STRIATAL DYSREGULATIONS AND ASSOCIATED PATHOLOGIES

Get Complete Project Material File(s) Now! »

History of glutamate identification as a neurotransmitter

It is well established since the fifties that glutamate is highly abundant and plays an important role in the CNS (Krebs 1935, Krebs et al. 1949, Schwerin et al. 1950). The first evidence that glutamate was involved in neurotransmission came in the early 1950s. First, glutamate was shown to be involved in the occurrence of seizures in patients suffering from petit mal epilepsia (Goodman et al. 1946, Wager 1946, Pond and Pond 1951). Later on, Hayashi revealed the ability of glutamate to trigger convulsions in humans, monkeys and dogs (Hayashi 1952, 1954). Interestingly this study also reported a toxic effect of high dose of glutamate. About the same time, glutamate was identified as the precursor for GABA (Roberts and Frankel 1950). Hayashi demonstrated that GABA antagonizes glutamate‑ induced seizures in dogs (Hayashi 1959). In 1954, Cole & Oikemus showed that intraperitoneal injection (i.p.) of glutamate in anaesthetized mouse has a stimulant effect (Cole and Oikemus 1954). In the late fifties, it was found that glutamate has the ability to depolarize and excite individual neurons in the spinal cord (Curtis et al. 1959, 1960). This established that glutamate is a potent major excitatory neurotransmitter in the CNS. Glutamate was then found to be enriched in synaptosomes and transported into synaptic vesicles (Kuhar and Snyder 1970, Disbrow et al. 1982).

Metabotropic glutamate receptors

Metabotropic glutamate receptors are coupled to trimeric G‑ proteins (G‑ protein‑ coupled receptor, GPCR) and second messenger systems. They are seven-transmembrane-domain proteins, with a large extracellular N-terminal domain (NTD) that binds their ligand, and an intracellular C-terminal domain (CTD) that is linked to different G‑ proteins.
The mGluRs recruit and activate G‑ proteins and downstream signaling cascades, resulting in short‑ term effects (post‑ translational modifications of pre‑ existing proteins), and long‑ term effects (recruitment of transcription factors and gene activation); overall, their kinetics of activation is slower than iGluRs (Figure A.7). mGluRs are involved in synaptic plasticity through long‑ term modifications of receptors and synapses, neurotransmitter release and neuronal excitability tuning (Swanson et al. 2005).
NMDARs and AMPARs exhibit a fast response to glutamate whereas mGluR kinetics is slower (time scale of hundreds of milliseconds to seconds) due to intracellular signaling cascades recruitment. The mGluRs were identified in the mid‑ eighties, when it was demonstrated that L‑ glutamate could stimulate phosphatidylinositol hydrolysis and intracellular Ca2+ mobilization, effect not reversed by the then‑ known iGluRs antagonists. Since then, eight mGluRs have been discovered (mGluR1‑ 8), and are divided in three groups, based on their structure, pharmacology and signal transduction. These receptors are located on both glutamatergic and non‑ glutamatergic neurons. They can be pre‑ or post‑ synaptic and also perisynaptic. Postsynaptic mGluRs modulate ion channel activity and therefore neuronal excitability. On the other hand, presynaptic mGluRs inhibit neurotransmitter release (Pinheiro and Mulle 2008). Studies indicate that mGluRs have an important role in anxiety‑ related disorders, and group I antagonists / group II agonists are proposed as potential anxiolytics (Swanson et al. 2005). From a functional point of view, these receptors can be stabilized by an inter-subunit disulfide bridge (Rondard et al. 2011).

Mechanistic of vesicular transport

All VNTs achieve an uptake against the nT electrochemical gradient. Therefore, this transport requires an energetic fuel. In the early nineties, it was established that nT transport is a secondary active transport, driven by an electrochemical proton gradient (Maycox et al. 1990). The ATPase responsible for generating the energy gradient was identified in 1987 (Nelson 1987). All synaptic vesicles carry a copy of the vacuolar H+-ATPase (Takamori et al. 2006). This primary transporter belongs to the vacuolar class of proton pumps. V‑ ATPases use the energy released by ATP hydrolysis to direct a flow of H+ into SVs. The influx of H+ generates an overall electrochemical proton gradient (∆µH+) through the SV membrane, which can be decomposed into:
 a chemical component (∆pH) because the intra-vesicular compartment is acidified.
 an electrical component (∆ψ) because the accumulation of positive charges inside the SV generates a membrane potential. (Ozkan and Ueda 1998).
VNTs then use this proton gradient to exchange lumenal protons for cytoplasmic transmitters. However, all VNTs do not display the same dependence on the two components of the electrochemical proton gradient (Figure A.16). Indeed, the transport of monoamines and acetylcholine primarily depends on the chemical component (∆pH), whereas the transport of glutamate depends predominantly on the electrical component (∆ψ). Accumulation of the inhibitory transmitters GABA and glycine relies on both ∆pH and ∆ψ. Regarding transport stoichiometry, the species that move across SV membrane depend on the imported neurotransmitter (mainly its charge).

