phillips and LeDoux, 1992), the hippocampus was found to have a selective role in fear to contextual stimuli (Kim and Fanselow, 1992; Phillips and LeDoux, 1992), although the putative necessity of the hippocampus in discrete fear conditioning is still highly debated (Calandreau et al., 2006; Trifilieff et al., 2006).
Once the conditioned freezing response is acquired, it is possible to inhibit it by a procedure called extinction. Extinction of conditioned freezing is induced by a prolonged or repeated CS
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 85 GENERAL INTRODUCTION presentation in the absence of the US (Pavlov, 1927; Rescorla, 1968; Cain et al., 2003). Extinction of conditioned fear is generally not considered as forgetting of the previously acquired CS-US association. Indeed, a previously extinguished fear response can recover with the passage of time (spontaneous recovery). Moreover, extinction is context-specific; if extinction is induced in a different context than acquisition, the freezing response is again high when the animals are re-exposed to the acquisition context (renewal). In the same way, if the US is again delivered after extinction completion, the conditioned fear response should reinstate only in the context in which US occurred (reinstatement; Myers and Davis, 2007). Therefore, extinction would be the consequence of a decreased contingency between the CS and the US. An important question in fear extinction mechanisms is its associative nature; although it is argued that extinction involves the formation of a new competing inhibitory CS-US association (new learning), non-associative mechanisms have been also proposed to account for the freezing inhibition observed following the extinction procedure (Myers and Davis, 2007). For instance, mice can exhibit strong freezing response to a neutral, unconditioned tone if they previously experienced a strong footshock in a distinct context (sensitized fear). The freezing response then declines during tone re-exposure in a way similar to that observed during extinction of conditioned freezing. In this sense, extinction of freezing would also involve habituation processes (Kamprath and Wotjak, 2004). This is an important point as CB 1 receptors have been particularly involved in this component of fear adaptation (Kamprath et al., 2006).
b. The conditioned freezing response
Rats or mice primarily tend to freeze in the absence of an escape route (Blanchard et al., 1976). The measurement of freezing behavior is non-invasive and performable inexpensively. Indeed, visual observation and scoring of freezing behavior is a reliable and commonly used index of conditioned fear. However, intense US or many conditioning trials have been shown to reduce the expression of freezing (Fanselow, 1984). It has been thus proposed that the level of fear determines the nature of the defensive behavior that is engaged in response to a threat following the “ predatory imminence continuum ” (Fanselow and Bolles, 1988). Accordingly, moderate levels of fear associated with a distal predator might induce freezing behavior to allow threat detection. However, particularly high levels of fear associated with contact with the predator might induce active defensive behaviors (fighting / escape attempt), thereby reducing freezing behaviors. Shock probability in fear conditioning settings might essentially model the predator distance in the “ predatory imminence continuum ” (Fanselow and Lester, 1988). Recently, selective inhibition of a subset of neurons within the CeA was shown to switch the response of conditioned mice form freezing to intense digging and rearing behaviors. This switch was interpreted as an increase of active coping behaviors (Gozzi et al., 2010). Hence, an absence or reduction of freezing associated with a neural manipulation may not necessarily imply a loss of fear, but rather a shift of the nature of the fear response engaged by the animal.
c. Neuronal mechanisms
The neurobiological literature on fear conditioning and fear extinction is impressively rich and comprises regional, cellular and molecular description of the brain mechanisms supporting these phenomena (LeDoux, 2000; Kim and Jung, 2006; Myers and Davis, 2007; Quirk and Mueller, 2007;
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 86 Part IV - CB1 receptors and memory
Ehrlich et al., 2009; Makkar et al., 2010; Sotres-Bayon and Quirk, 2010; Johansen et al., 2011). In sake of brevity, in this section, I will specifically focus on the mechanisms that will be useful for understanding the role of the ECS in the regulation of fear responses.
The amygdala is a critical centre for both the integration of relevant fearful stimuli and the organization of the neuronal outputs leading to the expression of the fear responses (Figure 17). The amygdalar activity is also important for proper extinction of conditioned freezing. In the case of auditory fear conditioning, both CS and US sensory inputs mainly converge to the lateral nucleus of the amygdala (LA) either directly from the somatosensory cortex or indirectly from the thalamus (LeDoux et al., 1990; Amaral and Insausti, 1992). Indeed, the LA is a site of convergence of CS and US information, and lesions confined to the LA abolish auditory fear conditioning (LeDoux et al., 1990). Accordingly, fear conditioning is accompanied by an enhancement of synaptic transmission at excitatory auditory input synapses in the LA. Auditory stimuli elicit field potentials in the LA of awake, freely behaving rats, and induction of LTP in the thalamic input to the LA leads to an enhancement of these auditory responses (Rogan and LeDoux, 1995). Additionally, fear conditioning in freely behaving rats results in a potentiation of excitatory field potentials recorded from LA (Rogan et al., 1997). These studies suggest that associative auditory fear learning occurs, at least in part, through changes in synaptic activity in the LA. The LA projects to the CeA, both directly and through the basal nucleus of the amygdala (BA). The intrinsic connexion between the LA and CeA involve complex local excitatory and inhibitory circuits (Ehrlich et al., 2009; Ciocchi et al., 2010). Damage to the CeA interferes with the expression of conditioned freezing (LeDoux et al., 1988; Ciocchi et al., 2010). The CeA is mainly composed of GABAergic neurons which are spatially and functionally organized to encode acquisition and expression of conditioned freezing. Indeed, the lateral subdivision of the CeA is necessary for acquisition of conditioned freezing, whereas its expression is mediated by GABAergic neurons of the medial subdivision of the CeA (Ciocchi et al., 2010). BA-restricted lesions do not prevent the acquisition or expression of conditioned freezing (Amorapanth et al., 2000; Nader et al., 2001). In turn, the CeA projects to brainstem areas that control the expression of conditioned fear responses, including autonomous, endocrine and behavioral (e.g. freezing) responses (Figure 17).
Amygdala nuclei are also under the control of other brain structures, including the mPFC and the hippocampus that are thought to ensure contextual, temporal and mnemonic modulations of conditioned fear expression (LeDoux, 2000). For instance, the activity of the prelimbic portion of the mPFC is necessary for conditioned freezing expression and has been proposed to participate in the encoding and integration of emotionally salient information (Laviolette et al., 2005; Quirk and Mueller, 2008).
Extinction of conditioned freezing also requires amygdalar activity. Its implication in extinction has been mainly shown by acute pharmacological inactivation of targeted nuclei since amygdalar lesion primarily prevents conditioned fear acquisition and expression (Myers and Davis, 2007). Inhibition of MAPK within the BLA prevented within-session freezing extinction (Herry et al., 2006) and both group I mGluR and NMDAR signallings in the BLA are also necessary for acquisition of extinction (Kim et al., 2007; Sotres-Bayon et al., 2007), suggesting that synaptic plasticity events occur in the
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 87 GENERAL INTRODUCTION
BLA during extinction. Morgan et al. (1993) observed in rats that mPFC lesions delayed extinction of conditioned freezing. Accordingly, the infralimbic (IL) region of the mPFC is a potential site of extinction consolidation, as IL lesions left within-session acquisition of extinction intact, but impaired extinction retrieval on the following day (Quirk et al., 2000). Electrophysiological evidence suggests that the IL inhibits the expression of conditioned freezing during extinction through reciprocal connections with the amygdala (Milad and Quirk, 2002; Quirk et al., 2003). Therefore, inhibition of conditioned freezing during extinction may occur through IL activation of the amygdala. Evidence also suggests that hippocampal projections to the mPFC and the amygdala mediate the context-dependent expression of extinction (Ji and Maren, 2005). Interestingly, Herry et al. (2008) could discriminate, in behaving mice, specific circuitries in the BA whose activity correlates with acquisition of extinction of auditory fear conditioning, or with high fear maintenance, respectively. These circuits depicted specific electrophysiological signatures, revealing distinct connexions with both the hippocampus and the mPFC. In particular, only “extinction neurons” are characterized by reciprocal connexions with the mPFC, emphasizing the importance of its bidirectional connections with the amygdala to ensure extinction (Milad and Quirk, 2002; Quirk et al, 2003; Herry et al., 2008). These results also suggest that close local microcircuits can switch between distinct behavioral states.
IV.3.2.2 Avoidance learning
a. Behavioral procedures
One of the most common behaviors correlated with the occurrence of aversive event is the tendency of organism to escape from stimuli associated with those events. Avoidance learning paradigms were used well before classical fear conditioning to study fear in laboratory animals (Estes and Skinner, 1941; Blanchard and Blanchard, 1969; LeDoux, 2012). Originally, Estes and Skinner (1941), developed the conditioned suppression of feeding behavior task in which food restricted rats trained to press a lever for food reinforcement decreased their rate of responding when a warning cue (i.e. CS) was presented and predicted the delivery of an electric shock (i.e. US). Later on, Sidman and colleagues observed that animals rapidly learn to engage behaviors to escape from a place where they were presented to a conditioned fear stimulus (active avoidance; Sidman, 1962). The symmetric procedure, passive avoidance, in which animals learn to avoid approaching behaviors toward a place that was previously associated with an aversive stimulus, was then established (Mineka, 1979). Indeed, avoidance learning is considered as the behavioral consequence of an instrumental (operant) conditioning in which a predictable aversive event (e.g. electric shock) does not occur contingent upon the occurrence or non-occurrence of a specified response (Bolles, 1970). In the active form, the avoidance contingency depends on the occurrence of a specific, on-going response; in the passive form, the avoidance contingency depends on the suppression of a specified response. In both forms the conditioned avoidance behavior results in the prevention of the punishment (Mineka, 1979). In each case, the conditioned avoidance response was taken as a measurement of fear learning (Estes and Skinner, 1941; Sidman et al., 1962; McAllister et al., 1971; Mineka, 1979).
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 88 Part IV - CB1 receptors and memory
Because conditioned avoidance tasks can be carried out in various ways (active, passive, signaled, unsignaled), each primarily involving the learning of a Pavlovian (CS-US) association and an instrumental association (Amorapanth et al., 2000; Cain and LeDoux, 2007; Choi et al., 2010; Lázaro- Muñoz et al., 2010), the behavioral complexity of avoidance conditioning was not fully understood. Thus, classical conditioning is the initial phase of avoidance conditioning. After the subjects rapidly undergo Pavlovian conditioning, they then learn avoidance responding using the CS as a warning signal. Failures to separate these components probably impeded the understanding of the brain mechanisms of conditioned avoidance. From the 1960’s, Pavlovian conditioning was considered as a more direct way to study fear processing, leading to its progressive success to assess the brain mechanisms of fear memories, at the depend of conditioned avoidance paradigms (LeDoux, 2012).
b. Neuronal mechanisms
Another reason accounting for the neglect of avoidance learning models is that no consensus about the role of the amygdala was found (Sarter and Markowitsch, 1985), likely due to the little appreciation of the anatomical complexity of the amygdalar subnuclei at that time (Amaral et al., 1992). In 2000, Amorapanth et al. performed selective electrolytic lesions on different amygdalar subnuclei in rats and evaluated the consequences of such lesions on both the freezing response induced by classical fear conditioning and active avoidance induced by the same CS-US pairing. Whereas lesions of the LA prevented acquisition of both conditioned freezing and active avoidance, CeA lesion specifically impaired acquisition of conditioned freezing, leaving active avoidance learning intact. Conversely, BA lesion had no effect on conditioned freezing whereas it blocked acquisition of active avoidance. Both LA and BA, but not CeA, are also necessary for the expression of already learned active avoidance responding (Choi et al., 2010). These findings indicate that the neural pathways within the amygdala mediating the ability of a CS to elicit conditioned freezing responses and those enabling active avoidance learning can be dissociated. The LA would be necessary for the acquisition and expression of the CS-US association; the LA-CeA projections would be part of the output system that responds to stimuli predicting danger by eliciting freezing response, whereas the LA-BA pathway would be part of an output system through which active fear responses are acquired and maintained to minimize exposure to a threatening stimulus (Amorapanth et al., 2000; Choi et al., 2010; Figure 17).
Individual variability in the rate of active avoidance learning has been often reported in rodents (Blanchard and Blanchard, 1969; Steimer et al., 1997; Koolhaas et al., 1999, 2010; Lázaro-Muñoz et al., 2010; Vicens-Costa et al., 2011; Díaz-Morán et al., 2012). Animals that show poor active avoidance performances tend to express persistent freezing responses even though freezing fails to avoid the aversive US, suggesting that conditioned freezing could be a competing defensive behavior that can interfere with active avoidance responses. Moreover, rat lines have been bred and behaviorally characterized on the basis of good and poor performances in active avoidance tasks, suggesting that the ability to acquire and perform active avoidance is a trait-like characteristic (Steimer et al., 1997). Interestingly, the CeA is not only dispensable for acquisition of active avoidance, but its focal lesion also rescued active avoidance learning in poor performer rats that expressed high levels
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 89 GENERAL INTRODUCTION of freezing during training, suggesting that CeA activity can constrain active avoidance learning in these animals (Lázaro-Muñoz et al., 2010; Choi et al., 2010). The amygdala sends neuronal projections to both dorsal and ventral portions of the striatum (Price, 2003). Recent fMRI studies in humans have correlated active avoidance learning with strong activation of the striatum, and it has been thus suggested that BA-striatal output would play a role in active avoidance learning (Schiller and Delgado, 2010; Delgado et al., 2009; Hartley and Phelps, 2010; Li et al., 2011; Figure 17).
CS
US
Sensory Thalamus LA
Somato -sensory CeA BA STRIATUM Cortex
PAG LH PVN Active avoidance
Freezing Blood Hormone pressure release
Figure 17. Representation of US and SC neuronal pathways in fear conditioning Adapted from Medina et al., 2002; Fanselow and Poulos, 2005; Hartley and Phelps, 2010.
IV.3.3 How CB 1 receptors modulate aversive memories
IV.3.3.1 Role of the endogenous CB 1 receptor signalling in learned fear responses
In classical cued fear conditioning, most of the published data indicate that the systemic pharmacological blockade of CB 1 receptors with SR141716A (3 mg/kg) or their genetic deletion in constitutive CB 1-KO mice deriving from C57BL/6N background induce little or no effect on acquisition or retrieval of the conditioned freezing response when tested 24 h after the conditioning phase (Marsicano et al., 2002; Cannich et al., 2004; Kamprath et al., 2006; Plendl and Wotjak, 2010).
However, another CB 1 antagonist (AM251, 3 mg/kg) administered prior to conditioning enhances acquisition of the freezing response in rat (Arenos et al., 2006). In contextual fear conditioning,
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 90 Part IV - CB1 receptors and memory administration of AM251 prior to conditioning is able to decrease conditioned freezing in rats (Arenos et al., 2006; Mikics et al., 2006). Similarly, constitutive CB 1-KO mice deriving from CD1 background (Ledent et al., 1999) does not display any freezing response when re-exposed to the conditioning context, suggesting that the ECS signalling is important for hippocampus-dependent fear conditioning (Mikics et al. 2006). However, SR141716A (1-10 mg/kg) does not disturb acquisition of the conditioned freezing response to context (Suzuki et al., 2004; Pamplona et al., 2006). Indeed, the endogenous role of CB 1 receptors in acquisition of the conditioned freezing response is not clear and the results seem to vary according to the experimental conditions. These results indicate different consequences following CB 1 receptor blockade using either SR141617A or AM251. Interestingly, both of these compounds can differentially affect glutamatergic and GABAergic transmissions in a species- dependent manner (i.e. rats versus mice; Haller et al., 2007), suggesting that these discrepancies could be explained, at least partially, by the different pharmacological properties of the drugs. Moreover, intense fear conditioning procedures can induce sustained freezing expression that might reach a ceiling effect in control animals, thus masking potential increase of freezing acquisition following manipulation of CB 1 receptor signalling.
However, the importance of the ECS in extinction of conditioned freezing is well accepted.
Indeed, it was first reported in 2002 that constitutive CB 1-KO mice failed in adapting their freezing response when exposed to repeated or prolonged CS presentation as compared to their wild-type littermates (Marsicano et al., 2002; Cannich et al., 2004; Kamprath et al., 2006). The acute injection of SR141716A (3 mg/kg) before extinction training in wild-type mice confirmed that eCBs, through the activation of CB 1 receptors, play a major role in extinction of cued conditioned freezing. Importantly, the necessity of the ECS signalling in extinction of conditioned freezing was later demonstrated for contextual fear conditioning (Suzuki et al., 2004; Pamplona et al., 2006), but also for other fear conditioned responses including analgesia, startle reflex potentiation and inhibitory avoidance (Finn et al., 2004a, 2004b; Chhatwal et al., 2005; Niyuhire et al., 2007). Moreover, increasing eCBs availability by administering the inhibitor of eCBs uptake and breakdown AM404, enhances extinction of the conditioned startle reflex in rats (Chhatwal et al., 2005). It is important to note that CB 1 receptors signalling is not necessary for extinction of conditioned response to appetitive stimuli, suggesting a specific involvement of the ECS in adaptation of conditioned aversive stimuli (Hölter et al., 2005; Niyuhire et al., 2007; Harloe et al., 2008).
The endogenous role of CB 1 receptors has been also evaluated in the two-way active avoidance task. This test is carried-out in a shuttle box composed of two identical compartments separated by a door. Animals learn to flee into the other compartment at the onset of a cue (e.g. a light or tone signal) to avoid a punishment (e.g. an electric footshock). CB 1-KO mice were shown to learn this task better than their wild-type littermates (Martin et al., 2002). However, Bura et al. (2007) were not able to replicate these data using a pharmacological approach. Additional studies are therefore needed to understand the role of the ECS in this task.
As mentioned previously, extinction of conditioned freezing can be attributed to both associative (new learning) and non-associative processes (habituation; Myers and Davis, 2007). In an attempt to
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 91 GENERAL INTRODUCTION
verify the role of CB 1 receptors in the non-associative component of extinction, Kamprath et al. (2006) reported that sensitized CB 1-KO mice by a strong footshock delivery, were impaired in within-session habituation to a neutral tone presented 24 h later, suggesting that the ECS mediates extinction of freezing at least in part through habituation-like processes. Interestingly, the specific deletion of CB 1 receptors on cortical glutamatergic neurons leads to a similar delayed habituation of the freezing response following sensitization, suggesting that the CB 1 receptors-mediated control of cortical glutamatergic transmission is important for the non-associative component of extinction (Kamprath et al., 2009). However, the contribution of CB 1 receptors on GABAergic and glutamatergic neurons in conditioned fear is unknown.
IV.3.3.2 Neuronal mechanisms of CB 1-dependent modulation of conditioned fear responses
Marsicano et al., (2002) reported that both AEA and 2-AG concentrations were increased in the amygdala of fear conditioned wild-type mice following the first CS re-exposure. Moreover, the impaired freezing extinction in mice was associated to the absence of LTDi in the BLA of CB 1-KO mice (see section II.4.3.1), pointing-out the crucial role of the amygdalar eCB tone in the process of extinction. While freezing extinction induces the activation of ERK1/2 and the phosphatase calcineurin in wild- type animals (Mansuy et al., 1998; Lin et al., 2001), this effect was strongly reduced in many brain regions of the CB 1-KO mice, especially in the BLA and in the prefrontal cortex (Cannich et al. 2004).
Furthermore, it was suggested that the facilitatory effect of the endogenous CB 1 receptor activation involved inhibition of CCK release from GABAergic neurons within the BLA (Chhatwal et al., 2009).
Recently, a distinct contribution of the CeA and the BLA in the CB 1 receptors-mediated modulation of freezing extinction was proposed (Kamprath et al., 2011). Intra-CeA administration of AM251 in mice impaired within-session extinction of freezing response 24 h following conditioning. Conversely, intra- BLA application of the same drug selectively blocked extinction of conditioned freezing on the subsequent days. These data were associated to a time-dependent facilitation of DSE and DSI in the CeA, suggesting that the ECS acts on both excitatory and inhibitory transmissions within the amygdala to ensure appropriate adaption of learned fear.
Cumulative evidence indicates that the mPFC is a critical site for CB 1 receptors-dependent modulation of conditioned fear. Exposition to an odor previously paired with a footshock strongly increases the bursting activity of a subpopulation of neurons in the mPFC receiving monosynaptic inputs from the BLA (Laviolette et al., 2005). In the same olfactory fear-conditioning procedure, a CB1 receptor antagonists applied into the mPFC blocked the acquisition of conditioned freezing. This effect was associated to an impairment of the associative firing of single neurons in the mPFC, but also to an altered LTP at BLA-prelimbic cortex synapses (Laviolette and Grace, 2006; Tan et al., 2010, 2011), indicating that CB 1 receptor transmission within the BLA-mPFC pathway is necessary for encoding olfactory fear conditioning. However, intra-mPFC administration of AM251 facilitated the acquisition of conditioned freezing and cardiovascular responses in a contextual fear conditioning procedure (Lisboa et al., 2010), suggesting that the control of CB 1 receptors into the mPFC may differ according to the
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 92 Part IV - CB1 receptors and memory nature of the conditioned stimuli. Nevertheless, the importance of the ECS signalling into the mPFC for extinction of conditioned responses has never been assessed.
The endogenous activity of CB 1 receptors in the hippocampus is necessary for extinction of contextual fear conditioning (de Oliveira Alvares et al., 2008) and CB 1 receptor blockade impaired the induction of LTP in CA1 pyramidal neurons of hippocampal slices that was prevented by selective blockade of GABA A receptors (Lin et al., 2011). These results suggest that activation of CB 1 receptors might mediate extinction of conditioned freezing to context by promoting induction of LTP via a GABA A receptor-mediated mechanism.
CB 1 receptors are abundant in the dorsal column of the PAG (dlPAG) that modulate defensive responses (Carrive et al., 1997, Figure 17). Enhancing eCBs availability into the dlPAG by local administration of AM404 attenuates the recall of conditioned freezing to context in a CB 1 receptors- dependent manner (Resstel et al., 2008). Interestingly, activation of CCK1 receptors inhibits
GABAergic synaptic transmission via activation of CB 1 receptors (Mitchell et al., 2011), suggesting that the ECS modulates conditioned freezing expression by regulating, at least in part, GABAergic transmission within the dlPAG. Taken together, these studies suggest a complex and region- dependent involvement of the endogenous CB 1 receptor signalling in the control of conditioned fear.
IV.3.4 CB 1 receptors in other forms of memories
Several studies reported that blocking the endogenous CB 1 receptor signalling resulted in better learning performances in different memory tasks including recognition memory and spatial memory. Recognition memory is based on the innate preference of rodents for exploring novel places, objects or congeners as compared to familiar ones. Two common protocols evaluating recognition memory are the object recognition and the social recognition tasks. Pharmacological blockade of CB 1 receptors resulted in a facilitation of short-term olfactory memory in a social recognition task, and reduced the deficits observed in aged rats and mice (Terranova et al., 1996; de Bruin et al., 2010). While pre-test administration of SR141716A (0.3 to 10 mg/kg) did not induce any effects in CD1 mice nor did post- training AM251 (2.5 and 5.0 mg/kg) in rats in the object recognition paradigm, the lowest dose of AM251 (1.0 mg/kg) significantly improved consolidation of recognition memory (Bura et al., 2007; Clarke et al., 2008; Bialuk and Winnicka, 2011). Altogether, these data suggest that, under certain conditions, decreasing the CB 1 receptor signalling can enhance consolidation of object recognition memory. Conversely, administration of exogenous cannabinoid agonists is known to impair consolidation of object recognition memory (Clarke et al., 2008; Puighermanal et al., 2009). This effect is mediated, at least in part, by CB 1 receptors located on GABAergic neurons through an NMDA receptors-dependent mechanism (Puighermanal et al., 2009). Interestingly, enhancing AEA availability, but not 2-AG, leads to object recognition memory failure (Busquets-Garcia et al., 2011; Pan et al., 2011), pointing-out a cellular and ligand-specific control of the ECS in recognition memory.
The most frequently used cognitive tests for assessing hippocampus-dependent reference memory in rodents are the spatial memory tasks. In the radial-maze, administration of the CB 1 receptor antagonist SR141716A before the acquisition phase improves rat performances (Lichtman,
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 93 GENERAL INTRODUCTION
2000). The same group confirmed that SR141716A injection before acquisition or retrieval of the same task is able to decrease the number of errors of the rats (Wise et al., 2007). Moreover, the effect of the
CB 1 antagonist was synergistic with that of an inhibitor of the acetylcholinesterase, suggesting that the ECS and the cholinergic system may interact in this form of learning. In the 8-arm-radial maze, rats are required to retrieve several elements of information during the test session; the four arms that were baited during the acquisition phase are to be avoided, whereas the four other previously blocked arms are to be visited. In these conditions, blockade of CB 1 receptors immediately after the acquisition phase could improve consolidation processes (Wolff and Leander, 2003). Furthermore, the role of the ECS has repeatedly been assessed in the long-term memory version of the Morris Water Maze. While
CB 1-KO mice learn the task as well as the wild-type controls, they are impaired in the reversal learning phase where the platform was moved to another location in the pool. Indeed, they repeatedly went to the previous location showing increased and non-adapted perseverance and a significant deficit in learning the new location, reflecting the involvement of the ECS in extinction and / or flexibility of spatial aversive learning in adult mice (Varvel and Lichtman, 2002; Varvel et al., 2005b). Importantly,
Albayram et al. (2011) reported that young, 2 months-old CB 1-KO mice showed improved acquisition of reference memory in the Morris Water Maze, but older, 12 months-old, CB 1-KO mice were severely impaired in both acquisition and reversal learning of the task. Old CB 1-KO mice also displayed an increased neuro-inflammation and astrocytes proliferation in the hippocampus, suggesting an age- related function of the ECS in spatial memory performances.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 94
Part V RESEARCH GOALS
CB 1 receptors are abundantly expressed in the brain where they are present in different cell populations. In neurons, endogenous activation of CB 1 receptors leads to a general inhibition of neurotransmitter release, including both GABA and glutamate release, the main inhibitory and excitatory neurotransmitters in the brain, respectively. CB 1 receptors are also present on astrocytes. The available literature indicates that these receptors participate in many astroglial functions, and, in particular, the endogenous signalling of astroglial CB 1 receptors facilitates excitatory transmission in the hippocampus. Indeed, understanding the cellular mechanisms by which CB 1 receptors control brain functions requires experimental tools to discriminate their particular roles in specific cell populations. Because electrophysiological approaches allow such discrimination, the implication of
CB 1 receptors on distinct synaptic plasticity processes within neuronal microcircuits have been characterized these last years.
Synaptic plasticity is considered as an important neuro-chemical foundation of learning and memory functions. Consistently, CB 1 receptors are key modulator of memory processing. In particular, exogenous activation of CB 1 receptors impairs working memory both in human and animals. In aversive memory settings, endogenous activation of CB 1 receptors is needed to ensure a proper adaptation of acquired fear responses. However, the relative contribution of specific cellular subtypes regulated by CB 1 receptors activity in memory processing has never been studied. The general aim of this thesis work is to characterize the brain cellular mechanisms by which CB 1 receptors control memory functions.
Using the constitutive and conditional mutagenesis of CB 1 receptors in mice, we analyzed the behavioral consequences of the lack of CB 1 receptors on particular cells of the brain in different memory tasks.
95 GENERAL INTRODUCTION
First, we focused on the description of the cellular mechanisms by which CB 1 receptors modulate working memory.
In particular, we addressed the cellular populations on which (i) endogenous CB 1 receptor signalling is necessary for working memory under basal conditions, and (ii) exogenous cannabinoids impair working memory performances. In this study, we used conditional CB 1-KO mice lacking CB 1 receptors on either forebrain GABAergic neurons or cortical glutamatergic neurons that were already characterized (GABA-CB 1-KO and Glu-CB 1-KO; see General Introduction section II.4.5). Moreover, a new CB 1-KO mouse line has been generated bearing an inducible deletion of CB 1 receptors on astrocytes. These “neuronal” and “astroglial” CB 1-KO mice were submitted to a spatial working memory (SWM) procedure of the Morris Water Maze paradigm in which their performances were analyzed before and after an acute systemic administration of THC. The cellular mechanisms by which CB 1 receptors mediate cannabinoid-induced SWM impairments were analyzed by our collaborators in this work. In vivo electrophysiological recordings in mice and rats hippocampus, combined with behavioral pharmacological approaches in wild type animals allowed us to describe a cellular pathway by which exogenous CB 1 receptor agonists impair SWM. This study represents the first part of the results presented in this thesis work (Part I CB1 receptors and working memory) and was published as referenced below:
Han J, Kesner P, Metna-Laurent M , Duan T, Xu L, Georges F, Koehl M, Abrous DN, Mendizabal-Zubiaga J, Grandes P, Liu Q, Bai G, Wang W, Xiong L, Ren W, Marsicano G, Zhang X (2012) Acute Cannabinoids Impair Working Memory through Astroglial CB1 Receptor Modulation of Hippocampal LTD. Cell 148, 1039-1050. Han J, Kesner P and Metna-Laurent M share the first authorship. Marsicano G and Zhang X share the senior authorship.
My contribution to this study was to perform the preliminary experiments to validate the delayed- matching-to-place protocol of the Morris Water Maze test and to reproduce the THC effects in wild- type mice (unpublished data). I performed and analyzed the behavioral characterization of the conditional CB 1-KO mice in this task as well as participated in the revisions of the manuscript.
Second, we studied the neuronal mechanisms involved in the regulation of learned fear responses by CB 1 receptors.
We first extensively analyzed the different behavioral responses determining fear coping strategies in classical fear conditioning and avoidance learning tasks in wild-type mice. We then characterized the contribution of CB 1 receptors in forebrain GABAergic neurons as well as cortical glutamatergic neurons in the control of fear coping by analyzing the behavioral phenotypes of the constitutive CB 1-KO mice, GABA-CB 1-KO and Glu-CB 1-KO mice in fear conditioning and active avoidance paradigms. Furthermore, we employed a local virus-mediated gene transfer strategy to restore CB 1 receptor expression in the amygdala of constitutive CB 1-KO mice in order to determine the sufficiently of amygdalar CB 1 receptors in the regulation of fear coping strategies. This study represents the second part of the results presented in this thesis work (Part II CB1 receptors and fear responses) and was published as referenced below:
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 96 Part V - Research goals
Metna-Laurent M , Soria-Gomez E, Verrier D, Conforzi M, Jégo P, Lafenêtre P, Marsicano G
(2012) Bimodal Control of Fear-Coping Strategies by CB 1 Cannabinoid Receptors. The Journal of Neuroscience, in press . Metna-Laurent M and Soria-Gomez E share the first authorship.
My contribution to this work was to design, perform and analyze the behavioral experiments as well as writing earlier and the present version of the manuscript.
.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 97
RESULTS
99
Part I
CB 1 RECEPTORS AND WORKING MEMORY
I.1 Résumé de l’article 1: Les cannabinoïdes perturbent la mémoire de travail par une modulation de la LTD hippocampique exercée par les récepteurs CB 1 astrogliaux
Les propriétés pharmacologiques des cannabinoïdes sont très nombreuses. Dans quelques pays comme le Canada ou le Royaume-Uni, des médicaments à base de THC (Sativex®, Marinol®) sont prescrits contre les douleurs neuropathiques, les nausées et vomissements, ainsi que comme stimulants de l’appétit. Le THC est également étudié pour ses qualités anti-convulsivantes et anti- inflammatoires. Cependant, la consommation de THC a des conséquences indésirables comme ses effets psychotropes. Comme nous l’avons vu, l’administration de cannabinoïdes chez l’homme et chez l’animal induit des déficits de mémoire de travail. La mémoire de travail est nécessaire pour le stockage temporaire et le traitement de l’information. Elle est à l’œuvre dans toutes nos tâches quotidiennes telles qu’entretenir une discussion, lire et comprendre un texte ou se repérer dans l’espace. Chez les rongeurs, l’administration aigüe de cannabinoïdes exogènes comme le THC ou le
HU210 diminue les performances de mémoire de travail spatiale par la stimulation des récepteurs CB 1 localisés dans l’hippocampe. Cependant, les mécanismes cellulaires impliqués dans ces effets des cannabinoïdes sont inconnus. Les récepteurs CB 1 sont abondants dans l’hippocampe et contrôlent l’activité des neurones glutamatergiques, GABAergiques ainsi que celle des astrocytes. Ces transmissions neuronales et astrocytaires sont plastiques et elles sont impliquées dans les processus de mémoire spatiale chez l’animal. L’objectif de cette étude était de déterminer les contributions de ces populations cellulaires dans les effets des cannabinoïdes exogènes sur la mémoire de travail spatiale.
Afin d’étudier les conséquences spécifiques de l’administration de cannabinoïdes sur ces types cellulaires, nous avons utilisé le modèle de mutation conditionnelle des récepteurs CB 1 chez les souris
101 Results
GABA-CB 1-KO et Glu-CB 1-KO, caractérisées par une délétion de ces récepteurs sur les neurones GABAergiques du prosencéphale et sur les neurones glutamatergiques corticaux. Nous avons généré une nouvelle lignée de souris mutantes caractérisée par une délétion inductible des récepteurs CB 1 sur les astrocytes (GFAP-CB 1-KO). Ces souris ont été générées par le croisement des souris de la 2 lignée CB 1-flox avec des souris de la lignée GFAP-CreERT (Hirrlinger et al., 2006). Chez ces souris, la protéine cre, associée au promoteur GFAP, est fusionnée avec un domaine de liaison des récepteurs aux œstrogènes muté et sensible au tamoxifène. En l’absence de tamoxifène, ce complexe de fusion maintient la protéine cre dans le cytoplasme. L’administration de tamoxifène chez les souris GFAP-CB 1-KO libère la cre du complexe de fusion, lui permettant d’entrer dans le noyau des cellules exprimant la protéine GFAP et d’exciser, par recombinaison homologue, la séquence codant pour les récepteurs CB 1 floxée. Une caractérisation anatomique par double immuno- histochimie dirigée contre les protéines CB 1 et GFAP observée par microscopie électronique au niveau de l’hippocampe a permis de confirmer la suppression des récepteurs CB 1 sur les astrocytes, et le maintien de l’expression de ces récepteurs sur les neurones chez les souris GFAP-CB 1-KO après le traitement au tamoxifène. Certaines expériences pharmacologiques ont été effectuées chez le rat adulte Sprague-Dawley. Dans cette étude, nous avons utilisé deux tests comportementaux de mémoire de travail spatiale: la procédure de delayed-matching-to-position dans le labyrinthe aquatique de Morris et le test d’alternation spatiale dans le labyrinthe en T. Des expériences d’électrophysiologie in vivo ont permis l’enregistrement des potentiels de champs post-synaptiques excitateurs ( field evoked post-synaptic potentials, fEPSP) au niveau de la région CA1 de l’hippocampe.
Les principaux résultats de cette étude sont les suivants : l’administration de cannabinoïdes (THC et HU210) induit une diminution durable des fEPSP du CA1 de l’hippocampe in vivo dont l’induction, mais non la maintenance, est dépendante de l’activité des récepteurs CB 1, indiquant que ces cannabinoïdes exogènes induisent une forme de LTD (CB-LTD). Cette CB-LTD est intacte chez les souris GABA-CB 1-KO et Glu-CB 1-KO. Cependant le THC n’induit aucune modification des fEPSP enregistrés chez les souris GFAP-CB 1-KO, indiquant que les cannabinoïdes induisent cette forme de
LTD par la stimulation des récepteurs CB 1 localisés sur les astrocytes et non sur ceux exprimés sur les neurones. Cette CB-LTD est dépendante de la signalisation des récepteurs NMDA et indépendante de celle des récepteurs mGluR de type 1. Cette CB-LTD est supportée par l’endocytose des récepteurs AMPA. Les expériences comportementales ont tout d’abord permis de reproduire les déficits de mémoire de travail spatiale induits par les cannabinoïdes administrés au niveau systémique ou directement dans la région CA1 de l’hippocampe, chez le rat et la souris de type sauvage. Nous avons observé que le THC altère les performances de mémoire de travail spatiale chez les souris
GABA-CB 1-KO et Glu-CB 1-KO. A l’inverse, les souris GFAP-CB 1-KO sont insensibles aux effets du THC sur cette forme de mémoire. De plus, les déficits de mémoire de travail spatiale induits par le THC sont compensés par la co-administration d’un antagoniste des récepteurs NMDA ainsi que part un inhibiteur de l’endocytose des récepteurs AMPA.
Ces résultats montrent que les récepteurs CB 1 exprimés sur les astrocytes, mais pas sur les neurones, sont nécessaires aux effets des cannabinoïdes exogènes sur la mémoire de travail
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 102 Part I - CB1 receptors and working memory spatiale. De plus, les cannabinoïdes induisent une forme de LTD au niveau des neurones du CA1 de l’hippocampe in vivo dont les mécanismes cellulaires sont similaires à ceux supportant les déficits de mémoires de travail observés par l’analyse comportementale, suggérant que les cannabinoïdes altèrent la mémoire de travail spatiale par l’induction d’une LTD dans l’hippocampe par la stimulation des récepteurs CB 1 astrogliaux.
Ce travail a été effectué en collaboration avec plusieurs équipes de recherche et a fait l’objet d’une publication dont les références sont les suivantes :
Han J, Kesner P, Metna-Laurent M , Duan T, Xu L, Georges F, Koehl M, Abrous DN, Mendizabal-Zubiaga J, Grandes P, Liu Q, Bai G, Wang W, Xiong L, Ren W, Marsicano G, Zhang X (2012) Acute Cannabinoids Impair Working Memory through Astroglial CB1 Receptor Modulation of Hippocampal LTD. Cell 148, 1039-1050. Han J, Kesner P et Metna-Laurent M partagent le statut de premier auteur. Marsicano G et Zhang X partagent le statut de dernier auteur.
Ma contribution à cette étude a été de réaliser les expériences préliminaires pour valider le protocole de delayed-matching-to-position dans le labyrinthe aquatique de Morris et reproduire les effets du THC chez des souris de type sauvage (données non publiées), la conduite des expériences et l’analyse des données relatives à la caractérisation comportementale des souris mutantes dans ce test, ainsi que la participation aux révisions du manuscrit. La version intégrale de cet article est présentée ci-après.
I.2 Article 1: Acute Cannabinoids Impair Working Memory through Astroglial CB 1 Receptor Modulation of Hippocampal LTD
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 103
Acute Cannabinoids Impair Working Memory through Astroglial CB1 Receptor Modulation of Hippocampal LTD
Jing Han,1,13 Philip Kesner,2,13 Mathilde Metna-Laurent,3,4,13 Tingting Duan,5,6 Lin Xu,5 Francois Georges,4,7 Muriel Koehl,3,4 Djoher Nora Abrous,3,4 Juan Mendizabal-Zubiaga,8 Pedro Grandes,8 Qingsong Liu,9 Guang Bai,10 Wei Wang,11 Lize Xiong,12 Wei Ren,1 Giovanni Marsicano,3,4,14,* and Xia Zhang2,14,* 1College of Life Sciences and Key Laboratory of Modern Teaching Technology, Shaanxi Normal University, Xian 710062, China 2University of Ottawa Institute of Mental Health Research at The Royal, Department of Psychiatry, and Department of Cellular and Molecular Medicine, Ottawa K1Z 7K4, Canada 3INSERM, Neurocentre Magendie 4University of Bordeaux, Neurocentre Magendie Physiopathologie de la Plasticite´ Neuronale, U862, F-33000 Bordeaux, France 5Key Lab of Animal Models and Human Disease Mechanisms, Kunming Institute of Zoology, Chinese Academy of Science, Kunming 650223, China 6School of Life Sciences, University of Science and Technology of China, Hefei 230027, China 7CNRS, Interdisciplinary Institute for Neuroscience, UMR 5297, Bordeaux 33000, France 8Department of Neurosciences, Faculty of Medicine and Dentistry, Basque Country University, Leioa 48940, Spain 9Department of Pharmacology and Toxicology, Medical College of Wisconsin, Milwaukee, WI 53226, USA 10Department of Neural and Pain Sciences, Dental School, Program in Neuroscience, University of Maryland, Baltimore, MD 21201, USA 11Department of Neurology, Tongji Hospital, Tongji Medical College, Huazhong University of Science and Technology, Wuhan, Hubai 430030, China 12Department of Anesthesiology, Xijing Hospital, Fourth Military Medical University, Xian 710032, China 13These authors contributed equally to this work 14These authors contributed equally to this work *Correspondence: [email protected] (G.M.), [email protected] (X.Z.) DOI 10.1016/j.cell.2012.01.037
SUMMARY INTRODUCTION
Impairment of working memory is one of the most The treatments of pain, nausea, seizures, ischemia, cerebral important deleterious effects of marijuana intoxica- trauma and tumors in humans and/or animals are some of the tion in humans, but its underlying mechanisms are potential therapeutic applications of derivatives of the plant presently unknown. Here, we demonstrate that Cannabis sativa (marijuana) or synthetic cannabinoids (Lem- the impairment of spatial working memory (SWM) berger, 1980; Robson, 2001; Brooks, 2002; Carlini, 2004; Hall and in vivo long-term depression (LTD) of synaptic et al., 2005). However, the potential therapeutic use of cannabis is limited by important side-effects associated with its use strength at hippocampal CA3-CA1 synapses, (Pacher et al., 2006). One of the major side effects of marijuana induced by an acute exposure of exogenous can- intoxication is the impairment of working memory in humans nabinoids, is fully abolished in conditional mutant (Ranganathan and D’Souza, 2006) and animals (Lichtman and mice lacking type-1 cannabinoid receptors (CB1R) Martin, 1996; Hampson and Deadwyler, 2000; Nava et al., in brain astroglial cells but is conserved in mice 2001; Varvel and Lichtman, 2002; Fadda et al., 2004; Hill et al., lacking CB1R in glutamatergic or GABAergic neu- 2004; Wise et al., 2009), but the cellular mechanisms of this rons. Blockade of neuronal glutamate N-methyl- effect are presently not known. D-aspartate receptors (NMDAR) and of synaptic Working memory is the ability to transiently hold and process trafficking of glutamate a-amino-3-hydroxy-5- information for reasoning, comprehension and learning, such methyl-isoxazole propionic acid receptors (AMPAR) as active thinking. Baddeley introduced a multicomponent model of human working memory with a central executive also abolishes cannabinoid effects on SWM and system responsible for information integration and coordination LTD induction and expression. We conclude that of two subsystems (Baddeley, 2003). One subsystem, the the impairment of working memory by marijuana phonological loop, stores the sound of language while the and cannabinoids is due to the activation of astroglial other subsystem, the visuo-spatial sketch pad, stores visual CB1R and is associated with astroglia-dependent (e.g., color) and spatial information (i.e., location). This theory hippocampal LTD in vivo. suggests a key role of spatial processing in working memory
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. 1039 performance. Spatial working memory (SWM) in humans RESULTS and animals requires online processing of information within many brain regions including the hippocampus (Hassabis Cannabinoids Induce In Vivo LTD at CA3-CA1 Synapses et al., 2007; Kesner, 2007). The hippocampal excitatory In vivo recordings of field excitatory postsynaptic potentials CA3-CA1 synapses, which connect glutamatergic axons of (fEPSP) from CA3-CA1 synapses in anesthetized rats revealed CA3 pyramidal neurons, including the ipsilateral Schaffer that an i.p. injection of HU210 (0.05 or 0.1 mg/kg), a potent collaterals and contralateral commissural fibers, with dendrites synthetic cannabinoid, or D9-tetrahydrocannabinol (THC, of CA1 pyramidal neurons (Witter and Amaral, 2004), have 5 mg/kg), the major psychoactive ingredient of marijuana, been proposed to play a key role in SWM (Rolls and Kesner, decreased fEPSP amplitude to approximately 40% of the 2006). baseline levels (Figures 1A and 1G). Similar results were Multiple forms of memory are likely subserved by activity- or obtained after an intra-CA1 infusion of HU210 (Figures S1A experience-dependent long-term potentiation (LTP) and depres- and S1C). In studies hereafter, animals received an i.p. injection sion (LTD) of synaptic strength (Malenka and Bear, 2004). of 0.05 mg/kg of HU210 or 5 mg/kg of THC if not otherwise Chronic exposure of rats to cannabinoids impairs both LTP stated. induction at CA3-CA1 synapses and hippocampal-dependent Cannabinoid-induced depression of synaptic transmission at SWM (Hill et al., 2004), suggesting a link between LTP impair- CA3-CA1 synapses in brain slices is not defined as LTD, ment and SWM impairment. This idea is supported by recent because it is fully reversed by application of CB1R antagonists data that knockout of the AMPAR GluR1 subunit impairs both 10 min after cannabinoid application (Chevaleyre et al., 2006; LTP induction at CA3-CA1 synapses and SWM (Sanderson Hajos et al., 2001; Kawamura et al., 2006). This indicates the et al., 2008). If LTP at CA3-CA1 synapses indeed contributes requirement of a continuous activation of CB1R for cannabinoid to SWM, LTD at these synapses may play a role in SWM impair- depression of transmission at CA3-CA1 synapses, a character- ment, because LTD could counteract LTP at the same synapses istic of transient synaptic depression but not of LTD (Chevaleyre (Han et al., 2011). et al., 2006). However, we observed that the decreased EPSP
Cannabinoid type-1 receptor (CB1R), one of the most abun- amplitude was blocked by injection of the selective CB1R dant G protein-coupled receptors in the brain (Herkenham antagonist AM281 (3 mg/kg, i.p.) (Cui et al., 2001) 10 min before, et al., 1990), is found in both GABAergic and glutamatergic but not 10 min after HU210 administration (Figures 1B and 1G), neurons in the hippocampal CA1 region (Herkenham et al., thus indicating LTD induction by cannabinoid exposure in vivo 1990; Kawamura et al., 2006; Marsicano and Lutz, 2006). Its (hereafter referred to as CB-LTD). This idea is further supported main neuronal action is to inhibit presynaptic neurotransmitter by two lines of evidence. First, while synaptic transmission release (Kano et al., 2009; Marsicano and Lutz, 2006). Indeed, depression can be transient (in min) or long-lasting (i.e., LTD cannabinoids can depress excitatory transmission at CA3-CA1 lasting > 24 h), a HU210 injection (0.1 mg/kg, i.p.) induced synapses in brain slices via activation of CB1R(Misner and CB-LTD at CA3-CA1 synapses for > 24 hr in freely moving rats Sullivan, 1999; Hajos et al., 2001; Kawamura et al., 2006; (Figures 1E and 1G), at a time where the acute effects of the Marsicano and Lutz, 2006; Takahashi and Castillo, 2006; Bajo drug should be decreased. Second, while the maintenance of et al., 2009; Serpa et al., 2009; Hoffman et al., 2010). Thus, late-phase LTD, but not early-phase LTD or transient synaptic cannabinoid-induced decrease of excitatory transmission might transmission depression, requires new protein synthesis (Kel- be related to SWM impairment. It is entirely unknown, however, leher et al., 2004), administration of inhibitors of protein transla- whether cannabinoids are able to induce LTD at CA3-CA1 tion (anisomycin, 18 mg/kg, i.p.) (Puighermanal et al., 2009) synapses in living animals and whether such in vivo LTD might or RNA transcription (actinomycin-D, 72 mg/12 ml, i.c.v.) contribute to SWM impairment induced by exogenous cannabi- (Manahan-Vaughan et al., 2000) 2 hr before HU210 injection noids. In addition to the presence in neurons, CB1R is also found selectively reversed the late-phase expression of CB-LTD in hippocampal astroglial cells and its activation, by stimulating (Figures 1C and 1G). 2+ Ca -dependent release of glutamate, potentiates synaptic To identify if CB1R expressed in the CA1 area contributes transmission at CA3-CA1 synapses in brain slices (Navarrete to CB-LTD at CA3-CA1 synapses, we applied adenoviral and Araque, 2010). However, the roles of astroglial CB1Rin vectors-containing shRNA against CB1R into the CA1 region the modulation of behavior and synaptic plasticity in living 4 days prior to HU210 injection. shRNA CB1R specifically animals are not known. knocked down CA1 expression of CB1R(Figure 1F) and sup- In this study, we employed conditional mutagenesis, in vivo pressed CB-LTD at CA3-CA1 synapses (Figures 1D and 1G). electrophysiology and behavioral tests to study the mechanism Interestingly, the cannabinoid effect seems to be specific for underlying the effect of cannabinoids on hippocampal-depen- the CA3-CA1 pathway, because systemic HU210 did not induce dent SWM. Surprisingly, we found that activation of astroglial CB-LTD at synapses of the perforant path onto dentate gyrus
CB1R, but not neuronal CB1R, by exogenous cannabinoids neurons (Figures S1B and S1C). Thus, in vivo cannabinoid expo- mediates SWM impairment and LTD induction at CA3-CA1 sure induces an in vivo LTD at CA3-CA1 synapses. synapses in vivo. Our data reveal an unanticipated hippocampal pathway linking astroglial activity, synaptic plasticity and Neuronal CB1R Is Dispensable for CB-LTD at CA3-CA1 memory processing, and define the specific mechanisms likely Synapses underlying cannabinoid-induced impairment of SWM in living Glutamatergic presynaptic membranes of CA3-CA1 synapses animals. contain CB1R(Kawamura et al., 2006). To test whether CB-LTD
1040 Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. Figure 1. Cannabinoids Induce In Vivo LTD at CA3-CA1 Synapses (A–E) Plots of normalized fEPSP slopes in anes- thetized rats (A–D) or freely moving rats (E) show that cannabinoid injection at 0 min elicits CA1 LTD lasting for > 2 hr (A–D) or > 24 hr (E), which is blocked by AM281 administration 10 min before, but not 10 min after, HU210 injection (B), or by
intra-CA1 infusion of shRNA CB1R (D), and that anisomycin (An) and actinomycin-D (AMD) selectively reverse the late-phase expression of HU210-elicited LTD (C). Representative fEPSP traces before (1) and after (2) vehicle or cannabi- noid injection are shown above each plot. (F) Graph (top) and immunoblotting photos
(bottom) show a reduction of CA1 CB1R expres-
sion by shRNA CB1R. (G) Histogram summarizes the average percent change of fEPSP slope before (1) and after (2) vehicle or cannabinoid injection as depicted in panels (A)–(E). All summary graphs show means ± standard error of the mean (SEM); n = numbers of animals re- corded in each group (A–E) or numbers of experi- ments conducted (F) in each group. *p < 0.01 versus vehicle control, Bonferronni post-hoc test
after one-way ANOVA (A: F3,13 = 56.560, p < 0.01;
B: F3,10 = 39.001, p < 0.01; C: F3,8 = 47.210, p <
0.01; F: F2,6 = 34.990, p < 0.01) or t test. See also Figure S1.
littermates (Figures 2A and 2C). We then determined the induction of CB-LTD in mutant mice carrying a selective deletion
of the CB1R gene in brain GABAergic neurons (GABA-CB1R-KO), including CA1 GABAergic neurons (Monory et al., 2006; Bellocchio et al., 2010). Again, THC induced a CB-LTD at CA3-CA1 synapses that was indistinguishable between wild-type mice and GABA-
CB1R-KO littermates (Figures 2A and 2C). Thus, CB1R expressed in glutama- tergic or GABAergic neurons does not participate in this in vivo form of CB-LTD in the hippocampal CA1 region.
