The role of the dopamine D1-D2 receptor heteromer in brain reward function: Relevance

to drug addiction and depression.

by

Maurice Yen Fu Shen

A thesis submitted in comformity with the requirements for the degree of Doctor of Philosophy

Graduate Department of Pharmacology and Toxicology

University of Toronto

© Copyright by Maurice Yen Fu Shen (2015)

ii

The role of the dopamine D1-D2 receptor heteromer in brain reward function: Relevance

to drug addiction and depression.

Maurice Yen Fu Shen, Doctor of Philosophy,

Department of Pharmacology and Toxicology, University of Toronto, 2015

Abstract

We have identified a novel dopamine-mediated signaling complex, the D1-D2 receptor

heteromer, which couples to Gq protein to elicit phospholipase C-mediated intracellular Ca2+

release. Activation of the D1-D2 heteromer was further shown to modulate the expression of

proteins implicated in reward in the nucleus accumbens, such as calcium calmodulin kinase II

(CaMKII) and brain derived neurotrophic factor (BDNF), a finding which may be of relevance

in neuropsychiatric disorders that exhibit altered reward perception such as addiction and

depression. Therefore, the purpose of current study was to investigate a role for the D1-D2

heteromer in reward-related behaviours using animal models of cocaine addiction and

depression. The results show that D1-D2 heteromer stimulation by agonist SKF 83959

attenuated addiction-related behaviours including cocaine conditioned place preference (CPP),

the maintenance and reinstatement of cocaine self-administration (SA), and the expression of

locomotor sensitization to cocaine, whereas D1-D2 heteromer inactivation by a selective

disrupting peptide TAT-D1 consistently augmented cocaine-induced behaviours, whereas a

control scrambled peptide had no such effects. Moreover, D1-D2 heteromer stimulation alone

was shown to induce conditioned place aversion, whereas its selective inactivation induced

CPP. In the tests for depression, D1-D2 heteromer stimulation by SKF 83959 induced a pro-

depressive and anxiogenic state in the forced swim test, novelty-induced hypophagia, and the

elevated plus maze, effects that were abolished by pretreatment with TAT-D1 peptide. Lastly,

iii

it was shown that the D1-D2 heteromer exerts dual modulation on neuronal activity in certain

regions of brain, increasing activity in the NAc, and exerting tonic inhibition in the prelimbic

and infralimbic cortex, orbitofrontal cortex, lateral habenula, and the thalamus, which likely

reflects the dual excitatory/inhibitory capabilities of the GABA/Glu-co-expressing D1R/D2R

MSNs. Collectively, these findings indicate that the D1-D2 heteromer may be a single

molecular entity that could bidirectionally modulate brain reward function depending on its

state of activation. Such an unprecedented function of a receptor complex makes the D1-D2

heteromer an attractive and novel therapeutic target for cocaine addiction and major depression,

two reward-related neuropsychiatric disorders that are currently without effective treatments.

iv

Acknowledgments

First and foremost, I would like to extend my most sincere gratitude to my supervisor

Dr. Susan George for giving me the opportunity to pursue a doctoral degree in her laboratory.

Her expertise and guidance has been the key to my success during my studies, and I believe the

lessons that I have learned from her will be immensely beneficial to my future career. It has

been an honor and privilege to work under her supervision.

I would like to thank the funding agencies that have supported this research: the

Canadian Institute of Health Research and the National Institure of Health.

Thank you to members of my committee, Dr. Paul Fletcher and Dr. Jose Nobrega, for

providing me with different and insightful perspectives to this work. Particularly, I would like

to thank Dr. Fletcher for lending me the equipment and his expertise for the drug self-

administration study.

I would also like to thank members of the George lab, Dr. Melissa Perreault, Dr.

Ahmed Hasbi, Dr. Gabriela Novak, Dr. Vaneeta Verma, Dr. Brian O’Dowd, Tuan Nguyen,

Theresa Fan, Ryan To, Marco Cheung, and David Yeung for creating a friendly and

inspirational work environment. I loved every moment in the lab with them around. Special

thanks to Dr. Perreault for guiding me with regards to all aspects of my study, ranging from

experimental design, grant applications, to manuscript submissions. My doctoral study

wouldn’t have been as smooth and successful as it has been without her invaluable support.

Also special thanks to Theresa for her technical expertise on animal surgeries, and to Zhao-xia

Li and Tony Ji for their technical support during my time at the Center for Addiction and

Mental Health for the self-administration work.

v

Last but not least, I would like to express my utmost gratitude to my family and friends

for always being there for me, and never doubting my ability to finish a prestigious degree.

vi

TABLE OF CONTENTS

i. Abstract 4

ii. Acknowledgments 6

iii. List of Publications 8

iv. List of Figures 9

v. List of Abbreviations 11

1. General Introduction 16

1.1 G protein-Coupled Receptors (GPCRs) 16

1.1.1 GPCR Structure 16

1.1.2 GPCR Activation 18

1.1.3 G-proteins 19

1.1.4 GPCR Oligomerization 20

1.2 Dopamine Receptors 22

1.2.1 The Genetic and Structural Properties of Dopamine Receptors 22

1.2.2 Dopamine Receptor Expressions 24

1.2.3 Dopamine Receptor Signalling 29

1.2.3.1 The Gαs-cAMP-PKA-DARPP-32 Pathway 29

1.2.3.2 The MEK-ERK1/2 Pathway 32

1.2.3.3 Calcium Signalling via Gαq-coupling 33

1.2.4 Dopamine Receptor Oligomerization 35

1.3 The Dopamine D1-D2 Receptor Heteromer: A Novel Dopamine Receptor Coupled

to Gq-PLC-Ca2+

36

1.3.1 The D1R and D2R Form Heteromeric Complex in vitro 38

1.3.2 Expression of the D1-D2 Heteromer in the Brain 40

1.3.3 A Unique Functional Role for the D1-D2 Heteromer 42

1.3.4 Regulation of the D1-D2 Heteromer Signalling 44

1.3.5 Generation of a Novel Antagonist for the D1-D2 Heteromer: The TAT-D1

Peptide 45

1.3.6 Physiological Effects of the D1-D2 Heteromer 46

1.4 The Role of Dopamine in the Brain Reward System 49

1.4.1 The Neurocircuitry of the Brain Reward System 50

1.4.1.1 The Nucleus Accumbens (NAc) 52

1.4.1.2 The Ventral Tegmental Area (VTA) 54

1.4.2 The Role of Mesolimbic Dopamine in Depression and Drug Addiction 58

1.5 The D1-D2 Heteromer as a Modulator of Brain Reward Function: Relevance to

Addiction and Depression 60

1.5.1 CaMKII 61

1.5.2 BDNF 62

1.5.3 GAD67 63

vii

1.6 Research Rational and Objectives 65

1.6.1 Hypotheses 66

2. The role of the dopamine D1-D2 receptor heteromer in cocaine reward. 67

3. Rapid anti-depressant and anxiolytic activities following dopamine D1-D2

receptor heteromer inactivation. 110

4. The dopamine D1-D2 receptor heteromer exerts tonic inhibitory effect on the

expression of amphetamine induced locomotor sensitization. 139

5. Regulation of c-fos expression by the dopamine D1-D2 receptor heteromer. 164

6. General Discussion 184

6.1 The D1-D2 heteromer is a negative modulator of psychostimulant and natural

reward 185

6.2 D1-D2 heteromer stimulation induced a depressive-like state 188

6.3 The D1-D2 heteromer modulation of reward: A potential role for BDNF and

GAD67 191

6.4 The D1-D2 heteromer is a novel molecular substrate for aversions 195

6.5 Significance and Conclusion 197

7. Future Directions 200

8. References 203

viii

List of Publications

1. Shen M.Y.F., Perreault M.L., Fan T., George S.R. The dopamine D1-D2 receptor

heteromer exerts a tonic inhibitory effect on the expression of amphetamine-induced

locomotor sensitization. Pharmacol Biochem Behav. 2015 Jan;128:33-40.

2. Perreault M.L.*, Shen M.Y.F.*, Fan T., George S.R. Regulation of c-fos expression by the

dopamine D1-D2 receptor heteromer. Neuroscience. 2015 Jan 29;285:194-203. *These

authors contributed equally to this work

3. Shen M.Y.F.*, Perreault M.L.*, Bambico F.R., Jones-Tabah J., Cheung M., Fan T.,

Nobrega J.N., George S.R. Rapid antidepressant and anxiolytic actions following

dopamine D1-D2 heteromer inactivation. Eur. J. Neuropsychopharm. Manuscript under

revision. *These authors contributed equally to this work

4. Hasbi A., Perreault M.L., Shen M.Y.F., Zhang L., To R., Fan T., Nguyen T., Ji X.,

O'Dowd B.F., George S.R. A peptide targeting an interaction interface disrupts the

dopamine D1-D2 receptor heteromer to block signaling and function in vitro and in vivo:

effective selective antagonism. FASEB J. 2014 Nov;28(11):4806-20.