VGLUTs cellular and subcellular distributions

The mRNA coding for VGLUT1 and VGLUT2 and the two proteins have complementary patterns of expression (Aihara et al. 2000, Fujiyama et al. 2001, Herzog et al. 2001, Boulland et al. 2004). VGLUT1 and VGLUT2 define two distinct classes of excitatory synapses in the CNS and can be used as bona fide markers for glutamatergic neurons (Fremeau et al. 2001, Kaneko and Fujiyama 2002). In a nutshell, VGLUT1 is expressed in cortical regions (i.e. cerebral and cerebellar cortices, hippocampus), whereas VGLUT2 is found in subcortical areas of the brain (i.e. diencephalon & brainstem). This pattern is highly conserved between rodents and human (Vigneault et al. 2015). For a more detailed expression pattern see Table A.3.
A slight note of caution needs to be put forward: in the brain, there is a coexpression of VGLUT1 and ‑ 2 to some extent. Mixed VGLUT1/VGLUT2 mRNA‑ expressing neurons were identified in the lateral olfactory tract and some thalamic nuclei, but proteins do not colocalize (Herzog et al. 2001). VGLUT1 and ‑ 2 can also be transiently co‑ expressed during development, as explained later (1.5: VGLUTs during development).

VGLUTs during development

Glutamatergic pathways undergo major adjustments during development. A highly differentiated spatial and age‑ dependent expression of the three VGLUTs is observed (Figure A.22). VGLUT1 concentration is very low at birth. Its expression increases steadily throughout postnatal development. It reaches a peak after postnatal day 14 (P14). Interestingly, VGLUT1 progressively replaces VGLUT2 in several regions such as the cortex or the hippocampus.
VGLUT2 represents the major VGLUT isoform during embryonic and early postnatal phases. Its expression occurs early in the embryonic life and reaches its maximum level at P7. VGLUT2 is first distributed homogeneously all over the cerebral cortex, but in the adult brain the labeling becomes more concentrated in cortical layers IV and VI. This has been named the “VGLUT2 – VGLUT1 switch” (Wojcik et al. 2004).
VGLUT3 expression increases progressively during postnatal development. It is increasingly present in the rostral brain (striatum & hippocampus) throughout post‑ natal development. VGLUT3 is strongly and transiently expressed in the caudal brain (cerebellum & superior olive complex) during early post‑ natal life (Schafer et al. 2002, Boulland et al. 2004, Gras et al. 2005). Then it disappears from these areas after P20 (Gras et al. 2005).
VGLUTs are essential regulators of vital functions in mammals. Deletion of VGLUT1 or VGLUT2 is lethal. VGLUT2‑ lacking mice die immediately after birth as a result of respiratory failure (Wallen-Mackenzie et al. 2010). VGLUT1‑ lacking mice die toward the end of the postnatal third week, precisely when the “VGLUT2 – VGLUT1 switch” occurs most likely due to diminished functioning of their cortex (Wojcik et al. 2004).
VGLUT3 shows a transient expression in several cell types. It is found in progenitor‑ like cells in the paraventricular zone at early developmental stages (Boulland et al. 2004). Moreover, cell cultures enriched with progenitor cells seem to have a high level of coexpression between VGLUT3 and Nestin (marker for progenitor cell). VGLUT3 is also transiently expressed in Purkinje cells in the cerebellum (Boulland et al. 2004, Gras et al. 2005).

Serotonin‑ glutamate dual release

The first clue for a dual transmitter release in 5‑ HT neurons was that extracellular stimulation of raphe nuclei leads to non‑ serotonergic excitatory post-synaptic currents (EPSCs) in the striatum, followed by longer latency serotonergic potentials (Park et al. 1982). Later, another study stated that cultured serotonergic neurons have the ability to elicit fast AMPAR‑ mediated post‑ synaptic currents (Johnson 1994). Then, it was demonstrated that activation of raphe nuclei leads to i) fast AMPAR‑ and ionotropic 5‑ HT3R‑ mediated ESPCs in hippocampal GABAergic interneurons and ii) a slow‑ rising putatively metabotropic 5‑ HT1AR‑ mediated inhibitory post‑ synaptic currents (IPSCs) in CA1 hippocampal neurons, or indirect inhibition via GABAergic interneurons (Varga et al. 2009). Recently, in vivo studies using optogenetic activation of serotonergic neurons from the DRN showed that AMPAR‑ mediated EPSCs are elicited in the VTA and AcbSh, comforting the idea that there is indeed a serotonin‑ glutamate dual release (Liu et al. 2014).