Astroglial CB1R Mediates CB-LTD at CA3-CA1 Synapses
CB1R is also functionally expressed in CA1 astrocytes (Navarrete and Araque,
2008). Therefore, astroglial CB1R might play a role in CB-LTD at CA3-CA1 synapses. To directly address this issue, depends on ‘‘glutamatergic’’ CB1R, we examined mutant mice we generated tamoxifen-inducible conditional mutant mice carrying a selective deletion of the CB1R gene in cortical and specifically lacking CB1R expression in astrocytes. ‘‘Floxed’’ hippocampal glutamatergic principal neurons (Glu-CB1R-KO) CB1R mutant mice (Marsicano et al., 2003) were crossed with (Monory et al., 2006; Bellocchio et al., 2010). Surprisingly, THC transgenic mice expressing the inducible version of the Cre induced a CB-LTD at CA3-CA1 synapses that was recombinase CreERT2 under the control of the promoter of indistinguishable between wild-type mice and Glu-CB1R-KO the human glial fibrillary acidic protein (GFAP-CreERT2 mice,
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. 1041 Figure 2. Cannabinoids Elicit CA1 LTD via Astroglial CB1R but Not Neuronal CB1R
(A and B) Plots of normalized fEPSP slopes in anesthetized mice show that THC injection at 0 min elicits CA1 LTD in wild-type (WT), Glu-CB1R-KO and GABA-
CB1R-KO mice (A), but not in GFAP-CB1R-KO mice (B). Representative fEPSP traces before (1) and after (2) treatment are shown above each plot. (C) Histogram summarizes the average percent changes of fEPSP slope before (1) and after (2) treatment.
(D and E) Histograms summarize the percentage of CB1R-labeled astrocytes and axons/terminals in GFAP-CB1R-WT mice, GFAP-CB1R-KO mice and CB1R-KO mice.
(F) Electron microscopic images show a high density of CB1R immunopositive silver grains (small arrows) in axons/terminals of both tamoxifen-treated GFAP-
CB1R-WT and GFAP-CB1R-KO mice, and a low density of silver grains (large arrow) in DAB-stained astrocytes (arrowheads) of GFAP-CB1R-WT mice but not of
GFAP-CB1R-KO littermates. The scale bar represents 500 nm.
(G) An electron microscopic image shows an absence of CB1R immunopositive silver grains in astrocytes stained with peroxidase/DAB and axons. The scale bar represents 500 nm. All summary graphs show means ± SEM; n = numbers of animals recorded (A, B) or numbers of positive immunoreactive profiles counted (D, E) in each group. *p <
0.01 versus control, Bonferronni post-hoc test after one-way ANOVA (A: F2,6 = 68.603, p = 0.884; B: F2,8 = 42.009, p < 0.01) or square Chi test (D).
1042 Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. Figure 3. Cannabinoids Induce NMDAR- Dependent LTD at CA3-CA1 Synapses (A–E) Plots of normalized fEPSP slopes in anes- thetized rats are presented with representative fEPSP traces (above plots) before (1) and after (2) vehicle or drug injection. An i.c.v. injection of TBOA induces LTD (A). E4CPG, but not vehicle, blocks LTD induced by DHPG injection at 0 min (B) without significant effects on LTD induced by HU210 (C). Intra-CA1 application of AP-5 suppresses HU210-induced LTD (D). Systemic administration of Ro25-6981 and ifenprodil, but not NVP-AAM077, prevents HU210-induced LTD (E). (F) Histogram summarizes the average percent change of fEPSP slope before (1) and after (2) drug or vehicle injection. All summary graphs show means ± SEM; n = numbers of animals recorded in each group. *p < 0.01 versus control, Bonferronni post-hoc test
after one-way ANOVA (B: F2,7 = 36.090, p < 0.01;
E: F3,10 = 40.409, p < 0.01) or t test.
Mechanisms of CB-LTD at CA3-CA1 Synapses Cannabinoids are able to activate hippo-
campal astroglial CB1R to increase extracellular glutamate levels (Navarrete and Araque, 2008). If a similar mechanism is involved in CB-LTD, LTD should be induced by the glutamate-uptake inhibitor DL-threo-b-benzyloxyaspartate (TBOA). Indeed, an i.c.v. injection of TBOA (10 nmol) (Wong et al., 2007) induced in vivo LTD at CA3-CA1 syn- apses (Figures 3A and 3F). If increased extracellular levels of glutamate induce LTD at CA3-CA1 synapse, postsyn- aptic metabotropic glutamate receptor (mGluR) may be responsible for this LTD induction, because postsynaptic mGluR activation produces LTD (Chevaleyre et al., 2006; Lovinger, 2008). However, the selective group I/group II mGluR
Hirrlinger et al., 2006) to eventually obtain the GFAP-CB1R-KO antagonist ethyl-4-carboxyphenylglycine (E4CPG, 35 nM/ mouse line. As compared to tamoxifen-treated wild-type 3.5 ml, i.c.v.) completely blocked in vivo LTD induced by the littermate controls (GFAP-CB1R-WT), GFAP-CB1R-KO mice group I mGluR agonist dihydroxyphenylglycine (DHPG, displayed a 79% reduction (p < 0.01) in the number of CA1 100 nM/5 ml, i.c.v.), but did not alter CB-LTD (Figures 3B, 3C, astrocytes labeled with a CB1R antibody (Figures 2D and 2F), and 3F). Surprisingly, CB-LTD was fully blocked by the selective whereas only background levels were observed in constitutive NMDAR antagonist AP-5 (50 mM, intra-CA1 iontophoretic
CB1R-KO mice (Figures 2E and 2G). Conversely, no difference ejection at À20 nA for 10 min) (Maalouf et al., 1998)(Figures (p = 0.2293) was observed between GFAP-CB1R-WT and 3D and 3F), and by the NR2B-preferring NMDAR antagonists GFAP-CB1R-KO mice in the number of CB1R-labeled CA1 Ro25-6981 (6 mg/kg, i.p.) (Fox et al., 2006) and ifenprodil neuronal axons/terminals (Figures 2D and 2F). THC elicited (5 mg/kg, i.p.) (Higgins et al., 2005)(Figures 3E and 3F). How- CB-LTD at CA3-CA1 synapses in tamoxifen-treated wild-type ever, the NR2A-preferring NMDAR antagonist NVP-AAM077 mice but not in GFAP-CB1R-KO mutant littermates (Figures 2B (1.2 mg/kg, i.p.) (Fox et al., 2006) did not alter CB-LTD in the and 2C). Therefore, cannabinoid exposure in vivo elicits same conditions (Figures 3E and 3F). Thus, in vivo cannabinoid
CB-LTD at CA3-CA1 synapses through CB1R expressed in exposure induces CB-LTD at CA3-CA1 synapses via activation astroglial cells. of NR2B-containing NMDAR.
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. 1043 Figure 4. Cannabinoids Induce AMPAR Endocytosis-Dependent Expression of CA1 LTD (A and B) Graphs and immunoblotting (bottom photos) show a decrease of GluR1 and GluR2 at the synaptic surface of CA1 neurons after HU210 injection, which is blocked by pretreatment with Tat-GluR2 but not Tat-GluR2S. (C) Plot of normalized fEPSP slopes in anes- thetized rats shows that injection of Tat-GluR2, but not Tat-GluR2S, 2 hr before HU210 injection at 0 min blocks HU210-induced LTD. Representative fEPSP traces before (1) and after (2) HU210 injection are shown above the plot. (D) Histogram summarizes the average percent change of fEPSP slope before (1) and after (2) HU210 injection (C) or Tat-GluR2 injection (E). (E) Plot of normalized slopes of fEPSPs in anes- thetized rats shows both naive rats and rats receiving Tat-GluR2 injection at 0 min display similar fEPSPs at CA3-CA1 synapses for 4 hr. Representative fEPSP traces recorded during À10–0 min (1) and 230–240 min (2) are shown below the slopes. All summary graphs show means ± SEM; n = numbers of experiments conducted (A and B) or numbers of animals recorded (C and E) in each group. *p < 0.05 versus control, t test.
(Figure 4B) in the CA1 and CB-LTD (Fig- ures 4C and 4D). Tat-GluR2 (1.5 mmol/kg, i.p.) did not significantly change the fEPSP amplitude at CA3-CA1 synapses for 4 hr after injection (Figures 4D and 4E). Altogether, these data strongly sug- gest that postsynaptic endocytosis of GluR1/GluR2 mediates the expression of CB-LTD at CA3-CA1 synapses.
Cannabinoid Impairment of Working Memory Shares the Same Mechanisms of CB-LTD CB-LTD is characterized by (1) activa-
tion of astroglial CB1R, (2) activation of NMDAR, and (3) internalization of AMPAR. These mechanisms were as- The expression of NMDAR-mediated LTD requires facilitated sessed in different behavioral models of cannabinoid impair- endocytosis of postsynaptic AMPAR (Collingridge et al., 2010). ment of spatial working memory (SWM).
AMPAR in CA1 pyramidal cells consists of 81% of GluR1/ The role of astroglial CB1R in cannabinoid impairment of GluR2 at synaptic membranes (Lu et al., 2009). The surface SWM was assessed by examining SWM performance of levels of GluR1/GluR2 in synaptosomes isolated from the CA1 tamoxifen-treated GFAP-CB1R-WT and GFAP-CB1R-KO litter- region significantly decreased after HU210 injection (Figure 4A), mates with a delayed-matching-to place (DMTP) version of the suggesting endocytosis of AMPAR in postsynaptic CA1 pyra- Morris water maze test (Steele and Morris, 1999). No significant midal cells following cannabinoid exposure in vivo. The adminis- differences were observed between wild-type and mutant tration of the brain-penetrating version of a peptide able to littermates during training (Figure S2A). In agreement with a block GluR2 endocytosis (‘‘Tat-GluR2’’ peptide, 1.5 mmol/kg, previous study (Varvel and Lichtman, 2002), THC impaired i.p.), but not of its scrambled analog (Tat-GluR2S) (Brebner SWM performance in GFAP-CB1R-WT mice, as evidenced by et al., 2005; Wong et al., 2007; Collingridge et al., 2010), specif- a significant decrease of both latency saving ratios (Figure 5A) ically blocked both HU210-induced GluR1/GluR2 endocytosis and path saving ratios (Figure 5B). In contrast, THC did not
1044 Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. Figure 5. Astroglial CB1R, NMDAR, and AMPAR Mediate Cannabinoid Impairment of SWM (A–D) Mouse DMTP version of the Morris water maze test. THC reduces both latency saving ratio (A) and path saving
ratio (B) in wild-type mice (A – D) and GABA-CB1R-KO
littermates but not in GFAP-CB1R-KO littermates. While
vehicle-treated Glu-CB1R-KO littermates show a signifi- cant decrease of both latency saving ratio and path saving ratio relative to vehicle-treated wild-type mice (C and D), THC reduces latency saving ratio (C) but not path saving
ratio (D) in Glu-CB1R-KO littermates. (E and F) Rat DMTP version of the Morris water maze test. HU210 reduces both path saving ratio (E) and latency saving ratio (F), which are prevented by i.p. pretreatment with Ro25-6981 or Tat-GluR2, while neither Ro25-6981 nor Tat-GluR2 significantly affects the ratio in the absence of HU210. All summary graphs show means ± SEM; n = numbers of animals tested in each group. *p < 0.05 versus control, Bonferronni post-hoc test after repeated-measure two-
way ANOVA ([A] F1,22 = 13.010, p < 0.01; [B] F1,22 = 7.999,
p < 0.01; [C] treatment: F1,30 = 37.28, p < 0.001; genotype x
treatment F2,30 = 2.92, p > 0.05; [D] treatment: F1,30 =
30.01, p < 0.001; genotype x treatment F2,30 = 4.25,
p < 0.05) or one-way ANOVA ([E] F5,36 = 19.307, p < 0.01;
[F] F5,36 = 13.110, p < 0.01). See also Figures S2 and S5.
pretreated with Ro25-6981 or ifenprodil, two NR2B-preferring NMDAR antagonists, or NVP- AAM077, a NR2A-prefering NMDAR antag- onist. The results show that NR2B- but not produce significant effects on GFAP-CB1R-KO littermates NR2A-preferring NMDAR antagonists abrogated HU210-in- (Figures 5A and 5B). While Glu-CB1R-KO littermates showed duced impairment of SWM performance (Figure 6A). Thus, a significant impairment of the acquisition of SWM (Figure S2B) activation of NR2B-containing NMDAR is necessary for the and subsequent poor performance of SWM in comparison cannabinoid-induced impairment of SWM. with wild-type mice (Figures 5C and 5D), THC impaired SWM The effects of the blockade of AMPAR internalization on performance (Figure 5C). Both GABA-CB1R-KO littermates and cannabinoid-induced SWM impairment was also tested in the control wild-type mice showed similar acquisition of SWM DNMTST paradigm. After 6 daily training sessions (Figure S3B), (Figure S2B), and THC impaired SWM performance (Figures rats received Tat-GluR2 or Tat-GluR2S (1.5 mmol/kg, i.p.) (Breb- 5C and 5D). THC treatment did not alter swim speed of GFAP- ner et al., 2005; Wong et al., 2007) 2 hr before HU210 injection
CB1R-WT and GFAP-CB1R-KO mice (Figure S2C), but slightly on each of the two testing days. Tat-GluR2, but not Tat-GluR2S, decreased this parameter in Glu- and GABA-CB1R-KO mice abolished HU210 impairment of SWM performance (Figure 6B). and WT littermates (Figure S2D). However, this slight effect To determine the specific role of the CA1 region, after 6 daily was equal for all genotypes (Figure S2D) and was equally distrib- training sessions (Figure S3C), Tat-GluR2 or Tat-GluR2S was uted among different trials (data not shown), thereby excluding infused bilaterally within the dorsal CA1 region (15 pmol/per its involvement in the altered SWM performance of the mice. injection) (Brebner et al., 2005; Wong et al., 2007)(Figure 6C)
Thus, CB1R in glutamatergic neurons, but not CB1Rin 60 min before each HU210 injection on each testing day. Intra- GABAergic neurons or astroglial cells, is necessary for mice to CA1 infusion of Tat-GluR2, but not Tat-GluR2S, blocked HU210 acquire SWM. Notably, however, astroglial CB1R, but not gluta- impairment of SWM performance (Figure 6D). Neither systemic matergic or GABAergic neuronal CB1R, is necessary to produce nor intra-CA1 administration of Tat-GluR2 significantly affected the detrimental effects of THC on SWM. basal locomotor activity, anxiety level or motor balance (Figures To test if NMDAR activation plays a role in cannabinoid im- S4A–S4E). Thus, AMPAR internalization in the CA1 hippocampal pairment of SWM, rats were tested in a T-maze using a delayed region is necessary for cannabinoid-induced alteration of SWM. nonmatching to sample protocol (DNMTST) (Kelsey and If intra-CA1 infusion of HU210 is able to induce CB-LTD Vargas, 1993). After 6 daily training sessions to ensure that at CA3-CA1 synapses (Figures S1A and S1C), a bilateral the task was mastered (>80% correct choices, Figure S3A), intra-CA1 infusion of HU210 should impair SWM. As expected, rats received 2 daily test sessions 30 min after injection of after six daily training sessions (Figure S3D), HU210 (0.1 mg/ HU210 or vehicle. Ten min before HU210 injection, rats were 0.5 ml/side) impaired rat SWM performance (Figure 6E).
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. 1045 Figure 6. NMDAR and AMPAR Mediate Cannabinoid Impairment of SWM (A) Rat DNMTS T-maze. HU210 suppresses SWM performances, which is prevented by i.p. pretreatment with Ro25-6981 and ifenprodil, but not with NVP- AAM077. (B and D) Rat DNMTS T-maze. Systemic (B) and intra-CA1 administration (D) of Tat-GluR2, but not Tat-GluR2S, blocks HU210 impairment of SWM performance. (C) Photograph (left) shows location of an intra-CA1 cannula, and histograms (right) show reconstructions of histology sections illustrating CA1 injection sites of Tat-GluR2 (solid circle) and Tat-GluR2S (open circle). (E) Intra-CA1 injection of HU210, but not vehicle, impairs SWM performance. All summary graphs show means ± SEM; n = numbers of animals tested in each group. *p < 0.01 versus control, Bonferronni post-hoc test after one-way ANOVA
(A: F5,36 = 59.070, p < 0.01; B: F3,28 = 54.220, p < 0.01; D: F3,32 = 41.562, p < 0.01; E: F1,12 = 36.090, p < 0.01). See also Figures S3 and S4.
Finally, we tested if the results obtained with the DNMTST SWM, is due to activation of astroglial CB1R. Furthermore, paradigm were reproducible with the DMTP water maze para- a novel form of cannabinoid-induced long-term synaptic plas- digm. One day after five daily training sessions to establish the ticity in the hippocampus appears to mechanistically underlie baseline levels of SWM (Figure S5A), rats received a test session this effect of cannabinoids in vivo. Our results are consistent of four trials. HU210 treatment before the test session impaired with a scenario (Figure 7), in which cannabinoid exposure in vivo
SWM performance, which was blocked by pretreatment with activates astroglial CB1R to increase ambient glutamate, which Ro25-6981 or Tat-GluR2 (Figures 5E and 5F). Neither Ro25- in turn activates NR2B-containing NMDAR to trigger AMPAR 6981 nor Tat-GluR2 administration alone significantly changed internalization at CA3-CA1 synapses. These events ultimately saving ratios (Figures 5E and 5F), suggesting that neither induce CB-LTD at these synapses, altering the function of hippo- NR2B-preferring NMDAR antagonists nor Tat-GluR2 interferes campal circuits that likely become unable to process SWM with basal SWM performance. Swim speeds during the SWM (Figure 7). task were not influenced by different treatments (Figure S5B). Early studies demonstrate that CB1R is expressed at high Thus, cannabinoid administration alters SWM performance in levels by neurons throughout the whole brain (Herkenham different behavioral tasks through the same mechanisms. et al., 1990; Matsuda et al., 1993; Tsou et al., 1998). More recent
Altogether, these data show that the same mechanisms studies show that CB1R is more abundant in GABAergic inter- underlying CB-LTD at hippocampal CA3-CA1 synapses (activa- neurons than in glutamatergic principal neurons (Kawamura tion of astroglial CB1R, activation of NMDAR and removal of et al., 2006). In the hippocampal CA1 area, CB1R density on AMPAR from the synaptic surface) also mediate cannabinoid- GABAergic presynaptic membranes is at least 10–20 times induced alterations of hippocampal-dependent SWM. higher than that on glutamatergic presynaptic membranes (Kawamura et al., 2006; Bellocchio et al., 2010). Cannabinoid DISCUSSION depression of in vitro excitatory or inhibitory synaptic transmis-
sion has been consistently shown to require CB1R in either This study shows that one of the most common effects of canna- glutamatergic or GABAergic presynaptic terminals, respectively binoid intoxication in humans and animals, the impairment of (Misner and Sullivan, 1999; Chevaleyre et al., 2006; Kawamura
1046 Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. that CB1R is expressed and quantifiable in hippocampal astro- cytes. We have further showed here that in vivo CB-LTD at CA3-CA1 synapses was not detectable in tamoxifen-inducible
conditional mutant mice specifically lacking CB1R expression in astrocytes (i.e., GFAP-CB1R-KO littermates). Our results strongly suggest a requirement of astroglial CB1R for CB-LTD at CA3-CA1 synapses in living animals. However, we also found that THC exposure in vivo did not
significantly alter basal synaptic transmission in GFAP-CB1R- KO littermates. These data, together with the finding that the
density of presynaptic CB1R at CA3-CA1 synapses is just above the background levels (Kawamura et al., 2006), suggest a negli-
gible role of presynaptic CB1R in excitatory transmission in vivo at CA3-CA1 synapses in response to exogenous cannabinoid exposure. Thus, in vitro cannabinoid application decreases excitatory synaptic transmission at CA3-CA1 synapses via acti-
vation of ‘‘glutamatergic’’ CB1R, whereas in vivo cannabinoid administration induces CB-LTD via astroglial CB1R without significant effects on presynaptic CB1R. The exact reason for this apparent mechanistic discrepancy between in vitro and in vivo effects of cannabinoids on synaptic transmission and plasticity is not known. Nevertheless, it is important to note that intact astroglial networks play prominent roles in brain functioning (Giaume et al., 2010). Indeed, astrocytes are more Figure 7. Proposed Model for In Vivo LTD Production at CA3-CA1 associated in networks than neurons due to the presence Synapses and Subsequent Working Memory Impairment of high levels of gap junctions and direct intercellullar com- CB1R exists in CA1 astrocytes (Figures 2D–2G) and presynaptic membranes munications (Giaume et al., 2010). It is therefore possible that with 10- to 20-fold of CB1R density in GABAergic membranes than gluta- the unavoidable disruption of these networks by slicing proce- matergic membranes (Kawamura et al., 2006). GABAergic and glutamatergic dures might alter the impact of astroglial CB1R signaling terminals containing CB1R synapse with dendrites and spines of CA1 pyra- midal cells, respectively (Kawamura et al., 2006). In vitro activation of in vitro. Meanwhile, slicing procedures might also upregulate presynaptic CB1R by cannabinoids reduces the release of glutamate and the number or function of presynaptic CB1R, leading to GABA from glutamatergic and GABAergic membranes, respectively. a decrease of glutamatergic transmission upon its activation However, cannabinoid exposure in vivo sequentially activates astroglial CB1R by exogenous cannabinoids. This idea is supported by the and postsynaptic NR2B-containing NMDAR, which elicits AMPAR endocy- evidence that although CB1R density is at least 10-20 times tosis-mediated expression of in vivo LTD at CA3-CA1 synapses, resulting in higher on inhibitory than excitatory terminals in the CA1 region working memory impairment. (Kawamura et al., 2006; Bellocchio et al., 2010), application of a saturating concentration of WIN22,212-2 (2 mM) to hippo- et al., 2006; Takahashi and Castillo, 2006; Navarrete and campal slices produced similar depression (50%) of EPSC Araque, 2008, 2010; Bajo et al., 2009). Indeed, cannabinoids (Kawamura et al., 2006) and IPSC (Hajos and Freund, 2002)in fail to reduce excitatory or inhibitory synaptic transmission in the CA1 area. Because brain slice preparations are extensively hippocampal slices of conditional mutant mice lacking CB1R used for studying alterations of synaptic strength following expression in either glutamatergic or GABAergic hippocampal in vitro application of other drugs of abuse, it is worthwhile to neurons, respectively (Domenici et al., 2006; Monory et al., explore whether astrocytes play a key role in the in vivo effects 2006). Unexpectedly, we observed here that in vivo exposure of these drugs of abuse that are different from their in vitro to exogenous cannabinoids induced full CB-LTD at excitatory effects. CA3-CA1 synapses in both wild-type mice and mutant litter- Recent studies with brain slices show that endocannabinoids mates lacking CB1R in either CA1 glutamatergic or GABAergic activate CA1 astroglial CB1R to increase extracellular glutamate neurons. These data do not support an involvement of glutama- levels, which in turn activate presynaptic mGluR to induce LTP at tergic or GABAergic CB1R in in vivo CB-LTD at CA3-CA1 CA3-CA1 synapses (Navarrete and Araque, 2008, 2010). synapses. However, we show here that cannabinoids activate astroglial
The presence of CB1R has also been suggested in brain cells to induce in vivo LTD at CA3-CA1 synapses. It is currently astrocytes (Moldrich and Wenger, 2000; Rodriguez et al., 2001; unknown why activation of astroglial CB1R by in vitro endocan- Salio et al., 2002), but the extremely low levels of CB1R expres- nabinoid and in vivo cannabinoid induces, respectively, in vitro sion in this cell population did not allow reaching the same LTP and in vivo LTD at CA3-CA1 synapses. It is possible that conclusive evidence of functional data (Navarrete and Araque, activation of astroglial CB1R in brain slices with disrupted astro- 2008, 2010). The use of double immunostaining applied to glial networks might produce lower levels of interstitial glutamate wild-type and conditional or constitutive CB1R mutant mice al- than those produced in living animals with intact astroglial lowed us to provide conclusive electron microscopic evidence networks, which then activate presynaptic mGluR in vitro and
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. 1047 postsynaptic NMDAR in vivo, respectively, to induce in vitro LTP findings strongly suggest a causative role of CB-LTD at CA3- and in vivo LTD at CA3-CA1 synapses. CA1 synapses in cannabinoid-induced impairment of SWM This study confirmed the consistent finding that HU210 and and reveal novel mechanistic views of the role of astrocytes in THC impair SWM in rodents (Lichtman and Martin, 1996; Hamp- learning and memory processes and of the memory-disruptive son and Deadwyler, 2000; Nava et al., 2001; Varvel and Licht- effects of marijuana intoxication. man, 2002; Fadda et al., 2004; Hill et al., 2004; Wise et al., 2009). Although a recent study claimed the inability of systemic EXPERIMENTAL PROCEDURES HU210 injection to impair SWM tested with the DMTP water maze paradigm (Robinson et al., 2007), this study failed to use Generation of Mutant Mice Constitutive CB R-KO mice and conditional Glu-CB R-KO and GABA-CB R- the ‘saving ratio’ analysis as we and others (Varvel and Lichtman, 1 1 1 KO mice were generated and genotyped as described (Marsicano et al., 2002; 2002) have successfully used to identify the detrimental effects Monory et al., 2006). GFAP-CB1R-KO mice were generated using the Cre/loxP f/f of HU210 and THC on rodent SWM performance. system. Mice carrying the ‘‘floxed’’ CB1R gene (CB1 )(Marsicano et al., 2003) While glutamatergic axonal CB1R is in part responsible for were crossed with GFAP-CreERT2 mice (Hirrlinger et al., 2006), using a three- f/f;GFAP-CreERT2 f/f cannabinoid-elicited locomotor suppression, catalepsy and step backcrossing procedure to obtain CB1R and CB1R litter- hypothermia (Monory et al., 2007), hippocampal GABAergic mates, called GFAP-CB1R-KO and GFAP-CB1R-WT, respectively. axonal CB R likely plays a key role in cannabinoid impairment 1 Immunohistochemistry for Electron Microscopy of long-term memory (Puighermanal et al., 2009). Our data using Animals were transcardially fixed with 0.1% glutaraldehyde, 4% formaldehyde GABA-CB1R-KO mice clearly show that ‘‘GABAergic’’ CB1Ris and 0.2% picric acid or with 2% formaldehyde and 8% picric acid. Hippo- fully dispensable both for basal performance of the SWM task campal vibrosections were cut for double preembedding staining of CB1R and, most importantly in this context, for the acute effect of and GFAP with silver-intensified immunogold method and immunoperoxidase method. Tissue preparations were photographed for quantification of positive exogenous cannabinoids. By showing that Glu-CB1R-KO mice are impaired in basal performance of the SWM task, our data immunoreactive profiles. Detailed procedures are described in the extended methods in the SOMs. suggest that CB1R expressed in cortical glutamatergic neurons participates in the endogenous control of SWM. This control Adenovirus Preparation and Administration might be exerted acutely by endogenous mobilization of Recombinant adenoviruses were prepared as described (Liu et al., 2010). After endocannabinoids during the task or can also be due to devel- intra-CA1 infusion of adenoviral vectors (1010 plaque-forming units/ml/injec- tion), the CA1 area surrounding the injection tract was dissected 4 days later opmental effects of CB1R deletion in this cell population (Mulder et al., 2008). However, exogenous THC treatment of for quantification of CB1R protein with procedures as described (Ji et al., 2006; Liu et al., 2010). Glu-CB1R-KO mice is still able to further reduce their poor performance, strongly suggesting the dispensable role of ‘‘gluta- Synaptosomal Surface AMPAR Measurement matergic’’ CB1R in the acute effects of exogenous cannabinoids Biotinylation experiments for the CA1 area on hippocampal slices were per- on SWM performance. formed as described (Kim et al., 2007). Protein fractions were transferred
Conversely, by showing that GFAP-CB1R-KO mice display onto nitrocellulose membranes, which were probed with primary antibodies normal learning of SWM, but totally fail to respond to THC, the to GluR1 (1:250, Millipore, Billerica, MA) or GluR2 (1:500, Millipore, Billerica, present study provides striking evidence for the necessary role MA) overnight at 4 C. Bands were analyzed by densitometry, and receptor ratios for AMPAR subunits were determined by dividing the surface intensity of astroglial CB R in SWM impairment induced by exogenous 1 by the total intensity. cannabinoids. Cannabinoid-induced LTD and impairment of SWM share not Electrophysiology Analysis only the dependency on astroglial CB1R but also a whole Under anesthesia, rats or mice received implantation of stimulating and series of well-defined molecular mechanisms. Thus, the phar- recording electrodes into the CA1 region. fEPSPs were evoked by applying macological blockade of NR2B-containing NMDAR, but not single pulses of stimulation at 0.067 Hz. Stimulus pulse intensities were 20- 60 nA with a duration of 500 ms. Spike2 software was utilized to record data. NR2A-containing NMDAR, prevented both CB-LTD at CA3- Procedures for fEPSP recordings from freely moving rats were generally CA1 synapses and cannabinoid impairment of SWM. Moreover, similar to those from anaesthetized rats with the exception of allowing rats the Tat-GluR2 peptide can selectively block the facilitated to recover for 2 weeks after surgery for electrode implantation. Detailed endocytosis of AMPAR (Collingridge et al., 2010), the final step procedures are described in the extended methods in the SOMs. of the expression of NMDAR-dependent LTD (Collingridge et al., 2010), without significant effects on LTP induction or basal Behavioral Tests Water Maze Test synaptic transmission (Collingridge et al., 2010). Both systemic Mice were tested in a DMTP version of the Morris water maze paradigm (Steele and intra-CA1 application of the Tat-GluR2 peptide not only and Morris, 1999). Briefly, after a habituation session of 3 trials without spatial disrupted the expression of CB-LTD at CA3-CA1 synapses but cues, mice received daily training sessions of 4 trials each with the maximal also cannabinoid impairment of SWM, as assessed with both escape latency of 60 s, and 30 min before each of the sessions 6 through the DMTP version of the Morris water maze test and the DNMTS 12 and before the 13th session, mice were treated with vehicle and THC T-maze test. (5 mg/kg, i.p.), respectively. Performances of individual SWMs were calculated Collectively, at least three key molecular mechanisms are using the ‘‘saving ratio’’ procedure (Varvel and Lichtman, 2002) and calculated as follows: path saving ratio = (path-length trial - path-length trial ) / (path- shared by CB-LTD and cannabinoid-induced impairment of 1 4 length trial1 + path-length trial4); and latency saving ratio = (escape latency SWM: (1) activation of astroglial CB1R by the exogenous canna- trial1 - escape latency trial4) / (escape latency trial1 + escape latency trial4). binoid; (2) increase of local glutamate and activation of NR2B- Procedures for rat water maze test were generally similar to mouse water containing NMDAR; (3) endocytosis of AMPAR (Figure 7). These maze test with the exception that rats received 5 daily sessions of SWM
1048 Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. training 1 day before a testing session of 4 trials with the maximal escape Brooks, J.W. (2002). Cannabinoids and analgesia. Curr. Anaesth. Crit. Care 13, latency of 90 s. Detailed procedures are described in the extended methods 215–220. in the SOMs. Carlini, E.A. (2004). The good and the bad effects of (-) trans-delta-9-tetrahy- Other Behavioral Tests drocannabinol (Delta 9-THC) on humans. Toxicon 44, 461–467. Rats were examined with the DNMTS T-maze test (Kelsey and Vargas, 1993), Chevaleyre, V., Takahashi, K.A., and Castillo, P.E. (2006). Endocannabinoid- locomotor activity test (Ji et al., 2006), elevated-plus-maze test (Ji et al., 2006) mediated synaptic plasticity in the CNS. Annu. Rev. Neurosci. 29, 37–76. and motor balance tests (Ji et al., 2006). Collingridge, G.L., Peineau, S., Howland, J.G., and Wang, Y.T. (2010). Long- Statistical Analysis term depression in the CNS. Nat. Rev. Neurosci. 11, 459–473. Results were reported as mean ± SEM. Statistical analysis of the data was per- Cui, S.S., Bowen, R.C., Gu, G.B., Hannesson, D.K., Yu, P.H., and Zhang, X. formed using a student t test, square Chi test, one-way ANOVA, or one-way or (2001). Prevention of cannabinoid withdrawal syndrome by lithium: involve- two-way ANOVA for repeated-measures, followed by Bonferronni post-hoc ment of oxytocinergic neuronal activation. J. Neurosci. 21, 9867–9876. test. Statistical significance was set at p < 0.05. Domenici, M.R., Azad, S.C., Marsicano, G., Schierloh, A., Wotjak, C.T., Dodt, H.U., Zieglga¨ nsberger, W., Lutz, B., and Rammes, G. (2006). Cannabinoid SUPPLEMENTAL INFORMATION receptor type 1 located on presynaptic terminals of principal neurons in the fore- brain controls glutamatergic synaptic transmission. J. Neurosci. 26, 5794–5799. Supplemental Information includes Extended Experimental Procedures and Fadda, P., Robinson, L., Fratta, W., Pertwee, R.G., and Riedel, G. (2004). five figures and can be found with this article online at doi:10.1016/j.cell. Differential effects of THC- or CBD-rich cannabis extracts on working memory 2012.01.037. in rats. Neuropharm 47, 1170–1179.
ACKNOWLEDGMENTS Fox, C.J., Russell, K.I., Wang, Y.T., and Christie, B.R. (2006). Contribution of NR2A and NR2B NMDA subunits to bidirectional synaptic plasticity in the We thank J.-C. Maillet and X. Li for technical assistance; D. Gonzales and the hippocampus in vivo. Hippocampus 16, 907–915. personnel of Animal Facility of the NeuroCentre Magendie for animal care and Giaume, C., Koulakoff, A., Roux, L., Holcman, D., and Rouach, N. (2010). genotyping; Drs. D. Lagace (University of Ottawa), L. Renaud (University of Astroglial networks: a step further in neuroglial and gliovascular interactions. Ottawa), Z. Jia (University of Toronto), P. Albert (University of Ottawa), B. Nat. Rev. Neurosci. 11, 87–99. Lutz (University of Mainz), and P.V. Piazza (INSERM) for discussion and Hajos, N., and Freund, T.F. (2002). Pharmacological separation of cannabinoid comments. We also thank K. Mackie for providing the CB1R antibody, ob- sensitive receptors on hippocampal excitatory and inhibitory fibers. Neuro- tained with the NIH grant DA011322, as well as J. Rubenstein, M. Ekker, pharm. 43, 503–510. K. Nave, and F. Kirchhoff for providing the Cre-expressing mice used to generate conditional mutant mice. Support from China-Canada Joint Hajos, N., Ledent, C., and Freund, T.F. (2001). Novel cannabinoid-sensitive Research grant (to L. Xu and X.Z.), The Basque Country Government grant receptor mediates inhibition of glutamatergic synaptic transmission in the (GIC07/70-IT-432-07, to P.G.), Ministerio de Ciencia e Innovacio´ n (SAF2009- hippocampus. Neuroscience 106, 1–4. 07065, to P.G.), Red de Trastornos Adictivos, RETICS, Instituto de Salud Car- Hall, W., Christie, M., and Currow, D. (2005). Cannabinoids and cancer: causa- los III, MICINN (RD07/0001/2001, to P.G.), NIH grants (DA024741, to Q.L.; tion, remediation, and palliation. Lancet Oncol. 6, 35–42. NS38077, to G.B.), grants from National Natural Science Foundation of China Hampson, R.E., and Deadwyler, S.A. (2000). Cannabinoids reveal the (NNSFC, 30725019, 81030021, to W.W.), 973 Progress (2009CB941300, to necessity of hippocampal neural encoding for short-term memory in rats. L. Xu), NNSFC for distinguished young scholars (30725039, to L.X.), Major J. Neurosci. 20, 8932–8942. Program of NNSFC (30930091, to L.X.), National High Technology Research Han, J., Liu, Z., Ren, W., and Zhang, X. (2011). Counteractive effects of canna- and Development Program of China (863, 2007AA02Z310, to W.R.), binoid and nicotine addictive behavior. Neuroreport 22, 181–184. INSERM/Avenir (to G.M.), INSERM (to D.N.A.), European Research Council (ERC-2010-StG-260515, to G.M.), EU-FP7 (REPROBESITY, HEALTH-F2- Hassabis, D., Kumaran, D., Vann, S.D., and Maguire, E.A. (2007). Patients with 2008-223713, to G.M.), Fondation pour la Recherche Medicale (to G.M.), hippocampal amnesia cannot imagine new experiences. Proc. Natl. Acad. Sci. Region Aquitaine (to M.M.-L., G.M., and D.N.A.), Canadian Institutes of Health USA 104, 1726–1731. Research (to X.Z.), Natural Sciences and Engineering Research Council of Herkenham, M., Lynn, A.B., Little, M.D., Johnson, M.R., Melvin, L.S., de Costa, Canada (to X.Z.), Cheung Kong Scholar (Chang Jiang Scholar) Award (to B.R., and Rice, K.C. (1990). Cannabinoid receptor localization in brain. Proc. X.Z.), and Canadian Foundation for Innovation (to X.Z.). Natl. Acad. Sci. USA 87, 1932–1936. Higgins, G.A., Ballard, T.M., Enderlin, M., Haman, M., and Kemp, J.A. (2005). Received: June 15, 2011 Evidence for improved performance in cognitive tasks following selective Revised: October 21, 2011 NR2B NMDA receptor antagonist pre-treatment in the rat. Psychopharm. Accepted: January 11, 2012 179, 85–98. Published: March 1, 2012 Hill, M.N., Froc, D.J., Fox, C.J., Gorzalka, B.B., and Christie, B.R. (2004). Pro- REFERENCES longed cannabinoid treatment results in spatial working memory deficits and impaired long-term potentiation in the CA1 region of the hippocampus in vivo. Baddeley, A. (2003). Working memory: looking back and looking forward. Nat. Eur. J. Neurosci. 20, 859–863. Rev. Neurosci. 4, 829–839. Hirrlinger, P.G., Scheller, A., Braun, C., Hirrlinger, J., and Kirchhoff, F. (2006). Bajo, M., Roberto, M., and Schweitzer, P.J. (2009). Differential alteration Temporal control of gene recombination in astrocytes by transgenic expression of hippocampal excitatory synaptic transmission by cannabinoid ligands. of the tamoxifen-inducible DNA recombinase variant CreERT2. Glia 54, 11–20. Neurosci. Res. 87, 766–775. Hoffman, A.F., Laaris, N., Kawamura, M., Masino, S.A., and Lupica, C.R. (2010). Bellocchio, L., Lafeneˆ tre, P., Cannich, A., Cota, D., Puente, N., Grandes, P., Control of cannabinoid CB1 receptor function on glutamate axon terminals by Chaouloff, F., Piazza, P.V., and Marsicano, G. (2010). Bimodal control of stimu- endogenous adenosine acting at A1 receptors. J. Neurosci. 30, 545–555. lated food intake by the endocannabinoid system. Nat. Neurosci. 13,281–283. Ji, S.P., Zhang, Y., Van Cleemput, J., Jiang, W., Liao, M., Li, L., Wan, Q., Back- Brebner, K., Wong, T.P., Liu, L., Liu, Y., Campsall, P., Gray, S., Phelps, L., Phillips, strom, J.R., and Zhang, X. (2006). Disruption of PTEN coupling with 5-HT2C A.G., and Wang, Y.T. (2005). Nucleus accumbens long-term depression and the receptors suppresses behavioral responses induced by drugs of abuse. Nat. expression of behavioral sensitization. Science 310, 1340–1343. Med. 12, 324–329.