ix

List of Figures

Figure 1 The basal ganglia circuitry in rats. 26

Figure 2 A potential third neuronal pathway within the basal ganglia. 28

Figure 3 The regulation and function of DARPP-32 phophorylations. 31

Figure 4 The activation of the MEK-ERK pathway is dependent on the

stimulation of both the D1R and the NMDAR. 34

Figure 5 The D1R and the D2R interact via electrostatic interactions to form the

D1-D2 heteromer. 39

Figure 6 The neurocircuitry of the brain reward system. 51

Figure 7 The proposed role of the D1-D2 heteromer in the regulation of brain

reward function. 199

Figure I-1 The effects of D1-D2 heteromer stimulation and inactivation on basal

conditioned place preference. 80

Figure I-2 The effects of D1-D2 heteromer stimulation and inactivation on cocaine-

induced CPP. 81

Figure I-3 The effect of Cdk5 inhibitor roscovitine on SKF 83959-induced CPA. 82

Figure I-4 The effects of D1-D2 heteromer stimulation and inactivation on

locomotion induced by acute and repeated cocaine treatment. 84

Figure I-5 The effects of D1-D2 heteromer stimulation and inactivation on the

expression of cocaine-induced locomotor sensitization. 86

Figure I-6 The effect of D1-D2 heteromer stimulation on cocaine self-

administration under the FR5 schedule. 88

Figure I-7 The lever-pressing behaviour during extinction training. 90

Figure I-8 The effects of D1-D2 heteromer stimulation and inactivation on cocaine-

induced reinstatement. 91

Figure I-9 The effect of D1-D2 heteromer stimulation on cue-induced

reinstatement. 93

Figure II-1 The effects of D1-D2 heteromer stimulation or inactivation on the

latency to and total immobility time in the forced swim test in rats. 122

Figure II-2 D1-D2 heteromer stimulation or inactivation had bidirectional effects on

anxiety-like behaviours in the elevated plus maze. 124

x

Figure II-3 D1-D2 heteromer stimulation or inactivation had bidirectional effects on

anxiety-like behaviours in the novelty-induced hypophagia test. 126

Figure II-4 The effects of TAT-D1 peptide treatment on anhedonia-like reactivity

induced by chronic unpredictable stress. 128

Figure II-5 The effects of TAT-D1 peptide treatment on anxiety-like reactivity

induced by chronic unpredictable stress. 130

Figure III-1 The effects acute D1-D2 heteromer stimulation and inactivation on basal

and amphetamine-induced locomotor activity. 148

Figure III-2 The effects of subchronic D1-D2 heteromer stimulation and inactivation

on basal and amphetamine-induced locomotor activity. 150

Figure III-3 The time course of basal or amphetamine-induced locomotor responses

following D1-D2 heteromer stimulation or inactivation during the course

of repeated treatment.

153

Figure III-4 The effects of D1-D2 receptor heteromer stimulation and inactivation on

the expression of amphetamine-induced locomotor sensitization. 155

Figure IV-1 Regulation of c-fos expression in rat nucleus accumbens by the

dopamine D1-D2 receptor heteromer. 171

Figure IV-2 Grooming induced by SKF 83959 in rats is mediated by the dopamine

D1-D2 heteromer. 173

Figure IV-3 Regulation of c-fos expression in rat caudate putamen by the dopamine

D1-D2 receptor heteromer. 174

Figure IV-4 Effects of dopamine D1-D2 heteromer disruption on c-fos

immunoreactivity in rat cortex. 176

Figure IV-5 Effect of dopamine D1-D2 heteromer disruption on c-fos expression in

rat lateral habenula and thalamus. 178

xi

List of Abbreviations

7-TM Seven Transmembrane

AAV Adeno-associated Virus

AC Adenylyl Cyclase

AH α-helical

AMPA α-amino-3-hydroxy-5-methyl-4-isoxazolepropionic acid

AP-1 Activator Protein-1

BAC Bacterial Artificial Chromosomes

BDNF Brain-derived Neurotrophic Factor

BLA Basolateral Amygdala

BRET Bioluminescence Resonance Energy Transfer

CaMKII Calcium Calmodulin Kinase II

cAMP Cyclic Adenosine Mono-phosphate

Cdk5 Cyclin-dependent Kinase 5

CeA Central Amygdaloid Nucleus

CK1 Casein Kinase 1

CK2 Casein Kinase 2

Co-IP Co-immunoprecipitation

CP Caudate Putamen

CPA Conditioned Place Aversion

CPP Conditioned Place Preference

CREB cAMP Response Element Binding Protein

DAG Diacylglycerol

DARPP-32 Dopamine and cAMP-regulated Protein Phosphatase (32kDa)

xii

DAT Dopamine Reuptake Transporter

DREADDs Designer Receptors Exclusively Activated by a Designer Drug

DYN Dynorphin

ECL Extracellular Loop

ENK Enkephalin

EPN Entopeduncular Nucleus

ER Endoplasmic Reticulum

ERK Extracellular Signal-related Kinase

FR Fixed-Ratio

FRET Fluorescent Resonance Energy Transfer

GABA γ-Aminobutyric Acid

GAD Glutamate Decarboxylase

GAP GTPase Accelerating Protein

GDP Guanine Nucleotide Di-phosphate

GP Globus Pallidus

GPCR G Protein-coupled Receptor

GRK G Protein-coupled Receptor Kinase

GSK-3 Glycogen Synthase Kinase 3

GTP Guanine Nucleotide Tri-phosphate

HEK Human Embryonic Kidney

HSV Herpes Simplex Virus

ICL Intracellular Loop

ICSS Intracranial Self-stimulation

IFC Infralimbic Cortex

xiii

IGF Insulin-like Growth Factor

IP3 Inositol 1,4,5-trisphosphate

LDTg Laterodorsal Tegmentum

LHb Lateral Habenula

LTD Long-term Depression

LTP Long-term Potentiation

MAPK Mitogen-activated Protein Kinases

MEK Mitogen-activated Protein Kinase Kinase

MSN Medium Spiny Neurons

mTOR Rictor-mammalian Target of Rapamycin

NAc Nucleus Accumbens

NLS Nuclear Translocation Sequence

NMDA N-methyl-D-aspartate

OFC Orbitofrontal Cortex

PD Parkinson’s Disease

PDK1 Phosphoinositide-dependent Kinase-1

PFC Prefrontal Cortex

PH Plekstrin Homology

PI3K Phosphoinositide 3 Kinases

PIP2 Phosphatidylinositol 4,5-bisphosphate

PKA Protein Kinase A

PKC protein Kinase C

PLCβ Phospholipase C β

PP-1 Protein Phosphatase-1

xiv

PP2A Protein Phosphatase 2A

PP2B Protein Phosphatase 2B

PPTg Pedunculopontine Tegmentum

PR Progressive Ratio

PTX Pertussis Toxin

RGS Regulators of G-protein Signalling

RH Regulator of G-protein Signalling Homology

RH RGS Homology

RKIP Raf-kinase Inhibitor Protein

RMTg Rostromedial Tegmental Nucleus

SA Self-administration

SKF 81297 (±)-6-Chloro-2,3,4,5-tetrahydro-1-phenyl-1H-3-benzazepine hydrobromide

SKF 83822 6-Chloro-2,3,4,5-tetrahydro-1-(3-methylphenyl)-3-(2-propenyl)-1H-3-benzazepine-7,8-diol

hydrobromide

SKF 83959 6-Chloro-2,3,4,5-tetrahydro-3-methyl-1-(3-methylphenyl)-1H-3-benzazepine-7,8-diol

SNc Substantia Nigra pars compacta

SNr Substantia Nigra pars reticulata

SRF Serum Response Factor

STEP Striatal-enriched Tyrosine Phosphatase

STN Subthalamic Nucleus

TH Tyrosine Hydroxylase

TM Transmembrane

VGLUT-1 Vesicular Glutamate Transporter-1

VGLUT-2 Vesicular Glutamate Transporter-2

xv

VP Ventral Pallidum

vSub Ventral Subiculum

VTA Ventral Tegmental Area

1

1. General Introduction

1.1 G-protein Coupled Receptors (GPCRs)

G-protein coupled receptors (GPCR), also known as seven transmembrane receptors,

are crucially involved in the transduction of various extracellular stimuli including light,

hormones, and neurotransmittors into specific cellular responses (Müller et al., 2008;

Rosenbaum et al., 2009). Abberant GPCR signalling has been implicated in various disease

states, ranging from cardiovascular diseases to neuropsychiatric disorders (Moreno et al., 2013;

Zalewska et al., 2014), and more than 30% of commercially available drugs target GPCRs

(Williams and Hill, 2009), making these receptors ideal candidates for the discovery of novel

therapeutic targets. As a result GPCRs have been widely studied targets for drug discovery

over several decades.

1.1.1 GPCR Structure

GPCRs are composed of seven transmembrane (7-TM) α-helices connected by three

extracellular loops (ECL) and three intracellular loops (ICL). The ECLs, which includes the N-

terminus, form part of the binding pocket for certain receptor ligands but small molecules like

the catecholamines bind deep within the transmembrane bundle. Particularly, the ECL2 adopts

a β-hairpin conformation that keeps the binding pocket open, and thus is crucial for ligand

binding to GPCRs (White et al., 2012). On the other hand, the ICLs are responsible for G-

protein coupling and the binding of proteins such as the arrestins, both of which are responsible

for initiating discrete cell signaling cascades (Katritch et al., 2012). The ICLs also includes the

C-terminus and the cytoplasmic helix H8, which is important for receptor stabilization and its

2

proper functioning (Maeda et al., 2010). The ECLs feature high structural diversity amongst

GPCRs in order to accommodate various types of ligands, while the G-protein-binding ICLs

are generally more conserved between GPCRs (Katritch et al., 2012). The α-helices of the 7-

TM core are frequently tilted due to the presence of helix-breakers amino acids such as proline,

serine, threonine and glycine residues (von Heijne, 1991; Senes et al., 2000; Deupi et al., 2004).