GABA‑ glutamate dual release

Several cases involve GABA / Glu co-release. The first description of a VGLUT3‑ dependent GABA / Glu dual release was reported in the developing auditory system at inhibitory synapse that are also VGLUT3‑ positive (Gillespie et al. 2005). This transient expression (~ until P7) during the auditory map peak sharpening has been hypothesized to help eliminate or strengthen connections between medial nucleus of the trapezoid body and the lateral superior olive.
There is some ultrastructural evidence for VGLUT3 and VIAAT colocalization on the same SVs membranes in the cortex and hippocampus, but not in the striatum (Stensrud et al. 2013). This is a clue for an in vivo storage of GABA and Glu within the same SVs, and for a subsequent co‑ release (Figure A.27). More recently, our team revealed that CCK-positive basket cells in the hippocampus signal with GABA and use glutamate as a secondary transmitter to retro-inhibit GABA signaling onto pyramidal cells (Fasano et al. 2017). In this situation, VGLUT3-dependent glutamate is acting upon mGluR.

Table of contents :

LIST OF TABLES
ABBREVIATION LIST
INTRODUCTION
PART I. GLUTAMATE: A FUNDAMENTAL NEUROTRANSMITTER 
1. Generalities about neurotransmission
2. History of glutamate identification as a neurotransmitter
3. Glutamate receptors and related signaling
3.1 Ionotropic glutamate receptors
3.2 Metabotropic glutamate receptors
4. Glutamate transporters
4.1 Plasma membrane glutamate transporters
4.2 Vesicular neurotransmitter transporters
PART II. VESICULAR GLUTAMATE TRANSPORTER TYPE 3 (VGLUT3) 
1. Vesicular glutamate transporters
1.1 A history of VGLUTs
1.2 VGLUTs cellular and subcellular distributions
1.3 VGLUTs structure
1.4 Mechanism of glutamate uptake by VGLUTs
1.5 VGLUTs during development
1.6 VGLUTs pharmacology
2. Vesicular glutamate transporter type 3 (VGLUT3)
2.1 Anatomical distribution
2.1.1 Regional distribution
2.1.2 Ultrastructural distribution
2.1.3 Neuronal subtypes distribution
2.2 Functional roles: overview
2.2.1 Dual release of transmitters
2.2.2 Vesicular synergy
2.2.3 Related phenotypes
PART III. THE STRIATUM 
1. Anatomy of the striatum
1.1 Matrix versus striosomes
1.2 Functional compartmentalization
1.3 Striatal cytoarchitecture
1.3.1 Medium spiny neurons
1.3.2 Striatal interneurons
2. Striatal connectivity
2.1 Striatal afferents
2.1.1 Glutamatergic innervation
2.1.2 Midbrain DA neuromodulation
2.1.3 Serotonergic innervation
2.1.4 Cholinergic innervation
2.1 Striatal activity modulation
2.1.1 DAergic modulation
2.1.2 Cholinergic modulation
2.1.3 Acetylcholine / dopamine activity during behavior
2.2 Striatal efferent to basal ganglia
2.2.1 Basal ganglia
2.2.2 The striatonigral (direct) pathway
2.2.3 The striatopallidal (indirect) pathway
2.2.4 The striosomal pathway
2.2.1 Current view on striatal function
PART IV. STRIATAL DYSREGULATIONS AND ASSOCIATED PATHOLOGIES 
1. The reward system
1.1 Reward circuitry
1.2 Mode of action of drugs
1.3 Drug addiction
1.4 Molecular basis of drug addiction
1.5 Role of VGLUT3 in addiction
1.6 Amphetamine
2. Dorsal striatum defects
2.1 Parkinson’s disease
2.2 Obsessive-compulsive disorders
3. Stereotypies
3.1 Definition and general considerations
3.2 Animal models to elicit stereotypies
3.2.1 L‑DOPA‑induced dyskinesia