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. 1049 Kano, M., Ohno-Shosaku, T., Hashimotodani, Y., Uchigashima, M., and Monory, K., Massa, F., Egertova´ , M., Eder, M., Blaudzun, H., Westenbroek, R., Watanabe, M. (2009). Endocannabinoid-mediated control of synaptic Kelsch, W., Jacob, W., Marsch, R., Ekker, M., et al. (2006). The endocannabi- transmission. Physiol. Rev. 89, 309–380. noid system controls key epileptogenic circuits in the hippocampus. Neuron Kawamura, Y., Fukaya, M., Maejima, T., Yoshida, T., Miura, E., Watanabe, M., 51, 455–466. Ohno-Shosaku, T., and Kano, M. (2006). The CB1 cannabinoid receptor is the Mulder, J., Aguado, T., Keimpema, E., Baraba´ s, K., Ballester Rosado, C.J., major cannabinoid receptor at excitatory presynaptic sites in the hippocampus Nguyen, L., Monory, K., Marsicano, G., Di Marzo, V., Hurd, Y.L., et al. and cerebellum. J. Neurosci. 26, 2991–3001. (2008). Endocannabinoid signaling controls pyramidal cell specification and Kelleher, R.J., III, Govindarajan, A., and Tonegawa, S. (2004). Translational long-range axon patterning. Proc. Natl. Acad. Sci. USA 105, 8760–8765. regulatory mechanisms in persistent forms of synaptic plasticity. Neuron 44, Nava, F., Carta, G., Colombo, G., and Gessa, G.L. (2001). Effects of chronic 59–73. Delta(9)-tetrahydrocannabinol treatment on hippocampal extracellular acetylcholine concentration and alternation performance in the T-maze. Kelsey, J.E., and Vargas, H. (1993). Medial septal lesions disrupt spatial, but Neuropharm. 41, 392–399. not nonspatial, working memory in rats. Behav. Neurosci. 107, 565–574. Navarrete, M., and Araque, A. (2008). Endocannabinoids mediate neuron- Kesner, R.P. (2007). Behavioral functions of the CA3 subregion of the hippo- astrocyte communication. Neuron 57, 883–893. campus. Learn. Mem. 14, 771–781. Navarrete, M., and Araque, A. (2010). Endocannabinoids potentiate synaptic Kim, J., Lee, S., Park, K., Hong, I., Song, B., Son, G., Park, H., Kim, W.R., Park, transmission through stimulation of astrocytes. Neuron 68, 113–126. E., Choe, H.K., et al. (2007). Amygdala depotentiation and fear extinction. Proc. Natl. Acad. Sci. USA 104, 20955–20960. Pacher, P., Ba´ tkai, S., and Kunos, G. (2006). The endocannabinoid system as an emerging target of pharmacotherapy. Pharmacol. Rev. 58, 389–462. Lemberger, L. (1980). Potential therapeutic usefulness of marijuana. Annu. Rev. Pharmacol. Toxicol. 20, 151–172. Puighermanal, E., Marsicano, G., Busquets-Garcia, A., Lutz, B., Maldonado, R., and Ozaita, A. (2009). Cannabinoid modulation of hippocampal long-term Lichtman, A.H., and Martin, B.R. (1996). D9-tetrahydrocannabinol impairs memory is mediated by mTOR signaling. Nat. Neurosci. 12, 1152–1158. spatial memory through a cannabinoid receptor mechanism. Psychopharm. Ranganathan, M., and D’Souza, D.C. (2006). The acute effects of cannabi- 126, 125–131. noids on memory in humans: a review. Psychopharm. 188, 425–444. Liu, Z., Han, J., Jia, L., Maillet, J.C., Bai, G., Xu, L., Jia, Z., Zheng, Q., Zhang, Robinson, L., Goonawardena, A.V., Pertwee, R.G., Hampson, R.E., and Riedel, W., Monette, R., et al. (2010). Synaptic neurotransmission depression in G. (2007). The synthetic cannabinoid HU210 induces spatial memory deficits ventral tegmental dopamine neurons and cannabinoid-associated addictive and suppresses hippocampal firing rate in rats. Br. J. Pharmacol. 151, 688–700. learning. PLoS ONE 5, e15634. Robson, P. (2001). Therapeutic aspects of cannabis and cannabinoids. Br. J. Lovinger, D.M. (2008). Presynaptic modulation by endocannabinoids. Hand- Psychiatry 178, 107–115. book Exp. Pharmacol. 184, 435–477. Rodriguez, J.J., Mackie, K., and Pickel, V.M. (2001). Ultrastructural localization Lu, W., Shi, Y., Jackson, A.C., Bjorgan, K., During, M.J., Sprengel, R., Seeburg, of the CB1 cannabinoid receptor in m-opioid receptor patches of the rat P.H., and Nicoll, R.A. (2009). Subunit composition of synaptic AMPA receptors caudate putamen nucleus. J. Neurosci. 21, 823–833. revealed by a single-cell genetic approach. Neuron 62, 254–268. Rolls, E.T., and Kesner, R.P. (2006). A computational theory of hippocampal Maalouf, M., Dykes, R.W., and Miasnikov, A.A. (1998). Effects of D-AP5 and function, and empirical tests of the theory. Prog. Neurobiol. 79, 1–48. NMDA microiontophoresis on associative learning in the barrel cortex of Salio, C., Doly, S., Fischer, J., Franzoni, M.F., and Conrath, M. (2002). Neuronal awake rats. Brain Res. 793, 149–168. and astrocytic localization of the cannabinoid receptor-1 in the dorsal horn of Malenka, R.C., and Bear, M.F. (2004). LTP and LTD: an embarrassment of the rat spinal cord. Neurosci. Lett. 329, 13–16. riches. Neuron 44, 5–21. Sanderson, D.J., Good, M.A., Seeburg, P.H., Sprengel, R., Rawlins, J.N., and Manahan-Vaughan, D., Kulla, A., and Frey, J.U. (2000). Requirement of Bannerman, D.M. (2008). The role of the GluR-A (GluR1) AMPA receptor translation but not transcription for the maintenance of long-term depression subunit in learning and memory. Prog. Brain Res. 169, 159–178. in the CA1 region of freely moving rats. J. Neurosci. 20, 8572–8576. Serpa, A., Ribeiro, J.A., and Sebastia˜ o, A.M. (2009). Cannabinoid CB1 Marsicano, G., Goodenough, S., Monory, K., Hermann, H., Eder, M., Cannich, and adenosine A1 receptors independently inhibit hippocampal synaptic A., Azad, S.C., Cascio, M.G., Gutie´ rrez, S.O., van der Stelt, M., et al. (2003). transmission. Eur. J. Pharmacol. 623, 41–46. CB1 cannabinoid receptors and on-demand defense against excitotoxicity. Steele, R.J., and Morris, R.G. (1999). Delay-dependent impairment of a Science 302, 84–88. matching-to-place task with chronic and intrahippocampal infusion of the Marsicano, G., and Lutz, B. (2006). Neuromodulatory functions of the endo- NMDA-antagonist D-AP5. Hippocampus 9, 118–136. cannabinoid system. J. Endocrinol. Invest. 29, 27–46. Takahashi, K.A., and Castillo, P.E. (2006). The CB1 cannabinoid receptor Marsicano, G., Wotjak, C.T., Azad, S.C., Bisogno, T., Rammes, G., Cascio, mediates glutamatergic synaptic suppression in the hippocampus. Neurosci- M.G., Hermann, H., Tang, J., Hofmann, C., Zieglga¨ nsberger, W., et al. ence 139, 795–802. (2002). The endogenous cannabinoid system controls extinction of aversive Tsou, K., Brown, S., Sanudo-Pena, M.C., Mackie, K., and Walker, J.M. (1998). memories. Nature 418, 530–534. Immunohistochemical distribution of cannabinoid CB1 receptors in the rat Matsuda, L.A., Bonner, T.I., and Lolait, S.J. (1993). Localization of cannabinoid central nervous system. Neuroscience 83, 393–411. receptor mRNA in rat brain. J. Comp. Neurol. 327, 535–550. Varvel, S.A., and Lichtman, A.H. (2002). Evaluation of CB1 receptor knockout Misner, D.L., and Sullivan, J.M. (1999). Mechanism of cannabinoid effects on mice in the Morris water maze. J. Pharmacol. Exp. Ther. 301, 915–924. long-term potentiation and depression in hippocampal CA1 neurons. J. Neuro- Wise, L.E., Thorpe, A.J., and Lichtman, A.H. (2009). Hippocampal CB(1) recep- sci. 19, 6795–6805. tors mediate the memory impairing effects of Delta(9)-tetrahydrocannabinol. Moldrich, G., and Wenger, T. (2000). Localization of the CB 1 cannabinoid Neuropsychopharmacology 34, 2072–2080. receptor in the rat brain. An immunohistochemical study. Peptides 21, 1735– Witter, M.P., and Amaral, D.G. (2004). In The Rat Nervous System, G. Paxinos 1742. (Academic Press, Elsevier, ed. 3), chap. 21., 637–704. Monory, K., Blaudzun, H., Massa, F., Kaiser, N., Lemberger, T., Schu¨ tz, G., Wong, T.P., Howland, J.G., Robillard, J.M., Ge, Y., Yu, W., Titterness, A.K., Wotjak, C.T., Lutz, B., and Marsicano, G. (2007). Genetic dissection of Brebner, K., Liu, L., Weinberg, J., Christie, B.R., et al. (2007). Hippocampal behavioral and automatic effects of D9-tetrahydrocannabinol in mice. PLoS long-term depression mediates acute stress-induced spatial memory retrieval Biol. 5, e269. impairment. Proc. Natl. Acad. Sci. USA 104, 11471–11476.
1050 Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. Supplemental Information
EXTENDED EXPERIMENTAL PROCEDURES
Animal Procedure Approval All procedures were performed in accordance with the guidelines established by the Canadian Council on Animal Care as approved by Animal Care Committee of the University of Ottawa Institute of Mental Health Research (IMHR ACC). Specifically, the IMHR ACC approved the present study (ACC-2010-003). Similar procedures approved by the IMHR ACC for conducting rat electrophysiology were also approved by the Chinese Academy of Sciences and Shaanxi Normal University. Procedures for rodent electrophysiology and behavioral testing were approved by French Agricultural Ministry (Authorization number A3310035).
Generation of Mutant Mice
Constitutive CB1R-KO mice and conditional Glu-CB1R-KO and GABA-CB1R-KO mice were generated and genotyped as described (Marsicano et al., 2002; Monory et al., 2006). The anatomical pattern of CB1R expression in these mutant mice was reported (Monory et al., 2006; Bellocchio et al., 2010). GFAP-CB1R-KO mice were generated using the Cre/loxP system. Mice carrying the ‘‘floxed’’ f/f CB1R gene (CB1 )(Marsicano et al., 2003) were crossed with GFAP-CreERT2 mice (Hirrlinger et al., 2006), using a three-step back- f/f;GFAP-CreERT2 f/f crossing procedure to obtain CB1R and CB1R littermates, called GFAP-CB1R-KO and GFAP-CB1R-WT, respectively. Due to the fact that CreERT2 protein is inactive in the absence of tamoxifen treatment (Hirrlinger et al., 2006), deletion of the CB1R gene was obtained in adult mice (8 weeks) by 8 daily injections of tamoxifen (1 mg, i.p.), dissolved in 90% sunflower oil, 10% ethanol to a final concentration of 10 mg/ml (Hirrlinger et al., 2006). The animals were used at least 4 weeks after tamoxifen treatment. The absence of the CB1R protein in the hippocampus of tamoxifen-treated GFAP-CB1R-KO mice was verified by double immunohisto- chemistry for electron microscopy (see below and Figures 2D–2G of the main text). All the mutant mice used in this study were in a mixed genetic background, with a predominant contribution of the C57BL6-N strain (at least 5-6 backcrossing generations). All experimental animals were littermates.
Double CB1R-GFAP Immunohistochemistry for Electron Microscopy GFAP-CB1R-WT, GFAP-CB1R-KO and CB1R-KO mice were deeply anesthetized by an i.p. injection of a mixture of Nembutal (5 mg/ 100 g body weight; Abbott Laboratories Inc., IL, USA) and urethane (130 mg/100 g body weight; Sigma-Aldrich, St. Louis, MO, USA). One animal of each condition was transcardially perfused with PBS (0.1 M, pH 7.4) and then fixed with 500 ml of 0.1% glutaraldehyde, 4% formaldehyde and 0.2% picric acid in PBS. Two other animals were perfused with Zamboni’s fixative solution (2% formaldehyde and 8% picric acid in PBS). Perfusates were used at 4 C. Tissue blocks were extensively rinsed in 0.1 M PBS (pH 7.4). Coronal hippo- campal vibrosections were cut at 50 mm and collected in 0.1 M PBS (pH 7.4) at room temperature (RT). Sections were preincubated in a blocking solution of 10% bovine serum albumin (BSA), 0.1% sodium azide and 0.02% saponin prepared in Tris-HCl buffered saline (TBS, pH 7.4) for 30 min at RT. A preembedding silver-intensified immunogold method and an immunoperoxidase method were used for double staining of CB1R and GFAP in hippocampal sections. They were incubated in primary polyclonal rabbit anti-CB1R(2mg/ml, kindly provided by Dr. K. Mackie, Indiana University, Bloomington, IN) and monoclonal mouse anti-GFAP (1:1,000; Sigma Chemical Company St. Louis, MO, USA) antibodies both in 10% BSA/TBS containing 0.1% sodium azide and 0.004% saponin on a shaker for 2 days at 4 C. After several washes in 1% BSA/TBS, tissue sections were incubated in a secondary 1.4 nm gold-labeled anti-rabbit
IgG (Fab’ fragment, 1:100, Nanoprobes Inc., Yaphank, NY, USA) for the detection of CB1R, and in a biotinylated anti-mouse IgG (1:200, Vector Laboratories Burlingame, CA, USA) for the detection of GFAP, both in 1% BSA/TBS with 0.004% saponin on a shaker for 4 hr at RT. Tissue was washed in 1% BSA/TBS and processed by a conventional avidin-biotin horseradish peroxidase complex method (ABC; Elite, Vector Laboratories Burlingame, CA, USA). Thereafter, tissue was washed again in 1% BSA/TBS overnight at 4 C and postfixed in 1% glutaraldehyde in TBS for 10 min at RT. Following washes in double-distilled water, gold particles were silver-intensified with a HQ Silver kit (Nanoprobes Inc., Yaphank, NY, USA) for about 12 min in the dark and then washed in 0.1 M PB (pH 7.4). Sections were incubated subsequently with 0.05% DAB in 0.1 M PB and 0.01% hydrogen peroxide for 5 min and washed in 0.1 M PB for 2 hr at RT. Stained sections were osmicated (1% OsO4 in 0.1 M PB, pH 7.4, 20 min), dehydrated in graded alcohols to propylene oxide and plastic embedded flat in Epon 812.80 nm ultrathin sections were collected on mesh nickel grids, stained with uranyl acetate and lead citrate, and examined in a PHILIPS EM2008S electron microscope. Tissue preparations were photographed by using a digital camera coupled to the electron microscope. Figure compositions were scanned at 500 dots per inch (dpi). Labeling and minor adjustments in contrast and brightness were made using Adobe Photoshop (CS, Adobe Systems, San Jose, CA, USA). For semiquantitative analysis, 50-mm-thick hippocampal sections (from each animal condition described above) showing good and reproducible DAB immunoreaction and silver-intensified gold particles were cut at 80 nm. Image-J (version 1.36) was used to measure the membrane length. Electron micrographs (10,000–28,000X) were taken from grids (132 mm side) containing DAB immu- nodeposits; all of them showed a similar DAB labeling intensity indicating that selected areas were at the same depth. Furthermore, to avoid false negatives, only ultrathin sections in the first 1.5 mm from the surface of the tissue block were examined. Positive labeling was considered if at least one immunoparticle was within approximately 30 nm from the membrane. Metal particles on membranes and positive immunoreactive profiles were visualized and counted. Percentages of GFAP-positive profiles and synaptic terminals with CB1R were analyzed and displayed using a statistical software package (GraphPad Prism 4, GraphPad Software Inc, San Diego, USA).
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. S1 Adenovirus Preparation and In Vivo Gene Knocking Down
With our established protocol (Liu et al., 2010), the oligonucleotides encoding scramble or rat CB1R-targeted shRNAs were synthe- sized, annealed and ligated into the green fluorescent protein (GFP)-expressing adenoviral shuttle vector pAdshRNA/H1. The target- 0 0 0 ing sequences of shRNAs are as follows: CB1R-1, 5 -GATGAACAAGCTTATCAAG-3 and CB1R-2, 5 -CTGCAAGAAGCTGCAATCT- 30; scramble, 50-GAGAGTACTACGATCATAA-30. Recombinant adenoviruses were prepared using the pAdEasy system. After intra-CA1 microinfusion (B/P 3.8 mm, M/L ± 2.5 mm, D/V 2.5 mm) of adenoviral vectors (1010 plaque-forming units/ml/injection), the CA1 region surrounding the injection tract was dissected 4 days later for the quantification of CB1R protein using immunoblotting procedures as described (Ji et al., 2006; Liu et al., 2010). Briefly, membranes were probed with primary antibodies to CB1R (1:500, Santa Cruz Biotechnology) overnight at 4 C. Immunoreactive bands were visualized using enhanced chemiluminescence and captured by a CCD digital camera. Bands were analyzed by densitometry (Alpha Innotech, Santa Clara, CA).
Synaptosomal Surface AMPAR Measurement Male Sprague-Dawley rats (Charles River) weighing about 200 g were killed for preparation of hippocampal slices (400 mm in thick- ness) 4 hr after injection of vehicle (i.p.) or HU210 (50 mg/kg, i.p.) with or without TAT-GluR2 or TAT-GluR2S (GL Biochem, Shanghai, China; 1.5 mmol/kg, i.p.) pretreatment 2 hr before HU210 injection. Biotinylation experiments were performed as described in Kim et al. (2007) with modifications. The CA1 region was microdissected from the hippocampus slices, pooled together and incubated with aCSF containing 1mg/ml of sulfosuccinimidyl-6-(biotinamido) hexanoate (Pierce Chemical, Rockford Il.) for 30 min on ice and quenched by three successive washes with ice cold TBS (50 mM Tris, pH 7.4, 150 mM NaCl). Microdissected CA1 slices were lysed in ice-cold homogenization buffer (10 mM Tris pH 7.6, 320 mM sucrose, 5 mM NaF, 1 mM Na3VO4, 1 mM EDTA, and 1 mM EGTA, 0.5 mM AEBSF, 50 mM Antipain HCl, 25 mM leupeptine and 0.5 mg/ml pepstatin). Homogenates were centrifugated at 1,000 x g at 4 C for 10 min to remove nuclei and large debris. After centrifugation, an aliquot of 50 ml was taken and designated as ‘‘total protein frac- tion.’’ The supernatant was further centrifugated at 10,000 x g at 4 C to obtain a crude synaptosomal fraction. The supernatant was then discarded, and the pellet was sonnicated on ice in lysis buffer (1% Triton X-100, 0.1% SDS, 50 mM Tris pH 7.4, 150 mM NaCl, 2 mM EDTA, 0.5 mM AEBSF, 50 mM Antipain HCl, 25 mM leupeptine and 0.5 mg/ml pepstatin). Samples were subsequently centri- fugated at 16,000 x g at 4 C for 20 min. In order to isolate biotinylated proteins, 150 mg of protein was incubated with 50 ml of Neu- travidin agarose (Pierce Chemical, Rockford Il.) for 4 hr at 4 C. The Neutravidin beads were subsequently washed 3 times with lysis buffer and proteins were eluted off the beads using 40 ml of 2 X SDS-PAGE loading buffer and boiling for 5 min. Both ‘‘surface’’ and ‘‘total’’ protein fractions were separated on 10% SDS-PAGE gels, transferred onto Nitrocellulose membranes (Thermo Scientific, Ottawa, ON). Membranes were probed with primary antibodies to GluR1 (1:250, Millipore, Billerica, MA) or GluR2 (1:500, Millipore, Billerica, MA) overnight at 4 C. After visualization, bands were analyzed by densitometry (FluroChem SP software, Alpha Innotech). Surface/total ratios for AMPAR subunits were determined by dividing the surface intensity by the total intensity. In order calculate the percent difference in cell surface expression of AMPAR subunits the surface/total ratio for the HU210 condition was divided by the surface/total ratio of the vehicle condition.
Rat Surgery Male Sprague-Dawley rats (Charles River) weighing 250-300 g received isoflurane anesthesia, followed by placing the rat onto a stereotaxic frame (Model 1730, David Kopf Instruments) on the surgical table. A longitudinal incision of 1.5-2.5 cm in length was made in the midline of the scalp, followed first by exposing the skull, including its Sigma point (i.e., the joint point of the coronal line with the sagital line), and then by drilling 1 or 2 holes of 0.3 cm in diameter on the skull. Next, 1 or 2 guide cannulae (26 Ga) were implanted into the unilateral or bilateral CA1 area (B/P 3.9 mm, M/L ± 2.6 mm, D/V 2.45 mm) or above the lateral ventricle (B/P 0.92 mm, M/L 1.4 mm, D/V 2.6 mm), followed first by affixing the cannula to the skull with screws and dental acrylic and then by filling the cannula with a obturator. After closing the wound with skin sutures, the rat was allowed to recover from anesthesia and returned to its home cage. Rats were allowed to recover for at least 7 days before intra-CA1 injection, i.c.v. microinjection, elec- trophysiology or behavioral testing. The intra-CA1 and i.c.v. microinjections were performed by insertion of a needle through the cannula and subsequent injection at a speed of approximately 0.1 ml/min and 1 ml/min, respectively. After behavioral testing, rats were deeply anesthetized with sodium pentobarbital (100 mg/kg, i.p.) and perfused transcardially with 150 ml of 0.1 M PBS, pH 7.4, followed by 250 ml of freshly prepared 4% paraformaldehyde in PBS. Then, rat brains were cut for cresyl violet staining to confirm correct cannula placement.
Electrophysiology Analysis Electrophysiology in Anesthetized Rats and Mice Male Sprague-Dawley rats (Charles River) weighing about 250 g were anaesthetized using 40% urethane (3 ml/kg, i.p.) comple- mented with 4% chloral hydrate (3 ml/kg, i.p.). The rats were then fixed to a stereotaxic frame (Model 1730, David Kopf Instruments), followed by drilling a hole in the skull and inserting two electrodes into the CA1 area or the angular bundle and dentate gyrus with the following coordinates: (I) CA1 area, B/P 4 mm, M/L 2.8 mm, D/V 2.4 mm for recording electrode, and B/P +3.5 mm, M/L +3.84 mm, D/V 2.4 mm for stimulating electrode; (II) dentate gyrus, B/P 2.8 mm, M/L 1.05 mm, D/V 4.1 mm for recording electrode, and angular bundle, B/P +8 mm, M/L +5.42 mm, D/V 2.98 mm for stimulating electrode. The stimulating electrode was placed into the correct region by hand while the glass microelectrode full of 2 M NaCl was moved first by hand to the brain surface and then
S2 Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. by Hydraulic Micropositioner (Model 2650, David Kopf Instruments) to ensure accuracy. The two electrodes were then adjusted to find the best wave. The recordings were performed with a Model MDA-3 AC Differential Amplifier (BAK Electronics, Inc.). With a concentric bipolar electrode, fEPSPs were evoked by applying single pulses of stimulation at 0.067 Hz. Stimulus pulse intensities were typically 20-60 mA with a duration of 500 ms. A stimulus intensity that yielded approximately 60% of the maximal response was selected for baseline measurements – using a Stimulator (Model S88, Grass Telefactor) and a Stimulus Isolator (Model A385, World Precision Instruments). Spike2 software was utilized to record data. Once the ideal placement of the electrodes was established, baseline fEPSPs were recorded for at least 20 min, followed by administration of HU210 (Sigma, St. Louis, USA; 0.05 or 0.1 mg/kg, i.p.; 50 mg/ml, iontophoretic ejection at 40 nA for 30 s, intra-CA1 infusion; dissolved in 2 x DMSO:1 x Tween 80:27 x physiological saline), THC (Sigma, St. Louis, USA; 5 mg/kg, i.p.; dissolved in 1 x ethanol:1 x Tween 80:10 x physiological saline), TBOA (Sigma, St. Louis, USA; 10 nmol, i.c.v.; dissolved in 0.1 M PBS) (Wong et al., 2007) or vehicle before a continuous recording of fEPSPs for 120 min. In agreement with a recent study (Wong et al., 2007), an i.c.v. administration of vehicle (0.1 M PBS) for TBOA slightly increased fEPSP amplitude, for unknown reasons, within 60 min after treatment (Figure 3A). Other groups of rats received different treatments before fEPSP recording for 120 min: (I) AM281 (Tocris Bioscience, Ellisville, USA; 3 mg/kg, i.p.; dissolved in 2 x DMSO:1 x Tween 80:37 x physiological saline) (Cui et al., 2001) or vehicle 10 min before or after a HU210 injection (50 mg/kg, i.p.); (II) anisomycin (Sigma, St. Louis, USA; 18 mg/kg, i.p.; dissolved in saline) (Puighermanal et al., 2009), actinomycin-D (Sigma, St. Louis, USA; 72 mg/ 12 ml, i.c.v.; dissolved in 2 x DMSO:1 x Tween 80:37 x physiological saline) (Manahan-Vaughan et al., 2000) or vehicle 2 hr before a HU210 injection (50 mg/kg, i.p.); (III) adenoviral vectors containing shRNA CB1R or its scrambled sequence were injected through a previously implanted guide cannula into the CA1 pyramidal cell layer (1010 plaque-forming units/ml/injection) 4 days prior to a place- ment of the recording electrode into the same region through the same guide cannula, followed by HU210 injection (50 mg/kg, i.p.); (IV) E4CPG injection (Tocris Bioscience, Ellisville, USA; 35 nM/3.5 ml, i.c.v.; dissolved in 1 x ethanol:1 x Tween 80:10 x physiological saline) 5 min either before DHPG injection (Tocris Bioscience, Ellisville, USA; 100 nM/5 ml, i.c.v.; dissolved in 1 x ethanol:1 x Tween 80:10 x physiological saline) or HU210 injection (50 mg/kg, i.p.); (V) AP-5 (50 mM, 20 nA for 10 min, intra-CA1 iontophoretic ejection) (Maalouf et al., 1998), Ro25,6981 (Sigma, St. Louis, USA; 6 mg/kg, i.p.) (Fox et al., 2006), ifenprodil (Sigma, St. Louis, USA; 5 mg/kg, i.p.) (Higgins et a., 2005), NVP-AAM077 (Fox et al., 2006) (Novartis Pharma AG; 1.2 mg/kg, i.p.) or vehicle 30 min prior to HU210 injec- tion (50 mg/kg, i.p.); or (VI) TAT-GluR2 or TAT-GluR2S (GL Biochem, Shanghai, China) injection (1.5 mmol/kg, i.p.) (Brebner et al., 2005; Wong et al., 2007; Ge et al., 2010) 2 hr before HU210 injection (50 mg/kg, i.p.). Additional two groups of rats received TAT-GluR2 or TAT-GluR2S injection (1.5 mmol/kg, i.p.), followed by recording of fEPSP for 4 hr. At the end of the experiment, a high frequency stim- ulation (HFS, 100 Hz train for one sec x 3 times with 20 s apart) was performed. If the HFS induced LTP the neurons were assumed to be in good condition and the observed LTD was not due to damage or fatigue of the neurons. Procedures for fEPSP recordings from anaesthetized mice were generally similar to those from anaesthetized rats with the following two exceptions: (a) after initial anes- thetization in a box with 4% halothane, a mouse was placed onto a stereotaxic frame with its nose placed in an anesthetic mask, through which 1.5% halothane was continuously provided for the duration of the experiment; (b) recording electrode was placed into the CA1 area with the coordinates of B/P 1.94 mm, M/L 1.2 mm, D/V 1.33 mm, and stimulating electrode had the coordinates of B/P +1.82 mm, M/L +2.13 mm, D/V 1.7 mm. Electrophysiology in Freely Moving Rats Male Sprague-Dawley rats weighing 250-300 g were anesthetized with pentobarbitone sodium (60 mg/kg, i.p.). Two electrodes made by gluing together a pair of twisted Teflon-coated 90% platinum/10% iridium wires (100 mm diameter, World Precision Instruments, USA) were placed into the CA1 area with following coordinates: B/P 3.8 mm, M/L 2.8 mm, D/V 2.4 mm for recording electrode, and B/P +4.8 mm, M/L +3.8 mm, D/V –2.4 mm for stimulating electrode. The optimal placement of the electrodes was adjudged to find the best wave. The whole assembly was sealed and fixed to the skull using dental acrylic. Rats were then placed back into their home cages individually for recovery for at least 2 weeks. Animals were then handled and allowed to fully adapt to the recording chamber, followed by recordings of stable fEPSPs at CA3-CA1 synapses for at least three consecutive days. Test fEPSPs were evoked at a frequency of 0.01667 Hz and at an intensity to evoke half of the maximum fEPSPs. After recording of stable baseline fEPSP for at least 40 min, HU210 (0.1 mg/kg, i.p.) and vehicle were injected, followed by recording of the fEPSPs at 2 hr, 3 hr, 6 hr and 24 hr. During the recording intervals, rats had free access to water and food in their home cages.
Behavioral Tests Water Maze Test
GFAP-CB1R-KO (tamoxifen-treated, n = 14), GFAP-CB1R-WT littermates (tamoxifen-treated, n = 10), and Glu-CB1R-KO (n = 11), GABA-CB1R-KO (n = 12) and wild-type (WT) littermates (n = 10) were tested in a delayed-matching to place version of the Morris Water Maze paradigm assessing spatial short-term memory (Steele and Morris, 1999). The maze consisted of a circular pool (150 cm diameter, 60 cm high) filled with water (19 –20 C) rendered opaque with nontoxic cosmetic white paint. The pool was flanked with black curtains on which different visual cues were posted. A white platform (14 cm diameter) was hidden 1 cm below the surface of the water. A computerized tracking system (VIDEOTRACK , Viewpoint, Lyon, France) allowed recording the path length of the animals and latency to reach the platform. Twenty four hours before training, mice were submitted to a habituation session consisting of 3 trials without spatial cues in order to acquire the basic rules of the task (i.e., finding the platform is the only escape). Mice were first released into the maze for 60 s without the platform. In a second trial, animals were placed on the platform at the level of the water surface in the center of the pool for 20 s, removed and then allowed to swim to the visible platform from one side of the pool. Animals
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. S3 that did not reach the platform within 60 s were manually guided to it. Mice were then removed after 30 s on the platform. The third trial was conducted in the same way except that the platform was submerged. During spatial learning training, the location of the submerged platform was changed daily somewhere halfway between the wall and the center of the tank. Daily training sessions con- sisted of 4 trials separated by inter-trial intervals of 30 s. Mice were released from 3 different starting points at each trial. Latency to reach the platform was recorded. If the mice did not reach the platform within 60 s, they were manually guided to it. To habituate the animals to the treatment, all the mice were injected with saline (0.1 ml, i.p) prior to the first 5 training sessions. Thirty min before sessions 6 to 12, all animals were treated with vehicle used to dissolve THC (1 x ethanol:1 x Cremophor:50 x saline; 10 ml/kg, i.p.). The average latencies to reach the platform during sessions 6 to 12 were considered as baseline learning performances for all genotypes. Thirty min before the thirteenth session, all the animals were treated with THC (5 mg/kg, i.p.; Sigma Aldrich, France). In this procedure, a decrease of escape latencies between the first and the following trials was assumed as indication of spatial working memory (SWM). Performances of individual SWMs were calculated using the ‘‘saving ratio’’ procedure (Varvel and Lichtman,
2002) between trial 1 and trial 4 and calculated as follows: path saving ratio = (path-length trial1 - path-length trial4) / path-length trial1 + path-length trial4); and latency saving ratio = (escape latency trial1 - escape latency trial4) / escape latency trial1 + escape latency trial4). Data are presented as absolute latencies (sessions 6-12) and as saving ratios (session 13) under vehicle or THC treatment for all mice tested. Latencies were analyzed with repeated-measure one- (within genotype) and two-way ANOVA (between genotypes), whereas saving ratios and swim speed were analyzed by two-way ANOVA, using treatment and genotype as variables. Procedures for rat water maze test were generally similar to mouse water maze test with the following exceptions: (a) male Sprague-Dawley rats (Charles River) weighing 250-300 g were handled for 3-5 days by the experimenter before 5 daily sessions of SWM training, followed 1 day later by testing sessions consisting of 4 trials; (b) the maximal escape latency was 90 s; (3) rats were injected with HU210 (0.05 mg/kg, i.p.), vehicle, Ro25-6981 (6 mg/kg, i.p.), Tat-GluR2 (1.5 mmol/kg, i.p.) or HU210 plus Ro25-6981 or Tat-GluR2 pretreat- ment 4 hr (HU210 and its vehicle or Ro25-6981), 2 hr (Tat-GluR2) or 20 min (Ro25-6981) before testing session. Delayed Nonmatching-to-Sample T-Maze Test Male Sprague-Dawley rats (Charles River) weighing 300-350 g received the DNMTST test described (Kelsey and Vargas, 1993). In a room with light on between 7:00 am and 7:00 pm, rats were housed individually in clear, plastic cages with ad lib access to food and water. One week after arrival, rats were gently handled twice a day for 3 days, during which the reward Graham Crumbs (‘‘Christie’’ brand from Kraft Canada Inc.) were introduced to their home cages once daily at 6:00 pm. During testing, rats were given 5g/100 g body weight of chow inside the cage so as to maintain their body weight at about 85% of their free-feeding body weight. Rats’ spatial working memory (SWM) was examined in a wooden T-shape box (75-cm-long start alley connected to two 30-cm-long goal arms with 15 cm in width and 30 cm in height). The inside wall of the maze and both sides of three doors were painted black. DNMTST test consisted of pretest training, acquisition training and performance test. On day 1 of pretest training, rats were allowed to explore the open maze and eat the Graham Crumbs from a dish placed at the end of one goal arm. On subsequent 2 days, the rats were allowed to leave the start box and enter a goal box to receive the Graham Crumbs, resulting in a gradual habituation to the guil- lotine doors. To ensure that rats have an equal experience in both goal boxes, the entrance to one goal box was occasionally blocked by the guillotine door. Training continued in this fashion until the rats were reliably traversing the maze in less than 2 min. The acqui- sition training lasted 6 days. Rats were brought in pairs to the maze room in their home cages and trained for 10 trials per day. Each trial included a sample run and a choice run. During the initial sample run, access to one goal box was blocked. Five seconds after placing the rat into the start box, the door from the start box was opened, enabling the rat to enter the open goal box to eat the Gra- ham Crumbs. Once finished eating, the rat was returned to the start box for a 5 s retention interval. On the subsequent choice run, all three doors were raised, and the rat received Graham Crumbs reward only after entering the goal box opposite to the food box on the preceding sample run. An entry was judged to have occurred when all four legs were in the goal box. Once entering the incorrect goal box, the rat received no Graham Crumbs and was required to spend 25 s in that goal box. The rat was then returned to its home cage for the 3 min inter-trial interval. The goal box that the rats were forced to enter on the sample run was selected randomly each day with the requirement that each goal box was selected on half of the 10 trials each day, but on no more than 3 trials in a row. The maze was cleaned of feces and urine between trials. The performance test lasted 2 days. Beginning 1 day following the acquisition training, rats were tested for the same 10 trials per day for 2 days as during acquisition with the exception that the retention interval between the sample and choice runs was 30 s. The effects of systemic or intra-CA1 HU210 were examined by giving an i.p. or intra-CA1 admin- istration of HU210 (0.05 mg/kg, i.p.; 0.1 mg/0.5 ml/each side, intra-CA1) or vehicle 10 min before the first trial of each testing day. The averaged values of these two-day data represent the ability of rat performance of SWM. The rats were placed back into their home cages during the retention intervals. To examine the effects of the LTD-blocking Tat-GluR2 peptide (GL Biochem, Shanghai, China) on SWM performance, rats received i.p. or bilateral intra-CA1 injections of TAT-GluR2 or TAT-GluR2S (1.5 mmol/kg, i.p.; 15 pmol/ 1.0ml/side, intra-CA1) 2 hr (i.p. injection) or 1 hr (intra-CA1 injection) (Brebner et al., 2005; Wong et al., 2007; Ge et al., 2010) before HU210 administration (50 mg/kg, i.p.). To examine the effects of NMDAR antagonists on SWM performance, rats received Ro25-6981 (6 mg/kg, i.p.) (Fox et al., 2006), ifenprodil (5 mg/kg, i.p.) (Higgins et a., 2005) and NVP-AAM077 (1.2 mg/kg, i.p.) (Fox et al., 2006) 10 min before HU210 injection (0.05 mg/kg, i.p.). On each of the 2 testing days, rats received HU210 or vehicle injection 30 min before test. Locomotor Activity Test Locomotor activity test was conducted as described (Ji et al., 2006). Locomotor activity levels were examined by computerized monitoring of photo-beam breaks. The square maze measured 45-cm long and 31-cm vertical. Two hours after an i.p. injection of
S4 Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. Tat-GluR2 (1.5 mmol/kg) or one hour after a bilateral intra-CA1 injection of Tat-GluR2 (15 pmol/1.0ml/side) through implanted guide cannulae, rats were allowed to investigate the maze for 5 min, during which the total number of movements was recorded. Elevated-Plus-Maze Test Elevated-plus-maze test was performed as described (Ji et al., 2006). Briefly, the Maze constructed with hard wood consists of two opposite open arms (50 cm 3 10 cm) and two opposite enclosed arms (50 cm 3 10 cm 3 40 cm) with the open arms surrounded by a 1 cm high Plexiglas ledge, set up 50 cm above the floor. The junction area of the four arms (central platform) measures 10 cm 3 10 cm. Before each trial, the maze was cleaned thoroughly with a solution of 60% alcohol. Immediately after the locomotor activity test, the rat was removed from the activity box and placed onto the central platform of the plus-maze facing an open arm, and allowed to explore the apparatus for 5 min. Time spent in and entries into open arms and enclosed arms were recorded. An arm entry was recorded when all four paws of the rat were in the arm. Measures of anxiety included entry ratio, the ratio of closed arm to open arm entries, and dwell ratio, the ratio of dwell time in the closed arms to the dwell time in all four arms, with higher values indicating higher levels of anxiety. Motor Balance Tests Motor balance tests were performed as described (Ji et al., 2006). The bar catalepsy test was conducted immediately after the elevated-plus-maze test. The forepaws of rats were placed on a 1-cm-diameter bar fixed horizontally at 10 cm from the bench surface for a maximum of 10 s, during which the time (in seconds) for the animal to remove both forepaws was recorded. Balance and locomotor deficits were assessed on an elevated beam-walking test. One day prior to Tat-GluR2 injection, rats were trained to traverse the elevated wooden beam of 3.0-cm width and 1-m length from the open end to the darkened reward box. The following day after the bar catalepsy test, the time for rats to traverse the beam or falling was assessed in 5 separate trials.
Statistical Analysis Results were reported as mean ± SEM. Statistical analysis of the data was performed using a student t test, square Chi test, one-way ANOVA, two-way ANOVA or two-way ANOVA for repeated-measures, followed by Bonferronni post-hoc test. Statistical significance was set at p < 0.05.
SUPPLEMENTAL REFERENCES
Bellocchio, L., Lafeneˆ tre, P., Cannich, A., Cota, D., Puente, N., Grandes, P., Chaouloff, F., Piazza, P.V., and Marsicano, G. (2010). Bimodal control of stimulated food intake by the endocannabinoid system. Nat. Neurosci. 13, 281–283. Brebner, K., Wong, T.P., Liu, L., Liu, Y., Campsall, P., Gray, S., Phelps, L., Phillips, A.G., and Wang, Y.T. (2005). Nucleus accumbens long-term depression and the expression of behavioral sensitization. Science 310, 1340–1343. Cui, S.S., Bowen, R.C., Gu, G.B., Hannesson, D.K., Yu, P.H., and Zhang, X. (2001). Prevention of cannabinoid withdrawal syndrome by lithium: involvement of oxytocinergic neuronal activation. J. Neurosci. 21, 9867–9876. Fox, C.J., Russell, K.I., Wang, Y.T., and Christie, B.R. (2006). Contribution of NR2A and NR2B NMDA subunits to bidirectional synaptic plasticity in the hippo- campus in vivo. Hippocampus 16, 907–915. Ge, Y., Dong, Z., Bagot, R.C., Howland, J.G., Phillips, A.G., Wong, T.P., and Wang, Y.T. (2010). Hippocampal long-term depression is required for the consol- idation of spatial memory. Proc. Natl. Acad. Sci. USA 107, 16697–16702. Higgins, G.A., Ballard, T.M., Enderlin, M., Haman, M., and Kemp, J.A. (2005). Evidence for improved performance in cognitive tasks following selective NR2B NMDA receptor antagonist pre-treatment in the rat. Psychopharm. 179, 85–98. Hirrlinger, P.G., Scheller, A., Braun, C., Hirrlinger, J., and Kirchhoff, F. (2006). Temporal control of gene recombination in astrocytes by transgenic expression of the tamoxifen-inducible DNA recombinase variant CreERT2. Glia 54, 11–20. Ji, S.P., Zhang, Y., Van Cleemput, J., Jiang, W., Liao, M., Li, L., Wan, Q., Backstrom, J.R., and Zhang, X. (2006). Disruption of PTEN coupling with 5-HT2C recep- tors suppresses behavioral responses induced by drugs of abuse. Nat. Med. 12, 324–329. Kelsey, J.E., and Vargas, H. (1993). Medial septal lesions disrupt spatial, but not nonspatial, working memory in rats. Behav. Neurosci. 107, 565–574. Kim, J., Lee, S., Park, K., Hong, I., Song, B., Son, G., Park, H., Kim, W.R., Park, E., Choe, H.K., et al. (2007). Amygdala depotentiation and fear extinction. Proc. Natl. Acad. Sci. USA 104, 20955–20960. Liu, Z., Han, J., Jia, L., Maillet, J.C., Bai, G., Xu, L., Jia, Z., Zheng, Q., Zhang, W., Monette, R., et al. (2010). Synaptic neurotransmission depression in ventral tegmental dopamine neurons and cannabinoid-associated addictive learning. PLoS ONE 5, e15634. Maalouf, M., Dykes, R.W., and Miasnikov, A.A. (1998). Effects of D-AP5 and NMDA microiontophoresis on associative learning in the barrel cortex of awake rats. Brain Res. 793, 149–168. Manahan-Vaughan, D., Kulla, A., and Frey, J.U. (2000). Requirement of translation but not transcription for the maintenance of long-term depression in the CA1 region of freely moving rats. J. Neurosci. 20, 8572–8576. Marsicano, G., Goodenough, S., Monory, K., Hermann, H., Eder, M., Cannich, A., Azad, S.C., Cascio, M.G., Gutie´ rrez, S.O., van der Stelt, M., et al. (2003). CB1 cannabinoid receptors and on-demand defense against excitotoxicity. Science 302, 84–88. Marsicano, G., Wotjak, C.T., Azad, S.C., Bisogno, T., Rammes, G., Cascio, M.G., Hermann, H., Tang, J., Hofmann, C., Zieglga¨ nsberger, W., et al. (2002). The endogenous cannabinoid system controls extinction of aversive memories. Nature 418, 530–534. Monory, K., Massa, F., Egertova´ , M., Eder, M., Blaudzun, H., Westenbroek, R., Kelsch, W., Jacob, W., Marsch, R., Ekker, M., et al. (2006). The endocannabinoid system controls key epileptogenic circuits in the hippocampus. Neuron 51, 455–466.
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. S5 Puighermanal, E., Marsicano, G., Busquets-Garcia, A., Lutz, B., Maldonado, R., and Ozaita, A. (2009). Cannabinoid modulation of hippocampal long-term memory is mediated by mTOR signaling. Nat. Neurosci. 12, 1152–1158. Steele, R.J., and Morris, R.G. (1999). Delay-dependent impairment of a matching-to-place task with chronic and intrahippocampal infusion of the NMDA-antag- onist D-AP5. Hippocampus 9, 118–136. Varvel, S.A., and Lichtman, A.H. (2002). Evaluation of CB1 receptor knockout mice in the Morris water maze. J. Pharmacol. Exp. Ther. 301, 915–924. Wong, T.P., Howland, J.G., Robillard, J.M., Ge, Y., Yu, W., Titterness, A.K., Brebner, K., Liu, L., Weinberg, J., Christie, B.R., et al. (2007). Hippocampal long-term depression mediates acute stress-induced spatial memory retrieval impairment. Proc. Natl. Acad. Sci. USA 104, 11471–11476.
S6 Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. Figure S1. HU210 Induces In Vivo LTD at CA3-CA1 Synapses but Not at PP-DG Synapses, Related to Figure 1 (A and B) Plots of normalized fEPSP slopes in anesthetized rats with representative fEPSP traces before (1) and after (2) vehicle or HU210 injection above each plot. Intra-CA1 infusion of HU210 (50 mg/ml, iontophoretic ejection at 40 nA for 30 s), but not vehicle, at 0 min elicits in vivo LTD at CA3-CA1 synapses lasting for > 2 hr (A), whereas an i.p. injection of HU210 (0.05 mg/kg) or vehicle at 0 min does not induce significant LTD at PP-DG synapses (B). (C) Histogram summarizes the average percent change of fEPSP slope before (1) and after (2) vehicle or cannabinoid injection in A and B. All summary graphs show means ± SEM; n = numbers of animals recorded in each group. *p < 0.01 versus vehicle control (A), p = 0.198 (B), t test.
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. S7 Figure S2. SWM Acquisition and Swimming Speed of Mice with DMTP Water Maze Paradigm, Related to Figure 5
(A and B) Pooled data on sessions 6–12 show that with the exception of Glu-CB1R-KO littermates displaying impairment of SWM acquisition, there is no significant genotype effect on SWM acquisition although all wild-type, GFAP-CB1R-KO and GABA-CB1R-KO mice significantly acquire SWM during trial 1. (C and D) Swim speeds during session 13 were not significantly different between wild-type and mutant mice receiving vehicle or THC. All summary graphs show means ± SEM; n = numbers of animals tested in each group. Data were analyzed by two- (between genotypes) or one-way ANOVA
(within genotype) for repeated-measures. (A) Two-way ANOVA: genotype F1,22 = 1.400, p = 0.249; trial F3,22 = 39.280, p < 0.001; genotype x trial: F3,66 = 1.891, p = 0.139; One-way ANOVA: GFAP-CB1R-WT: F3,27 = 3.743, p < 0.05; GFAP-CB1R-KO: F3,39 = 13.003, p < 0.001. (B) two-way ANOVA: genotype F2,30 = 5.075, p < 0.05; trial: F3,30 = 31.54, p < 0.001; genotype x trial: F6,90 = 4.129, p = 0.001; One-way ANOVA: WT: F3,27 = 5.147, p < 0.001; GABA-CB1R-KO: F3,33 = 10.96, p <
0.001; GLU-CB1R-KO: F3,30 = 22.07, p = 0.089. (C) Two-way ANOVA: genotype F1,22 = 0.909, p = 0.351; treatment F1,22 = 0.612, p = 0.442, genotype x treatment
F1,22 = 3,947, p = 0.06; (D) genotype F2,30 = 2.492, p = 0.100; treatment F2,30 = 4.179, p = 0.025).
S8 Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. Figure S3. Learning Curves of Naive Rats during DNMTST Test, Related to Figure 6 All naive rats receive the same 6-daily training sessions to improve their ability of making correct choices from approximately 60% on day 1 to > 80% on day 6. All summary graphs show means ± SEM; n = numbers of animals tested in each group. *p < 0.01 versus day 1, Bonferronni post-hoc test after one-way ANOVA for repeated-measure (A: F5,215 = 288.005, p < 0.01; B: F5,155 = 321.444, p < 0.01; C: F5,165 = 359.707, p < 0.01; D: F5,65 = 292.554, p < 0.01).
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. S9 Figure S4. Tat-GluR2 Peptides Do Not Significantly Change Rat Behavior, Related to Figure 6 (A–E) Rats receiving Tat-GluR2 treatment (1.5 mmol/kg, i.p.; 15 pmol/intra-CA1 injection) show similar locomotor activity (A) similar entry ratios (B) and dwell ratios (C) in elevated-plus-maze test, similar latency in catalepsy bar test (D), and spend similar walking time in elevated bean-walking test (E).
All summary graphs show means ± SEM; n = numbers of animals tested in each group. One-way ANOVA: (A) F2,19 = 6.430, p = 0.881; (B) F2,19 = 14.446, p = 0.770;
(C) F2,19 = 11.776, p = 0.810; (D) F2,19 = 9.301, p = 0.833; (E) F2,19 = 9.009, p = 0.827).
S10 Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. Figure S5. SWM Acquisition and Swimming Speed of Rats in the DMTP Water Maze Paradigm. Related to Figure 5 (A) All naive rats significantly acquire SWM during trial 1. (B) Swim speeds during SWM water maze test is not significantly different among groups of rats receiving different treatments. All the summary graphs show means ± SEM; n = numbers of animals tested in each group. *p < 0.01 versus trial 1 (A), Bonferronni post-hoc test after one-way
ANOVA for repeated-measure ([A] F4,164 = 727.667, p < 0.0001) or one-way ANOVA ([B] F5,36 = 9.331, p = 0.802).
Cell 148, 1039–1050, March 2, 2012 ª2012 Elsevier Inc. S11
Part II
CB 1 RECEPTORS AND FEAR RESPONSES
II.1 Résumé de l’article 2: contrôle bimodal des stratégies d’adaptation à la peur par les récepteurs cannabinoïdes CB 1
La peur est une émotion normale exprimée face à un danger. Elle permet d’engager des réponses visant à maintenir l’intégrité de l’organisme. Différentes réponses peuvent être adoptées face à une menace, ce sont les stratégies d’adaptation ou de coping à la peur. Chez l’homme et l’animal, des différences individuelles ont été observées dans le type de stratégie de coping adopté face à un stimulus menaçant. Chez le rongeur, ces comportements défensifs ont été catégorisés en fonction de leur caractère actif ou passif. Les stratégies de coping actif relèvent des comportements ayant pour but de supprimer la source du danger, comme la fuite. Les stratégies de coping passif sont caractérisées par l’inhibition comportementale, comme l’immobilité. Des différences individuelles dans le type de réponse adoptée ont été observées lors de la présentation de stimuli conditionnés. Comme nous l’avons détaillé dans l’introduction, l’activité endogène des récepteurs CB 1 est nécessaire à l’adaptation des réponses de peur conditionnées. Leur signalisation régule les neurotransmissions inhibitrices GABAergiques et excitatrices glutamatergiques dans des structures cérébrales impliquées dans le contrôle de la peur. Nous avons donc supposé que les récepteurs CB 1 contrôlent l’adoption des stratégies d’adaptation à la peur par une modulation des neurones GABAergiques et glutamatergiques.
