Imaging of Camp/PKA Dynamics Induced by Catecholamines in the Neocortical Network Shinobu Nomura

Imaging of cAMP/PKA dynamics induced by catecholamines in the neocortical network Shinobu Nomura

To cite this version:

Shinobu Nomura. Imaging of cAMP/PKA dynamics induced by catecholamines in the neocortical network. Neurons and Cognition [q-bio.NC]. Université Pierre et Marie Curie - Paris VI, 2014. English. NNT : 2014PA066440. tel-01374733

HAL Id: tel-01374733 https://tel.archives-ouvertes.fr/tel-01374733 Submitted on 1 Oct 2016

HAL is a multi-disciplinary open access L’archive ouverte pluridisciplinaire HAL, est archive for the deposit and dissemination of sci- destinée au dépôt et à la diffusion de documents entific research documents, whether they are pub- scientifiques de niveau recherche, publiés ou non, lished or not. The documents may come from émanant des établissements d’enseignement et de teaching and research institutions in France or recherche français ou étrangers, des laboratoires abroad, or from public or private research centers. publics ou privés.

Thèse de doctorat de l’Université Pierre et Marie Curie, Paris 6

Spécialité : Neurosciences Ecole doctorale Cerveau-Cognition-Comportement (ED3C)

Présentée par NOMURA Shinobu

Pour obtenir le grade de Docteur de l’Université Paris 6

Sujet de la thèse : Imaging of cAMP/PKA dynamics induced by catecholamines in the neocortical network

Soutenue le 26 septembre 2014

Devant le jury composé de :

Dr Denis Hervé, Président

Dr Emmanuel Valjent, Rapporteur

Pr Philippe De Deurwaerdère, Rapporteur

Dr Régine Hepp, Directeur de thèse

Invités exceptionnels

Dr Nicolas Gervasi

Dr Thierry Gallopin

Dr. Bertrand Lambolez

Remerciements

Je voudrais tout d'abord remercier mon directeur de thèse, Bertrand Lambolez, pour m’avoira ccueilli pour mon stage de M2 puis de m’avoir proposé ce projet sur la transmission catécholaminergique pour poursuivre une thèse à l’UPMC. Je le remercie pour son encadrement et pour ses conseils avisés mais également pour ses encouragements. Grâce à Bertrand, j’ai eu la chance de suivre une formation de qualité et de profiter pleinement de la vie parisienne. Cette expérience a été déterminante pour mon avenir professionnel qui se poursuivra désormais à Okinawa.

Je souhaite également remercier ma co-directrice de thèse Régine Hepp, pour avoir accepté de co-diriger ce travail, ainsi que pour ses remarques et son implication dans la rédaction de ce manuscrit. Je lui suis reconnaissante pour sa patience et sa générosité dans le travail qui ontpermis de surmonter les difficultés entre autres d’ordre linguistique.

Merci à Bruno Cauli pour son aide, ses conseils, son très grand humour et surtout son filtre infra rouge !

Je remercie pour leur accueil dans leur unités respectives Danièle Tritsch et Jean Mariani, directeurs de l’unité UMR7102-NPA et Hervé Chneiweiss, directeur de l’UMR 8246-NPS.

Je tiens à exprimer ma vive reconnaissance aux membres du jury Denis Hervé, Philippe De Deurwaerdère et Emmanuel Valjent qui ont accepté de juger mon travail et de faire ledéplacement pour la soutenance.

Je remercie également Thierry Gallopin et Nicolas Gervasi pour leur présence à ma soutenance.

Je voudrais remercier The Nakajima Foundation, pour son soutien financier pendant quatre ans. Cette aide m’a permis d’entreprendre mes études de manière sereine.

J’adresse mes remerciements les plus sincères à Elvire Guiot et Ludovic Tricoire pour leur investissement à me former à la microscopie biphotonique et à l’optogénétique. Leur aide a été cruciale notamment quand cela ne fonctionnait pas.

Merci à Ivan Cohen pour nous avoir aidé à donner une seconde vie à la photostimulation Merci à tous les autres membres de l’équipe pour leur gentillesse, pour m’avoir appris les gros mots en français et de m’avoir fait des blagues que je ne comprends toujours pas. Merci à Agnès, Audrey, Juliette Lim-Anna, Najate, Maud, Alexandre, Antoine, Matthieu, Xavier, Annastasios, Fabrice et puis ceux que j’oublie mais ce n’est pas exprès.

Je remercie tous les membres des équipes du 5ème étage et du 6ème, pour le prêt de matériel, pour leur aide et leurs conseils mais surtout pour les moments de convivialité partagés au sein du laboratoire.

Je voudrais aussi remercier tous les personnels des services généraux du laboratoire et le personnel de l’animalerie pour leur implication dans la vie quotidienne du laboratoire et le bien être des animaux. Le travail n’en a été que plus agréable.

Un grand merci au personnel de la plateforme d’imagerie pour leur disponibilité et leur gentillesse.

Merci à tous les membres de NPA/NPS qui m’ont aidé de près ou de loin dans mon travail, en particulier Odile Poirel, Frank Louis et Salah El Mestikawi pour leur aide et leur collaboration.

Merci à ma famille et mes proches au Japon pour leur soutien sans faille. Ils ont toujours encouragé mes grandes aventures à l’étranger. Merci également à mes amis en France pour leurs encouragements constants et pour tous ces bons moments passés ensemble.

Un grand merci à tous, vous qui avez rendu mon séjour à Paris mémorable.

中島国際交流財団の皆様には四年という長期間に渡るご支援をいただき、ここに博士学位論文としてまとめることが出来たことを心から感謝いたします。財団の方々・審査員の先生方からは経済的支援のみならず、沢山の応援をいただきました。重ねて御礼申し上げます。

最後に、私の選択を暖かく見守ってくれる家族へ。おかげさまで無事にフランス研究留学を終えることができました。本当にどうもありがとう。

Abstract

Neurotransmission mediated by the catecholamines dopamine (DA) and noradrenaline (NA) is critically involved in multiple brain functions and pathologies. NA and DA modulate neuronal excitability and synaptic transmission via Gs and Gi- coupled receptors and cAMP/PKA signaling. NA is released in the entire neocortex by axons of locus coeruleus (LC) neurons. In contrast, DA projections to the neocortex, which originate in the ventral tegmental area (VTA), are present in all layers of restricted medial and lateral areas but only in deep layers of other areas. According to binding assay and immunohistochemistry, dopamine receptors are nonetheless expressed in the entire cortex. The aim of this project is to characterize NA and DA effect using two-photon imaging of a PKA biosensor expressed in pyramidal neurons of cortical slice. We also aim to characterize cAMP/PKA signals elicited by the release of endogenous catecholamines in the neocortex.

I have examined the cortical distribution of NA/DA-induced PKA signals in rat brain slices using bath application of NA (10μM), the β-adrenoceptor agonist Isoproterenol (1μM), DA (10μM) and the D1/D5 receptor agonist SKF 38393(1 μM). Responses to these drugs were observed in layers 2/3 and 5 pyramidal neurons throughout the entire rostro-caudal axis (i.e. in frontal, parietal and occipital cortices) of regions with scarce VTA fibers. Responses to these agonists were inhibited by their respective antagonists (β-adrenoceptor antagonist propranolol, 50μM; D1/D5 antagonist SCH 23390, 1 μM). The negative contribution of D2- like or α-adrenoceptor to responses elicited by DA or NA was also assessed using specific antagonists. The results show a widespread distribution of both NA and DA receptors in the neocortex that extends far beyond the territory of VTA fiber innervation.

In order to investigate the PKA signals induced by the release of endogenous catecholamines, we combined optogenetics and 2-photon PKA imaging. Channelrhodopsin2 was expressed in LC neurons using conditional viral transfer in DBH Cre mice. The PKA sensor was also expressed by viral transfer in the cortex of the same mice. PKA activity in pyramidal cells was reversibly increased by photostimulation (470nm, 5ms pulse, 10Hz, for 5 seconds) and reproducible responses were obtained upon repetitive stimulation. Responses to photostimulation were inhibited by β-adrenoceptor antagonists. The NA transporter inhibitor Riboxetine (100 nM) increased responses to photostimulation. These results demonstrate that endogenous NA released from LC fibers by optogenetic stimulation activates the cAMP/PKA pathway in cortical slices, providing a means to characterize catecholaminergic transmission events.

Résumé

La neurotransmission portée par les catécholamines dopamine (DA) et noradrénaline (NA) joue un rôle primordial dans de nombreuses fonctions cérébrales et est altérée dans de multiples pathologies. La NA et la DA modulent l’excitabilité neuronale et la transmission synaptique par l’intermédiaire de récepteurs couplés aux protéines Gs et Gi qui régulent la voie de signalisation cAMP/PKA. La NA est libérée dans l’ensemble du cortex cérébral par les axones issus des neurones du locus coeruleus (LC). Les projections DA sont originaires des neurones de l’aire tegmentale ventrale (VTA). Elles sont présentes dans toutes les couches de régions restreintes du cortex frontal, médian et latéral, mais uniquement dans les couches profondes des autres régions corticales. Pourtant, des expériences de liaison et d’immunohistochimie ont démontré la présence des récepteurs DA dans tout le cortex. Le but de ce projet de thèse est de caractériser les effets de l’application de NA et de DA exogènes, mais aussi de la libération de catécholamines endogènes sur la voie AMPc/PKA. J’ai réalisé ce travail en combinant l’utilisation de la channelrhodopsine, un canal ionique photosensible, à l’imagerie biphotonique de biosenseurs fluorescents de l’activité PKA exprimés dans les neurones pyramidaux du cortex.

J’ai examiné les signaux PKA induits par l’application de la NA (10 μM), l’isoprotérénol (agoniste β-adrénergique, 1μM), la DA (10 μM) et le SKF 38393 (agoniste des récepteurs D1/D5 dopaminergiques, 1 μM) en tranches de cortex de rat. Toutes ces drogues ont stimulé l’activité PKA dans les cellules pyramidales des couches II/III et V du cortex, notamment dans les régions frontales, pariétales et occipitales peu innervées par les fibres DA. J’ai observé que les réponses NA et isoprotérénol sont inhibées par les antagonistes β - adrénergiques (propranolol, 50μM, CGP 20712 100 nM) et les réponses DA par un antagoniste des récepteurs D1/D5 (SCH 23390, 1 μM). La contribution inhibitrice des récepteurs de type D2 dopaminergiques et des récepteurs α2-adrénergiques a été également mise en évidence par des antagonistes spécifiques. Les résultats montrent que les récepteurs NA et DA sont fonctionnellement exprimés dans de vastes régions du cortex, bien au delà du territoire innervé par les fibres DA issue de la VTA. *

Afin d’étudier les signaux PKA induits par la libération des catécholamines endogènes, j’ai exprimé sélectivement la Channelrhodopsine 2 dans les neurones NA du LC par transfert viral conditionnel dans des souris DBHcre. Le senseur PKA a également été exprimé par transfert viral dans le cortex des mêmes souris. J’ai montré que l’activité PKA est augmentée de façon réversible et reproductible par photostimulation (470nm, 5ms pulse, 10Hz, durant 5 secondes) des fibres NA en tranches corticales. Ces réponses sont inhibées par les antagonistes β- adrénergiques. La riboxetine (100 nM), un bloquant de la recapture de la NA augmente l’amplitude des réponses photo-induites. Ces résultats montrent que la NA endogène libérée par photostimulation des fibres NA active la voie de signalisation AMPc/PKA. Cette approche permet donc de caractériser en temps réel la transmission catécholaminergique avec une résolution subcellulaire.

Abbreviations list AADC – Aromatic amio acid decarboxylase AKAR – A kinase Activity reporter AR – Adrenergic Receptor cAMP – cyclic Adenosine Monophosphate CAT – catecholamine CB – calbindin CCK – cholecystokinine CFP – Cyan Fluorescence Protein CNS – Central Nervous System COMT – Catechol-O-methyltransferase CR – calretinine DA – Dopamine DAT – Dopamine transporter DBH – dopamine β hydroxylase DIO – Double-floxed inverted open reading frame DR – Dopaminergic Receptor EF1α - Elongation factor 1α EMT – Extraneuronal monoamine transporter FSK – Forskolin GABA – Gamma-Aminobutyric acid GAG - CMV enhancer/β-globin chimeric promoter GFP – Green Fluorescent Protein GPCR – G-Protein coupled Receptor LC – Locus coeruleus L-Dopa – dyhydroxypheneyalanine MAO – Monoamine oxydase NA – Noradrenaline NET - Norepinephrine transporter NPY – Neuropeptide Y OCT – Organic cation transporter PKA – Protein Kinase A PKC – Protein Kinase C PMAT – Plasma membrane monoamine transporter PNMT – phenylethanolamine N- methyltransferase PV – parvalbumin SS – somatostatine TH – Tyrosine hydroxylase VIP – vaso-actif intestinal peptide VMAT - vesicles monoamine transporters VTA – Ventral Tegmental Area WPRE element (Woodchuck hepatitis virus post-transcriptional regulatory element) YFP – Yellow fluorescent Protein

Table of contents

I: INTRODUCTION ...... 10 I.1. Catecholamines ...... 10 I.1.1 Catecholamines from a historical point of view...... 10 I.1.2 Catecholamine’s synthesis ...... 12 I.1.3 Storage and release of catecholamines ...... 13 I.1.4 Reuptake and catabolism of catecholamines ...... 13 I.1.4.1 Transporters associated with Uptake1 and 2 ...... 14 I.1.4.2 Catabolism of catecholamines ...... 15 I.2. Noradrenergic and dopaminergic receptors and signal transduction ...... 16 I.2.1. The noradrenegic receptors ...... 17 I.2.1.1 Alpha adrenoceptors...... 17 I.2.1.2 Beta receptors ...... 17 I.2.2. Dopamine receptors ...... 18 I.2.2.1 D1-like receptors ...... 18 I.2.2.2 D2-like receptors ...... 18 I.2.3. Alternative signaling processes ...... 19 I.2.3.1. Receptor signaling through Gq and beta/gamma subunits ...... 19 I.2.3.2. Alternative signaling through oligomerization ...... 20 I.3. The neocortex ...... 22 I.3.1. External anatomy of the neocortex ...... 22 I.3.2. Neurons in the cortex ...... 23 I.3.1.1 Interneurons ...... 23 I.3.1.1 Pyramidal neurons...... 25 I.3.3. Laminar and columnar organization of the cortex ...... 26 I.3.4 Innervation of the cerebral cortex by noradrenergic and dopaminergic neurons ...... 29 I.3.4.1 The locus coeruleus ...... 29 I.3.4.2 The ventral tegmental area ...... 30 I.3.5 Distribution of DA and NA receptors in the cerebral cortex ...... 32 I.3.5.1 Cortical distribution of D1 and D5 receptors ...... 32 I.3.5.2 Cortical distribution of D2-like receptors ...... 33 I.3.5.3 Cortical distribution of alpha adrenergic receptors ...... 34 I.3.5.4.Cortical distribution of beta adrenergic receptors ...... 35 I.4. Optogenetic tools to study NA and DA neurotransmission ...... 36

I.4.1. cAMP/PKA imaging using genetically encoded sensors...... 36 I.4.1.1 Single wavelength measurements ...... 37 I.4.1.2. Dual wavelength measurements for ratiometric imaging...... 38 I.4.2 PKA sensors ...... 39 I.4.2. Channelrhodopsins: light gated cation channels to activate neurons...... 41 I.4.2.1. Molecular aspects of channelrhodopsins...... 41 I.4.2.2. Light induced photocurrents and optimization of channelrhodopsins...... 42 I.4.3 Expressing optogenetic tools in specific neurons using viruses and transgenic animals ...... 44 I.4.3.1 Specific expression using the Cre-Lox system ...... 44 I.4.3.1 Sindbis virus ...... 46 II. RESULTS: CATECHOLAMINERGIC TRANSMISSION IN THE NEOCORTEX...... 49 II.1 Summary of the article: “Noradrenalin and dopamine receptors both control cAMP-PKA signaling throughout the cerebral cortex” by Nomura et al...... 50 II.1.1 Distribution of LC and VTA fibers in the cerebral cortex ...... 50 II.1.2 NA and DA receptors widely trigger PKA activation in cortical pyramidal neurons ...... 51 II.1.3 Gs-coupled receptors ...... 52 II.1.4 Gi-coupled receptors ...... 52 II.2 Responses to endogenous catecholamines released by optogenetic stimulation of LC fibers in the neocortex ...... 54 II.2.1 Expression of channelrhodopsin in cortical LC fibers ...... 54 II.2.2 Expression of a PKA sensor in cortical pyramidal neurons from adult mice ...... 56 II.2.3 Photostimulation of LC fibers induces PKA activation in pyramidal cells ...... 57 II.2.4 Light-induced PKA responses are mediated by β adrenoceptors but do not involve D1/D5 DA receptors ...... 59 II.2.5 Inhibition of NA re-uptake potentiates light-induced PKA responses ...... 60 II.2.6 Light-induced PKA responses do not involve D2-like DA receptors ...... 62 II.2.7 Discussion ...... 62 II.3 Conclusion and perspectives ...... 64 III. MATERIALS AND METHODS ...... 66 III.1.Animals and brain slices ...... 66 III.1.1 Source and housing ...... 66 III.1.2 Viral production and stereotactic injections ...... 66 III.1.3 Brain slice preparation for imaging and drugs ...... 67 III.1.3 Immunohistochemistry ...... 67 III.2. Two photon imaging and photostimulation ...... 68

III.2.1. Two photon imaging...... 68 III.2.2. Light stimulation of ChR2 ...... 68 IV. APPENDIX: Hepp et al. 2014 ...... 70 V. REFERENCES ...... 71

I: INTRODUCTION

I.1. Catecholamines The catecholamines (CAT) constitute a large family of molecules comprising the endogenous substances dopamine (DA), noradrenalin (NA, also termed norepinephrine) and adrenaline (epinephrine),as well as many artificial molecules such as isoproterenol. Their discovery and investigation constitutes a major field in the history of physiology.

I.1.1 Catecholamines from a historical point of view The premises of the research about CAT can be found in the asthma book by Henry Hyde Salter (Hyde 1864) when he describes ″The cure of asthma by violent emotion is more sudden and complete than by any other remedy whatever; ….The cure … takes no time; it is instantaneous…”.″ ″Cure″ due to release of adrenaline from the adrenals is the interpretation that can be given today. In the early 1890s, the German pharmacologist Carl Jacobj (Jacobj 1892) showed that peristaltis of the intestine induced by electrical stimulation of the vagus nerve was abolished by electrical stimulation of the adrenals. A few years later, Oliver and Schäfer (Oliver and Schäfer 1894) discovered the cardiovascular action of suprarenal extract and the concept of hormones was starting to emerge.

The chemist John Jacob Abel (Abel 1899), was the first to partially purify a CAT from adrenal glands in Baltimore in 1898. He named it epinephrine. In 1900, the same epinephrine was totally purified and crystallized by the japanese chemist Jokiti Takamine (Takamine 1901) who called it Adrenalin (A). Noradrenalin (NA), also called norepinephrine was later discovered in peripheral nerves by Ulf von Euler in 1946 and a little latter found in the brain. Dopamine (DA) was identified ten years after the discovery of NA by Arvid Carlsson (Carlsson 1959) as yet another neurotransmitter in the brain and not just the NA precursor.

Studies about the synthesis of the CAT took place in the late 30’s and Hermann Blashko proposed the following synthesis pathway in 1938: tyrosine→ dopa→ DA→ NA→ A. This pathway was only confirmed in the mid 50’s. The inactivation of CAT by monoamine oxydase (MAO) and by Catechol-O- methyl Transferase (COMT) was shown respectively by Blashko and the biochemist Julius Axelrod (Blaschko et al. 1937; Blaschko 1939; Axelrod 1962). Axelrod did also a very important discovery that was the reuptake of CAT in the late 50’s and early 60’s, which is now accepted as a general principle of neurotransmitter

termination. Plasma membrane and vesicular transporters of CAT were only cloned in the 90’s.

The concept of receptor was proposed by Dale in the early 1900. He called these molecules “receptive substances” but could not go much further in the characterization of CAT receptors. Only 40 years later, did Ahlquist (Ahlquist 1948) propose that one CAT could act on two classes of receptors. The development of new pharmacological products confirmed Ahlquist’s postulate in the mid 60’s. The first gene coding for a CAT receptor was cloned in 1986 by Richard et al. Genes for all mammalian catecholamine receptors have now been cloned. There are nine adrenoceptor genes α1A, α1B, α1D, α2A, α2B, α2C, β1, β2 and β3 and five

DA receptor genes D1, D2, D3, D4 und D5.

I.1.2 Catecholamine’s synthesis The CATs are organic molecules containing a catechol and monoamine groups. CATs are synthesized from L-tyrosine in CATergic neuroendocrine cells and in neurons of the peripheral and central nervous systems (Fig. 1). L-Tyrosine is converted to dihydroxyphenylalanine (L-Dopa) by the cytosolic enzyme tyrosine hydroxylase (TH). This reaction represents the rate-limiting step in catecholamine biosynthesis. Aromatic amino acid decarboxylase (AADC) then catalyzes the decarboxylation of L-DOPA to DA which is translocated from the cytoplasm into vesicular storage granules of the cells. In dopaminergic neurons of the central nervous system, DA serves as the end product neurotransmitter. The DA formed in noradrenergic neurons and chromaffin cells is converted to NA by dopamine β- hydroxylase (DBH), an enzyme that is found only in the vesicles of cells that synthesize NA and A. In adrenal medullary chromaffin cells, NA is metabolized by the cytosolic enzyme phenylethanolamine N-methyltransferase (PNMT) to form A. A is then translocated into chromaffin granules, where the amine is stored while awaiting release (for more details see Kvetnansky et al. 2009; Jameson et al. 2010)

Figure 1. The catecholamine synthesis pathway (From Kvetnansky et al. 2009)

I.1.3 Storage and release of catecholamines The storage of CATs in vesicles and granules requires vesicular monoamine transporters (VMAT, for review see Wimalasena 2011). Two forms of VMAT have been cloned but VMAT2 is the only form found in neurons. The energy for vesicular uptake by VMAT is provided by the proton gradient established by H+ transporters. Vesicular stores of CATs are not static and exist in a highly dynamic equilibrium with the cytoplasm, as inward transport by VMATs is counterbalanced by outward leakage (Eisenhofer et al. 2004).

The release of CATs is stimulated by an influx of Ca2+, which in neurons is controlled primarily by membrane depolarization. This process occurs at the level of dilated portions of the axon called varicosities (Fig.2) in which vesicles aggregates can be observed (Seguela et al. 1990). The demonstration that most of these varicosities are non-junctional led Agnati and Fuxe to propose the existence of volume transmission complementary to the well known synaptic transmission in the CNS (Agnati and Fuxe 1986). Note that in spite of the absence of clear CATergic synapses, the terms presynaptic, post-synaptic and synaptic will be often used in this manuscript as a convenient way to designate the release sites on axon varicosities (i.e. the CAT source), the juxtaposed cellular membrane comprising receptors (i.e. the CAT target) and the extracellular space between these elements.

Figure 2. Varicosities on the NA fibers of the cortex. Electron micrographs of a longitudinally cut NA axon of the frontal labeled for NA. Note the small transverse diameter of this axon and the elongated shape of its two varicosities (dilations containing aggregated synaptic vesicles). It also outlines and occasionally opacifies synaptic vesicles x 21,000.scale bar 1µm.On the right, a non- junctional NA varicosity from the occipital cortex. No portion of this varicosity exhibits any suggestion of a membrane specialization. (From Seguela et al. 1990)

I.1.4 Reuptake and catabolism of catecholamines The biological effects of released CAT are terminated rapidly by uptake back into the axon varicosity and/or the target cells, or by conversion of CAT to inactive metabolites. Two

uptake mechanisms have been described in the frame of synaptic transmission: uptake1 and uptake2. Uptake1 occurs at the presynaptic nerve terminal to remove the neurotransmitter from the synapse. Uptake 2 occurs at postsynaptic and peripheral cells to prevent the neurotransmitter from diffusing laterally (Fig.3).

Figure 3. Representation of the main neurotransmission steps at a synapse. The neurotransmitter is synthesized in the presynaptic neuron, stored in synaptic vesicles and released by exocytosis. The neurotransmitter then acts at metabotropic and/or ionotropic receptors and is removed from the synaptic cleft by uptake in presynaptic or postsynaptic neurons and/or glial cells. The uptake process is carried out by plasma membrane-bound neurotransmitter transporters (PNT). (From Masson et al 1999)

I.1.4.1 Transporters associated with Uptake1 and 2 The NA transporter NET and the DA transporter DAT belong to the class of sodium-chloride Na+/Cl–-dependent transporters underlying uptake1, which is a high-affinity and low-capacity transport system. They both can transport NA and DA. The reuptake of NA and DA is essential in regulating the concentration of monoamine neurotransmitters in the extracellular space. The transporter also helps maintain homeostasis of the source neuron. NET expression is restricted to NA neurons and is not present on neurons that release DA. The transporters can be found along the cell body, axons, and dendrites of the neuron (Schroeter et al. 2000). NETs are located away from the synapse where NA is released, thus, to be reuptaken, NA needs to diffuse away from its release site (Torres et al. 2003). Similar features have been observed for DAT. The study by Hersch et al. in 1997 in the striatum showed that DAT is located in the plasma membrane of axon varicosities but not in active synaptic zones

suggesting that striatal DA reuptake may occur outside of the synaptic cleft following DA diffusion. The same study also showed dendritic expression of DAT in the subtantia nigra.

Uptake 2 is a sodium and chloride-independent, low-affinity and high-capacity transport system. In the brain, the extra-neuronal monoamine transporter (EMT; also called organic cation transporter, OCT3 in the rat) and the plasma membrane monoamine transporter (PMAT) have been described as the main molecular correlates of uptake 2 (Grundemann et al. 1998; Koepsell, 2004, Duan and Wang 2010 ). EMT/OCT3 and PMAT are mainly found in neurons and are rare in astrocytes (Vialou et al. 2008; Dhalin et al. 2007).

I.1.4.2 Catabolism of catecholamines While reuptake of catecholamines plays an important role in regulating their extracellular levels, degradation also contributes significantly to the termination of catecholamine neurotransmission (Fig. 4). CATs can be degraded by two main enzymes: monoamine oxidase (MAO) and catechol-o-methyl transferase (COMT). Respectively, these enzymes oxidize monoamines (including catecholamines) and methylate the hydroxyl groups of the phenyl moiety of CATs. The toxic aldehyde intermediates generated in the MAO reaction is either rapidly reduced to an alcohol (by cytosolic aldehyde reductase and/or aldose reductase) or oxidized to an acid (by mitochondrial aldehyde dehydrogenase).

Figure 4. Simplified diagram illustrating catecholamine metabolic pathways. The metabolites produced by MAO can be further metabolized by COMT and inversely. The actions of aldose or aldehyde reductase, aldehyde and alcohol dehydrogenaase and the pathways of sulfate conjugation are not shown. DBH, dopamine β-hydroxylase; PNMT, phenylethanolamine-N-methyltransferase; MAO, monoamine oxidase; COMT, catechol-O-methyltransferase; DHPG, 3,4- dihydroxyphenylglycol. VMA,vanillylmandelic acid; HVA Homovanillic acid.(From Zianni et al. 2004)

Separate genes encode two isoforms of MAO (types A and B), which are located on the outer membranes of mitochondria. In the brain, MAO-A is preferentially located in dopaminergic and noradrenergic neurons while MAO-B appears to be the major form present in serotonergic neurons and glia (for review see Shih et al. 1999).

Membrane-bound COMT is widely expressed and appears to be located principally in postsynaptic neurons, although a soluble form with lower affinity for catecholamines is present in glia, and is also widely distributed outside the brain (Myöhänen et al. 2010)

I.2. Noradrenergic and dopaminergic receptors and signal transduction CATs exert their effects through G protein coupled receptors. There are three major types of adrenergic/noradrenergic receptors (AR): α1, α2 and β, each divided in 3 subclasses, and five dopaminergic receptor (DR) grouped in two classes: D1-like and D2-like (Fig. 5)

Figure 5: Receptors for NA (top) and DA(bottom) Modified from Bylund 1992 and Beaulieu and Gainetdinov 2011.

I.2.1. The noradrenegic receptors

I.2.1.1 Alpha adrenoceptors

The α1-ARs are considered as postsynaptic and not presynaptic receptors, but like other postsynaptic receptors in the brain, they can cause modulation of the release of neurotransmitters. α1-ARs are generally considered excitatory. In the somato-sensory cortex,

α1-ARs activation has been found to increase the excitation seen after administration of glutamate or acetylcholine (Mouradian et al., 1991). α1-ARs can also directly enhance neurotransmitter release from glutamate terminals that innervate layer V pyramidal cells of the prefrontal cortex (Marek and Aghajanian, 1999). α1-ARs may affect many brain functions via non-neuronal mechanisms since they are also localized to glial cells. The activation of α1- ARs has been found to increase calcium transients in hippocampal astrocytes and Bergmann glial (Kulik et al., 1999). α1 receptors are generally coupled to Gq proteins, and can thus activate phospholipase C and phosphatidyl inositol intracellular signaling, resulting in activation of protein kinase C (PKC) and the release of intracellular calcium via inositol 1,4,5- triphosphate (Duman & Nestler, 1995).

NA binds α2-ARs with the highest affinity. The α2A and the α2C subtypes are found pre- synaptically on NA cells and terminals, while all three α subtypes are found post-synaptically (MacDonald et al., 1997). The most classical physiological function defined for α2-ARs is to mediate presynaptic feedback inhibition of neurotransmitter release from noradrenergic varicosities (Gilsbach et al. 2011). Due to the widespread expression of α receptors on post synaptic cells, α2-ARs are also involved in many post-synaptic and extra-synaptic actions (reviewed in Gilsbach et al.2011). The α2 receptors are generally coupled to Gi proteins (Duman & Nestler, 1995), which can reduce intracellular cyclic adenosine monophosphate (cAMP) production by inhibiting adenylyl cyclases.

I.2.1.2 Beta receptors Beta ARs are widely expressed both in the central nervous system and in the periphery. Three subtype of the beta receptor were cloned and called β1 to 3. β1 and β2 are abundant in the brain. In the rat brain, β1-ARs are found abundantly in the cortex, in the striatum and in the hippocampus. β2 receptors were mostly found in the cerebellum and hippocampus. Using RT- PCR, Summers et al. (1995) showed that β3 mRNA is also expressed in various brain regions. Two splice variants were described for the β3 receptors, which differ in their C-terminal tail. The β3b is the predominant form in the brain (Evans et al. 1999). However, the cellular

distribution and the role of β3 receptors in the brain remain largely unknown. All three β adrenoceptors are coupled to Gs proteins and stimulate the cAMP/PKA pathway. It has been reported that β2 can also couple to Gi protein, thereby inhibiting the Gs-driven activation of the cAMP/PKA pathway (Xiao et al. 1999).

I.2.2. Dopamine receptors

I.2.2.1 D1-like receptors The D1-like subfamily of DA receptor comprises the D1 and the D5 receptors originally termed D1A and D1B. Besides these two functional genes, two pseudogenes were identified for the human D5 (Grandy et al., 1991). Upon activation, D1 and D5 receptor stimulate the cAMP/PKA pathway as they are coupled to Gs proteins (Golf in the striatum). D1-like receptors are widely expressed in the brain and in the periphery. In the brain, D1 receptors are expressed at a high level in the nigrostriatal, mesolimbic, and mesocortical areas. D1 and D5 receptor are expressed at both the presynaptic and the post synaptic levels. At the presynaptic level it has been shown that D1-like receptors facilitate neurotransmitter release (Cameron and Williams 1993).

I.2.2.2 D2-like receptors The D2-like group comprises the D2, D3 and D4 receptors. D2-like receptors couple to the Gi/o family of G proteins and thus induce inhibition of cAMP/PKA signaling. D1-like and D2-like receptors differ in the structure of their genes. The D1 and D5 dopamine receptor genes do not contain introns in their coding regions, but the genes that encode D2-like receptors have several introns. (Gingrich and Caron, 1993). Alternatively spliced isoforms have been described for D2 and D3 receptors. The D2-short and the D2-long forms of the D2 receptor are functional. D2S has been shown to be mostly expressed pre-synaptically and to be primarily involved in autoreceptor functions, whereas D2L seems to be a predominantly postsynaptic isoform (Usiello et al., 2000; De Mei et al., 2009). Splice variants of the D3 dopamine receptor have also been described, and some of the encoding proteins have been shown to be nonfunctional (Giros et al., 1991).

D2-like receptors are widely expressed in the brain. The highest level of D2 receptor is found in the striatum, the nucleus accumbens and the olfactory tubercle (Mansour et al. 1990).The D3 receptor is strongly expressed in limbic areas. The D4 receptor has a widespread expression in the brain but exhibits the lowest expression level of D2-like receptors. The strongest levels are observed in the cortex and the striatum (Khan et al. 1998). D2-like

receptors are expressed both pre- and post-synaptically. Presynaptic autoreceptors can provide an important negative feedback that adjusts neuronal firing rate, synthesis, and release of the neurotransmitter in response to changes in extracellular neurotransmitter levels (Wolf and Roth, 1990; Sibley, 1999)

I.2.3. Alternative signaling processes DA and NA receptors, like other G coupled protein receptor were shown to form multimers or signaling complexes with a variety of receptors and channels. The importance of these complexes in regulating intracellular signals is briefly described below.

I.2.3.1. Receptor signaling through Gq and beta/gamma subunits There is evidence that, in addition to their effects on cAMP-regulated signaling, dopamine receptors can also couple to Gq to regulate phospholipase C (for review see Beaulieu and Gainetdinov 2011; and Hasbi et al 2010). This is especially true for the D5 receptor as the activation of that pathway by DA can be observed in D5-transfected cells but not in D5 knock-out mice (So et al. 2009; Sahu et al. 2009). The molecular mechanism leading to the activation of the Gq pathway is not fully understood and the possible interaction of the receptors with specific proteins or other GPCRs is still under investigation. Both D2 receptors and α2 receptors were shown to regulate calcium signaling through the beta/gamma subunit of their cognate G protein (for review see Beaulieu and Gainetdinov 2011; Delmas et al. 1999). In the striatum, it was shown that these subunits activate phospholipase C, leading to inhibition of L-type calcium channels in enkephalin-expressing medium spiny neurons (Hernandez Lopez et al. 2000). Perforated patch experiments performed in sympathetic neurons showed that α2 adrenoceptors inhibit N-type calcium currents via both Gi and Go proteins with beta and gamma subunits as the most probable final transducer (Delmas et al. 1999).

I.2.3.2. Alternative signaling through oligomerization Oligomerization is an important process in GPCR biology, which can change the coupling of receptors to intracellular signaling pathways. If considered as monomers, D1/D5 receptors will activate the cAMP pathway while D2-like receptors will inhibit it. Interestingly, it has been shown that D1-like and D2-like receptors can form heterodimers. In contrast to monomers, these dimers activate the phospholipase C pathway and thus regulate intracellular calcium levels (Fig. 6, Lee et al., 2004, for review see Hasbi et al. 2010).

Figure 6. D1-D2 receptor dimerisation and its implication for signaling. Dopamine D1 and D2 receptor homomers modulate adenylyl cyclase activity and cAMP accumulation in opposite ways through Gs/olf and Gi/o proteins, respectively (left panel). D1–D2 receptor heteromer activation leads to a novel signaling pathway mediated through Gq/11 protein activation, to a rapid, transient, intracellular calcium mobilization from endoplasmic reticulum (right panel). From Hasbi t al. 2010.

In addition to forming oligomers between different members of the same family, DA and NA receptors were shown to interact with GPCRs belonging to different families. The heteromers thus formed can regulate different signaling pathways and often possess a specific pharmacological profile. It has been recently shown that β1 or β2 receptors can heterodimerize with Gi coupled adenosine A1 receptors in recombinant systems but also in native heart cells. The oligomers exhibit a reduced affinity for β receptors ligands but not for those of the A2 receptor. The inhibitory effect of the A1 receptors on cAMP production was abrogated in both A1-β1 and A1-β2 expressing cells in response to A1 agonist (Chandrasekera et al. 2013). Adenosine A2a receptors and D2 receptor oligomers were also demonstrated (Fuxe et al. 1998).