The degree of tilt and the rotation angles of these α-helices differ depending on the types of

GPCRs and their activation state (Yohannan et al., 2004; Meruelo et al., 2011; Latek et al.,

2012). Nevertheless, the 7-TM core contains well conserved motifs characteristic of GPCRs

including the “E/DRY” on TM3 (Rasmussen et al., 2011), “CwxP” on TM6 (Vaidehi et al.,

2014), and “nPxxy” on TM7 (Rosenbaum et al., 2009). The Arg residue of the “E/DRY” motif

interacts with a glutamic acid residue on TM6 to form an “ionic-lock” that was proposed to

stabilize the GPCR in the inactivated state (Rasmussen et al., 2011). In addition, a disulphide

bridge forms between ECL2 and TM3 to maintain structural stability (Hanson et al., 2012). The

crystal structures of several GPCR have been reported to date, including those for the A2A

adenosine receptor, β1 and β2 adrenergic receptors, CXCR4 chemokine receptor, D3 dopamine

receptor, H1 histamine receptor, M2 and M3 muscarinic receptors, µ, δ, and κ opioid receptors,

5HT1B serotonin receptor, rhodopsin, CRF1 corticotrophin receptor, and the glucagon receptor

(Reviewed in: Vaidehi et al. 2014). Of note, the recent release of high-resolution crystal

structures of the µ-opioid and the β1-adrenergic receptors demonstrated that GPCRs exist as

oligomers, or receptor complexes, via hydrophobic interactions between the TM domains

(Manglik et al., 2012; Huang et al., 2013). Specifically, an interface between TM1, TM2, and

helix H8 was found in both µ-opioid and β1-adrenergic receptor oligomers (Manglik et al.,

3

2012; Huang et al., 2013), suggesting oligomerization of other GPCRs (Section 1.9) may occur

via a similar mechanism.

1.1.2 GPCR Activation

Under basal conditions, GPCRs exist in both active and inactive conformational states

that are in equilibrium, with a higher number of receptors in the inactive state when no ligand

is present (Niesen et al., 2011; Vaidehi and Bhattacharya, 2011). Stimulation of a GPCR by

ligand binding to the ECLs triggers a slight change in receptor conformation through the 7-TM

domain, which ultimately leads to structural alterations in the ICLs that permit receptor binding

to its cognate G-protein (Rasmussen et al., 2011). Upon binding to the GPCR, the G-protein

undergoes nucleotide exchange from GDP to GTP in its Gα subunit, thereby dissociating the

trimeric G-protein into Gα and Gβγ subunits, each capable of mediating its own signal

transduction pathways (De Lean et al., 1980; Leff, 1995; Freire, 1998). The C-terminus of a

stimulated GPCR is subjected to phosphorylation by G-protein receptor kinases (GRKs), which

then triggers the binding of arrestins to the GPCR that are responsible for the blockade of G-

protein activation and rapid homologous desensitization of the GPCR and its associated

signalling (Kohout and Lefkowitz, 2003; Reiter and Lefkowitz, 2006). The binding of arrestins

also results in clathrin-mediated endocytosis of the GPCR, which then may be subjected to

lysosomal degradation or resensitization and re-expression on the plasma membrane

(Palczewski et al., 1989; Freedman and Lefkowitz, 1996). Interestingly, arrestins were also

recently shown to induce their own non-canonical signalling cascades (Reviewed in: Maudsley

et al. 2013). In addition, the termination of the signalling can also be achieved via the intrinsic

4

GTPase activity of the Gα subunit that hyrdolyzes the GTP to GDP in the Gα subunit, resulting

in the re-association of Gα and Gβγ, and thus inactivation of the G-protein (Gilman, 1987).

During the process of GPCR activation, the intracellular portion of the TM6 tilts

outward towards TM5, and the Arg within the “E/DRY” motif on TM3 extends towards the

GPCR core to interact with the tyrosine residue of the “nPxxY” motif on TM7, thereby

breaking the “ionic lock” between TM3 and TM6 and turning the receptor into its active

conformation (Rasmussen et al., 2011).

1.1.3 G-proteins

The Gα subunit activates various types of signal transduction pathways, and can be

divided into four families, Gαs, Gαi/o, Gαq/11, and Gα12/13, based on the α subunit sequence

and functional characteristics. The Gαs and Gαi/o function to respectively, activate or inhibit

various types of adenylyl cyclases (AC) to modulate the intracellular concentration of cyclic

adenosine monophosphate (cAMP) (Wettschureck and Offermanns, 2005). The Gαq/11 protein

couples the GPCR to the β-isoform of phospholipase C (PLCβ) to increase intracellular

calcium levels (Wettschureck and Offermanns, 2005). In contrast, the functions of Gα12/13 are

still unclear due to the lack of specific inhibitors, although studies have shown that Gα12/13 are

activated by GPCRs that couple to Gαq/11 (Strathmann and Simon, 1990; Dhanasekaran and

Dermott, 1996). The Gα subunit has two structural domains: a nucleotide-binding, Ras-like

domain and an α-helical (AH) domain that partially occludes the bound nucleotide, and the two

domains are interconnected by two flexible linkers (Oldham et al., 2006). In addition, the Gα

5

subunit binds the ligand-bound GPCR via its C-terminus and associates with Gβγ through N-

terminal myristoylation (Resh, 1999).

1.1.4 GPCR Oligomerization

Classically, a GPCR was thought to exist and function as a monomeric entity that was

activated by a single ligand. However, studies using biophysical and molecular techniques

including co-immunoprecipitation (Co-IP), bioluminescence resonance energy transfer (BRET)

and fluorescent resonance energy transfer (FRET) have demonstrated that GPCRs can form

and function as oligomers both in vitro and in vivo (Reviewed in: Rivero-Müller et al. 2013).

Such findings have thus added another layer of complexity to our understanding of the

physiology, signalling and pharmacology of GPCRs.

Evidence has shown that GPCRs of the glutamate family exist as obligatory dimers, as

is the case for the GABAB receptor and the metabotropic glutamate receptors (mGluRs).

Specifically, the GABAB1 protomer is able to bind to its ligand but unable to transduce a signal,

whereas the GABAB2 protomer could couple to the G-protein but is incapable of ligand binding

(Kaupmann et al., 1998). Correct trafficking and signalling of the GABAB receptor only occurs

following the heterodimerization of the two protomers via coiled-coil interactions between

their C-termini (Comps-Agrar et al., 2011). Similarly, mGluRs were found to express in the

membrane as strict dimers via a disulphide bond linkage in the ECLs, and their proper

signalling requires intersubunit rearrangement between the two protomers (Tsuji et al., 2000;

El Moustaine et al., 2012). Furthermore, studies also suggest that GPCRs of the rhodopsin

family form homoligomers in the endoplasmic reticulum (ER) during protein synthesis and

6

maturation prior to cell surface delivery (Salahpour et al., 2004; Kong et al., 2006), although

the interactions between the protomers occur mainly via hydrophobic interactions between the

TM domains rather than the ECLs and ICLs like the glutamate family GPCRs. For instance,

homodimers of both the α1b-adrenergic receptor or dopamine D2 receptor are formed via

symmetrical interfaces between the TM1 and TM4 of the two protomers (Lee et al., 2003;

Lopez-Gimenez et al., 2007; Guo et al., 2008). In addition, different types of GPCRs can also

form heteroligomers primarily via electrostatic interactions between the ICLs (Navarro et al.,

2010). As an example, the heterodimer between the dopamine D2 and cannabinoid CB1

receptors is formed via electrostatic interactions between phosphorylated Thr321

–Ser322

in ICL3

of the CB1 receptor and an Arg-rich epitope in ICL3 of the D2 receptor.

At the physiological level, GPCR oligomerization could lead to the generation of novel

signalling cascades, the amplification or inhibition of an existing signalling, alteration in ligand

binding affinity, receptor delivery to cell surface, and heterologous receptor desensitization or

internalization (Rivero-Müller et al., 2013). To date, numerous GPCR homo- and

heteroligomers have been identified (Reviewed in: George et al. 2002; Moreno et al. 2013;

Gonzalez et al. 2014), many of which are implicated in various disease states ranging from

neuropsychiatric to cardiovascular disorders due to the ability of oligomerization to modulate

and diversify the signalling generated by GPCRs.

However, it should be noted that basic functioning of GPCRs may not require

oligomerization, since monomeric rhodopsin and β2-adrenergic receptor are capable of G-

protein coupling, GRK-mediated phosphorylation, and arrestin-binding when they are

reconstituted into high-density lipoprotein particles (Whorton et al., 2007, 2008). However, this

is an entirely artificial system for study and may not be physiologically relevant. In addition,

7

the crystal structure of an active state complex consisted of a ligand-bound, monomeric β2-

adrenergic receptor-T4-lysozyme fusion protein, a nucleotide-free Gs heterotrimer, and a

nanobody has also recently been solved (Rasmussen et al., 2011). Nevertheless, whether the

experimental setup of these studies reflects physiological conditions is still a matter of debate.

1.2 Dopamine Receptors

Dopamine (3-hydroxytyramine) is a catecholaminergic neurotransmitter that is critically

involved in the regulation of important physiological functions including voluntary movement,

feeding, affect, reward, sleep, attention, learning and memory (Missale et al., 1998; Sibley,

1999; Iversen and Iversen, 2007). Thus, disturbances in dopaminergic system in the brain have

been linked to a wide array of disorders including major depression and drug addiction, as well

as Parkinson’s disease (PD), Huntington’s disease, and schizophrenia, amongst others

(Ehringer and Hornykiewicz, 1960; Seeman et al., 1976; Jakel and Maragos, 2000; Koob and

Volkow, 2010a). Therefore, dopamine has been the target of active research since its discovery

more than 50 years ago.