3.2.1 Drug‑induced stereotypies
3.3 Functional and biochemical correlates of stereotypies
3.4 Methods to assess drug‑induced stereotypies
3.5 Role of VGLUT3 in LID
MATERIALS AND METHODS 
1. Animal models
1.1 VGLUT3—/—
1.2 VGLUT3LoxP/LoxP
1.3 VGLUT3 conditional knock‑out: genetic approach
1.3.1 SERT‑Cre conditional knock‑out
1.3.2 ChAT‑IRES‑Cre conditional knock‑out
1.4 VGLUT3 conditional knock‑out: viral approach
1.4.1 Surgical procedure
1.4.2 Cre‑expressing virus
2. Behavioral experiments
2.1 Spontaneous locomotor activity
2.2 Anxiety tests
2.2.1 Open field
2.2.2 O maze
2.3 Behavioral sensitizations
2.3.1 Behavioral sensitization to amphetamine
2.3.2 Locomotor sensitization to cocaine
2.4 Stereotypy scoring
2.4.1 Categorial scoring
2.4.2 Rating‑scale‑based scoring of general stereotypies
2.4.3 Rating‑scale‑based scoring of orofacial stereotypies
3. Drug treatments
4. Anatomical studies by immunohistochemistry
4.1 Immunofluorescence
4.1.1 Conditional knock‑outs validation: VGLUT3, VAChT and 5‑HT
4.1.2 ΔFosB immunofluorescence
4.2 Immunoautoradiography
5. Statistics
6. Experiments color code
RESULTS
PART I. VGLUT3 FULL KNOCK‑OUT AND AMPHETAMINE 
1. Basal characterization
1.1 Number of VGLUT3–/– used
1.2 VGLUT3–/– mice weight
1.3 Spontaneous locomotion
2. VGLUT3–/– and amphetamine 1 mg/kg
2.1 Acute injection of amphetamine 1 mg/kg
2.1 Repeated injections of amphetamine 1 mg/kg
3. VGLUT3–/– and amphetamine 3 mg/kg
3.1 Acute injection of amphetamine 3 mg/kg
3.1 Repeated injections of amphetamine 3 mg/kg
4. VGLUT3–/– and amphetamine 5 mg/kg
4.1 Effect on locomotion
4.1.1 Acute injection of amphetamine 5 mg/kg
4.1.2 Repeated injections of amphetamine 5 mg/kg
4.2 Effect on stereotypies
4.2.1 General stereotypies
4.2.2 Orofacial stereotypies
5. Discussion
PART II. CONTRIBUTION OF VGLUT3‑POSITIVE SYSTEMS IN THE RESPONSE TO AMPHETAMINE 
1. The VGLUT3‑positive serotonergic drive of the striatum .
1.1 Anatomical characterization of cKO-VGLUT35-HT
1.2 Behavioral characterization of cKO-VGLUT35-HT
1.2.1 Weight
1.2.2 Basal locomotion
1.2.3 Anxiety tests
1.2.4 Spontaneous locomotion before sensitization
1.3 cKO-VGLUT35-HT and amphetamine 5 mg/kg
1.3.1 Effect on locomotion
1.3.2 Effect on stereotypies
2. The VGLUT3‑positive cholinergic drive of the striatum
2.1 Anatomical characterization of cKO-VGLUT3ACh
2.2 Behavioral characterization of cKO-VGLUT3ACh
2.2.1 Weight
2.2.2 Basal locomotion
2.2.3 Anxiety tests
2.2.4 Spontaneous locomotion before sensitization
2.3 cKO-VGLUT3ACh and amphetamine 5 mg/kg
2.3.1 Effect on locomotion
2.3.2 Effect on stereotypies
2.4 cKO-VGLUT3ACh and cocaine 10 mg/kg
3. Discussion
PART III. STUDY OF THE ROLE OF VGLUT3 IN THE NUCLEUS ACCUMBENS IN THE RESPONSE TO COCAINE 
1. Development of viral injections
1.1 Choice of virus
1.2 Experimental groups
2. Anatomical validation
3. Behavioral characterization
3.1 Mice weight
3.2 Basal locomotion
3.3 Anxiety levels
4. Locomotor sensitization to cocaine
5. Discussion
GENERAL CONCLUSION
APPENDICES II
1. Temporal course of AMPH injections
1.1 VGLUT3–/– and AMPH 1 mg/kg ii
1.2 VGLUT3–/– and AMPH 3 mg/kg iii
1.3 VGLUT3–/– and AMPH 5 mg/kg iv
1.4 cKO-VGLUT35-HT and AMPH 5 mg/kg v
1.5 cKO-VGLUT3ACh and AMPH 5 mg/kg vi
2. Publications
REFERENCES 

GET THE COMPLETE PROJECT