Interactions of NA and DA receptors with GPCRs regulating different signaling pathways were demonstrated. Thus DA receptors were shown to interacts with 5HT2 serotonin

receptors, metabotropic glutamate receptors of type 1 or 5, neurotensine receptors, cholecystokinine receptors and others (for review see Beaulieu and Gainetdinov 2011; Fuxe 2014). Similarly NA receptors were shown to form dimers with olfactory receptors, opioïd receptors, angiotensin receptors, cannabinoïd receptors and others (for review see Cotecchia et al. 2012). Besides the modification of the pharmacological properties and the alteration of the signaling pathway, it is noteworthy that oligomerization also indirectly modulates cell signaling by regulating receptor trafficking and internalization (Fig.7).

Figure 7: Dimerization regulates internalization of DA receptors: (A) HEK 293 cells individually transfected with D1 or D3-GFP cDNAs were exposed to either the D1/D5 agonist SKF 81297 or to the D2-like agonist quinpirole. The D1 agonist induces internalization of D1 receptors but quinpirole does not induce internalization of D3 receptors.(B). HEK cells were co-transfected with D1 and D3 receptor. Here the D1 agonist does not allow internalization of the D1 receptor. When both SKF81297 and quinpirole are applied, both receptors internalize. DA which activated both D1 and D3 allows internalization of both receptors. From Fiorentini et al. 2008.

I.3. The neocortex The neocortex is the most superficial structure of the brain and from a phylogenetical point of view it is the latest structure that appeared in the brain. The neocortex is involved in higher functions of the nervous system, including voluntary movement, cognition, sensory perception and language. These functions are based on complex cellular networks which are regulated by various neuromodulators including DA and NA.

I.3.1. External anatomy of the neocortex The neocortex is divided in two hemispheres, themselves comprising four lobes delimited by fissures. Each of these lobes has specialized function: the frontal lobe regulates the voluntary movements, the temporal lobe is implicated in audition, memory and emotions, the parietal lobe is specialized in somesthesia and the occipital lobe is involved in vision (Fig. 8).

Figure 8: The four lobes of the cerebral cortex (lateral view). http://falnaufal17.blogspot.fr/2011/03/what-does-cerebrum-do.html

In humans and higher mammals, the cerebral cortex is folded into many gyri and sulci, which have allowed the cortex to expand its surface without large changes of its thickness (Fig.9).

Figure 9. The external aspect of the cerebral cortex in various species (De Felipe 2011).

I.3.2. Neurons in the cortex Three major cells types are found in the brain including the cortex: Glial cells which comprise microglial cells, oligodendrocytes, and astrocytes; cells from the cerebral vasculature and neurons which can be divided into interneurons and pyramidal cells. Only the neurons will be described in the next paragraphs with a special emphasis on pyramidal neurons.

I.3.1.1 Interneurons Inhibitory interneurons can be found throughout the cortex. They represent about 20% of the cortical neurons. Interneurons were originally described by Cajal as neurons with short axons (Cajal 1995 translation). Interneurons are thus mainly involved in local circuit. Interneurons exert inhibitory action in the cortex as their neurotransmitter is Gamma-aminobutyric acid (GABA). Interneurons display a variety of morphologies among which can be found fusiform, neurogliaform, bipolar, basket cells or chandelier cells (Fig. 10)

Figure 10. morphological diversity of interneurons in the neocortex. The figure shows a reconstruction of several classes of neocortical interneurons. Axons are in blue and dendrites in red. Cortical layers are indicated on the left of the figure. BPC, bipolar cell; ChC, chandelier cell; DBC, double bouquet cell; LBC, large basket cell; MC, Martinotti cell; NBC, nested basket cell; NGC, neurogliaform cell; SBC, small basket cell. (Adapted from Huang et al. 2007)

Besides their variety of morphologies, interneurons also display various profiles in their firing and in their content in biochemical markers: calcium binding proteins, peptides or enzymes

such as the nitric oxide synthase (Fig. 11). Although inhibitory interneurons represent only a small fraction of the cortical neurons, their functions are essential in regulating the cortical excitability and shaping network activities.

Figure 11. Electrophysiological and biochemical characteristics of four types of neocortical interneurons. (Left) The firing properties of the four neurons in response to depolarization or hyperpolarization steps. (Right) Agarose gel showing the expression of various markers present in the four studied interneurons and analyzed by single cell RT-PCR. The markers include GABA synthesizing enzymes GAD65 and GAD67; calcium binding protein calbindin (CB), parvalbumine (PV) calretinine (CR); and neuropeptides: neuropeptide Y (NPY), vasoactive intestinal peptide (VIP); somatostatine (SS) and cholecystokinine (CCK). (Cauli et al. 2000)

I.3.1.1 Pyramidal neurons. Pyramidal neurons represent ~80% of the cortical neurons. Pyramidal neurons exert excitatory action in the cortex as their neurotransmitter is glutamate. Pyramidal neurons are so called because of the morphology of their soma which has a conical shape (Fig. 12). A long apical dendrite emerges from the soma which is primarily oriented towards the superficial layers, in which it densely ramifies and runs parallel to the cortical surface. Several dendrites, of variable length, also emerge laterally from the soma and divide rapidly. The axon usually emerges from the base of the soma or from a basal dendrite. It runs towards the deep layers of the cortex where it ramifies before entering the white matter (Cajal reprint 1995). Pyramidal neurons are covered with thousands of dendritic spines (Fig. 12) that are the site of most excitatory glutamatergic synapses.

A B

Figure 12. Morphologic characteristics of pyramidal neurons of the cortex. Top left: Schematic diagram showing a typical pyramidal cell. (From De Felipe et al 1992). Right: Confocal microscopy image of a portion from the apical (A) and a basal (B) dendrite of an intracellular injected layer III pyramidal neuron of the human cingulate cortex. Many spines can be seen (from Benavides-Piccione 2013).

Despite their stereotypical shape, pyramidal cells are heterogeneous with regard to soma size and shape, dendritic branching, spine density, pattern of axon collaterals or electrophysiological properties (Fig. 13.).

Figure 13. Pyramidal neuron diversity. A-B: Two biocytin labeled pyramidal neurons of the layer V of the somato-sensory cortex of the rat with different morphologies (Schubert et al. 2001). Right: Responses of three different neocortical layer V pyramidal neurons to low (bottom) and high (top) current injections. Adapted from Williams and Stuart 1999.

I.3.3. Laminar and columnar organization of the cortex The neocortex comprises the grey matter which localizes just beneath the meninges and above the while matter. The grey matter is composed of many cells types while the white matter is essentially composed of myelinated fibers. In the rest of the introduction, only the neurons of the cortical grey matter will be considered. The cortical grey matter can be divided into six layers (Fig.14, (Paxinos 2004). The different cortical layers each contain a characteristic distribution of neuronal cell types and connections with other cortical and subcortical regions. All layers contain GABAergic interneurons that are not described below. Layer I is the only layer devoid of excitatory neurons.

Layer I also called molecular or plexiform layer is the most superficial layer. It is mainly composed of nervous fibers which are extensions of apical dendrites of glutamatergic pyramidal neurons from layer II/III and V and horizontally oriented axons, as well as glial cells and sparse interneurons.

Layers II and III, also called external granular layer and external pyramidal layer, respectively, contain many small and medium size pyramidal neurons .Layers I to III are the

main target of inter-hemispheric cortico-cortical afferents, and layer III is the principal source of cortico-cortical efferents.

Layer IV, the internal granular layer, contains excitatory stellate neurons and modified pyramidal neurons, and is a major target of thalamo-cortical afferents as well as intra- hemispheric cortico-cortical afferents.

Layer V, the internal pyramidal layer, contains large pyramidal neurons which give rise to axons leaving the cortex towards many subcortical structures.

Finally, layer VI, the polymorphic or multiform layer, contains pyramidal neurons in its upper part and polymorphic non-pyramidal excitatory neurons in its lower part. Layer VI sends efferent fibers to the thalamus, establishing a precise reciprocal interconnection between the cortex and the thalamus.

Figure 14. Six-layered structure of the cerebral cortex. On the left , the cells are labeled with a Golgi staining (for identification of neuronal and glial cells and their processes), in the middle cell bodies are showed after a Nissl

staining, and on the right only fibers are labeled (From Heimer 1994)

The thickness of individual layers varies among cortical areas. For example, layer IV is almost absent in frontal areas whereas it is very thick in sensory areas. Layer V is most developed in motor areas.

The columnar functional organization which superimposes the laminar organization of the cortex was originally proposed by Vernon Mountcastle in 1957 (Mountcastle 1957). He

suggested that neurons that are horizontally more than 500 µm from each other do not have overlapping sensory receptive fields. The cellular components of a column are vertically interconnected along the total thickness of the cortex (Fig. 15). Cortical columns are especially clear in primary sensory cortices. In rodents, cortical columns are well visible in layer IV of the somato-sensory cortex corresponding to the representation of vibrissae, forming structures called barrels (Fig. 15).

Figure 15. Cortical columns. Cortical columns contain vertically wired cells (left). In the somato- sensory cortex corresponding to the vibrissae representation on rodents Nissl staining (A) show a high density of cells in layer IV delimitating a barrel structure in which thalamus projects (B). Many cells neurites remain confined in the barrel (C). The axons and dendrites of pyramidal neurons from layer II/III and V are preferentially localized at the border of the barrel (D). (http://cs.brown.edu/~tld/projects/cortex/ and From Waite and Tracey 1995)

I.3.4 Innervation of the cerebral cortex by noradrenergic and dopaminergic neurons Noradrenergic and dopaminergic axons found in the cerebral cortex stem from noradrenergic neurons of the locus coeruleus (LC) and dopaminergic neurons of the ventral tegmental area (VTA).

I.3.4.1 The locus coeruleus The LC is a pontine nucleus located bilaterally near the pontomesencephalic junction, on each side of the fourth ventricle. In the rat the LC contains about 1500 neurons. Most of these neurons have NA as a neurotransmitter and a few possess GABA, however, the role of these latter is not well understood. NA neurons have extensively branched axons and project widely throughout the brain. NA neurons can be divided in three subclasses according to their morphology (Cintra et al. 1982; Fig. 16). Each subclass was shown to have specific efferent targets. Likewise, the fusiform neurons project to the cortex and the hippocampus, the multipolar neurons innervate the cerebellum and the spinal cord and the small round-shaped neuron target the hypothalamus (Loughlin et al. 1986).

Figure 16. Noradrenergic neurons and projections in the rodent brain. Top: Several noradrenergic nuclei can be found in the pons (A1 to A7). The most important of them in term of size and projections is locus coeruleus (A6, projections marked in red). Adapted from Zeman 2001. Bottom: three NA cell types were described in the LC. left: multipolar neurons, middle: fusiform neuron, right: round shape neurons (Cintra et al. 1982).

It has been demonstrated that changes in LC activity anticipate changes in behavioral states: LC neurons fire as a function of vigilance and arousal. LC neurons also increase their firing rate in response to salient stimulation such as flashes of light, intense tones, novel stimuli. Neurons of the LC discharge with phasic bursts of activity superimposed on highly regular tonic discharge (Fig.17; for review see Sara 2009; Berridge and Winterhouse 2003; Bennaroch 2009).

Figure 17: Firing rates of the LC. (left) Activity of an LC neuron in a behaving rat. The firing rate decreases when the animal becomes drowsy. (right) Activity of an LC neuron at the onset of a light (Bouret and Sara 2010)

I.3.4.2 The ventral tegmental area Dopaminergic neurons innervating the brain are located in the A8 to A10 cell groups of the midbrain (Fig. 18). The VTA corresponds to the A10 cell group and is part of the basal ganglia complex.

Figure 18. Location of the ventral tegmental area in the midbrain. The VTA is shown in red. The substantia nigra in grey (From Paxinos brain atlas).

The limits of the VTA are not well defined and has been determined by the location of surrounding nuclei. The VTA is heterogeneous and comprises multiple subregions (Fig.19). The VTA is located next to another well studied DA cell group, the substantia nigra (Fig 18).

Figure 19: Three cell types are present in the ventral tegmental area A-B: Glutamatergic neurons (red) are present in the VTA and a subset is DAergic (TH positive neurons in green). GABAergic interneurons (purple labeling) are present throughout the subregions of the VTA (right panel). None of them is DAergic (From Taylor et al. 2014).

The VTA is the origin of the DA cell bodies of the mesocorticolimbic DA system and is widely implicated in the reward circuitry of the brain. The VTA is also associated with learning, motivation and addiction. DA neurons of the VTA represent 50 to 70% of all the neurons present in the structure, a subset of which express the glutamate transporter VGluT2 and release glutamate. Besides the DA neurons, the VTA also comprises an important GABAergic interneuron network (30 to 50% of the neurons) and a few pure glutamatergic neurons (2%) (Fig. 19, Margolis et al 2006; Johnson et al. 1992; Nair-Roberts 2008 ; Taylor et al. 2014).

Figure 20. Dopaminergic nuclei and projections in the rodent brain. Several dopaminergic nuclei are found in the midbrain (A8-A10). The cortex is innervated by the A10 cell group of the ventral tegmental area. DA projections from the VTA are marked in red). Adapted from Zeman 2001.

In contrast to the NA system, DA neurons from the VTA do not innervate the entire cortex. In rodents, DA fibers present in frontal areas and can be observed in all the layers of medial and lateral areas. In contrast, only few DA fibers can be found in other cortical areas where they are restricted to the deep layer (Fig. 20). In primates and humans, DA afferents are more widespread and extend more caudally (Berger et al. 1991).

I.3.5 Distribution of DA and NA receptors in the cerebral cortex Many in situ hybridization, binding and immunohistochemistry experiments demonstrate that DA and NA receptors are widely expressed in the cerebral cortex, each exhibiting specific pattern of expression. In this section, expression of receptors will be mainly examined. in the cortex of rodents.

I.3.5.1 Cortical distribution of D1 and D5 receptors D1 receptors are widely expressed in the cortex with higher levels in regions such as the medial prefrontal cortex, the cingulate cortex, the insular cortex or the perirhinal cortex, which are areas densely innervated by DA fibers. D1 receptors are mainly found on the soma and in dendrites of neurons of the layer II/III and V/VI with a slight enrichment in layer V. (Fig. 21; Ariano and Sibley 1991; Fremeau et al. 1991; Huang et al. 1992, Luedkte et al. 1999; Mansour et al. 1990). A quantitative analysis performed in the medial prefrontal cortex showed that 20% of the pyramidal neurons and 30% of the interneurons express D1 receptors (Santana et al. 2009). These data were confirmed in Drd1a-tdTomato transgenic mice expressing the fluorophore tdTomato under the control of the D1 receptor promoter (Zhang et al. 2010).

Figure 21. In situ hybridization of D1 dopaminergic receptors in the cortex. Dark- field autoradiograms of D1 receptor binding.

cc, corpus callosum; DG: dentate gyrus.(From Mansour et al. 1990)

D5 receptor are widely expressed in the cortex and found on pyramidal and non pyramidal neurons. The number of D5 immunopositive neurons was determined to be larger than D1 positive neurons (Luedtke et al.1999). Immunostaining can be observed in the somata and the dendrites. Laminar differences in D5 immunoreactivity were observed and varied across cortical regions (Cialax et al. 2000; Khan et al. 2000) For example, D5 immunoreactive cells were observed in all layers of the medial prefrontal cortex in contrast to the motor cortex where was more intense in upper layers (Fig. 22)

Figure 22. D5 receptor immunocytochemistry in various regions of rat cerebral cortex. A: Prefrontal cortex (Cg3). B: Motor cortex (Fr1). C: Somatosensory (Adapted from Ciliax et al. 2000)

In the monkey cortex, D1 and D5 receptor were found to be expressed by many cortical neurons. At the electron microscopy level D1 and D5 receptors were found to be expressed at pre- and post-synaptic levels with a preferred post-synaptic distribution. Despite their co- expression, D1 and D5 receptors exhibit complementary subcellular distributions: D1 is found mainly found in spines whereas D5 receptors are found in dendritic shafts (Smiley et al. 1994; Bergson et al. 1995).

I.3.5.2 Cortical distribution of D2-like receptors D2, D3 and D4 were all detected in the neurons of the cortex, each having a specific pattern of expression. Immunohistochemistry experiments showed that among the D2-like receptor D4 is enriched in the cortex compared to D2 or D3 (Fig. 23 and 24) and is strongly present in layer II/III (Khan et al. 1998; Rivera et al. 2008, Defagot et al. 1997).

Figure 23. Rat brain immunocytochemistry with antibodies against D2 (D), D3 (C), and D4 (A) receptor proteins. D4 immunolabeling is concentrated in the cerebral cortex, hippocampus, caudate putamen, and inferior colliculus, whereas D3 and D2 receptors are both enriched in subcortical forebrain areas. Ac, nucleus accumbens; cc, corpus callosum; Cg, cingulate cortex; Ctx, cerebral cortex; Cb, cerebellum; CPu, caudate-putamen; Fr, frontal cortex; GP, globus pallidus; H, hippocampus; Hp, hypothalamus; IC, inferior colliculus; M, medulla; OB, olfactory bulb; OT, olfactory tubercle; P, parietal cortex; Pir, piriform cortex; S, septum; SC, superior colliculus; Th, thalamus. Scale bar 2 mm in A,C,D.(Adapted from Khan et al. 1998).

Figure 24. Rat parietal cerebral cortex immunostained with anti-D4 (A), anti-D3 (B), and anti-D2 (C) dopamine receptor antibodies. D4 immunoreactivity is much more intense than D3 and D2. Note strong D4-immunoreaction product in cell bodies and dendrites of pyramidal and nonpyramidal cortical neurons. I–V represent the cortical layers. Scale bar 67 μm (applies to A–C).Adapted from Khan

et al. 1998.

In contrast to the widely expressed D2 and D4, D3 receptors have a more discrete distribution (Fig.23). Immunohistochemistry data on DA receptors are generally congruent with receptor binding and in situ hybridization experiments, although mismatches between mRNA and protein patterns are occasionally noticed (Weiner et al. 1991; Lesveques et al 1992. Mengod 1992, Martres et al. 1985, Lidow 1989, Bouthenet 1987).

I.3.5.3 Cortical distribution of alpha adrenergic receptors Both α1 and α2 receptors are expressed in the cerebral cortex, with specific expression patterns for their subtypes. Alpha1 are mostly found post-synaptically while α2 receptors are

found both pre- and post-synaptically. Only α2 distribution will be described here since α1 has not been studied in the frame of my thesis. Immunohistochemical studies performed in the visual cortex showed that the α2A receptor is detected in pyramidal neurons, in interneurons and glial cells in all layers of the cortex (fig. 25). This is in accordance with in situ hybridization experiments revealing a ubiquitous labeling in the cortex with stronger labeling in layer VI than in other layers according to some authors (Venkatesan et al. 1996; Nicholas et al. 1993; McCune et al. 1993). However, the use of transgenic mice, in which the reporter gene LacZ is under the control of the α2B receptor promoter indicates that α2B receptors are expressed in the cortex with a preferred expression in layer II/III and VI (Wang et al. 2002, Fig 25), consistent with Some in situ hybridization studies (Tavares et al. 1996; Winzer- Serhan 1997). Alpha2C receptors have a widespread distribution in the cortex. They are found in all the cortical layers. In situ hybridization experiments suggest a higher expression in the forebrain compared to the occipital area (Nicholas et al. 1993, McCune et al. 1993).

Figure 25: Cortical distribution of α2 receptors. Left: Alpha2A receptor immunoreactivity in adult visual cortex as visualized by light microscopy. Few labeled neurons are observed in layer I, a higher density of labeled cells can be seen in layers II/III and V. Perikaryal labeling is also present in layers VI. Middle: Adra2b-NN–directed LacZ expression in cerebral cortex. By P14, reporter β- galactosidase is expressed in cells of layers II/III and VI of the parietal cortex. Right: Localization of α2C mRNAs in the adult rat brain using in situ hybridization. Alpha 2C receptors are widely expressed in the cortex. mRNAs can be found in all the layers (Nicholas et al. 1993; Wang et al. 2002 and McCune et al. 1993)

I.3.5.4.Cortical distribution of beta adrenergic receptors In situ hybridization and autoradiographic experiments have shown that β receptors exhibit a widespread expression in the cortex (Fig. 26). The strongest labeling can be observed for β1 receptors, especially in upper layers (Nicholas et al. 1993b; Lorton and Davis 1987, Rainbow et al. 1984). The extensive expression of β1 receptors was confirmed by immunohistochemistry (Paschalis et al. 2009). In the rat cortex, β2 autoradiographic signals

are stronger in layer IV than other layers (Rainbow et al. 1984). Compared to β1 and β2 receptors, very low levels of β3 were detected in human and rat cortex using PCR techniques (Rodriguez et al 1995; Summers et al. 1995) and the distribution of β3 receptors remains unclear.

Figure 26: Autoradio-graphic distribution of β1 (left) and β2 (right) receptors in the brain. (Rainbow et al. 1984).

I.4. Optogenetic tools to study NA and DA neurotransmission Optogenetics is a general word that applies to the genetically encoded molecules that are sensitive to light. The optogenetic toolbox comprises light activated channels such as channelrhodopsin or halorhodopsin, photoswitchable receptors and biosensors. The next paragraphs will focus on PKA sensors and channelrhodopsin used during the course of this thesis.

I.4.1. cAMP/PKA imaging using genetically encoded sensors. The cAMP/PKA intracellular signaling cascade regulates various processes ranging from the regulation of ion channels to that of gene expression. As seen earlier, NA and DA both regulate this pathway. Hence, recording the dynamics of cAMP and/or PKA activity in intact cells is a potent method to report the effects of exogenous and endogenous catecholamines.

The first attempts to measure directly cAMP concentration in cells were performed by using fluorescently labeled PKA subunits microinjected in cells (Adams et al. 1991). Microinjection was however a major drawback at that time to study cAMP variation in small vertebrate neurons. A major breakthrough in the field of neuroimaging was achieved with the design of genetically encoded optical sensors by the Tsien laboratory. The first of these sensors was designed to report calcium (Miyawaki et al. 1997), but biosensors were engineered subsequently to monitor PKA activity (Zhang et al. 2001).

Genetically encoded fluorescent sensors usually comprise one, two or three fluorescent proteins derived from the jellyfish green fluorescent protein GFP, or from the red fluorescent protein DsRed from the coral Discosoma Sp.(Fig.27).

Figure 27. Structure of the fluorescent proteins. Left: Three-dimensional structure of GFP in two projections. The light emitting chromophore is shown in green, the central helix bearing the chromophore is depicted in yellow. On the right, the aminoacids of the GFP (a) and DsRed (b) chromophores (Stepanenko et al. 2013).

These fluorescent proteins are linked to a protein domain sensitive to the molecule or the enzymatic activity of interest. The biological signal induces a change in the optical properties of the fusion protein, which can be detected by optical microscopy. The genetically encoded sensors are powerful tools as their expression can be targeted to virtually any cell or subcellular compartment.

I.4.1.1 Single wavelength measurements The most straightforward method is to monitor changes in fluorescence intensity at one wavelength. This measurement method applies to various chemical calcium indicators such as fluo-3, and to a series of biosensors whose brilliance depends on various biological signals.

Figure 28: Examples of biosensors based on a single fluorophore (Zhang et al. 2002)

Single wavelength sensors may use the intrinsic sensitivity of a fluorescent protein to a given biological parameter, e.g. pH or chloride, but generally comprise a sensing domain, as for the GCamp series of calcium biosensors (Fig.28). The main drawback in the use of single wavelength sensors is that fluorescence emission can result from various artifacts: fluctuations in illumination intensity, changes in cell volume (thus of sensor concentration) bleaching of the biosensor.

I.4.1.2. Dual wavelength measurements for ratiometric imaging Ratiometric quantification of fluorescence variation can alleviate some of the artifacts related to single wavelength measurements. The major optical property of the sensor to perform ratiometric imaging is that the fluorescence observed at two wavelengths varies with the biological signal in opposite direction. A typical radiometric sensor is the calcium indicator Fura-2 (Fig.29, Grynkiewicz et al., 1985).

Figure 29. Optical properties of Fura-2. Radiometric calcium measurements can be performed using fura-2 as the excitation properties at 340nm and 370nm change in opposite directions with the binding of calcium (indicated by the arrow). From The Molecular probes handbook (Life technologies)

Most of the genetically-encoded sensors used in ratiometric imaging contain two fluorescent proteins. The ratiometric properties of those sensors are due to an energy transfer between an excited fluorophore called the donor and a partner fluorophore called the acceptor. This process is known as fluorescence/Förster energy transfer or FRET. There are three main conditions to obtain FRET: 1) The emission spectrum of the donor fluorophore must strongly overlap the excitation spectrum of the acceptor fluorophore, 2) the distance between the two fluorophores must be below 10 nm and 3) the fluorophores must have a parallel orientation in the protein. The variation of the level of the molecule of interest is measured by calculating the variation of the ratio of fluorescence between the donor and the acceptor fluorophore in the recorded cell. When FRET occurs, the illumination of the donor will result in an increase in the emission of the acceptor and a diminution of the emission of the donor. One of the most widely used fluorophore FRET pairs consists of a blue variant of GFP such as CFP,

mTurquoise or Cerulean and a yellow variant of GFP such as EYFP, Citrine or Venus (Fig. 30).

Figure 30: Spectral properties of the FRET pair CFP/YFP. There is a massive overlap between the emission spectrum of CFP and the absorption spectrum of YFP. Source:

Semrock website

Semrock web site

I.4.2 PKA sensors I studied the effects of DA and NA on cortical neurons using AKAR3EV and GAKdYmut, two PKA sensors derived from the original “A kinase activity reporter” AKAR2 (Zhang et al. 2005) and AKAR3 (Allen and Zhang 2006). Both AKAR2 and AKAR3 are composed of a tandem of CFP and YFP derivatives flanking a PKA sensitive domain (Fig. 31). This domain in composed of a synthetic substrate peptide created with a consensus phosphorylation site specific for PKA and a protein domain that binds to this peptide when it is phosphorylated. Here the domain binding to the phosphopeptide is a forkhead-associated (FHA) domain. In the case of AKAR2, the FRET pair is composed of the enhanced CFP (ECFP) and of citrine, a yellow variant of GFP showing little sensitivity to pH fluctuations, In AKAR2, the substrate peptide and the FHA domain are linked together with a few aminoacid linker and directly to the fluorophores. When used in neurons in brain slices, a maximal ratio change of about 20% can be measured in response to the application of 13µM of the adenylate cyclase activator forskolin (Gervasi et al. 2007). The multiple mutations performed on the fluorophores led to the improvement of their biophysical properties. Higher FRET efficiency could be observed with certain GFP mutants, which were introduced into the AKAR2 sensor to create AKAR3, in which the citrine was replaced by a yellow circularly permutated cpVenus. The increased dynamic range observed for AKAR3 was not only due to the modification of the yellow fluorophore but also to the introduction of a 14 amino acid flexible linker between the PKA substrate peptide and the FHA domain (Allen and Zhang 2006).

Figure 31: AKAR sensors. A variety of AKAR sensor exist. Most are derived from the FRET sensor AKAR2. A few non FRET sensors were developped. LumAKAR is a luminescent AKAR based on complementation of two parts of luciferase upon phosphorylatiobn. GAkdYmut contains two fluorophores one GFP and a dead fluorophore.

AKAR3 was further improved by replacing the blue fluorophores ECFP by the brighter variant Cerulean to generate AKAR4 (Depry et al. 2011). Besides the work of the Jin Zhang’s group in generating improved AKAR sensors, the group of Michiyuki Mastuda improved AKAR3 by changing the fluorophores but also by introducing a long flexible linker of 116 amino acids between the peptide and the FHA domain thus reducing basal intramolecular FRET (Komatsu et al. 2011). The introduction of the long linker associated with the FRET pair ECFP/Ypet gave rise to the efficient AKAR3EV. In transfected HEK cells, the maximal response obtained for AKAR3EV and AKAR4 when applying 13 µM forskolin was ~50% and ~25% respectively (unpublished data from R. Hepp).

The GAKdYmut sensor is a single GFP sensor, which has been engineered from AKAR2 by my host laboratory (Bonnot et al., 2014). This sensor was originally designed to image PKA signals using two-photon microscopy. It comprises a GFP derived from the AKAR2 CFP as sole fluorophore, while the YFP has been inactivated (Fig.32). The maximal response to forskolin of the original AKAR2 sensor in cortical neurons from brain slices is a ratio variation of ~20% change, while we observe up to four fold increases in GAKdYmut fluorescence intensity in the same experimental conditions (Bonnot et al., 2014).

Figure 32. The GAKdYmut sensor. Top: structure of the PKA sensor AKAR2 containing CFP (blue) and YFP (yellow). Underneath is depicted the structure of the PKA sensor GAKdYmut. The CFP was mutated back into a GFP and dYmut is a dead YFP fluorophore which does not absorb or emit light. Bottom: Upon PKA phosphorylation, the phoshorylated sensor changes conformation resulting in an increase of the fluorescence of the GFP. Excitation wavelengths correspond to two photon excitation. Adapted from Bonnot et al. 2014.

I.4.2. Channelrhodopsins: light gated cation channels to activate neurons. Chlamydomonas reinhardtii is a single-cell green alga that swims with two flagella. The light induced behavior of this alga is mediated by light-activated channels called Channelrhodopsin 1 (ChR1) and Channelrhodopsin 2 (ChR2), which were originally identified by three independent groups (Nagel et al. 2002, 2003; Sineshchekov et al. 2002; Suzuki et al. 2003). The demonstration that ChRs act as light-gated ion channels was followed by major advances in neuroscience, where ChRs are used for neuronal depolarization simply via application of light pulses.

I.4.2.1. Molecular aspects of channelrhodopsins Channelrhodopsins, whose variants can also be found in other algae such as Volvox carteri, belong to the microbial-type rhodopsin protein family. The core structure is composed of seven transmembrane domains that form the ion channel and a retinal molecule that covalently binds to a consensus Lys residue at the middle of TM7 (Fig. 33). Photoisomerization of an all-trans retinal to a 13-cis configuration follows light absorption. This conformational change allows the channel pore to become permeable to cations, such as Na+, K+, Ca2+ and H+ in a few milliseconds (Fig. 33, Nagel et al. 2002 and 2003).

Figure 33. ChR2 activation by light. Top left: Schematic representation of the ChR2 structure. On the bottom left: isomerisation of retinal by blue light (http://www.biophys.mpg.de/en/bamberg.html). Top right. Photocurrents of a HEK293 cell, transiently expressing ChR2. Photocurrents were determined at –100, –50, 0, +50, and +100 mV (Nagel et al. 2003).

I.4.2.2. Light induced photocurrents and optimization of channelrhodopsins. ChR1 is best activated at a wavelength of 500 nm in physiological conditions and is highly permeable to protons (Nagel et al. 2002). The nature of the ions flowing through the channel makes ChR1 currents sensitive to intracellular and extracellular pH. ChR1 is not currently used in neuroscience due to the small induced photocurrent compared to its homologue ChR2 (for review see Nagel et al. 2005). In contrast to ChR1, ChR2 is a non selective cationic channel (Nagel et al. 2003). The excitation spectrum of ChR2 shows a peak around 473nm, in the blue light range. This channel opens rapidly after absorption of a photon to generate a large conductance for monovalent cations and some divalent cations (e.g. calcium but not magnesium). ChR2 desensitizes in the continuous presence of light to a smaller but still important steady-state conductance (Fig 33, Nagel et al. 2003). The desensitization of the currents was shown to be dependent on pH and membrane potential.

ChR2 was the first ChR to be used to excite neurons using light. It was shown that the stimulation was reliable and allows millisecond control of neuronal spiking without interfering with the endogenous electrophysiological properties of the neurons (Boyden et al. 2005). The main downsides of ChR2 are the level of desensitization, which strongly

diminishes the current in neurons. and the slow recovery from desensitization (about 20 seconds in the dark), which precludes high frequency stimulation. Besides, high levels of expression induce the formation of intracellular aggregates and thus incorrect targeting to the plasma membrane. Several variants of ChR2 were produced using site directed mutagenesis to improve ChR2 properties. During my work, I used ChR2 bearing a single mutation H134R (Fig.34, Nagel et al. 2005). ChR2-H134R has a modest reduction in desensitization, a slight increase in light sensitivity and slower channel closing compared with ChR2 These changes increase photocurrents, but the slower kinetics makes ChR2/H134R less temporally precise than ChR2 (Nagel et al. 2005). Besides ChR2-H134R, several other variants are now available with faster kinetics but reduced photocurrents (for example ChETA (ChR2-E123T); Fig.35 Gunaydin et al. 2010), strong photocurrents but slower kinetics (for example ChR2- T159C, Berndt et al. 2011, Fig.34). To improve ChR properties chimeras between ChR1 and ChR2 were also generated (ChED/ChEF/ChIEF) and some exhibit an increased steady state response with kinetics similar to ChR2 (for review see Lin 2012, Fig. 34). Most of the ChRs available are excited by bluish light although some have a red shifted excitation such as ChEF and ChETA (490 nm) or VChR1 (530–570nm).

Figure 34. Comparison of light-induced ChR responses to long light pulses for various ChR variants. Depicted are ChR photoresponses to 500ms

470nm light pulses. Each variant has different levels of desensitization and kinetics (from Lin 2012)

I.4.3 Expressing optogenetic tools in specific neurons using viruses and transgenic animals The increasing use of optogenetic tools in neuroscience is driven by the fact that they can be targeted to specific cell-types within complex neural networks to understand their role in complex structures. Classical transfection methods such as lipofection or electroporation cannot be used to drive specific expression. The common techniques used for expression in animals are transgenic animals, viral vectors, and in utero electroporation. In this manuscript, the use of the Cre-Lox system will be described in combination with the use of adeno- associated and Sindbis viruses.

I.4.3.1 Specific expression using the Cre-Lox system Cre-Lox recombination is known as a site-specific recombinase technology, and is widely used to carry out deletions, insertions, translocations and inversions at specific sites in the DNA of cells. It allows the DNA modification to be targeted to a specific cell type or be triggered by a specific external stimulus. Cre-Lox recombination involves the targeting of a specific DNA sequence and splicing it with the enzyme Cre recombinase. This enzyme is a site specific recombinase which catalyses the recombination between identical loxP sequences (Fig. 35). LoxP sequences were identified in bateriophages P1 and contain 34 bp with two inverted recognition sites sandwiching an 8bp spacer region (Missirlis et al. 2006).

No expression Recombinaisonrecombinaison

Expression

Figure 35: Example of Cre-Lox recombination to induce gain of function.

To achieve specific expression of the protein of interest, Cre and loxP mice strains are developed separately and are later crossed according to the needs. Cre expressing strains contain a transgene that expresses Cre under the control of a widespread (general) or tissue- specific (conditional) promoter. LoxP-flanked (floxed) strains contain loxP sites flanking (on each side of) a critical portion of a target gene or genomic region of interest. For a gain of

function insertion, such as the expression of ChR or a sensor, the gene interest can be inactivated by a LoxP sequences on each side of stop sequence that will be lost upon recombination (Fig. 35). This method can be efficient but was also shown to be potentially leaky inducing expression in non Cre expressing cells (Sohal et al. 2009). Depending on the promoter, expression of the protein of interest can be slow and/or low and regulated over time. Imaging fluorescent sensors especially those with a low brilliance or expressing sufficient ChR to trigger a neuron’s response is often compromised which such an approach.

The most commonly used strategy to date for the expression of ChR2 in brain tissue is through viral transfer. Viral vectors driving ChR2 expression can be delivered directly into specific brain regions with robust transduction efficacy and limited tissue damage using stereotactic injections. Adeno-associated viruses (AAVs) provide extensive spatial spread and high expression levels (Taymans et al. 2007). Recent studies have used 'double-floxed' inverted open reading frame (DIO) viral vectors to achieve high specificity and expression level of ChR in neurons using AAVs. This system was called FLEX switch (Fig. 36. Atasoy et al. 2008, Sohal et al. 2009).