1.2.1 The Genetic and Structural Properties of Dopamine Receptors

Dopamine exerts its physiological functions by binding to dopamine receptors, which

are GPCRs of the rhodopsin subfamily. Therefore, dopamine receptors possess the seven-

transmembrane-spanning structural phenotype that is iconic of GPCRs and signal through G-

proteins. There are five different subtypes of dopamine receptors, which are classified as either

8

D1-like (D1R and D5R) or D2-like (D2R, D3R and D4R) on the basis of their structural,

biochemical and pharmacological properties (Spano et al., 1978; Kebabian and Calne, 1979;

Andersen et al., 1990; Tiberi et al., 1991). Members of the same dopamine receptor subfamily

share high sequence homology in their TM domain, with D1R being 80% homologous to the

D5R, and the D3R and the D4R being 75% and 53% homologous to the D2R, respectively

(Gingrich and Caron, 1993; Missale et al., 1998). The N-terminal domains of all dopamine

receptors are similar in length, while the C-termini of the D1-like receptors are seven times

longer than that for the D2-like receptors (Gingrich and Caron, 1993; Missale et al., 1998). To

date, the crystal structure has only been reported for the D3R (Chien et al., 2010).

The D1-like receptors couple to the Gαs/olf proteins to stimulate cAMP production by

AC, whereas the D2-like receptors activate the Gαi/o proteins to negatively modulate cAMP

production (Spano et al., 1978; Kebabian and Calne, 1979). In addition, evidence also suggests

that the D5R can couple to cAMP (Sunahara et al., 1991) and the Gαq protein to increase

intracellular calcium concentration via the activation of the IP3 receptor (Sahu et al., 2009; So

et al., 2009). The two dopamine receptor subfamilies are also distinct in their gene structures.

The coding regions of the D1R and D5R do not contain introns, whereas several introns have

been identified in the genes that encode the D2-like receptors, with six introns in the D2R, five

in the D3R and three in the D4R genes (Gingrich and Caron, 1993). As a result, genes of the

D2-like receptors can be alternatively spliced to produce receptor variants. For instance, there

are two D2R isoforms termed D2R-long (D2L) and D2R-short (D2S) that differ by 29 amino

acids in their third ICL (Usiello et al., 2000). Similarly, splice variants for the D3R and D4R

have also been described (Giros et al. 1991; Van Tol et al. 1992). For the D4R, its isoforms are

characterized by a 48-base-pair repeat sequence in the third ICL, the the number of repeats

9

varying from 2 up to 11 repeats (Van Tol et al. 1992). In general, each splice variant of the

D2-like receptors possesses distinct anatomical, physiological and pharmacological properties

(De Mei et al., 2009).

1.2.2 Dopamine Receptor Expressions

Dopaminergic neurons are much more prominent in the brain compared to peripheral

areas, and are subdivided into four major pathways in brain: the nigrostriatal, mesolimbic,

mesocortical, and tuberoinfundibular pathways (Anden et al., 1964; Dahlstroem and Fuxe,

1964). The D1-like receptors are expressed exclusively on the postsynaptic membrane, whereas

the D2R and D3R are found both pre- and postsynaptically (Sokoloff et al., 1992). The D2S

isoform is selectively expressed on the presynaptic terminals and is thought to be involved in

autoreceptor functions that provide negative feedback modulation on synaptic dopamine

release (Usiello et al., 2000; De Mei et al., 2009). In contrast, the D2L isoform is mostly found

in postsynaptic sites where it inhibits intracellular cAMP concentration via the coupling to

Gαi/o protein (Usiello et al., 2000; De Mei et al., 2009). In addition to their differences in the

expression sites, the D2S and D2L also differ in their affinity to dopamine receptor agonists,

with the D2S being activated at a lower dose of dopamine agonists than necessary to activate

postsynaptic receptors (De Mei et al., 2009).

In the brain, the highest density of the D1R is found in the caudate putamen (CP),

nucleus accumbens (NAc), ventral tegmental area (VTA), substantia nigra, olfactory bulb,

amygdala, and the frontal cortex. The D1R is also identified at a lower level in the

hippocampus, cerebellum, thalamic and hypothalamic areas (Levey et al., 1993; Missale et al.,

10

1998). The expression of the D2R largely overlaps with that of the D1R, which includes the CP,

NAc, VTA, substantia nigra, olfactory bulb, hypothalamus, frontal cortex, septum, amygdala

and hippocampus (Levey et al., 1993; Missale et al., 1998; Seeman, 2006). For the basal

ganglia, studies using transgenic mice with bacterial artificial chromosomes (BAC) that

express fluorescent proteins under the control of specific promoters have demonstrated distinct

segregation of D1R- and D2R-containing medium spiny neurons (MSNs) predominantly in the

CP, and to a lesser degree in the NAc (Shuen et al., 2008; Valjent et al., 2009). Specifically,

MSNs that project to the substantia nigra pars reticulata (SNr) and the entopeduncular nucleus

(EPN) comprise the direct striatonigral pathway that selectively expresses the D1R. In contrast,

the MSNs that project to the globus pallidus (GP) selectively expresses the D2R and are known

as the indirect striatopallidal pathway, since they indirectly reach the SNr/EPN via synaptic

relays in the GP and the subthalamic nucleus (STN) (Figure 1).

Nevertheless, recent studies have also identified a subpopulation of MSNs in the NAc

(17-30%) and CP (~6%) of BAC transgenic mice that co-express both the D1R and D2R

(Bertran-Gonzalez et al., 2008; Matamales et al., 2009; Gangarossa et al., 2013b). Using highly

validated selective antibodies for the D1R and the D2R, we have demonstrated the co-

expression of both receptors in a number of nuclei within the rat basal ganglia, including the

NAc core (~25% of D1R-expressing neurons), NAc shell (~35%), CP (~7%), ventral pallidum

(VP) (~30%), GP (~60%), and STN (~50%) (Perreault et al., 2010), suggesting that, in ventral

striatum, the direct and indirect pathways are not completely segregated, and there likely exists

a third neuronal pathway within the basal ganglia circuitry that co-express both the D1R and

the D2R (Perreault et al., 2011) (Figure 2). Additionally, we showed that D1R and D2R co-

expression were found in both the cell bodies and exclusively on the presynaptic, but not the

11

Figure 1: The basal ganglia circuitry in rats. A schematic depicting the connections between

the nuclei within the basal ganglia circuitry. The D1R and D2R-expressing medium spiny

neurons (MSNs) were thought to be largely segregated and respectively comprise the direct

striatonigral and indirect striatopallidal pathways.

SNc: substantia nigra pars compacta, SNr: substantia nigra pars reticulata, VTA: ventral

tegmental area, GP: globus pallidus, VP: ventral pallidum, STN: subthalamic nucleus, EPN:

entopeduncular nucleus

Cortex

Striatum

D2R D1R

SNc/VTA

GP/VP

STN

EPN/SNr

Thalamus

Functional

Output

Ind

irect d

irect

Inhibitory

Excitatory

12

postsynaptic, terminals, as indicated by the D1R and D2R co-expression with the presynaptic

marker synaptophysin, but not with the postsynaptic marker PSD95 (Perreault et al., 2010).

Since presynaptic dopamine receptors on MSN terminals have been shown to modulate both

the excitatory and inhibitory postsynaptic currents (Ford, 2014; Zhang et al., 2014), we

postulated that the neurons within the basal ganglia that co-express both the D1R and the D2R,

which terminate in regions within both the direct and indirect pathways, may function to

maintain homeostatic balance between the two neuronal pathways, perhaps via the regulation

of GABA and glutamate expression (Perreault et al., 2011, 2012a). Indeed, we have later

revealed that striatal D1R/D2R co-expressing neurons also exhibited a unique GABA/glutmate

co-expressing phenotype (Perreault et al., 2012a). In addition to the striatum, co-expression of

the D1R and D2R has also been observed in 20-25% of the pyramidal neurons in the PFC of

BAC transgenic mice, with almost all of the D1R-expressing neurons in the PFC also co-

expressing the D2R (Zhang et al., 2010), although the potential function of the D1R/D2R co-

expressing neurons in the PFC has yet to be elucidated.

The D3R has a relatively more limited distribution pattern in the brain, with the highest

expression being in the NAc shell, the olfactory tubercle, and the islands of Calleja (Sokoloff et

al., 1992). Amongst the dopamine receptor subtypes, the D4R has the lowest expression level

in the brain, which includes the frontal cortex, amygdala, hippocampus, hypothalamus, GP,

SNr and the thalamus (Missale et al., 1998). Lastly, the expression of the D5R has been

reported at high level in the pyramidal neurons of the PFC (Oda et al., 2010), and at lower level,

in the premotor cortex, the cingulated cortex, the entorhinal cortex, SN, hypothalamus,

hippocampus, the dentate gyrus, and in the interneurons of the striatum (Missale et al., 1998).

13

D1R, D2R Pathways

Striatum

D1R/D2R

SNc VTA

SNr

EPN VP GP

Putative D1R/D2R

Pathways

Figure 2: A potential third neuronal pathway within the basal ganglia. A schematic

depicting the known neuronal connections between the nuclei within the basal ganglia that

express exclusively the D1R or the D2R, and the putative D1R/D2R co-expressing pathways.