Nous avons analysé le comportement des souris de type sauvage de souche C57BL/6N, des souris CB 1-KO caractérisées par une délétion constitutive des récepteurs CB 1, ainsi que des souris
GABA-CB 1-KO et Glu-CB 1-KO caractérisées par une délétion de ces récepteurs sur les neurones GABAergiques du prosencéphale et sur les neurones glutamatergiques corticaux, respectivement. Trois tests de peur conditionnée décrits précédemment ont été utilisés: le conditionnement classique de peur au son et les procédures de conditionnement instrumental d’évitement : l’évitement actif et
105 Results l’évitement passif. Au cours du conditionnement classique de peur, différentes réponses ont été mesurées chez les animaux lors de la présentation du SC. La réponse de freezing (absence de tout mouvement excepté ceux dédiés à la respiration), le comportement de digging (creusement dans la litière) ou de rearing (posture debout sur les membres postérieurs). Les réponses de freezing et d’évitement passif sont considérées comme des stratégies de coping passif tandis que les réponses de digging , de rearing et d’évitement actif sont considérées comme des stratégies de coping actif (Koolhaas et al., 1999; De Boer and Koolhaas, 2003; Gozzi et al., 2010; Lázaro-Muñoz et al., 2010; Vicens-Costa et al., 2010). Il a été proposé que l’administration aigüe de THC par voie systémique agit préférentiellement sur les récepteurs CB 1 localisés sur les neurones GABAergiques du prosencéphale ou glutamatergiques corticaux en fonction de la dose (Bellocchio et al., 2010). Cette discrimination pharmacologique a été évaluée chez des souris de type sauvage soumises au conditionnement classique de peur afin de distinguer les conséquences de la stimulation aigüe des récepteurs CB 1 sur ces populations neuronales. La réexpression des récepteurs CB 1 a été induite dans l’amygdale des souris CB 1-KO constitutives afin de vérifier l’implication de cette structure dans le contrôle des stratégies de coping par les récepteurs CB 1.
Chez des souris de type sauvage, nous avons tout d’abord montré que le type de stratégie adopté à la présentation du SC au cours du conditionnement classique de peur prédit les performances d’évitement actif chez le même animal. Cependant, la délétion constitutive des récepteurs CB 1 favorise la réponse de freezing lors du conditionnement classique et facilite les performances d’évitement passif et actif. L’analyse comportementale des souris GABA-CB 1-KO et
Glu-CB 1-KO indique que l’adoption des stratégies de coping passif chez les souris CB 1-KO constitutives est expliquée par la délétion des récepteurs CB 1 sur les neurones glutamatergiques corticaux tandis que la facilitation des stratégies de coping actif est due à la délétion des récepteurs
CB 1 sur les neurones GABAergiques du prosencéphale. Ces résultats sont confirmés par notre approche pharmacologique. Enfin, la réexpression spécifique des récepteurs CB 1 dans l’amygdale des souris CB 1-KO restaure un pattern comportemental similaire à celui observé chez les souris de type sauvage au cours du conditionnement classique de peur.
Ces résultats mettent en évidence que les stratégies de coping à la peur sont soumises à des variations individuelles spontanées qui peuvent être observées au cours des tests de conditionnement de peur. L’adoption des stratégies de coping actif et passif est déterminée, au moins en partie, par le contrôle des transmissions glutamatergiques et GABAergiques exercé par les récepteurs CB 1 du cerveau Nos données suggèrent également que la signalisation des récepteurs CB 1 influence ces stratégies par un contrôle de l’activité de l’amygdale.
Ce travail a fait l’objet d’une publication actuellement sous presse dont les références sont les suivantes :
Metna-Laurent M , Soria-Gomez E, Verrier D, Conforzi M, Jégo P, Lafenêtre P, Marsicano G
(2012) Bimodal Control of Fear-Coping Strategies by CB 1 Cannabinoid Receptors. The Journal of Neuroscience, in press . Metna-Laurent M et Soria-Gomez E partagent le statut de premier auteur.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 106 Part II - CB1 receptors and fear responses
Ma contribution à cette étude à été la conception, la réalisation et l’analyse des expériences comportementales publiées, ainsi que la rédaction des différentes versions du manuscrit. La version intégrale de cet article est présentée ci-après.
II.2 Article 2: Bimodal control of fear-coping strategies by CB 1 cannabinoid receptors
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 107
Bimodal control of fear-coping strategies by CB 1 cannabinoid receptors
Abbreviated title: CB 1 and fear coping strategies
Mathilde Metna-Laurent 1,2,3 , Edgar Soria-Gomez 1,2,3 , Danièle Verrier 1,2, , Martina Conforzi 1,2 , Pierrick Jégo 1,2 , Pauline Lafenêtre 1,2 and Giovanni Marsicano 1,2,*
1INSERM U862, NeuroCentre Magendie, EndoCannabinoids and NeuroAdaptation 2University of Bordeaux, Bordeaux, France. 3Equally contributed
* Correspondence: Dr Giovanni Marsicano Neurocentre Magendie INSERM U862 146, rue Léo Saignat 33076 Bordeaux cedex – France Phone number: ++33 557573756 Fax number: ++33 557573751 Email: [email protected]
Number of pages: 30 Number of figures: 6 Number of words: Abstract (246), Introduction (566), Discussion (1442)
Conflict of Interest: none
Acknowledgements: We thank the staff of the Animal and Genotyping Facilities of the NCM for mouse caring and genotyping and J. Lourenço, C.T. Wotjak, S. Caillé and the members of Marsicano laboratory for valuable suggestions. Support from: INSERM/AVENIR, Region Aquitaine, Fyssen Foundation, CONACyT, EU-FP7 (REPROBESITY, HEALTH-F2-2008-223713), European Research Council (ENDOFOOD, ERC-2010-StG-260515), Fondation pour la Recherche Medicale and UFA/DFH (G2R-FA-151-07). Abstract
To maximize their chances of survival, animals need to rapidly and efficiently respond to aversive situations. These responses can be classified as active or passive and depend on the specific nature of threats, but also on individual fear coping styles. In this study, we show that the control of excitatory and inhibitory brain neurons by type-1 cannabinoid (CB 1) receptors is a key determinant of fear coping strategies in mice. In classical fear conditioning, a switch between initially predominant passive fear responses (freezing) and active behaviors (escape attempts and risk assessment) develops over time. Constitutive genetic deletion of CB 1
-/- receptors in CB 1 mice disrupted this pattern by favoring passive responses. This phenotype can be ascribed to endocannabinoid control of excitatory neurons, because it was reproduced in conditional mutant mice lacking CB 1 receptors from cortical glutamatergic neurons. CB 1 receptors deletion from GABAergic brain neurons led to the opposite phenotype,
9 characterized by the predominance of active coping. The CB 1 receptors agonist ∆ - tetrahydrocannabinol (THC) exerted a biphasic control of fear coping strategies, with lower and higher doses favoring active and passive responses, respectively. Finally, viral re-
-/- expression of CB 1 receptors in the amygdala of CB 1 mice restored the normal switch between the two coping strategies. These data strongly suggest that CB 1 receptor signaling bimodally controls the spontaneous adoption of active or passive coping strategies in individuals. This primary function of the endocannabinoid system (ECS) in shaping individual behavioral traits should be considered when studying the mechanisms of physiological and pathological fear.
1 Introduction
Humans and animals must adopt appropriate fear responses when exposed to threatening situations (LeDoux, 2000; Blanchard et al., 2001). The ability to efficiently cope with potential dangers strongly influences the consequences of aversive stimuli on the organisms
(Hartley and Phelps, 2009). Human and animal studies revealed individual differences in the way of mastering imminent environmental challenges (Schlund et al., 2010; Lázaro-Muñoz et al., 2010; Blanchard et al., 2001). In rodents, defensive behaviors were clustered in either passive (or reactive) coping, or active (or proactive) coping, respectively (Koolhaas et al.,
1999). The predominant response to aversive stimuli of passively-coping animals is immobility, whereas active “copers” tend to adopt behaviors aiming at removing the danger source (De Boer and Koolhaas, 2003; Cain and LeDoux, 2007). These different behavioral strategies rely on distinct physiological mechanisms (Koolhaas et al., 2010). However, most studies in fear conditioning only rely on the analysis of passive coping behaviors. Gozzi et al.
(2010) recently reported that conditioned freezing in mice (i.e. passive coping) was inhibited to favor active coping by pharmaco-genetic manipulations of amygdalar activity, suggesting that inhibition of conditioned freezing may reflect not only a quantitative attenuation of fear but also a qualitative change of fear response.
The endogenous activity of CB 1 receptors is necessary for extinction of freezing in fear conditioning and of passive avoidance learning (Marsicano et al., 2002; Lafenêtre et al., 2007;
Dubreucq et al., 2010), indicating that CB 1 signaling controls passive fear responses. In contrast, the deletion of the CB 1 gene facilitates active avoidance learning (Martin et al.,
2002).
The brain circuitries controlling conditioned freezing have been extensively described. The medial prefrontal cortex (mPFC), the hippocampus and the amygdala as well as downstream nuclei, including the hypothalamus and the periaqueducal grey (PAG), act in concert to
2 mediate appropriate freezing responses (Laviolette et al., 2005; Myers and Davis, 2007;
Roozendaal et al., 2009; Resstel et al., 2009; Sotres-Bayon and Quirk, 2010). CB 1 receptors are enriched in these brain structures and they modulate conditioned freezing in a region- dependent manner (Laviolette and Grace, 2006; Resstel et al., 2009; Kamprath et al., 2010;
Terzian et al., 2011; Dubreucq et al., 2012). However, the mechanisms by which CB 1 receptors modulate active versus passive fear coping strategies have been poorly studied so far.
Interestingly, the behavioral functions of CB 1 receptors also depend on their cell-type localization (Lafenêtre et al., 2009; Bellocchio et al., 2010) and, in particular, on their ability to negatively regulate both GABAergic and glutamatergic neurotransmissions (Chevaleyre et al., 2006). CB 1 receptors may thus influence fear coping strategies acting at glutamatergic and
GABAergic neurons.
By submitting constitutive or cell type-specific CB 1 mutant mice to classical and instrumental fear conditioning, we reveal that the ECS determines the fear coping strategies by exerting a specific control on glutamatergic or GABAergic neurons. In addition, CB 1 expression in the amygdaloid area is sufficient to guarantee normal fear coping strategies, suggesting that the
ECS drives fear coping styles at least in part by modulating the activity of neurons located in the amygdala.
3 Material and Methods
Animals
Experiments were in agreement with the Committee on Animal Health and Care of INSERM and French Ministry of Agriculture and Forestry (authorization number, 3306369).
2- to 4-month old male C57BL/6N (JANVIER, France) and constitutive or conditional CB1- mutant mice and their wild-type littermates were maintained on a 12 h light/dark cycle (light on, 7 am) with food and water ad libitum . All the mutant lines were in a mixed genetic background, with a predominant C57BL/6NCrl contribution (6-7 backcrossing generations).
All the animals used in experiments involving mutant mice were littermates. For constitutive
-/- CB 1 mice (Marsicano et al., 2002), the parents of the experimental animals were always heterozygous for the mutation. Conditional mutant mice were obtained as described by
flox/flox crossing mice carrying "floxed" CB 1 alleles ( CB 1 ; Marsicano et al., 2003) with the Cre- expressing transgenic mouse lines NEX-Cre and Dlx5/6-Cre (Goebbels et al., 2006; Monory
flox/flox;NEX-Cre -/- flox/flox;Dlx5/6-Cre et al., 2006) to obtain CB 1 (called Glu-CB 1 ) and CB 1 mice (called
-/- GABA-CB 1 ), respectively (Monory et al., 2006; Monory et al., 2007; Bellocchio et al.,
-/- -/- 2010). To allow a direct comparison of GABA-CB 1 and Glu-CB 1 mice with the same wild-
-/- -/- +/+ type littermate mice, GABA-CB 1 and Glu-CB 1 and wild-type GABA-Glu-CB 1 littermates (called WT) derived from a double mutant line (Bellocchio et al., 2010). Briefly,
-/- -/- GABA-CB 1 mice were mated with Glu-CB 1 mice in order to obtain a first generation of double mutant mice lacking CB 1 on both GABAergic and glutamatergic neurons
-/- flox/flox (GABA/Glu-CB 1 ), which were then crossed with CB 1 females (phenotypically wild- type), in order to avoid potential influence of the mother’s genotype on the offspring phenotype of the experimental animals (Bellocchio et al., 2010). Mutant mice were genotyped at the age of 2 weeks and re-genotyped after the experiments by PCR on tail tissue as described (Marsicano et al., 2002; Monory et al., 2006; Bellocchio et al., 2010) using the
4 flox/flox following primer sets: CB 1 : 5’-GCTGTCTCTGGTCCTCTTAAA, 5´-
GGTGTCACCTCTGAAAACAGA and 5´-CCTACCCGGTAGAATTAGCTT. Cre forward for NEX-Cre (for Glu-CB1 -/-): 5’-TCTTTTTCATGTGCTCTTGG; Cre forward for Dlx5/6-
Cre (for GABA-CB1 -/-): 5’-AGCAATCGCACTCACAACAGA, Cre reverse for both lines:
5’-CGCGCCTGAAGATATAGAAGA. Previous extensive anatomical characterizations showed that the mutant mice used in the present study carry deletions of CB 1 receptors (i)
-/- from all the cells of the body ( CB 1 mice; Marsicano et al., 2002), (ii) mainly from cortical glutamatergic neurons in the dorsal telencephalon, including neurons located in neocortex, paleocortex, archicortex, hippocampal formation and cortical portions of the amygdala (Glu -
-/- CB 1 ; Monory et al., 2006; Monory et al., 2007; Bellocchio et al., 2010), or (iii) mainly from
-/- forebrain GABAergic neurons (GABA -CB 1 ; Monory et al., 2006; Monory et al., 2007;
Bellocchio et al., 2010).
All experiments took place during the light phase. For fear conditioning and shock sensitivity, mice were single housed 7 days before testing. For active and passive avoidance, mice were maintained in groups (2 to 4 per cage).
Drugs
In the pharmacological approach, ∆9-tetrahydrocannabinol (THC, Sigma-Aldrich, France) was dissolved with 2% ethanol, 2% cremophor in sterile saline solution. Naive C57BL/6N mice were weighted and received an intra-peritoneal injection of vehicle, 0.3 mg/kg, 1 mg/kg, or 3 mg/kg of THC one hour before the CS re-exposure session.
AAV-mediated re-expression of CB 1 receptors into the amygdala of constitutive mutant
-/- CB 1 mice
AAV vector synthesis
5 The AAV-CB 1 constructions were generated according to Klugmann et al. (2011). Briefly, the cDNA encoding a hemagglutinin (HA)-tagged rat CB 1-receptor was cloned into an AAV expression cassette containing the 1.1kb CMV immediate early enhancer/chicken β-actin hybrid promoter (CBA), the woodchuck hepatitis virus post-transcriptional regulatory element
(WPRE), and the bovine growth hormone polyadenylation sequence (bGHpA) flanked by
AAV2 inverted terminal repeats (pAAV-CB 1). Packaging of pseudotyped AAV1/2 chimeric vectors with equal ratios of AAV1 and AAV2 capsid proteins was performed as described
(Klugmann et al., 2005) and genomic titers were determined by quantitative real-time PCR of vector genomes using primers for WPRE (During et al., 2003).
Intra-amygdala AAV-CB 1 injection
-/- Constitutive CB 1 mice were deeply anaesthetized with a ketamine (0.0125 mg/ml) and xylazine (0.001 mg/ml) mixture (0.2 ml/mouse, i.p). 0.5 µl of either an empty viral construct
11 (EV) or AAV-CB 1 (6 × 10 viral genomes/ml) was injected bilaterally aiming at the amygdala (AP: -2.0 mm, ML: ±2.0 mm, DV: -2.0 mm, from bregma). Vector delivery was performed at a rate of 0.1 µl/min using a mini-pump (Harvard Apparatus, France) with 33G injectors (Plastics One, VA, USA) in a stereotaxic frame (David Kopf Instruments, CA,
USA). The injectors were left in place for an additional minute to allow vector diffusion.
Mice were submitted to the fear conditioning and active avoidance procedures 5 and 6 weeks following virus administration, respectively.
Fear conditioning
Classical fear conditioning was carried out as described (Dubreucq et al., 2010) in a square conditioning box (Imetronic, France) made of grey Perspex (length: 26 cm; width: 18 cm; height: 25 cm) with a metal grid floor and located in a soundproof chamber (length: 55 cm; width: 60 cm; height: 50 cm). A video camera placed above the conditioning box allowed
6 observation of animals’ behaviors. On the conditioning day, each mouse was placed into the conditioning chamber and left free to explore for 2 min. A single footshock (0.5 mA, 1 s,
“squared” scrambler) was then co-emitted during the last second of a 20 sec-tone (1.5 kHz, 60 dB). Twenty-four hours after conditioning, mice were placed back in the chamber in their home cage in order to attenuate novelty-induced exploratory behaviors. They were then exposed to the tone (conditioned stimulus, CS) for 8 min preceded by 2-min pre-tone.
Freezing (i.e. lack of movements except those associated with breathing), and active coping
(i.e. digging, rearing and wall-sniffing/rearing) were scored (De Boer and Koolhaas, 2003;
Gozzi et al., 2010) by an experimenter blind to mouse genotypes, and expressed as percentage of time. To ensure that the observed behaviors were specifically induced by the CS, the maximal freezing and active coping percent times per min induced by the tone were compared to pre-tone levels for each mouse, revealing a consistent and robust increase of both behaviors during CS presentation (p<0.001 for all genotypes, paired t-test, data not shown).
Active and passive avoidance
The two-way active and passive avoidance tests took place in two-compartment shuttle boxes
(40x10x15cm, Imetronic, France) located in dark soundproof cubicles supplied with infrared sensors and video cameras. The floor of the shuttle boxes was made of steel cabled grid releasing electric pulses (“squared” scrambler). Tone generators were fixed on the top of each compartment. On day 1, animals were allowed to explore both compartments for 15 min.
Twenty-four hours later, mice underwent four daily sessions made of fifty active or passive avoidance trials. In the active avoidance procedure, a trial began with the tone start, which was accompanied 10 s later by the footshock (0.2 mA) until the animal changed of compartment (maximal shock duration: 15 s). Passive avoidance was conducted in the same conditions, except that changing of compartment within tone-on duration induced an acute footshock (2 s). These symmetric tasks allowed a direct comparison of the mouse groups in
7 active or passive avoidance responses. The subsequent trial began from the compartment where the mouse was detected after a 20 s inter-trial interval (ITI). In active avoidance, a correct response was assigned when the mouse transited to the other compartment before shock delivery. In passive avoidance, a correct response was assigned when the mouse remained in the original compartment during tone presentation. Avoidance performances were expressed in percentage of correct responses.
Shock sensitivity testing
Shock sensitivity was examined in each mutant line. Naïve mice were introduced into the fear conditioning chamber (see above) and submitted to 5 footshocks (1 s) of increasing intensities
(from 0.1 to 0.5 mA) every 30 s. The first shock intensity at which flinching, vocalizing and jumping reactions appeared was taken as sensitivity threshold.
In situ hybridization of CB 1 mRNA
-/- Following the behavioral tasks, control mice (n=10), and AAV-CB 1-treated CB 1 mice
-/- -/- (n=13) as well as naive GABA-CB 1 , GLU-CB 1 and WT littermates mice were sacrificed by cervical dislocation, their brain quickly removed, frozen on dry ice and stored at -80º C until sectioning in a cryostat (14 µm, Microm HM 500M, Microm Microtech). The DIG- labeled riboprobes against mouse CB 1 receptors were prepared as described (Marsicano and
Lutz, 1999; Marsicano, 2003; Monory et al., 2006). In situ hybridization of CB 1 mRNA was performed according to the standard procedure used in the laboratory (Bellocchio et al., 2010;
Lourenço et al., 2010; Dubreucq et al., 2012). Signal amplification was achieved using the
TSA™ Plus System Cyanine 3/Fluorescein kit (Perkin Elmer, France). Blocking and wash buffers were prepared according to the manufacturer’s protocol. Slides were analyzed by epifluorescence microscopy at 5X (Leica, France).
8 -/- Only the AAV-CB 1-treated CB 1 mouse brains showing a bilateral major expression of CB 1 mRNA in the basolateral amygdala (BLA) and central amygdala (CeA) nuclei were included in the final analysis (n=8 out of 13).
Statistical analysis
Total freezing and active coping time percentages during CS presentation were compared using t-test (constitutive CB 1 mutants) and 1-way ANOVA (conditional CB 1 mutants). Time- course data were analyzed using 2-way ANOVA for repeated measures. Correlations were assessed using the Pearson’s r linear regression analysis. To facilitate between genotype analyses of avoidance learning in the conditional mutant mice, the area under the curve
(AUC) values were also calculated for each animal (trapezoid rule; Bura et al., 2007) and compared using 1-way ANOVA. Bonferroni’s post-hoc test was applied when appropriate.
Active and passive avoidance responses of wild-type mice were also compared between each sessions using 1-way ANOVA (data not shown), ensuring that both avoidance trainings induced an increase of performances.
9 Results
Behavioral analysis of the coping strategies adopted in fear conditioning and avoidance learning in C57BL/6N mice
In classical fear conditioning in rodent, re-exposure to a CS previously paired with an aversive stimulus induces a strong freezing response (LeDoux, 2000). As CS presentation continues, the freezing rate progressively declines, due to extinction and/or habituation processes (Myers and Davis 2007). However, CS presentation can also elicit proactive behaviors (Gozzi et al., 2010) considered as attempts to actively cope over the danger source
(Koolhaas et al., 1999; Blanchard et al., 2001; De Boer and Koolhaas, 2003).
The first experiment aimed at describing the temporal expression of both freezing and active coping induced by a conditioned tone in C57BL/6N mice. Twenty-four hours following conditioning, the 8-min tone presentation provoked a behavioral pattern that can be distinguished into 3 temporal phases: (i) Immediate tone re-exposure induced a strong freezing response associated with a weak active coping rate ( behavior X time interaction:
F(7,126) =26,22, p<0.001; minute 1, p<0.001; Fig. 1A; "phase 1"). (ii) During "phase 2", simultaneous decrease of freezing and increase of active coping were observed, resulting in an equivalent expression of both response types ( minute 2-3, p>0.05; Fig. 1A). (iii) In "phase 3", active coping eventually overcame freezing ( minute 4-8, p<0.001; Fig. 1A). Importantly, freezing and active coping represented only a portion of the whole observation (<60%) indicating that these responses were not mutually exclusive. Thus, the coordinated decrease of freezing and increase of active behaviors seems to be the consequence of a "temporal switch" in the individual CS-induced coping strategy.
Avoidance learning in rodents was proposed to depend on the individual predominant coping style (Koolhaas et al., 1999). For instance, rats showing a strong tendency to freeze in fear conditioning were those performing the poorer in active avoidance learning (Lázaro-Muñoz et
10 al., 2010; Vicens-Costa et al., 2010). Previously fear-conditioned mice were submitted to one active avoidance session. Mice displayed 50±7% of total avoidance responses, reaching ~80% of correct responses at the end of the session ( F(4,32) =4.67, p<0.01 ; Fig. 1B).
Interestingly, active avoidance responses were negatively correlated with individual freezing and positively related to active coping during "phase 2" of tone presentation, when the levels of the two behaviors reached similar levels (Fig. 1C). Thus, behavioral tendencies during fear conditioning can predict active avoidance levels.
Constitutive deletion of CB 1 receptors prevents the temporal switch of coping strategies in classical fear conditioning
The total percentages of freezing and active behaviors during CS presentation were compared
-/- +/+ between constitutive mutant CB 1 mice and CB 1 littermate controls. CS presentation
-/- induced stronger freezing ( t=3.17; p<0.01 ; Fig. 2A) and weaker active coping in CB 1 mice
+/+ +/+ (t =3.81; p<0.001 ; Fig. 2B) as compared to CB 1 littermates. CB 1 mice displayed the three temporal "phases" leading to the switch from freezing to active coping ( behavior X time
-/- interaction: F (7,154) =42.29, p<0.001 ; Fig. 2C). In contrast, CB 1 mice displayed a prolonged
"phase 1", maintaining higher freezing as compared to active coping until the third/fourth minute of tone exposure ( behavior X time interaction: F (7,154) =11.18, p<0.001 ; minute 1-2, p<0.01; Fig. 2D). Interestingly, mutant mice lacked "phase 3", as freezing and active coping subsequently overlapped until the end of the session ( minute 3-8, p>0.05; Fig. 2D), indicating that the switch between freezing and active coping did not occur. These data suggest that CB 1 signaling controls the expression and the temporal relationship between passive and active responses to fear-conditioned stimuli.
CB 1 receptors bimodally control fear coping strategies
11 To assess whether the site of CB 1 neuronal expression influences the type of fear coping
-/- -/- strategies, we tested the Glu-CB 1 , GABA-CB 1 and WT littermates mice in our fear conditioning procedure. The mouse genotype influenced the overall freezing levels
-/- -/- (F(2,42) =4.49; p<0.05 ; Fig. 3A), with GABA-CB 1 freezing less than Glu-CB 1 ( p<0.05 ).
-/- Conversely, the CS induced stronger active coping behaviors in GABA-CB 1 as compared to
-/- both WT and Glu-CB 1 littermates ( F(2,42) =8.58; p<0.001; p<0.01 for both comparisons; Fig.
3B).
WT animals displayed a similar time-course behavioral pattern as C57BL/6N mice, with the presence of the three temporal "phases" of tone-induced responses ( behavior X time
-/- interaction: F (7,168) =20.45, p<0.001; Fig. 3C). In contrast, Glu-CB 1 mice never displayed dominant active coping responses to tone presentation, lacking the "phase 3" of the temporal behavioral expression ( behavior X time interaction: F (7,238) =36.62; p<0.001 ; minute 3-8, p>0.05; Fig. 3D). The early CS re-exposure did not elicit a dominant freezing response in
-/- GABA-CB 1 mice (lack of "phase 1"), which instead promptly adopted highly dominant active coping ( behavior X time interaction: F (7,182) =12.66; p<0.001 ; minute 1, p>0.05; minute
2-8, p<0.001; Fig. 3E).
The state of fear has been conceived as the set of defensive behaviors elicited by a threat
(Blanchard et al., 2001). In our conditions, the sum of CS-induced freezing and active coping might therefore provide an indication of the overall state of fear. Interestingly, these cumulated values did not differ between genotypes (Fig. 3F). Altogether, these data suggest that CB 1 receptor expression in GABAergic or cortical glutamatergic brain neurons does not significantly impact on overall fear learning and expression, but it determines the strategy to cope with conditioned fear stimuli.
Bimodal control of CB 1 receptors on fear avoidance learning
12 -/- Ubiquitous deletion of CB 1 receptors in CB 1 mice strengthened both active ( genotype main effect: F (1,16) =5.66; p<0.05 ; Fig. 4A) and passive ( genotype main effect: F (1,20) =4.99; p<0.05 ;
Fig. 4B) avoidance learning. In conditional mutant mice, the genotype affected both active avoidance ( genotype main effect: F (2,60) =4.80; p<0.05 ; Fig. 4C) and passive avoidance performances ( genotype main effect: F (2,71) =13.11; p<0.001 ; Fig. 4D). AUC analyses
-/- uncovered an improved active avoidance performance in GABA-CB 1 mice as compared to
-/- WT littermates ( F(2,60) =4.46; p<0.05; GABA-CB 1 vs WT, p<0.05; Fig. 4C). Conversely,
-/- Glu-CB 1 exhibited higher passive avoidance responses as compared to controls
-/- (F(2,71) =12.49; p<0.001 ; Glu-CB 1 vs WT, p<0.01; Fig. 4D).
Importantly, the observed phenotypes could not be assigned to altered locomotion or pain perception as mutants and controls displayed similar initial numbers of ITI transitions and equivalent pain responses to footshocks (data not shown). These data reveal the differential impact of CB 1-dependent control of glutamatergic and GABAergic neurons on active and passive avoidance learning, respectively.
THC dose-dependently alters the coping style in classical fear conditioning.
Cannabinoid receptor agonists such as THC often dose-dependently induce opposite behavioral effects (Moreira and Lutz, 2008; Bellocchio et al., 2010). We thus assessed whether an acute administration of THC at 0.3, 1 or 3 mg/kg could lead to distinct coping styles when mice were re-exposed to the CS. THC biphasically affected the total amount of freezing ( F(3,40) =13.12; p<0.001 ; Fig. 5A) and of active coping ( F(3,40) =8.47; p<0.001 ; Fig.
5B). Mice receiving the lowest dose of THC showed a reduced freezing response ( p<0.05,
Fig. 5A) as compared to vehicle-injected mice. While the freezing time of animals treated with the intermediate dose of THC (1 mg/kg) did not differ from controls ( p>0.05 ), mice injected with 3 mg/kg of THC increased overall freezing time as compared to controls
13 (p<0.05, Fig. 5A). However, the CS induced weaker active coping behaviors in the THC (3 mg/kg)-treated mice as compared to both vehicle- ( p<0.01 ) and THC (0.3 mg/kg)-treated mice ( p<0.001; Fig. 5B). Again, THC administered at 1 mg/kg did not alter the total amount of active coping behaviors as compared to vehicle (p>0.05; Fig. 5B).
Vehicle-treated animals displayed the three temporal "phases" of CS-induced responses
(behavior X time interaction: F (7,140) =44.61, p<0.001; Fig. 5C). Strikingly, the low dose of
THC (0.3 mg/kg) prevented the initial dominant freezing response to the tone (lack of "phase
1") and favored early prevailing active coping behaviors ( behavior X time interaction:
F(7,140) =11.38; p<0.001 ; minute 1, p>0.05; minute 2-8, p<0.001; Fig. 5D). Mice submitted to the intermediate dose of THC (1 mg/kg) showed the three temporal "phases" of CS-induced responses, albeit the shift from dominant freezing to active coping behaviors was slightly delayed ( behavior X time interaction: F (7,140) =15.69; p<0.001 ; Fig. 5E). In contrast, animals receiving the highest dose of THC (3 mg/kg) displayed a prolonged "phase 1", keeping a longer freezing response as compared to active coping until the sixth minute of CS exposure
(behavior X time interaction: F (7,140) =5.44; p<0.001 ; minute 1-6, p<0.05; Fig. 5F) and lacked the "phase 3" ( minute 7-8, p>0.05 ), thereby indicating that the switch between freezing and active coping did not occur.
Overall, these data show that the exogenous activation of CB 1 receptors exerts a biphasic effect on both the intensity and the prevailing type of fear-elicited behaviors along CS exposure.
-/- Viral re-expression of CB 1 receptors in the amygdala of CB 1 mice restores the transition from passive to active coping responses.
The amygdala is a key structure governing the expression of defensive responses to aversive stimuli (LeDoux, 2000; Myers and Davis, 2007; Ehrlich et al., 2009; Gozzi et al., 2010). CB 1
14 receptor mRNA is expressed at high levels on GABAergic interneurons and at lower levels on glutamatergic neurons of the BLA, whereas it is present at low levels in GABAergic neurons of the CeA (Marsicano and Lutz, 1999; Monory et al., 2007; Bellocchio et al., 2010; Fig. 6A).
As expected, CB 1 mRNA expression was absent in GABAergic neurons of the amygdala of
-/- -/- GABA-CB 1 mice (including CeA neurons and BLA interneurons, Fig. 6A). In Glu-CB 1 mice, CB 1 mRNA was absent in glutamatergic neurons of the BLA (Fig. 6A). Notably, the control exerted by CB 1 receptors signaling on both inhibitory and excitatory amygdalar neurotransmissions has been recently associated with the temporal adaptation of conditioned freezing responses (Kamprath et al., 2010) and in the control of synaptic plasticity (Marsicano et al., 2002; Azad et al., 2004). Thus, we asked whether the selective expression of CB 1 receptors within the amygdala would be sufficient to induce a pattern of fear responses similar to that observed in wild-type animals. To this aim, we used the AAV-technology to re-express
-/- the CB 1 receptor gene in amygdalar cells of constitutive CB 1 mice. In situ hybridization analysis revealed that the injection of AAV-CB 1 determined the expression of CB 1 mRNA in a portion of the amygdala complex, including lateral, basolateral, medial and central amygdala (Fig. 6B).
-/- In the fear conditioning test, AAV-CB 1-treated CB 1 mice showed a reduced total freezing
-/- time during the CS presentation as compared to control CB 1 mice injected with an empty
-/- virus (AAV-EV-treated-CB 1 t=2.15; p<0.05 ; Fig. 6C). However, the treatment did not influence the overall active coping behaviors expression time ( t=0.40; p>0.05 ; Fig.6D). As
-/- expected, the control AAV-EV-treated-CB 1 mice did not display the temporal transition from the CS-induced freezing to active coping behaviors ( behavior X time interaction:
F(7,126) =10.91; p<0.001; minute 1, p<0.001; minute 2-8, p>0.05; Fig. 6E), lacking the "phase
3" of the fear responses (see Fig. 2). Notably, the temporal switch was restored in AAV-CB 1-
-/- treated CB 1 mice ( behavior X time interaction: F (7,98) =15.08; p<0.001; minute 1, p<0.001;
15 minute 2-4, p>0.05; minute 5, 7-8, p<0.01; Fig. 6F), which exhibited a behavioral pattern
+/+ very similar to C57BL6/N and CB 1 mice (Fig. 1A and 2C).
Thus, re-expression of CB 1 receptors in the amygdaloid complex is sufficient to rescue wild-
-/- type-like fear coping strategies in constitutive CB 1 mice.
16 Discussion
In natural settings, “active” or “passive” strategies represent the most efficacious responses to avoid potential dangers (Blanchard et al., 2010). The adoption of such coping styles depends on the nature of the threat, but they are also largely subserved by individual propensity towards active or passive fear responses (Koolhaas et al., 2010). The present data confirm that these individual strategies are observable in fear conditioning experiments (Cain and LeDoux,
2007; Gozzi et al., 2010) and show (i) that the balanced control of excitatory and inhibitory neurons by CB 1 receptors is a determinant of the individual preference towards active or passive coping with aversive situations, (ii) that pharmacological CB 1 receptor activation exerts a biphasic control of fear coping strategies, and (iii) that CB 1 receptors in the amygdala play a key role in the balance between passive and active coping strategies.
Conditioned fear responding is influenced by individual variations in defensive strategies
Extinction of conditioned freezing in wild-type mice was accompanied by an increase of active coping responses. This might just indicate that the decrease in freezing occurs when fear attenuates, with animals progressively returning to normal activities. This interpretation does not hold for several reasons. First, digging, rearing and wall-sniffing are considered defensive behaviors that rodents express when confronted with potentially threatening situations (Koolhaas et al., 1999; De Boer and Koolhaas, 2003; Cain and LeDoux, 2007;
Blanchard et al., 2010). Indeed, in our conditions, these responses were clearly distinguishable from the home-cage behaviors of mice in the absence of any aversive CS.
Second, freezing and active coping durations reached less than 60% of the total observation time, indicating that they did not represent the whole behavioral repertoire of the animals.
Third, both freezing and active behaviors reached higher levels during tone presentation than
17 during the pre-tone period in all animals, indicating that they were tone-induced (i.e. fear- elicited) behavioral responses. Finally, as previously reported in rats (Lázaro-Muñoz et al.,
2010; Vicens-Costa et al., 2010), we were able to measure the relationship between the type of coping adopted by individual mice in fear conditioning and active avoidance performances.
Altogether, these observations indicate that passive or active fear coping strategies likely represent individual behavioral traits maintained over different experimental conditions.
Freezing is considered a reliable index of fear memory in fear conditioning experiments
(Maren, 2008). However, conditioned fear can be expressed through a wide variety of behavioral responses and some individuals can still adopt predominant active strategies to cope with the CS, suggesting that weak freezing does not necessarily imply a low fear state and/or lower fear memory (Maren, 2008; Gozzi et al., 2010). Indeed, “low freezer” animals in classical fear conditioning are the best performers in active avoidance (Lázaro-Muñoz et al.,
2010; Vicens-Costa et al., 2010; present results), implying paradoxically that animals with
"lower" levels of fear and/or memory in one test would display higher fear and/or memory in the other test. Great and obvious differences exist between Pavlovian fear conditioning and instrumental avoidance learning (Hartley and Phelps, 2009). However, our and others' observations suggest that the performances in these fear-based memory tests might be not exclusively due to the ability to process fear memory, but also to intrinsic "coping styles"
(Koolhaas et al., 2010; Gozzi et al., 2010; Vicens-Costa et al., 2010).
CB 1 receptor signaling determines the coping style to fear conditioned stimuli by acting upon GABAergic and glutamatergic neurons.
In line with previous reports, ubiquitous CB 1 receptor inactivation led to a predominant freezing response in classical fear conditioning, and enhanced passive and active avoidance performances (Marsicano et al., 2002; Martin et al., 2002; Lafenêtre et al., 2007; Resstel et
18 al., 2009; Dubreucq et al., 2010). Thus, the consequences of a total CB 1 receptor deletion on the potentiation of either passive or active fear coping strategies are task-specific. This could argue against the existence of preserved individual coping styles among threat situations.
However, in classical fear conditioning, the deletion of CB 1 receptors in GABAergic neurons favored active coping whereas the mutation restricted to cortical glutamatergic neurons promoted passive coping. The two specific mutations induced higher performances in the active and passive versions of avoidance learning, respectively. Therefore, our data reveal that the CB 1 receptor-dependent control of inhibitory and excitatory brain neuronal activity is a key determinant of fear coping strategies. Accordingly, Kamprath et al. (2009) reported that the suppression of CB 1 receptors from glutamatergic neurons strengthened freezing after a fear-sensitization procedure, supporting that the endogenous control of CB 1 receptor signaling on excitatory neurotransmission mediates adaptation of passive fear responding. Recent studies pointed out very specific roles of CB 1 receptors expressed on other restricted neuronal populations in the expression of conditioned fear responses. In particular, mice bearing a CB 1 receptor deletion in the hypothalamus and mediobasal amygdala showed a dominant active coping strategy (i.e. digging behavior) in tone-cued fear conditioning (Dubreucq et al., 2012).
Conversely, the specific suppression of CB 1 receptors on type-1 dopamine receptors- expressing cells (Monory et al., 2007), including the medium spiny neurons of the striatum, favored freezing responses in both tone-cued and contextual fear conditioning settings
(Terzian et al., 2011). Thus, the endocannabinoid control of fear responses is exerted at different brain sites. However, our data strongly suggest that CB 1 receptors, by balancing inhibitory and excitatory brain neuronal activity, are one of the biological determinants of individual fear coping styles.
Exogenous THC administration exerted a biphasic effect on fear coping strategies, with low doses favoring active coping and higher doses promoting passive responses in classical fear
19 conditioning, respectively. Such biphasic effects of cannabinoids were also reported on other behavioral dimensions including unconditioned anxiety and food intake (Moreira and Lutz,
2008; Bellocchio et al., 2010). Interestingly, the low and high doses of THC induced opposite
-/- -/- fear coping styles than those observed respectively in the Glu-CB 1 and GABA-CB 1 mice, suggesting that these biphasic effects might be mediated by cell-type specific CB 1 receptors as previously shown (Puighermanal et al., 2009; Bellocchio et al., 2010; Piet et al., 2011).
Overall, the bidirectional effects of acute exogenous stimulation of CB 1 receptors support a balanced control of the ECS on fear coping strategies.
CB 1 receptors in the amygdala are sufficient to guarantee normal fear coping strategies.
The functional neuroanatomy of fear learning and processing has been extensively characterized in the last decades (LeDoux, 2000; Myers and Davis, 2007; Ehrlich et al., 2009;
Hartley et Phelps, 2009). Much less, conversely, is known on the neuroanatomical and neuroendocrine substrates of individual coping styles (Koolhaas et al., 2010), rendering difficult the identification of the site(s) where the ECS controls active or passive fear responses.
The amygdaloid complex is a major integration site for learning and adaptation of passive and active responses to fear conditioned stimuli (LeDoux, 2000; Myers and Davis, 2007; Hartley and Phelps, 2009; Choi et al., 2010; Lázaro-Muñoz et al., 2010), and amygdalar CB 1 receptors are necessary for a normal expression of conditioned freezing (Marsicano et al., 2002;
Kamprath et al., 2010; Tan et al., 2010, 2011). The virus-mediated re-expression of CB 1
-/- receptors in CB 1 mice show that amygdalar CB 1 receptor signaling is sufficient to mediate adaptation of conditioned freezing and, importantly, to restore a normal switch between passive and active coping.
20 BLA neurons contain CB 1 receptors both at glutamatergic neurons and GABAergic interneurons (Marsicano and Lutz, 1999; Katona et al., 2006; Bellocchio et al., 2010).
Amygdalar CB 1 receptor signaling is necessary for acquisition of conditioned freezing to olfactory CS, likely through a control of the firing activity of principal neurons projecting to the prelimbic division of the mPFC (Laviolette and Grace, 2006; Tan et al., 2010, 2011).
Thus, due to its main presynaptic terminal localization (Piomelli, 2003), the re-expression of
CB 1 receptor might reinstate normal fear coping by acting at intra-amygdala circuits and/or at
BLA-mPFC projections.
The CeA is emerging as a key subcortical structure mediating the adoption of active and passive fear conditioned responses (Choi et al., 2010; Gozzi et al., 2010; Ehrlich et al., 2009).
CB 1 receptors are expressed in the CeA and their endogenous signaling is required for short- term extinction of conditioned freezing (Kamprath et al., 2010). The ECS might thus control the expression of active and passive fear strategies through a modulation of GABAergic and glutamatergic neurotransmissions within the CeA.
The amygdala is also connected to downstream structures, including the PAG and the hypothalamus, that are crucially involved in the expression of defensive responses and accompanying neuroendocrine signals (Roozendaal et al., 2009, Resstel et al., 2009).
Interestingly, the local stimulation of CB 1 receptors in the PAG attenuates freezing expression in contextual fear conditioning (Resstel et al., 2008; Moreira et al., 2009). Therefore, presynaptic CB 1 receptors expressed on BLA-PAG projections might also participate in the behavioral consequences of our viral-mediated reinstatement of CB 1 gene expression.
Conclusion
Inefficient fear coping characterizes the pathophysiology of important psychiatric diseases,
21 including phobias and post-traumatic stress disorders. Our data highlight a complex, task- dependent regulation of the ECS on defensive strategies that is driven, at least in part, by a bimodal control of CB 1 receptors on glutamatergic and GABAergic neuronal activities.
Overall, this study suggests that the differential impact on active or passive fear-coping strategies should be considered for designing therapeutic approaches involving the modulation of CB 1 signaling.
22 References
Azad SC, Monory K, Marsicano G, Cravatt BF, Lutz B, Zieglgänsberger W, Rammes G (2004) Circuitry for associative plasticity in the amygdala involves endocannabinoid signaling. J Neurosci 24:9953-9961.
Bellocchio L, Lafenêtre P, Cannich A, Cota D, Puente N, Grandes P, Chaouloff F, Piazza PV, Marsicano G (2010) Bimodal control of stimulated food intake by the endocannabinoid system. Nat Neurosci 13:281-283.
Blanchard DC, Griebel G, Blanchard RJ (2001) Mouse defensive behaviors: pharmacological and behavioral assays for anxiety and panic. Neurosci Biobehav Rev 25:205–218.
Blanchard DC, Griebel G, Pobbe R, Blanchard RJ (2010) Risk assessment as an evolved threat detection and analysis process. Neurosci Biobehav Rev 35:991-998.
De Boer SF, Koolhaas JM (2003) Defensive burying in rodents: ethology, neurobiology and psychopharmacology. Eur J Pharmacol 463:145–161.
Bura SA, Castane A, Ledent C, Valverde O, Maldonado R. (2007) Genetic and pharmacological approaches to evaluate the interaction between the cannabinoid and cholinergic systems in cognitive processes. Br J Pharmacol 150:758–765.
Cain CK, LeDoux JE (2007) Escape from fear: A detailed behavioral analysis of two atypical responses reinforced by CS termination. J Exp Psychol Anim Behav Process 33:451-463.
Chevaleyre V, Takahashi KA, Castillo PE (2006) Endocannabinoid-mediated synaptic plasticity in the CNS. Annu Rev Neurosci 29:37–76.
Choi JS, Cain CK, LeDoux JE (2010) The role of amygdala nuclei in the expression of auditory signaled two-way active avoidance in rats. Learn Mem 17:139–147.
Dubreucq S, Kambire S, Conforzi M, Metna-Laurent M, Cannich A, Soria-Gomez E, Richard E, Marsicano G, Chaouloff F (2012) Cannabinoid type 1 receptors located on single-minded 1–expressing neurons control emotional behaviors. Neuroscience 204:230-244.
Dubreucq S, Koehl M, Abrous DN, Marsicano G, Chaouloff F (2010) CB1 receptor deficiency decreases wheel-running activity: Consequences on emotional behaviours and hippocampal neurogenesis. Exp Neurol 224:106–113.
During MJ, Young D, Baer K, Lawlor P, Klugmann M (2003) Development and optimization of adeno-associated virus vector transfer into the central nervous system. Methods Mol Med 76:221-236.
Ehrlich I, Humeau Y, Grenier F, Ciocchi S, Herry C, Lüthi A (2009) Amygdala Inhibitory Circuits and the Control of Fear Memory. Neuron 62:757-771.
Goebbels S, Bormuth I, Bode U, Hermanson O, Schwab MH, Nave KA (2006) Genetic targeting of principal neurons in neocortex and hippocampus of NEX-Cre mice. Genesis 44: 611–621.
23 Gozzi A, Jain A, Giovanelli A, Bertollini C, Crestan V, Schwarz AJ, Tsetsenis T, Ragozzino D, Gross CT, Bifone A (2010) A Neural Switch for Active and Passive Fear. Neuron 67:656- 666.
Hartley CA, Phelps EA (2009) Changing Fear: The Neurocircuitry of Emotion Regulation. Neuropsychopharmacology 35:136-146.
Kamprath K, Plendl W, Marsicano G, Deussing JM, Wurst W, Lutz B, Wotjak CT (2009) Endocannabinoids mediate acute fear adaptation via glutamatergic neurons independently of corticotropin-releasing hormone signaling. Genes Brain Behav 8:203–211.
Kamprath K, Romo-Parra H, Häring M, Gaburro S, Doengi M, Lutz B, Pape HC (2010) Short-term adaptation of conditioned fear responses through endocannabinoid signaling in the central amygdala. Neuropsychopharmacology 36:652–663.