Figure 36: Principle of a FLEX switch for ChR2mCherry using two different loxP sequences. The sequence of interest is inserted in the viral vector with an inverted orientation and flanked by two different and alternate loxP sites (loxP and lox2272). Cre recombinase will only recombine identical site. FLEX switch recombination sequence for stable inversion proceeds in two steps: (1) inversion followed by (2) excision. (From Atasoy et al. 2008)

Currently used ChR2-AAV vectors are optimized for strong expression. They usually contain a strong promoter such as GAG (CMV enhancer/β-globin chimeric promoter) or EF1α (Elongation factor 1α) and a WPRE element (Woodchuck hepatitis virus post-transcriptional regulatory element) which will optimize the export of unspliced RNA into the cytoplasm (Fig 37, Klein et al. 2006). The viral vector also contains two ITR (inverted terminal repeat) sequences present in the wild type viral genome which are required for efficient multiplication

of the AAV genome, for integration of the AAV ssDNA into the host cell genome and rescue from it and efficient encapsidation of the AAV single stranded DNA. A polyadenylation cassette is usually also inserted (Fig. 37). For best packaging, the genome size should not exceed 5kb (Wu et al. 2010), limiting transgene size to 2 kb.

Figure 37: Double-floxed Cre-dependent AAV vector design for the expression of YFP-ChR2 (Sohal et al. 2009)

The production of recombinant AAV viruses requires triple co-transfection of the ITR and transgene containing plasmid with a plasmid containing replication and capsid sequences and a third plasmid containing helper genes necessary for replication. Various capsid genes can be chosen to fix the serotype of the viral particle, which is critical for the virus tropism. Production of AAV is time consuming as obtaining high titer viral solution additionally requires purification and concentration. According to the use and the needs, the delay for optimal protein expression can range from days to several weeks.

I.4.3.1 Sindbis virus The delay between AAV or lentiviral infection and proper protein expression can be a drawback for certain experiments needing strong expression with a short delay. The size of the transgene can also be limiting for the use of AAV or lentiviruses. In the laboratory we routinely use Sindbis viruses, which belong to the alphavirus family, for neuronal expression of genetically encoded fluorescent sensors in brain slices. The genome of wild type alphaviruses is constituted by a ~12kb ssRNA+ and thus allows subcloning of transgenes up to 4kbs (Schlesinger and Dubensky 1999, Ehrengruber 2002, Marie 2013 ). Sindbis viruses or Semliki viruses, which are viruses from the same family, are of special interest as their production is easy and fast. Sindbis and Semliki viruses have a wide tropism which includes many types of neurons but nor all. For example, recombinant Sindbis viruses barely transduce interneurons or dopaminergic neurons (unpublished results from the laboratory). Recombinant Sindbis viruses are produced from two plasmids. One plasmid contains non-structural proteins such as replicases and the transgene and the other plasmid contains structural protein

sequences for the capsid production (Fig. 38). The virus production needs co-transfection of RNA produced from the two plasmids into baby hamster kidney cells. Viral proteins and RNA will be very rapidly produced and recombinant RNA package into viral particles which bud out of the cells into the culture medium that can be collected and used the next day (see method section).

Figure 38. Generation of recombinant Sindbis viruses. (A) Vector RNA encoding nsP1–4 and the transgene under the control of the subgenomic promoter (broad arrow), and defective helper RNA encoding the structural proteins downstream of the subgenomic promoter are obtained by in vitro transcription and co-transfected into BHK cells. (B) Within the cytoplasm, vector RNA replication occurs through the action of nsP1–4.The helper RNA is also replicated and transcribed by nsP1–4. Capsid proteins package only vector RNA containing the transgene (dashed) while spike proteins are incorporated into the BHK cell membrane. Nucleocapsids dock to the cell membrane where spike proteins have been incorporated, thus allowing the budding of Sindbis particles (C). Upon infection of a neuron, nucleocapsids are released into the cytoplasm and the vector RNA is liberated. The replicase complex composed of nsP1–4 amplifies the vector RNA and the transgene is translated into the recombinant protein (from Ehrengruber 2002).

Transduction of cells using Sindbis viruses results in a rapid and strong expression of the transgene as viral RNA multiplies in the host cell. The major limitation of Sindbis is its toxicity due to the properties of the non structural protein nsp2 to inhibit cellular transcription. Mutations in nsp2 were done to reduce these cytotoxic effects (Kim et al. 2004; Garmashova et al 2006). Sindbis transduction in the brain was also demonstrated to decrease with the age of the animal. Thus, transduction in adult tissue is less efficient than in young animals (Johnson et al. 1972, Chen et al. 2000).

II. RESULTS: CATECHOLAMINERGIC TRANSMISSION IN THE NEOCORTEX The catecholamines noradrenalin (NA) and dopamine (DA) widely modulate the activity and plasticity of brain circuits and their involvement in multiple normal and pathological behaviors is well-established. However, their cellular effects have not been extensively studied. In particular, little is known on neurotransmission events mediated by endogenous catecholamines. This is probably due to the fact that catecholamines modulate neuronal excitability and fast synaptic transmission but do not trigger major membrane currents, which limits the sensitivity of electrophysiological detection of their effects. Conversely, catecholamines exert powerful effects on cAMP/PKA signaling, for which my host laboratory has developed/implemented sensitive imaging tools over the past years for the study of neocortical networks (Gervasi et al. 2007, Drobac et al. 2010, Hu et al. 2011, Bonnot et al. 2014). The general aim of my PhD project was thus to characterize catecholaminergic transmission in the neocortex using PKA imaging as a functional readout resolved in time and space. Catecholamines are released in the neocortex from the axons of locus coeruleus (LC) NA neurons and ventral tegmental area (VTA) DA neurons. LC fibers innervate the entire cortical mantle, whereas VTA fibers exhibit a restricted distribution (see introduction). Nonetheless, several studies indicate that DA and NA can be co-released from LC fibers where DA is present as the biosynthetic precursor of NA (see introduction). I thus also tested the hypothesis that DA and NA can be co-released from LC fibers and can both contribute to catecholaminergic effects in cortical areas outside the VTA projection territory.

In the first part of my PhD, I re-examined the distribution of LC and VTA fibers in the rodent cortex using specific genetic labeling of catecholaminergic neurons. I next characterized the functional impact of Gs- and Gi-coupled DA and NA receptors on cAMP/PKA signaling in cortical regions with scarce VTA fibers. This was achieved using 2-photon imaging of a genetically-encoded PKA sensor expressed via ex-vivo viral transfer in brain slices from young rats, as previously established in my host laboratory (Gervasi et al. 2007, Drobac et al. 2010, Hu et al. 2011, Bonnot et al. 2014). This study is described in section II.1, which incorporates the Nomura et al. manuscript entitled “Noradrenalin and dopamine receptors both control cAMP-PKA signaling throughout the cerebral cortex” that is in revision at Frontiers in Cellular Neuroscience.

In the second part of my PhD, I set the conditions for specific release of catecholamines in neocortical slices using optogenetic tools and a DBH Cre-driver mouse line. To this end, I implemented in vivo viral delivery for combined expression of a PKA sensor in the cortex and

of the light-sensitive channelrhodopsin (ChR2) in the LC, as well as imaging in adult rodent brain slices. I then characterized cAMP/PKA responses elicited by photostimulation of channelrhodopsin-expressing LC fibers using combined 2-photon imaging and fibered illumination in neocortical slices from adult mice. This study is described in section II.2 and will be part of a future publication with my co-authorship.

In parallel, I also took part in characterizing the expression pattern of the GluD1 and GluD2 subunits of the glutamate receptor delta family in the mouse brain. My contribution to this project was to perform in situ hybridization of the GluD1 mRNA in the mouse brain under supervision of Odile Poirel and Salah El Mestikawy. The report of this study (Hepp et al.) Glutamate receptors of the delta family are widely expressed in the adult brain. Brain Struct Funct. In press) in included as an appendix to the present manuscript.

II.1 Summary of the article: “Noradrenalin and dopamine receptors both control cAMP-PKA signaling throughout the cerebral cortex” by Nomura et al. The effects of DA and NA are mediated by G protein-coupled receptors. The five DA receptors belong to the D1/D5 or D2-like classes, which activate or inhibit cAMP/PKA signaling via Gs or Gi proteins, respectively. NA also regulates the cAMP/PKA pathway by activating Gs-coupled beta1-3 receptors or Gi-coupled alpha2 adrenoceptors (see introduction). The broad distribution of NA receptors in the rodent cortex is consistent with that of LC fibers (Nicholas et al. 1996; Venkatesan et al. 1996; Papay et al. 2006; Paschalis et al. 2009). In contrast, a mismatch exists between the widespread expression also reported for DA receptors (Ariano and Sibley 1994; Khan et al. 1998; Luedtke et al. 1999; Lemberger et al. 2007; Rivera et al. 2008; Oda et al. 2010), and the restricted distribution of VTA fibers in the rodent cortex. This study addresses the question of the relative functional impact of DA and NA receptors in cortical regions devoid of VTA fibers, in relation with the reported co- release of DA and NA by LC fibers (Devoto et al. 2005 and 2008).

II.1.1 Distribution of LC and VTA fibers in the cerebral cortex I first re-examined the distribution of LC and VTA fibers in the rodent cortex using genetic labeling, which allows the mapping of neuronal projections with exquisite sensitivity and selectivity. Site and cell-type-specific labeling of corticopetal catecholaminergic neurons was achieved through stereotactic injection of a Cre-dependent adeno-associated virion (AAV) encoding a membrane-bound channelrhodopsin-GFP fluorescent fusion protein that efficiently labels axons in combination with anti-GFP immunohistochemistry. The conditional

AAV was injected in the VTA of DAT-Cre mice to label corticopetal DA neurons and in the LC of DBH-Cre mice to label NA neurons. Site and cell-type-specific transduction of catecholaminergic neurons in the VTA or LC was assessed using anti-GFP and anti-TH immunohistochemistry. The overall distribution of GFP-labeled DA and NA fibers we observed in the mouse cortex is largely congruent with maps obtained in the mouse or rat cortex using other techniques (Morrison et al. 1978 and 1979; Descarries et al. 1987; Berger et al. 1991; Nicholas et al. 1996; Venkatesan et al. 1996; Latsari et al. 2002; Papay et al. 2006; Paschalis et al. 2009). NA fibers were present in all layers and areas of the cortex, although their density exhibited regional and laminar variations, such as an overall lower density in frontal than other regions and a higher density in upper layers of cingulate, somato-sensory, rhinal and visual cortices than in other layers of these regions. Conversely, DA fibers were present in all layers of two medial and lateral rostro-caudal bands comprising frontal, cingulate, rhinal cortices as well as the agranular area of the insular cortex but were confined to deep layers of other cortical areas. The restricted distribution of cortical DA fibers is consistent with the hypothesis that DA in large cortical areas is not released from DA fibers, but from NA fibers (Devoto et al. 2005 and 2008).

II.1.2 NA and DA receptors widely trigger PKA activation in cortical pyramidal neurons I then studied the effects of NA and DA on cAMP/PKA signaling in parasagittal slices of medio-lateral cortex containing scarce DA fibers. This was performed using 2-photon imaging in brain slices of young rats (P15) expressing the fluorescent PKA sensor GAkdYmut (Bonnot et al., 2014) after overnight incubation with a recombinant Sindbis virus. Using this paradigm, the virus transduces pyramidal neurons of layers II/III and V with high efficiency and selectivity and expression does not conspicuously affect membrane or signaling properties of the neurons (Drobac et al., 2010; Hu et al., 2011, Bonnot et al., 2014). I found that NA and DA both activate the cAMP/PKA pathway in layers II/III and V pyramidal neurons throughout the rostro-caudal axis of the neocortex. These results indicate that NA released by LC fibers widely influences the function of neural networks in the entire cortical mantle. The same conclusion applies to DA, regardless of its source. We found that NA responses were generally larger, with faster onsets and exhibited less regional and cell-to-cell variability than DA responses. Furthermore, measurements of EC50s for PKA activation revealed that NA effects occurred at ~20 fold lower dose than DA effects.

II.1.3 Gs-coupled receptors NA-induced PKA activation was mimicked by the beta agonist isoproterenol and primarily mediated by beta 1 receptors, as shown by the powerful inhibitory effect of the specific antagonist CGP20712 on NA responses. These results are consistent with the predominant expression of beta1 over beta2-3 receptors in the cortex and with the widespread expression of beta1 receptors in layers and areas of the rodent cortex (Nicholas et al. 1996; Paschalis et al. 2009). DA effects were similar in layer II/III and layer V, but both the amplitude and velocity of DA responses increased from rostral to caudal regions. DA responses were mimicked by the D1/D5 agonist SKF38393 and prevented by the D1/D5 receptor antagonist SCH23390, pointing to the involvement of D1 and/or D5 receptors in DA-induced PKA activation. Regional and laminar differences of D1 and D5 expression levels have been reported in the rodent cortex (Ariano et al. 1994; Ciliax et al. 2000). However, the lack of specific pharmacological tools differentiating these receptors makes it difficult to assign the observed functional variations to the regional levels of these receptors. The possibility that cortical D1/D5 receptors forms heteromers with beta1 adrenoceptors further complicates the interpretation of regional variations of DA responses. As mentioned above, the beta blocker propranolol and the beta1 antagonist CGP20712 both inhibited cortical NA responses. Surprisingly, these antagonists also prevented DA responses and largely reduced SKF38393 responses suggesting that DA-induced PKA activation in the cortex may involve beta1 adrenergic receptors. DA receptors are known to form heteromers with a variety of receptors including beta adrenoceptors (Rebois et al. 2012; Fuxe et al. 2014). Moreover, cross antagonism by specific antagonist has been demonstrated for the beta1-D4 receptor heteromer (Gonzales et al. 2012). Hence, D1/D5 receptors may form heteromers with beta1 adrenoceptors in the cortex, thereby exhibiting the specific pharmacological pattern we observed.

II.1.4 Gi-coupled receptors Block of alpha2 adrenoceptors by yohimbin and of D2-like receptors by haloperidol increased the amplitude and velocity of NA and DA responses. This is consistent with an inhibitory effect of these Gi-coupled receptors on cAMP/PKA signaling. Hence, DA and NA exert a balanced control on cortical cAMP/PKA signaling by activating both Gs- and Gi-coupled receptors. This is consistent with previous reports showing widespread immunoreactivity for alpha2 and D4 receptors in the cortex (Venkatesan et al. 1996; Khan et al. 1998; Rivera et al. 2008) and for D2 receptors in pyramidal cells of the medial prefrontal cortex (Zhang et al.

2010). In the case of DA, the effect of Gi-coupled receptors was so powerful that their block by haloperidol largely increased the mean response to DA and decreased the proportion of cells showing low or no responses to DA. The EC50 values we measured were ~20 fold higher for DA than for NA and it is known that NA and DA Gi-coupled receptors exhibit higher affinities than their cognate Gs-coupled receptors (for a detailed comparison of pharmacological properties of these receptors, see the International Union of Basic and Clinical Pharmacology database: http://www.guidetopharmacology.org/). This suggests that, at low catecholamine concentration, DA may essentially trigger cAMP/PKA inhibition whereas NA effects may be more balanced. Conversely, large catecholamine increases may be required for a significant contribution of DA to cAMP/PKA stimulation. Interestingly, DA levels in basal conditions or upon LC stimulation are comparable to those of NA in various cortical regions, but markedly lower than DA levels in basal ganglia (Devoto 2005 and 2008). Assuming co-release of NA and DA from LC fibers in the cortex, DA may thus expand the dynamic range of LC effects on cortical cAMP/PKA signaling.

Published article: Nomura et al. 2014, Frontiers in Cellular Neuroscience, in press

ORIGINAL RESEARCH ARTICLE published: 21 August 2014 doi: 10.3389/fncel.2014.00247 Noradrenalin and dopamine receptors both control cAMP-PKA signaling throughout the cerebral cortex

Shinobu Nomura1,2,3 , Maud Bouhadana1,2,3 , Carole Morel 1,2,3 , Philippe Faure1,2,3 , Bruno Cauli 1,2,3 , Bertrand Lambolez 1,2,3 and Régine Hepp1,2,3 * 1 Sorbonne Universités, UPMC Université Paris 06, UM CR 18, Neuroscience Paris Seine, Paris, France 2 Centre National de la Recherche Scientiﬁque (CNRS), UMR 8246, Paris, France 3 Institut National de la Santé et de la Recherche Médicale (INSERM), U 1130, Paris, France

Edited by: Noradrenergic fibers innervate the entire cerebral cortex, whereas the cortical distribution Nicholas C. Spitzer, University of of dopaminergic fibers is more restricted. However, the relative functional impact of California, San Diego, USA noradrenalin and dopamine receptors in various cortical regions is largely unknown. Using Reviewed by: a specific genetic label, we first confirmed that noradrenergic fibers innervate the entire Ursula Felderhoff-Müser, University of Duisburg-Essen, Germany cortex whereas dopaminergic fibers were present in all layers of restricted medial and De-Lai Qiu, YanBian University, China lateral areas but only in deep layers of other areas. Imaging of a genetically encoded sensor *Correspondence: revealed that noradrenalin and dopamine widely activate PKA in cortical pyramidal neurons Régine Hepp, Sorbonne Universités, of frontal, parietal and occipital regions with scarce dopaminergic fibers. Responses to UPMC Université Paris 06, UM CR 18, noradrenalin had higher amplitude, velocity and occurred at more than 10-fold lower dose Neuroscience Paris Seine, 9 Quai Saint Bernard, Case Courrier 16, than those elicited by dopamine, whose amplitude and velocity increased along the antero- 75005 Paris, France posterior axis.The pharmacology of these responses was consistent with the involvement e-mail: [email protected] of Gs-coupled beta1 adrenergic and D1/D5 dopaminergic receptors, but the inhibition of both noradrenalin and dopamine responses by beta adrenergic antagonists was suggestive of the existence of beta1-D1/D5 heteromeric receptors. Responses also involved Gi- coupled alpha2 adrenergic and D2-like dopaminergic receptors that markedly reduced their amplitude and velocity and contributed to their cell-to-cell heterogeneity. Our results reveal that noradrenalin and dopamine receptors both control cAMP-PKA signaling throughout the cerebral cortex with moderate regional and laminar differences. These receptors can thus mediate widespread effects of both catecholamines, which are reportedly co-released by cortical noradrenergic fibers beyond the territory of dopaminergic fibers.

Keywords: catecholamines, GPCR, protein kinase A, imaging, cerebral cortex

INTRODUCTION the cerebral cortex stem from DA neurons of the ventral tegmen- The catecholamines dopamine (DA) and noradrenalin (NA) are tal area (VTA) and NA neurons of the locus coeruleus (LC). LC neurotransmitters that widely modulate brain circuits and behav- fibers innervate the entire cortical mantle, whereas VTA fibers dis- iors. NA is involved in arousal, attention, memory, and stress tribute in all layers of medio-frontal and ventro-lateral cortices, whereas DA is implicated in learning, reward, attention, and but are restricted to deep layers in other areas of the rodent cor- movement control. Likewise, catecholaminergic dysfunctions are tex (Morrison et al., 1978, 1979; Descarries et al., 1987; Berger associated with cognitive, emotional, and motor disorders and et al., 1991; Latsari et al., 2002). The broad distribution of NA catecholaminergic transmission is the target of multiple drugs receptors in the rodent cortex is consistent with that of LC fibers used in therapy of human brain disorders. Catecholamines are (Nicholas et al., 1996; Venkatesan et al., 1996; Papay et al., 2006; synthesized in discrete brainstem nuclei via a common pathway Paschalis et al., 2009). In contrast, a mismatch exists between the involving tyrosine hydroxylase (TH) that leads to DA production, widespread expression also reported for DA receptors (Ariano and which is converted to NA by DA beta hydroxylase (DBH). The Sibley, 1994; Khan et al., 1998; Luedtke et al., 1999; Lemberger effects of DA and NA are mediated by G protein-coupled recep- et al., 2007; Rivera et al., 2008; Oda et al., 2010), and the restricted tors. The five DA receptors belong to the D1/D5 or D2-like classes, distribution of VTA fibers in the rodent cortex. While LC fibers are which activate or inhibit cAMP/protein kinase A (PKA) signal- a plausible source of cortical DA outside the VTA projection areas ing via Gs or Gi proteins, respectively (Beaulieu and Gainetdinov, (Devoto et al., 2005, 2008), the question of the relative functional 2011). NA also regulates the cAMP/PKA pathway by activating Gs- impact of DA and NA receptors in various cortical regions has not coupled beta1–3 receptors or Gi-coupled alpha2 adrenoceptors. been addressed. NA additionally activates the phospholipase C pathway through In the present study, we first examined the distribution of Gq-coupled alpha1 receptors (Bylund, 1992; Cotecchia, 2010; LC and VTA fibers in the rodent cortex using site and cell- Evans et al., 2010). type-specific labeling of catecholaminergic neurons with green Dopamine- and NA-containing fibers exhibit wide, but distinc- fluorescent protein (GFP) via conditional viral transfer. We then tive, distributions in the brain. Catecholaminergic projections to characterized the functional impact of Gs- and Gi-coupled DA

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 1 Nomura et al. Catecholaminergic control of cortical cAMP/PKA

and NA receptors on cAMP/PKA signaling in layers II/III and with PBS complemented with 0.2% fish skin gelatin, 0.25% tri- V of the frontal, parietal, and occipital cortex using 2-photon ton X-100 (PBS-GT) for2hatroomtemperature.Brainslices imaging of a genetically encoded PKA sensor in rat brain slices were incubated overnight at 4◦C with primary antibody diluted (Bonnot et al., 2014). Our results confirm the differential dis- in PBS-GT (monoclonal mouse anti TH: MAB318 (Clone LNC1), tribution of LC and VTA fibers in the cortex and reveal that Millipore, 1/2000; polyclonal chicken anti-GFP: GFP-1020 (Aves both NA and DA receptors control cAMP-PKA signaling through- Labs, 1/2000). After 6 ×10 min washes in PBS-GT, slices were out the cerebral cortex with moderate regional and laminar incubated for3hatroomtemperaturewithfluorescentlylabeled differences. secondary antibody diluted at 1/1000 in PBS-GT (goat anti chicken alexa488, Invitrogen A11039; goat anti mouse Alexa555, Invitro- MATERIALS AND METHODS gen A21422). Slices were mounted in fluoromount (Clinisciences) ANIMALS after extensive washes in PBS-GT followed by PBS. Images were All the experiments were performed according to the guide- obtained using an AXIO Zoom.V16 macroscope (Zeiss). Labeled lines of the French Ministry of Agriculture, Food Processing fibers were drawn in black (value of the pixel: 255) on white back- Industry and Forestry for handling animals (decree 2013-118). ground (value of the pixel: 0) using Image J and Photoshop Cs2 Transgenic DBH-Cre mice (DBH-cre) were a gift from Bruno (Adobe). The density of fibers was obtained from drawings using Giros [McGill University, Canada, MMRRC line: Tg(Dbh-cre) the line plot profile tool of ImageJ on lanes of 200 pixels (260 μm) KH212Gsat/Mmucd, stock number 032081-UCD (Gong et al., width covering the vertical extent of the cortex from pia to white 2007)]. DA transporter-Cre mice (DAT-cre) were a gift from Uwe matter. The fiber density corresponds to the averaged value of Maskos [Institut Pasteur, France, (Tg)BAC-DATiCrefto (Turiault black and white pixels in a 200 pixel line projected on the vertical et al., 2007)]. Male Wistar rats (12–15 days old) were obtained axis along the lane. from Janvier Labs. Animals were maintained in a 12 h light–12 h dark cycle, in stable conditions of temperature (22◦C), with food BRAIN SLICE PREPARATION AND VIRAL TRANSDUCTION FOR IMAGING and water available ad libitum. PKA ACTIVITY Rats were killed by decapitation. Brains were quickly removed and AAV PRODUCTION AND STEREOTACTIC INJECTION immersed in ice-cold artificial cerebrospinal fluid (ACSF) con- Site- and cell-type-specific labeling of NA or DA neurons was taining (in mM): 126 NaCl, 2.5 KCl, 1.25 NaH2PO4,2CaCl2,1 achieved by stereotactic injection of a viral vector into the LC of MgCl2, 26 NaHCO3,20D-glucose, 5 Na pyruvate, 1 kynurenic DBH-Cre or the VTA of DAT-Cre mice (1–3 month-old). The viral acid, and saturated with 5% CO2/95% O2. Parasagittal slices ◦ vector was a recombinant adeno-associated virus (AAV) driving (300 μm thick) of cortex were cut at an angle of 10 using a Leica Cre-dependent expression of a fusion protein containing channel- VT 1000S Vibratome (Leica). Slices were kept at room tempera- rhodopsin 2 (ChR2) and a yellow variant (YFP) of the GFP from ture for 30 min in the same solution. Brain slices were placed onto Aequorea victoria. The Cre-inducible vector AAV2/1-EF1a-DIO- a millicell-CM membrane (Millipore) with culture medium (50% hChR2(H134R)-EYFP-WPRE-HGHpA (titer: 3 × 1011 gc/ml) was minimum essential medium, 50% Hanks’ balanced salt solution, produced from Addgene plasmid #20298 at the vector core facil- 6.5 g/L glucose, and 100 U/ml penicillin/100 μg/ml streptomycin; ity of Nantes University (UMR 1089 IRT1, France). Aliquots of Invitrogen). Transduction was performed by adding ∼5 × 105 the pseudovirion were stored at −80◦C before stereotactic injec- particles per slice of sindbis virus encoding the GAkdYmut sen- tion. Mice (six DAT-Cre and six DBH-Cre) were anesthetized sor that reports PKA activation through a reversible increase of with isoflurane and placed on a small animal stereotactic frame. fluorescence intensity (Bonnot et al., 2014). Slices were incubated ◦ For transduction of LC NA neurons, AAV-EF1α-DIO-ChR2- overnight at 35 Cin5%CO2. The next morning, brain slices YFP pseudovirion was bilaterally injected adjacent to the LC were equilibrated in ACSF for 1 h and then placed into the record- at coordinates from bregma: antero-posterior (AP), −5.45 mm; ing chamber and perfused continuously at 2 ml/min with ACSF at ◦ medio-lateral (ML), ±1 mm; dorso-ventral (DV) −3.65 mm). 32 C. Under these conditions, sindbis viral transduction efficiently For DA neuron transduction, pseudovirion was bilaterally injected and selectively targets cortical pyramidal neurons, and leaves their into the VTA (AP: −3.4 mm, ML: ±0.5 mm, DV: −4.4 mm). functional properties essentially unaffected (Drobac et al., 2010; Injections were performed through an internal canula at a rate Hu et al., 2011). of 0.1 μl/min for 10 min (total 1 μl per site). The canula was held in place for 15 min before retraction out of the OPTICAL RECORDINGS brain. Two-photon images were obtained with a custom-built 2-photon laser scanning microscope as described (Bonnot et al., 2014), IMMUNOHISTOCHEMISTRY based on an Olympus BX51WI upright microscope (Olympus, Eight weeks after injection, mice were deeply anesthetized with Tokyo, Japan) with ×40 (0.8 NA) or ×60 (0.9 NA) water- 10 mg/ml ketamine and 0.1% xylazine before transcardiac perfu- immersion objectives and a titanium:sapphire laser (MaiTai HP; sion with 50 ml of 4% paraformaldehyde in 0.12 mM phosphate Spectra Physics, Ellicot City, MD, USA). Two-photon excitation buffer (pH 7.4). Brains were removed and post-fixed with the was performed at 920 nm for GFP. Images were acquired as z same solution for 2 h, cryoprotected with 30% sucrose and cut stacks and analyzed using ImageJ (U.S. National Institutes of with a freezing microtome (Leica) at a thickness of 40 μm. Slices Health, Bethesda, MD, USA; http://rsbweb.nih.gov/ij/). Occa- were washed overnight with PBS, then blocked and permeabilized sional x, y, and z drifts were corrected using custom macros

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 2 Nomura et al. Catecholaminergic control of cortical cAMP/PKA

developed from ImageJ plugins TurboReg, StackReg (Thevenaz (Figures 1B,C). Quantification of NA fibers (see Materials and et al., 1998), MultiStackReg, and Image CorrelationJ (Chinga and Methods) revealed moderate laminar and regional differences, Syverud, 2007). Fluorescence variations were measured at the such as an overall lower density in frontal than other regions and soma of pyramidal neurons. Fluorescence intensity of regions of a higher density in upper layers in cingulate, somatosensory, rhi- interest (ROIs) was calculated for each time point from average nal, and visual than in other layers of these regions (Figure 2). intensity projection of 3–5 frames by averaging pixel inten- Similar results were obtained on six LC-injected DBH-Cre mice. sity. Variations of fluorescence intensity in a given ROI were DA fibers had a more contrasted distribution. The highest den- expressed as the ratio F/F0 and calculated according to the sity of fibers was observed in a medial rostro-caudal band that formula (F − F0)/F0. F corresponds to the fluorescence inten- comprised all layers of the prelimbic cortex and extended cau- sity in the ROI at a given time point, and F0 corresponds to dally to the cingulated cortex, but did not reach the retrosplenial the mean fluorescence intensity in the same ROI during control occipital area (Figures 1D and 2). A lateral rostro-caudal band baseline prior to drug application. Pseudocolor hue saturation of lower fiber density comprised all layers of the lateral part of value (HSV) encoding of fluorescence intensity was performed the frontal association cortex, of the agranular insular cortex and using IGOR Pro (WaveMetrics) custom procedures. Pseudocolor of the rhinal occipital area. Between these two bands, DA fibers images were obtained by dividing, pixel-by-pixel a raw fluo- were restricted to deep layers in a large region extending from the rescence image F by the F0 image averaged over several time medial part of the frontal association to the visual cortex through points prior to drug application. Color coding displays the ratio the somatosensory area (Figures 1D and 2). Similar results were F/F0 (in hue) and the fluorescence F (in value). The EC50 obtained on six VTA-injected DAT-Cre mice. The striatum con- for NA and DA were obtained by fitting the values with the tained a dense network of DA fibers but was sparsely innervated Hill equation using IgorPro6. The same equation was used to by NA fibers (Figures 1C,D), as expected from their known dis- determine the 10–90% rise time of the agonist induced PKA tributions (Aston-Jones, 2004; Gerfen, 2004). These data confirm activation. that NA fibers widely innervate the cortical mantle, whereas the distribution of DA fibers is more restricted and exhibits area and DRUGS layer specificity. All drugs were bath applied. SKF38393, SCH23390; Propranolol, Yohimbin, CGP20712, and forskolin were purchased from Tocris. NA AND DA RECEPTORS ARE CO-EXPRESSED AND MEDIATE PKA NA, DA, haloperidol, and isoproterenol were from Sigma-Aldrich. ACTIVATION IN CORTICAL PYRAMIDAL NEURONS We examined responses of cortical neurons to NA, DA or spe- STATISTICS cific receptor agonists in frontal, parietal, and occipital areas In this report, N represents the number of animals or brain slices where DA fibers are restricted to deep layers (i.e., lateral frontal, tested while n represents the number of cells on which measure- somato-sensory, and visual cortices; see above). We used 2-photon ments were performed. Values are expressed as mean ± SEM. imaging to monitor responses of layers II/III and V pyramidal Statistical significance was assessed with Student’s t-test. p value neurons expressing the PKA sensor GAkdYmut (Bonnot et al., < 0.05 was considered statistically significant. 2014) following sindbis viral transfer in acute rat brain slices (see Materials and Methods and Drobac et al., 2010; Hu et al., RESULTS 2011). Response amplitudes measured at the soma were normal- DISTRIBUTION OF LC AND VTA FIBERS IN THE CEREBRAL CORTEX ized to those elicited by a maximally effective concentration of In order to compare LC and VTA projections to the rodent cortex, the adenylate cyclase activator forskolin (FSK, 12 μM; Gervasi we selectively expressed GFP in DA or NA neurons using Cre- et al., 2007). As illustrated in Figure 3, successive bath applica- dependent viral transduction (see Materials and Methods). DBH- tion of DA (10 μM) and NA (10 μM) triggered PKA activation Cre mice (N = 6) and DAT-Cre mice (N = 6) were injected in the in most pyramidal neurons of the parietal cortex, showing that LC or in the VTA, respectively, with a recombinant pseudovirion DA and NA receptors can be co-expressed in individual neu- driving GFP expression at the membrane of Cre-positive neurons. rons. To estimate the proportion of responsive cells, we arbitrarily Anti-GFP and anti-TH immunohistochemistry showed that GFP defined a threshold at 5% of the maximal fluorescence variation was widely expressed in the LC of DBH-Cre mice and in the VTA obtained with FSK. Using this criterion, we found that virtually and substantia nigra of DAT-Cre mice following viral transduction all pyramidal neurons responded to NA (n = 45, N = 2) and (Figure 1A). In LC-injected DBH-Cre mice, 83.9 ± 2.7% (n = 227; among those, 83 ± 4% responded to DA (39 out of 45 cells). N = 5) of TH-positive neurons in the LC expressed GFP.Similarly, These effects were mimicked by bath application of the D1/D5 86.1 ± 1.2% (n = 158; N = 3) of TH-positive neurons in the VTA agonist SKF38393 (1 μM) or the beta adrenoceptor agonist iso- co-expressed GFP. Conversely, LC neurons of VTA-injected DAT- proterenol (1 μM; Figure 3). We found that 97% of the cells Cre mice and VTA neurons of LC-injected DBH-Cre mice were (57 out of 59) responded to isoproterenol and 75% to SKF (44 GFP-negative (Figure 1A). These results show that the present out of 59), suggesting that most pyramidal neurons co-express protocol of conditional viral transduction enabled efficient site- D1/D5 and beta-adrenergic receptors. Dose–response relation- and cell-type-specific GFP expression in LC NA neurons or VTA ships of normalized NA and DA effects showed that NA was DA neurons. We next examined the distribution of NA and DA more potent than DA in activating PKA (Figure 3). Indeed, the fibers in the cerebral cortex. GFP labeling revealed a high density EC50 of NA responses was estimated at 25 ± 2 nM and that of of NA fibers throughout all cortical regions and layers examined DA at 460 ± 30 nM from the fit of their dose–response curves

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 3 Nomura et al. Catecholaminergic control of cortical cAMP/PKA

FIGURE 1 | Catecholaminergic projections to the cortex. (A) Immuno- immunolabeling of GFP positive fibers in coronal brain sections of LC-injected labeling against TH and GFP in the LC and VTA 8 weeks after bilateral injection DBH-Cre and VTA-injected DAT-Cre mice (antero-posterior from left to right). of a Cre-dependent AAV expressing GFP in the LC of DBH-Cre mice or the Note the different distribution of NA and DA fibers in cortical areas and VTA of DAT-Cre mice. Note that GFP labeling in VTA-injected DAT-Cre mice caudate-putamen (CPu). Ai: agranular insular cortex; Au: auditory cortex; Cg: extends to neighbor DA neurons of the substantia nigra pars compacta (SNc) cingulate cortex; DLO: dorso-lateral orbital cortex; FrA: frontal association but is absent from NA neurons of the LC. (B) Immunolabeling of GFP- cortex; M1-M2-M: primary or secondary motor cortex; MO: medial orbital expressing fibers in the somato-sensory parietal cortex of DBH-Cre and cortex; Pir: piriform cortex; PrL: prelimbic cortex; Rh: rhinal cortex; RS: DAT-Cre mice. (C,D) Drawings of NA or DA fibers obtained from retrosplenial cortex; S1: primary somatosensory cortex; V: visual cortex.

(see Materials and Methods). Furthermore, maximal responses to occipital cortices (Figure 4 and Table 1). The mean amplitude NA and DA were 60 ± 1% and 47 ± 2% of the FSK response, of NA responses varied between 54 ± 4% in occipital layer V respectively. (n = 43; N = 4) and 75 ± 4% of the FSK effect in frontal layer II/III (n = 37, N = 2). Isoproterenol-induced responses varied NA AND DA ACTIVATE PKA ACROSS CORTICAL REGIONS AND LAYERS similarly between 44 ± 3% in frontal layer V (n = 50, N = 4) We next compared PKA activation elicited by DA, NA, SKF38393, and 56 ± 2% of the FSK effect in frontal layer II/III (n = 76, and isoproterenol in layers II/III and V of frontal, parietal, and N = 4). DA and SKF38393 both caused PKA activation in all

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 4 Nomura et al. Catecholaminergic control of cortical cAMP/PKA

FIGURE 2 | Density of noradrenergic and dopaminergic fibers in various along the lane plotted on the x-axis of the graphs from the pial surface to areas of the cortex. The density of fibers was obtained from the drawings indicated depths. NA fibers are widespread in the cortex and are enriched shown in Figure 1 on lanes of 200 pixels (260 μm) width orthogonal to the (arrows) in upper layers of cingulated (Cg), somato-sensory (S), visual (V), and pial surface, as depicted in images of the upper panel (gray bars). The fiber rhinal (Rh) cortices. Note that DA fibers are absent from most layers of a density, measured in three regions of anterior (A), intermediate (B), and rostro-caudal band comprising the medio-lateral part of the frontal association posterieor (C) coronal brain sections, corresponds to the averaged pixel value (FrA), S and V cortices, except for a thin lamina in deep layers.