SNc: substantia nigra pars compacta, SNr: substantia nigra pars reticulata, VTA: ventral

tegmental area, GP: globus pallidus, VP: ventral pallidum, STN: subthalamic nucleus, EPN:

entopeduncular nucleus

14

1.2.3 Dopamine Receptor Signalling

The signalling initiation and modulation of dopamine receptors follows that of a typical

GPCR as described earlier in Section 1. In brief, the binding of dopamine induces

conformational changes in the dopamine receptor structure to promote G-protein coupling,

after which the guanine nucleotide exchange occurs on the Gα subunit, leading to the

dissociation of the trimeric G-protein into Gα and Gβγ, each capable of eliciting its own

signalling cascade (Gingrich and Caron, 1993). Activated dopamine receptors are then

subjected to phosphorylation by the GRKs, which in turn results in β-arrestin recruitment to the

receptor that induces receptor desensitization and internalization (Gainetdinov et al., 2003).

The RGS proteins also accelerate GTP-hydrolysis on the Gα subunit to terminate dopamine

receptor signalling (Dohlman and Thorner, 1997). This section will describe in detail both G-

protein-dependent and G-protein-independent signalling cascades that are mediated by

dopamine receptors in the brain.

1.2.3.1 The Gα-cAMP-PKA-DARPP-32 Pathway

As described in Section 2.1, D1-like receptors couple to the Gαs/olf protein to stimulate,

whereas D2-like receptors are coupled to the Gαi/o protein to inhibit, cAMP production and

PKA activity (Spano et al., 1978; Kebabian and Calne, 1979). Amongst the substrates for PKA,

the dopamine and cAMP-regulated protein phosphatase (DARPP-32) is one of the most

extensively studied effectors of dopaminergic signalling. DARPP-32 is highly expressed on the

MSNs in the striatum where it functions as an integrator of various cell signalling responses

(Svenningsson et al., 2004). PKA phosphorylates DARPP-32 at the Thr34 site, which promotes

15

its action as a potent inhibitor of protein phosphatase-1 (PP-1). It has been shown that DARPP-

32 is differentially phosphorylated at this site in the two subpopulations of MSNs in the

striatum following dopamine receptor stimulation, with increased Thr34 phosphorylation being

found selectively in the D1R-expressing neurons and reduced Thr34 phosphorylation in the

D2R-expressing neurons as a result of the opposing modulation of PKA activity by D1R and

D2R (Bateup et al., 2008). In addition, DARPP-32 Thr34 phosphorylation can be

dephosphorylated by the calmodulin-dependent protein phosphatase 2B (PP2B) following D2R

stimulation (Nishi et al., 1997). Another site of phosphorylation on DARPP-32 is Thr75 by

cyclin-dependent kinase 5 (Cdk5) which results in its action as an inhibitor of PKA, and

thereby releasing the inhibitory control of DARPP-32 on PP-1 activity (Bibb et al., 1999).

Furthermore, DARPP-32 is also phosphorylated at Ser137 by casein kinase 1 (CK1) and at

Ser97/102 by casein kinase 2 (CK2), which respectively reduces and enhances DARPP-32

Thr34 phosphorylation by PKA (Girault et al., 1989; Desdouits et al., 1995) (Figure 3). The

knockout of of CK2 selectively in D1R-expressing MSNs has recently been shown to alter the

biochemical and behavioural effects mediated by the D1R (Rebholz et al., 2013). Lastly, the

dephosphorylation of Ser97/102 by protein phosphatase 2A (PP2A) accumulates DARPP-32 in

the nucleus where it prevents dephosphorylaion of histone H3 by PP-1 to increase gene

expression in response to D1R stimulation (Stipanovich et al., 2008).

In the MSNs, the phosphorylation state of several PKA substrates such as AMPA and

NMDA glutamate receptors are governed by the equilibrium between DARPP-32-mediated

PP-1 and PKA activity (Snyder et al., 2000; Nishi et al., 2002, 2005). The inhibition of PP-1 by

DARPP-32 Thr34 phosphorylation shifts the equilibrium towards the phosphorylated state and

enhances the efficacy of D1R-PKA-mediated signalling. However, although DARPP-32

16

Figure 3: The regulation and function of DARPP-32 phophorylations. DARPP-32 can be

phosphorylated and dephosphorylated at four distinct sites respectitvely by different protein

kinases and protein phosphatases. Each phosphorylation carries out its own unique function.

PKA: protein kinase A, Cdk5: cyclin-dependent kinase 5, CK1&2: casein kinase 1&2,

PP2A/B/C: protein phosphatase 2A/B/C.

NH2 COOH Thr75 Ser97 Ser130 Thr34

P P P P

PKA

PP2B

PP2A

Cdk5

PP2A

CK2 CK1

PP2A PP2A

PP2C

Inhibits

PP-1

Inhibits

PKA

Enhances

PKA

Enhances

PP2B

17

functions as a crucial modulator or dopaminergic signalling, it is not the only effector that

regulates dopamine-mediated behaviours. For example, studies using DARPP-32 knockout

mice have demonstrated that behavioural responses to dopaminergic drugs such as cocaine and

apomorphine are only partially disrupted in the mutant mice, indicating that signalling

mechanisms other than the D1R-PKA-DARPP-32 cascade are involved in dopamine-related

behaviours (Fienberg et al., 1998; Nally et al., 2004). Nevertheless, the role of DARPP-32 in

various neuropsychiatric diseases involving a dysfunctional dopamine system has been

extensively studied, particularly in the drug addiction research (Reviewed in: Nairn et al. 2004;

Svenningsson et al. 2005; Yger and Girault 2011).

1.2.4.2 The MEK-ERK1/2 Pathway

Mitogen-activated protein (MAP) kinases have been shown to be involved in

dopamine-mediated signalling and behaviours (Berhow et al., 1996). Studies have

demonstrated that dopamine receptors can regulate ERK1 and ERK2 as administration of

indirect dopamine receptor agonists such as cocaine and amphetamine, as well as D2R

antagonist haloperidol have all been shown to enhance ERK1/2 phosphorylation and activation

(Valjent et al., 2000; Pozzi et al., 2003; Beaulieu et al., 2006). Specifically, genetic and

pharmacological manipulations determined that D1R is essential for striatal ERK1/2 activation,

whereas D2-like receptors, particularly the D3R, inhibit ERK1/2-mediated signalling in the

striatum (Zhang et al., 2004; Valjent et al., 2005). In addition, D1R-mediated ERK1/2

activation is also dependent on the NMDA glutamate receptor, the stimulation of which results

in a calcium influx that activates ERK kinase MEK via Ras-GRF1, a guanine nucleotide

18

exchange factor (Farnsworth et al., 1995; Valjent et al., 2005). In the absence of D1R

stimulation, ERK1/2 phosphorylation by MEK is counteracted by the activity of striatal-

enriched tyrosine phosphatase (STEP), which results in a zero-sum equilibrium where the

overall activity of ERK1/2 remain unchanged. Activation of the D1R-PKA-DARPP-32

pathway leads to the inhibition of PP-1 and the subsequent inactivation of STEP, which then

shifts the equilibrium between MEK and STEP activity in favour of the activation of ERK1/2

by MEK (Valjent et al., 2000, 2005) (Figure 4). This co-regulation of ERK1/2 activity by the

D1R and NMDA receptor suggest that ERK may be a signal integrator of dopaminergic and

glutamatergic neurotransmissions, particularly during the development of behavioural

responses to drugs of abuse. Indeed, inhibition of ERK1/2 activity by MEK inhibitor SL327

has been shown to antagonize the locomotor-stimulating effect of cocaine and amphetamine

(Beaulieu et al., 2006; Valjent et al., 2006b). Moreover, studies have suggested that ERK-

mediated signalling cascades are essential for the activation of transcription factors such as

cAMP response element binding protein (CREB), Zif268, and c-fos that are associated with

long-term changes in gene expression and the development of behavioural adaptations to drugs

of abuse (Brami-Cherrier et al., 2005; Valjent et al., 2006a).

1.2.4.3 Calcium Signalling via Gαq-coupling

In addition to their classic modulatory effect on cAMP/PKA signalling, some evidence

also suggests that specific subtypes of dopamine receptors can couple to the Gαq protein to

regulate PLC activity (Felder et al., 1989; Friedman et al., 1997; Lee et al., 2004; Sahu et al.,

2009).

19

Figure 4: The activation of the MEK-ERK pathway is dependent on the stimulation of

both the D1R and the NMDAR. D1R stimulation leads to the inhibition of PP-1 and STEP by

phosphorylating DARPP-32 at Thr34 to relieve the inhibition of NMDAR-mediated activation

of MEK-ERK pathway. PP-1: Protein Phosphatase 1, STEP: striatal-enriched tyrosine

phosphatase

AC

D1R NMDAR

Gs

D2R

Gi

cAMP

PKA

Ca2+

Ras

MEK

ERK

DARPP-32

Thr34

PP-1

STEP

CREB

Gene Expression c-Fos

Zif268

BDNF

Plasma Membrane

Nucleus

Cytoplasm

20

Activation of PLC leads to the production of inositol triphosphate (IP3) and diacylglycerol

(DAG), which in turn results in the activation of protein kinase C (PKC) by DAG and an

increased intracellular calcium release following IP3 receptor stimulation. It has thus been

delineated that the dopamine-induced PLC-Ca2+

signalling is mediated in part by D5R

stimulation, since the IP3 accumulation elicited by a D1-like receptor agonist SKF 83959 was

attenuated in striatal neurons derived from transgenic mice with D5R gene deletion (Sahu et al.,

2009). Similarly, it has been shown that the expression of D1R in human embryonic kidney

(HEK) cells did not induce intracellular calcium signalling following exposure to D1R-like

dopamine receptor agonist SKF 81297, whereas the stable expression of D5R in HEK cells

produced extensive calcium mobilization following SKF 81297 treatment (So et al., 2009). In

addition, the heterodimer between the D1R and D2R, known as the dopamine D1-D2 receptor

heteromer, can also regulate PLC-mediated calcium signalling both in vitro and in vivo (Lee et

al., 2004; Rashid et al., 2007; Hasbi et al., 2010), which will be discussed in detail in Section 3.