Katona I, Rancz EA, Acsady L, Ledent C, Mackie K, Hajos N, Freund TF (2001) Distribution of CB1 cannabinoid receptors in the amygdala and their role in the control of GABAergic transmission. J Neurosci 21:9506-9518.
Klugmann M, Symes CW, Leichtlein CB, Klaussner BK, Dunning J, Fong D, Young D, During MJ (2005) AAV-mediated hippocampal expression of short and long Homer 1 proteins differentially affect cognition and seizure activity in adult rats. Mol Cell Neurosci 28:347-360.
Klugmann M, Goepfrich A, Friemel CM, Schneider M (2011) AAV-Mediated Overexpression of the CB1 Receptor in the mPFC of Adult Rats Alters Cognitive Flexibility, Social Behavior, and Emotional Reactivity. Front Behav Neurosci 5:37.
Koolhaas JM, De Boer SF, Coppens CM, Buwalda B (2010) Neuroendocrinology of coping styles: Towards understanding the biology of individual variation. Front Neuroendocrinol 31:307–321.
Koolhaas JM, Korte SM, De Boer SF, Van Der Vegt BJ, Van Reenen CG, Hopster H, De Jong IC, Ruis MAW, Blokhuis HJ (1999) Coping styles in animals: current status in behavior and stress-physiology. Neurosci Biobehav Rev 23:925–935.
Lafenêtre P, Chaouloff F, Marsicano G. (2007) The endocannabinoid system in the processing of anxiety and fear and how CB1 receptors may modulate fear extinction. Pharmacol Res 56:367-381.
Lafenêtre P, Chaouloff F, Marsicano G (2009) Bidirectional regulation of novelty-induced behavioral inhibition by the endocannabinoid system. Neuropharmacology 57:715–721.
Laviolette SR, Lipski WJ, Grace AA (2005) A subpopulation of neurons in the medial prefrontal cortex encodes emotional learning with burst and frequency codes through a dopamine D4 receptor-dependent basolateral amygdala input. J Neurosci 25:6066-6075.
Laviolette SR, Grace AA (2006) Cannabinoids Potentiate Emotional Learning Plasticity in Neurons of the Medial Prefrontal Cortex through Basolateral Amygdala Inputs. J Neurosci 26:6458-6468.
24 Lázaro-Muñoz G, LeDoux JE, Cain CK (2010) Sidman Instrumental Avoidance Initially Depends on Lateral and Basal Amygdala and Is Constrained by Central Amygdala-Mediated Pavlovian Processes. Biol Psychiatry 67:1120-1127.
LeDoux JE (2000) Emotion circuits in the brain. Annu Rev Neurosci 23:155–184.
Lourenço J, Cannich A, Carta M, Coussen F, Mulle C, Marsicano G (2010) Synaptic activation of kainate receptors gates presynaptic CB1 signaling at GABAergic synapses. Nat Neurosci 13:197–204.
Maren S (2008) Pavlovian fear conditioning as a behavioral assay for hippocampus and amygdala function: cautions and caveats. Eur J Neurosci 28:1661–1666.
Marsicano G, Goodenough S, Monory K, Hermann H, Eder M, Cannich A, Azad SC, Cascio MG, Gutiérrez SO, van der Stelt M, López-Rodriguez ML, Casanova E, Schütz G, Zieglgänsberger W, Di Marzo V, Behl C, Lutz B (2003) CB1 Cannabinoid Receptors and On- Demand Defense Against Excitotoxicity. Science 302:84–88.
Marsicano G, Lutz B (1999) Expression of the cannabinoid receptor CB1 in distinct neuronal subpopulations in the adult mouse forebrain. Eur J Neurosci 11:4213–4225.
Marsicano G, Wotjak CT, Azad SC, Bisogno T, Rammes G, Cascio MG, Hermann H, Tang J, Hofmann C, Zieglgänsberger W, Di Marzo V, Lutz B (2002) The endogenous cannabinoid system controls extinction of aversive memories. Nature 418:530–534.
Martin M, Ledent C, Parmentier M, Maldonado R, Valverde O (2002) Involvement of CB1 cannabinoid receptors in emotional behaviour. Psychopharmacology 159:379-387.
Monory K, Blaudzun H, Massa F, Kaiser N, Lemberger T, Schütz G, Wotjak CT, Lutz B, Marsicano G (2007) Genetic Dissection of Behavioural and Autonomic Effects of 9- Tetrahydrocannabinol in Mice. Plos Biol 5:e269.
Monory K, Massa F, Egertová M, Eder M, Blaudzun H, Westenbroek R, Kelsch W, Jacob W, Marsch R, Ekker M (2006) The Endocannabinoid System Controls Key Epileptogenic Circuits in the Hippocampus. Neuron 51:455-466.
Moreira FA, Lutz B (2008). The endocannabinoid system: emotion, learning and addiction. Addict Biol 13:196–212.
Myers KM, Davis M (2007) Mechanisms of fear extinction. Mol Psychiatry 12:120–150.
Piet R, Garenne A, Farrugia F, Le Masson G, Marsicano G, Chavis P, Manzoni OJ (2011) State-Dependent, Bidirectional Modulation of Neural Network Activity by Endocannabinoids. J Neurosci 31:16591–16596.
Puighermanal E, Marsicano G, Busquets-Garcia A, Lutz B, Maldonado R, Ozaita A (2009) Cannabinoid modulation of hippocampal long-term memory is mediated by mTOR signaling. Nat Neurosci 12:1152–1158.
Resstel LBM, Moreira FA, Guimarães FS (2009) Endocannabinoid System and Fear Conditioning. Vitam Horm 81:421-440.Roozendaal B, McEwen BS, Chattarji S (2009) Stress, memory and the amygdala. Nat Rev Neurosci 10:423-433.
25 Schlund MW, Siegle GJ, Ladouceur CD, Silk JS, Cataldo MF, Forbes EE, Dahl RE, Ryan ND (2010) Nothing to fear? Neural systems supporting avoidance behavior in healthy youths. NeuroImage 52:710-719.
Sotres-Bayon F, and Quirk GJ (2010) Prefrontal control of fear: more than just extinction. Curr Opin Neurobiol 20: 231–235.
Tan H, Lauzon NM, Bishop SF, Bechard MA, Laviolette SR (2010) Integrated Cannabinoid CB1 Receptor Transmission within the Amygdala–Prefrontal Cortical Pathway Modulates Neuronal Plasticity and Emotional Memory Encoding. Cereb Cortex 20:1486-1496.
Tan H, Lauzon NM, Bishop SF, Chi N, Bechard M, Laviolette SR (2011) Cannabinoid Transmission in the Basolateral Amygdala Modulates Fear Memory Formation via Functional Inputs to the Prelimbic Cortex. J Neurosci 31: 5300-5312.
Terzian AL, Drago F, Wotjak CT, Micale V (2011) The dopamine and cannabinoid interaction in the modulation of emotions and cognition: assessing the role of cannabinoid CB1 receptor in neurons expressing dopamine D1 receptors. Front Behav Neurosci 5:49.
Vicens-Costa E, Martinez-Membrives E, López-Aumatell R, Guitart-Masip M, Canete T, Blázquez G, Tobena A, Fernández-Teruel A (2010) Two-way avoidance acquisition is negatively related to conditioned freezing and positively associated with startle reactions: A dissection of anxiety and fear in genetically heterogeneous rats. Physiol Behav 103:148-156.
26 Figures legends
Figure 1.
Freezing and active coping responses to tone in classical fear conditioning predict active avoidance performances in C57BL/6N mice. A. Temporal expression of freezing and active coping responses to the conditioned tone. In the “phase 1” (P 1), mice show a strong freezing expression associated to low active coping. In the “phase 2” (P 2, grey background) freezing and active coping reached an equivalent expression percentage. In the “phase 3” (P 3), mice displayed dominant active coping behaviors and attenuated freezing response. B. Total percentage of correct response in a single active avoidance training session (left) and within session learning curve (blocks of 10 trials, right). C. Correlation of individual freezing and active coping time percentage observed at the second minute of CS presentation against the total active avoidance performances. Data are mean ± SEM expressed as percentage. *p<0.05;
*** p<0.001, n=10.
Figure 2.
Constitutive deletion of CB 1 receptors prevents the temporal shift between freezing and
-/- +/+ active coping in fear conditioning. A. Total freezing of CB 1 and CB 1 littermates during
-/- +/+ the 8-min CS presentation. B. Total active coping of CB 1 and CB 1 littermates during the
+/+ 8-min CS presentation. C. Time-course of freezing and active coping in CB 1 mice. D.
-/- Time-course of freezing and active coping in CB 1 mice. Data are mean ± SEM. *p<0.05;
** p<0.01; *** p<0.001, n=12 per genotype.
Figure 3.
Conditional deletion of CB 1 receptors in GABAergic or glutamatergic neurons differentially alters the shift between freezing and active coping in fear conditioning. A.
-/- -/- Total freezing of Glu-CB 1 (n=18), GABA-CB 1 (n=14) and WT (n=13) littermates during
-/- -/- the 8-min CS presentation. B. Total active coping of Glu-CB 1 , GABA-CB 1 and WT
27 littermates during the 8-min CS presentation. C. Time-course of freezing and active coping of
WT mice (note the presence of the 3 "phases" of behavior). D. Time-course of freezing and
-/- active coping of Glu-CB 1 mice (note the absence of "phase 3"). E. Time-course of freezing
-/- and active coping of GABA-CB 1 mice (note the absence of "phase 1"). F. Sum of individual freezing and active coping time scores as percentage of the total 8 min-tone exposure (left) or per 60s-bins (right). Data are mean ± SEM. *p<0.05; ** p<0.01; *** p<0.001.
Figure 4.
Neuronal type-specific regulation of active and passive avoidance learning by CB 1
-/- +/+ receptors. A, B. Time-course of avoidance responses of CB 1 and CB 1 littermates in the
-/- +/+ -/- +/+ active [ A; CB 1 (n=10), CB 1 (n=8)] and passive [ B; CB 1 (n=9), CB 1 (n=13)] versions of the 2-way avoidance paradigm. C, D . Time-course of behavioral responses of conditional
-/- -/- -/- CB 1 and WT in the active [ C; Glu-CB 1 (n=17), GABA-CB 1 (n=16), WT (n=30)] and
-/- -/- passive [ D; Glu-CB 1 (n=21), GABA-CB 1 (n=17), WT (n=36)] versions of the 2-way avoidance paradigm. Data are mean ± SEM expressed as percentage of avoidance responses at each training session. *p<0.05; ** p<0.01.
Figure 5.
Biphasic effects of systemic THC on freezing and active coping behaviors in classical fear conditioning. A. Total freezing in C57BL/6N mice treated with vehicle (n=11), THC 0.3 mg/kg (n=11), 1 mg/kg (n=11) or 3 mg/kg (n=11) during the 8-min CS presentation. B. Total active coping of the vehicle, THC 0.3 mg/kg, THC 1 mg/kg and THC 3 mg/kg groups during the 8-min CS presentation. C. Time-course of freezing and active coping of the vehicle- injected mice (note the presence of the 3 "phases" of behavior). D. Time-course of freezing and active coping of the THC 0.3 mg/kg group (note the absence of "phase 1"). E. Time- course of freezing and active coping of the THC 0.3 mg/kg group (note the presence of the 3
"phases" of behavior). F. Time-course of freezing and active coping of the THC 3 mg/kg
28 group (note the absence of the “phase 3”). Data are mean ± SEM. *p<0.05; ** p<0.01;
*** p<0.001.
Figure 6.
-/- AAV-mediated re-expression of CB 1 receptors in the amygdala of constitutive CB 1 mutant mice restores fear coping strategies. A,B. Representative photomicrograph (5X magnification) of CB 1 mRNA expression analyzed by fluorescent in situ hybridization in the
-/- -/- amygdala in WT, GABA-CB 1 and Glu-CB 1 littermate mice (A) and in control and AAV-
-/- CB 1-treated CB 1 mice (B). Note the expression of CB 1 in the central amygdaloid (CeA) and basolateral nuclei of the amygdala (BLA) in the AAV-CB 1-treated mouse as compared to control. opt, optic tractus. C. Total freezing in control (n=11) and AAV-CB 1-treated (n=8)
-/- CB 1 littermate mice during the 8-min CS presentation. D. Total active coping in control and
-/- AAV-CB 1-treated CB 1 mice during the 8-min CS presentation. E. Time-course of freezing
-/- and active coping of the control CB 1 mice (note the absence of the “phase 3”). F. Time-
-/- course of freezing and active coping of the AAV-CB 1-treated CB 1 mice (note the presence of the 3 “phases” of behavior). Data are mean ± SEM. *p<0.05; ** p<0.01; *** p<0.001.
29
GENERAL DISCUSSION
109
Part I RESUME DE LA DISCUSSION GENERALE
L’objectif de ce travail était de préciser les mécanismes cellulaires par lesquels le SEC module les processus de mémoire. D’une part, nous avons mis en évidence que les cannabinoïdes exogènes, incluant le THC, induisent des déficits de mémoire de travail spatiale par la stimulation des récepteurs
CB 1 exprimés sur les astrocytes et non sur les neurones glutamatergiques et GABAergiques. La stimulation de ces récepteurs par les cannabinoïdes induit une dépression à long-terme de la transmission excitatrice dépendante des récepteurs NMDA de l’hippocampe in vivo . Cette modification de la plasticité synaptique de l’hippocampe et les déficits de mémoire induits par les cannabinoïdes sont sous-tendus par des mécanismes cellulaires similaires, suggérant qu’une modification de la communication entre les astrocytes et les neurones participe à l’induction du déficit de mémoire de travail par les cannabinoïdes.
La discussion de ces résultats s’articule autour de trois points : nous proposons tout d’abord d’évaluer le lien de causalité suggéré par nos travaux entre le mécanisme de dépression à long-terme décrit in vivo et les conséquences de l’administration de cannabinoïdes sur la mémoire de travail observées au niveau comportemental. Notamment, nous proposons l’emploi d’une nouvelle stratégie génétique permettant, chez des souris dépourvues de l’expression constitutive des récepteurs CB 1, d’induire une réexpression des récepteurs CB 1 uniquement sur les cellules GFAP-positives, c'est-à-dire, principalement sur les astrocytes (Figure 18). La caractérisation électrophysiologique et comportementale de ces animaux permettra de tester si la présence des récepteurs CB 1 sur les astrocytes suffit à induire l’altération de la mémoire de travail par les cannabinoïdes et que ces effets s’accompagnent des modifications de la communication cellulaire décrites dans notre étude, et ainsi de mieux évaluer les relations entre ces deux phénomènes. Nos résultats montrent que les cannabinoïdes induisent une forme de dépression à long-terme dépendante de la signalisation des récepteurs NMDA dans l’hippocampe in vivo . Cependant, ces mécanismes diffèrent de ceux observés dans d’autres études in vitro (voir Introduction part II.4.3). Nous proposons que la destruction du réseau astrocytaire provoqué par la procédure de coupe des tranches de cerveau puisse expliquer
111 General Discussion ces différences observées in vivo et in vitro . En particulier, l’interface entre les vaisseaux sanguins et les neurones assurée par le réseau astrocytaire permet l’apport énergétique nécessaire aux fonctions neuronales (Hajos and Moddy, 2009; Ivanov et al., 2011). Ce processus métabolique astrocytaire n’est plus assuré in vitro mais substitué par un apport énergétique artificiel. Or, il a été montré que la composition du fluide cérébrospinal (par exemple sa concentration en glucose) influence la transmission et la plasticité neuronale enregistrée sur les tranches de cerveau. En perspective de ce travail, nous proposons donc d’évaluer l’effet de l’administration de cannabinoïdes in vivo à l’aide du modèle de souris double mutantes pour les connexines 43 et connexines 30 chez lequel le couplage astrocytaire est perturbé (Wallraff et al., 2006; Rouach et al., 2008). Le troisième point de cette discussion aborde le rôle endogène des récepteurs CB 1 dans les processus de mémoire. Notamment, nos résultats indiquent que ces récepteurs localisés sur les astrocytes ne sont pas nécessaires à l’apprentissage de la tâche de mémoire de travail spatiale dans nos conditions. Compte-tenu de l’importance des communications entre les astrocytes et les neurones dans le contrôle des fonctions de mémoire, nous supposons que la signalisation endogène des récepteurs CB 1 astrogliaux pourrait
être impliquée dans d’autres formes de mémoire. L’analyse du phénotype des souris GFAP-CB 1-KO au cours de tests évaluant d’autres processus mnésiques permettrait de tester cette hypothèse et d’apporter des informations importantes sur le rôle du SEC et des astrocytes dans les processus d’apprentissage et de mémoire.
Le second volet de ce travail a permis de mettre en évidence le rôle spécifique des récepteurs CB 1 exprimés sur les neurones glutamatergiques et GABAergiques dans le contrôle des réponses de peur apprises. Nos résultats indiquent d’une part que différentes stratégies de coping (passif ou actif) peuvent être observées chez des souris de type sauvage soumises à un conditionnement classique de peur. Le type de stratégie adopté par ces sujets prédit les performances au cours de l’apprentissage de l’évitement actif. Ces données suggèrent l’existence d’une variabilité individuelle spontanée dans l’adoption du type de réponse de peur apprise. Bien que la réponse de freezing soit un index de peur conditionnée très étudié, d’autres comportements, comme les réponses de digging et de rearing , peuvent être induits par des stimuli aversifs innés ou conditionnés (Blanchard et al., 1990; Dielenberg et al., 2001; De Boer and Koolhaas, 2003; Cain and Ledoux, 2007; Blanchard et al., 2011). D’autres études ont montré que l’adoption de ces réponses peut être déterminée par des variations individuelles stables (Driscoll and Bättig, 1982 ; Weshler, 1995; Ursin and Olff, 1993; Koolhaas et al., 1999). Par exemple, des lignées de rats ont été sélectionnées en fonction du type de réponses adoptées au cours de tests de conditionnement aversif. Les rats RLA ( Roman Low Avoidance ) ont de faibles performances d’apprentissage de l’évitement actif et montrent une réponse de freezing intense lorsqu’ils sont soumis à un conditionnement classique de peur. A l’inverse, les rats RHA ( Roman High Avoidance ) montrent des performances d’évitement actif élevées tandis qu’ils sont caractérisés par une réponse de freezing de faible intensité lors du conditionnement classique de peur. Ainsi, ces données supportent l’idée selon laquelle la variabilité des stratégies défensives adoptées par les individus peut influencer l’interprétation des réponses observées lors de test de conditionnement de peur.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 112 Part I - Résumé de la discussion générale
Nos données suggèrent que l’augmentation de la transmission GABAergique induite par la délétion des récepteurs CB 1 sur les neurones GABAergiques pourrait favoriser les stratégies de coping actif tandis que l’augmentation de la transmission glutamatergique provoquée par leur délétion sur les neurones glutamatergiques pourrait favoriser les stratégies de coping passif, indiquant que le SEC est un déterminant physiologique important dans l’adoption du type de stratégie de coping à la peur. La réexpression des récepteurs CB 1 restreinte à l’amygdale restaure un pattern de réponse similaire à celui des souris de type sauvage, suggérant que le SEC influence l’adoption des réponses de peur par un contrôle de l’activité de l’amygdale. D’autres études ont montré que l’administration d’agonistes des récepteurs GABA A facilite l’adoption de stratégies de coping actif (Lal and Forster, 1990;
Fernandez-Teruel et al., 1991; Escorihuela et al., 1993). De plus, la stimulation des récepteurs CB 1 dans l’amygdale empêche l’apprentissage de l’évitement passif, cet effet étant réversé par la co- administration de NMDA. Ainsi, l’ensemble de ces données supporte un contrôle bimodal des récepteurs CB 1 sur les transmissions excitatrices et inhibitrices dans les stratégies de réponses à la peur apprise et le rôle clef de l’amygdale dans cette fonction. Enfin, nous proposons d’étudier le comportement des lignées de souris CB 1-KO dans le paradigme de l’enfouissement défensif (defensive burying ) qui est très utilisé pour distinguer l’adoption des stratégies de coping actif et passif (Pinel et al., 1980; Koolhaas et al., 1999; De Boer and Koolhaas, 2003). Dans ce test, une sonde électrifiée est introduite dans la cage de l’animal. En réponse à un contact avec la sonde, l’animal peut adopter un comportement de freezing ( coping passif) ou une réponse de digging afin d’enfouir la sonde ( coping actif). Ainsi, l’emploi de ce test permettrait de confirmer l’implication des récepteurs CB 1 dans l’adoption des stratégies défensives et également d’en préciser les mécanismes cellulaires.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 113
Part II CELLULAR MECHANISMS INVOLVED IN THE CONTROL OF
WORKING MEMORY BY CB 1 RECEPTORS
The first objective of this thesis work was to precise on which brain cell type(s) CB 1 receptors are necessary for normal SWM performances and mediate the deleterious effect of cannabinoids on this function.
We could discriminate the contribution of astroglial and neuronal CB 1 receptors in WM by generating the GFAP-CB 1-KO mouse line, bearing a specific deletion of CB 1 receptors on astrocytes.
Importantly, we provided a quantification of CB 1 receptor expression levels on the GFAP-expressing astrocytes in the hippocampus. Indeed, our data show that CB 1 receptors are detectable on approximately 40 % of these cells.
In our SWM test, we observed that CB 1 receptors expressed on cortical glutamatergic neurons are necessary for normal WM performances, whereas CB 1 receptors located on astrocytes and on forebrain GABAergic neurons are dispensable for such performances. However, we showed that the
WM impairment induced by acute THC in mice does not require CB 1 receptors expressed neither on cortical glutamatergic neurons nor on forebrain GABAergic neurons, but instead depends on the presence of CB 1 receptors expressed on astrocytes.
Furthermore, we proposed a possible mechanism accounting for the THC-induced WM impairment by the stimulation of astroglial CB 1 receptors. Exogenous cannabinoids, including THC, were able to induce an NMDA receptors-dependent LTD at hippocampal CA1 synapses of living wild- type mice and rats. Similarly, this CB-LTD is not dependent on the “neuronal” CB 1 receptors but it requires CB 1 receptors expression on astrocytes. We described several mechanisms underlying CB- LTD, including the involvement of AMPA receptors endocytosis and of NR2B-containing NMDA receptors subunits. Drugs interfering with both AMPA receptors endocytosis and NR2B-containing NMDA receptors subunits also prevent the cannabinoid-induced WM impairments in wild-type mice and rats. Thus, CB-LTD and cannabinoid-induced WM deficits share common brain mechanisms,
115 General Discussion
suggesting that acute cannabinoids impair WM by inducing LTD in CA1 through astroglial CB 1 receptors.
II.1 Cannabinoid-induced LTD and working memory impairment: looking for causality
We showed that cannabinoid-induced LTD and SWM impairments depend on astroglial CB 1 receptor activation, NMDA receptor activation and AMPA receptor endocytosis. Although these shared molecular and cellular mechanisms strongly suggest that CB-LTD may cause SWM, it is not a sufficient argument for stating that such a causal relationship actually exists. Establishing irrefutable causal relationships between events is a complex and long task supposing to address several conditions. For instance, Hill (1965) proposed several criteria outlining the minimal conditions needed to establish a causal relationship between two events. For instance, one condition is the consistency of the results, referring to an association between two events that must be replicated in studies using different settings or methods. The GFAP-CB 1-KO mouse line developed in our laboratory is the only currently available model to specifically manipulate astroglial CB 1 receptors activity in vivo . Therefore, due to technological limitations, it is extremely difficult to meet this criterion at present.
Importantly, a causative link between CB-LTD and cannabinoid impairment of SWM would necessarily imply a temporal relationship between both phenomena (Hill, 1965). Our study did not explore this point. For instance, it might be proposed to assess the simultaneous presence of cannabinoid-induced LTD and impairment of SWM by recordings fEPSP in the CA1 of freely moving animals during the SWM task. However, any further manipulation able to alter both these phenomena, including the deletion of CB 1 receptors on astrocytes, could be due to independent mechanisms, thereby impeding to establish an irrefutable causal link between them.
Another argument that supports causality is the strength of the relation between the two events
(Hill, 1965). We showed that several factors, including the presence of astroglial CB 1 receptors, are necessary to observe cannabinoid-induced LTD and SWM. Is the presence of CB 1 receptors on astrocytes a sufficient condition to observe CB-LTD and SWM impairment by cannabinoids? We are currently developing a mouse line in order to obtain such an astrocyte-specific genetic “rescue” of CB1 receptor expression in constitutive CB 1-KO mice. The genetic “rescue” strategy derived from the Cre/loxP system (Sauer and Henderson, 1989). Briefly, a “stop cassette” flanked by LoxP sites
(“floxed-stop”) is introduced immediately upstream of the coding sequence of the CB 1 gene by homologous recombination to generate a mutant mouse line by standard procedures (Figure 18). Mice obtained by this approach (so-called “ CB 1-stop”) bear no expression of CB 1 receptors because the presence of the “floxed-stop” cassette will block the transcriptional activity of the promoter region of the gene. Therefore, these mice will be phenotypically very similar to the constitutive CB 1-KO. The 2 crossing of the CB 1-stop mice with the GFAP-CreERT transgenic mice (Hirrlinger et al., 2006) will induce a Cre-mediated deletion of the “stop cassette” restricted to GFAP-expressing cells. By this way, CB 1 receptors will be “re-expressed” in astrocytes of “rescued” mice. Current “rescue” strategies of genetic deletions are based on viral or transgenic re-expression of genes in KO mice. However,
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 116 Part II - Cellular mechanisms involved in the control of working memory by CB1 receptors these approaches do not allow controlling the site and the levels of expression of the “rescued” gene, very likely leading to the presence of the targeted protein in other cell types and at different levels than 2 the endogenous one. Importantly, the crossing of the “CB 1-stop” mice with the GFAP-CreERT mice will induce a re-expression of CB 1 receptors only in the cells where the regulatory sequences of CB 1 receptors are normally activated, and the expression levels will be the same as in wild-type animals.
Indeed, if it is possible to achieve a specific re-expression of CB 1 receptors restricted to astrocytes using this strategy, the administration of cannabinoids should induce CB-LTD and SWM impairment in these animals in a similar manner as in wild-type animals. If so, we could conclude that CB 1 receptors expressed on astrocytes, but not on other cells of the body, are necessary and sufficient to mediate the cannabinoid-induced LTD and SWM impairment, thereby increasing the possibility that these two vents are indeed causally linked.
GFAP-expressing cells
Lox/P STOP Lox/P CB1 CB1
Cre Cre Lox/P CB1 CB1
Figure 18. Representation of the genetic strategy employed to induce a specific rescue of CB 1 receptors expression in GFAP-expressing cells of constitutive CB 1-KO mice.
Hill (1965) also proposed the criterion of coherence following which it is necessary to evaluate claims of causality within the context of the current state of knowledge within related fields. We proposed that the induction of a NMDA receptor-dependant LTD in CA1 by cannabinoids leads to SWM impairment. In this context, it is relevant to discuss the relationship between hippocampal LTD and WM.
There is evidence for a parallel processing of information between the prefrontal cortex and the hippocampus in WM tasks (Floresco et al., 1997; Izaki et al., 2001; Lee and Kesner, 2003; Porter et al., 2000). For example, only the double inactivation of the prefrontal cortex and the hippocampus in a delayed-non-matching-to-place task with ITI of 10 s provoked a deficit in SWM performance, suggesting that the consequences of the inactivation of one structure can be compensated by the activity of the other (Lee and Kesner, 2003). However, when these delays exceeded 10 s, the single inactivation of the hippocampus, but not of the prefrontal cortex, was sufficient to impair SWM in rats. These results indicate that the integrity of the hippocampus is necessary for maintaining and process information in WM for delays longer than 10 s in rats (Lee and Kesner, 2003). We found that local HU210 administration in the hippocampus impairs delayed-non-matching-to-place with ITI of 30 s. Our
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 117 General Discussion
results are thus consistent with this view, although we cannot rule out an involvement of CB 1 receptors located in the prefrontal cortex in the cannabinoid-induced SWM impairment.
While hippocampal LTP is largely correlated with long-term reference memory, little is known about the memory correlates of hippocampal LTD (Zeng et al., 2001; Kemp and Manahan-Vaughan, 2004; Neves et al., 2008; Collingridge et al., 2010). One proposed reason is that it is difficult to selectively alter LTD without modifying LTP (Kemp and Manahan-Vaughan, 2004; Collingridge et al., 2010). However, genetic studies could uncover proteins that are especially involved in LTD. For instance, the inhibition of calcineurin in the forebrain of mice causes both impairment in SWM and a decrease in hippocampal LTD in hippocampal CA1 slices although the mechanisms underlying this form of LTD are not fully investigated (Zeng et al., 2001). Similarly, genetic suppression of the phosphatase PP2A in the mouse forebrain impairs NMDA receptor-dependent LTD in hippocampal CA1 slices; this effect being associated with a decrease of WM performances in the T-maze (Nicholls et al., 2008). Exogenous D-serine, which binds to the glycine site of NMDA receptors, strengthens NR2B-NMDA receptor-dependant LTD at CA1 synapses in mouse brain slices and is also known as a WM enhancer (Duffy et al., 2007; Zhang et al., 2008; Bado et al., 2011). These results suggest that NMDA receptor-dependant LTD has a positive role in WM. In this view, LTD would not be a mechanism that constrains WM. This is not consistent with our interpretation that CB-LTD mediates the cannabinoid-induced SWM impairment. However, LTD facilitation has also been associated with memory defects. For instance, pathogenic β-amyloid treatment enhanced LTD that has been correlated with WM failures occurring in Alzheimer disease (Huntley and Howard, 2010). Furthermore, cognitive impairments observed following stress restrain also reduces LTD induction threshold (Yang et al., 2005; Wong et al., 2007). Overall, the link between LTD and memory is not clear. There are several types of LTD that may differ according to their induction protocol or expression mechanisms (Kemp and Manahan-Vaughan, 2004; Collingridge et al., 2010). Most of the studies evaluating the relationship between NMDA receptor-dependent LTD and WM used electrical stimulation (e.g. Low frequency stimulation) as induction protocols (Duffy et al. 2007; Zhang et al., 2008; Nicholls et al., 2008). Conversely, we described an NMDA receptor-dependent LTD induced by acute chemical stimulation of astroglial CB 1 receptors. Indeed, this form of LTD could be negatively linked to WM.
Again, the genetic “rescue” of CB 1 receptor expression on astrocytes might be useful to challenge this hypothesis. Another crucial distinction between most of these studies and ours is the use of in vivo recordings.
II.2 Roles of CB 1 receptors in vitro versus in vivo
Our study uncovered that the modulation of CB 1 receptor expression leads to distinct mechanistic consequences in in vivo electrophysiological recordings as compared to those reported in recordings from brain slices (Kawamura et al., 2006; Navarrete et Araque, 2010; see General
Introduction section II.2). Indeed, the activation of CB 1 receptors decreases excitatory transmission in both hippocampal and cerebellar brain slices. This effect is thought to depend on pre-synaptic CB 1 receptors at glutamatergic synapses (Kawamura et al., 2006; Monory et al., 2006; Navarrete and
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 118 Part II - Cellular mechanisms involved in the control of working memory by CB1 receptors
Araque, 2010). Moreover, the stimulation of CB 1 receptors in slices induces several forms of LTD that are not dependent on post-synaptic NMDA receptors (Chevaleyre and Castillo, 2006; Kano et al., 2009). Navarrete and Araque (2010) pointed to a synaptic potentiation of excitatory transmission following activation of CB 1 receptors on astrocytes at CA3-CA1 synapses in brain slices. Conversely, we found in vivo that (i) exogenous activation of CB 1 receptors on astrocytes induce a long-term decrease of excitatory transmission, (ii) this CB-LTD is not prevented following deletion of CB 1 receptors on glutamatergic terminals at these synapses, and (iii) CB-LTD depends on NR2B- containing NMDA receptors.
Functional discrepancies have been already reported between in vivo and in vitro assessments of synaptic plasticity. Indeed, dissociation between LTP in vitro and neural plasticity in vivo has been shown (Hensch et al., 1998; Gruart and Delgado-García, 2007). For instance, genetic manipulation of PKA leads to an absence of in vitro LTP recorded in the visual cortex but failed to do so in intact animals (Hensch et al., 1998). Similarly, pharmacological blockade of NR2B-NMDA receptors prevents hippocampal LTD in vivo in both young and adult rats (Fox et al., 2006; our study) but not in hippocampal slices of young rats (Bartlett et al., 2007).
Few studies assessed the effect of cannabinoids on synaptic transmission in the hippocampus in vivo. Heyser et al. (1993) recorded extracellular spiking activity of hippocampal neurons from the CA1 pyramidal layer in rats performing a DMTS task. The THC (2.0 mg/kg)-induced impairment of DMTS was associated with a decrease of 50% of the firing rate of these neurons, consistently with the inhibition of excitatory transmission observed in our study following cannabinoid administration. Interestingly, acute systemic THC (1.0 mg/kg) administration in mice increases ERK1/2 phosphorylation in hippocampal neurons in a CB 1 receptor-dependant way. This effect was abolished by co-administration of MK801 (0.1 mg/kg), suggesting that THC effects on ERK1/2 pathway depends on NMDA receptors activity (Derkinderen et al., 2003). However, the similar potentiation of ERK1/2 phosphorylation following CB 1 receptor stimulation observed in hippocampal slices was resistant to MK801, indicating that this effect was not dependant on NMDA receptors in this preparation (Derkinderen et al., 2003). Therefore, this study argues for a similar distinction regarding the contribution of NMDA receptors in the effects of THC in vitro and in vivo. Given the importance of ERK1/2 phosphorylation for several forms of synaptic plasticity and memory (Atkins et al., 1998), these distinct in vivo and in vitro pathways may be involved in the differential effects of CB 1 receptor stimulation on synaptic plasticity as assessed in living animals versus in hippocampal slices.
We proposed that slicing procedures might lead to these different consequences due to a disruption of the network organization of neurons, but most importantly of astrocytes, possibly altering
CB 1 receptor-dependent astrocytes-neurons communication processes occurring in vivo. The neuro- vascular coupling by astrocytes in the intact brain is disrupted in brain slices. Therefore, the brain energetic demands are not ensured in vitro , but instead depend on artificial experimental conditions that determine, for instance, slice oxygenation and the composition of the energetic substrates in artificial cerebrospinal fluid including glucose concentration (Hájos and Mody, 2009; Ivanov and Zilberter, 2011). Both oxygenation conditions and energetic supply strongly influence synaptic
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 119 General Discussion functions, but also the ability of neuronal networks to generate oscillations in brain slices (Hájos et al., 2009; Ivanov et al., 2011). It has been shown that excitatory transmission in mouse hippocampal slices depends on the rate of oxygen perfusion (Ivanov and Zilberter, 2011). Moreover, most studies use glucose as a unique energetic substrate for maintaining neuronal activity in slices. However, glucose alone, even at high concentrations, is not sufficient to mimic in vivo synaptic functions (Zilberter et al., 2010). Astrocytes provide lactate to neurons as a main energy source for neuronal functions in vivo (see General Introduction section III.3.1) and supplying artificial cerebrospinal fluid with lactate enhances synaptic plasticity (Ivanov and Zilberter, 2011), suggesting that the conditions of slices preparation in vitro could mask synaptic events occurring in vivo by altering astrocytic metabolic functions. Importantly, astrocytes control glutamatergic synaptic transmission by the supply of energetic metabolites through astrocytes networks. Preventing astroglial gap junctional connections in the astrocyte-specific double-KO mice for Cx43 and Cx30 increases glutamatergic transmission (Wallraff et al., 2006; Rouach et al., 2008). CB-LTD depends on a tight regulation of astrocytes- neurons excitatory transmission in the hippocampus. Therefore, it is plausible that CB-LTD was not observed in vitro because the astrocyte network itself is disrupted. Indeed, one would expect that preventing network astrocytic communications in vivo would also prevent CB-LTD. To directly test this possibility, one could propose to assess the effect of cannabinoids on hippocampal excitatory transmission in conditions where astrocytic gap junctional coupling is blocked, for instance in the astrocyte-specific double-KO mice for Cx43 and Cx30 (Wallraff et al., 2006).
How could cannabinoid-induced LTD at the synaptic level lead to WM impairment? As previously mentioned, it has been showed that cannabinoids induce impairment of spatial alternation performances associated with a de-synchronization of neuronal oscillatory rhythms in the hippocampus (Robbe et al., 2006; Robbe and Buzsaki, 2009), suggesting that CB-LTD mediated by astroglial CB 1 receptors might be involved in the perturbation of the rhythm generation by cannabinoids. Consistently, both hippocampal LTD and astrocyte signalling have been implicated in neuronal network activities. Astrocytes are in contact with thousands of synapses and are able to release gliotransmitters which regulate neuronal excitability, synaptic transmission and plasticity. It is thus likely that they strongly affect neural network properties. Thus, astrocytes can simultaneously activate populations of neurons in CA1 as shown by paired recordings (Fellin et al., 2004). Astrocytic Ca 2+ elevation and subsequent glutamate release lead to the synchronous excitation of clusters of pyramidal neurons through an NMDA receptor dependent signalling, suggesting that excitatory gliotransmission may contribute to neuronal synchronization (Fellin et al., 2004; Carmignoto and Fellin, 2006). In addition, adenosine synthesized from ATP released by astrocytes control cortical slow oscillations, a fundamental network activity that characterizes non-REM (rapid eye movements) sleep which is necessary for proper memory functions (Fellin et al., 2009; Halassa et al., 2009). The role of slow-wave sleep is thought to downscale synaptic strength through shared molecular mechanisms involved in LTD, including the frequency of occurrence / induction and the dependence of glutamatergic transmission (Tononi and Cirelli, 2006). Interestingly, AEA administration in rats increases forebrain dialyzed adenosine levels associated to an increased slow-wave sleep through a
CB 1 receptor-dependent mechanism (Murillo-Rodriguez et al., 2003). Therefore, it is likely that
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 120 Part II - Cellular mechanisms involved in the control of working memory by CB1 receptors exogenous cannabinoids dampen WM performances by disturbing astrocytes’ ability to control neuronal network activity.
II.3 Endogenous CB 1 receptor signalling and memory
Our study reveals that the presence of astroglial CB 1 receptors is necessary for the impairment of WM by THC, whereas the expression of CB 1 receptors on glutamatergic and on GABAergic neurons is dispensable for this effect. Furthermore, mice lacking CB 1 receptors on astrocytes as well as mice lacking CB 1 receptors on GABAergic neurons learned the SWM task as well as their respective wild-type littermate controls before THC treatment. Hence, astroglial CB 1 receptors and
CB 1 receptors expressed on GABAergic neurons are not necessary for SWM. Conversely, we observed that mice lacking CB 1 receptors on glutamatergic neurons were impaired in the SWM task.
Even though the performances of these animals were still sensitive to THC, indicating that CB 1 receptors located on cortical glutamatergic neurons are dispensable for the cannabinoid-induced impairment of SWM, these results indicate that endogenous cortical glutamatergic CB 1 receptors control SWM learning. Therefore, the effect of THC on SWM does not reflect a potentiation of the endogenous CB 1 receptor signalling but acts through separated mechanisms. This distinction might rely on the different spatial and temporal ways of activation of CB 1 receptors by exogenous cannabinoids and endogenously released cannabinoids. Systemic or local intra-cerebral administration of cannabinoids leads to an extended activation of CB 1 receptors and the duration of this activation depends on the pharmacokinetics of the drug. Conversely, the “on demand” action of eCBs (Piomelli, 2003; see General Introduction section II.3) supposes a tightly regulated endogenous activation of CB 1 receptors in terms of space and time. Indeed, few cells might be under the control of the physiological activation of CB 1 receptors whereas the neighboring ones might remain outside the influence of eCBs. It is therefore possible that, during specific ongoing neuronal activity such as WM processing, CB 1 receptors are activated in a limited number of cells, thus accounting for the fact that astroglial CB1 receptors and glutamatergic CB 1 receptors exert distinct contributions to the modulation of SWM performances following exogenous and endogenous CB 1 receptor activation.
Our data reveal that glutamatergic CB 1 receptors are necessary for SWM learning. However, many studies reported that acute complete pharmacological or constitutive genetic blockade of CB 1 receptors results in none to facilitatory effects on several WM memory paradigms (Lichtman, 2000;
Wise et al., 2009). Unlike in the GFAP-CB 1-KO mice, the deletion of CB 1 receptors in the Glu-CB 1-KO mice cannot be induced during adulthood, thereby leaving the possibility that the lack of CB 1 receptors during development leads to the observed phenotype. It is currently impossible to distinguish the contribution of cortical glutamatergic CB 1 receptors at the moment of WM from potential long-term developmental alterations induced by this non-inducible conditional genetic manipulation. Indeed, evidence exists regarding the role of the ECS on cortical pyramidal neurons growth at the early developmental stages (Mulder et al., 2008). Newborn Glu-CB 1-KO mice display an altered axonal elongation that is not observed in the constitutive CB 1-KO mice, suggesting the existence of
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 121 General Discussion compensatory signalling pathways that may account for the distinct phenotypic consequences observed during adulthood.
So far, the GFAP-CB 1-KO mouse line is the unique tool for studying the role of astroglial CB 1 receptors in vivo . Although our results uncovered its necessity for the THC-induced WM deficits, the endogenous role of these receptors needs to be investigated. Preliminary results obtain in our laboratory suggest that astroglial CB 1 receptors are necessary for object recognition memory (Busquets-Garcia, Robin, unpublished data). Hence, a thorough evaluation of the behavior of the
GFAP-CB 1-KO mice in other memory paradigms will provide important information regarding the endogenous role of astroglial CB 1 receptor signalling in learning and memory. Moreover, learning and memory processes are regulated by adult neurogenesis (Kempermann et al., 1997). In adult animals, particular brain regions such as the subventricular zone are known as neurogenic niches that contain different types of neural stem cells (Bordey, 2006). In particular, the subventricular zone contains protoplasmic astrocytes expressing GFAP (also called type B cells) that can generate intermediate progenitors that will give rise to new born neurons (Bordey, 2006). CB 1 receptors are expressed on GFAP-expressing cells of the subventricular zone (Moldrich and Wenger, 2000) and are necessary for normal adult neurogenesis in this area through unknown mechanisms (Jin et al., 2004). It is thus possible that CB 1 receptors expressed on these astrocytes proliferative progenitors control neuronal fate. As a consequence, the inhibition of neurogenesis could be involved in the effect of CB 1 receptor deletion on GFAP-expressing cells on memory functions. Therefore, it would be of interest to evaluate the neurogenesis capacity of the GFAP-CB 1-KO mice.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 122
Part III NEURONAL SUBTYPE -SPECIFIC CONTROL OF LEARNED
FEAR RESPONSES BY CB 1 RECEPTORS
The second objective of this thesis work was to precise the neuronal mechanisms involved in the regulation of learned fear responses by CB 1 receptors.
We first described the relationships between different fear responses evoked by a conditioned tone in wild-type C57BL/6N mice. In classical fear conditioning, we described the temporal expression of the freezing response (passive coping; Blanchard et al., 1976; Koolhaas et al., 1999) and active coping behaviors (rearing, digging, wall rearing / sniffing; Koolhaas et al., 1999; De Boer and Koolhaas, 2003; Gozzi et al., 2010). The decrease of freezing over tone presentation is associated with a concomitant increase of active coping behaviors which become the dominant response pattern at the end of the session. We then confirmed that the amount of freezing is negatively associated with active avoidance performances in the same animals (Lázaro-Muñoz et al., 2010; Vicens-Costa et al., 2011). Conversely, we observed that the amount of active coping behaviors is positively correlated with active avoidance performances. Thus, the type of response adopted in classical fear conditioning predicts individual active avoidance performances, suggesting that conditioned fear responses are subjected to individual variability.
The constitutive deletion of CB 1 receptors in CB 1-KO mice leads to strong freezing expression that prevents the development of active coping behaviors in classical conditioning. However, the same deletion induces both higher passive and active avoidance learning as compared to wild-type littermates, indicating that the relationship between the fear coping strategies adopted in classical fear conditioning and avoidance learning performances is disturbed following constitutive CB 1 receptors inactivation. We found that the dominant freezing response adopted in classical fear conditioning and higher passive avoidance learning displayed by CB 1-KO mice are likely accounted by the deletion of
CB 1 receptors on cortical glutamatergic neurons. Conversely, the deletion of CB 1 receptors on forebrain GABAergic neurons in GABA-CB 1-KO leads to the immediate adoption of active coping
123 General Discussion behaviors in classical fear conditioning and to a facilitation of active avoidance learning. Acute low and high doses of THC has been proposed to preferentially act at CB 1 receptors on glutamatergic neurons and GABAergic neurons, respectively (Bellocchio et al., 2010). Interestingly, we observed a dose- dependent biphasic effect of THC on the coping styles in wild-type C57BL/6N mice in classical fear conditioning, with low doses favoring active responses and higher doses promoting passive behaviors.
Considering that low and high doses of THC have been suggested to act primarily through CB 1 receptors expressed on glutamatergic and GABAergic neurons, respectively (Bellocchio et al., 2010),
Altogether, these data point to a cell type-specific control of CB 1 receptors over the behavioral strategy engaged in response to fear conditioned stimuli.
Local re-expression of CB 1 receptors in the amygdala of constitutive CB 1-KO restores the temporal relationship between freezing and active coping in classical fear conditioning by decreasing freezing responses, suggesting that CB 1 receptors in the amygdala participate in the adoption of the coping style adopted by animals in our conditions.
III.1 Individual variability of the fear responses
Although freezing behavior is a well-established index of conditioned fear, other behaviors may compete with freezing under a variety of conditions. It has been proposed that rodents confronted by potential dangers such as a predator odor or a shock probe can show behaviors oriented toward the threat to facilitate both visual and / or olfactory detection (Blanchard and Blanchard, 1989; McNaughton and Corr, 2004; Blanchard et al., 2011). These behaviors, including defensive burying and rearing, are thus considered as attempts to investigate the potential danger when, for instance, its source is ambiguous (Blanchard et al., 1990; Dielenberg et al., 2001; De Boer and Koolhaas, 2003; Cain and Ledoux, 2007; Blanchard et al., 2011). For instance, if sawdust is present in a cage with an aversive encounter, rodents can actively dig to bury the danger source with the bedding (Hebb et al., 2002; De Boer and Koolhaas, 2003; Blanchard et al., 2011). Rodents show rearing behaviors when introduced in novel environments (Platel and Porsolt, 1982). However, rearing behaviors are also induced in rats, together with an increase of blood pressure, by the introduction of a cat odor in a familiar place (Dielenberg et al., 2001). Interestingly, both increases of rearing and blood pressure are also observed if rats are returned in the previously cat odor-associated context (Dielenberg et al., 2001). Moreover, rearing can compete with freezing when rodents are introduced to inescapable contexts previously associated with shocks and rearing can be learned as a conditioned avoidance response (Ulrich & Azrin, 1962; De Boer and Koolhaas, 2003; Cain and Ledoux, 2007). Indeed, our study and others suggest that freezing is just one, although likely the most prominent, of the many behaviors that rodents might exhibit in response to a fear conditioned stimulus.