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 5 Nomura et al. Catecholaminergic control of cortical cAMP/PKA

FIGURE 3 | Continued NA and DA activate the cAMP/PKA pathway in pyramidal neurons. 2-photon imaging of pyramidal neurons expressing the GAkdYmut fluorescent PKA sensor in parietal cortical slices. (A) Responses to bath application of DA, NA, and the adenylate cyclase activator FSK. Traces show the variation of fluorescence intensity measured at the soma of individual neurons indicated by arrows on pseudocolor images below. Black arrowheads indicate time points corresponding to pseudocolor images. (B) Responses to the D1/D5 agonist SKF38393 and the non-specific beta adrenergic agonist isoproterenol. (C) Dose–response curves of PKA activation by NA and DA. Each data point corresponds to the mean response measured in 10–140 individual neurons from 2 to 10 different brain slices. Curves were obtained by fitting the values with the Hill equation. NA was more potent than DA in activating PKA.

areas examined but responses were smaller than those elicited by NA or isoproterenol (Figure 4A and Table 1). DA responses were signiﬁcantly smaller in both layers of the frontal cortex than in more caudal regions. Indeed, DA responses in layer II/III were 46 ± 4% in occipital (n = 38, N = 2) and 33 ± 3% of the FSK effect in the frontal cortex (n = 39, N = 3, p = 0.006). Simi- larly, DA responses in layer V were 44 ± 4% in occipital (n = 37, N = 3) and 29 ± 2% of the FSK effect in the frontal cortex (n = 44, N = 3, p = 0.01). Similar results were obtained with the D1/D5 agonist SKF38393, although amplitudes were smaller than those obtained with DA (Figure 4A and Table 1, p = 0.04 in layer II/III and p = 0.02 in layer V). We also observed that less than 5% of the recorded pyramidal neurons were unresponsive to NA or isoproterenol, regardless of the layer or cortical region examined. For DA, a similar result was obtained in layer II/III. However, the proportion of unresponsive cells was higher in layer V and reached 12% in the parietal cortex. For SFK38393, the proportion of unresponsive cells was above 5% in all cases with more unresponsive cells in layer V than in layer II/III. The highest value was observed in layer V of the parietal cortex with 27% of recorded neurons not responding to SFK38393. These results indicate that NA and isoproterenol induce overall larger and more homogeneous cortical responses than DA and SKF38393, whose effects increased along the antero-posterior axis. We also examined the onset kinetics of PKA responses by determining their 10–90% rise time (Figure 4B and Table 1). Kinetics of NA and isoproterenol responses exhibited moderate regional differences. Conversely, DA and SKF38393 response onsets were faster in occipital than in frontal regions with intermediate values in the parietal cortex. This was observed in both layers II/III and V (0.01 < p ≤ 0.003 for DA, p < 0.001 for SKF). As a consequence of these regional variations, the onset kinetics of DA-receptor mediated responses were markedly slower than those of NA-receptor mediated responses in the frontal cortex, but were comparable in the occipital cortex. These results conﬁrm that NA and isoproterenol induce overall more homogeneous cortical responses and indicate that DA and SKF38393 effects increase in amplitude and velocity along the antero-posterior axis.

CORTICAL EFFECTS OF NA AND DA INVOLVE BOTH Gs- AND Gi-COUPLED RECEPTORS Gs-coupled NA and DA receptors involved in PKA activation were FIGURE 3 | Continued characterized in parieto-cortical layer II/III pyramidal neurons

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 6 Nomura et al. Catecholaminergic control of cortical cAMP/PKA

FIGURE 4 | NA and DA activate PKA across cortical regions and layers. variations normalized to the response to FSK. (B) Onset kinetics of responses Data were obtained by 2-photon imaging of PKA activity in pyramidal neurons was analyzed by calculating the 10–90% rise time. Tables indicate significant as illustrated in Figure 3. Imaging was performed in frontal, parietal, and (s) or non-significant (ns) differences between values measured in the areas occipital regions of parasagittal slices of the medio-lateral cortex that exhibits (F-P: frontal vs. parietal; P-O: parietal vs. occipital; F-O: frontal vs. occipital). scarce DA fibers restricted to deep layers. (A) Amplitude of fluorescence Note the antero-posterior gradient of DA responses amplitude and velocity.

Table 1 | NA and DA activate PKA across cortical region and layers.

Frontal Parietal Occipital

Drugs (n/N) Layer Amplitude (% FSK) T10–90% (s) Amplitude (% FSK) T10–90% (s) Amplitude (% FSK) T10–90% (s)

DA II/III 32.5 ± 3.3 (39/3) 109.9 ± 14.1 (22/3) 49.5 ± 1.8 (242/12) 110.5 ± 4.8 (181/12) 46.4 ± 3.9 (38/2) 61.7 ± 7.1 (27/2) V 28.5 ± 2.4 (44/3) 170.0 ± 13.1 (30/3) 33.7 ± 3.1 (105/6) 143.4 ± 11.9 (53/6) 44.1 ± 3.8 (37/3) 119.3 ± 14.0 (37/3) NA II/III 74.5 ± 3.7 (37/2) 57.8 ± 3.0 (32/2) 61.9 ± 2.9 (114/7) 74.3 ± 2.0 (84/7) 61.3 ± 2.5 (77/2) 53.3 ± 2.8 (62/2) V57.6± 3.5 (26/3) 90.2 ± 4.9 (21/3) 72.5 ± 5.0 (29/2) 48.8 ± 2.8 (26/2) 53.9 ± 3.5 (43/4) 68.4 ± 3.2 (35/4) SKF II/III 21.1 ± 3.4 (31/4) 147.4 ± 9.9 (39/4) 19.9 ± 1.4 (196/10) 134.9 ± 7.9 (81/10) 34.5 ± 2.6 (40/4) 55.2 ± 3.4 (31/3) V 22.0 ± 3.6 (36/4) 192.4 ± 20.2 (24/3) 31.4 ± 3.8 (60/4) 106.3 ± 8.9 (32/4) 32.9 ± 2.8 (69/5) 64.6 ± 5.9 (28/3) ISO II/III 55.8 ± 2.0 (76/4) 69.2 ± 3.5 (70/4) 48.1 ± 1.5 (172/7) 45.7 ± 1.2 (98/7) 52.3 ± 3.1 (40/3) 47.5 ± 2.5 (32/3) V 43.5 ± 2.5 (50/4) 54.0 ± 3.8 (46/4) 48.9 ± 3.8 (60/4) 65.9 ± 4.1 (47/4) 46.8 ± 2.1 (135/5) 62.7 ± 4.3 (50/5)

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 7 Nomura et al. Catecholaminergic control of cortical cAMP/PKA

using specific antagonists (Figure 5 and Table 2). The non- receptors. Responses to DA (10 μM) and to SKF38393 (1 μM) selective beta-adrenergic receptor antagonist propranolol (50 μM, were almost abolished in the presence of the D1/D5 receptor N = 2, n = 22) and the specific beta1 antagonist CGP20712 antagonist SCH23390 (1 μM; Table 2). These results indicate (100 nM, N = 3, n = 40) reduced PKA activation by NA (10 μM) that DA-induced PKA activation in cortical pyramidal neurons by more than 90% (Figure 5). Responses to isoproterenol (1 μM) is mediated by D1/D5 receptors. NA and isoproterenol responses were also reduced by propranolol to a similar extent (Table 2). were not significantly altered by the D1/D5 antagonist SCH23390 These results indicate that NA-induced PKA activation in corti- (1 μM), showing that NA effects did not involve DA receptors. cal pyramidal neurons is primarily mediated by beta1-adrenergic Surprisingly, we found that responses to DA and SKF38393 were

FIGURE 5 | NA and DA receptors involved in PKA activation. Graphs show isoproterenol and the D1/D5 agonist SKF38393. ND: not determined; but the effects of antagonists propranolol (beta adrenergic), CGP20712 Castro et al. (2010) reported inhibition of isoproterenol responses in cortical (beta1-specific) and SCH23390 (D1/D5) on responses to NA and DA of layer pyramidal neurons by CGP20712 in similar experimental conditions. Values II/III pyramidal neurons of the parietal cortex. Gray traces correspond to the represent the mean response of 20–200 neurons recorded in 2–12 different fluorescence intensity normalized to the baseline value measured on brain slices. Statistical significance relative to control responses is indicated individual neurons expressing the PKA sensor in parietal cortical slices by an asterisk. Note that DA responses were prevented by both beta (averaged traces are in black). Histograms summarize the effects of these adrenergic antagonists whereas NA responses were insensitive to the D1/D5 antagonists on responses to NA, DA, the beta adrenergic agonist antagonist.

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 8 Nomura et al. Catecholaminergic control of cortical cAMP/PKA

Table 2 | PKA stimulation involves beta1 adrenergic and D1/D5 dopaminergic receptors.

NA Amplitude (% FSK) ISO Amplitude (% FSK) DA Amplitude (% FSK) SKF Amplitude (% FSK)

Control (n/N) 61.9 ± 2.9 (114/7) 48.1 ± 1.5 (172/7) 49.5 ± 1.8 (242/12) 19.9 ± 1.4 (196/10) SCH23390 (n/N) 51.8 ± 3.9 (56/2) 60.5 ± 2.3 (66/2) 3.5 ± 1.4 (56/2) 4.0 ± 0.1(49/4) Propranolol (n/N) 6.0 ± 0.7 (22/2) 4.1 ± 0.8 (21/2) 9.7 ± 1.3 (57/2) 8.5 ± 2.1(24/2) CGP20712 (n/N) 8.8 ± 1.9 (40/3) ND 11.0 ± 2.7 (54/2) 11.3 ± 1.2 (101/3)

inhibited by beta adrenergic antagonists. Indeed, propranolol and medial and lateral cortical areas but only in deep layers of other CGP20712 similarly reduced responses to DA (10 μM) by ∼80% areas. Imaging of PKA activity in frontal, parietal, and occipital and responses to SKF38393 (1 μM) by ∼50% (Figure 4 and regions with scarce DA fibers revealed that layers II/III and V pyra- Table 2), suggesting that DA effects may involve beta1-D1/D5 midal neurons widely respond to NA and DA. The EC50 of NA heteromeric receptors. response was more than 10-fold lower than for DA. Responses to We next examined the contribution of Gi-coupled receptors NA and isoproterenol had higher amplitude, velocity and were in the response of parieto-cortical layer II/III pyramidal neurons more homogeneous than those elicited by DA and SKF38393, to NA and DA (Figure 6 and Table 3). Inhibition of the alpha2 whose amplitude and velocity increased along the antero-posterior adrenoceptor by yohimbin (1 μM) significantly increased the axis. NA- and DA-induced PKA activation was mediated by Gs- mean amplitude of the NA response by 17% (p = 0.007). Similarly, coupled beta1 and D1/D5 receptors. Beta adrenergic antagonists the D2-like receptors antagonist haloperidol (1 μM) significantly inhibited both NA and DA responses whereas the effect of a increased the DA response by 30% (p < 0.001). Haloperidol and D1/D5 antagonist was selective of DA responses. NA and DA yohimbin also increased the velocity of DA and NA responses. responses also involved Gi-coupled alpha2 adrenergic and D2- Indeed, in the presence of haloperidol, the 10–90% rise time of like dopaminergic receptors that markedly reduced the amplitude DA responses decreased significantly from 111 ± 5s(n = 181, and velocity of responses and contributed to their cell-to-cell N = 12) to 76 ± 3s(n = 84, N = 4, p < 0.001). Yohimbin appli- heterogeneity. cation similarly resulted in a decreased rise time of NA responses from 74 ± 2(n = 84, N = 7) to 47 ± 2s(n = 119, N = 5; DISTRIBUTION OF CATECHOLAMINERGIC FIBERS IN THE CORTEX p < 0.001; Figure 6 and Table 3). We previously noted a large het- The overall distribution of GFP-labeled DA and NA fibers in erogeneity in the amplitudes of DA responses (Figures 3 and 5). the cortex we describe is largely congruent with previous maps We thus examined whether antagonists of Gi-coupled receptors obtained using histochemistry, radiolabeling or immunohisto- could modify the distribution of DA and NA response amplitudes chemical approaches (Morrison et al., 1978, 1979; Descarries et al., (Figures 6B,C). The plot of DA response amplitudes in control 1987; Berger et al., 1991; Nicholas et al., 1996; Venkatesan et al., condition had a broad distribution, with almost half of the cells 1996; Latsari et al., 2002; Papay et al., 2006; Paschalis et al., 2009). exhibiting responses below 40% of the FSK effect, and only 6% NA fibers were present in all layers and areas of the cortex, responding above 90%. The distribution of DA response ampli- although their density exhibited regional and laminar variations. tudes was shifted to higher values in the presence of haloperidol, Conversely, DA fibers were present in all layers of two medial with only 10% of the cells responding below 40% of the FSK and lateral rostro-caudal bands comprising frontal, cingulated, effect, and 22% responding above 90%. Although NA response rhinal cortices as well as the agranular area of the insular cor- amplitudes were less variable, yohimbine also markedly shifted tex but were confined to deep layers of other cortical areas. The their distribution toward higher values. These results indicate that present study shows that DA receptors are functionally expressed Gi-coupled receptors are co-activated with Gs-coupled receptors in areas with scarce or no DA fibers, consistent with the pres- by DA and NA, and significantly limit DA- and NA-induced PKA ence of their cognate mRNAs and proteins (Ariano and Sibley, activation in cortical pyramidal neurons. 1994; Khan et al., 1998; Luedtke et al., 1999; Lemberger et al., 2007; Rivera et al., 2008; Oda et al., 2010). The source and nature of the DISCUSSION endogenous ligand activating these DA receptors is still unclear but Genetic labeling confirmed that NA neurons innervate the entire several studies indicate that DA can be co-released from NA fibers cortex whereas DA fibers were present in all layers of restricted where it is present as the biosynthetic precursor of NA. Indeed,

Table 3 | NA and DA effects involve alpha2 adrenergic and D2-like dopaminergic receptors.

DA DA + Haloperidol NA NA +Yohimbin

Amplitude (% FSK) (n/N) 49.5 ± 1.8 (242/12) 69.7 ± 1.9 (143/6) 61.9 ± 2.9 (114/7) 72.3 ± 1.7 (133/6) T10−90% (s) (n/N) 110.5 ± 4.8 (181/12) 75.9 ± 3.0 (84/4) 74.3 ± 2.0 (84/7) 46.8 ± 1.9 (119/5)

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 9 Nomura et al. Catecholaminergic control of cortical cAMP/PKA

Devoto et al. (2005, 2008) showed that electrical stimulation of the LC results in VTA-independent NA and DA increases in the frontal cortex but also in cortices devoid of DA ﬁbers. Similar conclusions were drawn from a study of amphetamine-induced DA release in the dorsal hippocampus, which is prevented by TH knock-down using siRNAs in the LC but not in the VTA (Smith and Greene, 2012). Hence, LC ﬁbers are a plausible source of endogenous DA that provides a rationale for the widespread functional expression of Gs- and Gi-coupled DA receptors in the cortex.

NA AND DA RECEPTORS WIDELY ACTIVATE cAMP/PKA IN THE CORTEX Our results indicate that NA and DA receptors mediate activation of the cAMP/PKA pathway in pyramidal neurons throughout the cortex, even in areas with scarce or no DA fibers. Indeed, we observed robust PKA activation by NA and DA in layer II/III of fronto-lateral, parietal and occipital regions of the cortex where DA fibers are restricted to deep layers. These results indicate that NA released by LC fibers widely influences the function of neural networks in the entire cortical mantle. The same conclusion applies to DA, regardless of its source. We found that NA responses were generally larger, with faster onsets and exhibited less regional variability than DA responses. NA-induced PKA activation was mimicked by the beta agonist isoproterenol and primarily mediated by beta 1 receptors, as shown by the powerful inhibitory effect of the specific antagonist CGP20712 on NA responses. These results are consistent with the predominant expression of beta1 over beta2–3 receptors in the cortex and with the widespread expression of beta1 receptors in layers and areas of the rodent cortex (Nicholas et al., 1996; Paschalis et al., 2009). DA-induced PKA activation exhibited similar properties in layer II/III and layer V, but both the amplitude and velocity of DA responses increased from rostral to caudal regions. DA responses were mimicked by the D1/D5 agonist SKF38393 and prevented by the D1/D5 receptor antagonist SCH23390, pointing to the involvement of D1 and/or D5 receptors in DA-induced PKA activation and its ros- trocaudal variations. Regional and laminar differences of D1 and D5 expression levels have been reported in the rodent cortex (Ariano and Sibley, 1994; Ciliax et al., 2000). However, the lack of specific pharmacological tools differentiating these receptors makes it difficult to assign the present functional variations to the regional levels of these receptors. The possibility that D1/D5 receptors forms heteromers with beta1 adrenoceptors further complicates the interpretation of DA response regional variations. We found that the beta blocker propranolol and the beta1 FIGURE 6 | Gi-coupled receptors dampen NA- and DA-induced PKA antagonist CGP20712 inhibited NA responses. Surprisingly, these activation. (A) Mean amplitude and rise time of NA- and DA-induced (10 μM each) PKA activation measured on layer II/III pyramidal neurons of antagonists also prevented DA responses and largely reduced the parietal cortex in the absence or presence of the D2-like antagonist SKF38393 responses suggesting that DA-induced PKA activation haloperidol (1 μM) or the alpha2 adrenergic antagonist yohimbin (1 μM). in the cortex may involve beta1 adrenergic receptors. Previ- (B,C) Distribution of FSK-normalized DA or NA response amplitudes ous studies established that DA effects can occur through the measured in the same sample of neurons as in (A) in absence or presence of haloperidol or yohimbin. The left axis corresponds to the histogram of the activation of adrenergic receptors (Aguayo and Grossie, 1994; proportion of cells for each amplitude interval and the right axis to the curve Rey et al., 2001; Cornil et al., 2002). The present observa- of cumulated frequency over the amplitude intervals. Note the shift of NA tions that DA effects were mimicked by SKF38393 and that and DA response amplitude distributions toward higher amplitudes in the presence of antagonists of Gi-coupled receptors. response to these agonists were prevented by the D1/D5 antagonist SCH23390 imply a direct effect of DA through D1/D5,

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 10 Nomura et al. Catecholaminergic control of cortical cAMP/PKA

but not beta adrenergic receptors. DA receptors are known the cortex, DA may thus expand the dynamic range of LC effects to form heteromers with a variety of receptors including beta on cortical cAMP/PKA signaling. adrenoceptors (Rebois et al., 2012; Fuxe et al., 2014). Moreover, cross antagonism by specific antagonist has been demonstrated ACKNOWLEDGMENTS for the beta1-D4 receptor heteromer (Gonzalez et al., 2012). The authors thank Bruno Giros, Uwe Maskos, Julie Catteau, Hence, these reports substantiate the hypothesis that D1/D5 Ludovic Tricoire, Elvire Guiot, and the IFR83 Cell Imaging Facility receptors exist as heteromers with beta1 adrenoceptors in the for their valuable help. Shinobu Nomura is recipient of a Naka- cortex, thus exhibiting the specific pharmacological pattern we jima Foundation fellowship. This work was supported by Centre observed. National de la Recherche Scientifique, Université Pierre et Marie Curie-P6 and by grants from Ecole des Neurosciences de Paris NA AND DA CONTROL CORTICAL cAMP/PKA SIGNALING VIA Gs- and (“Network for Viral Transfer”) and Fondation pour la Recherche Gi-COUPLED RECEPTORS sur le Cerveau/Rotary Club de France. Our results indicate that DA and NA exert a balanced control on cortical cAMP/PKA signaling by activating both Gs- and Gi- REFERENCES coupled receptors. We found that Gs-coupled beta1 and D1/D5 Aguayo, L. G., and Grossie, J. (1994). Dopamine inhibits a sustained calcium receptors are co-expressed in virtually all layers II/III and V pyra- current through activation of alpha adrenergic receptors and a GTP-binding protein in adult rat sympathetic neurons. J. Pharmacol. Exp. Ther. 269, midal neurons throughout the antero-posterior axis of the cortex. 503–508. The effects of yohimbin and haloperidol on NA and DA response Ariano, M. A., and Sibley, D. R. (1994). Dopamine receptor distribution in in pyramidal neurons of parietal layer II/III suggest that alpha2 the rat CNS: elucidation using anti-peptide antisera directed against D1A adrenoceptors and D2-like receptors are also broadly expressed and D3 subtypes. Brain Res. 649, 95–110. doi: 10.1016/0006-8993(94) in the cortex. This is in agreement with previous reports show- 91052-9 Aston-Jones, G. (2004). “Locus coeruleus, A5 and A7 noradrenergic cell groups,” ing widespread immunoreactivity for alpha2 and D4 receptors in The Rat Nervous System, 3rd Edn, Chap. 11, ed. G. Paxinos (San Diego, CA: in the cortex (Venkatesan et al., 1996; Khan et al., 1998; Rivera Elsevier Academic Press), 259–294. et al., 2008) and for D2 receptors in pyramidal cells of the medial Beaulieu, J. M., and Gainetdinov, R. R. (2011). The physiology, signaling, and prefrontal cortex (Zhang et al., 2010). pharmacology of dopamine receptors. Pharmacol. Rev. 63, 182–217. doi: We showed that kinetics and amplitudes of catecholaminer- 10.1124/pr.110.002642 Berger, B., Gaspar, P., and Verney, C. (1991). Dopaminergic innervation of the gic signals vary between cortical areas. It was previously shown cerebral cortex: unexpected differences between rodents and primates. Trends in pyramidal neurons of the prefrontal cortex that intracellular Neurosci. 14, 21–27. doi: 10.1016/0166-2236(91)90179-X elements of the cAMP/PKA cascade, such as phosphodiesterases, Bonnot, A., Guiot, E., Hepp, R., Cavellini, L., Tricoire, L., and Lambolez, B. (2014). phosphatases, or the coupling between receptor and adenylate Single-fluorophore biosensors based on conformation-sensitive GFP variants. cyclases are essential in controlling the kinetics and shape of FASEB J. 28, 1375–1385. doi: 10.1096/fj.13-240507 Bylund, D. B. (1992). Subtypes of alpha 1- and alpha 2-adrenergic receptors. FASEB D1-like responses (Castro et al., 2013). J. 6, 832–839. We found that block of alpha2 adrenoreceptors and D2-like Castro, L. R., Brito, M., Guiot, E., Polito, M., Korn, C. W., Hervé, D., et al. (2013). receptors by yohimbin and haloperidol increased the amplitude Striatal neurones have a specific ability to respond to phasic dopamine release. J. and velocity of NA and DA responses. This is consistent with an Physiol. 591(Pt 13), 3197–3214. doi: 10.1113/jphysiol.2013.252197 Castro, L. R., Gervasi, N., Guiot, E., Cavellini, L., Nikolaev, V. O., Paupardin- inhibitory effect of these receptors on cAMP/PKA signaling that Tritsch, D., et al. (2010). Type 4 phosphodiesterase plays different integrating limits DA- and NA-induced PKA activation in cortical pyramidal roles in different cellular domains in pyramidal cortical neurons. J. Neurosci. 30, neurons. In the case of DA, the balance of Gs- and Gi-coupled 6143–6151. doi: 10.1523/JNEUROSCI.5851-09.2010 receptors effects was such that many cells were not responsive Chinga, G., and Syverud, K. (2007). Quantification of paper mass distribu- to a saturating DA concentration. Hence, regulation of the bal- tions within local picking areas. Nordic Pulp Paper Res. J. 22, 441–446. doi: 10.3183/NPPRJ-2007-22-04-p441-446 ance between Gs- and Gi-coupled receptor activities can largely Ciliax, B. J., Nash, N., Heilman, C., Sunahara, R., Hartney,A., Tiberi, M., et al. (2000). influence the net effect of catecholamines on cortical neurons and Dopamine D(5) receptor immunolocalization in rat and monkey brain. Synapse networks. Interestingly, DA levels in basal conditions or upon 37, 125–145. doi: 10.1002/1098-2396(200008)37:2<125::AID-SYN7>3.0. LC stimulation are comparable to those of NA in various corti- CO;2-7 cal regions, but markedly lower than DA levels in basal ganglia Cornil, C. A., Balthazart, J., Motte, P.,Massotte, L., and Seutin, V. (2002). Dopamine activates noradrenergic receptors in the preoptic area. J. Neurosci. 22, 9320–9330. (Devoto et al., 2005, 2008). Our measurements of EC50 values are Cotecchia, S. (2010). The alpha1-adrenergic receptors: diversity of signaling net- indicative of a ∼20-fold lower affinity of DA for D1/D5 recep- works and regulation. J. Recept. Signal Transduct. Res. 30, 410–419. doi: tors than of NA for beta1 receptors, while NA and DA Gi-coupled 10.3109/10799893.2010.518152 receptors exhibit higher affinities than their cognate Gs-coupled Descarries, L., Lemay, B., Doucet, G., and Berger, B. (1987). Regional and laminar receptors (for a detailed comparison of pharmacological proper- density of the dopamine innervation in adult rat cerebral cortex. Neuroscience 21, 807–824. doi: 10.1016/0306-4522(87)90038-8 ties of receptors, see the International Union of Basic and Clinical Devoto, P., Flore, G., Saba, P., Fa, M., and Gessa, G. L. (2005). Stimulation of Pharmacology database: http://www.guidetopharmacology.org/). the locus coeruleus elicits noradrenaline and dopamine release in the medial This suggests that, at low catecholamine concentration, DA may prefrontal and parietal cortex. J. Neurochem. 92, 368–374. doi: 10.1111/j.1471- essentially trigger cAMP/PKA inhibition whereas NA effects may 4159.2004.02866.x be more balanced. Conversely, large catecholamine increases may Devoto, P., Flore, G., Saba, P., Castelli, M. P., Piras, A. P., Luesu, W., et al. (2008). 6-Hydroxydopamine lesion in the ventral tegmental area fails to reduce extra- be required for a significant contribution of DA to cAMP/PKA cellular dopamine in the cerebral cortex. J. Neurosci. Res. 86, 1647–1658. doi: stimulation. Assuming co-release of NA and DA from LC fibers in 10.1002/jnr.21611

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 11 Nomura et al. Catecholaminergic control of cortical cAMP/PKA

Drobac, E., Tricoire, L., Chaffotte, A. F., Guiot, E., and Lambolez, B. (2010). Calcium Papay, R., Gaivin, R., Jha, A., McCune, D. F., McGrath, J. C., Rodrigo, M. imaging in single neurons from brain slices using bioluminescent reporters. J. C., et al. (2006). Localization of the mouse alpha1A-adrenergic receptor (AR) Neurosci. Res. 88, 695–711. doi: 10.1002/jnr.22249 in the brain: alpha1AAR is expressed in neurons, GABAergic interneurons, Evans, B. A., Sato, M., Sarwar, M., Hutchinson, D. S., and Summers, R. J. (2010). and NG2 oligodendrocyte progenitors. J. Comp. Neurol. 497, 209–222. doi: Ligand-directed signalling at beta-adrenoceptors. Br. J. Pharmacol. 159, 1022– 10.1002/cne.20992 1038. doi: 10.1111/j.1476-5381.2009.00602.x Paschalis, A., Churchill, L., Marina, N., Kasymov, V., Gourine, A., and Ackland, Fuxe, K., Tarakanov, A., Romero, F. W., Ferraro, L., Tanganelli, S., Filip, G. (2009). Beta1-adrenoceptor distribution in the rat brain: an immuno- M., et al. (2014). Diversity and bias through receptor-receptor interactions histochemical study. Neurosci. Lett. 458, 84–88. doi: 10.1016/j.neulet.2009. in GPCR heteroreceptor complexes. Focus on examples from dopamine 04.023 D2 receptor heteromerization. Front. Endocrinol. (Lausanne) 5:71. doi: Rebois, R. V., Maki, K., Meeks, J. A., Fishman, P. H., Hebert, T. E., and Northup, 10.3389/fendo.2014.00071 J. K. (2012). D2-like dopamine and beta-adrenergic receptors form a signaling Gerfen, C. R. (2004). “Basal ganglia,”in The Rat Nervous System, 3rd Edn, Chap. 18, complex that integrates Gs- and Gi-mediated regulation of adenylyl cyclase. Cell. ed. G. Paxinos (San Diego, CA: Elsevier Academic Press), 455–508. Signal. 24, 2051–2060. doi: 10.1016/j.cellsig.2012.06.011 Gervasi, N., Hepp, R., Tricoire, L., Zhang, J., Lambolez, B., Paupardin-Tritsch, D., Rey, E., Hernandez-Diaz, F. J., Abreu, P., Alonso, R., and Tabares, L. et al. (2007). Dynamics of protein kinase a signaling at the membrane, in the (2001). Dopamine induces intracellular Ca2+ signals mediated by alpha1B- cytosol, and in the nucleus of neurons in mouse brain slices. J. Neurosci. 27, adrenoceptors in rat pineal cells. Eur. J. Pharmacol. 430, 9–17. doi: 10.1016/S0014- 2744–2750. doi: 10.1523/JNEUROSCI.5352-06.2007 2999(01)01250-X Gong, S., Doughty, M., Harbaugh, C. R., Cummins, A., Hatten, M. E., Heintz, Rivera, A., Penafiel, A., Megias, M., Agnati, L. F., Lopez-Tellez, J. F., Gago, B., N., et al. (2007). Targeting Cre recombinase to specific neuron populations et al. (2008). Cellular localization and distribution of dopamine D(4) receptors with bacterial artificial chromosome constructs. J. Neurosci. 27, 9817–9823. doi: in the rat cerebral cortex and their relationship with the cortical dopaminergic 10.1523/JNEUROSCI.2707-07.2007 and noradrenergic nerve terminal networks. Neuroscience 155, 997–1010. doi: Gonzalez, S., Moreno-Delgado, D., Moreno, E., Perez-Capote, K., Franco, R., Mallol, 10.1016/j.neuroscience.2008.05.060 J., et al. (2012). Circadian-related heteromerization of adrenergic and dopamine Smith, C. C., and Greene, R. W. (2012). CNS dopamine transmission D4 receptors modulates melatonin synthesis and release in the pineal gland. PLoS mediated by noradrenergic innervation. J. Neurosci. 32, 6072–6080. doi: Biol. 10:e1001347. doi: 10.1371/journal.pbio.1001347 10.1523/JNEUROSCI.6486-11.2012 Hu, E., Demmou, L., Cauli, B., Gallopin, T., Geoffroy, H., Harris-Warrick, R. Thevenaz, P., Ruttimann, U. E., and Unser, M. (1998). A pyramid approach to M., et al. (2011). VIP, CRF, and PACAP act at distinct receptors to elicit dif- subpixel registration based on intensity. IEEE Trans. Image Process 7, 27–41. doi: ferent cAMP/PKA dynamics in the neocortex. Cereb. Cortex 21, 708–718. doi: 10.1109/83.650848 10.1093/cercor/bhq143 Turiault, M., Parnaudeau, S., Milet, A., Parlato, R., Rouzeau, J. D., Lazar, M., et al. Khan, Z. U., Gutierrez, A., Martin, R., Penafiel, A., Rivera, A., and De La, C. (2007). Analysis of dopamine transporter gene expression pattern – generation A. (1998). Differential regional and cellular distribution of dopamine D2-like of DAT-iCre transgenic mice. FEBS J. 274, 3568–3577. doi: 10.1111/j.1742- receptors: an immunocytochemical study of subtype-specific antibodies in rat 4658.2007.05886.x and human brain. J. Comp. Neurol. 402, 353–371. doi: 10.1002/(SICI)1096- Venkatesan, C., Song, X. Z., Go, C. G., Kurose, H., and Aoki, C. (1996). Cellular and 9861(19981221)402:3<353::AID-CNE5>3.0.CO;2-4 subcellular distribution of alpha 2A-adrenergic receptors in the visual cortex of Latsari, M., Dori, I., Antonopoulos, J., Chiotelli, M., and Dinopoulos, A. (2002). neonatal and adult rats. J. Comp. Neurol. 365, 79–95. doi: 10.1002/(SICI)1096- Noradrenergic innervation of the developing and mature visual and motor cortex 9861(19960129)365:1<79::AID-CNE7>3.0.CO;2-G of the rat brain: a light and electron microscopic immunocytochemical analysis. Zhang, Z. W., Burke, M. W., Calakos, N., Beaulieu, J. M., and Vaucher, E. J. Comp. Neurol. 445, 145–158. doi: 10.1002/cne.10156 (2010). Confocal analysis of cholinergic and dopaminergic inputs onto pyra- Lemberger, T., Parlato, R., Dassesse, D., Westphal, M., Casanova, E., Turiault, M., midal cells in the prefrontal cortex of rodents. Front. Neuroanat. 4:21. doi: et al. (2007). Expression of Cre recombinase in dopaminoceptive neurons. BMC 10.3389/fnana.2010.00021 Neurosci. 8:4. doi: 10.1186/1471-2202-8-4 Luedtke, R. R., Griffin, S. A., Conroy, S. S., Jin, X., Pinto, A., and Sesack, Conflict of Interest Statement: The authors declare that the research was conducted S. R. (1999). Immunoblot and immunohistochemical comparison of murine in the absence of any commercial or financial relationships that could be construed monoclonal antibodies specific for the rat D1a and D1b dopamine recep- as a potential conflict of interest. tor subtypes. J. Neuroimmunol. 101, 170–187. doi: 10.1016/S0165-5728(99) 00142-3 Received: 19 June 2014; paper pending published: 15 July 2014; accepted: 05 August Morrison, J. H., Grzanna, R., Molliver, M. E., and Coyle, J. T.(1978). The distribution 2014; published online: 21 August 2014. and orientation of noradrenergic fibers in neocortex of the rat: an immunofluo- Citation: Nomura S, Bouhadana M, Morel C, Faure P, Cauli B, Lambolez B rescence study. J. Comp. Neurol. 181, 17–39. doi: 10.1002/cne.901810103 and Hepp R (2014) Noradrenalin and dopamine receptors both control cAMP- Morrison, J. H., Molliver, M. E., and Grzanna, R. (1979). Noradrenergic innervation PKA signaling throughout the cerebral cortex. Front. Cell. Neurosci. 8:247. doi: of cerebral cortex: widespread effects of local cortical lesions. Science 205, 313– 10.3389/fncel.2014.00247 316. doi: 10.1126/science.451605 This article was submitted to the journal Frontiers in Cellular Neuroscience. Nicholas, A. P., Hokfelt, T., and Pieribone, V. A. (1996). The distribution and Copyright © 2014 Nomura, Bouhadana, Morel, Faure, Cauli, Lambolez and Hepp. significance of CNS adrenoceptors examined with in situ hybridization. Trends This is an open-access article distributed under the terms of the Creative Commons Pharmacol. Sci. 17, 245–255. doi: 10.1016/0165-6147(96)10022-5 Attribution License (CC BY). The use, distribution or reproduction in other forums is Oda, S., Funato, H., Adachi-Akahane, S., Ito, M., Okada, A., Igarashi, H., et al. permitted, provided the original author(s) or licensor are credited and that the original (2010). Dopamine D5 receptor immunoreactivity is differentially distributed in publication in this journal is cited, in accordance with accepted academic practice. GABAergic interneurons and pyramidal cells in the rat medial prefrontal cortex. No use, distribution or reproduction is permitted which does not comply with these Brain Res. 1329, 89–102. doi: 10.1016/j.brainres.2010.03.011 terms.