1.2.5 Dopamine Receptor Oligomerization

Similar to most GPCRs, it is now well accepted that dopamine receptors exist and

function as oligomers rather than monomers. For instance, both D1R and D2R require the

appropriate homomeric conformation for their proper signalling and cell surface trafficking

(Lee et al., 2003; Kong et al., 2006). In addition, dopamine receptors have been shown to form

heteromers between different dopamine receptor subtypes, as well as with other GPCRs and

ion channels (Reviewed in: Gomes et al. 2013). This thesis, however, will focus primarily on

one dopamine receptor heteromer, the dopamine D1-D2 receptor heteromer, a receptor

21

complex that has recently been identified, which may play a role in neuropsychiatric disorders

such as drug addiction and depression due to its unique signalling properties (Lee et al., 2004;

Rashid et al., 2007; Hasbi et al., 2009; Perreault et al., 2011).

1.3 The Dopamine D1-D2 Receptor Heteromer: A Novel Dopamine Receptor Complex

Coupled to Gq-PLC-Ca2+

In addition to the canonical cAMP/AC/PKA signalling mediated by the D1R and the

D2R, comprehensive behavioural studies conducted by Waddington et al using various

synthetic dopamine receptor agonists have revealed the existence of a “D1R-like dopamine

receptor” that was capable of AC-independent signalling (Deveney and Waddington, 1995;

Adachi et al., 1999; Tomiyama et al., 2002; O’Sullivan et al., 2005). Additional studies

performed by Friedman et al later revealed that this D1-like dopamine receptor signalled

through the PLCβ/IP3 cascade to release calcium from intracellular stores in the striatum and

the PFC in vivo (Undie et al., 1994; Wang et al., 1995; Jin et al., 2003; Zhen et al., 2004).

However, the exact nature of the D1R-like dopamine receptor that mediated such calcium

signaling in the brain remained elusive for many years. Although it had been demonstrated that

the D5R could couple to the Gαq protein to induce a PLC-mediated calcium signal in both the

striatum and the PFC (Sahu et al., 2009; Perreault et al., 2012b), its minimal expression in the

striatum necessitated the search for other possible mechanisms that could mediate the

production of a calcium signal by dopamine in this region.

Studies have shown that when the D1R alone was expressed in heterologous cell

systems such as Cos7, HEK293, CHO, BHK cells, intracellular calcium signalling in response

22

to dopamine treatment was not evident (Dearry et al., 1990; Pedersen et al., 1994; Lee et al.,

2004). This led to the idea that perhaps the calcium signal induced by dopamine may have

resulted from the co-activation of D1R with another GPCR also present in native striatal tissue

but not in cells. Amongst the possible GPCRs that could interact with the D1R, it was

postulated the D2R as a likely candidate as it was also highly expressed in the striatum

(Missale et al., 1998). Although the D1R and D2R were thought to be predominantly

segregated in striatum, many studies had reported that certain D1R- and D2R-mediated

biochemical, electrophysiological, and behavioural effects were synergistic upon the co-

activation of both receptors. For instance, coactivation of the D1R and D2R was shown to

potentiate immediate early gene expression, to synergistically inhibit NAc neuronal activity,

and to produce maximal locomotor behaviours, when compared to the stimulation of either

receptor alone (White et al., 1988; Hu et al., 1990; Keefe and Gerfen, 1995). There were three

possible explanations for these findings: 1) D1R and D2R on different neurons were interacting

through inter-neuronal circuitry communication, 2) D1R and D2R in the same neurons were

functionally interacting through their downstream signalling cascades, 3) D1R and D2R in

same neurons formed a receptor complex with unique signalling properties. However, the fact

that the D1R and D2R opposingly regulated the cAMP/PKA pathway (Spano et al., 1978;

Kebabian and Calne, 1979) suggested that these synergistic effects were not mediated simply

by the concurrent activation of individual Gαs-coupled D1R and Gαi/o-coupled D2R, but

rather may have been due to a novel dopamine receptor complex composed of D1R and D2R.

Indeed, our laboratory has thus far conducted thorough in vitro and in vivo analyses and

identified such a dopamine receptor complex, which was termed the dopamine D1-D2 receptor

heteromer.

23

1.3.1 The D1R and D2R form a Heteromeric Complex in vitro

Evidence for the existence of a heteromeric complex comprised of the D1R and the

D2R initially came from a study that showed the D1R and the D2R could be co-

immunoprecipitated from both primary-cultured striatal neurons and the rat striatum (Lee et al.,

2004). While such a finding suggested that the D1R and D2R were part of the same protein

complex, it did not indicate a direct physical interaction as the two receptors may have been

linked through scaffolding proteins. In later studies a direct physical interaction between the

two receptors was shown using a bioluminescence resonance energy transfer (BRET) technique

to assay D1R and D2R proximity in HEK cells stably expressing the D1R and D2R (Verma et

al., 2010; Hasbi et al., 2014). In these studies it was demonstrated that the D1R and D2R are in

physical contact with each other as indicated by the robust BRET signal elicited by the energy

transfer between D1R-Rluc and D2R-GFP. Furthermore, by tagging the D2R with a nuclear

translocation sequence (NLS) that induced D2R trafficking to the nucleus, we observed in

HEK cells the co-translocation of D1R with D2R to the nucleus, further demonstrating that the

D1R and D2R oligomerize into a heteromeric receptor complex (O’Dowd et al., 2005, 2012).

Additional studies using the NLS and BRET techniques from our laboratory subsequently

revealed that the D1R interacted with both the D2S and D2L isoforms, and that the D1-D2

heteromer was formed in the ER through the electrostatic interactions between two glutamic

acid residues in the C-terminal tail of D1R and two arginine residues in the ICL-3 of D2R

(O’Dowd et al., 2005, 2012; Hasbi et al., 2014) (Figure 5). Nevertheless, a separate study

reported that the D1R interacted with part of the additional 29 amino acids sequence that is

present in the D2L but not in the D2S isoform to form the D1-D2 heteromer (Pei et al., 2010),

however our laboratory and others (Chun et al., 2013) could not replicate this finding.

24

Fig

ure

5: T

he D

1R

an

d th

e D2

R in

tera

ct v

ia electro

static

intera

ction

s to fo

rm

the D

1-D

2 h

etero

mer. T

he electro

static

interactio

ns b

etween

the tw

o g

lutam

ic acid resid

ues (4

04-4

05

) in th

e C-term

inal o

f D1R

and th

e two arg

inin

e residues (2

74

-

275) in

the th

ird in

tracellular lo

op o

f D2R

interact fo

rm th

e D1

-D2 h

eterom

er.

25

1.3.2 Expression of the D1-D2 Heteromer in the Brain

Having confirmed the existence of the D1-D2 heteromer in vitro, we next wanted to

determine whether this receptor complex was present in vivo. Using highly specific

fluorophore-tagged antibodies against the D1R and the D2R (validated in D1-/- and D2-/- gene

deleted striatal tissue and in HEK cells expressing each of the 5 dopamine receptors

individually), the co-expression and co-localization of the two receptors was observed in adult

rat NAc core, NAc shell, CP, GP, VP, and the EPN to various degrees (Perreault et al., 2010).

Specifically, in both neonatal striatal neurons and in the dorsal and ventral striatum (CP and

NAc respectively), our antibodies were able to detect the co-localization of D1R and D2R with

each other, as well as with dynorphin (DYN) and enkephalin (ENK), well-known neuronal

markers for D1R- and D2R-expressing neurons respectively, thus validating the specificity of

our antibodies (Perreault et al., 2010). However, the co-localization of the D1R and D2R does

not necessarily mean the two receptors are physically interacting. Therefore, using the confocal

FRET microscopy in striatal slices in situ, which evaluates the proximity of the endogenous

D1R and D2R using fluorophore-tagged antibodies, we demonstrated a robust FRET efficiency

(~0.20) in the NAc core and shell, indicating that the two receptors were less than 100Ǻ apart,

with the majority of D1R/D2R co-expressing neurons (~90%) exhibiting D1-D2 heteromer

expression (Perreault et al., 2010). Since approximately 30% of D1R-expressing MSNs in the

NAc co-express the D2R, results from the FRET study indicated that roughly 27% of MSNs in

the NAc exhibited D1-D2 heteromer formation. Additionally, the expression of the D1-D2

heteromer in the NAc shell was subsequently found to increase with age, with the heteromer

expression being lower in juvenile rats compared to adults (Perreault et al., 2013, 2014). A

similarly high FRET efficiency (~0.22) was also observed between the D1R and the D2R in the

26

GP, suggesting a high degree of D1-D2 heteromer formation in this region as well, although

only a small number of neurons in the GP co-express both the D1R and the D2R (Perreault et

al., 2010, 2011). In contrast, while we were also able to detect a FRET signal between the D1R

and the D2R in the CP, the efficiency was much lower (~0.05), with only approximately 25%

of the D1R/D2R co-expressing neurons in the CP expressing the D1-D2 heteromer (Perreault et

al., 2010). Since only 4-6% of MSNs in the CP co-expressed the D1R and D2R, the FRET data

suggests merely 1-2% of total MSNs in the CP express the D1-D2 heteromer. Furthermore,

immunohistochemical analysis revealed localization of the D1-D2 heteromer selectively in cell

bodies and presynaptic terminals in the NAc and CP, indicating a possible role for the receptor

complex in neurotransmitter release from MSNs (Perreault et al., 2010).