It has been proposed that freezing and escape behaviors are part of the innate species-specific defense responses of the rodent behavioral repertoire that are engaged by aversive stimuli (Bolles, 1970; Cain and Ledoux, 2007), but the conditions in which a particular behavior is selected are not known. The test situation can determine the type of defensive response adopted by the subjects. For instance, classical fear conditioning settings are characterized by the absence of escape routes and it
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 124 Part III - Neuronal subtype-specific control of learned fear responses by CB1 receptors has been proposed that rodents naturally tend to freeze in this condition (Blanchard et al., 1976). As noticed in the General Introduction (section IV.3.2.1b), an alternative account is that the threat proximity determines the type of response adopted (Fanselow and Lester, 1988; McNaughton and Corr, 2004). A third hypothesis is that the type of defensive behaviors can be determined by stable individual trait-like characteristics (Weshler, 1995; Ursin and Olff, 1993; Koolhaas et al., 1999). Indeed, even within inbred populations of mice or rats exposed to aversive situations, subpopulations can be distinguished by the type of response prevalently adopted. For instance, several studies reported that some individuals within a population of wild-type, naive rats or mice show very low conditioned freezing response after classical fear conditioning, and some individuals never acquire active avoidance even after an extended training (Lehner et al., 2008, Lázaro-Muñoz et al., 2010; Vicens- Costa et al., 2010). Moreover, rat have been selected and bred with regard to their defensive behavioral profiles in a variety of testing situations, so as to generate strains that are characterized, for instance, by their ability to learn active avoidance (Martin et al., 1982). Indeed, the Roman High Avoidance (RHA) and Low Avoidance (RLA) are characterized by good and poor active avoidance performances, respectively. Interestingly, these animals also differ in the amount of conditioned freezing induced by classical fear conditioning, with RHA showing a decreased freezing response and RLA displaying increased freezing response to CS (Aguilar et al., 2002; Steimer and Driscoll, 2003). Consistently, conditioned freezing during the initial stages of two-way active avoidance learning is negatively correlated to the efficiency in the acquisition of the task (Vicens-Costa et al., 2010; see below), supporting that freezing tendency to CS runs against the appearance of active avoidance responses. Our results also indicate that active coping behaviors measured in classical fear conditioning are positively related to active avoidance performances in wild-type animals. Accordingly, comparative studies have proposed a positive link between defensive digging and two-way active avoidance performances (Koolhaas et al., 2010). However, it is important to remind that classical fear conditioning and instrumental conditioning such as avoidance learning involve quite different learning processes (see General Introduction section IV.3.2). Avoidance conditioning has long been viewed as a two-stage learning process (Mowrer and Lamoreaux, 1946; McAllister et al., 1971) in which animals initially undergoes Pavlovian conditioning to form an association between the shock and the CS in the apparatus. Subsequently, the subjects learn the instrumental response to avoid the shock. Further, the “fear” aroused by the presence of the CS motivates learning of the instrumental response. Following this theory, “fear” reduction is associated with successful avoidance and has been proposed to reinforce avoidance learning. In this case, “fear” is defined as the presence of Pavlovian defensive responses. However, if one considers fear as a central state that can result from exposure to innate or learned stimuli (McAllister et al., 1971), without assumption about the relationship between the type of learning and fear emotional state, fear could be indicated in animals by either classical or instrumental conditioned fear responses. Therefore, our study supports the idea that individual variability in defensive strategies might bias the interpretation of performances observed in a conditioned aversive test.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 125 General Discussion
III.2 The ECS as a determinant of the types of fear coping strategies
We observed that the deletion of CB 1 receptors on cortical glutamatergic neurons and in forebrain GABAergic neurons leads to distinct coping strategies in classical fear conditioning and consistently enhanced passive and active conditioned avoidance responding, respectively. We thus proposed that CB 1 receptors signalling is an important physiological determinant of fear coping styles and that suppression of the inhibitory control on GABAergic and glutamatergic neurotransmissions likely facilitates the adoption of active or passive fear coping strategies, respectively. Consistently, increasing GABAergic tone by systemic pharmacological administration of GABA A receptor agonists including benzodiazepines enhances acquisition of shuttle box active avoidance (Lal and Forster, 1990; Fernández-Teruel et al., 1991; Escorihuela et al., 1993). Moreover, it has been recently shown that the acute stimulation of CB 1 receptors into the CeA impairs consolidation of one-trial dark-light passive avoidance, an effect reversed by co-administration of NMDA (Ghiasvand et al., 2011). In classical fear conditioning, Lerner et al. (2008) measured extracellular GABA concentration by in vivo microdialysis in the BLA of two rat groups selected upon their conditioned freezing response intensity (i.e. “high freezer” versus “low freezer”). Although GABA levels were identical prior and following the conditioning session in both rat groups, they observed an increase of extracellular GABA during a 10- minutes CS re-exposure only in low freezer animals, suggesting that high GABA release in the BLA might decrease freezing responses. Unfortunately, active coping behaviors were not analyzed in this study. Altogether, these data are in agreement with our proposed bimodal control of CB 1 receptors on glutamatergic and GABAergic tones in fear coping styles and further support a key role of the amygdala in this mechanism.
Innate tendencies to passively or actively cope with fearful stimuli can be assessed in other behavioral paradigms that allow animals to choose between different patterns of responses efficient for avoiding a harmful stimulus. For instance, in the defensive burying paradigm, animals are confronted with an electrified probe inserted into their home cage. In response to a brief contact with the probe, animals are observed either to actively bury the probe with the bedding or to display freezing (Beninger et al., 1980; Koolhaas et al., 1999; De Boer and Koolhaas, 2003). Both response patterns can be considered as successful coping because they lead to shock avoidance. A positive correlation between shuttle box active avoidance and burying behavior in the defensive burying and a negative correlation between active avoidance and freezing behavior to the probe insertion were observed in the RLA / RHA rat strains (Boersma et al., 2009). Furthermore, since both coping behaviors can be expressed within the task, the shock-probe burying paradigm is well suited to describe the neural circuitry of active and passive coping strategies. Indeed, one study assessed the involvement of CB 1 receptor signalling in this test (Degroot and Nomikos, 2004). In the constitutive
CB 1-KO mice, a decrease of burying time but no change in freezing time was observed, and mutant mice made fewer contacts with the electrified probe. In addition, the acute pharmacological blockade of CB 1 receptors leads to similar consequences only at a middle dose-range (i.e. 3 mg/kg but not 1 and 10 mg/kg), suggesting that the coping strategies adopted in this test might vary as a function of
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 126 Part III - Neuronal subtype-specific control of learned fear responses by CB1 receptors
the amount of CB 1 receptors targeted. Interestingly, individual differentiation in behaviors adopted in the defensive burying test has been associated with distinct patterns of both endocrine reactivity and brain regions activity as assessed by immediate-early genes expression (Koolhaas et al., 2010). Thus, exploring the phenotype of the CB 1-KO mouse lines in the defensive burying test would be an interesting perspective to further understand the brain mechanisms underlying the control of fear coping strategies by the ECS.
Interestingly, a recent positron emission tomography (PET) study in healthy humans revealed a strong inverse correlation between the brain binding of CB 1 receptors and personality traits linked to novelty-seeking (Van Laere et al., 2009), which have been associated to proactive coping (Steimer and Driscoll, 2003). Considering that the very large majority of CB 1 protein in the brain is expressed by GABAergic neurons (Katona et al., 2000; Kawamura et al., 2006; Monory et al., 2006; Bellocchio et al., 2010), the decreased levels of CB 1 binding in "novelty-seekers" observed in the PET study are likely due to reduced expression levels in inhibitory circuits. Thus, the PET data, together with the present and previous results using conditional mutant mice (Lafenêtre et al., 2009) suggest that the differential impact of CB 1 receptors on fear coping strategies might be extended to human subjects.
III.3 Conclusion
Brain functions are one of the most complex phenomena of the world. Understanding their mechanisms challenged mankind since thousands of years through the activity of philosophers, theologists and scientists. Modern neurosciences aim at this goal through the scientific approach, which is mainly based on the detailed dissection of specific elements of the brain mechanisms to eventually determine the basic principles of its activity. This approach led in the last centuries to impressive progresses. However, there are intrinsic limits of this approach. On one hand, our capacity to dissect small elements of the brain is limited by the experimental tools available to observe them. It is the typical case, in which the “light of the microscope alters the shape of the specimen”. Every time we try to observe the details of brain functions, our interventions are likely modifying the functions themselves. On the other hand, the brain is a complex system, in which each single element has its proper individual characteristics and is tightly integrated with billions of other elements. Therefore, the dissection of each element and its observation brings the risk of loosing the global vision of the whole system. In other words, we risk knowing everything about a specific ion channel, but failing to understand the network, in which the ion channel is integrated.
The study of the ECS presents some advantages to partially overcome these limitations. It allows scientists to focus on few elements (a receptor, few ligands and proteins), which can be “dissected” in their specific functions. On the other hand, the ECS is involved in a plethora of brain processes. This obliges the researchers to keep always in mind the multitude of different aspects possibly involved in the observed phenomena.
In our opinion, these peculiar features make of the ECS a well-suited subject to progress in the general understanding of brain functions. With all its limits, by pointing out the importance of astroglial
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 127 General Discussion signalling in memory and the different components of fear coping strategies, I believe that this thesis work is one example of this specific advantage provided by the study of the ECS.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 128
REFERENCES
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 129
A
Abel, E.L. (1971). Marihuana and memory: acquisition or retrieval? Science 173 , 1038–1040. Abush, H., and Akirav, I. (2010). Cannabinoids modulate hippocampal memory and plasticity. Hippocampus 20 , 1126–1138. Adams, R. (1940). MARIHUANA. Science 92 , 115–119. Aguado, T., Palazuelos, J., Monory, K., Stella, N., Cravatt, B., Lutz, B., Marsicano, G., Kokaia, Z., Guzmán, M., and Galve-Roperh, I. (2006). The endocannabinoid system promotes astroglial differentiation by acting on neural progenitor cells. J. Neurosci. 26 , 1551–1561. Aguilar, R., Gil, L., Flint, J., Gray, J.A., Dawson, G.R., Driscoll, P., Giménez-Llort, L., Escorihuela, R.M., Fernández-Teruel, A., and Tobeña, A. (2002). Learned fear, emotional reactivity and fear of heights: a factor analytic map from a large F(2) intercross of Roman rat strains. Brain Res. Bull. 57 , 17–26. Akagi, K., Sandig, V., Vooijs, M., Van der Valk, M., Giovannini, M., Strauss, M., and Berns, A. (1997). Cre-mediated somatic site-specific recombination in mice. Nucleic Acids Res. 25 , 1766–1773. Albayram, O., Alferink, J., Pitsch, J., Piyanova, A., Neitzert, K., Poppensieker, K., Mauer, D., Michel, K., Legler, A., Becker, A., et al. (2011). Role of CB1 cannabinoid receptors on GABAergic neurons in brain aging. Proc. Natl. Acad. Sci. U.S.A. 108 , 11256–11261. Alger, B.E., and Kim, J. (2011). Supply and demand for endocannabinoids. Trends Neurosci. 34 , 304–315. Ali, A.B. (2007). Presynaptic Inhibition of GABAA receptor-mediated unitary IPSPs by cannabinoid receptors at synapses between CCK-positive interneurons in rat hippocampus. J. Neurophysiol. 98 , 861–869. Altevogt, B.M., and Paul, D.L. (2004). Four classes of intercellular channels between glial cells in the CNS. J. Neurosci. 24 , 4313–4323. Alvarez-Buylla, A., and Lim, D.A. (2004). For the long run: maintaining germinal niches in the adult brain. Neuron 41 , 683–686. Amaral, D.G., and Insausti, R. (1992). Retrograde transport of D-[3H]-aspartate injected into the monkey amygdaloid complex. Exp Brain Res 88 , 375–388. Amorapanth, P., LeDoux, J.E., and Nader, K. (2000). Different lateral amygdala outputs mediate reactions and actions elicited by a fear-arousing stimulus. Nat. Neurosci. 3, 74–79. Anderson, C.M., and Swanson, R.A. (2000). Astrocyte glutamate transport: review of properties, regulation, and physiological functions. Glia 32 , 1–14. André, A., and Gonthier, M.P. (2010). The endocannabinoid system: Its roles in energy balance and potential as a target for obesity treatment. The International Journal of Biochemistry & Cell Biology 42 , 1788–1801. Araque, A., Martín, E.D., Perea, G., Arellano, J.I., and Buño, W. (2002). Synaptically released acetylcholine evokes Ca2+ elevations in astrocytes in hippocampal slices. J. Neurosci. 22 , 2443– 2450. Araque, A., Parpura, V., Sanzgiri, R.P., and Haydon, P.G. (1999). Tripartite synapses: glia, the unacknowledged partner. Trends Neurosci. 22 , 208–215. Arenos, J.D., Musty, R.E., and Bucci, D.J. (2006). Blockade of cannabinoid CB1 receptors alters contextual learning and memory. Eur. J. Pharmacol. 539 , 177–183. Atkins, C.M., Selcher, J.C., Petraitis, J.J., Trzaskos, J.M., and Sweatt, J.D. (1998). The MAPK cascade is required for mammalian associative learning. Nat. Neurosci. 1, 602–609.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 131
Atkinson, R., and Shiffrin, R. (1968). « Human memory: A proposed system and its control processes ». In The Psychology of Learning and Motivation: Advances in Research and Theory, (New York: K. Spence and J. Spence), pp. 89–195. Auclair, N., Otani, S., Soubrie, P., and Crepel, F. (2000). Cannabinoids modulate synaptic strength and plasticity at glutamatergic synapses of rat prefrontal cortex pyramidal neurons. J. Neurophysiol. 83 , 3287–3293. Azad, S.C., Kurz, J., Marsicano, G., Lutz, B., Zieglgansberger, W., and Rammes, G. (2008). Activation of CB1 specifically located on GABAergic interneurons inhibits LTD in the lateral amygdala. Learning & Memory 15 , 143–152. Azad, S.C., Monory, K., Marsicano, G., Cravatt, B.F., Lutz, B., Zieglgänsberger, W., and Rammes, G. (2004). Circuitry for associative plasticity in the amygdala involves endocannabinoid signaling. J. Neurosci. 24 , 9953–9961. B
Bacci, A., Huguenard, J.R., and Prince, D.A. (2003). Functional autaptic neurotransmission in fast-spiking interneurons: a novel form of feedback inhibition in the neocortex. J. Neurosci. 23 , 859– 866. Bacci, A., Huguenard, J.R., and Prince, D.A. (2004). Long-lasting self-inhibition of neocortical interneurons mediated by endocannabinoids. Nature 431 , 312–316. Bacci, A., Huguenard, J.R., and Prince, D.A. (2005). Modulation of neocortical interneurons: extrinsic influences and exercises in self-control. Trends Neurosci. 28 , 602–610. Baddeley, A. (2010). Working memory. Curr. Biol. 20 , R136–140. Baddeley, A., and Hitch, G. (1974). Working Memory. In The Psychology of Learning and Motivation: Advances in Research and Theory, (New York: G.A. Bower), pp. 47–89. Baddeley, A.D., and Baddeley, A.D. (2007). Working memory, thought, and action (Oxford; New York: Oxford University Press). Bado, P., Madeira, C., Vargas-Lopes, C., Moulin, T.C., Wasilewska-Sampaio, A.P., Maretti, L., de Oliveira, R.V., Amaral, O.B., and Panizzutti, R. (2011). Effects of low-dose D-serine on recognition and working memory in mice. Psychopharmacology (Berl.) 218 , 461–470. Balthasar, N., Dalgaard, L.T., Lee, C.E., Yu, J., Funahashi, H., Williams, T., Ferreira, M., Tang, V., McGovern, R.A., Kenny, C.D., et al. (2005). Divergence of melanocortin pathways in the control of food intake and energy expenditure. Cell 123 , 493–505. Bari, M., Battista, N., Fezza, F., Finazzi-Agrò, A., and Maccarrone, M. (2005). Lipid rafts control signaling of type-1 cannabinoid receptors in neuronal cells. Implications for anandamide-induced apoptosis. J. Biol. Chem. 280 , 12212–12220. Barrett, K.E., and Ganong, W.F. (2010). Ganong’s review of medical physiology (New York: McGraw-Hill Medical). Bartlett, T.E., Bannister, N.J., Collett, V.J., Dargan, S.L., Massey, P.V., Bortolotto, Z.A., Fitzjohn, S.M., Bashir, Z.I., Collingridge, G.L., and Lodge, D. (2007). Differential roles of NR2A and NR2B-containing NMDA receptors in LTP and LTD in the CA1 region of two-week old rat hippocampus. Neuropharmacology 52 , 60–70. Bass, N.H., Hess, H.H., Pope, A., and Thalheimer, C. (1971). Quantitative cytoarchitectonic distribution of neurons, glia, and DNa in rat cerebral cortex. J. Comp. Neurol. 143 , 481–490. Bekar, L.K., He, W., and Nedergaard, M. (2008). Locus coeruleus alpha-adrenergic-mediated activation of cortical astrocytes in vivo. Cereb. Cortex 18 , 2789–2795.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 132
Bellocchio, L., Lafenêtre, P., Cannich, A., Cota, D., Puente, N., Grandes, P., Chaouloff, F., Piazza, P.V., and Marsicano, G. (2010). Bimodal control of stimulated food intake by the endocannabinoid system. Nat Neurosci 13 , 281–283. Beltramo, M. (1997). Functional Role of High-Affinity Anandamide Transport, as Revealed by Selective Inhibition. Science 277 , 1094–1097. Bénard, G., Massa, F., Puente, N., Lourenço, J., Bellocchio, L., Soria-Gómez, E., Matias, I., Delamarre, A., Metna-Laurent, M., Cannich, A., et al. (2012). Mitochondrial CB(1) receptors regulate neuronal energy metabolism. Nat. Neurosci. 15 , 558–564. Beninger, R.J., MacLennan, A.J., and Pinel, J.P. (1980). The use of conditioned defensive burying to test the effects of pimozide on associative learning. Pharmacol. Biochem. Behav. 12 , 445– 448. Benito, C., Núñez, E., Tolón, R.M., Carrier, E.J., Rábano, A., Hillard, C.J., and Romero, J. (2003). Cannabinoid CB2 receptors and fatty acid amide hydrolase are selectively overexpressed in neuritic plaque-associated glia in Alzheimer’s disease brains. J. Neurosci. 23 , 11136–11141. Benito, C., Romero, J.P., Tolón, R.M., Clemente, D., Docagne, F., Hillard, C.J., Guaza, C., and Romero, J. (2007). Cannabinoid CB1 and CB2 receptors and fatty acid amide hydrolase are specific markers of plaque cell subtypes in human multiple sclerosis. J. Neurosci. 27 , 2396–2402. Bezzi, P., and Volterra, A. (2001). A neuron-glia signalling network in the active brain. Curr. Opin. Neurobiol. 11 , 387–394. Bialuk, I., and Winnicka, M.M. (2011). AM251, cannabinoids receptors ligand, improves recognition memory in rats. Pharmacol Rep 63 , 670–679. Bidaut-Russell, M., Devane, W.A., and Howlett, A.C. (1990). Cannabinoid receptors and modulation of cyclic AMP accumulation in the rat brain. J. Neurochem. 55 , 21–26. Bisogno, T., Ligresti, A., and Di Marzo, V. (2005). The endocannabinoid signalling system: biochemical aspects. Pharmacol. Biochem. Behav. 81 , 224–238. Bisogno, T., MacCarrone, M., De Petrocellis, L., Jarrahian, A., Finazzi-Agrò, A., Hillard, C., and Di Marzo, V. (2001). The uptake by cells of 2-arachidonoylglycerol, an endogenous agonist of cannabinoid receptors. Eur. J. Biochem. 268 , 1982–1989. Blanchard, D.C., Griebel, G., and Blanchard, R.J. (2001). Mouse defensive behaviors: pharmacological and behavioral assays for anxiety and panic. Neurosci Biobehav Rev 25 , 205–218. Blanchard, D.C., Griebel, G., and Blanchard, R.J. (2003). The Mouse Defense Test Battery: pharmacological and behavioral assays for anxiety and panic. European Journal of Pharmacology 463 , 97–116. Blanchard, D.C., Griebel, G., Pobbe, R., and Blanchard, R.J. (2011). Risk assessment as an evolved threat detection and analysis process. Neurosci Biobehav Rev 35 , 991–998. Blanchard, R.J., and Blanchard, D.C. (1969). Passive and active reactions to fear-eliciting stimuli. J Comp Physiol Psychol 68 , 129–135. Blanchard, R.J., and Blanchard, D.C. (1989). Antipredator defensive behaviors in a visible burrow system. J Comp Psychol 103 , 70–82. Blanchard, R.J., Blanchard, D.C., Rodgers, J., and Weiss, S.M. (1990). The characterization and modelling of antipredator defensive behavior. Neurosci Biobehav Rev 14 , 463–472. Blanchard, R.J., Fukunaga, K.K., and Blanchard, D.C. (1976). Environmental control of defensive reactions to footshock. Bulletin of the Psychonomic Society 8, 40. Blankman, J.L., Simon, G.M., and Cravatt, B.F. (2007). A comprehensive profile of brain enzymes that hydrolyze the endocannabinoid 2-arachidonoylglycerol. Chem. Biol. 14 , 1347–1356. Blázquez, C., Sánchez, C., Daza, A., Galve-Roperh, I., and Guzmán, M. (1999). The stimulation of ketogenesis by cannabinoids in cultured astrocytes defines carnitine palmitoyltransferase I as a new ceramide-activated enzyme. J. Neurochem. 72 , 1759–1768. De Boer, S.F., and Koolhaas, J.M. (2003). Defensive burying in rodents: ethology, neurobiology and psychopharmacology. European Journal of Pharmacology 463 , 145–161.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 133
Boersma, G.J., Scheurink, A.J.W., Wielinga, P.Y., Steimer, T.J., and Benthem, L. (2009). The passive coping Roman Low Avoidance rat, a non-obese rat model for insulin resistance. Physiol. Behav. 97 , 353–358. Bojnik, E., Turunç, E., Arma ğan, G., Kanıt, L., Benyhe, S., Yalçın, A., and Borsodi, A. (2012). Changes in the cannabinoid (CB1) receptor expression level and G-protein activation in kainic acid induced seizures. Epilepsy Res. 99 , 64–68. Bolles, R.C. (1970). Species-specific defense reactions and avoidance learning. Psychological Review 77 , 32. Bontempi, B., Laurent-Demir, C., Destrade, C., and Jaffard, R. (1999). Time-dependent reorganization of brain circuitry underlying long-term memory storage. Nature 400 , 671–675. Bordey, A. (2006). Adult neurogenesis: basic concepts of signaling. Cell Cycle 5, 722–728. Bouaboula, M., Bourrié, B., Rinaldi-Carmona, M., Shire, D., Le Fur, G., and Casellas, P. (1995a). Stimulation of cannabinoid receptor CB1 induces krox-24 expression in human astrocytoma cells. J. Biol. Chem. 270 , 13973–13980. Bouaboula, M., Poinot-Chazel, C., Bourrié, B., Canat, X., Calandra, B., Rinaldi-Carmona, M., Le Fur, G., and Casellas, P. (1995b). Activation of mitogen-activated protein kinases by stimulation of the central cannabinoid receptor CB1. Biochem. J. 312 ( Pt 2) , 637–641. Bouzier-Sore, A.-K., Voisin, P., Canioni, P., Magistretti, P.J., and Pellerin, L. (2003). Lactate is a preferential oxidative energy substrate over glucose for neurons in culture. J. Cereb. Blood Flow Metab. 23 , 1298–1306. Bowser, D.N., and Khakh, B.S. (2004). ATP excites interneurons and astrocytes to increase synaptic inhibition in neuronal networks. J. Neurosci. 24 , 8606–8620. Brandon, E.P., Idzerda, R.L., and McKnight, G.S. (1997). PKA isoforms, neural pathways, and behaviour: making the connection. Curr. Opin. Neurobiol. 7, 397–403. Brito, G.N., and Thomas, G.J. (1981). T-maze alternation, response patterning, and septo- hippocampal circuitry in rats. Behav. Brain Res. 3, 319–340. Brown, A.J. (2007). Novel cannabinoid receptors. Br. J. Pharmacol. 152 , 567–575. Brown, A.M. (2004). Brain glycogen re-awakened. J. Neurochem. 89 , 537–552. Brown, A.M., and Ransom, B.R. (2007). Astrocyte glycogen and brain energy metabolism. Glia 55 , 1263–1271. Brown, J.E., Kassouny, M., and Cross, J.K. (1977). Kinetic studies of food intake and sucrose solution preference by rats treated with low doses of delta9-tetrahydrocannabinol. Behav Biol 20 , 104– 110. Brown, T.M., Brotchie, J.M., and Fitzjohn, S.M. (2003). Cannabinoids decrease corticostriatal synaptic transmission via an effect on glutamate uptake. J. Neurosci. 23 , 11073–11077. de Bruin, N.M.W.J., Prickaerts, J., Lange, J.H.M., Akkerman, S., Andriambeloson, E., de Haan, M., Wijnen, J., van Drimmelen, M., Hissink, E., Heijink, L., et al. (2010). SLV330, a cannabinoid CB1 receptor antagonist, ameliorates deficits in the T-maze, object recognition and Social Recognition Tasks in rodents. Neurobiol Learn Mem 93 , 522–531. Bura, S.A., Castañé, A., Ledent, C., Valverde, O., and Maldonado, R. (2007). Genetic and pharmacological approaches to evaluate the interaction between the cannabinoid and cholinergic systems in cognitive processes. Br. J. Pharmacol. 150 , 758–765. Bushong, E.A., Martone, M.E., Jones, Y.Z., and Ellisman, M.H. (2002). Protoplasmic astrocytes in CA1 stratum radiatum occupy separate anatomical domains. J. Neurosci. 22 , 183–192. Busquets-Garcia, A., Puighermanal, E., Pastor, A., de la Torre, R., Maldonado, R., and Ozaita, A. (2011). Differential role of anandamide and 2-arachidonoylglycerol in memory and anxiety-like responses. Biol. Psychiatry 70 , 479–486.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 134
C
Cadas, H., Gaillet, S., Beltramo, M., Venance, L., and Piomelli, D. (1996). Biosynthesis of an endogenous cannabinoid precursor in neurons and its control by calcium and cAMP. J. Neurosci. 16 , 3934–3942. Cain, C.K., Blouin, A.M., and Barad, M. (2003). Temporally massed CS presentations generate more fear extinction than spaced presentations. J Exp Psychol Anim Behav Process 29 , 323–333. Cain, C.K., and LeDoux, J.E. (2007). Escape from fear: a detailed behavioral analysis of two atypical responses reinforced by CS termination. J Exp Psychol Anim Behav Process 33 , 451–463. Calandreau, L., Trifilieff, P., Mons, N., Costes, L., Marien, M., Marighetto, A., Micheau, J., Jaffard, R., and Desmedt, A. (2006). Extracellular hippocampal acetylcholine level controls amygdala function and promotes adaptive conditioned emotional response. J. Neurosci. 26 , 13556–13566. Calignano, A., La Rana, G., Giuffrida, A., and Piomelli, D. (1998). Control of pain initiation by endogenous cannabinoids. Nature 394 , 277–281. Cannich, A., Wotjak, C.T., Kamprath, K., Hermann, H., Lutz, B., and Marsicano, G. (2004). CB1 cannabinoid receptors modulate kinase and phosphatase activity during extinction of conditioned fear in mice. Learn. Mem. 11 , 625–632. Carlim, E., and Kramer, C. (1965). Effects of Cannabis sativa (marihuana) on maze performance of the rat. Psychopharmacology 7, 175–181. Carlini, E., Karniol, I., Renault, P., and Schuster, C. (1974). Effects of marihuana in laboratory animals and in man. British Journal of Pharmacology 50 , 299.
Carlini, E.A. (2004). The good and the bad effects of ⊖ trans-delta-9-tetrahydrocannabinol ([Delta] 9-THC) on humans. Toxicon 44 , 461–467. Carmignoto, G., and Fellin, T. (2006). Glutamate release from astrocytes as a non-synaptic mechanism for neuronal synchronization in the hippocampus. J. Physiol. Paris 99 , 98–102. Carrive, P., Leung, P., Harris, J., and Paxinos, G. (1997). Conditioned fear to context is associated with increased Fos expression in the caudal ventrolateral region of the midbrain periaqueductal gray. Neuroscience 78 , 165–177. Casanova, E., Fehsenfeld, S., Mantamadiotis, T., Lemberger, T., Greiner, E., Stewart, A.F., and Schütz, G. (2001). A CamKIIalpha iCre BAC allows brain-specific gene inactivation. Genesis 31 , 37– 42. Di Castro, M.A., Chuquet, J., Liaudet, N., Bhaukaurally, K., Santello, M., Bouvier, D., Tiret, P., and Volterra, A. (2011). Local Ca2+ detection and modulation of synaptic release by astrocytes. Nat. Neurosci. 14 , 1276–1284. Chakrabarti, A., Onaivi, E.S., and Chaudhuri, G. (1995). Cloning and sequencing of a cDNA encoding the mouse brain-type cannabinoid receptor protein. DNA Seq. 5, 385–388. Chanda, P.K., Gao, Y., Mark, L., Btesh, J., Strassle, B.W., Lu, P., Piesla, M.J., Zhang, M.-Y., Bingham, B., Uveges, A., et al. (2010). Monoacylglycerol lipase activity is a critical modulator of the tone and integrity of the endocannabinoid system. Mol. Pharmacol. 78 , 996–1003. Chari-Bitron, A., and Bino, T. (1971). Effect of 1-tetrahydrocannabinol on ATPase activity of rat liver mitochondria. Biochem. Pharmacol. 20 , 473–475. Chen, J.-K., Chen, J., Imig, J.D., Wei, S., Hachey, D.L., Guthi, J.S., Falck, J.R., Capdevila, J.H., and Harris, R.C. (2008). Identification of novel endogenous cytochrome p450 arachidonate metabolites with high affinity for cannabinoid receptors. J. Biol. Chem. 283 , 24514–24524. Chevaleyre, V., and Castillo, P.E. (2003). Heterosynaptic LTD of hippocampal GABAergic synapses: a novel role of endocannabinoids in regulating excitability. Neuron 38 , 461–472.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 135
Chevaleyre, V., and Castillo, P.E. (2004). Endocannabinoid-mediated metaplasticity in the hippocampus. Neuron 43 , 871–881. Chevaleyre, V., Heifets, B.D., Kaeser, P.S., Südhof, T.C., Purpura, D.P., and Castillo, P.E. (2007). Endocannabinoid-mediated long-term plasticity requires cAMP/PKA signaling and RIM1alpha. Neuron 54 , 801–812. Chevaleyre, V., Takahashi, K.A., and Castillo, P.E. (2006). Endocannabinoid-mediated synaptic plasticity in the CNS. Annu. Rev. Neurosci. 29 , 37–76. Chhatwal, J.P., Davis, M., Maguschak, K.A., and Ressler, K.J. (2005). Enhancing cannabinoid neurotransmission augments the extinction of conditioned fear. Neuropsychopharmacology 30 , 516– 524. Chhatwal, J.P., Gutman, A.R., Maguschak, K.A., Bowser, M.E., Yang, Y., Davis, M., and Ressler, K.J. (2009). Functional interactions between endocannabinoid and CCK neurotransmitter systems may be critical for extinction learning. Neuropsychopharmacology 34 , 509–521. Childers, S.R., and Deadwyler, S.A. (1996). Role of cyclic AMP in the actions of cannabinoid receptors. Biochem. Pharmacol. 52 , 819–827. Chiu, C.Q., Puente, N., Grandes, P., and Castillo, P.E. (2010). Dopaminergic modulation of endocannabinoid-mediated plasticity at GABAergic synapses in the prefrontal cortex. J. Neurosci. 30 , 7236–7248. Choi, J.-S., Cain, C.K., and LeDoux, J.E. (2010). The role of amygdala nuclei in the expression of auditory signaled two-way active avoidance in rats. Learn. Mem. 17 , 139–147. Ciocchi, S., Herry, C., Grenier, F., Wolff, S.B.E., Letzkus, J.J., Vlachos, I., Ehrlich, I., Sprengel, R., Deisseroth, K., Stadler, M.B., et al. (2010). Encoding of conditioned fear in central amygdala inhibitory circuits. Nature 468 , 277–282. Citri, A., and Malenka, R.C. (2008). Synaptic plasticity: multiple forms, functions, and mechanisms. Neuropsychopharmacology 33 , 18–41. Clarke, J.R., Rossato, J.I., Monteiro, S., Bevilaqua, L.R.., Izquierdo, I., and Cammarota, M. (2008). Posttraining activation of CB1 cannabinoid receptors in the CA1 region of the dorsal hippocampus impairs object recognition long-term memory. Neurobiology of Learning and Memory 90 , 374–381. Collingridge, G.L., Peineau, S., Howland, J.G., and Wang, Y.T. (2010). Long-term depression in the CNS. Nat. Rev. Neurosci. 11 , 459–473. Collins, D.R., Pertwee, R.G., and Davies, S.N. (1995). Prevention by the cannabinoid antagonist, SR141716A, of cannabinoid-mediated blockade of long-term potentiation in the rat hippocampal slice. Br. J. Pharmacol. 115 , 869–870. Compton, D.R., Gold, L.H., Ward, S.J., Balster, R.L., and Martin, B.R. (1992). Aminoalkylindole analogs: cannabimimetic activity of a class of compounds structurally distinct from delta 9- tetrahydrocannabinol. J. Pharmacol. Exp. Ther. 263 , 1118–1126. Corcoran, M.E., McCaughran, J.A., Jr, and Wada, J.A. (1973). Acute antiepileptic effects of 9- tetrahydrocannabinol in rats with kindled seizures. Exp. Neurol. 40 , 471–483. Cornell-Bell, A.H., Finkbeiner, S.M., Cooper, M.S., and Smith, S.J. (1990). Glutamate induces calcium waves in cultured astrocytes: long-range glial signaling. Science 247 , 470–473. Cota, D., Marsicano, G., Tschöp, M., Grübler, Y., Flachskamm, C., Schubert, M., Auer, D., Yassouridis, A., Thöne-Reineke, C., Ortmann, S., et al. (2003). The endogenous cannabinoid system affects energy balance via central orexigenic drive and peripheral lipogenesis. J. Clin. Invest. 112 , 423–431. Cota, D., Steiner, M.-A., Marsicano, G., Cervino, C., Herman, J.P., Grubler, Y., Stalla, J., Pasquali, R., Lutz, B., Stalla, G.K., et al. (2006). Requirement of Cannabinoid Receptor Type 1 for the Basal Modulation of Hypothalamic-Pituitary-Adrenal Axis Function. Endocrinology 148 , 1574–1581. Cravatt, B.F., Giang, D.K., Mayfield, S.P., Boger, D.L., Lerner, R.A., and Gilula, N.B. (1996). Molecular characterization of an enzyme that degrades neuromodulatory fatty-acid amides. Nature 384 , 83–87.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 136
Crawley, J.N., Corwin, R.L., Robinson, J.K., Felder, C.C., Devane, W.A., and Axelrod, J. (1993). Anandamide, an endogenous ligand of the cannabinoid receptor, induces hypomotility and hypothermia in vivo in rodents. Pharmacol. Biochem. Behav. 46 , 967–972. D
D’Souza, D.C., Perry, E., MacDougall, L., Ammerman, Y., Cooper, T., Wu, Y.-T., Braley, G., Gueorguieva, R., and Krystal, J.H. (2004). The psychotomimetic effects of intravenous delta-9- tetrahydrocannabinol in healthy individuals: implications for psychosis. Neuropsychopharmacology 29 , 1558–1572. Darley, C.F., Tinklenberg, J.R., Hollister, T.E., and Atkinson, R.C. (1973). Marihuana and retrieval from short-term memory. Psychopharmacologia 29 , 231–238. Darwin, C. (1872). The expression of the emotions in man and animals (London: John Murray). Deacon, R.M.J., and Rawlins, J.N.P. (2006). T-maze alternation in the rodent. Nat Protoc 1, 7– 12. Degroot, A., and Nomikos, G.G. (2004). Genetic deletion and pharmacological blockade of CB1 receptors modulates anxiety in the shock-probe burying test. Eur. J. Neurosci. 20 , 1059–1064. Delgado, M.R., Jou, R.L., Ledoux, J.E., and Phelps, E.A. (2009). Avoiding negative outcomes: tracking the mechanisms of avoidance learning in humans during fear conditioning. Front Behav Neurosci 3, 33. Demuth, D.G., and Molleman, A. (2006). Cannabinoid signalling. Life Sci. 78 , 549–563. Derkinderen, P., Valjent, E., Toutant, M., Corvol, J.C., Enslen, H., Ledent, C., Trzaskos, J., Caboche, J., and Girault, J.A. (2003). Regulation of extracellular signal-regulated kinase by cannabinoids in hippocampus. The Journal of Neuroscience 23 , 2371–2382. Dermietzel, R., Gao, Y., Scemes, E., Vieira, D., Urban, M., Kremer, M., Bennett, M.V., and Spray, D.C. (2000). Connexin43 null mice reveal that astrocytes express multiple connexins. Brain Res. Brain Res. Rev. 32 , 45–56. Devane, W.A., Dysarz, F.A., 3rd, Johnson, M.R., Melvin, L.S., and Howlett, A.C. (1988). Determination and characterization of a cannabinoid receptor in rat brain. Mol. Pharmacol. 34 , 605– 613. Devane, W.A., Hanus, L., Breuer, A., Pertwee, R.G., Stevenson, L.A., Griffin, G., Gibson, D., Mandelbaum, A., Etinger, A., and Mechoulam, R. (1992). Isolation and structure of a brain constituent that binds to the cannabinoid receptor. Science 258 , 1946–1949. Di, S., Boudaba, C., Popescu, I.R., Weng, F.-J., Harris, C., Marcheselli, V.L., Bazan, N.G., and Tasker, J.G. (2005). Activity-dependent release and actions of endocannabinoids in the rat hypothalamic supraoptic nucleus. J. Physiol. (Lond.) 569 , 751–760. Di, S., Malcher-Lopes, R., Halmos, K.C., and Tasker, J.G. (2003). Nongenomic glucocorticoid inhibition via endocannabinoid release in the hypothalamus: a fast feedback mechanism. J. Neurosci. 23 , 4850–4857. Diagnostic and statistical manual of mental disorders: (2000). DSM-IV-TR (Washington, DC: American Psychiatric Association). Díaz-Morán, S., Palència, M., Mont-Cardona, C., Cañete, T., Blázquez, G., Martínez- Membrives, E., López-Aumatell, R., Tobeña, A., and Fernández-Teruel, A. (2012). Coping style and stress hormone responses in genetically heterogeneous rats: comparison with the Roman rat strains. Behav. Brain Res. 228 , 203–210. Dielenberg, R.A., Carrive, P., and McGregor, I.S. (2001). The cardiovascular and behavioral response to cat odor in rats: unconditioned and conditioned effects1. Brain Research 897 , 228–237.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 137
Dienel, G.A., Ball, K.K., and Cruz, N.F. (2007). A glycogen phosphorylase inhibitor selectively enhances local rates of glucose utilization in brain during sensory stimulation of conscious rats: implications for glycogen turnover. J. Neurochem. 102 , 466–478. Dinh, T.P., Kathuria, S., and Piomelli, D. (2004). RNA interference suggests a primary role for monoacylglycerol lipase in the degradation of the endocannabinoid 2-arachidonoylglycerol. Mol. Pharmacol. 66 , 1260–1264. Domenici, M.R., Azad, S.C., Marsicano, G., Schierloh, A., Wotjak, C.T., Dodt, H.-U., Zieglgänsberger, W., Lutz, B., and Rammes, G. (2006). Cannabinoid receptor type 1 located on presynaptic terminals of principal neurons in the forebrain controls glutamatergic synaptic transmission. J. Neurosci. 26 , 5794–5799. Dragoi, G., Harris, K.D., and Buzsáki, G. (2003). Place representation within hippocampal networks is modified by long-term potentiation. Neuron 39 , 843–853. Drew, G.M., Mitchell, V.A., and Vaughan, C.W. (2008). Glutamate spillover modulates GABAergic synaptic transmission in the rat midbrain periaqueductal grey via metabotropic glutamate receptors and endocannabinoid signaling. J. Neurosci. 28 , 808–815. Dubreucq, S., Kambire, S., Conforzi, M., Metna-Laurent, M., Cannich, A., Soria-Gomez, E., Richard, E., Marsicano, G., and Chaouloff, F. (2012a). Cannabinoid type 1 receptors located on single-minded 1-expressing neurons control emotional behaviors. Neuroscience 204 , 230–244. Dubreucq, S., Matias, I., Cardinal, P., Häring, M., Lutz, B., Marsicano, G., and Chaouloff, F. (2012b). Genetic Dissection of the Role of Cannabinoid Type-1 Receptors in the Emotional Consequences of Repeated Social Stress in Mice. Neuropsychopharmacology: Official Publication of the American College of Neuropsychopharmacology. Duffy, S., Labrie, V., and Roder, J.C. (2007). D-serine augments NMDA-NR2B receptor- dependent hippocampal long-term depression and spatial reversal learning. Neuropsychopharmacology 33 , 1004–1018. E
Ehrlich, I., Humeau, Y., Grenier, F., Ciocchi, S., Herry, C., and Lüthi, A. (2009). Amygdala inhibitory circuits and the control of fear memory. Neuron 62 , 757–771. Ehrlich, L.C., Peterson, P.K., and Hu, S. (1999). Interleukin (IL)-1beta-mediated apoptosis of human astrocytes. Neuroreport 10 , 1849–1852. Ekman, P. (2003). Emotions Inside Out: 130 Years after Darwin’s The Expression of the Emotions in Man and Animals (New York). Escorihuela, R.M., Fernández-Teruel, A., Zapata, A., Núñez, J.F., and Tobeña, A. (1993). Flumazenil prevents the anxiolytic effects of diazepam, alprazolam and adinazolam on the early acquisition of two-way active avoidance. Pharmacol. Res. 28 , 53–58. Esposito, G., Iuvone, T., Savani, C., Scuderi, C., De Filippis, D., Papa, M., Di Marzo, V., and Steardo, L. (2007). Opposing control of cannabinoid receptor stimulation on amyloid-beta-induced reactive gliosis: in vitro and in vivo evidence. J. Pharmacol. Exp. Ther. 322 , 1144–1152. Estes, W.K., and Skinner, B.F. (1941). Some quantitative properties of anxiety. Journal of Experimental Psychology 29 , 390–400.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 138
F