Frontiers in Cellular Neuroscience www.frontiersin.org August 2014 | Volume 8 | Article 247 | 12

II.2 Responses to endogenous catecholamines released by optogenetic stimulation of LC fibers in the neocortex Little is known on neurotransmission events mediated by endogenous catecholamines. The aim of this study was to characterize catecholaminergic transmission events at the cellular level using real-time 2-photon imaging paradigms in acute cortical slice similar to those described in the above study. I studied the effects of endogenous catecholamines released from LC fibers, which presumably enable characterizing catecholaminergic transmission in any cortical layer and region due to their widespread distribution. Because LC fibers release NA, but possibly also DA, my secondary aim was to test the respective contributions of NA and DA to catecholaminergic transmission from LC fibers. I used an optogenetic approach to achieve selective and time-resolved stimulation of LC fibers in cortical slices. Indeed, 2- photon laser scanning does not result in sufficient channelrhodopsin (ChR2) excitation (because of the small 2-photon excitation volume and the low occurrence of ChR2 molecules) to trigger neuronal excitation (Rickgauer and Tank, 2009). Expressions of ChR2 in LC fibers and of a PKA sensor in cortical pyramidal neurons were both performed via viral transfer in vivo. Because of this dual injection protocol and of the necessity to work in adult slices for sufficient ChR2 expression, setting the conditions for combined optogenetics and 2-photon imaging required time and efforts, as briefly described below.

II.2.1 Expression of channelrhodopsin in cortical LC fibers Expression of channelrhodopsin (ChR2) in LC fibers was performed via bilateral injection of a Cre-dependent AAV encoding a ChR2-GFP (ChR2-YFP) fusion protein into the LC of 1 month-old DBH-Cre mice (as in the above Nomura et al. study). As described above, this protocol resulted in selective and efficient expression of ChR2-YFP in LC neurons and cortical LC fibers, as judged from anti-GFP immunohistochemistry (Fig.1 and Fig.2). Since the same AAV stock has been used in a previous study of my host laboratory and shown to result in functional ChR2 expression in the soma and corticopetal axons of basal forebrain cholinergic neurons (Hay et al., in revision), no further characterization was performed at this stage. Alternative route of ChR2 expression via the use of Ai32 mice bearing a floxed ChR2 transgene was not tested because it was not found suitable for the release of acetylcholine upon photostimulation of cortical cholinergic fibers (Hay et al., unpublished).

Figure 1: Expression of ChR2 in LC neurons. Upper: Low magnification picture showing the bilateral expression of ChR after a YFP immunohistochemistry. NA neurons of the LC are identified using immunohistochemistry against TH. Middle: Higher magnification from A. Lower: correspond to the inset in middle left panel. Membrane labeling for ChR2-YFP can be observed.

Figure 2: Expression of ChR2 in LC fibers in the cortex. YFP immunolabeling shows numerous ChR2 expressing fibers throughout the cortex

II.2.2 Expression of a PKA sensor in cortical pyramidal neurons from adult mice I chose to express the PKA sensor using tested stocks of recombinant Sindbis pseudovirions, which provide rapid and high level transduction of cortical pyramidal neurons. An 8-10 weeks delay post LC injection is required for high axonal expression of ChR2 in our experimental conditions. As a result, viral transduction of cortical neurons with a PKA sensor had to be performed in vivo in ~3 month-old mice, the day preceding combined imaging and optogenetic experiments in brain slices. At this age, standard protocols of slice preparation lead to significant neuronal death and tissue opacity. To address these issues, I tested two protocols for slice preparation from adult brains based either on intra-cardiac perfusion of ice- cold ACSF prior to slicing (used in the lab for electrophysiology) or on the use of a high- sucrose/low-calcium slicing solution (as kindly advised by Romain Pigeat and Dr. Nathalie Leresche, NPS). Furthermore, transduction efficiency with Sindbis virus is markedly decreased in adult as compared to younger animals (Johnson et al 1972). This resulted in low basal fluorescence of cells expressing the GAkdYmut PKA sensor, which exhibits large variations of fluorescence intensity but relatively low fluorescence in resting conditions (Bonnot et al., 2014). I thus also tested Sindbis viral transfer of the FRET PKA sensor AKAR3-EV endowed with bright fluorophores and large dynamic range (Komatsu et al., 2011). In my hands, the combination of AKAR3-EV expression and slicing in high sucrose/low calcium solution gave the best results and proved suitable for combined optogenetics and PKA imaging in adult brain slices. An example of AKAR3-EV expression 10 h after stereotactic injection of recombinant Sindbis virions in the parietal cortex of a 3 month-old mouse is shown in Fig.3. I noticed that extending the post-injection incubation period did not improve expression and actually resulted in apparently lesser survival of AKAR3-EV-expressing cells upon slicing at 18 h post injection (not shown). I thus routinely performed brain slicing after 10-12 h (overnight) incubation post injection. I chose an injection site 0.5 mm below the pial surface, because more superficial injection resulted in lesser transduction, presumably due to virions escaping to the cortical surface through the tract of the injection pipette. As visible in Fig.3, the AKAR3-EV sensor was expressed in pyramidal cells of layers II to VI, but imaging was performed on layer 5 pyramidal cells, which were more densely labeled. The functionality of the AKAR3-EV sensor in adult brain slices was tested using application of a maximally effective concentration of the adenylate cyclase activator forskolin (FSK, 12 µM; Gervasi et al., 2007). FSK elicited a maximal ratio value when AKAR3-EV is maximally phosphorylated. The mean response expressed as ΔR/R0 was 23 ±1% (n=84 cells, mean ± SEM)

Figure 3: Expression of AKAR3EV in the cortex of a 3 months old mouse 10h after injection of a recombinant Sindbis virus. AKAR3EV is expressed in all cortical layers. Slices were counterstained with DAPI.

II.2.3 Photostimulation of LC fibers induces PKA activation in pyramidal cells Implementation of the 2-photon set-up to accommodate fibered 1-photon stimulation of ChR2 (470 nm) was performed by Ludovic Tricoire and Elvire Guiot. I tested several photostimulation protocols for their ability to induce PKA activation. Photostimulation consisted in trains of light pulses of variable length (2 to 10 ms) applied at 10 Hz for different overall durations (5 to 20 s). All protocols elicited reversible PKA activation. However, photostimulation consisting in 5 s trains of 5 ms light pulses applied at 10 Hz yielded the most robust and reproducible responses (see example in Fig.4A) and was used in all subsequent experiments. Conversely, PKA responses of variable amplitudes were elicited by photostimulation with shorter (2 ms) light pulses (not shown) and responses induced using longer light pulses (10 ms) or trains (20 s) decreased upon repetitive stimulation (see example in Fig.4B). As shown in Fig.4A, photostimulation (5 ms pulse at 10 Hz for 5 s) induced peaks of PKA activity of sustained amplitude upon repetitive stimulation (interval between photostimulus trains, ~10 min). All responses were measured at the soma of layer 5 pyramidal neurons and were normalized to the first response obtained in each cell. Responses elicited by five successive photostimulus trains exhibited roughly stable amplitudes that ranged between 86 ± 14 % and 100 ± 10 % of the first response (n=23 cells, N=3 slices from 3 mice). These results indicate that selective photostimulation of cortical LC fibers triggers the release of endogenous catecholamines that activate cAMP/PKA signaling in pyramidal neurons. Furthermore, the selected photostimulation protocol yielded reproducible responses upon repetitive stimulation, thereby allowing the pharmacological characterization of these responses.

Figure 4. ChR2 Photostimulation elicits PKA responses in AKAR3EV expressing neurons imaged using two photon microscopy. A. Photostimulation induced repetitive PKA activation with similar amplitudes. B. Prolonged illumination resulted in decreasing response amplitudes. Grey traces correspond to the mean ratio value measured in the soma of the cells shown below. The red trace is the mean trace of all recorded cells. The histograms show the values of 5 consecutive responses normalized to the first response. Responses observed in A are shown in pseudocolour images. The calibration square shows the ratio value in colour and the pixel intensity coded in hue. Image time points correspond to the black arrows above traces.

II.2.4 Light-induced PKA responses are mediated by β adrenoceptors but do not involve D1/D5 DA receptors I next characterized Gs-coupled catecholamine receptors involved in PKA responses elicited by photostimulation of LC fibers using specific antagonists of these receptors. Before drug application, I performed two photostimulation trains to assess the stability of the PKA response.

Figure 5. Light- induced PKA responses are mediated by β1-ARs. A. Application of the β1 antagonist CGP20712 prior to photostimulation inhibits the PKA response. B. Inhibition of D1-like receptors prior to photostimulation was ineffective. Application of the specific β1 adrenoceptor antagonist CGP 20712 (100µM) almost completely prevented responses to a third photostimulation train (Fig.5A). The amplitude of

the residual response observed in the presence of CGP 20712 was 10 ± 4 % of the first response obtained in control condition (n=11, N=3 slices from 3 mice). These results indicate that light-evoked PKA responses were essentially mediated by β1 adrenoceptors, pointing to NA as the major catecholamine released by LC fibers in our experimental conditions. Conversely, application of the D1/D5 receptor antagonist SCH 23390 (1 µM) did not decrease light-induced responses (117 ± 11 % of the first response in control condition; n=25, N=3 slices from 3 mice; Fig.5B). These results rule out the involvement of D1/D5 receptors in light-induced PKA activation and confirm that NA, but not DA, is the major catecholamine released by LC fibers in our experimental conditions.

II.2.5 Inhibition of NA re-uptake potentiates light-induced PKA responses The clearance of released catecholamines involves recapture into catecholaminergic axons through relatively selective high-affinity transporters: the norepinephrine transporter (NET) and the dopamine transporter (DAT), which are thought to limit the neurotransmission signal in space and time. I used the same protocol as described in the above paragraph to test the effect of catecholamine reuptake inhibitors on light-evoked PKA responses. I found that application of the specific NET inhibitor riboxetine (100 nM) largely increased the response to photostimulation (Fig.6A). The mean response amplitude obtained in the presence of riboxetine was more than twice the amplitude of the first response measured in control conditions (250 ± 23%, n=27 cells, N=5 slices from 4 mice). These results confirm that NA is the major catecholamine released by LC fibers in our experimental conditions. In contrast, application of the DAT inhibitor GBR 12893 (100 nM) did not significantly modify the amplitude of the PKA response (99 ± 15 % of the first response in control condition; n=6 cells, N=2 slices from 2 mice; Fig.6B). These results indicate that photostimulation did not induce DA release from LC fibers, or at concentrations below the threshold for activation of D1/D5 receptors.

Figure 6. Inhibition of NET but not DAT transporters enhance light induced PKA responses. A. Application of the NET antagonist Riboxetine prior to photostimulation strongly enhances the light induced PKA response .B. Inhibition of DAT transporters with GBR12783 prior to photostimulation does not modify PKA responses.

II.2.6 Light-induced PKA responses do not involve D2-like DA receptors We previously found that inhibitory effects of D2-like receptors robustly contribute to cortical responses to exogenous DA, and the affinity these Gi-coupled receptors is reportedly higher those of D1/D5 receptors (see paragraph II.1.4 and Nomura et al. paper). I thus tested the possible involvement of D2-like receptors in cortical PKA responses induced by photostimulation of LC fibers. Using the same protocol as above, I found that application of the D2-like receptor antagonist haloperidol (10 µM) did not notably modify the amplitude of the PKA response (101 ± 12 %; n=10 cells, N=2 slices from 2 mice; Fig.7). These results indicate that D2-like receptors did not contribute to light-induced PKA responses that appear to be essentially mediated by NA receptors, confirming that NA, but not DA, is the major catecholamine released by LC fibers in our experimental conditions.

Figure 7. Light induced PKA responses are not mediated by D2-like receptors. Application of the D2-like antagonist haloperidol prior to photostimulation does not alter light induced PKA responses.

II.2.7 Discussion This study provides a proof of principle that neurotransmission events mediated by endogenous catecholamines can be studied with relevant spatiotemporal resolution using optogenetic stimulation of catecholaminergic fibers and 2-photon imaging of a PKA biosensor in brain slices. This is an important step towards the characterization of catecholaminergic transmission at the cellular and the subcellular levels.

As they stand, the results demonstrate phasic noradrenergic neurotransmission time-locked on axonal stimulation and occurring on a several minutes time scale. It is noteworthy that the decay of the PKA response is probably faster than reported by the AKAR-EV sensor, since AKAR probes are slowly dephosphorylated (i.e. deactivated) due to high affinity of the FHA

domain for the PKA substrate peptide (Dr Jin Zhang, unpublished observations). This suggests that catecholaminergic transmission can modulate brain networks on the minute time-scale. The results also show that NA released from LC fibers activates β1 adrenoceptors on pyramidal neurons, giving rise to a cAMP/PKA signal of sufficient amplitude to be clearly detectable at the single cell level. The amplitude of the response was dependent on NET activity, suggesting that NA reuptake controls the spatiotemporal extent of phasic noradrenergic transmission. Hence, the present results suggest that cortical noradrenergic transmission can operate with higher spatial and temporal precision that generally thought of a diffuse neuromodulatory system relying on volume transmission.

Our results also rule out the involvement of DA receptors in light-induced PKA activation and indicate that NA, but not DA, is the major catecholamine released by LC fibers in our experimental conditions. These results are at odds with the observations by Devoto et al. (2005 and 2008) that electrical stimulation of the LC results in cortical increases of both NA and DA that are resistant to lesions of midbrain DA neurons. This discrepancy has several plausible explanations. First, the stimulation paradigms differ in several respects. In particular, the duration of the electrical stimulation trains used by Devoto et al. (20 min) exceeds by far that of the present photostimulation (5 s). This may be a key factor in determining the proportion of NA and DA co-released from LC fibers, since they proceed from the same biosynthetic pathway and can be accumulated in the same vesicles. Second, our data were obtained in slice preparation using selective stimulation of local axons, whereas Devoto et al. studies were performed in vivo, in conditions where multiple other factors potentially influence catecholamine release. The importance of such factors is illustrated by recent reports showing that striatal cholinergic interneurons locally exert a strong control over striatal DA release that can override the remote effect of DA neurons somatic firing (Threlfell S et al., 2012; Cachope R et al., 2012). Hence, further work is required to confirm or dismiss the hypothesis that DA and NA can be co-released from LC fibers and can both contribute to catecholaminergic effects in cortical areas outside the VTA projection territory.

II.3 Conclusion and perspectives The main results of my thesis indicate that both NA and DA receptors are functionally expressed throughout the rostro-caudal axis of the neocortex and that phasic noradrenergic transmission can operate at the minute time-scale and have a significant impact at the level of single cortical neurons. These results also highlight the unanswered question of how dopaminergic transmission proceeds in cortical regions devoid of DA fibers. From a technical viewpoint, the present work represents an important step towards the characterization of catecholaminergic transmission at the cellular and the subcellular levels. Some of the perspectives opened by the present results are briefly discussed below along two lines of investigation. Other possible developments, such as the study of putative beta1-D1/D5 heteromeric receptors, are also of interest but fall beyond the scope of my thesis.

Spatio-temporal properties of noradrenergic transmission: The current protocol should allow testing the contribution of Gi-coupled receptors to light-induced responses. Some contribution is indeed expected but the effect of Gi-coupled receptors on the dynamics of the response may not be trivial. This protocol can also readily be used to explore the impact of noradrenergic transmission in different subcellular compartments of pyramidal neurons. The above results stem from measurements at the level of neuronal somata, but studies at the level of thin dendrites or spines are also of interest. Indeed, these compartments are major targets of catecholaminergic effects on dendritic integration and synaptic plasticity and their cAMP/PKA signaling properties differ from those of the soma (Castro et al., 2010, Bonnot et al., 2014). The current protocol can be easily modified to achieve higher temporal resolution using the high affinity cAMP biosensor Epac2-camps300, which is available in house and reports cAMP signals with high sensitivity and temporal fidelity (Norris et al., 2009; Castro et al., 2010). The use of this biosensor should provide a better estimate of the time frame of noradrenergic transmission. It is noteworthy that our light-induced PKA responses have amplitudes similar to those observed upon VIP application, which translate into a substantial decrease of the PKA-sensitive slow after hyperpolarization (Hu et al., 2011). Hence, an electrophysiological study of the effects of LC fibers photostimulation should help evaluate the impact of noradrenergic transmission on the excitability of pyramidal neurons. Studies of “presynaptic” factors modulating noradrenalin release at the level of LC fibers (e.g. via mu- opioid receptors (Matsumoto et al., 1994; Trendelenburg et al., 2000) can also be envisioned using the current experimental setup. Conversely, characterizing neurotransmission microdomains in relation to noradrenalin release sites, diffusion distance and spatial extent of

cAMP/PKA activation would require substantial efforts to increase the proportion of cortical neurons express the biosensor (e.g. though the use of AAV-mediated transfer) and to target photostimulation to single LC fibers (e.g. through structured illumination [Papagiakoumou E et al., 2010]).

DA release from LC fibers: LC fibers are the most plausible source of DA in cortical regions that are devoid of VTA fibers but nonetheless exhibit substantial levels of extracellular DA. As stated above at paragraph II.2.7, our experimental conditions differ from those used to evidence DA release from LC fibers in the neocortex (Devoto et al., 2005 and 2008). These differences may explain that our photostimulation paradigm failed to release effective amounts of DA from LC fibers. The electrical stimulation trains used by Devoto et al. last for 20 min, whereas I only tested the contribution of DA to PKA responses elicited by 5 s photostimulation. It is nonetheless possible that prolonged photostimulation can empty vesicles from their NA content and favor vesicular uptake of DA. This hypothesis may be tested by applying brief (5 s) photostimulation episodes following continuous photostimulation for several minutes. However, the cross-antagonism by beta adrenoceptor antagonists, which precludes isolating DA effects on PKA activation, may not allow reaching firm conclusions on this point. Alternative experiments relying on direct measurement of DA release evoked by photostimulation of LC fibers in slices and in vivo may thus be required to demonstrate DA release from LC fibers and characterize the conditions of this release.

III. MATERIALS AND METHODS Materials and methods relative to the Nomura et al. paper can be found in the corresponding section of the joined publication manuscript. The present section is dedicated to the material and methods associated with the optogenetic stimulation of LC fibers in the cortex.

III.1.Animals and brain slices

III.1.1 Source and housing All the experiments were performed according to the guidelines of the French Ministry of Agriculture, Food Processing Industry and Forestry for handling animals (decree 2013-118). Transgenic DBH-Cre mice (DBH-Cre) were a gift from Bruno Giros (McGill University, Canada, MMRRC line: Tg(Dbh-cre)KH212Gsat/Mmucd, stock number 032081-UCD (Gong et al. 2007)). Animals were maintained in a 12 h light–12 h dark cycle, in stable conditions of temperature (22°C), with food and water available ad libitum.

III.1.2 Viral production and stereotactic injections Site- and cell-type-specific labeling of NA neurons was achieved by stereotactic injection of a viral vector into the LC of DBH-Cre (1 month-old). The viral vector was a recombinant adeno-associated virus (AAV) driving Cre-dependent expression of a fusion protein containing channelrhodopsin 2 (ChR2) and a yellow variant (YFP) of the GFP from Aequorea victoria. The Cre-inducible vector AAV2/1-EF1a-DIO-hChR2(H134R)-EYFP-WPRE- HGHpA (titer: 3.1011 gc/ml) was produced from Addgene plasmid #20298 at the vector core facility of Nantes University (UMR 1089 IRT1, France). Aliquots of the pseudovirion were stored at −80 °C before stereotactic injection. The plasmid coding for AKAR3EV was provided by M. Matsuda (Kyoto University, Japan). The coding sequence of AKAR3EV was subcloned into the viral vector pSinRep5 (Invitrogen). This construct and the helper plasmid pDH26S (Invitrogen) were transcripted in vitro using the Megascript SP6 kit (Ambion). Baby hamster kidney cells were electroporated with the two RNAs (2.10 7 cells, 950 µF, 230 V) and grown for 24 h at 37°C in 5%CO2 in DMEM supplemented with 5% fetal calf serum before collecting cell supernatant containing the viruses (titer: ~109 ml, i.p/ml.).

Mice were anesthetized with isoflurane and placed on a small animal stereotactic frame (KOPF). A small hole was made in the skull using a 26G needle. For transduction of LC NA neurons, AAV-EF1α-DIO-ChR2-YFP pseudovirion was bilaterally injected adjacent to the

LC at coordinates from bregma: anteroposterior (AP), -5.45 mm; mediolateral (ML), ± 1 mm; dorsoventral (DV) -3.65 mm). Injections were performed through an internal canula at a rate of 0.1 μl/min for 10 min (total 1 µl per site). The canula was held in place for 15 minutes before retraction out of the brain. 8-10 weeks after the AAV EF1α ChR2-YFP injection in LC, mice were anesthetized with isoflurane and placed on the stereotaxic frame. A small hole was made in the skull using a 26G needle. 2 µl of Sindbis AKAR3EV mixed with Sindbis- mCherry (from Katia Boutourlinsky, diluted 1:40 in Sindbis AKAR3EV solution) was unilaterally or bilaterally injected in the parietal cortex (anteroposterior (AP), 1.5 mm; mediolateral (ML), ± 2 mm; dorsoventral (DV) 0.4-0.5 mm) through a glass pipette at a flow rate of 0.1 µl /min. The pipette was then held in place for approximately 15 minutes before completely retracting out of the brain. Cortical slices for imaging experiments ware prepared after 10-12 h of sindbis virus injection.Red fluorescent of mCherry was used for searching injection site on the cortical slice without activating ChR2 by the light.

III.1.3 Brain slice preparation for imaging and drugs Twelve hours after sindbis injection, mice were deeply anesthetized with 10 mg/ml ketamine and 0.1% xylazine before transcardiac perfusion with 20 ml ice cold sucrose ACSF solution containing : 73 sucrose, 85 NaCl, 2.5 KCl, 1.25 NaHPO4, 0.4 CaCl2, 7 MgCl2, 26 NaHCO3, 20 D-glucose, 5 Na pyruvate, 2 kynurenic acid (mM), 50 minocycline hydrochloride (50nM) saturated with 5% 5% CO2/95% O2. The brain was dissected and immersed in the same solution. Coronal slices (300µm thick) of neocortex were cut using a Leika VT 1000S Vibratome (Leica). Slices were kept at 33° for 30 min in the ACSF containing 126 NaCl, 2.5

KCl, 1.25 NaHPO4, 2 CaCl2, 1 MgCl2, 26 NaHCO3, 20 D-glucose, 5 Na pyruvate, 2 kynurenic acid (mM), 50 minocycline hydrochloride (50nM) with 5% CO2/95% O2 then transfer in the ACSF at room temperature for 1 hour before imaging experiment. All the drugs were from Sigma or Tocris and were bath applied (SCH23390 hydrochloride (1µM); Haloperidol (10µM); CGP20712 hydrochloride (100 nM); Riboxetine mysylate (100 nM); GBR 12783 dihydrochloride (100nM); Forskolin (12.5µM)).

III.1.3 Immunohistochemistry Eight weeks after injection, mice were deeply anesthetized with 10 mg/ml ketamine and 0.1% xylazine before transcardiac perfusion with 50 ml of 4% paraformaldehyde in 0.12 mM phosphate buffer (pH7.4). Brains were removed and post fixed with the same fixative for 2 hours, cryoprotected with 30% sucrose and cut with a freezing microtome (Leica) at a thickness of 40 µm. Slices were washed overnight with PBS, then blocked and permeabilized

with PBS complemented with 0.2% fish skin gelatin, 0.25% triton X-100 (PBS-GT) for 2h at room temperature. Brain slices were incubated overnight at 4°C with primary antibody diluted in PBS-GT (monoclonal mouse anti TH: MAB318 (Clone LNC1), Millipore, France 1/2000; polyclonal chicken anti-GFP: GFP-1020 (Aves Labs, USA,1/2000). After 6x10min wash in PBS-GT, slices were incubated for 3h at room temperature with fluorescent labeled secondary antibody diluted at 1/1000 in PBS-GT (goat anti chicken alexa488 (Invitrogen A11039); goat anti mouse 555 (Invitrogen A21422). DAPI (1/1000) was added in the last washing solution. Slices were mounted in fluoromount (Clinisciences, France) after extensive washes in PBS- GT followed by PBS.

III.2. Two photon imaging and photostimulation

III.2.1. Two photon imaging. Two-photon images were obtained with a custom built two-photon laser scanning microscope based on Olympus BX51WI upright microscope (Olympus) with x60 (0.9 NA) water- immersion objectives and a titanium:sapphire laser (MaiTai HP; Spectra Physics). For FRET imaging, two photon excitation was performed at 850 nm for CFP. Galvanometric mirrors (model 6210; Cambridge Technology) were used for bidirectional line scanning at 1.4 frame / s, and piezo-driven objective scanner (P-721 PIFOC; physic Instrument GmbH) was for z- stack image acquisition. A two-photon emission filter was used to reject residual excitation light (E700 SP, Chroma Technology). A fluorescence cube containing 479/40 and 542/50 emission filters and a 506 nm dichroic beamsplitter (FF01-479/40, FF01-542/50, and FF506- Di02-25x36 Brightline Filters; Semrock) was used for the orthogonal separation of the two fluorescence signals. Two imaging channels (H9305 photomultipliers, Hamamatsu) were set up for simultaneous detection of the two types of fluorescence emission. The 512x512-pixel images were exported using MATLAB (The Mathworks) and analyzed with custom made macros using ImageJ (U.S. National Institute of Health). Fluorescence changes were measured as fluorescence ratio R=F535/F480, were F535 is the fluorescence of YFP and F 480 is the fluorescence of CFP. Pseudocolor hue saturation value (HSV) encoding of fluorescence intensity and ratio was performed using MATLAB custom procedures written by Hirokazu Tanimoto (Institute Jacques Monod).

III.2.2. Light stimulation of ChR2 A fiber optic cable (400µm diameter, ThorLabs) was installed on the 2-photon microscope stage and controlled with micromanipulators (ROE-200, SUTTER INSTRUMENT). Fiber-

coupled High-Power LED and LED driver (LEDD1B, Thorlabs) was used to deliver the 470 nm light pulses for ChR2 activation. A mechanical shutter (DSS25 and VDM1000 Sutter driver UNIBLITZ) was installed on the top of the objective holder to protect PMTs from LED light interference during photostimulation. Light pulse trains and shutter opening were controlled by custom software written in LabVIEW (National Instruments). The tip of the optical fiber was placed just above the slice surface and positioned 400 µm away from the center of the imaging field with and angle of 30°. Various stimulations were tested (see Results section). Up to five identical successive responses were obtained with 5ms light flashes with a pulse frequency of 10 Hz and total time of pulse train of 5s. Amplitude of successive light induced responses were compared to the pre-light stimulation baseline and calculated as (R-R0)/R0 with R0 being the mean baseline ratio prior to the photostimulation Values were normalized to the first response amplitude. Statistical analyses were performed with MATLAB (MathWorks) One-way ANOVA followed by Tukey‘s test were used to determine statistical significance. The data are presented as mean ± SEM.

IV. APPENDIX: Hepp et al. 2014 Hepp R, Hay YA, Aguado C, Lujan R, Dauphinot L, Potier MC, Nomura S, Poirel O, El Mestikawy S, Lambolez B, Tricoire L. Glutamate receptors of the delta family are widely expressed in the adult brain. Brain Struct Funct. 2014 Aug 20. [Epub ahead of print]

Brain Struct Funct DOI 10.1007/s00429-014-0827-4

ORIGINAL ARTICLE

Glutamate receptors of the delta family are widely expressed in the adult brain

Re´gine Hepp • Y. Audrey Hay • Carolina Aguado • Rafael Lujan • Luce Dauphinot • Marie Claude Potier • Shinobu Nomura • Odile Poirel • Salah El Mestikawy • Bertrand Lambolez • Ludovic Tricoire

Received: 29 January 2014 / Accepted: 17 June 2014 Ó Springer-Verlag Berlin Heidelberg 2014

Abstract Recent reports point to critical roles of gluta- structures, but not in non-nervous tissues examined. Both mate receptor subunit delta2 (GluD2) at excitatory syn- delta subunits were upregulated during postnatal develop- apses and link GluD1 gene alteration to schizophrenia but ment. Widespread neuronal expression of the GluD1 the expression patterns of these subunits in the brain protein was confirmed using immunohistochemistry. remain almost uncharacterized. We examined the distri- Examination at the electron microscopic level in the hip- bution of GluD1–2 mRNAs and proteins in the adult rodent pocampus revealed that GluD1 was mainly localized at brain, focusing mainly on GluD1. In situ hybridization postsynaptic density of excitatory synapses on pyramidal revealed widespread neuronal expression of the GluD1 cells. Control experiments performed using mice carrying mRNA, with higher levels occurring in several forebrain deletion of the GluD1- or the GluD2-encoding gene con- regions and lower levels in cerebellum. Quantitative RT- firmed the specificity of the present mRNA and protein PCR assessed differential GluD1 expression in cortex and analyses. Our results support a role for the delta family of cerebellum, and revealed GluD2 expression in cortex, glutamate receptors at excitatory synapses in neuronal albeit at markedly lower level than in cerebellum. Like- networks throughout the adult brain. wise, a high GluD1/GluD2 mRNA ratio was observed in cortex and a low ratio in cerebellum. GluD1 and GluD2 Keywords Delta glutamate receptors Cortex mRNAs were co-expressed in single cortical and hippo- Cerebellum GRID1 GRID2 campal neurons, with a large predominance of GluD1. Western blots using GluD1- and GluD2-specific antibodies showed expression of both subunits in various brain Introduction

Glutamate is the main excitatory neurotransmitter in the B. Lambolez and L. Tricoire contributed equally. vertebrate central nervous system. During the quest for

R. Hepp Y. A. Hay S. Nomura O. Poirel C. Aguado R. Lujan S. El Mestikawy B. Lambolez L. Tricoire (&) Departamento de Ciencias Me´dicas, Facultad de Medicina, Sorbonne Universite´s, UPMC Univ Paris 06, UM CR 18, Instituto de Investigacio´n en Discapacidades Neurolo´gicas Neuroscience Paris Seine, 75005 Paris, France (IDINE), Universidad Castilla-La Mancha, Campus Biosanitario, e-mail: [email protected] C/Almansa 14, 02008 Albacete, Spain

R. Hepp Y. A. Hay S. Nomura O. Poirel L. Dauphinot M. C. Potier S. El Mestikawy B. Lambolez L. Tricoire Sorbonne Universite´s, UPMC Univ Paris 06, Inserm, CNRS, Centre National de la Recherche Scientiﬁque (CNRS), UM 75, U1127, UMR 7225, ICM, 75013 Paris, France UMR 8246, Paris, France S. El Mestikawy R. Hepp Y. A. Hay S. Nomura O. Poirel Department of Psychiatry, Douglas Hospital Research Center, S. El Mestikawy B. Lambolez L. Tricoire McGill University, Montreal, Quebec H4H 1R3, Canada Institut national de la Sante´ et de la Recherche Me´dicale (INSERM), UMR-S 1130, Paris, France

123 Brain Struct Funct ionotropic glutamate receptors (iGluRs) two peculiar can- family of glutamate receptors in neuronal networks didates, GluRdelta1 (GluD1) and GluRdelta2 (GluD2), throughout the adult brain. were cloned by sequence homology with iGluRs subunits of the AMPA, Kainate, and NMDA subtypes (Araki et al. 1993; Lomeli et al. 1993; Yamazaki et al. 1992). However, Materials and methods delta subunits are unresponsive to glutamate and progress in identifying their functions has been slower than for other Animals iGluRs (Schmid and Hollmann 2008). Much of what is known on GluDs derives from studies All animal handling procedures were performed in accor- of GluD2, which is prominently expressed in cerebellar dance with European Commission guidelines (2010/63/ Purkinje cells (Araki et al. 1993; Lomeli et al. 1993) and is UE). Genotyping of GluD1 KO mice (generous gift from required for proper development and function of the cere- Jian Zuo, Memphis, TE) mice was performed by end-point bellum (Kakegawa et al. 2008; Kashiwabuchi et al. 1995). PCR as described previously [Table 1; Fig. 1, (Gao et al. GluD2 acts as a synapse organizer via interactions with 2007)]. GluD2 hotfoot (HO) mice (HO-Nancy line, a postsynaptic scaffold and signaling proteins, and with generous gift from Carole Levenes, Paris, France) exhibit a presynaptic parallel fiber terminals (Kakegawa et al. 2008; deletion of the exons 9–12 (Lalouette et al. 2001). Geno- Matsuda et al. 2010; Uemura et al. 2010). Moreover, the typing of the GluD2 HO allele was performed by copy metabotropic glutamate receptor mGlu1 associates with number quantitative PCR. Genomic DNA was extracted GluD2 (Kato et al. 2012) and triggers the opening of the using the Agencourt DNAdvance kit (Beckman coulter, GluD2 channel, which is critically involved in the slow Villepinte, France). Exons 7, 10 and 11 were amplified glutamatergic current at the parallel fiber-to-Purkinje cell separately from 10 ng of genomic DNA using specific synapse (Ady et al. 2014; Kato et al. 2012). primers (Table 1; Fig. 1) in a total reaction volume of Recombinant GluD1 is also endowed with a functional 10 ll using the LC480 Sybr Green I Master mix on a channel pore domain and promotes synapse formation LightcyclerÒ 480 System (Roche). One step at 95 °C for in vitro (Kuroyanagi et al. 2009;Orthetal.2013; Ryu et al. 5 min followed by 45 cycles (95 °C, 15 s; 60 °C, 60 s) 2012; Yadav et al. 2011). Several reports implicate the were performed. Each PCR reaction was performed in GRID1geneinschizophrenia(Fallinetal.2005; Guo et al. duplicate according to manufacturer instructions. Relative 2007; Treutlein et al. 2009). Likewise, GRID1 knockout abundance of exon 10 compared to wild-type (WT) ani- (KO) mice exhibit behavioral correlates of schizophrenia mals was calculated using the formula 2^([Ct(exon7) - Ct symptoms, such as hyperaggressiveness and deficits in social (exon10)] - [Ct(exon7WT) - Ct(exon10WT)]) where interaction (Gao et al. 2007; Yadav et al. 2012, 2013), mean Ct values were normalized to the mean Ct value of thereby establishing the requirement of GluD1 for normal exon 7 and to those values obtained from WT animals. brain functions. However, little is known on the explicit Similar formula was used for quantification of exon 11 functions of GluD1 in neuronal networks, for its expression abundance. Typical Ct values were in the range 20–23. In pattern in the postnatal brain remains almost uncharacterized. heterozygous mice, exons 10 and 11 were twice less Original studies of delta subunits mRNA distribution in abundant than in WT mice. Homozygous mice were the rodent brain report selective GluD2 expression in identified by the absence of amplification of both exons 10 Purkinje cells and rapid postnatal decrease of GluD1 and 11 (typically Ct [32). expression down to low levels in the adult (Araki et al. 1993; Lomeli et al. 1993). However, subsequent reports In situ hybridization indicate that adult expression of both subunits is more widespread than originally described (Gao et al. 2007; Mouse brains of WT mice were frozen in isopentane at Mayat et al. 1995; Yadav et al. 2012; Yamasaki et al. -30 °C. Sections were prepared with a cryostat at -20 °C, 2011). This prompted us to re-examine their expression thaw-mounted on glass slides and stored at -80 °C until patterns across the brain with a primary focus on GluD1. use. Regional mRNA hybridization in situ was performed We used a combination of in situ hybridization, RT- as described (Gras et al. 2008). Antisense oligonucleotides PCR, western blot and immunohistochemistry to charac- (see Table 1; Fig. 1) were designed using Helios oligo terize the expression patterns of GluD1 and GluD2 in the design software (Helios Biosciences, France). Negative rodent brain. We found that both delta subunits are widely controls were performed using sense oligonucleotides expressed in the adult brain and co-localize at the single complementary to antisense probes. Oligonucleotides were cell level. The expression of both subunits increases after labeled with [S35]dATP (GE Healthcare, Chalfont St. birth, and we provide evidence for postsynaptic localiza- Giles, UK) using terminal transferase to a specific activity tion of GluD1. Our results support a role for the delta of 5 9 108 dpm/lg. The day of the experiment, slides were 123 Brain Struct Funct