The findings that demonstrated the abundant expression of the D1-D2 heteromer in the

NAc has thus been regarded as controversial, since it has traditionally been thought that the

expression of the D1R and D2R on striatal MSNs are largely segregated, with D1R residing

selectively on the direct striatonigral pathway and D2R localizing exclusively in the indirect

striatopallidal pathway (Gerfen and Surmeier, 2011). Nevertheless, several studies have also

identified the co-expression of D1R and D2R in striatal neuronal cultures and in native striatum,

thus supporting the existence of the D1-D2 heteromer in vivo (Bertran-Gonzalez et al., 2008;

Matamales et al., 2009; Perreault et al., 2010; Gangarossa et al., 2013b). For instance, using

BAC transgenic mice with eGFP-tagged to the promoter elements of D1R and D2R, a study by

Bertran-Gonzalez et al. (2008) showed that approximately 17% of NAc shell MSNs and 6% of

CP MSNs exhibit D1R/D2R co-localization. A separate study later validated the accuracy of

dopamine receptor expression and representation in the BAC transgenic mice by showing that

100% of MSNs in these mice expressed D1R, D2R or both, with the degree of co-localization

27

being similar to that was reported in the Bertran-Gonzalez study (Matamales et al., 2009).

Moreover, the subpopulation of MSNs with the highest co-expression of D1R and D2R was

found to be located specifically in the bundle-shaped area in the caudomedial NAc shell

(Gangarossa et al., 2013b). Combined with the FRET analyses, collectively these studies

strongly indicate that the D1R and D2R are co-expressed in a specific subset of MSNs in the

NAc and form the D1-D2 heteromer in these neurons.

1.3.3 A Unique Functional Role for the D1-D2 Heteromer

Using an in vitro cellular system, it was first demonstrated that the co-administration of

individual D1R and D2R selective agonists stimulated a calcium signal only in HEK293 cells

that stably co-expressed both the D1R and D2R, but not in cells that expressed the individual

D1R or D2R homoligomers alone (Lee et al., 2004). In addition, the calcium signal was

abolished by selective antagonists for the D1R or the D2R, as well as by inhibitors for Gαq,

PLC, and the IP3 receptor (Lee et al., 2004; Rashid et al., 2007; Hasbi et al., 2009), thus

indicating that the calcium signal was mediated by Gαq coupling, dependent on PLC activation,

resulted from an intracellular calcium source, and required the co-activation of both D1R and

D2R. In addition, the involvement of the Gαq protein in mediating this calcium signal was

further confirmed by the finding showing increased GTPγS incorporation into the Gαq protein

following D1R and D2R co-activation in HEK293 cells stably expressing both receptors

(Rashid et al., 2007).

A variety of dopamine receptor agonists were subsequently screened to determine their

effects at the D1-D2 heteromer. A synthetic benzazepine derivative known as SKF 83959 was

28

identified that could stimulate intracellular calcium mobilization in HEK293 cells and primary

cultured striatal neurons without affecting the D1R- or D2R-mediated AC activity (Rashid et

al., 2007; Hasbi et al., 2009; Verma et al., 2010). Additional analyses revealed that SKF 83959

increased the GTPγS incorporation into the Gαq protein in striatal tissue and generated a

calcium signal that was abolished by selective D1R or D2R antagonists, as well as by inhibitors

for Gαq, PLC, and the IP3 receptor in primary cultured striatal neurons (Rashid et al., 2007;

Hasbi et al., 2009). Additionally, SKF 83959-induced calcium signal was also absent in

primary cultured striatal neurons derived from transgenic mice gene-deleted for the D1R

presumably due to the lack of D1-D2 heteromers (Hasbi et al., 2009).

Further characterization of SKF 83959 demonstrated that within the D1-D2 heteromer

the compound acted as a full agonist at the D1R and as a partial agonist at the D2R (Rashid et

al., 2007). This was deduced from the findings showing SKF 83959 alone elicited a calcium

signal that was approximately 60-75% of that was generated by dopamine, but could be

enhanced upon the co-administration of D2R agonist quinpirole to the same extent as that was

induced by dopamine (Rashid et al., 2007; So et al., 2007). Moreover, SKF 83959 exhibited

high affinity binding to the D1R homoligomers and low affinity for the Gαi-coupled D2R

homoligomers, but did not activate either of these receptors (Andringa et al., 1999; Chun et al.,

2013). Nevertheless, the use of pertussis toxin (PTX) that renders the D2R to a state of low

affinity to agonist binding showed that SKF 83959 was able to bind to a PTX-insensitive D2R

with high affinity in rat striatum and in HEK293 cells stably expressing both the D1R and D2R,

but not in cells expressing either D1R or D2R alone (Rashid et al., 2007). This finding

suggested that SKF 83959 was able to bind to a specific D2R linked to the PTX-resistant Gαq

protein, such as the D2R within the D1-D2 heteromer, thus supporting the idea that SKF 83959

29

stimulated the D1-D2 heteromer by binding to both protomers within the receptor complex.

Collectively, these data suggested that SKF 83959 was an agonist for the D1-D2 heteromer,

and thus could be a useful tool to study the physiological role of the D1-D2 heteromer in vivo.

1.3.4 Regulation of the D1-D2 Heteromer Signalling

Similar to other GPCRs, the D1-D2 heteromer is subjected to signal desensitization and

receptor internalization following its activation. It was shown that D1-D2 heteromer-mediated

calcium signal was desensitized by prior exposure to dopamine, SKF 83959, the D1R-selective

agonist SKF 83822, or D2R-selective agonist quinpirole, albeit at different rates (So et al.,

2007; Verma et al., 2010), suggesting that agonist occupancy of either receptor within the D1-

D2 heteromer was sufficient to induce a conformation change that led to the desensitization of

the calcium signal even without heteromer stimulation. This desensitization occurred

independent of calcium storage depletion, exogenous calcium entry or receptor internalization,

and was specifically mediated by GRK2 via its RGS and catalytic domains (So et al., 2007;

Verma et al., 2010). Similarly, the D1-D2 heteromer was also found to internalize by the co-

activation of both D1R and D2R by dopamine or by the sole activation of either receptor within

the receptor complex (Verma et al., 2010). Since the D1-D2 heteromer was previously shown

to be formed in the ER (O’Dowd et al., 2005), the ability of agonist occupancy to either

constituent receptor to desensitize and internalize the D1-D2 heteromer indicates that the

integrity of the receptor complex was maintained throughout the typical GPCR life cycle from

the initial cell surface trafficking to the eventual signal termination.

30

1.3.5 Generation of a Novel Antagonist for the D1-D2 Heteromer: The TAT-D1 Peptide

Although SKF 83959 was shown to be an agonist for the D1-D2 heteromer, the

compound has been shown to have affinity for, or activate, a number of other receptors

including the D5R, the 5HT-2C receptor, and the α2c-adrenergic receptor (Andringa et al., 1999;

Sahu et al., 2009; Chun et al., 2013). In addition, there are also conflicting reports as to whether

SKF 83959 acts as an antagonist, partial agonist, or has no effect at the D1R (Andringa et al.,

1999; Jin et al., 2003; Rashid et al., 2007; Lee et al., 2014). The lack of specificity of SKF

83959 is a major caveat when the compound is used in studies that aim to address the

physiological effects of the D1-D2 heteromer, as systemic administration of SKF 83959 could

result in various off-target effects in addition to D1-D2 heteromer activation. In the past, SKF

83959 was used in combination with D1R and D2R antagonists or in gene-deleted mice to

validate D1-D2 heteromer-specific effects in vitro (Rashid et al., 2007; Hasbi et al., 2009;

Perreault et al., 2012b). However, such manipulations would also inevitably disrupt the

physiological function of the D1R and the D2R in vivo. It was therefore necessary to develop

an antagonist selective to the D1-D2 heteromer to establish specificity for SKF 83959-induced

effects.

Using serial deletions and point mutations of the D1R and the D2R, it was identified

that the two receptors heteroligomerize via electrostatic interactions between two glutamic acid

residues on the C-terminus of the D1R and two arginine residues on the ICL3 of the D2R

(O’Dowd et al., 2012; Hasbi et al., 2014). The loss of any of these four residues abolished the

BRET signals in HEK293 cells (Hasbi et al., 2014), and prevented the nuclear translocation of

D1R with D2R-NLS in HEK293 cells (O’Dowd et al., 2012). A short peptide was then

synthesized that was comprised of the same 16 amino acid sequence as the D1R C-terminus

31

that included the two Glu residues. The peptide was fused to a TAT sequence, a cell-

penetrating motif derived from human immunodeficiency virus, at its N-terminus to render it

cell permeable (Schwarze et al., 1999). The resulting TAT-D1 peptide was able to dose-

dependently reduce the BRET and FRET signals elicited by the D1-D2 heteromer when it was

incubated with HEK293 cells stably co-expressing both receptors and primary cultured striatal

neurons, respectively (Hasbi et al., 2014). In addition, the TAT-D1 peptide also reduced the co-

immunoprecipitation of D1R and D2R in cells as well as in native striatal tissues, and dose-

dependently attenuated SKF 83959-induced calcium mobilization in primary cultured striatal

neurons (Hasbi et al., 2014). The specificity of the TAT-D1 peptide towards the D1-D2

heteromer was confirmed by findings showing that the BRET signals elicited by D1R

homoligomer, D2R homoligomer, or D2-D5 heteromer were not affected by TAT-D1

administration (Hasbi et al., 2014). Furthermore, a TAT-scrambled control peptide that

contains the same amino acid residues as the TAT-D1 peptide but in random order had no

effect in the BRET, FRET, Co-IP and calcium mobilization experiments (Hasbi et al., 2014).