Fanselow, M.S. (1980). Conditioned and unconditional components of post-shock freezing. Pavlov J Biol Sci 15 , 177–182. Fanselow, M.S. (1984). Shock-induced analgesia on the formalin test: effects of shock severity, naloxone, hypophysectomy, and associative variables. Behav. Neurosci. 98 , 79–95. Fanselow, M.S., and Bolles, R.C. (1988). A functional behavioristic approach to aversely motivated behavior: predatory imminence as a determinant of the topography of defensive behavior. In Evolution and Learning, (Hillsdale, N.J., England: Lawrence Erlbaum Associates), pp. 185–212. Fanselow, M.S., and Poulos, A.M. (2005). The neuroscience of mammalian associative learning. Annu Rev Psychol 56 , 207–234. Farina, C., Aloisi, F., and Meinl, E. (2007). Astrocytes are active players in cerebral innate immunity. Trends Immunol. 28 , 138–145. Fellin, T., Halassa, M.M., Terunuma, M., Succol, F., Takano, H., Frank, M., Moss, S.J., and Haydon, P.G. (2009). Endogenous nonneuronal modulators of synaptic transmission control cortical slow oscillations in vivo. Proc. Natl. Acad. Sci. U.S.A. 106 , 15037–15042. Fellin, T., Pascual, O., Gobbo, S., Pozzan, T., Haydon, P.G., and Carmignoto, G. (2004). Neuronal synchrony mediated by astrocytic glutamate through activation of extrasynaptic NMDA receptors. Neuron 43 , 729–743. Fernández-Teruel, A., Escorihuela, R.M., Boix, F., and Tobena, A. (1991). Effects of different handling-stimulation procedures and benzodiazepines on two-way active avoidance acquisition in rats. Pharmacol. Res. 24 , 273–282. Ferrari, F., Ottani, A., Vivoli, R., and Giuliani, D. (1999). Learning impairment produced in rats by the cannabinoid agonist HU 210 in a water-maze task. Pharmacol. Biochem. Behav. 64 , 555–561. Finn, D.P., Beckett, S.R.G., Richardson, D., Kendall, D.A., Marsden, C.A., and Chapman, V. (2004a). Evidence for differential modulation of conditioned aversion and fear-conditioned analgesia by CB1 receptors. Eur. J. Neurosci. 20 , 848–852. Finn, D.P., Jhaveri, M.D., Beckett, S.R.G., Kendall, D.A., Marsden, C.A., and Chapman, V. (2004b). Cannabinoids modulate ultrasound-induced aversive responses in rats. Psychopharmacology (Berl.) 172 , 41–51. Fino, E., Paille, V., Cui, Y., Morera-Herreras, T., Deniau, J.-M., and Venance, L. (2010). Distinct coincidence detectors govern the corticostriatal spike timing-dependent plasticity. J. Physiol. (Lond.) 588 , 3045–3062. Fletcher, P.C., and Honey, G.D. (2006). Schizophrenia, ketamine and cannabis: evidence of overlapping memory deficits. Trends Cogn. Sci. (Regul. Ed.) 10 , 167–174. Florian, C., Vecsey, C.G., Halassa, M.M., Haydon, P.G., and Abel, T. (2011). Astrocyte-derived adenosine and A1 receptor activity contribute to sleep loss-induced deficits in hippocampal synaptic plasticity and memory in mice. J. Neurosci. 31 , 6956–6962. Földy, C., Neu, A., Jones, M.V., and Soltesz, I. (2006). Presynaptic, activity-dependent modulation of cannabinoid type 1 receptor-mediated inhibition of GABA release. J. Neurosci. 26 , 1465–1469. Fossat, P., Turpin, F.R., Sacchi, S., Dulong, J., Shi, T., Rivet, J.-M., Sweedler, J.V., Pollegioni, L., Millan, M.J., Oliet, S.H.R., et al. (2012). Glial D-serine gates NMDA receptors at excitatory synapses in prefrontal cortex. Cereb. Cortex 22 , 595–606. Fox, C.J., Russell, K.I., Wang, Y.T., and Christie, B.R. (2006). Contribution of NR2A and NR2B NMDA subunits to bidirectional synaptic plasticity in the hippocampus in vivo. Hippocampus 16 , 907– 915.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 139
Frankland, P.W., Bontempi, B., Talton, L.E., Kaczmarek, L., and Silva, A.J. (2004). The involvement of the anterior cingulate cortex in remote contextual fear memory. Science 304 , 881–883. Freund, T.F., Katona, I., and Piomelli, D. (2003). Role of endogenous cannabinoids in synaptic signaling. Physiol. Rev. 83 , 1017–1066. Fride, E., and Mechoulam, R. (1993). Pharmacological activity of the cannabinoid receptor agonist, anandamide, a brain constituent. Eur. J. Pharmacol. 231 , 313–314. Froger, N., Orellana, J.A., Cohen-Salmon, M., Ezan, P., Amigou, E., Sáez, J.C., and Giaume, C. (2009). Cannabinoids prevent the opposite regulation of astroglial connexin43 hemichannels and gap junction channels induced by pro-inflammatory treatments. J. Neurochem. 111 , 1383–1397. Fu, J., Bottegoni, G., Sasso, O., Bertorelli, R., Rocchia, W., Masetti, M., Guijarro, A., Lodola, A., Armirotti, A., Garau, G., et al. (2012). A catalytically silent FAAH-1 variant drives anandamide transport in neurons. Nat. Neurosci. 15 , 64–69. Fukudome, Y., Ohno-Shosaku, T., Matsui, M., Omori, Y., Fukaya, M., Tsubokawa, H., Taketo, M.M., Watanabe, M., Manabe, T., and Kano, M. (2004). Two distinct classes of muscarinic action on hippocampal inhibitory synapses: M2-mediated direct suppression and M1/M3-mediated indirect suppression through endocannabinoid signalling. Eur. J. Neurosci. 19 , 2682–2692. G
Gadea, A., and López-Colomé, A.M. (2001). Glial transporters for glutamate, glycine and GABA I. Glutamate transporters. J. Neurosci. Res. 63 , 453–460. Galante, M., and Diana, M.A. (2004). Group I metabotropic glutamate receptors inhibit GABA release at interneuron-Purkinje cell synapses through endocannabinoid production. J. Neurosci. 24 , 4865–4874. Gaoni, Y., and Mechoulam, R. (1964). Isolation, structure and partial synthesis of an active constituant of hashish. Journal of American Chemical Society 86 , 1646–1647. Garcia, D.E., Brown, S., Hille, B., and Mackie, K. (1998). Protein kinase C disrupts cannabinoid actions by phosphorylation of the CB1 cannabinoid receptor. J. Neurosci. 18 , 2834–2841. Garcia-Lopez, P., Garcia-Marin, V., and Freire, M. (2010). The histological slides and drawings of Cajal. Frontiers in Neuroanatomy 4,. Gatley, S.J., Gifford, A.N., Volkow, N.D., Lan, R., and Makriyannis, A. (1996). 123I-labeled AM251: a radioiodinated ligand which binds in vivo to mouse brain cannabinoid CB1 receptors. Eur. J. Pharmacol. 307 , 331–338. Gérard, C.M., Mollereau, C., Vassart, G., and Parmentier, M. (1991). Molecular cloning of a human cannabinoid receptor which is also expressed in testis. Biochem. J. 279 ( Pt 1) , 129–134. Gerdeman, G.L., Ronesi, J., and Lovinger, D.M. (2002). Postsynaptic endocannabinoid release is critical to long-term depression in the striatum. Nat. Neurosci. 5, 446–451. Ghiasvand, M., Rezayof, A., Zarrindast, M.R., and Ahmadi, S. (2011). Activation of cannabinoid CB1 receptors in the central amygdala impairs inhibitory avoidance memory consolidation via NMDA receptors. Neurobiol Learn Mem 96 , 333–338. Giaume, C., Koulakoff, A., Roux, L., Holcman, D., and Rouach, N. (2010). Astroglial networks: a step further in neuroglial and gliovascular interactions. Nat. Rev. Neurosci. 11 , 87–99. Giaume, C., Tabernero, A., and Medina, J.M. (1997). Metabolic trafficking through astrocytic gap junctions. Glia 21 , 114–123. Gibbs, M.E., Anderson, D.G., and Hertz, L. (2006). Inhibition of glycogenolysis in astrocytes interrupts memory consolidation in young chickens. Glia 54 , 214–222. Girardeau, G., Benchenane, K., Wiener, S.I., Buzsáki, G., and Zugaro, M.B. (2009). Selective suppression of hippocampal ripples impairs spatial memory. Nat. Neurosci. 12 , 1222–1223.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 140
Glaser, S.T., Abumrad, N.A., Fatade, F., Kaczocha, M., Studholme, K.M., and Deutsch, D.G. (2003). Evidence against the presence of an anandamide transporter. Proc. Natl. Acad. Sci. U.S.A. 100 , 4269–4274. Goebbels, S., Bormuth, I., Bode, U., Hermanson, O., Schwab, M.H., and Nave, K.A. (2006). Genetic targeting of principal neurons in neocortex and hippocampus of NEX-Cre mice. Genesis 44 , 611–621. Golgi, C. (1885). Sulla fina anatomia degli organi centrali del sistema nervoso. Riv. Sper. Fremiat. Med. Leg. Alienazioni Ment. 11 , 72–123. Goparaju, S.K., Ueda, N., Yamaguchi, H., and Yamamoto, S. (1998). Anandamide amidohydrolase reacting with 2-arachidonoylglycerol, another cannabinoid receptor ligand. FEBS Lett. 422 , 69–73. Gozzi, A., Jain, A., Giovannelli, A., Giovanelli, A., Bertollini, C., Crestan, V., Schwarz, A.J., Tsetsenis, T., Ragozzino, D., Gross, C.T., et al. (2010). A neural switch for active and passive fear. Neuron 67 , 656–666. Greenwald, M.K., and Stitzer, M.L. (2000). Antinociceptive, subjective and behavioral effects of smoked marijuana in humans. Drug Alcohol Depend 59 , 261–275. Gross, J.J. (1998). Antecedent- and response-focused emotion regulation: divergent consequences for experience, expression, and physiology. J Pers Soc Psychol 74 , 224–237. Gruart, A., and Delgado-García, J.M. (2007). Activity-dependent changes of the hippocampal CA3-CA1 synapse during the acquisition of associative learning in conscious mice. Genes Brain Behav. 6 Suppl 1 , 24–31. Grueter, B.A., Brasnjo, G., and Malenka, R.C. (2010). Postsynaptic TRPV1 triggers cell type- specific long-term depression in the nucleus accumbens. Nat. Neurosci. 13 , 1519–1525. Guzmán, M., and Blázquez, C. (2001). Is there an astrocyte-neuron ketone body shuttle? Trends Endocrinol. Metab. 12 , 169–173. H
Hájos, N., Ellender, T.J., Zemankovics, R., Mann, E.O., Exley, R., Cragg, S.J., Freund, T.F., and Paulsen, O. (2009). Maintaining network activity in submerged hippocampal slices: importance of oxygen supply. Eur. J. Neurosci. 29 , 319–327. Hájos, N., and Mody, I. (2009). Establishing a physiological environment for visualized in vitro brain slice recordings by increasing oxygen supply and modifying aCSF content. J. Neurosci. Methods 183 , 107–113. Halassa, M.M., Florian, C., Fellin, T., Munoz, J.R., Lee, S.-Y., Abel, T., Haydon, P.G., and Frank, M.G. (2009). Astrocytic modulation of sleep homeostasis and cognitive consequences of sleep loss. Neuron 61 , 213–219. Hall, W., and Solowij, N. (1998). Adverse effects of cannabis. Lancet 352 , 1611–1616. Haller, J., Mátyás, F., Soproni, K., Varga, B., Barsy, B., Németh, B., Mikics, E., Freund, T.F., and Hájos, N. (2007). Correlated species differences in the effects of cannabinoid ligands on anxiety and on GABAergic and glutamatergic synaptic transmission. Eur. J. Neurosci. 25 , 2445–2456. Hampson, R.E., and Deadwyler, S.A. (1999). Cannabinoids, hippocampal function and memory. Life Sciences 65 , 715–723. Hampson, R.E., Mu, J., and Deadwyler, S.A. (2000). Cannabinoid and kappa opioid receptors reduce potassium K current via activation of G(s) proteins in cultured hippocampal neurons. J. Neurophysiol. 84 , 2356–2364. Hampson, R.E., Simeral, J.D., and Deadwyler, S.A. (1999). Distribution of spatial and nonspatial information in dorsal hippocampus. Nature 402 , 610–614.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 141
Hanson, K.L., Winward, J.L., Schweinsburg, A.D., Medina, K.L., Brown, S.A., and Tapert, S.F. (2010). Longitudinal study of cognition among adolescent marijuana users over three weeks of abstinence. Addict Behav 35 , 970–976. Häring, M., Marsicano, G., Lutz, B., and Monory, K. (2007). Identification of the cannabinoid receptor type 1 in serotonergic cells of raphe nuclei in mice. Neuroscience 146 , 1212–1219. Harloe, J.P., Thorpe, A.J., and Lichtman, A.H. (2008). Differential endocannabinoid regulation of extinction in appetitive and aversive Barnes maze tasks. Learn. Mem. 15 , 806–809. Hartley, C.A., and Phelps, E.A. (2010). Changing fear: the neurocircuitry of emotion regulation. Neuropsychopharmacology 35 , 136–146. Hashimotodani, Y., Ohno-Shosaku, T., and Kano, M. (2007). Ca(2+)-assisted receptor-driven endocannabinoid release: mechanisms that associate presynaptic and postsynaptic activities. Curr. Opin. Neurobiol. 17 , 360–365. Hashimotodani, Y., Ohno-Shosaku, T., Tsubokawa, H., Ogata, H., Emoto, K., Maejima, T., Araishi, K., Shin, H.-S., and Kano, M. (2005). Phospholipase Cbeta serves as a coincidence detector through its Ca2+ dependency for triggering retrograde endocannabinoid signal. Neuron 45 , 257–268. Haydon, P.G. (2001). GLIA: listening and talking to the synapse. Nat. Rev. Neurosci. 2, 185– 193. Heifets, B.D., Chevaleyre, V., and Castillo, P.E. (2008). Interneuron activity controls endocannabinoid-mediated presynaptic plasticity through calcineurin. Proc. Natl. Acad. Sci. U.S.A. 105 , 10250–10255. Henneberger, C., Papouin, T., Oliet, S.H.R., and Rusakov, D.A. (2010). Long-term potentiation depends on release of D-serine from astrocytes. Nature 463 , 232–236. Hensch, T.K., Gordon, J.A., Brandon, E.P., McKnight, G.S., Idzerda, R.L., and Stryker, M.P. (1998). Comparison of plasticity in vivo and in vitro in the developing visual cortex of normal and protein kinase A RIbeta-deficient mice. J. Neurosci. 18 , 2108–2117. Herkenham, M., Lynn, A.B., Little, M.D., Johnson, M.R., Melvin, L.S., de Costa, B.R., and Rice, K.C. (1990). Cannabinoid receptor localization in brain. Proc. Natl. Acad. Sci. U.S.A. 87 , 1932–1936. Hermann, H., Marsicano, G., and Lutz, B. (2002). Coexpression of the cannabinoid receptor type 1 with dopamine and serotonin receptors in distinct neuronal subpopulations of the adult mouse forebrain. Neuroscience 109 , 451–460. Herry, C., Ciocchi, S., Senn, V., Demmou, L., Müller, C., and Lüthi, A. (2008). Switching on and off fear by distinct neuronal circuits. Nature 454 , 600–606. Herry, C., Trifilieff, P., Micheau, J., Lüthi, A., and Mons, N. (2006). Extinction of auditory fear conditioning requires MAPK/ERK activation in the basolateral amygdala. Eur. J. Neurosci. 24 , 261– 269. Heyser, C.J., Hampson, R.E., and Deadwyler, S.A. (1993). Effects of delta-9- tetrahydrocannabinol on delayed match to sample performance in rats: alterations in short-term memory associated with changes in task specific firing of hippocampal cells. Journal of Pharmacology and Experimental Therapeutics 264 , 294–307. Higuchi, S., Irie, K., Mishima, S., Araki, M., Ohji, M., Shirakawa, A., Akitake, Y., Matsuyama, K., Mishima, K., Mishima, K., et al. (2010). The cannabinoid 1-receptor silent antagonist O-2050 attenuates preference for high-fat diet and activated astrocytes in mice. J. Pharmacol. Sci. 112 , 369– 372. Hill, A.B. (1965). The environment and disease: association or causation? Proc. R. Soc. Med. 58 , 295–300. Hill, M.N., Patel, S., Campolongo, P., Tasker, J.G., Wotjak, C.T., and Bains, J.S. (2010). Functional interactions between stress and the endocannabinoid system: from synaptic signaling to behavioral output. J. Neurosci. 30 , 14980–14986. Hillard, C.J., Edgemond, W.S., Jarrahian, A., and Campbell, W.B. (1997). Accumulation of N- arachidonoylethanolamine (anandamide) into cerebellar granule cells occurs via facilitated diffusion. J. Neurochem. 69 , 631–638.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 142
Hillard, C.J., Harris, R.A., and Bloom, A.S. (1985). Effects of the cannabinoids on physical properties of brain membranes and phospholipid vesicles: fluorescence studies. J. Pharmacol. Exp. Ther. 232 , 579–588. Hirasawa, M., Schwab, Y., Natah, S., Hillard, C.J., Mackie, K., Sharkey, K.A., and Pittman, Q.J. (2004). Dendritically released transmitters cooperate via autocrine and retrograde actions to inhibit afferent excitation in rat brain. J. Physiol. (Lond.) 559 , 611–624. Hirrlinger, P.G., Scheller, A., Braun, C., Hirrlinger, J., and Kirchhoff, F. (2006). Temporal control of gene recombination in astrocytes by transgenic expression of the tamoxifen-inducible DNA recombinase variant CreERT2. Glia 54 , 11–20. Hoffman, A.F., and Lupica, C.R. (2000). Mechanisms of Cannabinoid Inhibition of GABAASynaptic Transmission in the Hippocampus. The Journal of Neuroscience 20 , 2470–2479. Hölter, S.M., Kallnik, M., Wurst, W., Marsicano, G., Lutz, B., and Wotjak, C.T. (2005). Cannabinoid CB1 receptor is dispensable for memory extinction in an appetitively-motivated learning task. Eur. J. Pharmacol. 510 , 69–74. Houades, V., Koulakoff, A., Ezan, P., Seif, I., and Giaume, C. (2008). Gap junction-mediated astrocytic networks in the mouse barrel cortex. J. Neurosci. 28 , 5207–5217. Howlett, A.C. (1984). Inhibition of neuroblastoma adenylate cyclase by cannabinoid and nantradol compounds. Life Sci. 35 , 1803–1810. Howlett, A.C. (2002). The cannabinoid receptors. Prostaglandins Other Lipid Mediat. 68-69 , 619–631. Howlett, A.C., Johnson, M.R., Melvin, L.S., and Milne, G.M. (1988). Nonclassical cannabinoid analgetics inhibit adenylate cyclase: development of a cannabinoid receptor model. Mol. Pharmacol. 33 , 297–302. Howlett, A.C., Qualy, J.M., and Khachatrian, L.L. (1986). Involvement of Gi in the inhibition of adenylate cyclase by cannabimimetic drugs. Mol. Pharmacol. 29 , 307–313. Hsieh, C., Brown, S., Derleth, C., and Mackie, K. (1999). Internalization and recycling of the CB1 cannabinoid receptor. J. Neurochem. 73 , 493–501. Huntley, J.D., and Howard, R.J. (2010). Working memory in early Alzheimer’s disease: a neuropsychological review. Int J Geriatr Psychiatry 25 , 121–132. I
Ivanov, A., Mukhtarov, M., Bregestovski, P., and Zilberter, Y. (2011). Lactate Effectively Covers Energy Demands during Neuronal Network Activity in Neonatal Hippocampal Slices. Front Neuroenergetics 3, 2. Ivanov, A., and Zilberter, Y. (2011). Critical state of energy metabolism in brain slices: the principal role of oxygen delivery and energy substrates in shaping neuronal activity. Front Neuroenergetics 3, 9. J
Jacob, A., and Todd, A.R. (1940). Cannabidiol and Cannabol, Constituents of Cannabis indica Resin. Nature 145 , 350–350. Jaffard, R., Etchamendy, N., Desmedt, A., Krazem, A., Cortes-Torrea, C., and Marighetto, A. (2000). [An animal model of human declarative (relational) memory and of its dysfunction]. Therapie 55 , 477–485.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 143
Jentsch, J.D., Andrusiak, E., Tran, A., Bowers, M.B., Jr, and Roth, R.H. (1997). Delta 9- tetrahydrocannabinol increases prefrontal cortical catecholaminergic utilization and impairs spatial working memory in the rat: blockade of dopaminergic effects with HA966. Neuropsychopharmacology 16 , 426–432. Ji, J., and Maren, S. (2005). Electrolytic lesions of the dorsal hippocampus disrupt renewal of conditional fear after extinction. Learn. Mem. 12 , 270–276. Jin, K., Xie, L., Kim, S.H., Parmentier-Batteur, S., Sun, Y., Mao, X.O., Childs, J., and Greenberg, D.A. (2004). Defective adult neurogenesis in CB1 cannabinoid receptor knockout mice. Mol. Pharmacol. 66 , 204–208. Johansen, J.P., Cain, C.K., Ostroff, L.E., and LeDoux, J.E. (2011). Molecular mechanisms of fear learning and memory. Cell 147 , 509–524. Jones, R.T. (1978). Behavioral tolerance: lessons learned from cannabis research. NIDA Res. Monogr. 118–126. K
Kaczocha, M., Glaser, S.T., and Deutsch, D.G. (2009). Identification of intracellular carriers for the endocannabinoid anandamide. Proc. Natl. Acad. Sci. U.S.A. 106 , 6375–6380. Kamprath, K., Marsicano, G., Tang, J., Monory, K., Bisogno, T., Di Marzo, V., Lutz, B., and Wotjak, C.T. (2006). Cannabinoid CB1 receptor mediates fear extinction via habituation-like processes. J. Neurosci. 26 , 6677–6686. Kamprath, K., Plendl, W., Marsicano, G., Deussing, J.M., Wurst, W., Lutz, B., and Wotjak, C.T. (2009). Endocannabinoids mediate acute fear adaptation via glutamatergic neurons independently of corticotropin-releasing hormone signaling. Genes Brain Behav. 8, 203–211. Kamprath, K., Romo-Parra, H., Häring, M., Gaburro, S., Doengi, M., Lutz, B., and Pape, H.-C. (2011). Short-term adaptation of conditioned fear responses through endocannabinoid signaling in the central amygdala. Neuropsychopharmacology 36 , 652–663. Kamprath, K., and Wotjak, C.T. (2004). Nonassociative learning processes determine expression and extinction of conditioned fear in mice. Learn. Mem. 11 , 770–786. Kandel, E.R. (2001). The molecular biology of memory storage: a dialogue between genes and synapses. Science 294 , 1030–1038. Kang, J., Jiang, L., Goldman, S.A., and Nedergaard, M. (1998). Astrocyte-mediated potentiation of inhibitory synaptic transmission. Nat. Neurosci. 1, 683–692. Kano, M., Ohno-Shosaku, T., Hashimotodani, Y., Uchigashima, M., and Watanabe, M. (2009). Endocannabinoid-mediated control of synaptic transmission. Physiol. Rev. 89 , 309–380. Kathuria, S., Gaetani, S., Fegley, D., Valiño, F., Duranti, A., Tontini, A., Mor, M., Tarzia, G., La Rana, G., Calignano, A., et al. (2003). Modulation of anxiety through blockade of anandamide hydrolysis. Nat. Med. 9, 76–81. Katona, I., Rancz, E.A., Acsady, L., Ledent, C., Mackie, K., Hajos, N., and Freund, T.F. (2001). Distribution of CB1 cannabinoid receptors in the amygdala and their role in the control of GABAergic transmission. J. Neurosci. 21 , 9506–9518. Katona, I., Sperlágh, B., Maglóczky, Z., Sántha, E., Köfalvi, A., Czirják, S., Mackie, K., Vizi, E.S., and Freund, T.F. (2000). GABAergic interneurons are the targets of cannabinoid actions in the human hippocampus. Neuroscience 100 , 797–804. Katona, I., Sperlágh, B., Sík, A., Käfalvi, A., Vizi, E.S., Mackie, K., and Freund, T.F. (1999). Presynaptically located CB1 cannabinoid receptors regulate GABA release from axon terminals of specific hippocampal interneurons. J. Neurosci. 19 , 4544–4558.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 144
Katona, I., Urbán, G.M., Wallace, M., Ledent, C., Jung, K.-M., Piomelli, D., Mackie, K., and Freund, T.F. (2006). Molecular composition of the endocannabinoid system at glutamatergic synapses. J. Neurosci. 26 , 5628–5637. Kawamura, Y., Fukaya, M., Maejima, T., Yoshida, T., Miura, E., Watanabe, M., Ohno-Shosaku, T., and Kano, M. (2006). The CB1 cannabinoid receptor is the major cannabinoid receptor at excitatory presynaptic sites in the hippocampus and cerebellum. J. Neurosci. 26 , 2991–3001. Kelly, T.H., Foltin, R.W., Emurian, C.S., and Fischman, M.W. (1990). Multidimensional behavioral effects of marijuana. Prog. Neuropsychopharmacol. Biol. Psychiatry 14 , 885–902. Kemp, A., and Manahan-Vaughan, D. (2004). Hippocampal long-term depression and long-term potentiation encode different aspects of novelty acquisition. Proc. Natl. Acad. Sci. U.S.A. 101 , 8192– 8197. Kempermann, G., Kuhn, H.G., and Gage, F.H. (1997). More hippocampal neurons in adult mice living in an enriched environment. Nature 386 , 493–495. Kesner, R.P., Hunt, M.E., Williams, J.M., and Long, J.M. (1996). Prefrontal cortex and working memory for spatial response, spatial location, and visual object information in the rat. Cereb. Cortex 6, 311–318. Kettenmann, H., and Ransom, B.R. (1995). Neuroglia (New York: Oxford University Press). Kettenmann, H., and Verkhratsky, A. (2008). Neuroglia: the 150 years after. Trends Neurosci. 31 , 653–659. Kim, J., and Alger, B.E. (2004). Inhibition of cyclooxygenase-2 potentiates retrograde endocannabinoid effects in hippocampus. Nat. Neurosci. 7, 697–698. Kim, J., Isokawa, M., Ledent, C., and Alger, B.E. (2002). Activation of muscarinic acetylcholine receptors enhances the release of endogenous cannabinoids in the hippocampus. J. Neurosci. 22 , 10182–10191. Kim, J., Lee, S., Park, K., Hong, I., Song, B., Son, G., Park, H., Kim, W.R., Park, E., Choe, H.K., et al. (2007). Amygdala depotentiation and fear extinction. Proc. Natl. Acad. Sci. U.S.A. 104 , 20955– 20960. Kim, J.J., and Fanselow, M.S. (1992). Modality-specific retrograde amnesia of fear. Science 256 , 675–677. Kim, J.J., and Jung, M.W. (2006). Neural circuits and mechanisms involved in Pavlovian fear conditioning: a critical review. Neurosci Biobehav Rev 30 , 188–202. Kimelberg, H.K. (2004). The problem of astrocyte identity. Neurochem. Int. 45 , 191–202. Kiplinger, G.F., and Manno, J.E. (1971). Dose-response relationships to cannabis in human subjects. Pharmacol. Rev. 23 , 339–347. Klein, T.W., Newton, C., Larsen, K., Lu, L., Perkins, I., Nong, L., and Friedman, H. (2003). The cannabinoid system and immune modulation. J. Leukoc. Biol. 74 , 486–496. Koob, A. (2009). The root of thought : unlocking glia--the brain cell that will help us sharpen our wits, heal injury, and treat brain disease (Upper Saddle River, N.J.: FT Press). Koolhaas, J.M., De Boer, S.F., Coppens, C.M., and Buwalda, B. (2010). Neuroendocrinology of coping styles: Towards understanding the biology of individual variation. Frontiers in Neuroendocrinology 31 , 307–321. Koolhaas, J.M., Korte, S.M., De Boer, S.F., Van Der Vegt, B.J., Van Reenen, C.G., Hopster, H., De Jong, I.C., Ruis, M.A.W., and Blokhuis, H.J. (1999). Coping styles in animals: current status in behavior and stress-physiology. Neuroscience & Biobehavioral Reviews 23 , 925–935. Kossoff, E.H., and Rho, J.M. (2009). Ketogenic diets: evidence for short- and long-term efficacy. Neurotherapeutics 6, 406–414. Kouznetsova, M., Kelley, B., Shen, M., and Thayer, S.A. (2002). Desensitization of cannabinoid- mediated presynaptic inhibition of neurotransmission between rat hippocampal neurons in culture. Mol. Pharmacol. 61 , 477–485.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 145
Kreitzer, A.C., and Malenka, R.C. (2005). Dopamine modulation of state-dependent endocannabinoid release and long-term depression in the striatum. J. Neurosci. 25 , 10537–10545. Kreitzer, A.C., and Regehr, W.G. (2001). Cerebellar depolarization-induced suppression of inhibition is mediated by endogenous cannabinoids. J. Neurosci. 21 , RC174. Kucewicz, M.T., Tricklebank, M.D., Bogacz, R., and Jones, M.W. (2011). Dysfunctional Prefrontal Cortical Network Activity and Interactions following Cannabinoid Receptor Activation. The Journal of Neuroscience 31 , 15560–15568. L
LaBar, K.S., LeDoux, J.E., Spencer, D.D., and Phelps, E.A. (1995). Impaired fear conditioning following unilateral temporal lobectomy in humans. J. Neurosci. 15 , 6846–6855. Labrie, V., Fukumura, R., Rastogi, A., Fick, L.J., Wang, W., Boutros, P.C., Kennedy, J.L., Semeralul, M.O., Lee, F.H., Baker, G.B., et al. (2009). Serine racemase is associated with schizophrenia susceptibility in humans and in a mouse model. Hum. Mol. Genet. 18 , 3227–3243. Van Laere, K., Goffin, K., Bormans, G., Casteels, C., Mortelmans, L., de Hoon, J., Grachev, I., Vandenbulcke, M., and Pieters, G. (2009). Relationship of type 1 cannabinoid receptor availability in the human brain to novelty-seeking temperament. Arch. Gen. Psychiatry 66 , 196–204. Lafourcade, M., Elezgarai, I., Mato, S., Bakiri, Y., Grandes, P., and Manzoni, O.J. (2007). Molecular Components and Functions of the Endocannabinoid System in Mouse Prefrontal Cortex. PLoS ONE 2, e709. Lal, H., and Forster, M.J. (1990). Flumazenil improves active avoidance performance in aging NZB/BlNJ and C57BL/6NNia mice. Pharmacol. Biochem. Behav. 35 , 747–750. Lan, R., Gatley, J., Lu, Q., Fan, P., Fernando, S.R., Volkow, N.D., Pertwee, R., and Makriyannis, A. (1999). Design and synthesis of the CB1 selective cannabinoid antagonist AM281: a potential human SPECT ligand. AAPS PharmSci 1, E4. Laroche, S., Davis, S., and Jay, T.M. (2000). Plasticity at hippocampal to prefrontal cortex synapses: dual roles in working memory and consolidation. Hippocampus 10 , 438–446. Laviolette, S.R., and Grace, A.A. (2006). Cannabinoids Potentiate Emotional Learning Plasticity in Neurons of the Medial Prefrontal Cortex through Basolateral Amygdala Inputs. J. Neurosci. 26 , 6458–6468. Laviolette, S.R., Lipski, W.J., and Grace, A.A. (2005). A subpopulation of neurons in the medial prefrontal cortex encodes emotional learning with burst and frequency codes through a dopamine D4 receptor-dependent basolateral amygdala input. J. Neurosci. 25 , 6066–6075. Lázaro-Muñoz, G., LeDoux, J.E., and Cain, C.K. (2010). Sidman instrumental avoidance initially depends on lateral and basal amygdala and is constrained by central amygdala-mediated Pavlovian processes. Biol. Psychiatry 67 , 1120–1127. Lazarus, R.S., and Folkman, S. (1984). Stress, appraisal, and coping (New York: Springer Pub. Co.). Ledent, C., Valverde, O., Cossu, G., Petitet, F., Aubert, J.F., Beslot, F., Böhme, G.A., Imperato, A., Pedrazzini, T., Roques, B.P., et al. (1999). Unresponsiveness to cannabinoids and reduced addictive effects of opiates in CB1 receptor knockout mice. Science 283 , 401–404. LeDoux, J.E. (2000). Emotion circuits in the brain. Annu. Rev. Neurosci. 23 , 155–184. LeDoux, J.E. (2012). Evolution of human emotion: a view through fear. Prog. Brain Res. 195 , 431–442. LeDoux, J.E., Cicchetti, P., Xagoraris, A., and Romanski, L.M. (1990). The lateral amygdaloid nucleus: sensory interface of the amygdala in fear conditioning. J. Neurosci. 10 , 1062–1069.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 146
LeDoux, J.E., Iwata, J., Cicchetti, P., and Reis, D.J. (1988). Different projections of the central amygdaloid nucleus mediate autonomic and behavioral correlates of conditioned fear. J. Neurosci. 8, 2517–2529. LeDoux, J.E., Sakaguchi, A., and Reis, D.J. (1984). Subcortical efferent projections of the medial geniculate nucleus mediate emotional responses conditioned to acoustic stimuli. J. Neurosci. 4, 683–698. Lee, I., and Kesner, R.P. (2003). Time-dependent relationship between the dorsal hippocampus and the prefrontal cortex in spatial memory. J. Neurosci. 23 , 1517–1523. Lehner, M., Taracha, E., Skórzewska, A., Turzynska, D., Sobolewska, A., Maciejak, P., Szyndler, J., Hamed, A., Bidzinski, A., Wislowska-Stanek, A., et al. (2008). Expression of c-Fos and CRF in the brains of rats differing in the strength of a fear response. Behavioural Brain Research 188 , 154–167. Lehner, M., Wislowska-Stanek, A., Taracha, E., Maciejak, P., Szyndler, J., Skórzewska, A., Turzynska, D., Sobolewska, A., Hamed, A., Bidzinski, A., et al. (2010). The effects of midazolam and d-cycloserine on the release of glutamate and GABA in the basolateral amygdala of low and high anxiety rats during extinction trial of a conditioned fear test. Neurobiology of Learning and Memory 94 , 468–480. Lemberger, T., Parlato, R., Dassesse, D., Westphal, M., Casanova, E., Turiault, M., Tronche, F., Schiffmann, S.N., and Schütz, G. (2007). Expression of Cre recombinase in dopaminoceptive neurons. BMC Neurosci 8, 4. von Lenhossek, M. (1891). Zur Kenntnis der Neuroglia des menschlichen Ruckenmarkes. Verh Anat Ges 5, 193–221. Li, J., Schiller, D., Schoenbaum, G., Phelps, E.A., and Daw, N.D. (2011). Differential roles of human striatum and amygdala in associative learning. Nature Neuroscience 14 , 1250–1252. Lichtman, A.H. (2000). SR 141716A enhances spatial memory as assessed in a radial-arm maze task in rats. Eur. J. Pharmacol. 404 , 175–179. Lichtman, A.H., Dimen, K.R., and Martin, B.R. (1995). Systemic or intrahippocampal cannabinoid administration impairs spatial memory in rats. Psychopharmacology (Berl.) 119 , 282–290. Ligresti, A., Morera, E., Van Der Stelt, M., Monory, K., Lutz, B., Ortar, G., and Di Marzo, V. (2004). Further evidence for the existence of a specific process for the membrane transport of anandamide. Biochem. J. 380 , 265–272. Lin, C.H., Yeh, S.H., Lin, C.H., Lu, K.T., Leu, T.H., Chang, W.C., and Gean, P.W. (2001). A role for the PI-3 kinase signaling pathway in fear conditioning and synaptic plasticity in the amygdala. Neuron 31 , 841–851. Lin, Q.-S., Yang, Q., Liu, D.-D., Sun, Z., Dang, H., Liang, J., Wang, Y.-X., Chen, J., and Li, S.-T. (2011). Hippocampal endocannabinoids play an important role in induction of long-term potentiation and regulation of contextual fear memory formation. Brain Res. Bull. 86 , 139–145. Lisboa, S.F., Reis, D.G., da Silva, A.L., Corrêa, F.M.A., Guimarães, F.S., and Resstel, L.B.M. (2010). Cannabinoid CB1 receptors in the medial prefrontal cortex modulate the expression of contextual fear conditioning. Int. J. Neuropsychopharmacol. 13 , 1163–1173. Little, P.J., Compton, D.R., Johnson, M.R., Melvin, L.S., and Martin, B.R. (1988). Pharmacology and stereoselectivity of structurally novel cannabinoids in mice. J. Pharmacol. Exp. Ther. 247 , 1046– 1051. Lograno, M.D., and Romano, M.R. (2004). Cannabinoid agonists induce contractile responses through Gi/o-dependent activation of phospholipase C in the bovine ciliary muscle. Eur. J. Pharmacol. 494 , 55–62. Long, J.Z., Li, W., Booker, L., Burston, J.J., Kinsey, S.G., Schlosburg, J.E., Pavón, F.J., Serrano, A.M., Selley, D.E., Parsons, L.H., et al. (2009a). Selective blockade of 2-arachidonoylglycerol hydrolysis produces cannabinoid behavioral effects. Nat. Chem. Biol. 5, 37–44. Long, J.Z., Nomura, D.K., Vann, R.E., Walentiny, D.M., Booker, L., Jin, X., Burston, J.J., Sim- Selley, L.J., Lichtman, A.H., Wiley, J.L., et al. (2009b). Dual blockade of FAAH and MAGL identifies
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 147
behavioral processes regulated by endocannabinoid crosstalk in vivo. Proc. Natl. Acad. Sci. U.S.A. 106 , 20270–20275. Van der Loos, H., and Glaser, E.M. (1972). Autapses in neocortex cerebri: synapses between a pyramidal cell’s axon and its own dendrites. Brain Res. 48 , 355–360. López Rodríguez, A.B., Mateos Vicente, B., Romero-Zerbo, S.Y., Rodriguez-Rodriguez, N., Bellini, M.J., Rodriguez de Fonseca, F., Bermudez-Silva, F.J., Azcoitia, I., Garcia-Segura, L.M., and Viveros, M.-P. (2011). Estradiol decreases cortical reactive astrogliosis after brain injury by a mechanism involving cannabinoid receptors. Cereb. Cortex 21 , 2046–2055. López-Gallardo, M., López-Rodríguez, A.B., Llorente-Berzal, Á., Rotllant, D., Mackie, K., Armario, A., Nadal, R., and Viveros, M.-P. (2012). Maternal deprivation and adolescent cannabinoid exposure impact hippocampal astrocytes, CB1 receptors and brain-derived neurotrophic factor in a sexually dimorphic fashion. Neuroscience 204 , 90–103. Loría, F., Petrosino, S., Hernangómez, M., Mestre, L., Spagnolo, A., Correa, F., Di Marzo, V., Docagne, F., and Guaza, C. (2010). An endocannabinoid tone limits excitotoxicity in vitro and in a model of multiple sclerosis. Neurobiol. Dis. 37 , 166–176. Lou, Z.-Y., Zhao, C.-B., and Xiao, B.-G. (2012). Immunoregulation of experimental autoimmune encephalomyelitis by the selective CB1 receptor antagonist. J. Neurosci. Res. 90 , 84–95. Lourenço, J., Cannich, A., Carta, M., Coussen, F., Mulle, C., and Marsicano, G. (2010). Synaptic activation of kainate receptors gates presynaptic CB1 signaling at GABAergic synapses. Nat Neurosci 13 , 197–204. Ludányi, A., Eross, L., Czirják, S., Vajda, J., Halász, P., Watanabe, M., Palkovits, M., Maglóczky, Z., Freund, T.F., and Katona, I. (2008). Downregulation of the CB1 cannabinoid receptor and related molecular elements of the endocannabinoid system in epileptic human hippocampus. J. Neurosci. 28 , 2976–2990. Lugaro, E (1907). Sulk funzioni della nevroglia. Riu. Pat. Neru. Ment. 12 , 225–233. M
Mackie, K. (2005). Cannabinoid receptor homo- and heterodimerization. Life Sci. 77 , 1667– 1673. Mackie, K. (2006). Cannabinoid receptors as therapeutic targets. Annu. Rev. Pharmacol. Toxicol. 46 , 101–122. Mackie, K., Devane, W.A., and Hille, B. (1993). Anandamide, an endogenous cannabinoid, inhibits calcium currents as a partial agonist in N18 neuroblastoma cells. Mol. Pharmacol. 44 , 498– 503. Maejima, T., Hashimoto, K., Yoshida, T., Aiba, A., and Kano, M. (2001). Presynaptic inhibition caused by retrograde signal from metabotropic glutamate to cannabinoid receptors. Neuron 31 , 463– 475. Magistretti, P.J., and Pellerin, L. (1999). Cellular mechanisms of brain energy metabolism and their relevance to functional brain imaging. Philos. Trans. R. Soc. Lond., B, Biol. Sci. 354 , 1155–1163. Magistretti, P.J., Sorg, O., Yu, N., Martin, J.L., and Pellerin, L. (1993). Neurotransmitters regulate energy metabolism in astrocytes: implications for the metabolic trafficking between neural cells. Dev. Neurosci. 15 , 306–312. Mahoney, J.M., and Harris, R.A. (1972). Effect of 9 -tetrahydrocannabinol on mitochondrial precesses. Biochem. Pharmacol. 21 , 1217–1226. Mailleux, P., and Vanderhaeghen, J.J. (1992). Distribution of neuronal cannabinoid receptor in the adult rat brain: a comparative receptor binding radioautography and in situ hybridization histochemistry. Neuroscience 48 , 655–668.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 148
Maione, S., Bisogno, T., de Novellis, V., Palazzo, E., Cristino, L., Valenti, M., Petrosino, S., Guglielmotti, V., Rossi, F., and Di Marzo, V. (2006). Elevation of endocannabinoid levels in the ventrolateral periaqueductal grey through inhibition of fatty acid amide hydrolase affects descending nociceptive pathways via both cannabinoid receptor type 1 and transient receptor potential vanilloid type-1 receptors. J. Pharmacol. Exp. Ther. 316 , 969–982. Makkar, S.R., Zhang, S.Q., and Cranney, J. (2010). Behavioral and Neural Analysis of GABA in the Acquisition, Consolidation, Reconsolidation, and Extinction of Fear Memory. Neuropsychopharmacology. Malenka, R.C., and Bear, M.F. (2004). LTP and LTD: an embarrassment of riches. Neuron 44 , 5–21. Mallet, P.E., and Beninger, R.J. (1998). The cannabinoid CB1 receptor antagonist SR141716A attenuates the memory impairment produced by delta9-tetrahydrocannabinol or anandamide. Psychopharmacology (Berl.) 140 , 11–19. Mansuy, I.M., Mayford, M., Jacob, B., Kandel, E.R., and Bach, M.E. (1998). Restricted and regulated overexpression reveals calcineurin as a key component in the transition from short-term to long-term memory. Cell 92 , 39–49. Maren, S. (1996). Synaptic transmission and plasticity in the amygdala. An emerging physiology of fear conditioning circuits. Mol. Neurobiol. 13 , 1–22. Marinelli, S., Pacioni, S., Bisogno, T., Di Marzo, V., Prince, D.A., Huguenard, J.R., and Bacci, A. (2008). The endocannabinoid 2-arachidonoylglycerol is responsible for the slow self-inhibition in neocortical interneurons. J. Neurosci. 28 , 13532–13541. Marinelli, S., Pacioni, S., Cannich, A., Marsicano, G., and Bacci, A. (2009). Self-modulation of neocortical pyramidal neurons by endocannabinoids. Nat. Neurosci. 12 , 1488–1490. Marrs, W.R., Blankman, J.L., Horne, E.A., Thomazeau, A., Lin, Y.H., Coy, J., Bodor, A.L., Muccioli, G.G., Hu, S.S.-J., Woodruff, G., et al. (2010). The serine hydrolase ABHD6 controls the accumulation and efficacy of 2-AG at cannabinoid receptors. Nat. Neurosci. 13 , 951–957. Marsch, R., Foeller, E., Rammes, G., Bunck, M., Kössl, M., Holsboer, F., Zieglgänsberger, W., Landgraf, R., Lutz, B., and Wotjak, C.T. (2007). Reduced anxiety, conditioned fear, and hippocampal long-term potentiation in transient receptor potential vanilloid type 1 receptor-deficient mice. The Journal of Neuroscience 27 , 832–839. Marsicano, G., Goodenough, S., Monory, K., Hermann, H., Eder, M., Cannich, A., Azad, S.C., Cascio, M.G., Gutiérrez, S.O., van der Stelt, M., et al. (2003). CB1 cannabinoid receptors and on- demand defense against excitotoxicity. Science 302 , 84–88. Marsicano, G., and Lutz, B. (1999). Expression of the cannabinoid receptor CB1 in distinct neuronal subpopulations in the adult mouse forebrain. Eur. J. Neurosci. 11 , 4213–4225. Marsicano, G., Wotjak, C.T., Azad, S.C., Bisogno, T., Rammes, G., Cascio, M.G., Hermann, H., Tang, J., Hofmann, C., Zieglgänsberger, W., et al. (2002). The endogenous cannabinoid system controls extinction of aversive memories. Nature 418 , 530–534. Martin, J.R., Oettinger, R., Driscoll, P., Buzzi, R., and Bättig, K. (1982). Effects of chlordiazepoxide and imipramine on maze patrolling within two different maze configurations by psychogenetically selected lines of rats. Psychopharmacology (Berl.) 78 , 58–62. Martin, L.A., Wei, D.S., and Alger, B.E. (2001). Heterogeneous susceptibility of GABA(A) receptor-mediated IPSCs to depolarization-induced suppression of inhibition in rat hippocampus. J. Physiol. (Lond.) 532 , 685–700. Martin, M., Ledent, C., Parmentier, M., Maldonado, R., and Valverde, O. (2002). Involvement of CB1 cannabinoid receptors in emotional behaviour. Psychopharmacology (Berl.) 159 , 379–387. Di Marzo, V., Fontana, A., Cadas, H., Schinelli, S., Cimino, G., Schwartz, J.C., and Piomelli, D. (1994). Formation and inactivation of endogenous cannabinoid anandamide in central neurons. Nature 372 , 686–691. Di Marzo, V., Melck, D., Bisogno, T., and De Petrocellis, L. (1998). Endocannabinoids: endogenous cannabinoid receptor ligands with neuromodulatory action. Trends Neurosci. 21 , 521– 528.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 149
Di Marzo, V., and Petrocellis, L.D. (2006). Plant, Synthetic, and Endogenous Cannabinoids in Medicine. Annu. Rev. Med. 57 , 553–574. Massa, F., Koehl, M., Koelh, M., Wiesner, T., Grosjean, N., Revest, J.-M., Piazza, P.-V., Abrous, D.N., and Oliet, S.H.R. (2011). Conditional reduction of adult neurogenesis impairs bidirectional hippocampal synaptic plasticity. Proc. Natl. Acad. Sci. U.S.A. 108 , 6644–6649. Mato, S., Chevaleyre, V., Robbe, D., Pazos, A., Castillo, P.E., and Manzoni, O.J. (2004). A single in-vivo exposure to delta 9THC blocks endocannabinoid-mediated synaptic plasticity. Nat. Neurosci. 7, 585–586. Mato, S., Lafourcade, M., Robbe, D., Bakiri, Y., and Manzoni, O.J. (2008). Role of the cyclic- AMP/PKA cascade and of P/Q-type Ca++ channels in endocannabinoid-mediated long-term depression in the nucleus accumbens. Neuropharmacology 54 , 87–94. Matsuda, L.A., Bonner, T.I., and Lolait, S.J. (1993). Localization of cannabinoid receptor mRNA in rat brain. J. Comp. Neurol. 327 , 535–550. Matsuda, L.A., Lolait, S.J., Brownstein, M.J., Young, A.C., and Bonner, T.I. (1990). Structure of a cannabinoid receptor and functional expression of the cloned cDNA. Nature 346 , 561–564. Mattson, M., Gleichmann, M., and Cheng, A. (2008). Mitochondria in Neuroplasticity and Neurological Disorders. Neuron 60 , 748–766. Mátyás, F., Yanovsky, Y., Mackie, K., Kelsch, W., Misgeld, U., and Freund, T.F. (2006). Subcellular localization of type 1 cannabinoid receptors in the rat basal ganglia. Neuroscience 137 , 337–361. McAllister, W.R., McAllister, D.E., and Douglass, W.K. (1971). The inverse relationship between shock intensity and shuttle-box avoidance learning in rats: A reinforcement explanation. Journal of Comparative and Physiological Psychology 74 , 426–433. McFarland, M.J., and Barker, E.L. (2004). Anandamide transport. Pharmacol. Ther. 104 , 117– 135. McNaughton, B.L. (1982). Long-term synaptic enhancement and short-term potentiation in rat fascia dentata act through different mechanisms. J. Physiol. (Lond.) 324 , 249–262. McNaughton, N., and Corr, P.J. (2004). A two-dimensional neuropsychology of defense: fear/anxiety and defensive distance. Neurosci Biobehav Rev 28 , 285–305. Mechoulam, R., Ben-Shabat, S., Hanus, L., Ligumsky, M., Kaminski, N.E., Schatz, A.R., Gopher, A., Almog, S., Martin, B.R., and Compton, D.R. (1995). Identification of an endogenous 2- monoglyceride, present in canine gut, that binds to cannabinoid receptors. Biochem. Pharmacol. 50 , 83–90. Mechoulam, R., Braun, P., and Gaoni, Y. (1972). Syntheses of. DELTA. 1-tetrahydrocannabinol and related cannabinoids. Journal of the American Chemical Society 94 , 6159–6165. Mechoulam, R., Breuer, A., Järbe, T.U., Hiltunen, A.J., and Glaser, R. (1990). Cannabimimetic activity of novel enantiomeric, benzofuran cannabinoids. J. Med. Chem. 33 , 1037–1043. Medina, J.F., Repa, J.C., Mauk, M.D., and LeDoux, J.E. (2002). Parallels between cerebellum- and amygdala-dependent conditioning. Nat. Rev. Neurosci. 3, 122–131. Melis, M., Pistis, M., Perra, S., Muntoni, A.L., Pillolla, G., and Gessa, G.L. (2004). Endocannabinoids mediate presynaptic inhibition of glutamatergic transmission in rat ventral tegmental area dopamine neurons through activation of CB1 receptors. J. Neurosci. 24 , 53–62. Middeldorp, J., and Hol, E.M. (2011). GFAP in health and disease. Prog. Neurobiol. 93 , 421– 443. Mikics, E., Dombi, T., Barsvári, B., Varga, B., Ledent, C., Freund, T.F., and Haller, J. (2006). The effects of cannabinoids on contextual conditioned fear in CB1 knockout and CD1 mice. Behav Pharmacol 17 , 223–230. Mikuriya, T.H. (1969). Marijuana in medicine: past, present and future. California Medicine 110 , 34.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 150