Table 1 PCR primers Application GenBank accession number Sequence Size (bp) scRT-PCR vGluT1 1st PCR: Sense, 361 GGCTCCTTTTTCTGGGGGTACa 259 NM_053859.1 Antisense, 600: CCAGCCGACTCCGTTCTAAG 2nd PCR: sense, 373: TGGGGGTACATTGTCACTCAGAa 201 Antisense, 553: ATGGCAAGCAGGGTATGTGAC GAD65 1st PCR: sense, 99: CCAAAAGTTCACGGGCGGa 375 NM_012563.1 Antisense, 454: TCCTCCAGATTTTGCGGTTG 2nd PCR: sense, 156: TGAGAAGCCAGCAGAGAGCGa 260 Antisense, 392: TGGGGTAATGGAAATCAATCACTT GAD67 1st PCR: sense, 83: ATGATACTTGGTGTGGCGTAGCa 253 NM_017007.1 Antisense, 314: GTTTGCTCCTCCCCGTTCTTAG 2nd PCR: Sense, 159: CAATAGCCTGGAAGAGAAGAGTCGa 177 Antisense, 314: GTTTGCTCCTCCCCGTTCTTAG GluD1 1st PCR: sense, 582: ACAAAAGGTGGACAAGAACATCAG 329 NM_024378.1 Antisense, 910: GGGAAGAGATGCGGTGGTTAT 2nd PCR: sense, 614: TCACCAGCCTCTTCACCACC 291 Antisense, 9034: AGATGCGGTGGTTATTCCTCA GluD2 1st PCR: sense, 1985: AGAGCTCCATCCAGTCTCTTCAGG 337 NM_024379.1 Antisense, 2321: AGTGCGATGCCATACCCCCG 2nd PCR: sense, 2037: TGGCACAGTCTTGGATTCCGCA 235 Antisense, 2271: GGAACAGTCGGGGTCATTGATGGC ISH GluD1 Antisense 1, 1286: GGGCCTCTCCTGTAGACTGCCATTCAAGCCCTTC – NM_008166.2 Antisense 2, 398: GCTAGGGTTCAGGTGGCAGGCAGTACGTGGTGAAC Antisense 3, 257: TCCCTGGGTCATGAGGTCACAGGCTTCTTGC Antisense 4, 2241: CTCCACCACGGCCACATCCCATAGAAAGGCGTAG Antisense 5, 2719: CCCCCATCTCTAGGGCAGAAAGCTCAATGGACGC RT-qPCR GluD1 Sense, 2582: ACCAGGAGACCCCCAAAG 77 NM_008166.2 Antisense, 2658: TTCATCCATGAGGCTGTTGA GluD2 Sense, 1958: CCGCTTTCCTCACTATCACC 65 NM_008167.2 Antisense, 2022: TTGCTTGGACAGATCCTGAA TBP Sense, -160: GGCGGTTTGGCTAGGTTT 83 NM_013684.3 Antisense, -78: GGGTTATCTTCACACACCATGA Ppib Sense, 454: TTCTTCATAACCACAGTCAAGACC 92 NM_011149.2 Antisense, 545: ACCTTCCGTACCACATCCAT HRPT Sense, 267: TGATAGATCCATTCCTATGACTGTAGA 126 NM_013556.2 Antisense, 392: AAGACATTCTTTCCAGTTAAAGTTGAG Actb Sense, 337: AAGGCCAACCGTGAAAAGAT 110 NM_007393.3 Antisense, 446: GTGGTACGACCAGAGGCATAC RT-cPCR GluD1/GluD2 Sense, 1171/1184: ATGTCCA(G/C)TTTGAAATCCTTGG 453/452 Antisense,1623/1635: Cy5.5-CACTGAGTAGTCCATGTAGCG Genotyping GluD1-KO Wild-type: sense: GCAAGCGCTACATGGACTACb 220 Antisense: GGCACTGTGCAGGGTGGCAG Mutant: sense: CCTGAATGAACTGCAGGACGb 500 Antisense: CGCTATGTCCTGATAGCGATC GluD2-HO Exon 7: sense: AGCTGGAGGACCGTAAGTGGCA 96 Antisense: TGGTCTCCAGCATGGACCGC Exon 10: sense: CCGAAGAAATACCAGGGCTTCTCCA 121 Antisense: TCCCATCCTCTTGTGGGCTTCCA Exon 11: sense: CCATCACTCCAGACCGGGAGAA 150 Antisense: CAGCAATGCAGGCCCACAGA Position 1, ﬁrst base of the start codon. Bold characters refer to GluD2 a Karagiannis et al. (2009) b Gao et al. (2007)

123 Brain Struct Funct

Fig. 1 Location of the probes used in this study on the GRID1 and arrows, and arrowheads indicate the positions of the oligonucleotide GRID2 genes. Intron–exon structure of GRID1 and GRID2 genes, in probes used to analyze GluD1 and GluD2 mRNA expression in the which open boxes refer to exon sequences and black boxes to present study. ISH in situ hybridization, scRT-PCR, RT-cPCR, and transmembrane domains of the GluD1 and GluD2 proteins. The RT-qPCR single cell, co-amplification and quantitative RT-PCR, horizontal bars indicate the coding sequence of the antigen used for respectively. RT-cPCR quantifies GluD1 and GluD2 mRNA ratios by the generation of the anti-GluD1 and anti-GluD2 antibodies. Asterisk, co-amplification of their cDNAs using a single primer pair

fixed in 4 % formaldehyde in PBS, washed with PBS, and Pachot et al. (2004) and normalization was performed rinsed with water, dehydrated in 70 % ethanol, and air- using the formula given in Hellemans et al. (2007). dried. Sections were then covered with 140 lL of hybridization medium (Helios Biosciences, France) containing Relative abundance of GluD1 and GluD2 mRNAs 3–5 9 105 dpm of the labeled oligonucleotide mix. Slides were incubated overnight at 42 °C, washed, dried, and The procedure was performed essentially as described by exposed to BAS-SR Fujifilm Imaging Plate for 3 weeks. (Lambolez et al. 1996), except that quantification was The plates were scanned with a Fujifilm BioImaging based on fluorescence instead of radioactivity (Angulo Analyzer BAS-5000. et al. 1997; Szabo et al. 2012). The quantification of transcripts for GluD1 and 2 was performed from four Quantitative RT-PCR 4-months old mice via co-amplification using a Cy5.5 fluorescently labeled antisense primer and repeated three Total RNAs were extracted from frozen individual neo- times for each animal. Fifty nanogram of cDNA prepared cortex and cerebellum and treated with DNAse using Nu- from individual mouse cortex and cerebellum (see above) cleospin RNA II kit (Macherey–Nagel, France) in were amplified using 20 pmol of each primer (Table 1; accordance with the manufacturer’s protocol. One micro- Fig. 1) in a total reaction volume of 100 ll according to gram of each RNA sample was individually reverse tran- manufacturer instruction. Half of the antisense primer was scribed into cDNAs for 2 h at 42 °C using the Verso cDNA 50 labeled with Cy5.5. Five cycles (94 °C, 30 s; 45 °C, kit (ThermoFisher Scientific, Waltham, USA) according to 30 s; ramp to 72 °C, 1 min 10 s; 72 °C, 30 s) followed by the manufacturer’s instructions. qPCR assays were per- 35 cycles (94 °C, 30 s; 49 °C, 30 s; 72 °C, 30 s) were formed in a LightcyclerÒ 480 System (Roche), in the performed. The PCR products were digested by BfaI and presence of 200 nM of each primer (Table 1; Fig. 1), HpaII, which selectively cut GluD1 and GluD2 PCR 100 nM of specific hydrolysis probe (designed with Uni- fragments, respectively. The restriction reactions were versal Probe Library, Roche Applied Science) and 1X performed overnight at 37 °Con2ll aliquots of the PCR Solaris qPCR Master mix (Thermo Scientific). Each product in a 5 ll volume, and then run in parallel on a 2 % amplification reaction was performed in triplicate. For the agarose gel. Fluorescence was acquired using the Odyssey quantification of GluD1 and 2 transcripts in WT versus infrared imaging system (LI-COR Biosciences) using the mutant animals, amplification data were obtained from four excitation at 700 nm. Quantification of fluorescence inte- 1-month old animals of each genotype and normalized to grals was performed on raw images using the gel analyzer those obtained with the reference genes TATA box binding plugin of ImageJ software. Because only the antisense protein (TBP) and peptidylprolyl isomerase B (Ppib, also primer was labeled, each restriction cut generated only one known as cyclophilin B). For the developmental analysis of labeled restriction fragment. For each run, corresponding to GluD1 and 2, 2–6 animals were used for each time point a subunit-specific cut, the integral of the fluorescence and TBP, Ppib, hypoxanthine guanine phosphoribosyl present in both the cut and the uncut peaks was normalized transferase (HRPT) and beta-actin (Actb) genes were used to 100 %. The percentage of fluorescence present in the cut for normalization. Validation of these genes for use as peak thus represented the percentage of the corresponding reference was performed by Vandesompele et al. (2002) subunit in the GluD1-2 amplified products. For establishing

123 Brain Struct Funct the calibration curves, PCR co-amplification of cloned The determination of the GluD1/GluD2 ratio at the sin- GluD1 and 2 cDNA was performed similarly. Plasmids gle cell level was performed as described previously containing the full length open reading frame of GluD1 and (Angulo et al. 1997; Lambolez et al. 1996). The GluD 2 were mixed at different ratio to a final DNA concentra- cDNAs present in the RT reaction were co-amplified with tion of 100 fg/ll. One microlitre (100 fg = 104 molecules the primer pair described in Table 1 and Fig. 1 using of GluD1 and 2 cDNA) of this mix was then processed as 20 pmol of each primer in a total reaction volume of 100 ll. above in quadruplicate. Five cycles (94 °C, 30 s; 45 °C, 30 s; ramp to 72 °C, 1 min 10 s; 72 °C, 30 s) followed by 35 cycles (94 °C, 30 s; Single cell RT-PCR 49 °C, 30 s; 72 °C, 30 s) were performed. The product of this first PCR was purified on a 2 % low-melting-point Experiments were done on acute slices prepared from 2- to agarose gel. Aliquots of the purified product from the first 3-week-old wistar rats or C57BL/6 J mice (Janvier Labs). PCR were reamplified with the same primer pair, half of the Electrophysiological characterization of rat neocortical or antisense primer being Cy5.5 fluorescently labeled at the 50 mouse hippocampal interneurons and pyramidal cells was end. Thirty-five cycles (94 °C, 30 s; 49 °C, 30 s; 72 °C, performed by whole-cell patch-clamp recording using a 45 s) were performed. The PCR products were then diges- potassium gluconate-based intracellular solution containing ted by the GluD1-selective enzyme BfaI, run on agarose gel biocytin for post hoc morphological processing of recorded and fluorescence was acquired and analyzed as described cells, as described (Karagiannis et al. 2009). At the end of above. A control digest by both GluD1- and GluD2-selec- the whole-cell recording, the cell cytoplasm was aspirated tive (HpaII) enzymes was performed for each cell to verify into the recording pipette while maintaining a tight seal. the absence of uncut PCR product. The pipette was removed delicately to allow outside-out patch formation. The content of the pipette was expelled Production of GluD1-specific antibody into a test tube and reverse transcription (RT) was performed in a final volume of 10 ll as described previously The portion of the mouse cDNA sequence encoding the (Lambolez et al. 1992). Next, two steps of polymerase C-terminal domain E852-S928 of GluD1 (UniprotKB chain reaction (PCR) were performed essentially as accession number: Q61627, Fig. 1) was subcloned into the described (Cauli et al. 1997). The cDNAs present in the RT pGEX-2T vector and the GST-GluD1Ct fusion protein reaction were first amplified simultaneously using the pri- produced using IPTG induced BL21 bacteria. The GST- mer pairs described in Table 1 (for each primer pair the GluD1Ct protein was purified from the bacterial lysate on sense and antisense primers were positioned on two dif- glutathione-Sepharose beads and dialyzed against PBS. ferent exons, see Fig. 1 for GluD1-2). Taq polymerase (2.5 Antisera recognizing GluD1 were generated by immuniz- U; Qiagen, Hilden, Germany) and 20 pmol of each primer ing three rabbits with the purified GST-GluD1Ct protein were added to the buffer supplied by the manufacturer (Agro-Bio, La Ferte´ Saint Aubin, France). For immuno- (final volume, 100 ll), and 21 cycles (94 °C for 30 s, histochemistry and electron microscopy, the antibodies 60 °C for 30 s, and 72 °C for 35 s) of PCR were run. were affinity purified on GST-GluD1Ct linked Affigel 15 Second rounds of PCR were performed using 2 ll of the (Biorad, France) and concentrated using Amicon filters first PCR product as template. In this second round, each with a 50 K cutoff (Millipore, France). cDNA was amplified individually with a second set of primer pairs internal to the first PCR primer pair (nested Western blot primers) and positioned on two different exons. Thirty-five PCR cycles were performed (as described above). Then Tissue, brain or transfected HEK cells lysates were 10 ll of each individual PCR were run on a 2 % agarose obtained by sonication in ice-cold PBS supplemented with gel, in parallel with molecular weight markers and stained protease inhibitors (Roche Diagnostics, France). Protein with ethidium bromide. The RT-PCR protocol was tested concentration was quantified using the Bradford’s method on 1 ng of total RNA purified from rat whole brain, and all using BSA as standard. Equal protein amounts (15 lg/lane the transcripts were detected. Sizes of the PCR-generated for non-nervous tissues, 5 lg/lane for brain and HEK cells fragments were as predicted by the mRNA sequences. A lysates) were separated by SDS-PAGE and transferred onto control for mRNA contamination from surrounding tissue nitrocellulose sheets. Protein loading was controlled by was performed by placing a patch pipette into the slice reversible Ponceau Red staining. Non-specific sites on without establishing a seal. Positive pressure was then nitrocellulose membranes were blocked for 1 h at room interrupted and, following the removal of the pipette, its temperature with PBS containing Tween-20 (0.1 %) and content was processed as described. No PCR product was 5 % non-fat dry milk. Membranes were incubated over- obtained using this protocol (n = 5). night with primary antibodies at 4 °C in PBS containing 123 Brain Struct Funct

Tween 20 (0.1 %), and 1 % non-fat dry milk. The primary (Tonnes et al. 1999). Briefly, horizontal cryostat sections antibodies used were mouse monoclonal against b-actin (1/ (10 lm) from mouse brain were apposed to nitrocellulose 10,000, Sigma; clones AC40 or AC15) or GAPDH (1/ membranes moistened with 48 mM Tris-base, 39 mM 10,000; Millipore-Calbiochem; clone 6C5) and rabbit glycine, 2 % (w/v) sodium dodecyl sulfate, and 20 % (v/v) polyclonal against GluD1 (crude serum 1/10,000) and methanol for 15 min at room temperature (*20 °C). After GluD2 (1/2,000; Frontier institute, Japan). Bound anti- blocking in 5 % (w/v) non-fat dry milk in phosphate-buf- bodies were detected with secondary anti-IgG antibodies fered saline, nitrocellulose membranes were treated with coupled to either IRDye 700DX or IRDye 800CW (diluted DNase I (5 U/mL), washed and incubated in 2 % (w/v) 1/5,000, Tebu-bio, France) and analyzed with the LI-COR sodium dodecyl sulfate and 100 mM b-mercaptoethanol in Odyssey detection system. For the quantification of GluD1 100 mM Tris–HCl (pH 7.0) for 60 min at 45 °C to remove and 2 transcripts in different organs and brain structures, adhering tissue residues. After extensive washing, the blots data were obtained from three 1-month old WT mice and were reacted with affinity-purified anti-GluD1 antibodies normalized to glyceralgehyde 3-phosphate dehydrogenase (0.5 lg/mL) in blocking solution overnight at 4 °C. The (GAPDH). For the developmental analysis of GluD1 and 2, bound primary antibodies were detected with alkaline 3 animals were used for each time point and Actb was used phosphatase-conjugated anti-rabbit or anti-rabbit IgG sec- for normalization. Both GAPDH and Actb have been ondary antibodies. shown to provide reliable reference for quantification in our experimental conditions (Weldon et al. 2008). Immunohistochemistry for light microscopy

Immunocytochemistry Immunohistochemical reactions at the light microscopic level were carried out using the immunoperoxidase method Twenty-four hours after transfection with plasmids encoding as described earlier (Lujan et al. 1996). Briefly, sections GluD1 or GluD2 or with an empty vector, HEK cells were were incubated in 10 % normal goat serum (NGS) diluted fixed with 4 % paraformaldehyde for 20 min and permeabi- in 50 mM Tris buffer (pH 7.4) containing 0.9 % NaCl lized with 0.25 % Triton X-100. Non-specific sites were (TBS), with 0.2 % Triton X-100, for 1 h. Sections were blocked using 2 g/L fish skin gelatin (Sigma-Aldrich, incubated in anti-GluD1 antibody (1–2 lg/ml diluted in France). Cells were incubated 2 h at room temperature with TBS containing 1 % NGS), followed by incubation in the affinity-purified rabbit anti-GST-GluD1Ct antibody biotinylated goat anti-rabbit IgG (Vector Laboratories, (1/2,000). After washing, cells were incubated for 1 h at room Burlingame, CA, USA) in TBS containing 1 % NGS. 0 temperature with Alexa Fluor 555-coupled F(ab )2 fragments Sections were then transferred into avidin-biotin-peroxi- of goat anti-rabbit IgGs (1/2,000; Invitrogen, France). Cells dase complex (ABC kit, Vector Laboratories). Bound were mounted in fluoromount (Clinisciences, France). peroxidase enzyme activity was revealed using 3, 30-di- aminobenzidine tetrahydrochloride (DAB; 0.05 % in TB, Tissue preparation for histoblotting pH 7.4) as the chromogen and 0.01 % H2O2 as the sub- and immunohistochemistry strate. Finally, sections were air-dried and coverslipped prior to observation in a Nikkon photomicroscope (Nikkon, For histoblotting, animals were deeply anaesthetized by Eclipse 80i) equipped with differential interference con- intraperitoneal injection of ketamine/xylazine 1:1 (0.1 mL/ trast optics and a digital imaging camera. kg b.w.) and the brains were quickly frozen in liquid nitrogen. For immunohistochemistry, animals were anaes- Immunohistochemistry for electron microscopy thetized and transcardially perfused with ice-cold fixative containing 4 % paraformaldehyde, with or without 0.05 % Immunohistochemical reactions at the electron micro- glutaraldehyde and 15 % (v/v) saturated picric acid made scopic level were carried out using the pre- and post- up in 0.1 M phosphate buffer (PB, pH 7.4). After perfusion, embedding immunogold methods as described earlier brains were removed and immersed in the same fixative for (Lujan et al. 1996). These two methods provide comple- 2 h or overnight at 4 °C. Tissue blocks were washed mentary information about the subcellular and subsynaptic thoroughly in 0.1 M PB. Coronal 60 lm thick sections location of proteins and are best used in combination to were cut on a Vibratome (Leica VT1000). achieve conclusive results. The pre-embedding immunogold is the most reliable method for the localization of Histoblotting proteins at extrasynaptic and perisynaptic sites, and the post-embedding for synaptic proteins at putative glutama- The regional distribution of GluD1 was analyzed in mouse tergic synapses (Lujan et al. 1996). Ultrastructural analyses brain, using an in situ blotting technique (histoblot) were performed in a Jeol-1010 electron microscope. 123 Brain Struct Funct

Pre-embedding immunogold method which was defined within 5–10 lm from the surface. Randomly selected areas were then photographed from the Briefly, free-floating sections were incubated in 10 % NGS selected ultrathin sections and printed with a final magni- diluted in TBS. Sections were then incubated in anti-GluD1 fication of 50,0009. Immunoparticles identified in each antibodies (4–5 lg/ml diluted in TBS containing 1 % reference area and present in different subcellular com- NGS), followed by incubation in goat anti-rabbit IgG partments (dendritic spines, dendritic shafts, and axon coupled to 1.4 nm gold (Nanoprobes Inc., Stony Brook, terminals) were counted. Data were expressed as percent- NY, USA). Sections were postfixed in 1 % glutaraldehyde age of immunoparticles in each subcellular compartment. and washed in double distilled water, followed by silver enhancement of the gold particles with a HQ Silver kit Controls for electron microscopy (Nanoprobes Inc.). Sections were then treated with osmium tetraoxide (1 % in 0.1 M PB), block-stained with uranyl As controls for method specificity, sections were incubated acetate, dehydrated in graded series of ethanol and flat- with the omission of the primary antibodies, and other embedded on glass slides in Durcupan (Fluka) resin. sections were incubated with 5 % normal rabbit serum Regions of interest were cut at 70–90 nm on an ultrami- replacing the rabbit primary antibodies. Under these con- crotome (Reichert Ultracut E, Leica, Austria) and collected ditions, immunoreactivity resembling that obtained using on 200-mesh copper grids. Staining was performed on the specific antibodies could not be detected. Using poly- drops of 1 % aqueous uranyl acetate followed by Rey- clonal rabbit antibodies to calretinin, no plasma membrane nolds’s lead citrate. labeling was observed with our post-embedding method, showing that the labeling of the plasma membrane is due to Post-embedding immunogold method the antibodies raised to the peptide sequence present in GluD1. Briefly, ultrathin sections 80-nm thick from Lowicryl- embedded blocks of hippocampus were picked up on Statistical analyses coated nickel grids and incubated on drops of a blocking solution consisting of 2 % Human Serum Albumin (HSA) All values presented in this study are mean ± S.E.M. in 0.05 M TBS and 0.03 % Triton X-100 (TBST). The Statistical significance was assessed with Student’s paired grids were incubated with anti-GluD1 antibodies (10 lg/ml t test. p value\0.05 was considered statistically significant. in TBST with 2 % HSA) at 28 °C overnight. The grids were incubated on drops of goat anti-rabbit IgG conjugated to 10 nm colloidal gold particles (Nanoprobes) in 2 % Results HSA and 0.5 % polyethylene glycol in TBST. The grids were then washed in TBS and counterstained for electron Expression pattern of the GluD1 mRNA in the adult microscopy with saturated aqueous uranyl acetate followed brain by lead citrate. Data were obtained from three young adult animals, three samples of tissue per animal for preparation Antisense oligonucleotide probes complementary to the of embedding blocks, totaling n = 9 blocks. GluD1 mRNA were chosen in regions of weak homology with the GluD2 mRNA and tested individually by in situ Quantification of immunogold labeling hybridization on parasagittal sections of adult mouse brain (see ‘‘Materials and methods’’). Each of the five selected To establish the relative frequency of GluD1 immunore- probes yielded a specific signal as judged from the weak activity in CA1 pyramidal cells, we used 60-lm coronal signal observed with corresponding control sense probes in slices processed for pre-embedding immunogold immu- the same experimental conditions (Fig. 2a). An equimolar nohistochemistry. The procedures were similar to those mix of the five antisense probes was used in all subsequent used previously (Lujan et al. 1996). Briefly, for each of in situ hybridization experiments to determine the expres- three young adult animals, three samples of tissue were sion pattern of the GluD1 mRNA in the adult mouse brain. obtained for preparation of embedding blocks, totaling Labeling of the gray matter was observed throughout the n = 9 blocks. To minimize false negatives, electron antero-posterior and dorso-ventral axes of parasagittal microscopic serial ultrathin sections were cut close to the sections, with strongest labeling in most forebrain regions, surface of each block, as immunoreactivity decreased with and weakest labeling in the cerebellum (Fig. 2a). Exami- depth. We estimated the quality of immunolabeling by nation of coronal sections revealed intense labeling of the always selecting areas with optimal gold labeling at granular layer of the main olfactory bulb, piriform cortex, approximately the same distance from the cutting surface, basal ganglia (dorsal striatum and nucleus accumbens) and 123 Brain Struct Funct

Fig. 2 Regional distribution of the GluD1 transcript. a Autoradio- matter areas such as the corpus callosum (cc) and internal capsule grams obtained by in situ hybridization performed with [S35]labeled (ic). Marked regional differences are also observed with intense antisense oligonucleotides specific of GluD1 nucleic sequence on labeling in some structures, e.g. piriform cortex (Pir) and hippocam- sagittal sections of adult mouse brain. #1 to #5 hybridization with pus, and faint labeling in others such as the substantia nigra pars individual probes, mix 1–5 hybridization with an equimolar mix of the reticulata (SNR). Acb nucleus accumbens, Cbcerebellum, cc corpus five probes. Note the low signal obtained with corresponding sense callosum, CPu caudate putamen, Cx cortex, DR dorsal raphe; oligonucleotides. b Autoradiograms obtained with the mix of the five hippocampal formation (S subiculum, CA1-3 hippocampal fields, antisense probes on coronal sections of adult mouse brain. Labeling of DG dentate gyrus), Hyp hypothalamus, ic internal capsule, IC inferior the gray matter is visible throughout the antero-posterior and dorso- colliculus, LS lateral septum, MnR median raphe nucleus, MS medial ventral axes of brain sections, with overall stronger signal in the septum; olfactory bulb (Acl anterior commissure, Epl external forebrain and comparatively weaker signal in the cerebellum, plexiform layer, GrO granule cell layer), PAG periaqueductal gray, consistent with a widespread neuronal expression of the GluD1 Pir piriform cortex, SC superior colliculus, SNR substantia nigra pars mRNA in the brain. In contrast, a low signal is detected in white reticulata, Thal thalamus hippocampal formation (Fig. 2b). In the hippocampal for- exception of layer IV. Labeling was particularly intense in mation, the GluD1 transcript was primarily detected in the retrosplenial cortex. Other areas where GluD1 was principal neurons (pyramidal and granular cells). The entire highly expressed notably include some thalamic and cortex abundantly expressed the GluD1 transcript with the hypothalamic nuclei, periaqueductal gray, dorsal raphe, and

123 Brain Struct Funct inferior colliculi. Conversely, GluD1 appeared less The relative proportions of GluD1 and GluD2 mRNAs expressed in few regions, such as the pars reticulata of the in the neocortex and the cerebellum were quantified using substantia nigra and the cerebellum. The low signal in white co-amplification of their cDNAs with common primers, matter areas (such as the corpus callosum and internal followed by subunit-specific restriction digest of the PCR capsule) suggests that the GluD1 mRNA is scarce or absent products (see methods). The ability of the PCR to maintain in glial cells. These results indicate that the GluD1 mRNA original proportions of GluD1 and GluD2 cDNAs was first is widely expressed in neurons of the adult brain. assessed by mixing known amounts of two plasmids containing their entire coding sequence (Fig. 3c, d). Using GluD1 and GluD2 mRNA are co-expressed either GluD1-specific or GluD2-specific digest, we in the cortex, hippocampus and cerebellum and increase obtained a linear relationship between the ratio of input postnatally GluD1 and GluD2 plasmids and the ratio measured on PCR products using this approach (GluD1-specific, slope = We next characterized the expression of GluD1 and GluD2 0.94 ± 0.06, r2 = 0.995; GluD2-specific, slope = 0.95 ± mRNAs in mouse neocortex and cerebellum using quan- 0.08, r2 = 0.992; Fig. 3e). The same protocol was then titative RT-PCR. The abundance of these mRNAs was applied to cDNAs obtained from neocortex and cerebellum normalized to those of a set of mRNA species showing of adult mice (Fig. 3f). Using GluD1-specific digest, we little variation across brain structures and postnatal ages found that the GluD1 transcript represented 80.7 ± 1.5 % (see methods). The expression level of the GluD1 mRNA of GluD transcripts in the neocortex (ratio GluD1/ in adult mice was 1.66 ± 0.13 fold higher in the neocortex GluD2 = 4.3 ± 0.4). In contrast, in the cerebellum, GluD2 than in the cerebellum (p \ 0.05, Fig. 3a). This moderate was the most abundant subunit (ratio GluD1/GluD2 = difference between the two structures is consistent with the 0.32 ± 0.01) and represented 75.8 ± 0.3 % of GluD above in situ hybridization results. The same experiments mRNAs. The same analyses performed with GluD2- performed on GRID1-/- mice yielded only minimal signal, specific digest yielded similar results (neocortex and cere- attesting to the specificity of GluD1 mRNA detection. This bellum, ratio GluD1/GluD2 = 4.62 ± 0.32 and 0.37 ± 0.01 residual signal can be attributed to the production of respectively). These results indicate that GluD1 and GluD2 truncated GluD1 mRNAs in the KO mice, in which only mRNAs are co-expressed at a high GluD1/GluD2 ratio in exons 11 and 12 of the GRID1 gene are deleted (Gao et al. the neocortex and a low GluD1/GluD2 ratio in the 2007). The expression level of the GluD2 mRNA in adult cerebellum. mice was 13.8 ± 0.8 fold higher in the cerebellum than in Co-expression of GluD1 and GluD2 mRNAs was next the neocortex (p \ 0.05, Fig. 3a), consistent with its high tested at the single cell level using single cell RT-PCR abundance in cerebellar Purkinje cells (Araki et al. 1993; (scRT-PCR) of neocortical neurons from acute brain slices Lomeli et al. 1993). No GluD2 mRNA was detected in (see ‘‘Materials and methods’’). Excitatory pyramidal neocortex or cerebellum of mutant mice carrying a neurons and inhibitory interneurons were visually selected homozygous deletion of the GRID2 gene encompassing the and electrophysiologically characterized based on their PCR amplicon [Hotfoot-Nancy strain, GRID2HO/HO; (Lal- action potential firing patterns (Cauli et al. 1997; Connors ouette et al. 2001)], confirming expression of GluD2 in the and Gutnick 1990). Subsequent scRT-PCR analyses neocortex of WT mice. Hence, GluD1 and GluD2 mRNAs revealed typical expression of vesicular glutamate trans- are both expressed in the neocortex and cerebellum of adult porter VGluT1 in pyramidal neurons and of glutamic acid mice. GluD1 mRNA levels moderately differ between decarboxylase GAD 65 and/or GAD 67 isoforms in inter- these two structures, whereas GluD2 mRNA level is neurons (Fig. 4a). Delta subunit mRNA expression was markedly higher in the cerebellum than in the neocortex. detected in all cells tested. GluD1 was expressed in all Postnatal variations of GluD1 and GluD2 mRNA expres- pyramidal neurons tested (n = 13), and additional expression were also examined (Fig. 3b). The expression levels of sion of GluD2 was detected in 4 of these neurons (Fig. 4a). GluD1 and GluD2 mRNAs in the neocortex smoothly The GluD1 mRNA was also found in a majority of inter- increased by a factor of *4 between P1 and P42 and non- neurons tested (n = 11 out of 13), with additional occur- significantly decreased between P42 and P56. In the cere- rence of the GluD2 mRNA in 3 of these neurons. Only bellum, GluD1 mRNA expression was stable before P14 GluD2 was detected in the two remaining interneurons and reached a plateau after a twofold increase between P14 (Fig. 4a). We also analyzed the expression of GluD subunit and P28. In contrast, the GluD2 mRNA continuously mRNAs in single hippocampal neurons because GluD1 increased after P7 to reach *9 fold its P1 level at P56. protein distribution in this structure was examined in detail Hence, GluD1 and GluD2 mRNAs are both up-regulated during the course of this study (see Figs. 5, 8 below). This during the postnatal development of the neocortex and the was performed on CA1 pyramidal cells (n = 10) and cerebellum. interneurons (n = 10) from stratum radiatum and stratum 123 Brain Struct Funct

Fig. 3 GluD1 and GluD2 a GluD1 GluD2 mRNA expression in cortex and WT -/- cerebellum. a Abundance of 1 GRID1 GRID2HO/HO 16 GluD1 and GluD2 normalized nce 0.8 to two reference genes as 12 determined by quantitative RT- 0.6 PCR and standardized to their 8 0.4 cortical levels. N.D. not detected. Note the low or 0.2 4 absence of signal in mice normalized abundance 0 N.D. N.D.

normalized abunda 0 lacking a given GluD subunit Cortex Cerebellum Cortex Cerebellum and the minimal change of b expression level of the GluD1 remaining subunit. Cortex GluD2 Cerebellum b Abundance of GluD1 and 4 8 GluD2 normalized to four reference genes analyzed by 3 6 quantitative RT-PCR during postnatal development and 2 4 standardized to their P1 levels. 1 c–e. Calibration of the relative 2

normalized abundance

quantification of GluD1 and normalized abundance 0 0 GluD2 mRNAs by cDNA co- P1 P7 P14 P28 P42 P56 P1 P7 P14 P28 P42 P56 amplification. c Fluorescence Age (day) Age (day) images of PCR fragments c GluD1 digest GluD2 digest amplified from a mix of GluD1- and GluD2-encoding plasmids %D1 %D1 using a Cy5.5-labeled primer and migrated on agarose gel u u after subunit specific restriction digest (u uncut). d Fluorescence D1 profiles of agarose gels (upper panels same gels as in c). No D2 undigested PCR product is observed upon GluD1?GluD2 d digest. e Ratios obtained using GluD1 digest GluD2 digest either GluD1 or GluD2 digest from variable proportion of GluD1- and GluD2-encoding plasmids. Note that the slope of the regression line derived from GluD1 digest is close to 1, %D1 %D1 indicating that GluD1 and u D1 u D2 GluD2 are amplified with similar efficiency. f Relative GluD1+2 digest No digest abundance of GluD1 and GluD2 transcripts determined from mouse brain extracts by cDNA co-amplification as in c–d. Both GluD1- and GluD2-specific digest indicate that GluD1 is the %D1 %D1 most abundant GluD subunit in D1 D2 u the cortex while GluD2 is the efGluD1 digest predominant subunit in the 100 cerebellum 80 Cerebellum Cortex slope = 0.94 60 u D1 40 GluD2 digest

Calculated %D1 20 GluD1 digest GluD2 digest 0 Cerebellum Cortex 0 20 40 60 80 100 Theoretical %D1 u D2

123 Brain Struct Funct

of GluD cDNAs obtained from single pyramidal cells and interneurons, respectively (Fig. 4b, c). These results indicate that GluD1 and GluD2 mRNA are widely expressed in neuronal types of the neocortex and hippocampus, and that their occurrences and proportions in single cells reﬂect the dominance of GluD1 observed in our analyses of cortical tissue samples (see above).

GluD1 and GluD2 proteins are widely expressed in the adult brain and increase postnatally

In order to study the expression of the GluD1 protein, we raised a polyclonal anti-GluD1 antibody against part of its C-terminal intracellular domain that shows low homology with GluD2 (see ‘‘Materials and methods’’). This antibody was tested on western blots of protein extracts from HEK cells mock-transfected or transfected with plasmids encoding GluD1, GluD2 or the NMDA receptor subunit GluN1 (Fig. 5a). A single band of molecular weight of *100 kD was immunolabeled in the GluD1-containing lane whereas other lanes were negative, consistent with specific detection of the 1,009 amino acids GluD1 protein. Conversely, the same experiment performed with a GluD2-specific antibody raised against the corresponding region of the GluD2 protein (Takayama et al. 1995) resulted in labeling of a single band of molecular weight *100 kD in the GluD2-containing lane (Fig. 5a). Anti-GluD1 immunocytochemistry of HEK cells expressing GluD1 resulted in labeling localized primarily at the plasma membrane, whereas no labeling was observed in mock-transfected or GluD2-expressing HEK Fig. 4 Expression of GluD1 and GluD2 transcripts in single cortical cells (Fig. 5c). Finally, the anti-GluD1 antibody labeled a and hippocampal neurons. a Expression of GluD subunit mRNAs and main band of *100 kD on western blots of mouse neo- neuron-type markers in cortical neurons. Left panels show typical cortical and cerebellar protein extracts (Fig. 5b). This band responses of a pyramidal cell and an interneuron to depolarizing and -/- Middle panels was not observed in extracts from GRID1 mice but was hyperpolarizing current pulses in neocortical slices. HO/HO agarose gels of single cell RT-PCR products obtained from the same present in extracts from GRID2 mice. The same neurons as in left panels. Right panel summary of GluD1 and GluD2 experiments performed with the anti-GluD2 antibody expression in excitatory pyramidal cells (VGLUT1?, n = 13) and labeled a *100 kD band, which was not detected in inhibitory interneurons (GAD?, n = 13). b, c Relative abundance of HO/HO -/- GluD subunit mRNAs determined by co-amplification of GluD1 and GRID2 but persisted in GRID1 mice extracts GluD2 cDNAs obtained from single hippocampal neurons. b Fluores- (Fig. 5b). These results indicate that the present anti-GluD1 cence images of PCR fragments amplified from GluD1- and GluD2- antibody enables specific and sensitive detection of the encoding plasmids (D1, D2) and from cDNAs obtained from single GluD1 protein, and confirm the reported specificity of the pyramidal cells (P1-4) and interneurons (I1-4). PCR was performed using a Cy5.5-labeled primer and products were migrated on agarose anti-GluD2 antibody (Takayama et al. 1995). gel after GluD1-specific restriction digest by Bfa I (u uncut, di primer We probed the expression of GluD1 and GluD2 proteins dimer). c Fluorescence profiles of agarose gels shown in b reveal in various brain structures and peripheral organs of adult expression of both GluD1 (cut) and GluD2 (uncut) in individual cells mice using similar western blot analyses (Fig. 5d). None of and the predominance of the GluD1 transcript these proteins were detected in any of the seven peripheral organs examined. Expression of GluD proteins was nor- lacunosum molecular using co-amplification of GluD1 and malized to that of the housekeeping enzyme glyceraldehyde GluD2 cDNAs obtained from single neurons harvested in 3-phosphate dehydrogenase (GAPDH). The GluD1 protein hippocampal slices (see above and ‘‘Materials and meth- was detected in all brain structures tested and was *1.5–2 ods’’). GluD1 and GluD2 mRNAs were both detected in all fold higher in cortex, hippocampus and striatum than in tested cells. However, GluD1 was the most abundant olfactory bulb, thalamus, mesencephalon and cerebellum subunit and represented 97.5 ± 0.9 % and 92.1 ± 3.0 % (Fig. 5d), in good agreement with above in situ hybridization 123 Brain Struct Funct

Fig. 5 Expression of GluD1 and GluD2 proteins a, b Western blot observed in GluD2-expressing cells. d Western blot analysis of GluD analysis of GluD1 and GluD2 in HEK cells expressing indicated expression in peripheral organs and brain structures. GluD1 and glutamate receptor subunits and in mouse brain extracts. aGluD: GluD2 staining intensity was normalized to that of GAPDH. antibody against GluD subunit. Note the absence of signal in controls Histogram shows the abundance of GluD1 and GluD2 in diverse lacking a given GluD subunit. c Fluorescence image of HEK brain structures as compared to their cortical levels. Protein amounts expressing GluD subunits following immunocytochemistry using loaded per lane: 15 lg for non-nervous tissues and 5 lg for brain and anti-GluD1 antibody. Labeling localized primarily at the plasma HEK cells lysates membrane in GluD1-expressing cells, whereas no labeling was and quantitative RT-PCR results. GluD2 protein expression We also examined the expression of GluD proteins was also detected in all brain regions examined, but with normalized to actin levels during postnatal development of more contrasted inter-structure differences than GluD1 the cortex and the cerebellum (Fig. 6a, b). Cortical (Fig. 5d). Indeed, GluD2 level exhibited moderate differ- expression of GluD1 and GluD2 increased *4.5 fold ences among forebrain regions, but was higher in the mes- (GluD1) or *6.5 fold (GluD2) between P0 and P28. encephalon and reached 27 ± 4 fold its cortical level in the Between P28 and P56, GluD1 decreased down to *3 fold cerebellum, consistent with quantitative RT-PCR data. its P0 level, whereas GluD2 level remained roughly stable. These results indicate that both GluD1 and GluD2 proteins In the cerebellum, GluD1 abundance continuously are widely expressed in the adult mouse brain. They further increased to reach *3.5 fold its P0 level at P42 and then indicate that the abundance of GluD1 across the brain is less slightly decreased at P56. GluD2 level was almost unde- variable than that of GluD2, which is markedly enriched in tectable at P0, then increased *4 fold between P15 and the cerebellum, presumably because of its high abundance in P42 and remained stable at P56. These results show overall Purkinje cells (Araki et al. 1993; Lomeli et al. 1993). consistency with those obtained above on GluD mRNAs

123 Brain Struct Funct

Fig. 6 Upregulation of GluD1 and GluD2 proteins during postnatal panels histograms show the abundance of GluD1 and GluD2 during development. Upper panels Western blot analysis of GluD cortical postnatal development as compared to P0 levels. Note the distinct y- and cerebellar expression at indicated postnatal days. GluD1 and axis used for GluD2 in cerebellum GluD2 staining intensity was normalized to that of b-actin. Lower and confirm, at the protein level, that GluD1 and GluD2 are adult brain, as observed at the mRNA level using in situ both up-regulated during postnatal development of the hybridization. neocortex and the cerebellum. To study in more detail the distribution patterns of GluD1, we carried out immunohistochemical studies at the Expression and localization patterns of the GluD1 light microscopic level, focusing on the forebrain. Simi- protein in the adult brain larly to histoblots, immunoreactivity for GluD1 was strong in the hippocampus, caudate putamen and neocortex We used the histoblot technique to determine the (Fig. 7b). Moderate immunoreactivity was observed in the regional distribution and expression levels of GluD1 in basolateral amygdaloid nucleus, in some thalamic nuclei the young adult mouse brain. This method is a reliable and in the lateral habenula (Fig. 7b). In the hippocampus, and convenient way to compare the regional distribution GluD1 was highest in the stratum lacunosum-moleculare of of different proteins in brain samples without compro- the CA1 and CA3 regions, and in the molecular layer of the mising the integrity of antibody-binding sites by tissue dentate gyrus, and lower in the strata oriens and radiatum, fixation, which is required for conventional immunohis- with no immunoreactivity detected in the stratum pyrami- tochemistry (Tonnes et al. 1999). Proteins transferred dale (Fig. 7b). To validate the specificity of the immuno- to nitrocellulose membranes were immunostained with reactions, GRID1-/- mice were used. The complete lack of purified GluD1-specific antibodies using conventional labeling in GRID1-/- mice (Fig. 7c) demonstrates that immunoblotting. The overall expression pattern of GluD1 immunolabeling at the light microscopic level is due to showed a strong immunoreactivity in the hippocampus, specific antibody-antigen interactions. caudate putamen and neocortex (Fig. 7a). Moderate We next used electron microscopy to study the subcel- immunoreactivity was detected in the granule cell layer lular distribution of GluD1 in the different subfields of the of the cerebellum and in the deep cerebellar nuclei, and CA1 region of the hippocampus. Using the pre-embedding weak in the thalamus, in midbrain nuclei, including the immunogold method, labeling for GluD1 in the strata inferior and superior colliculus, and brainstem nuclei oriens, radiatum and lacunosum-moleculare was found (Fig. 7a). No labeling was observed in white matter areas primarily along the plasma membrane, but also associated like the corpus callosum. These results indicate that the with the endoplasmic reticulum (ER) cisterna of dendritic GluD1 protein is widely expressed in neurons of the shafts and spine apparatus (Fig. 8a–f). To a lesser extent, 123 Brain Struct Funct

Fig. 7 Distribution of GluD1 protein in the adult mouse brain. forebrain using a pre-embedding immunoperoxidase method at the a GluD1 protein regional distribution was visualized on histoblots of light microscopic level. In the hippocampus, GluD1 immunoreactivity brain horizontal sections of adult mice using an affinity-purified anti- was low in the strata oriens and radiatum of the CA1 and CA3 GluD1 antibody. The receptor subunit exhibited a broad distribution regions. Immunoreactivity was highest in the stratum lacunosum- in the brain. In particular, strong immunoreactivity for GluD1 was moleculare of the CA1 and CA3 regions, and in the molecular layer of detected in the hippocampus (Hp), caudate putamen (CPu), cerebel- the dentate gyrus. In the GRID1-/- mice, the pattern of immunore- lum (Cb) and cortex (Cx), with the lowest intensity in the thalamus activity detected in the WT was missing. BL basolateral amygdala. (Thal) and deep cerebellar nuclei (dcn), and no labeling in areas like Scale bar a 0.5 cm, b, c 0.2 cm the corpus callosum (cc). b, c Immunoreactivity for GluD1 in the immunoparticles for GluD1 were also detected at presyn- protein detection techniques. GluD1 was expressed in aptic sites, along the plasma membrane of axon terminals neurons throughout the brain, with higher levels in the establishing excitatory synapses with spines of CA1 pyra- forebrain and lower levels in the cerebellum. GluD1 was midal cells. Quantitative analysis of immunogold particles localized at the postsynaptic density of excitatory syn- confirmed that most immunoparticles were found at post- apses on hippocampal pyramidal cells. GluD2 expression synaptic sites (96.3 %, n = 1,345) and only few (3.7 %, was also widespread but was markedly enriched in the n = 52) at presynaptic sites. Of the immunoparticles cerebellum. Likewise, the GluD1/GluD2 mRNA ratio was located at postsynaptic sites, 70.8 % (n = 942) were found high in the cortex and low in the cerebellum. Co- in the plasma membrane of dendritic spines and shafts, and expression of GluD1 and GluD2 mRNAs was assessed in 29.2 % (n = 393) at cytoplasmic site associated in intra- single cortical and hippocampal neurons. The expression cellular membranes. Finally, to determine if GluD1 is also of both subunits increased during postnatal development. located at postsynaptic membrane specialization we used Our results suggest that GluD1 and GluD2 contribute to post-embedding immunogold techniques in the stratum the function of neuronal networks throughout the adult lacunosum-moleculare, a region consisting primarily of tuft brain. dendrites from CA1 neurons and exhibiting prominent labeling for GluD1 (Fig. 8g, h). Using this approach, we GluD1 is widely expressed in neurons of the adult brain found that immunoparticles for GluD1 were mainly localized along the postsynaptic density of excitatory synapses Our results show that GluD1 is expressed in most regions in dendritic spines (Fig. 8g, h). Although single cell RT- of the adult brain. The converging dataset we obtained PCR analyses revealed expression of GluD1 in both CA1 using multiple probes of GluD1 mRNA and protein, and pyramidal cells and interneurons, spines observed in this control experiments performed using inactive probes or study are far more likely to be those of pyramidal cells GRID1-/- mice leave little doubt on the validity of the (densely spiny) than those of interneurons (less numerous present conclusion. Although GluD1 expression was not and sparsely spiny) (Freund and Buzsaki 1996). uniform across the brain, we observed only moderate variations of its regional abundance with maximal differences of *2 fold between the structures examined. A good Discussion overall correspondence was observed between GluD1 mRNA and protein levels. It is noteworthy that expression We probed the expression of GluD1 and GluD2 in the of the GluD1 mRNA was sufficiently high to be consis- adult rodent brain using a combination of mRNA and tently detected in individual cortical and hippocampal

123 Brain Struct Funct

Fig. 8 Subcellular localization of GluD1. Electron micrographs of plasma membrane (arrows) of some dendritic spines (s) and dendritic the hippocampus showing immunoparticles for GluD1 in different shafts (Den) of pyramidal cells, as well as at intracellular sites subﬁelds of CA1 region, as detected using immunogold methods in (crossed arrows). To a lesser extent, GluD1 immunoparticles were the adult mice. a, b Using the pre-embedding inmmunogold method, observed at presynaptic sites along the extrasynaptic plasma mem- only few immunoparticles for GluD1 were detected along the brane of axon terminals (at) establishing excitatory synapses with extrasynaptic plasma membrane (arrows) of some dendritic spines dendritic spines (s). f Speciﬁcity of the antibody was tested in (s) of CA1 pyramidal cells establishing asymmetrical synapses with GRID1-/- mice. Very few immunoparticles for GluD1 were observed axon terminal (at) in the strata oriens (a) and radiatum (b). Few associated with mitochondria and/or myelinated axons (double GluD1 immunoparticles were observed associated with intracellular arrows). g–h Using the post-embedding inmmunogold method, membranes (crossed arrows) of the dendritic shafts (Den). c–e The GluD1 immunoparticles were detected along the postsynaptic density stratum lacunosum-moleculare showed the highest density of immu- of dendritic spines (s) of CA1 pyramidal cells establishing asymmet- noparticles for GluD1, mainly detected along the extrasynaptic rical synapses with axon terminal (at). Scale bars a–h 0.5 lm

neurons using single cell RT-PCR, unlike low abundance possibility that GluD1 is expressed also in aminergic and mRNAs encoding neuropeptide receptors (Ferezou et al. peptidergic neurons. Closer examination in the hippocam- 2007; Gallopin et al. 2006). GluD1 expression was pri- pus revealed that GluD1 was mainly localized at the marily neuronal, as judged from low in situ hybridization plasma membrane along the postsynaptic density of and immunochemical signals in the white matter and from excitatory synapses in dendritic spines of pyramidal cells. the absence of protein detection in other organs examined. This localization is similar to that reported for GluD2 at the The strong labeling of essentially glutamatergic (e.g. hip- parallel ﬁber-to-Purkinje cell synapse (Landsend et al. pocampal pyramidal cell layer) or GABAergic (e.g. stria- 1997). These results suggest that GluD1 plays a role at tum) brain regions indicates that GluD1 expression is not glutamatergic synapses throughout the brain, possibly in restricted to either excitatory or inhibitory neurons, as synapse organization and/or in mediating excitatory cur- assessed by single cell RT-PCR experiments on cortical rents as reported for GluD2 (Ady et al. 2014; Kakegawa neurons. Furthermore, the intense labeling observed in the et al. 2008; Kato et al. 2012; Matsuda et al. 2010; Uemura dorsal raphe and some hypothalamic nuclei raises the et al. 2010).

123 Brain Struct Funct

GluD1 and GluD2 are widely co-expressed in the adult GluD1 and GluD2 are upregulated during postnatal brain development

We also found expression of GluD2 mRNA and protein in We found that the expression of GluD1 and GluD2 largely all adult brain regions examined, as assessed by control increase during the postnatal development of the cortex experiments on GRID2HO/HO mice. However, GluD2 and the cerebellum. This finding is at odds with the rapid expression levels were more contrasted than those of decrease of GluD1 expression after birth reported from GluD1. Indeed, GluD2 was markedly more abundant in in situ hybridization experiments (Lomeli et al. 1993). the cerebellum than in the forebrain, consistent with its This discrepancy can be explained by the difficulty to high enrichment in Purkinje cells (Araki et al. 1993; normalize in situ hybridization signals due to lack of Yamasaki et al. 2011). Interestingly, this distribution is internal control. In contrast, the present results derive reminiscent of that of mGlu1, which forms a complex with from careful normalization of mRNA and protein abun- GluD2 (Kato et al. 2012) and triggers GluD2-mediated dance to the expression of multiple reference genes, currents (Ady et al. 2014; Kato et al. 2012), and exhibits attesting to their validity. In the cortex, GluD1 and GluD2 similarly widespread expression with high enrichment in increased synchronously after birth to reach adult levels at Purkinje cells (Shigemoto et al. 1992). Expression of the postnatal day 28. This time-window is posterior to the GluD2 mRNA outside the cerebellum was nonetheless proliferation and migration of cortical neuron precursors sufficiently high to be detected in individual neurons using (Bayer and Altman 1991) and corresponds to the main single cell RT-PCR, as reported for mGlu1 (Cauli et al. period of cortical synapse formation (Blue and Parnavelas 2000). 1983). In the cerebellum, GluD1 and GluD2 increased These results show that GluD1 and GluD2 are widely asynchronously, consistent with their main expression in co-expressed in the brain. Quantification of the relative granule and Purkinje cells, respectively. GluD1 increase proportions of their mRNAs indicates that GluD1 is more occurred primarily after P14, at the onset of mossy fiber- abundant than GluD2 in the cortex, whereas GluD2 pre- granule cell synapse formation (Hamori and Somogyi dominates in the cerebellum. Regional western blots fur- 1983), whereas GluD2 already increased after P7 during ther suggest that GluD1 is also dominant in other forebrain the development of Purkinje cell dendrites and the for- regions tested, in which GluD1 and GluD2 levels are mation of parallel fiber synapses onto these cells. The similar to those measured in the cortex. Conversely, expression time-course of delta subunits and their sus- abundance of the two proteins may be more even in the tained expression in the adult are consistent with a role in mesencephalon where GluD1 level is lower, and GluD2 the maturation and adult function of neuronal networks level higher, than in the cortex. Interstingly, The anti- rather than in their embryonic development, as reported GluD1/2 antibody used by Mayat et al. (1995) produces for GluD2 (Ady et al. 2014; Kakegawa et al. 2008; intense labeling in the dorsal cochlear nucleus at a level Kashiwabuchi et al. 1995; Kato et al. 2012; Matsuda et al. similar to that observed in the cerebellum. GluD1–2 levels 2010; Uemura et al. 2010) and as suggested by the present and GluD1/GluD2 ratio in the dorsal cochlear nucleus were observation of GluD1 localization at postsynaptic not examined in the present study. Nonetheless, the cere- densities. bellum-like properties of this brainstem nucleus (Bell et al. 2008) suggest that a high GluD2 level may account for its GluD1 and GluD2 contribution to the function strong immunolabeling described by Mayat et al. (1995). of neuronal networks This suggests in turn that GluD2 may predominate over GluD1 in discrete brain regions or neuron types outside the The present results and those of earlier studies indicate that cerebellum. the major role of GluD subunits pertains to the organization It is noteworthy that regional co-expression does not and/or function of excitatory synapses (Ady et al. 2014; imply cellular co-expression. Although we detected Kakegawa et al. 2008; Kato et al. 2012; Matsuda et al. GluD1 and GluD2 co-expression in our sample of indi- 2010; Uemura et al. 2010). Although widespread regional vidual cortical and hippocampal neurons, they rather co-expression of GluD1 and GluD2 raises the possibility appear segregated to different neuronal types in the cer- that they co-assemble as heteromers, data obtained so far ebellum. Indeed, we found that GluD1 is present in suggest that co-assembly is not required for function. granule cells, which are reportedly devoid of GluD2 Indeed, GluD2 fulfills its functions at the parallel fiber- (Araki et al. 1993; Yamasaki et al. 2011). Conversely, we Purkinje cell synapse in the absence of GluD1 (Ady et al. observed no conspicuous expression of GluD1 in Purkinje 2014; Kakegawa et al. 2008; Matsuda et al. 2010; Uemura cells, which abundantly express GluD2 (Araki et al. et al. 2010). Furthermore, the marked differences in rela- 1993). tive abundance of the two subunits we observed at the 123 Brain Struct Funct regional level indicate that GluDs occur mainly as homo- Araki K, Meguro H, Kushiya E, Takayama C, Inoue Y, Mishina M mers. This suggests in turn that GluD1 and GluD2 have (1993) Selective expression of the glutamate receptor channel delta 2 subunit in cerebellar Purkinje cells. Biochem Biophys distinct contributions to brain functions. Indeed, mice Res Commun 197(3):1267–1276. doi:10.1006/bbrc.1993.2614 lacking GRID1 or GRID2 genes exhibited modest, non- Bayer S, Altman J (1991) Neocortical development. Raven Press, significant, changes of the remaining GluD subunit New York expression, indicating that each subunit does not effec- Bell CC, Han V, Sawtell NB (2008) Cerebellum-like structures and their implications for cerebellar function. Annu Rev Neurosci tively compensate for the absence of the other. Nonethe- 31:1–24. doi:10.1146/annurev.neuro.30.051606.094225 -/- -/- less, GRID1 and GRID2 mice exhibit changes in the Blue ME, Parnavelas JG (1983) The formation and maturation of expression of other genes related to excitatory transmission synapses in the visual cortex of the rat. I. Qualitative analysis. [e.g. other iGluRs, (Yadav et al. 2011, 2013; Yamasaki J Neurocytol 12(4):599–616 Cauli B, Audinat E, Lambolez B, Angulo MC, Ropert N, Tsuzuki K, et al. 2011)], which likely compensate for the absence of Hestrin S, Rossier J (1997) Molecular and physiological the cognate GluD subunit at glutamatergic synapses. Mice diversity of cortical nonpyramidal cells. J Neurosci 17(10): lacking functional GluD1 or GluD2 show different patterns 3894–3906 of behavioral and neurochemical alterations. As expected, Cauli B, Porter JT, Tsuzuki K, Lambolez B, Rossier J, Quenet B, Audinat E (2000) Classification of fusiform neocortical inter- GRID2 deletions prominently affects the cerebellum neurons based on unsupervised clustering. Proc Natl Acad Sci (Guastavino et al. 1990; Kashiwabuchi et al. 1995; Lal- USA 97(11):6144–6149 (97/11/6144) ouette et al. 2001; Yamasaki et al. 2011), but may result in Connors BW, Gutnick MJ (1990) Intrinsic firing patterns of diverse more subtle alterations of other structures that express neocortical neurons. Trends Neurosci 13(3):99–104 pii:0166- -/- 2236(90)90185-D lower levels of GluD2. Conversely, GRID1 mice Fallin MD, Lasseter VK, Avramopoulos D, Nicodemus KK, Woly- exhibit multiple alterations of cognitive, emotional and niec PS, McGrath JA, Steel G, Nestadt G, Liang KY, Huganir social behaviors (Yadav et al. 2012, 2013), in agreement RL, Valle D, Pulver AE (2005) Bipolar I disorder and with the association of human GRID1 gene mutations with schizophrenia: a 440-single-nucleotide polymorphism screen of 64 candidate genes among Ashkenazi Jewish case-parent trios. schizophrenia (Fallin et al. 2005; Gao et al. 2007; Guo Am J Hum Genet 77(6):918–936. doi:10.1086/497703 et al. 2007; Treutlein et al. 2009). Likewise, neurochemical Ferezou I, Hill EL, Cauli B, Gibelin N, Kaneko T, Rossier J, alterations have been assessed in several brain regions of Lambolez B (2007) Extensive overlap of mu-opioid and these mice (Yadav et al. 2012, 2013), consistent with nicotinic sensitivity in cortical interneurons. Cereb Cortex 17(8):1948–1957. doi:10.1093/cercor/bhl104 widespread GluD1 expression. Hence, our results provide Freund TF, Buzsaki G (1996) Interneurons of the hippocampus. molecular and anatomical bases for the involvement of Hippocampus 6(4):347–470. doi:10.1002/(SICI)1098-1063(1996) GluD subunits in multiple brain functions. Gallopin T, Geoffroy H, Rossier J, Lambolez B (2006) Cortical sources of CRF, NKB, and CCK and their effects on pyramidal Acknowledgments The authors thank Delphine Bouteiller, Thierry cells in the neocortex. Cereb Cortex 16(10):1440–1452. doi:10. Galli, Yannick Marie and Agathe Verraes for their valuable help. We 1093/cercor/bhj081 thank Jian Zuo and Carole Levenes for sharing GRID1 KO and Gao J, Maison SF, Wu X, Hirose K, Jones SM, Bayazitov I, Tian Y, GRID2 HO mouse line respectively. S.N. is a fellow of the Nakajima Mittleman G, Matthews DB, Zakharenko SS, Liberman MC, Zuo Foundation. This work was supported by grants from Agence Na- J (2007) Orphan glutamate receptor delta1 subunit required for tionale de la Recherche (ANR# BLAN-SVSE4-LS-110624, ‘‘IHU high-frequency hearing. Mol Cell Biol 27(12):4500–4512. Institut de Neurosciences Translationnelles de Paris’’; ‘‘Investisse- doi:10.1128/MCB.02051-06 ments d’avenir’’ ANR-10-IAIHU-06), Fondation pour la Recherche Gras C, Amilhon B, Lepicard EM, Poirel O, Vinatier J, Herbin M, sur le Cerveau/Rotary Club de France, and from the Spanish Ministry Dumas S, Tzavara ET, Wade MR, Nomikos GG, Hanoun N, of Education and Science (BFU2012-38348 and CONSOLIDER Saurini F, Kemel ML, Gasnier B, Giros B, El Mestikawy S CSD2008-00005) to R.L. (2008) The vesicular glutamate transporter VGLUT3 synergizes striatal acetylcholine tone. Nat Neurosci 11(3):292–300. doi:10. Conflict of interest The authors declare no competing financial 1038/nn2052 interests. Guastavino JM, Sotelo C, Damez-Kinselle I (1990) Hot-foot murine mutation: behavioral effects and neuroanatomical alterations. Brain Res 523(2):199–210 pii:0006-8993(90)91488-3 Guo SZ, Huang K, Shi YY, Tang W, Zhou J, Feng GY, Zhu SM, Liu References HJ, Chen Y, Sun XD, He L (2007) A case-control association study between the GRID1 gene and schizophrenia in the Chinese Ady V, Perroy J, Tricoire L, Piochon C, Dadak S, Chen X, Dusart I, Northern Han population. Schizophr Res 93(1–3):385–390. Fagni L, Lambolez B, Levenes C (2014) Type 1 metabotropic doi:10.1016/j.schres.2007.03.007 glutamate receptors (mGlu1) trigger the gating of GluD2 delta Hamori J, Somogyi J (1983) Differentiation of cerebellar mossy fiber glutamate receptors. EMBO Rep 15(1):103–109. doi:10.1002/ synapses in the rat: a quantitative electron microscope study. embr.201337371 J Comp Neurol 220(4):365–377. doi:10.1002/cne.902200402 Angulo MC, Lambolez B, Audinat E, Hestrin S, Rossier J (1997) Hellemans J, Mortier G, De Paepe A, Speleman F, Vandesompele J Subunit composition, kinetic, and permeation properties of (2007) qBase relative quantification framework and software for AMPA receptors in single neocortical nonpyramidal cells. management and automated analysis of real-time quantitative J Neurosci 17(17):6685–6696 PCR data. Genome Biol 8(2):R19. doi:10.1186/gb-2007-8-2-r19

123 Brain Struct Funct

Kakegawa W, Miyazaki T, Emi K, Matsuda K, Kohda K, Motohashi Pachot A, Blond JL, Mougin B, Miossec P (2004) Peptidylpropyl J, Mishina M, Kawahara S, Watanabe M, Yuzaki M (2008) isomerase B (PPIB): a suitable reference gene for mRNA Differential regulation of synaptic plasticity and cerebellar motor quantification in peripheral whole blood. J Biotechnol 114(1–2): learning by the C-terminal PDZ-binding motif of GluRdelta2. 121–124. doi:10.1016/j.jbiotec.2004.07.001 J Neurosci 28(6):1460–1468. doi:10.1523/JNEUROSCI.2553- Ryu K, Yokoyama M, Yamashita M, Hirano T (2012) Induction of 07.2008 excitatory and inhibitory presynaptic differentiation by GluD1. Karagiannis A, Gallopin T, David C, Battaglia D, Geoffroy H, Biochem Biophys Res Commun 417(1):157–161. doi:10.1016/j. Rossier J, Hillman EM, Staiger JF, Cauli B (2009) Classification bbrc.2011.11.075 of NPY-expressing neocortical interneurons. J Neurosci Schmid SM, Hollmann M (2008) To gate or not to gate: are the delta 29(11):3642–3659. doi:10.1523/JNEUROSCI.0058-09.2009 subunits in the glutamate receptor family functional ion chan- Kashiwabuchi N, Ikeda K, Araki K, Hirano T, Shibuki K, Takayama nels? Mol Neurobiol 37(2–3):126–141. doi:10.1007/s12035-008- C, Inoue Y, Kutsuwada T, Yagi T, Kang Y et al (1995) 8025-0 Impairment of motor coordination, Purkinje cell synapse Shigemoto R, Nakanishi S, Mizuno N (1992) Distribution of the formation, and cerebellar long-term depression in GluR delta 2 mRNA for a metabotropic glutamate receptor (mGluR1) in the mutant mice. Cell 81(2):245–252 pii:0092-8674(95)90334-8 central nervous system: an in situ hybridization study in adult Kato AS, Knierman MD, Siuda ER, Isaac JT, Nisenbaum ES, Bredt and developing rat. J Comp Neurol 322(1):121–135. doi:10. DS (2012) Glutamate receptor delta2 associates with metabo- 1002/cne.903220110 tropic glutamate receptor 1 (mGluR1), protein kinase Cgamma, Szabo A, Somogyi J, Cauli B, Lambolez B, Somogyi P, Lamsa KP and canonical transient receptor potential 3 and regulates (2012) Calcium-permeable AMPA receptors provide a common mGluR1-mediated synaptic transmission in cerebellar Purkinje mechanism for LTP in glutamatergic synapses of distinct neurons. J Neurosci 32(44):15296–15308. doi:10.1523/JNEUR hippocampal interneuron types. J Neurosci 32(19):6511–6516. OSCI.0705-12.2012 doi:10.1523/JNEUROSCI.0206-12.2012 Kuroyanagi T, Yokoyama M, Hirano T (2009) Postsynaptic glutamate Takayama C, Nakagawa S, Watanabe M, Mishina M, Inoue Y (1995) receptor delta family contributes to presynaptic terminal differ- Light- and electron-microscopic localization of the glutamate entiation and establishment of synaptic transmission. Proc Natl receptor channel delta 2 subunit in the mouse Purkinje cell. Acad Sci USA 106(12):4912–4916. doi:10.1073/pnas.090089 Neurosci Lett 188(2):89–92. pii:030439409511403J 2106 Tonnes J, Stierli B, Cerletti C, Behrmann JT, Molnar E, Streit P Lalouette A, Lohof A, Sotelo C, Guenet J, Mariani J (2001) (1999) Regional distribution and developmental changes of Neurobiological effects of a null mutation depend on genetic GluR1-flop protein revealed by monoclonal antibody in rat brain. context: comparison between two hotfoot alleles of the delta-2 J Neurochem 73(5):2195–2205 ionotropic glutamate receptor. Neuroscience 105(2):443–455. Treutlein J, Muhleisen TW, Frank J, Mattheisen M, Herms S, Ludwig pii:S0306452201001932 KU, Treutlein T, Schmael C, Strohmaier J, Bosshenz KV, Breuer Lambolez B, Audinat E, Bochet P, Crepel F, Rossier J (1992) AMPA R, Paul T, Witt SH, Schulze TG, Schlosser RG, Nenadic I, Sauer receptor subunits expressed by single Purkinje cells. Neuron H, Becker T, Maier W, Cichon S, Nothen MM, Rietschel M 9(2):247–258. pii:0896-6273(92)90164-9 (2009) Dissection of phenotype reveals possible association Lambolez B, Ropert N, Perrais D, Rossier J, Hestrin S (1996) between schizophrenia and glutamate receptor delta 1 (GRID1) Correlation between kinetics and RNA splicing of alpha-amino- gene promoter. Schizophr Res 111(1–3):123–130. doi:10.1016/j. 3-hydroxy-5-methylisoxazole-4-propionic acid receptors in neo- schres.2009.03.011 cortical neurons. Proc Natl Acad Sci USA 93(5):1797–1802 Uemura T, Lee SJ, Yasumura M, Takeuchi T, Yoshida T, Ra M, Landsend AS, Amiry-Moghaddam M, Matsubara A, Bergersen L, Taguchi R, Sakimura K, Mishina M (2010) Trans-synaptic Usami S, Wenthold RJ, Ottersen OP (1997) Differential interaction of GluRdelta2 and Neurexin through Cbln1 mediates localization of delta glutamate receptors in the rat cerebellum: synapse formation in the cerebellum. Cell 141(6):1068–1079. coexpression with AMPA receptors in parallel fiber-spine doi:10.1016/j.cell.2010.04.035 synapses and absence from climbing fiber-spine synapses. Vandesompele J, De Preter K, Pattyn F, Poppe B, Van Roy N, De J Neurosci 17(2):834–842 Paepe A, Speleman F (2002) Accurate normalization of real-time Lomeli H, Sprengel R, Laurie DJ, Kohr G, Herb A, Seeburg PH, quantitative RT-PCR data by geometric averaging of multiple Wisden W (1993) The rat delta-1 and delta-2 subunits extend internal control genes. Genome Biol 3(7):RESEARCH0034 the excitatory amino acid receptor family. FEBS Lett 315(3): Weldon S, Ambroz K, Schutz-Geschwender A, Olive DM (2008) 318–322. pii:0014-5793(93)81186-4 Near-infrared fluorescence detection permits accurate imaging of Lujan R, Nusser Z, Roberts JD, Shigemoto R, Somogyi P (1996) loading controls for Western blot analysis. Anal Biochem Perisynaptic location of metabotropic glutamate receptors 375(1):156–158. doi:10.1016/j.ab.2007.11.035 mGluR1 and mGluR5 on dendrites and dendritic spines in the Yadav R, Rimerman R, Scofield MA, Dravid SM (2011) Mutations in rat hippocampus. Eur J Neurosci 8(7):1488–1500 the transmembrane domain M3 generate spontaneously open Matsuda K, Miura E, Miyazaki T, Kakegawa W, Emi K, Narumi S, orphan glutamate delta1 receptor. Brain Res 1382:1–8. doi:10. Fukazawa Y, Ito-Ishida A, Kondo T, Shigemoto R, Watanabe M, 1016/j.brainres.2010.12.086 Yuzaki M (2010) Cbln1 is a ligand for an orphan glutamate Yadav R, Gupta SC, Hillman BG, Bhatt JM, Stairs DJ, Dravid SM receptor delta2, a bidirectional synapse organizer. Science (2012) Deletion of glutamate delta-1 receptor in mouse leads to 328(5976):363–368. doi:10.1126/science.1185152 aberrant emotional and social behaviors. PLoS One 7(3):e32969. Mayat E, Petralia RS, Wang YX, Wenthold RJ (1995) Immunopre- doi:10.1371/journal.pone.0032969 cipitation, immunoblotting, and immunocytochemistry studies Yadav R, Hillman BG, Gupta SC, Suryavanshi P, Bhatt JM, Pavuluri suggest that glutamate receptor delta subunits form novel R, Stairs DJ, Dravid SM (2013) Deletion of glutamate delta-1 postsynaptic receptor complexes. J Neurosci 15(3 Pt 2): receptor in mouse leads to enhanced working memory and deficit 2533–2546 in fear conditioning. PLoS One 8(4):e60785. doi:10.1371/ Orth A, Tapken D, Hollmann M (2013) The delta subfamily of journal.pone.0060785 glutamate receptors: characterization of receptor chimeras and Yamasaki M, Miyazaki T, Azechi H, Abe M, Natsume R, Hagiwara mutants. Eur J Neurosci 37(10):1620–1630. doi:10.1111/ejn.12193 T, Aiba A, Mishina M, Sakimura K, Watanabe M (2011)

123 Brain Struct Funct

Glutamate receptor delta2 is essential for input pathway- Yamazaki M, Araki K, Shibata A, Mishina M (1992) Molecular dependent regulation of synaptic AMPAR contents in cerebellar cloning of a cDNA encoding a novel member of the mouse Purkinje cells. J Neurosci 31(9):3362–3374. doi:10.1523/ glutamate receptor channel family. Biochem Biophys Res JNEUROSCI.5601-10.2011 Commun 183(2):886–892 pii:0006-291X(92)90566-4

123

V. REFERENCES A Abel J (1899). "Ueber den blutdrucksteigernden Bestandtheil der Nebenniere, das Epinephrin". Zeitschrift für physiologische Chemie 28: 318–361. Adams SR1, Harootunian AT, Buechler YJ, Taylor SS, Tsien RY., (1991) Fluorescence ratio imaging of cyclic AMP in single cells. Nature 349:694-7. Agnati, L.F., Fuxe, K. (2000). Volume transmission as a key feature of information handling in the central nervous system possible new interpretative value of the Turing’s B-type machine. Prog. Brain Res. 125, 3– 19. Ahlquist R.P. (1948). A study of the adrenotropic receptors. American Journal of Physiology 153: 586–600. Allen MD, Zhang J (2006) Subcellular dynamics of protein kinase A activity visualized by FRET-based reporters. Biochem Biophys Res Commun. 348(2):716-21. Ariano MA, Sibley DR. 1994. Dopamine receptor distribution in the rat CNS: elucidation using anti-peptide antisera directed against D1A and D3 subtypes. Brain Res 649:95-110. Atasoy D, AponteY, Hong Su H, Sternson SM (2008). A FLEX Switch Targets Channelrhodopsin-2 to Multiple Cell Types for Imaging and Long-Range Circuit Mapping. The Journal of Neuroscience, 9, 28(28): 7025- 7030 Axelrod J (1962). O-Methylation of epinephrine and other catechols. Science 126 (3270): 400–401.