Milad, M.R., and Quirk, G.J. (2002). Neurons in medial prefrontal cortex signal memory for fear extinction. Nature 420 , 70–74. Miller, G.A. (1960). Plans and the structure of behavior (New York: Holt). Milner, B., Squire, L.R., and Kandel, E.R. (1998). Cognitive neuroscience and the study of memory. Neuron 20 , 445–468. Min, R., Di Marzo, V., and Mansvelder, H.D. (2010). DAG lipase involvement in depolarization- induced suppression of inhibition: does endocannabinoid biosynthesis always meet the demand? Neuroscientist 16 , 608–613. Mineka, S. (1979). The role of fear in theories of avoidance learning, flooding, and extinction. Psychological Bulletin 86 , 985. Misner, D.L., and Sullivan, J.M. (1999). Mechanism of cannabinoid effects on long-term potentiation and depression in hippocampal CA1 neurons. J. Neurosci. 19 , 6795–6805. Mitchell, V.A., Jeong, H.-J., Drew, G.M., and Vaughan, C.W. (2011). Cholecystokinin exerts an effect via the endocannabinoid system to inhibit GABAergic transmission in midbrain periaqueductal gray. Neuropsychopharmacology 36 , 1801–1810. Miyamoto, E. (2006). Molecular mechanism of neuronal plasticity: induction and maintenance of long-term potentiation in the hippocampus. J. Pharmacol. Sci. 100 , 433–442. Mokler, D.J., Robinson, S.E., Johnson, J.H., Hong, J.S., and Rosecrans, J.A. (1986). Effects of postweaning administration of delta-9-tetrahydrocannabinol (THC) on adult behavioral and neuroendocrine function in Sprague-Dawley and Fischer-344 rats. Neurobehav Toxicol Teratol 8, 407–413. Moldrich, G., and Wenger, T. (2000). Localization of the CB1 cannabinoid receptor in the rat brain. An immunohistochemical study. Peptides 21 , 1735–1742. Molina-Holgado, F., Lledó, A., and Guaza, C. (1997). Anandamide suppresses nitric oxide and TNF-alpha responses to Theiler’s virus or endotoxin in astrocytes. Neuroreport 8, 1929–1933. Molina-Holgado, F., Molina-Holgado, E., Guaza, C., and Rothwell, N.J. (2002). Role of CB1 and CB2 receptors in the inhibitory effects of cannabinoids on lipopolysaccharide-induced nitric oxide release in astrocyte cultures. Journal of Neuroscience Research 67 , 829–836. Monory, K., Blaudzun, H., Massa, F., Kaiser, N., Lemberger, T., Schütz, G., Wotjak, C.T., Lutz, B., and Marsicano, G. (2007). Genetic Dissection of Behavioural and Autonomic Effects of ∆9- Tetrahydrocannabinol in Mice. Plos Biol 5, e269. Monory, K., Massa, F., Egertová, M., Eder, M., Blaudzun, H., Westenbroek, R., Kelsch, W., Jacob, W., Marsch, R., and Ekker, M. (2006). The Endocannabinoid System Controls Key Epileptogenic Circuits in the Hippocampus. Neuron 51 , 455–466. Moreau, J. (1845). Du haschish et de l’aliénation mentale : études psychologiques (Paris: Fortin, Masson). Moreira, F.A., Aguiar, D.C., Campos, A.C., Lisboa, S.F., Terzian, A.L., Resstel, L.B., and Guimarães, F.S. (2009). Antiaversive effects of cannabinoids: is the periaqueductal gray involved? Neural Plast. 2009 , 625469. Morgan, M.A., Romanski, L.M., and LeDoux, J.E. (1993). Extinction of emotional learning: contribution of medial prefrontal cortex. Neurosci. Lett. 163 , 109–113. Morris, R. (1984). Developments of a water-maze procedure for studying spatial learning in the rat. J Neurosci Methods 11 , 47–60. Mowrer, O.H., and Lamoreaux, R.R. (1946). Fear as an intervening variable in avoidance conditioning. J Comp Psychol 39 , 29–50. Mu, J., Zhuang, S.Y., Hampson, R.E., and Deadwyler, S.A. (2000). Protein kinase-dependent phosphorylation and cannabinoid receptor modulation of potassium A current (IA) in cultured rat hippocampal neurons. Pflugers Arch. 439 , 541–546. Mulder, J., Aguado, T., Keimpema, E., Barabás, K., Ballester Rosado, C.J., Nguyen, L., Monory, K., Marsicano, G., Di Marzo, V., Hurd, Y.L., et al. (2008). Endocannabinoid signaling controls
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 151
pyramidal cell specification and long-range axon patterning. Proceedings of the National Academy of Sciences 105 , 8760. Mulligan, S.J., and MacVicar, B.A. (2004). Calcium transients in astrocyte endfeet cause cerebrovascular constrictions. Nature 431 , 195–199. Munro, S., Thomas, K.L., and Abu-Shaar, M. (1993). Molecular characterization of a peripheral receptor for cannabinoids. Nature 365 , 61–65. Murillo-Rodriguez, E., Blanco-Centurion, C., Sanchez, C., Piomelli, D., and Shiromani, P.J. (2003). Anandamide enhances extracellular levels of adenosine and induces sleep: an in vivo microdialysis study. Sleep 26 , 943–947. Myers, K.M., and Davis, M. (2007). Mechanisms of fear extinction. Mol. Psychiatry 12 , 120–150. N
Nader, K., Majidishad, P., Amorapanth, P., and LeDoux, J.E. (2001). Damage to the lateral and central, but not other, amygdaloid nuclei prevents the acquisition of auditory fear conditioning. Learn. Mem. 8, 156–163. Nava, F., Carta, G., Battasi, A.M., and Gessa, G.L. (2000). D(2) dopamine receptors enable delta(9)-tetrahydrocannabinol induced memory impairment and reduction of hippocampal extracellular acetylcholine concentration. Br. J. Pharmacol. 130 , 1201–1210. Nava, F., Carta, G., Colombo, G., and Gessa, G.L. (2001). Effects of chronic Delta(9)- tetrahydrocannabinol treatment on hippocampal extracellular acetylcholine concentration and alternation performance in the T-maze. Neuropharmacology 41 , 392–399. Navarrete, M., and Araque, A. (2008). Endocannabinoids mediate neuron-astrocyte communication. Neuron 57 , 883–893. Navarrete, M., and Araque, A. (2010). Endocannabinoids potentiate synaptic transmission through stimulation of astrocytes. Neuron 68 , 113–126. Nedergaard, M., Ransom, B., and Goldman, S.A. (2003). New roles for astrocytes: redefining the functional architecture of the brain. Trends Neurosci. 26 , 523–530. Neves, G., Cooke, S.F., and Bliss, T.V.P. (2008). Synaptic plasticity, memory and the hippocampus: a neural network approach to causality. Nat. Rev. Neurosci. 9, 65–75. Nicholls, R.E., Alarcon, J.M., Malleret, G., Carroll, R.C., Grody, M., Vronskaya, S., and Kandel, E.R. (2008). Transgenic mice lacking NMDAR-dependent LTD exhibit deficits in behavioral flexibility. Neuron 58 , 104–117. Nicholson, R.A., Liao, C., Zheng, J., David, L.S., Coyne, L., Errington, A.C., Singh, G., and Lees, G. (2003). Sodium channel inhibition by anandamide and synthetic cannabimimetics in brain. Brain Res. 978 , 194–204. Niehaus, J.L., Liu, Y., Wallis, K.T., Egertová, M., Bhartur, S.G., Mukhopadhyay, S., Shi, S., He, H., Selley, D.E., Howlett, A.C., et al. (2007). CB1 cannabinoid receptor activity is modulated by the cannabinoid receptor interacting protein CRIP 1a. Mol. Pharmacol. 72 , 1557–1566. Niyuhire, F., Varvel, S.A., Thorpe, A.J., Stokes, R.J., Wiley, J.L., and Lichtman, A.H. (2007). The disruptive effects of the CB1 receptor antagonist rimonabant on extinction learning in mice are task-specific. Psychopharmacology (Berl.) 191 , 223–231.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 152
O
O’Shaughnessy, R. (1839b). Case of Tetanus, Cured by a Preparation of Hemp (the Cannabis indica.). 8 462–469. O’Shaughnessy, R. (1839a). On the Preparations of the Indian Hemp, or Gunjah, Transactions of the Medical and Physical Society of Bengal. Transactions of the Medical and Physical Society of Bengal 8, 421–461. Oberheim, N.A., Takano, T., Han, X., He, W., Lin, J.H.C., Wang, F., Xu, Q., Wyatt, J.D., Pilcher, W., Ojemann, J.G., et al. (2009). Uniquely hominid features of adult human astrocytes. J. Neurosci. 29 , 3276–3287. Oberheim, N.A., Wang, X., Goldman, S., and Nedergaard, M. (2006). Astrocytic complexity distinguishes the human brain. Trends Neurosci. 29 , 547–553. Oddi, S., Fezza, F., Pasquariello, N., D’Agostino, A., Catanzaro, G., De Simone, C., Rapino, C., Finazzi-Agrò, A., and Maccarrone, M. (2009). Molecular identification of albumin and Hsp70 as cytosolic anandamide-binding proteins. Chem. Biol. 16 , 624–632. Ohman, A. (2000). Fear and anxiety: Evolutionary, cognitive, and clinical perspectives. In Handbook of Emotions, Second Edition, (New York: M. Lewis and J. M. Haviland-Jones), pp. 573– 593. Ohno-Shosaku, T., Shosaku, J., Tsubokawa, H., and Kano, M. (2002a). Cooperative endocannabinoid production by neuronal depolarization and group I metabotropic glutamate receptor activation. Eur. J. Neurosci. 15 , 953–961. Ohno-Shosaku, T., Tsubokawa, H., Mizushima, I., Yoneda, N., Zimmer, A., and Kano, M. (2002b). Presynaptic cannabinoid sensitivity is a major determinant of depolarization-induced retrograde suppression at hippocampal synapses. J. Neurosci. 22 , 3864–3872. Oliet, S.H.R., Baimoukhametova, D.V., Piet, R., and Bains, J.S. (2007). Retrograde regulation of GABA transmission by the tonic release of oxytocin and endocannabinoids governs postsynaptic firing. J. Neurosci. 27 , 1325–1333. de Oliveira Alvares, L., Pasqualini Genro, B., Diehl, F., Molina, V.A., and Quillfeldt, J.A. (2008). Opposite action of hippocampal CB1 receptors in memory reconsolidation and extinction. Neuroscience 154 , 1648–1655. Olton, D.S. (1979). Mazes, maps, and memory. Am Psychol 34 , 583–596. Orellana, J.A., Shoji, K.F., Abudara, V., Ezan, P., Amigou, E., Sáez, P.J., Jiang, J.X., Naus, C.C., Sáez, J.C., and Giaume, C. (2011). Amyloid β-induced death in neurons involves glial and neuronal hemichannels. J. Neurosci. 31 , 4962–4977. Oropeza, V.C., Mackie, K., and Van Bockstaele, E.J. (2007). Cannabinoid receptors are localized to noradrenergic axon terminals in the rat frontal cortex. Brain Res. 1127 , 36–44. Ortega-Gutiérrez, S., Molina-Holgado, E., and Guaza, C. (2005). Effect of anandamide uptake inhibition in the production of nitric oxide and in the release of cytokines in astrocyte cultures. Glia 52 , 163–168. P
Pagotto, U. (2005). The Emerging Role of the Endocannabinoid System in Endocrine Regulation and Energy Balance. Endocrine Reviews 27 , 73–100.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 153
Pagotto, U., Cervino, C., Vicennati, V., Marsicano, G., Lutz, B., and Pasquali, R. (2006). How many sites of action for endocannabinoids to control energy metabolism? Int J Obes (Lond) 30 Suppl 1, S39–43. Pamplona, F.A., Menezes-de-Lima, O., Jr, and Takahashi, R.N. (2010). Aspirin-triggered lipoxin induces CB1-dependent catalepsy in mice. Neurosci. Lett. 470 , 33–37. Pamplona, F.A., Prediger, R.D.S., Pandolfo, P., and Takahashi, R.N. (2006). The cannabinoid receptor agonist WIN 55,212-2 facilitates the extinction of contextual fear memory and spatial memory in rats. Psychopharmacology (Berl.) 188 , 641–649. Pan, B., Wang, W., Long, J.Z., Sun, D., Hillard, C.J., Cravatt, B.F., and Liu, Q. (2009). Blockade of 2-arachidonoylglycerol hydrolysis by selective monoacylglycerol lipase inhibitor 4-nitrophenyl 4- (dibenzo [d][1, 3] dioxol-5-yl (hydroxy) methyl) piperidine-1-carboxylate (JZL184) enhances retrograde endocannabinoid signaling. Journal of Pharmacology and Experimental Therapeutics 331 , 591. Pan, B., Wang, W., Zhong, P., Blankman, J.L., Cravatt, B.F., and Liu, Q. (2011). Alterations of endocannabinoid signaling, synaptic plasticity, learning, and memory in monoacylglycerol lipase knock-out mice. J. Neurosci. 31 , 13420–13430. Panatier, A., Theodosis, D.T., Mothet, J.-P., Touquet, B., Pollegioni, L., Poulain, D.A., and Oliet, S.H.R. (2006). Glia-derived D-serine controls NMDA receptor activity and synaptic memory. Cell 125 , 775–784. Panatier, A., Vallée, J., Haber, M., Murai, K.K., Lacaille, J.-C., and Robitaille, R. (2011). Astrocytes are endogenous regulators of basal transmission at central synapses. Cell 146 , 785–798. Pannasch, U., Vargova, L., Reingruber, J., Ezan, P., Holcman, D., Giaume, C., Sykova, E., and Rouach, N. (2011). Astroglial networks scale synaptic activity and plasticity. Proceedings of the National Academy of Sciences 108 , 8467–8472. Pascual, O., Casper, K.B., Kubera, C., Zhang, J., Revilla-Sanchez, R., Sul, J.-Y., Takano, H., Moss, S.J., McCarthy, K., and Haydon, P.G. (2005). Astrocytic purinergic signaling coordinates synaptic networks. Science 310 , 113–116. Di Pasquale, E., Chahinian, H., Sanchez, P., and Fantini, J. (2009). The insertion and transport of anandamide in synthetic lipid membranes are both cholesterol-dependent. PLoS ONE 4, e4989. Pasti, L., Volterra, A., Pozzan, T., and Carmignoto, G. (1997). Intracellular calcium oscillations in astrocytes: a highly plastic, bidirectional form of communication between neurons and astrocytes in situ. J. Neurosci. 17 , 7817–7830. Păunescu, H., Coman, O.A., Coman, L., Ghi ă , I., Georgescu, S.R., Dr ăghia, F., and Fulga, I. (2011). Cannabinoid system and cyclooxygenases inhibitors. J Med Life 4, 11–20. Pavlov, I.P. (1927). Conditioned reflexes (London). Pekny, M., and Pekna, M. (2004). Astrocyte intermediate filaments in CNS pathologies and regeneration. J. Pathol. 204 , 428–437. Pellerin, L. (2005). How astrocytes feed hungry neurons. Mol. Neurobiol. 32 , 59–72. Pellerin, L., and Magistretti, P.J. (1994). Glutamate uptake into astrocytes stimulates aerobic glycolysis: a mechanism coupling neuronal activity to glucose utilization. Proc. Natl. Acad. Sci. U.S.A. 91 , 10625–10629. Perea, G., and Araque, A. (2005). Glial calcium signaling and neuron-glia communication. Cell Calcium 38 , 375–382. Perea, G., Navarrete, M., and Araque, A. (2009). Tripartite synapses: astrocytes process and control synaptic information. Trends Neurosci. 32 , 421–431. Pertwee, R.G. (1997). Pharmacology of cannabinoid CB1 and CB2 receptors. Pharmacol. Ther. 74 , 129–180. Phelps, E.A. (2004). Human emotion and memory: interactions of the amygdala and hippocampal complex. Curr. Opin. Neurobiol. 14 , 198–202. Phillips, R.G., and LeDoux, J.E. (1992). Differential contribution of amygdala and hippocampus to cued and contextual fear conditioning. Behav. Neurosci. 106 , 274–285.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 154
Piomelli, D. (2003). The molecular logic of endocannabinoid signalling. Nat. Rev. Neurosci. 4, 873–884. Pitcher, J.A., Freedman, N.J., and Lefkowitz, R.J. (1998). G protein-coupled receptor kinases. Annu. Rev. Biochem. 67 , 653–692. Pitler, T.A., and Alger, B.E. (1992). Postsynaptic spike firing reduces synaptic GABAA responses in hippocampal pyramidal cells. J. Neurosci. 12 , 4122–4132. Platel, A., and Porsolt, R.D. (1982). Habituation of exploratory activity in mice: a screening test for memory enhancing drugs. Psychopharmacology (Berl.) 78 , 346–352. Plendl, W., and Wotjak, C.T. (2010). Dissociation of within- and between-session extinction of conditioned fear. J. Neurosci. 30 , 4990–4998. Pope, H.G., Jr, Gruber, A.J., Hudson, J.I., Huestis, M.A., and Yurgelun-Todd, D. (2001). Neuropsychological performance in long-term cannabis users. Arch. Gen. Psychiatry 58 , 909–915. Price, J.L. (2003). Comparative aspects of amygdala connectivity. Ann. N. Y. Acad. Sci. 985 , 50–58. Puighermanal, E., Marsicano, G., Busquets-Garcia, A., Lutz, B., Maldonado, R., and Ozaita, A. (2009). Cannabinoid modulation of hippocampal long-term memory is mediated by mTOR signaling. Nat. Neurosci. 12 , 1152–1158. Q
Quarta, C., Bellocchio, L., Mancini, G., Mazza, R., Cervino, C., Braulke, L.J., Fekete, C., Latorre, R., Nanni, C., Bucci, M., et al. (2010). CB(1) signaling in forebrain and sympathetic neurons is a key determinant of endocannabinoid actions on energy balance. Cell Metab. 11 , 273–285. Quirk, G.J., Likhtik, E., Pelletier, J.G., and Paré, D. (2003). Stimulation of medial prefrontal cortex decreases the responsiveness of central amygdala output neurons. J. Neurosci. 23 , 8800– 8807. Quirk, G.J., and Mueller, D. (2007). Neural mechanisms of extinction learning and retrieval. Neuropsychopharmacology 33 , 56–72. Quirk, G.J., and Mueller, D. (2008). Neural mechanisms of extinction learning and retrieval. Neuropsychopharmacology 33 , 56–72. Quirk, G.J., Russo, G.K., Barron, J.L., and Lebron, K. (2000). The role of ventromedial prefrontal cortex in the recovery of extinguished fear. J. Neurosci. 20 , 6225–6231. R
Ramírez, B.G., Blázquez, C., Gómez del Pulgar, T., Guzmán, M., and de Ceballos, M.L. (2005). Prevention of Alzheimer’s disease pathology by cannabinoids: neuroprotection mediated by blockade of microglial activation. J. Neurosci. 25 , 1904–1913. Ranganathan, M., and D’Souza, D.C. (2006). The acute effects of cannabinoids on memory in humans: a review. Psychopharmacology (Berl.) 188 , 425–444. Ransom, B., Behar, T., and Nedergaard, M. (2003). New roles for astrocytes (stars at last). Trends Neurosci. 26 , 520–522. Razdan, R.K. (1986). Structure-activity relationships in cannabinoids. Pharmacol. Rev. 38 , 75– 149.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 155
Rescorla, R.A. (1968). Probability of shock in the presence and absence of CS in fear conditioning. J Comp Physiol Psychol 66 , 1–5. Resstel, L.B.M., Lisboa, S.F., Aguiar, D.C., Corrêa, F.M.A., and Guimarães, F.S. (2008). Activation of CB1 cannabinoid receptors in the dorsolateral periaqueductal gray reduces the expression of contextual fear conditioning in rats. Psychopharmacology (Berl.) 198 , 405–411. Rinaldi-Carmona, M., Barth, F., Héaulme, M., Shire, D., Calandra, B., Congy, C., Martinez, S., Maruani, J., Néliat, G., and Caput, D. (1994). SR141716A, a potent and selective antagonist of the brain cannabinoid receptor. FEBS Lett. 350 , 240–244. Del Río Hortega, P. (1918). Noticia de un nuevo y fácil método para la coloración de la neuroglia y el tejido conjuntivo. Trabajos Del Laboratorio De Investigaciones Biológicas 15 , 367–378. Robbe, D., and Buzsáki, G. (2009). Alteration of theta timescale dynamics of hippocampal place cells by a cannabinoid is associated with memory impairment. J. Neurosci. 29 , 12597–12605. Robbe, D., Kopf, M., Remaury, A., Bockaert, J., and Manzoni, O.J. (2002). Endogenous cannabinoids mediate long-term synaptic depression in the nucleus accumbens. Proc. Natl. Acad. Sci. U.S.A. 99 , 8384–8388. Robbe, D., Montgomery, S.M., Thome, A., Rueda-Orozco, P.E., McNaughton, B.L., and Buzsaki, G. (2006). Cannabinoids reveal importance of spike timing coordination in hippocampal function. Nature Neuroscience 9, 1526–1533. Rödner Sznitman, S., Olsson, B., Room, R., and European Monitoring Centre for Drugs and Drug Addiction (2008). A cannabis reader : global issues and local experiences : perspectives on cannabis controversies, treatment and regulation in Europe (Luxembourg: Office for Official Publications of the European Communities, c2008.). Rodriguez, J.J., Mackie, K., and Pickel, V.M. (2001). Ultrastructural localization of the CB1 cannabinoid receptor in mu-opioid receptor patches of the rat Caudate putamen nucleus. J. Neurosci. 21 , 823–833. Rogan, M.T., and LeDoux, J.E. (1995). LTP is accompanied by commensurate enhancement of auditory-evoked responses in a fear conditioning circuit. Neuron 15 , 127–136. Rogan, M.T., Stäubli, U.V., and LeDoux, J.E. (1997). Fear conditioning induces associative long-term potentiation in the amygdala. Nature 390 , 604–607. Romero, J., Hillard, C.J., Calero, M., and Rábano, A. (2002). Fatty acid amide hydrolase localization in the human central nervous system: an immunohistochemical study. Brain Res. Mol. Brain Res. 100 , 85–93. Ross, S.E., Greenberg, M.E., and Stiles, C.D. (2003). Basic helix-loop-helix factors in cortical development. Neuron 39 , 13–25. Rouach, N., Koulakoff, A., Abudara, V., Willecke, K., and Giaume, C. (2008). Astroglial metabolic networks sustain hippocampal synaptic transmission. Science 322 , 1551–1555. Rowitch, D.H., and Kriegstein, A.R. (2010). Developmental genetics of vertebrate glial-cell specification. Nature 468 , 214–222. Rozenfeld, R., Bushlin, I., Gomes, I., Tzavaras, N., Gupta, A., Neves, S., Battini, L., Gusella, G.L., Lachmann, A., Ma’ayan, A., et al. (2012). Receptor heteromerization expands the repertoire of cannabinoid signaling in rodent neurons. PLoS ONE 7, e29239. Rubino, T., Forlani, G., Viganò, D., Zippel, R., and Parolaro, D. (2004). Modulation of extracellular signal-regulated kinases cascade by chronic delta 9-tetrahydrocannabinol treatment. Mol. Cell. Neurosci. 25 , 355–362. Russo, E.B. (2007). History of cannabis and its preparations in saga, science, and sobriquet. Chemistry & Biodiversity 4, 1614–1648. Ryberg, E., Vu, H.K., Larsson, N., Groblewski, T., Hjorth, S., Elebring, T., Sjögren, S., and Greasley, P.J. (2005). Identification and characterisation of a novel splice variant of the human CB1 receptor. FEBS Lett. 579 , 259–264.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 156
S
Safo, P.K., and Regehr, W.G. (2005). Endocannabinoids control the induction of cerebellar LTD. Neuron 48 , 647–659. Sagan, S., Venance, L., Torrens, Y., Cordier, J., Glowinski, J., and Giaume, C. (1999). Anandamide and WIN 55212-2 inhibit cyclic AMP formation through G-protein-coupled receptors distinct from CB1 cannabinoid receptors in cultured astrocytes. Eur. J. Neurosci. 11 , 691–699. Sánchez, A.J., and García-Merino, A. (2012). Neuroprotective agents: cannabinoids. Clin. Immunol. 142 , 57–67. Sánchez, C., Galve-Roperh, I., Rueda, D., and Guzmán, M. (1998). Involvement of sphingomyelin hydrolysis and the mitogen-activated protein kinase cascade in the Delta9- tetrahydrocannabinol-induced stimulation of glucose metabolism in primary astrocytes. Mol. Pharmacol. 54 , 834–843. Sánchez, C., Rueda, D., Ségui, B., Galve-Roperh, I., Levade, T., and Guzmán, M. (2001). The CB(1) cannabinoid receptor of astrocytes is coupled to sphingomyelin hydrolysis through the adaptor protein fan. Mol. Pharmacol. 59 , 955–959. Sarter, M., and Markowitsch, H.J. (1985). Involvement of the amygdala in learning and memory: a critical review, with emphasis on anatomical relations. Behav. Neurosci. 99 , 342–380. Sauer, B., and Henderson, N. (1989). Cre-stimulated recombination at loxP-containing DNA sequences placed into the mammalian genome. Nucleic Acids Res. 17 , 147–161. Schiller, D., and Delgado, M.R. (2010). Overlapping neural systems mediating extinction, reversal and regulation of fear. Trends Cogn. Sci. (Regul. Ed.) 14 , 268–276. Schulze, G.E., McMillan, D.E., Bailey, J.R., Scallet, A., Ali, S.F., Slikker, W., Jr, and Paule, M.G. (1988). Acute effects of delta-9-tetrahydrocannabinol in rhesus monkeys as measured by performance in a battery of complex operant tests. J. Pharmacol. Exp. Ther. 245 , 178–186. Seifert, G., and Steinhäuser, C. (2011). Neuron-astrocyte signaling and epilepsy. Exp. Neurol. Selley, D.E., Stark, S., Sim, L.J., and Childers, S.R. (1996). Cannabinoid receptor stimulation of guanosine-5’-O-(3-[35S]thio)triphosphate binding in rat brain membranes. Life Sci. 59 , 659–668. Sheng, W.S., Hu, S., Min, X., Cabral, G.A., Lokensgard, J.R., and Peterson, P.K. (2005). Synthetic cannabinoid WIN55,212-2 inhibits generation of inflammatory mediators by IL-1beta- stimulated human astrocytes. Glia 49 , 211–219. Sherrington, C.S. (1906). The Integrative Action of the Nervous System (New Haven). Shim, J.-Y., Rudd, J., and Ding, T.T. (2011). Distinct second extracellular loop structures of the brain cannabinoid CB(1) receptor: implication in ligand binding and receptor function. Proteins 79 , 581–597. Shire, D., Calandra, B., Rinaldi-Carmona, M., Oustric, D., Pessègue, B., Bonnin-Cabanne, O., Le Fur, G., Caput, D., and Ferrara, P. (1996). Molecular cloning, expression and function of the murine CB2 peripheral cannabinoid receptor. Biochim. Biophys. Acta 1307 , 132–136. Shire, D., Carillon, C., Kaghad, M., Calandra, B., Rinaldi-Carmona, M., Le Fur, G., Caput, D., and Ferrara, P. (1995). An amino-terminal variant of the central cannabinoid receptor resulting from alternative splicing. J. Biol. Chem. 270 , 3726–3731. Shivachar, A.C. (2007). Cannabinoids inhibit sodium-dependent, high-affinity excitatory amino acid transport in cultured rat cortical astrocytes. Biochem. Pharmacol. 73 , 2004–2011. SIDMAN, M. (1962). Classical avoidance without a warning stimulus. J Exp Anal Behav 5, 97– 104. Sim-Selley, L.J. (2003). Regulation of cannabinoid CB1 receptors in the central nervous system by chronic cannabinoids. Crit Rev Neurobiol 15 , 91–119.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 157
Singla, S., Kreitzer, A.C., and Malenka, R.C. (2007). Mechanisms for synapse specificity during striatal long-term depression. J. Neurosci. 27 , 5260–5264. Sjöström, P.J., Turrigiano, G.G., and Nelson, S.B. (2001). Rate, timing, and cooperativity jointly determine cortical synaptic plasticity. Neuron 32 , 1149–1164. Sjöström, P.J., Turrigiano, G.G., and Nelson, S.B. (2003). Neocortical LTD via coincident activation of presynaptic NMDA and cannabinoid receptors. Neuron 39 , 641–654. Slangen, J.L., Earley, B., Jaffard, R., Richelle, M., and Olton, D.S. (1990). Behavioral models of memory and amnesia. Pharmacopsychiatry 23 Suppl 2 , 81–83; discussion 84. Smith, T.H., Sim-Selley, L.J., and Selley, D.E. (2010). Cannabinoid CB1 receptor-interacting proteins: novel targets for central nervous system drug discovery? Br. J. Pharmacol. 160 , 454–466. Sotres-Bayon, F., Bush, D.E.A., and LeDoux, J.E. (2007). Acquisition of fear extinction requires activation of NR2B-containing NMDA receptors in the lateral amygdala. Neuropsychopharmacology 32 , 1929–1940. Sotres-Bayon, F., and Quirk, G.J. (2010). Prefrontal control of fear: more than just extinction. Curr. Opin. Neurobiol. 20 , 231–235. Squire, L.R. (2004). Memory systems of the brain: a brief history and current perspective. Neurobiology of Learning and Memory 82 , 171–177. Squire, L.R., and Zola-Morgan, S. (1988). Memory: brain systems and behavior. Trends Neurosci. 11 , 170–175. Starowicz, K., Nigam, S., and Di Marzo, V. (2007). Biochemistry and pharmacology of endovanilloids. Pharmacol. Ther. 114 , 13–33. Stavrou, S., Feng, Z., Lemon, S.M., and Roos, R.P. (2010). Different strains of Theiler’s murine encephalomyelitis virus antagonize different sites in the type I interferon pathway. J. Virol. 84 , 9181– 9189. Steimer, T., and Driscoll, P. (2003). Divergent stress responses and coping styles in psychogenetically selected Roman high-(RHA) and low-(RLA) avoidance rats: behavioural, neuroendocrine and developmental aspects. Stress 6, 87–100. Steimer, T., la Fleur, S., and Schulz, P.E. (1997). Neuroendocrine correlates of emotional reactivity and coping in male rats from the Roman high (RHA/Verh)- and low (RLA/Verh)-avoidance lines. Behav. Genet. 27 , 503–512. Stella, N. (2010). Cannabinoid and cannabinoid-like receptors in microglia, astrocytes, and astrocytomas. Glia 58 , 1017–1030. Stella, N., and Piomelli, D. (2001). Receptor-dependent formation of endogenous cannabinoids in cortical neurons. Eur. J. Pharmacol. 425 , 189–196. Stella, N., Schweitzer, P., and Piomelli, D. (1997). A second endogenous cannabinoid that modulates long-term potentiation. Nature 388 , 773–778. Straiker, A., and Mackie, K. (2007). Metabotropic suppression of excitation in murine autaptic hippocampal neurons. J. Physiol. (Lond.) 578 , 773–785. Stühmer, T., Puelles, L., Ekker, M., and Rubenstein, J.L.R. (2002). Expression from a Dlx gene enhancer marks adult mouse cortical GABAergic neurons. Cereb. Cortex 12 , 75–85. Suárez, J., Romero-Zerbo, S.Y., Rivera, P., Bermúdez-Silva, F.J., Pérez, J., De Fonseca, F.R., and Fernández-Llebrez, P. (2010). Endocannabinoid system in the adult rat circumventricular areas: an immunohistochemical study. J. Comp. Neurol. 518 , 3065–3085. Sugiura, T., Kondo, S., Sukagawa, A., Nakane, S., Shinoda, A., Itoh, K., Yamashita, A., and Waku, K. (1995). 2-Arachidonoylglycerol: a possible endogenous cannabinoid receptor ligand in brain. Biochem. Biophys. Res. Commun. 215 , 89–97. Suzuki, A., Josselyn, S.A., Frankland, P.W., Masushige, S., Silva, A.J., and Kida, S. (2004). Memory reconsolidation and extinction have distinct temporal and biochemical signatures. J. Neurosci. 24 , 4787–4795.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 158
Suzuki, A., Stern, S.A., Bozdagi, O., Huntley, G.W., Walker, R.H., Magistretti, P.J., and Alberini, C.M. (2011). Astrocyte-neuron lactate transport is required for long-term memory formation. Cell 144 , 810–823. T
Tabernero, A., Medina, J.M., and Giaume, C. (2006). Glucose metabolism and proliferation in glia: role of astrocytic gap junctions. J. Neurochem. 99 , 1049–1061. Takahashi, K.A., and Linden, D.J. (2000). Cannabinoid receptor modulation of synapses received by cerebellar Purkinje cells. J. Neurophysiol. 83 , 1167–1180. Takata, N., Mishima, T., Hisatsune, C., Nagai, T., Ebisui, E., Mikoshiba, K., and Hirase, H. (2011). Astrocyte calcium signaling transforms cholinergic modulation to cortical plasticity in vivo. J. Neurosci. 31 , 18155–18165. Takuma, K., Baba, A., and Matsuda, T. (2004). Astrocyte apoptosis: implications for neuroprotection. Prog. Neurobiol. 72 , 111–127. Tan, H., Lauzon, N.M., Bishop, S.F., Bechard, M.A., and Laviolette, S.R. (2010). Integrated cannabinoid CB1 receptor transmission within the amygdala-prefrontal cortical pathway modulates neuronal plasticity and emotional memory encoding. Cereb. Cortex 20 , 1486–1496. Tan, H., Lauzon, N.M., Bishop, S.F., Chi, N., Bechard, M., and Laviolette, S.R. (2011). Cannabinoid transmission in the basolateral amygdala modulates fear memory formation via functional inputs to the prelimbic cortex. J. Neurosci. 31 , 5300–5312. Tanaka, K., Watase, K., Manabe, T., Yamada, K., Watanabe, M., Takahashi, K., Iwama, H., Nishikawa, T., Ichihara, N., Kikuchi, T., et al. (1997). Epilepsy and exacerbation of brain injury in mice lacking the glutamate transporter GLT-1. Science 276 , 1699–1702. Tanimura, A., Yamazaki, M., Hashimotodani, Y., Uchigashima, M., Kawata, S., Abe, M., Kita, Y., Hashimoto, K., Shimizu, T., Watanabe, M., et al. (2010). The endocannabinoid 2- arachidonoylglycerol produced by diacylglycerol lipase alpha mediates retrograde suppression of synaptic transmission. Neuron 65 , 320–327. Tekkök, S.B., Brown, A.M., Westenbroek, R., Pellerin, L., and Ransom, B.R. (2005). Transfer of glycogen-derived lactate from astrocytes to axons via specific monocarboxylate transporters supports mouse optic nerve activity. J. Neurosci. Res. 81 , 644–652. Terranova, J.P., Storme, J.J., Lafon, N., Pé ŕio, A., Rinaldi-Carmona, M., Le Fur, G., and Soubrié, P. (1996). Improvement of memory in rodents by the selective CB1 cannabinoid receptor antagonist, SR 141716. Psychopharmacology (Berl.) 126 , 165–172. Terzian, A.L., Drago, F., Wotjak, C.T., and Micale, V. (2011). The Dopamine and Cannabinoid Interaction in the Modulation of Emotions and Cognition: Assessing the Role of Cannabinoid CB1 Receptor in Neurons Expressing Dopamine D1 Receptors. Front Behav Neurosci 5, 49. Theodosis, D.T., and Poulain, D.A. (1993). Activity-dependent neuronal-glial and synaptic plasticity in the adult mammalian hypothalamus. Neuroscience 57 , 501–535. Tinklenberg, J.R., Melges, F.T., Hollister, L.E., and Gillespie, H.K. (1970). Marijuana and immediate memory. Nature 226 , 1171–1172. Tononi, G., and Cirelli, C. (2006). Sleep function and synaptic homeostasis. Sleep Med Rev 10 , 49–62. Touriño, C., Zimmer, A., and Valverde, O. (2010). THC Prevents MDMA Neurotoxicity in Mice. PLoS ONE 5, e9143. Trifilieff, P., Herry, C., Vanhoutte, P., Caboche, J., Desmedt, A., Riedel, G., Mons, N., and Micheau, J. (2006). Foreground contextual fear memory consolidation requires two independent phases of hippocampal ERK/CREB activation. Learn. Mem. 13 , 349–358.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 159
Tsou, K., Brown, S., Sañudo-Peña, M.C., Mackie, K., and Walker, J.M. (1998). Immunohistochemical distribution of cannabinoid CB1 receptors in the rat central nervous system. Neuroscience 83 , 393–411. U
Uchigashima, M., Narushima, M., Fukaya, M., Katona, I., Kano, M., and Watanabe, M. (2007). Subcellular arrangement of molecules for 2-arachidonoyl-glycerol-mediated retrograde signaling and its physiological contribution to synaptic modulation in the striatum. J. Neurosci. 27 , 3663–3676. Uchigashima, M., Yamazaki, M., Yamasaki, M., Tanimura, A., Sakimura, K., Kano, M., and Watanabe, M. (2011). Molecular and morphological configuration for 2-arachidonoylglycerol-mediated retrograde signaling at mossy cell-granule cell synapses in the dentate gyrus. J. Neurosci. 31 , 7700– 7714. Ursin, H., and Olff, M. (1993). Psychobiology of coping and defence strategies. Neuropsychobiology 28 , 66–71. V
Vadhan, N.P., Serper, M.R., and Haney, M. (2009). Effects of ∆-THC on Working Memory: Implications for Schizophrenia? Prim Psychiatry 16 , 51–99. Valjent, E., Caboche, J., and Vanhoutte, P. (2001). Mitogen-activated protein kinase/extracellular signal-regulated kinase induced gene regulation in brain: a molecular substrate for learning and memory? Mol. Neurobiol. 23 , 83–99. Varma, N., Carlson, G.C., Ledent, C., and Alger, B.E. (2001). Metabotropic glutamate receptors drive the endocannabinoid system in hippocampus. J. Neurosci. 21 , RC188. Varvel, S.A., Anum, E., Niyuhire, F., Wise, L.E., and Lichtman, A.H. (2005a). Delta(9)-THC- induced cognitive deficits in mice are reversed by the GABA(A) antagonist bicuculline. Psychopharmacology (Berl.) 178 , 317–327. Varvel, S.A., Anum, E.A., and Lichtman, A.H. (2005b). Disruption of CB(1) receptor signaling impairs extinction of spatial memory in mice. Psychopharmacology (Berl.) 179 , 863–872. Varvel, S.A., Hamm, R.J., Martin, B.R., and Lichtman, A.H. (2001). Differential effects of delta 9- THC on spatial reference and working memory in mice. Psychopharmacology (Berl.) 157 , 142–150. Varvel, S.A., and Lichtman, A.H. (2002). Evaluation of CB1 receptor knockout mice in the Morris water maze. J. Pharmacol. Exp. Ther. 301 , 915–924. Verkhratsky, A., Parpura, V., and Rodríguez, J.J. (2011). Where the thoughts dwell: the physiology of neuronal-glial “diffuse neural net.”Brain Res Rev 66 , 133–151. Vicens-Costa, E., Martinez-Membrives, E., López-Aumatell, R., Guitart-Masip, M., Canete, T., Blázquez, G., Tobena, A., and Fernández-Teruel, A. (2010). Two-way avoidance acquisition is negatively related to conditioned freezing and positively associated with startle reactions: A dissection of anxiety and fear in genetically heterogeneous rats. Physiology & Behavior. Vicens-Costa, E., Martínez-Membrives, E., López-Aumatell, R., Guitart-Masip, M., Cañete, T., Blázquez, G., Tobeña, A., and Fernández-Teruel, A. (2011). Two-way avoidance acquisition is negatively related to conditioned freezing and positively associated with startle reactions: a dissection of anxiety and fear in genetically heterogeneous rats. Physiol. Behav. 103 , 148–156. Villares, J. (2007). Chronic use of marijuana decreases cannabinoid receptor binding and mRNA expression in the human brain. Neuroscience 145 , 323–334.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 160
Virchow, R. (1858). Die Cellularpathologie in Ihrer Begründung auf Physiologische and Pathologische Gewebelehre. In Zwanzig Vorlesungen Gehalten Während Der Monate Februar, März Und April 1858 Im Pathologischen Institut Zu Berlin, 1st Edition, (Berlin), p. 440. Vischer, H.F., Watts, A.O., Nijmeijer, S., and Leurs, R. (2011). G protein-coupled receptors: walking hand-in-hand, talking hand-in-hand? Br. J. Pharmacol. 163 , 246–260. Volkow, N.D., Gillespie, H., Mullani, N., Tancredi, L., Grant, C., Valentine, A., and Hollister, L. (1996). Brain glucose metabolism in chronic marijuana users at baseline and during marijuana intoxication. Psychiatry Res 67 , 29–38. Volterra, A., and Meldolesi, J. (2005). Astrocytes, from brain glue to communication elements: the revolution continues. Nat. Rev. Neurosci. 6, 626–640. W
Wada, J.A., Sato, M., and Corcoran, M.E. (1973). Antiepileptic properties of 9 - tetrahydrocannabinol. Exp. Neurol. 39 , 157–165. Walker, E.A. (2010). Animal models. Adv. Exp. Med. Biol. 678 , 138–146. Wallraff, A., Köhling, R., Heinemann, U., Theis, M., Willecke, K., and Steinhäuser, C. (2006). The impact of astrocytic gap junctional coupling on potassium buffering in the hippocampus. J. Neurosci. 26 , 5438–5447. Walter, L. (2002). Astrocytes in Culture Produce Anandamide and Other Acylethanolamides. Journal of Biological Chemistry 277 , 20869–20876. Walter, L., and Stella, N. (2003). Endothelin-1 increases 2-arachidonoyl glycerol (2-AG) production in astrocytes. Glia 44 , 85–90. Walter, L., and Stella, N. (2004). Cannabinoids and neuroinflammation. Br. J. Pharmacol. 141 , 775–785. Walz, W. (2000). Role of astrocytes in the clearance of excess extracellular potassium. Neurochem. Int. 36 , 291–300. Wang, X., Lou, N., Xu, Q., Tian, G.-F., Peng, W.G., Han, X., Kang, J., Takano, T., and Nedergaard, M. (2006). Astrocytic Ca2+ signaling evoked by sensory stimulation in vivo. Nat. Neurosci. 9, 816–823. Weber, T., Böhm, G., Hermann, E., Schütz, G., Schönig, K., and Bartsch, D. (2009). Inducible gene manipulations in serotonergic neurons. Front Mol Neurosci 2, 24. Wechsler, B. (1995). Coping and coping strategies: a behavioural view. Applied Animal Behaviour Science 43 , 123–134. Weiss, J.M. (1968). Effects of coping responses on stress. J Comp Physiol Psychol 65 , 251– 260. Wenger, T., Moldrich, G., and Furst, S. (2003). Neuromorphological background of cannabis addiction. Brain Res. Bull. 61 , 125–128. Whyte, D.A., Al-Hammadi, S., Balhaj, G., Brown, O.M., Penefsky, H.S., and Souid, A.-K. (2010). Cannabinoids Inhibit Cellular Respiration of Human Oral Cancer Cells. Pharmacology 85 , 328–335. Wiegert, J.S., and Bading, H. (2011). Activity-dependent calcium signaling and ERK-MAP kinases in neurons: a link to structural plasticity of the nucleus and gene transcription regulation. Cell Calcium 49 , 296–305. Wilson, R.I., Kunos, G., and Nicoll, R.A. (2001). Presynaptic specificity of endocannabinoid signaling in the hippocampus. Neuron 31 , 453–462. Wilson, R.I., and Nicoll, R.A. (2001). Endogenous cannabinoids mediate retrograde signalling at hippocampal synapses. Nature 410 , 588–592.
Mathilde METNA – LAURENT – Doctoral Thesis – University of BORDEAUX 2 161
Wilson, W.H., Ellinwood, E.H., Mathew, R.J., and Johnson, K. (1994). Effects of marijuana on performance of a computerized cognitive-neuromotor test battery. Psychiatry Res 51 , 115–125. Wise, L.E., Iredale, P.A., Stokes, R.J., and Lichtman, A.H. (2007). Combination of rimonabant and donepezil prolongs spatial memory duration. Neuropsychopharmacology 32 , 1805–1812. Wise, L.E., Thorpe, A.J., and Lichtman, A.H. (2009). Hippocampal CB(1) receptors mediate the memory impairing effects of Delta(9)-tetrahydrocannabinol. Neuropsychopharmacology 34 , 2072– 2080. Wittmann, G., Deli, L., Kalló, I., Hrabovszky, E., Watanabe, M., Liposits, Z., and Fekete, C. (2007). Distribution of type 1 cannabinoid receptor (CB1)-immunoreactive axons in the mouse hypothalamus. J. Comp. Neurol. 503 , 270–279. Wolff, M.C., and Leander, J.D. (2003). SR141716A, a cannabinoid CB1 receptor antagonist, improves memory in a delayed radial maze task. Eur. J. Pharmacol. 477 , 213–217. Wong, T.P., Howland, J.G., Robillard, J.M., Ge, Y., Yu, W., Titterness, A.K., Brebner, K., Liu, L., Weinberg, J., Christie, B.R., et al. (2007). Hippocampal long-term depression mediates acute stress- induced spatial memory retrieval impairment. Proc. Natl. Acad. Sci. U.S.A. 104 , 11471–11476. Y
Yang, C.-H., Huang, C.-C., and Hsu, K.-S. (2005). Behavioral stress enhances hippocampal CA1 long-term depression through the blockade of the glutamate uptake. J. Neurosci. 25 , 4288–4293. Yao, B., and Mackie, K. (2009). Endocannabinoid receptor pharmacology. Curr Top Behav Neurosci 1, 37–63. Yi, C.-X., Habegger, K.M., Chowen, J.A., Stern, J., and Tschöp, M.H. (2011). A role for astrocytes in the central control of metabolism. Neuroendocrinology 93 , 143–149. Yoon, T., Okada, J., Jung, M.W., and Kim, J.J. (2008). Prefrontal cortex and hippocampus subserve different components of working memory in rats. Learn. Mem. 15 , 97–105. Z
Zajicek, J.P., and Apostu, V.I. (2011). Role of cannabinoids in multiple sclerosis. CNS Drugs 25 , 187–201. Zeng, H., Chattarji, S., Barbarosie, M., Rondi-Reig, L., Philpot, B.D., Miyakawa, T., Bear, M.F., and Tonegawa, S. (2001). Forebrain-specific calcineurin knockout selectively impairs bidirectional synaptic plasticity and working/episodic-like memory. Cell 107 , 617–629. Zhang, Z., Gong, N., Wang, W., Xu, L., and Xu, T.L. (2008). Bell-shaped D-serine actions on hippocampal long-term depression and spatial memory retrieval. Cerebral Cortex 18 , 2391. Zilberter, Y., Zilberter, T., and Bregestovski, P. (2010). Neuronal activity in vitro and the in vivo reality: the role of energy homeostasis. Trends Pharmacol. Sci. 31 , 394–401. Zimmer, A., Zimmer, A.M., Hohmann, A.G., Herkenham, M., and Bonner, T.I. (1999). Increased mortality, hypoactivity, and hypoalgesia in cannabinoid CB1 receptor knockout mice. Proc. Natl. Acad. Sci. U.S.A. 96 , 5780–5785.
Mathilde METNA-LAURENT – Doctoral Thesis – University of BORDEAUX 2 162