B Beaulieu JM, Gainetdinov RR. (2011) The physiology, signaling, and pharmacology of dopamine receptors. Pharmacol Rev. 63(1):182-217. Benarroch EE.(2009) The locus ceruleus norepinephrine system: functional organization and potential clinical significance. Neurology. Nov 17;73(20):1699-704. Benavides-Piccione R, Fernaud-Espinosa I, Robles V, Yuste R, DeFelipe J. (2013) Age-based comparison of human dendritic spine structure using complete three-dimensional reconstructions. Cereb Cortex. 23 (8):1798-810. Berger B, Gaspar P, Verney C. (1991). Dopaminergic innervation of the cerebral cortex: unexpected differences between rodents and primates. Trends Neurosci 14:21-27. Bergson C, Mrzljak L, Smiley JF, Pappy M, Levenson R, Goldman-Rakic PS.(1995) Regional, cellular, and subcellular variations in the distribution of D1 and D5 dopamine receptors in primate brain. J Neurosci. 15(12):7821-36. Berndt A, Schoenenberger P, Mattis J, Tye KM, Deisseroth K, Hegemann P, Oertner TG. High-efficiency channelrhodopsins for fast neuronal stimulation at low light levels. Proc Natl Acad Sci U S A. 2011 May 3;108(18):7595-600. Berridge CW, Waterhouse BD (2003) The locus coeruleus–noradrenergic system: modulation of behavioral state and state-dependent cognitive processes Brain Research Reviews, 42 (1) , 33-84 Blaschko H , Richter D. und Schlossmann H (1937). The oxidation of adrenaline and other amines. Biochemical Journal 31: 2187–2196. Blaschko H. (1939). "The specific action of l-dopa decarboxylase". The Journal of Physiology 96: 50–51 Bonnot A, Guiot E, Hepp R, Cavellini L, Tricoire L, Lambolez B. 2014. Single-fluorophore biosensors based on conformation-sensitive GFP variants. FASEB J 28:1375-1385. Bouthenet ML, Martres MP, Sales N, Schwartz JC. 1987. A detailed mapping of dopamine D-2 receptors in rat central nervous system by autoradiography with [125I]iodosulpride. Neuroscience 20:117-155. Bouret S and Sara SJ (2010), Scholarpedia, 5(3):2845. Boyden E, Zhang F, Bamberg E, Nagel G, Deisseroth K (2005) Millisecond-timescale, genetically targeted optical control of neural activity Nature Neuroscience 8, 1263 - 1268 Bylund DB. 1992. Subtypes of alpha 1- and alpha 2-adrenergic receptors. FASEB J 6:832-839.

C Cachope R, Mateo Y, Mathur BN, Irving J, Wang HL, Morales M, Lovinger DM, Cheer JF. Selective activation of cholinergic interneurons enhances accumbal phasic dopamine release: setting the tone for reward processing.Cell Rep. 2012 Jul 26;2(1):33-41. Cajal SR (reprint 1995) Histology of the Nervous System of man and vertebrate, translation by Swanson N S and Swanson LW). Oxford University Press Cameron DL and William JT. Dopamine D1 receptors facilitate transmitter release (1993). Nature 366:344-347 Carlsson A. (1959). The occurrence, distribution and physiological role of catcholamines in the nervous system. Pharmacological Reviews 11: 490–493.

Castro LR, Gervasi N, Guiot E, Cavellini L, Nikolaev VO, Paupardin-Tritsch D, Vincent P. Type 4 phosphodiesterase plays different integrating roles in different cellular domains in pyramidal cortical neurons. J Neurosci. 2010 Apr 28;30(17):6143-51. Cauli B, Porter JT, Tsuzuki K, Lambolez B, Rossier J, Quenet B, Audinat E.(2000) Classification of fusiform neocortical interneurons based on unsupervised clustering. Proc Natl Acad Sci U S A. 23;97(11):6144-9. Chandrasekera PC, Wan TC, Gizewski ET, Auchampach JA, Lasley RD. (2013) Adenosine A1 receptors heterodimerize with β1- and β2-adrenergic receptors creating novel receptor complexes with altered G protein coupling and signaling. Cell Signal. 25(4):736-42. Chen BE, Lendvai B, Nimchinsky EA, Burbach B, Fox K, Svoboda K.(2000) Imaging high-resolution structure of GFP-expressing neurons in neocortex in vivo. Learn Mem.;7(6):433-41. Ciliax BJ, Nash N, Heilman C, Sunahara R, Hartney A, Tiberi M, Rye DB, Caron MG, Niznik HB, Levey AI. (2000). Dopamine D(5) receptor immunolocalization in rat and monkey brain. Synapse 37:125-145. Cintra L, Dfaz-Cintra S, Kemper T. And P. J. Morgane PJ. (1982) Nucleus Locus Coeruleus: A Morphometric Golgi Study in Rats of Three Age Groups. Brain Research, 247 17-28 Cotecchia S, Stanasila L, Diviani D. Protein-protein interactions at the adrenergic receptors.(2012) Curr Drug Targets. 2012 Jan;13(1):15-27.

D Dahlin A, Xia L, Kong W, Hevner R, Wang J. (2007). Expression and immunolocalization of the plasma membrane monoamine transporter in the brain. Neuroscience. May 25;146(3):1193-211. Defagot MC, Malchiodi EL, Villar MJ, Antonelli MC. (1997). Distribution of D4 dopamine receptor in rat brain with sequence-specific antibodies. Brain Res Mol Brain Res 45:1-12. Defelipe J (2011) The evolution of the brain, the human nature of cortical circuits, and intellectual creativity. Front Neuroanat. 2011 May 16;5:29. Delmas P, Abogadie FC, Milligan G, Buckley NJ, Brown DA. (1999) Betagamma dimers derived from Go and Gi proteins contribute different components of adrenergic inhibition of Ca2+ channels in rat sympathetic neurones. J Physiol. 1999 Jul 1;518 ( Pt 1):23-36. De Mei C, Ramos M, Iitaka C, and Borrelli E (2009) Getting specialized: presynaptic and postsynaptic dopamine D2 receptors. Curr Opin Pharmacol 9:53–58. Depry C, Allen MD, Zhang J. Visualization of PKA activity in plasma membrane microdomains. Mol Biosyst. 2011 Jan;7(1):52-8 Descarries L, Lemay B, Doucet G, Berger B. 1987. Regional and laminar density of the dopamine innervation in adult rat cerebral cortex. Neuroscience 21:807-824. Devoto P, Flore G, Saba P, Fa M, Gessa GL. 2005. Stimulation of the locus coeruleus elicits noradrenaline and dopamine release in the medial prefrontal and parietal cortex. J Neurochem 92:368-374. Devoto P, Flore G, Saba P, Castelli MP, Piras AP, Luesu W, Viaggi MC, Ennas MG, Gessa GL. 2008. 6- Hydroxydopamine lesion in the ventral tegmental area fails to reduce extracellular dopamine in the cerebral cortex. J Neurosci Res 86:1647-1658. Drobac E1, Tricoire L, Chaffotte AF, Guiot E, Lambolez B. Calcium imaging in single neurons from brain slices using bioluminescent reporters. J Neurosci Res. 2010 Mar;88(4):695-711. Duan H, Wang J. (2010) Selective transport of monoamine neurotransmitters by human plasma membrane monoamine transporter and organic cation transporter 3. J Pharmacol Exp Ther. Dec; 335(3):743-53 Duman, R. S., & Nestler, E. J. (1995). Signal transduction pathways for catecholamine receptors. In F. E. Bloom, & D. J. Kupfer (Eds.), Psychopharmacology: The Fourth Generation of Progress (pp. 303−320). N.Y.: Raven Press.

E Ehrengruber MU. Alphaviral gene transfer in neurobiology.(2002) Brain Res Bull. 59(1):13-22. Eisenhofer G, Kopin IJ, Goldstein DS. (2004) Leaky catecholamine stores: undue waste or a stress response coping mechanism? Ann NY Acad Sci 1018: 224–230. Evans B.A., Papaioannou M., Hamilton S., Summers R.J. (1999). Alternative splicing generates two isoforms of the b3-adrenoceptor which are differentially expressed in mouse tissues. British Journal of Pharmacology 127, 1525-1531 F Fiorentini C, Busi C, Gorruso E, Gotti C, Spano P, Missale C. (2008) Reciprocal regulation of dopamine D1 and D3 receptor function and trafficking by heterodimerization. Mol Pharmacol. 74(1):59-69. Fremeau R, Duncan G, Fornarettol MG, Allen Dearry A, Gingrich J, Breese G, And Marc G. Caronii M (1991) Localization of D1 dopamine receptor mrRNA in brain supports a role in cognitive, affective, and neuroendocrine aspects of dopaminergic neurotransmission Proc. Natl. Acad. Sci. USA 88: 3772-3776.

Fuxe K,Ferre S,Zoli M,Agnati LF. (1998) Integrated events in central dopamine transmission as analyzed at multiple levels. Evidence for intramembrane adenosine A2A/dopamine D2 and adenosine A1/dopamine D1 receptor interactions in the basal ganglia. Brain Res Brain Res Rev 26:258–73. Fuxe K, Tarakanov A, Romero Fernandez W, Ferraro L, Tanganelli S, Filip M, Agnati LF, Garriga P, Diaz- Cabiale Z, Borroto-Escuela DO. (2014) Diversity and Bias through Receptor-Receptor Interactions in GPCR Heteroreceptor Complexes. Focus on Examples from Dopamine D2 Receptor Heteromerization. Front Endocrinol (Lausanne). 13;5:71.

G Garmashova N, Gorchakov R, Frolova E, Frolov I. Sindbis virus nonstructural protein nsP2 is cytotoxic and inhibits cellular transcription. J Virol. 2006 Jun;80(12):5686-96. Gervasi N, Hepp R, Tricoire L, Zhang J, Lambolez B, Paupardin-Tritsch D, Vincent P. Dynamics of protein kinase A signaling at the membrane, in the cytosol, and in the nucleus of neurons in mouse brain slices.J Neurosci. 2007 Mar 14;27(11):2744-50. Gilsbach R. Albarran-Juarez J and Hein L (2011) Pre- versus Postsynaptic Signaling by a2-Adrenoceptors, Current Topics in Membranes, 67:139-160. Giros B, Martres MP, Pilon C, Sokoloff P, and Schwartz JC (1991) Shorter variants of the D3 dopamine receptor produced through various patterns of alternative splicing. Biochem Biophys Res Commun 176:1584–1592. Gonzalez S, Moreno-Delgado D, Moreno E, Perez-Capote K, Franco R, Mallol J, Cortes A, Casado V, Lluis C, Ortiz J, Ferre S, Canela E, McCormick PJ. 2012. Circadian-related heteromerization of adrenergic and dopamine D(4) receptors modulates melatonin synthesis and release in the pineal gland. PLoS Biol 10:e1001347. Grandy DK, Zhang YA, Bouvier C, Zhou QY, Johnson RA, Allen L, Buck K, Bunzow JR, Salon J, and Civelli O (1991) Multiple human D5 dopamine receptor genes: a functional receptor and two pseudogenes. Proc Natl Acad Sci USA 88:9175–9179. Gingrich JA and Caron MG (1993) Recent advances in the molecular biology of dopamine receptors. Annu Rev Neurosci 16:299–321. Grundemann D, Schechinger B, Rappold GA, Schomig E. Molecular identification of the corticosterone- sensitive extraneuronal catecholamine transporter. Nature Neurosci 1998; 1: 349–351 Grynkiewicz G, Poenie M, Tsien RY.(1985). A new generation of Ca2+ indicators with greatly improved fluorescence properties.” J. Biol. Chem. 260 :3440-50. Gunaydin LA, Yizhar O, Berndt A, Sohal VS, Deisseroth K & Hegemann P (2010). Ultrafast optogenetic control. Nat Neurosci 13, 387–392.

H Han, X. et al. Millisecond-timescale optical control of neural dynamics in the nonhuman primate brain. Neuron 62, 191–198 (2009). Hasbi A, O'Dowd BF, George SR. (2010) Heteromerization of dopamine D2 receptors with dopamine D1 or D5 receptors generates intracellular calcium signaling by different mechanisms. Curr Opin Pharmacol. 10(1):93- 9. Heimer L. (1994) The Human Brain and Spinal Cord: Functional Neuroanatomy and Dissection Guide, 2nd ed. New York: Springer. Hepp R, Audrey Hay Y, Aguado C, Lujan R, Dauphinot L, Potier MC, Nomura S, Poirel O, El Mestikawy S, Lambolez B, Tricoire L. Glutamate receptors of the delta family are widely expressed in the adult brain. Brain Struct Funct. 2014 Jul 8. [Epub ahead of print] Herbst KJ, Allen MD, Zhang J. Luminescent kinase activity biosensors based on a versatile bimolecular switch. J Am Chem Soc. 2011 Apr 20;133(15):5676-9 Hernandez-Lopez S, Tkatch T, Perez-Garci E, Galarraga E, Bargas J, Hamm H, and Surmeier DJ (2000) D2 dopamine receptors in striatal medium spiny neurons reduce L-type Ca2+ currents and excitability via a novel PLC[beta]1-IP3- calcineurin-signaling cascade. J Neurosci 20:8987–8995. Hersch SM, Yi H, Heilman CJ, Edwards RH, Levey AI (1997). "Subcellular localization and molecular topology of the dopamine transporter in the striatum and substantia nigra". J. Comp. Neurol. 388 (2): 211–227. Hu E, Demmou L, Cauli B, Gallopin T, Geoffroy H, Harris-Warrick RM, Paupardin-Tritsch D, Lambolez B, Vincent P, Hepp R. VIP, CRF, and PACAP act at distinct receptors to elicit different cAMP/PKA dynamics in the neocortex. Cereb Cortex. 2011 Mar;21(3):708-18. Huang Q, Zhou D, Chase K, Gusella JF, Aronin N, DiFiglia M. (1992). Immunohistochemical localization of the D1 dopamine receptor in rat brain reveals its axonal transport, pre- and postsynaptic localization, and prevalence in the basal ganglia, limbic system, and thalamic reticular nucleus. Proc Natl Acad Sci U S A 89:11988-11992.

Huang Z J, Di Cristo G, Ango F. (.2007) Development of GABA innervations in the cerebral and cerebellar cortices Nature reviews 8:673-686 Hyde Salter H (1864) On Asthma: its pathology and therapy. Philadelphia, Blanchard.

I-J Jacobj C.(1892). "Beiträge zur physiologischen und pharmakologischen Kenntniss der Darmbewegungen mit besonderer Berücksichtigung der Beziehung der Nebenniere zu denselben". Archiv für experimentelle Pathologie und Pharmakologie 29 (3–4): 171–211. Jameson J.L , De Groot L (2010) Endocrinology: Adult and Pediatric 6th edition, Elsevier Johnson RT, McFarland HF, Levy SE. Age-dependent resistance to viral encephalitis: studies of infections due to Sindbis virus in mice. J Infect Dis. 1972 Mar;125(3):257-62. Johnson, SW; North, RA (May 1992). "Two types of neurone in the rat ventral tegmental area and their synaptic inputs.". The Journal of physiology 450: 455–68.

K Khan Z, Gutiérrez A, Martín R, Peñafiel A, Rivera A, De La Calle A.(1998) Differential regional and cellular distribution of dopamine D2-like receptors: an immunocytochemical study of subtype-specific antibodies in rat and human brain. J Comp Neurol. Dec 21;402(3):353-71. Khan Z, Gutierrez A, Martin R, Penafiel A., Rivera A.;De La Calle A.(2000) Dopamine D5 receptors of rat and human brain. Neuroscience 100( 4) 689–699 Kim J, Dittgen T, Nimmerjahn A, Waters J, Pawlak V, Helmchen F, Schlesinger S, Seeburg PH, Osten P. Sindbis vector SINrep(nsP2S726): a tool for rapid heterologous expression with attenuated cytotoxicity in neurons. J Neurosci Methods. 2004 Feb 15;133(1-2):81-90. Klein R, Ruttkowski B, Knapp E, Salmons B, Günzburg WH, Hohenadl C. WPRE-mediated enhancement of gene expression is promoter and cell line specificGene 372 (2006) 153–161 Koepsell H. (2004) Polyspecific organic cation transporters: their functions and interactions with drugs Trends Pharmacol. Sci., 25 pp. 375–381 Komatsu N, Aoki K, Yamada M, Yukinaga H, Fujita Y, Kamioka Y, Matsuda M. Development of an optimized backbone of FRET biosensors for kinases and GTPases. Mol Biol Cell. 2011 Dec;22(23):4647-56. Kulik A, Haentzsch A, Luckermann M, Reichelt W, Ballanyi K (1999) Neuron-glia signaling via α1- adrenoceptor-mediated Ca release in Bergmann glial cells in situ. J Neurosci 19:8401–8408. Kvetnansky R, Sabban EL, Palkovits M. (2009) Catecholaminergic systems in stress: structural and molecular genetic approaches. Physiol Rev. Apr;89(2):535-606.

L Latsari M, Dori I, Antonopoulos J, Chiotelli M, Dinopoulos A. 2002. Noradrenergic innervation of the developing and mature visual and motor cortex of the rat brain: a light and electron microscopic immunocytochemical analysis. J Comp Neurol 445:145-158. Lee SP, So CH, Rashid AJ, Varghese G, Cheng R, Lanca AJ, O’Dowd BF, George SR (2004) Dopamine D1 and D2 receptor coactivation generates a novel phospholipase C-mediated calcium signal. J Biol Chem 279:35671-35678. Lemberger T, Parlato R, Dassesse D, Westphal M, Casanova E, Turiault M, Tronche F, Schiffmann SN, Schutz G. 2007. Expression of Cre recombinase in dopaminoceptive neurons. BMC Neurosci 8:4. Lévesque D, Diaz J, Pilon C, Martres MP, Giros B, Souil E, Schott D, Morgat JL, Schwartz JC, Sokoloff P. (1992) Identification, characterization, and localization of the dopamine D3 receptor in rat brain using 7- [3H]hydroxy-N,N-di-n-propyl-2-aminotetralin. Proc Natl Acad Sci U S A. 1;89(17):8155-9. Lidow MS, Goldman-Rakic PS, Rakic P, Innis RB. (1989). Dopamine D2 receptors in the cerebral cortex: distribution and pharmacological characterization with [3H]raclopride. Proc Natl Acad Sci U S A 86:6412- 6416. Lin JY. (2012) Optogenetic excitation of neurons with channelrhodopsins: light instrumentation, expression systems, and channelrhodopsin variants. Prog Brain Res. 2012;196:29-47. Lorton D and Davis N. (1987) The distribution of beta1 and beta2 adrenergic receptors of normal and reeler mouse brain: an in vitro autoradiographic study. Neuroscience 23( 1), 199-210 Loughlin SE, Foote SL, Grzanna R.(1986) Efferent projections of nucleus locus coeruleus: morphologic subpopulations have different efferent targets. Neuroscience.18(2):307-19. Luedtke RR, Griffin SA, Conroy SS, Jin X, Pinto A, Sesack SR. 1999. Immunoblot and immunohistochemical comparison of murine monoclonal antibodies specific for the rat D1a and D1b dopamine receptor subtypes. J Neuroimmunol 101:170-187.

M Mansour A, Meador-Woodruff JH, Bunzow Civelli O, Akil H, and Watson SJ. (1990) Localization of D2 receptor mRNA and D1 and D2 receptor binding in the rat brain and pituitary: An in situ hybridization- receptor autoradiographic analysis. J. Neuro, 10(8): 2587-2600 Marek GJ, Aghajanian GK (1999) 5-HT2A receptor or alpha1-adrenoceptor activation induces excitatory postsynaptic currents in layer V pyramidal cells of the medial prefrontal cortex. Eur J Pharamcol 367:197– 206. Margolis EB, Lock H, Hjelmstad GO, Fields HL. 2006b. The ventral tegmental area revisited: Is there an electrophysiological marker for dopaminergic neurons? J. Physiol. 577(3) 907–24 Marie H. Sindbis Virus-Mediated In Vivo Expression of Recombinant CREB to Study Gene Function in Neuronal Plasticity and Behavior (2013) Multidisciplinary Tools for Investigating Synaptic Plasticity, Neuromethods, vol. 81,Elsevier Martres MP, Sales N, Bouthenet ML, Schwartz JC (1985).Localisation and pharmacological characterisation of D-2 dopamine receptors in rat cerebral neocortex and cerebellum using [125I]iodosulpride. Eur J Pharmacol. 3;118(3):211-9. Masson J, Sagné C, Hamon M, El Mestikawy S. Neurotransmitter transporters in the central nervous system. Pharmacol Rev. 1999 Sep;51(3):439-64. Matsumoto M, Yoshioka M, Togashi H, Hirokami M, Tochihara M, Ikeda T, Smith CB, Saito H. mu-Opioid receptors modulate noradrenaline release from the rat hippocampus as measured by brain microdialysis.Brain Res. 1994 Feb 4;636(1):1-8. McCune SK, Voigt MM, Hill JM. Expression of multiple alpha adrenergic receptor subtype messenger RNAs in the adult rat brain. Neuroscience. 1993 Nov;57(1):143-51. Mengod G, Villaró MT, Landwehrmeyer GB, Martinez-Mir MI, Niznik HB, Sunahara RK, Seeman P, O'Dowd BF, Probst A, Palacios JM. Visualization of dopamine D1, D2 and D3 receptor mRNAs in human and rat brain. Neurochem Int. 1992 Mar;20 Suppl:33S-43S. Missirlis, P. I.; Smailus, D. E.; Holt, R. A. (2006). "A high-throughput screen identifying sequence and promiscuity characteristics of the loxP spacer region in Cre-mediated recombination". BMC Genomics 7: 73. Miyawaki A, Llopis J, Heim R, McCaffery JM, Adams JA, Ikura M, Tsien RY., (1997), “Fluorescent indicators for Ca2+ based on green fluorescent proteins and calmodulin” Nature 388:882-7. Morrison JH, Grzanna R, Molliver ME, Coyle JT. 1978. The distribution and orientation of noradrenergic fibers in neocortex of the rat: an immunofluorescence study. J Comp Neurol 181:17-39. Morrison JH, Molliver ME, Grzanna R. 1979. Noradrenergic innervation of cerebral cortex: widespread effects of local cortical lesions. Science 205:313-316. Mountcastle VB (July 1957). "Modality and topographic properties of single neurons of cat's somatic sensory cortex". Journal of Neurophysiology 20 (4): 408–34. Mouradian, R. D., Seller, F. M., & Waterhouse, B. D. (1991). Noradrenergic potentiation of excitatory transmitter action in cerebrocortical slices: evidence of mediation by an alpha1-receptor-linked second messenger pathway. Brain Res 546, 83−95. Myöhänen TT, Schendzielorz N, Männistö PT (2010) Distribution of catechol-O-methyltransferase (COMT) proteins and enzymatic activities in wild-type and soluble COMT deficient mice. J Neurochem. 2010 Jun; 113(6):1632-43.

N Nagel, G., Ollig, D., Fuhrmann, M., Kateriya, S., Musti, A. M., Bamberg, E., Hegemann, P. (2002). Channelrhodopsin-1: a light-gated proton channel in green algae. Science 296, 2395–2398. Nagel, G., Szellas, T., Huhn, W., Kateriya, S., Adeishvili, N., Berthold, P., Ollig, D., Hegemann, P., Bamberg, E. (2003). Channelrhodopsin-2, a directly light-gated cation-selective membrane channel. Proc. Natl Acad. Sci. USA 100, 13940–13945. Nagel G, Brauner M, Liewald JF, Adeishvili N, Bamberg E, Gottschalk A (2005). Light activation of channelrhodopsin-2 in excitable cells of Caenorhabditis elegans triggers rapid behavioral responses. Curr Biol. 15(24):2279-84. Nagel G, Szellas T, Kateriya S, Adeishvili N, Hegemann P, Bamberg E. (2005b) Channelrhodopsins: directly light-gated cation channels Biochem Soc Trans.33(Pt 4):863-6.. Nair-Roberts RG, Chatelain-Badie SD, Benson E, White-Cooper H, Bolam JP, Ungless MA.Stereological estimates of dopaminergic, GABAergic and glutamatergic neurons in the ventral tegmental area, substantia nigra and retrorubral field in the rat. Neuroscience. 2008 Apr 9;152(4):1024-31. Nathanson, J.L., Yanagawa, Y., Obata, K. & Callaway, E.M. Preferential labeling of inhibitory and excitatory cortical neurons by endogenous tropism of adeno-associated virus and lentivirus vectors. Neuroscience 161, 441–450 (2009).

Nicholas AP, Pierribone V and Hokfelt T (1993) Distributions of mRNAs for Alpha-2 Adrenergic Receptor Subtypes in Rat Brain: An In Situ Hybridization Study. J Comp Neurol 328:575-594 Nicholas AP, Pierribone V and Hokfelt T (1993b) Cellular localization of the messenger RNA for beta1 and beta2 adrenergic receptors in the rat brain: an in situ hybridization study. Neuroscience 56(4):1023-1039 Nicholas AP, Hokfelt T, Pieribone VA. 1996. The distribution and significance of CNS adrenoceptors examined with in situ hybridization. Trends Pharmacol Sci 17:245-255.

O Oda S, Funato H, chi-Akahane S, Ito M, Okada A, Igarashi H, Yokofujita J, Kuroda M. 2010. Dopamine D5 receptor immunoreactivity is differentially distributed in GABAergic interneurons and pyramidal cells in the rat medial prefrontal cortex. Brain Res 1329:89-102. Oliver G, Schäfer E.A. (1894). "On the physiological action of extract of the suprarenal capsules". The Journal of Physiology 16: I–IV. PMC 1514529.

P Papagiakoumou E, Anselmi F, Bègue A, de Sars V, Glückstad J, Isacoff EY, Emiliani V. Scanless two-photon excitation of channelrhodopsin-2. Nat Methods. 2010 Oct;7(10):848-54. Papay R, Gaivin R, Jha A, McCune DF, McGrath JC, Rodrigo MC, Simpson PC, Doze VA, Perez DM. 2006. Localization of the mouse alpha1A-adrenergic receptor (AR) in the brain: alpha1AAR is expressed in neurons, GABAergic interneurons, and NG2 oligodendrocyte progenitors. J Comp Neurol 497:209-222. Paschalis A, Churchill L, Marina N, Kasymov V, Gourine A, Ackland G. 2009. beta1-Adrenoceptor distribution in the rat brain: an immunohistochemical study. Neurosci Lett 458:84-88. Paxinos G (2004) The Rat Nervous System, third edition. Elseviers

Q-R Rainbow TH, Parson B, Wolfe B (1984). Quantitative autoradiography of beta1 and beta2 adrenergic receptors. Proc. Natl. Acad. Sci. USA. 81, 1585-1589 Rebois RV, Maki K, Meeks JA, Fishman PH, Hebert TE, Northup JK. 2012. D2-like dopamine and beta- adrenergic receptors form a signaling complex that integrates Gs- and Gi-mediated regulation of adenylyl cyclase. Cell Signal 24:2051-2060. Richard A.F. Dixon R., Kobilka B. , Strader D. and 13 others (1986). "Cloning of the gene and cDNA for mammalian β-adrenergic receptor and homology with rhodopsin". Nature 321 (6065): 75–79. Rickgauer JP1, Tank DW. (2009)Two-photon excitation of channelrhodopsin-2 at saturation. Proc Natl Acad Sci U S A. 2009 Sep 1;106(35):15025-30. Rivera A, Penafiel A, Megias M, Agnati LF, Lopez-Tellez JF, Gago B, Gutierrez A, De La CA, Fuxe K. 2008. Cellular localization and distribution of dopamine D(4) receptors in the rat cerebral cortex and their relationship with the cortical dopaminergic and noradrenergic nerve terminal networks. Neuroscience 155:997-1010. S Sahu A, Tyeryar KR, Vongtau HO, Sibley DR, Undieh AS. (2009). D5 dopamine receptors are required for dopaminergic activation of phospholipase C. Mol Pharmacol. 75(3):447-53. Santana N, Mengod G, Artigas F.(2009) Quantitative analysis of the expression of dopamine D1 and D2 receptors in pyramidal and GABAergic neurons of the rat prefrontal cortex. Cereb Cortex. 2009 Apr;19(4):849-60. Sara SJ. (2009) The locus coeruleus and noradrenergic modulation of cognition. Nat Rev Neurosci. 2009 Mar;10(3):211-23 Schlesinger S, Dubensky TW. Alphavirus vectors for gene expression and vaccines. Curr Opin Biotechnol. 1999 Oct;10(5):434-9. Schroeter S, Apparsundaram S, Wiley RG, Miner LH, Sesack SR, Blakely RD (May 2000). "Immunolocalization of the cocaine- and antidepressant-sensitive l-norepinephrine transporter". J. Comp. Neurol. 420 (2): 211–32. Seguela P, Watkins KC, Geffard M, Descarries L. Noradrenaline axon terminals in adult rat neocortex: an immunocytochemical analysis in serial thin sections.Neuroscience. 1990;35(2):249-64. Shih JC, Chen K, Ridd MJ. (1999) Monoamine oxidase: from genes to behavior. Annu Rev Neurosci. 22:197- 217. Sibley DR (1999) New insights into dopaminergic receptor function using antisense and genetically altered animals. Annu Rev Pharmacol Toxicol 39:313–341. Sineshchekov, O. A., Jung, K. H. & Spudich, J. L. 2002. Two rhodopsins mediate phototaxis to low- and high- intensity light in Chlamydomonas reinhardtii. Proc. Natl Acad. Sci. U.S.A. 99, 8689–8694.

Smiley JF, Levey AI, Ciliax BJ, Goldman-Rakic PS.(1994) D1 dopamine receptor immunoreactivity in human and monkey cerebral cortex: predominant and extrasynaptic localization in dendritic spines. Proc Natl Acad Sci U S A. 91(12):5720-4. So CH, Verma V, Alijaniaram M, Cheng R, Rashid AJ, O'Dowd BF, George SR.(2009) Calcium signaling by dopamine D5 receptor and D5-D2 receptor hetero-oligomers occurs by a mechanism distinct from that for dopamine D1-D2 receptor hetero-oligomers. Mol Pharmacol. 75(4):843-54. Sohal VS1, Zhang F, Yizhar O, Deisseroth K.(2009) Parvalbumin neurons and gamma rhythms enhance cortical circuit performance. Nature. 2009 Jun 4;459(7247):698-702. Stepanenko OV1, Stepanenko OV, Kuznetsova IM, Verkhusha VV, Turoverov KK. Beta-barrel scaffold of fluorescent proteins: folding, stability and role in chromophore formation. Int Rev Cell Mol Biol. 2013;302:221-78. 2. Summers RJ, Papaioannou M, Harris S, Evans BA. Expression of beta 3-adrenoceptor mRNA in rat brain. Br J Pharmacol. 1995 Nov; 116(6):2547-8. Suzuki, T., Yamasaki, K., Fujita, S., Oda, K., Iseki, M., Yoshida, K., Watanabe, M., Daiyasu, H., Toh, H., Asamizu, E., Tabata, S., Miura, K., Fukuzawa, H., Nakamura, S. & Takahashi, T. 2003. Archaeal-type rhodopsins in Chlamydomonas: model structure and intracellular localization. Biochem. Biophys. Res. Commun. 301, 711–717.

T Takamine J (1901). "Adrenalin the active principle of the suprarenal glands and its mode of preparation". 'The American Journal of Pharmacy 73: 523–535. Tavares A, Handy DE, Bogdanova NN, Rosene DL, Gavras H.(1996) Localization of alpha 2A- and alpha 2B- adrenergic receptor subtypes in brain. Hypertension. 27(3 Pt 1):449-55. Taylor SR, Badurek S, Dileone RJ, Nashmi R, Minichiello L, Picciotto MR.(2014) GABAergic and Glutamatergic Efferents of the Mouse Ventral Tegmental Area. J Comp Neurol. 2014 Apr 9. Taymans JM1, Vandenberghe LH, Haute CV, Thiry I, Deroose CM, Mortelmans L, Wilson JM, Debyser Z, Baekelandt V. (2007) Comparative analysis of adeno-associated viral vector serotypes 1, 2, 5, 7, and 8 in mouse brain. Hum Gene Ther. ;18(3):195-206. Threlfell S, Lalic T, Platt NJ, Jennings KA, Deisseroth K, Cragg SJ. Striatal dopamine release is triggered by synchronized activity in cholinergic interneurons. Neuron. 2012 Jul 12;75(1):58-64. Torres GE, Gainetdinov RR, Caron MG (January 2003). "Plasma membrane monoamine transporters: structure, regulation and function". Nat. Rev. Neurosci. 4 (1): 13–25.

Trendelenburg AU, Cox SL, Schelb V, Klebroff W, Khairallah L, Starke K. Modulation of (3)H-noradrenaline release by presynaptic opioid, cannabinoid and bradykinin receptors and beta-adrenoceptors in mouse tissues. Br J Pharmacol. 2000 May;130(2):321-30.

U Usiello A, Baik JH, Rouge´-Pont F, Picetti R, Dierich A, LeMeur M, Piazza PV, and Borrelli E (2000) Distinct functions of the two isoforms of dopamine D2 receptors. Nature 408:199–203. V Vialou V, Balasse L, Callebert J, Launay JM, Giros B, Gautron S. (2008) Altered aminergic neurotransmission in the brain of organic cation transporter 3-deficient mice. J Neurochem. Aug;106(3):1471-82. Venkatesan C, Song XZ, Go CG, Kurose H, Aoki C.(1996) Cellular and subcellular distribution of alpha 2A- adrenergic receptors in the visual cortex of neonatal and adult rats. J Comp Neurol. 29;365(1):79-95.

W Waite PME and Tracey DJ (1995) Trigeminalsensory system. In the rat nervous system, G. axinos, ed (London Academic press) 705-724 Wang GS, Chang NC, Wu SC, Chang AC (2002). Regulated expression of alpha2B adrenoceptor during development. Dev Dyn. ;225(2):142-52. Weiner DM, Levey AI, Sunahara RK, Niznik HB, O'Dowd BF, Seeman P, Brann MR.(1991) D1 and D2 dopamine receptor mRNA in rat brain. Proc Natl Acad Sci U S A. Mar 1;88(5):1859-63. Williams SR, Stuart GJ. Mechanisms and consequences of action potential burst firing in rat neocortical pyramidal neurons. J Physiol. 1999 Dec 1;521 Pt 2:467-82. Wimalasena K. Vesicular monoamine transporters: structure-function, pharmacology, and medicinal chemistry.Med Res Rev. 2011 Jul;31(4):483-519. Wolf ME and Roth RH (1990) Autoreceptor regulation of dopamine synthesis. Ann NY Acad Sci 604:323–343. Wong AC1, Shetreat ME, Clarke JO, Rayport S (1999) D1- and D2-like dopamine receptors are co- localized on the presynaptic varicosities of striatal and nucleus accumbens neurons in vitro. 1999 Mar;89(1):221-33.

Wu Z, Yang H, Colosi P.Effect of genome size on AAV vector packaging. Mol Ther. 2010 Jan;18(1):80-6. doi: 10.1038/mt.2009.255

X Xiao RP, Avdonin P, Zhou YY, Cheng H, Akhter SA, Eschenhagen T, Lefkowitz RJ, Koch WJ, Lakatta EG. (1999) Coupling of beta2-adrenoceptor to Gi proteins and its physiological relevance in murine cardiac myocytes. Circ Res. Jan 8-22;84(1):43-52.

Y-Z Zhang J, Ma Y, Taylor SS, Tsien RY. Genetically encoded reporters of protein kinase A activity reveal impact of substrate tethering. Proc Natl Acad Sci U S A. 2001 Dec 18;98(26):14997-5002. Zhang J1, Campbell RE, Ting AY, Tsien RY. (2002)Creating new fluorescent probes for cell biology. Nat Rev Mol Cell Biol. Dec;3(12):906-18. Zhang J, Hupfeld CJ, Taylor SS, Olefsky JM, Tsien RY. Insulin disrupts beta-adrenergic signalling to protein kinase A in adipocytes. Nature. 2005 Sep 22;437(7058):569-73. Zhang ZW, Burke MW, Calakos N, Beaulieu JM, Vaucher E. (2010). Confocal Analysis of Cholinergic and Dopaminergic Inputs onto Pyramidal Cells in the Prefrontal Cortex of Rodents. Front Neuroanat 4:21. Zeman A. Consciousness. Brain. 2001 Jul;124(Pt 7):1263-89. Zianni D, Tzanela M, Klimopoulos S, Thalassinos NC. 2004 Symptomatic pheochromocytoma with normal urinary catecholamine metabolites. Hormones 2004, 3(2):132-137