Collectively these findings indicate that the TAT-D1 peptide was able to selectively disrupt the

D1-D2 heteromer formation and antagonize its signalling. Therefore, the use of the TAT-D1

peptide would allow us to exclusively determine D1-D2 heteromer-specific physiological

effects.

1.3.6 Physiological Effects of the D1-D2 Heteromer

A physiological role for the D1-D2 heteromer in vivo was first demonstrated in rat

striatal neurons and tissue which showed increased phosphorylation and total expression of

32

calcium calmodulin kinase II (CaMKII), specifically in the NAc, as a result of increased

intracellular calcium release following acute D1-D2 heteromer stimulation (Lee et al., 2004;

Rashid et al., 2007; Hasbi et al., 2009; Ng et al., 2010). CaMKII has been highly implicated in

the regulation of synaptic plasticity through the modulation of glutamatergic transmission

(Anderson et al., 2008; Jenkins and Traynelis, 2012), and has also been shown to play a role in

the transcriptional regulation of brain-derived neurotrophic factor (BDNF) through

phosphorylation of the transcriptional repressor MeCP2 (Zhou et al., 2006). Indeed, it was

shown that subsequent to CaMKII phosphorylation following acute D1-D2 heteromer

stimulation by SKF 83959 also increased the expression of BDNF in both the NAc and the

VTA, and promoted morphological maturation and differentiation of primary cultured striatal

neurons (Hasbi et al., 2009; Perreault et al., 2012a). In contrast, a reduction in BDNF levels

and CaMKII phosphorylation was observed in the SN following acute D1-D2 heteromer

stimulation, whereas no effect was observed in the PFC (Perreault et al., 2012a, 2013). The

involvement of the D1-D2 heteromer in the modulation of CaMKII phosphorylation and BDNF

expression in the striatum was further confirmed using primary cultured striatal neurons

derived from D1R or D5R gene-deleted transgenic mice, which showed that the absence of

D1R, but not the D5R, abolished the effect of SKF 83959 on CaMKII phosphorylation and

BDNF expression in striatal neurons (Hasbi et al., 2009), likely due to the lack of D1-D2

heteromer formation in the D1R knockout mice.

As described in Section 1.3.2, the D1-D2 heteromer is expressed in a specific subset of

MSNs in the NAc that co-express both the D1R and D2R. Intriguingly, we further

demonstrated that these D1-D2 heteromer-expressing MSNs also co-express both GABA and

glutamate (Perreault et al., 2012a), thus potentially allowing these neurons to exert dual

33

modulation on the neuronal activity of specific regions in the brain via their efferent

projections. Indeed, we showed that acute D1-D2 heteromer stimulation by SKF 83959

increased the expression of a major GABA-producing enzyme, GAD67, in the NAc and VTA,

whereas a reduction in GAD67 expression was observed in the SN (Perreault et al., 2012a). On

the other hand, the expression of vesicular glutamate transporter-2 (VGLUT2), which regulates

glutamate release into the synapse, was increased in the CP, VTA and SN, whereas VGLUT1

expression was increased in the SN only, by acute D1-D2 heteromer stimulation (Perreault et

al., 2012a). In addition, selective stimulation of the D1-D2 heteromer in the NAc shell, where

the heteromer is predominantly expressed (Section 1.3.2), increased the expression of GAD67

in the VTA, and enhanced the expression of both GAD67 and VGLUT1/2 in the SN (Perreault

et al., 2012a). Subsequent quantification of GABA and glutamate levels following SKF 83959

treatment using dot blots found that the GABA to glutamate ratio was increased in the NAc,

but was reduced in the CP, in response to D1-D2 heteromer stimulation (Perreault et al., 2012a),

indicating differential modulation of the inhibitory GABAergic tone by the D1-D2 heteromer

in the two regions. Collectively, the ability of the D1-D2 heteromer to regulate GABA and

glutamate activity in a region-specific manner supports the hypothesis that one function of the

D1R/D2R co-expressing neurons, which extend efferent projections to nuclei in both the direct

and indirect pathways, may be to fine-tune neuronal activity and maintain homeostatic balance

between the direct and the indirect pathways (Perreault et al., 2011).

Behaviourally, acute D1-D2 heteromer stimulation by SKF 83959 enhanced basal

locomotion and induced robust grooming in rats (Perreault et al., 2010). This grooming

response elicited by SKF 83959 was further confirmed to involve the D1-D2 heteromer as the

behaviour was not observed following selective D1R stimulation by SKF 83822, was abolished

34

by D2R antagonist raclopride (Perreault et al., 2010) and the D1-D2 heteromer disrupting

peptide TAT-D1 (Hasbi et al., 2014), was absent in transgenic mice gene-deleted for D1R, but

not D5R (Perreault et al., 2012a), and was attenuated in juvenile rats possibly as a result of

lower D1-D2 heteromer expression compared to adult rats (Perreault et al., 2013).

Lastly, in contrast to that was observed with acute stimulation, repeated D1-D2

heteromer stimulation by daily SKF 83959 treatment for 7 days was found to reduce total

CaMKII expression in the NAc without affecting its phosphorylation (Perreault et al., 2010).

Since CaMKII mediates the Ser831 phosphorylation of AMPA glutamate receptor subunit

GluA1 (Snyder et al., 2000), repeated D1-D2 heteromer stimulation concurrently resulted in

reduced GluA1 Ser831 phosphorylation in the NAc (Perreault et al., 2010). However, the exact

mechanism responsible for the opposing effects between acute and repeated D1-D2 heteromer

stimulation on CaMKII phosphorylation remain to be elucidated.

In summary, we have identified a novel dopamine receptor complex in the brain with a

novel signalling properties and a unique expression pattern. Furthermore, D1-D2 heteromer

stimulation in vivo modulated the expression of various proteins in the brain that have been

critically implicated both depression and drug addiction, such as BDNF and GAD67, which

will be discussed in detail in Section 1.5.

1.4 The Role of Dopamine in the Brain Reward System

A reward is a stimulus that, when given to humans or animals, will alter their behaviour

to reinforce them to actively acquire such stimulus. The brain reward system is critical for the

reward function in response to various external stimuli such as food, water, and sex, and is

35

therefore evolutionarily crucial for survival (Wise, 2004; Grace et al., 2007). A dysfunctional

brain reward system has been critically implicated in the pathophysiology of neuropsychiatric

disorders such as depression and drug addiction (Koob and Volkow, 2010b; Russo and Nestler,

2013). The primary reward pathway in the brain consists of the dopaminergic projection from

the VTA to the NAc, where the limbic information is translated into goal-directed behaviour

via the projections from the NAc to various nuclei of the basal ganglia motor circuitry (Sesack

and Grace, 2010). As the D1-D2 heteromer is predominantly expressed in the NAc, and since

its stimulation was shown to modulate the expression of various proteins in both the NAc and

the VTA, it possible that the D1-D2 heteromer may play an important role in the modulation of

brain reward function.

1.4.1 The Neurocircuitry of the Brain Reward System

Although the primary component of reward circuit is comprised of dopaminergic

projections from the VTA to the NAc, also known as the mesolimbic circuitry, the activity of

this pathway can be in turn modulated by inputs from nuclei of the cortical and limbic regions,

including the PFC, basolateral amygdala (BLA), the ventral subiculum (vSub) of the

hippocampus for the NAc, and PFC, pedunculopontine tegmentum (PPTg), laterodorsal

tegmentum (LDTg), rostromedial tegmental nucleus (RMTg), and the lateral habenula (Lhb)

for the VTA. Collectively, interconnections between these nuclei and the mesolimbic pathway

comprise the brain reward system (Figure 6).

36

Figure 6: The neurocircuitry of the brain reward system. The NAc receives dopaminergic

projections from the VTA and glutamatergic projections from the mPFC, BLA, and vSub. The

NAc itself sends out GABAergic projections to the VTA and the lateral hypothalamus. The

LDTg and RMTg respectively send glutamatergic and GABAergic projections to the VTA to

control dopamine release in the NAc. The RMTg in turn receives glutamatergic projections

from the Lhb.

NAc: Nucleus Accumbens, VTA: Ventral Tegemental Area, BLA: Basolateral Amygdala,

vSub: Ventral Subiculum, LDTg: Laterodorsal Tegmentum, RMTg: rostromedial tegmental

nucleus, LHb: Lateral Habenula. (Figure cited from Russo and Nestler 2013, Figure 1).

37

1.4.1.1 The Nucleus Accumbens

The NAc is critical for the regulation of reward-seeking behaviours (Mogenson et al.,

1980; Groenewegen et al., 1996). The NAc receives afferent projections from limbic regions

such as the VTA, BLA, and the vSub (Kelley and Domesick, 1982; Kelley et al., 1982;

Groenewegen et al., 1987), and sends efferent projections to nuclei of the basal ganglia

circuitry including the VP, STN, SNr, and the thalamus (Heimer et al., 1991; Usuda et al., 1998;

Dallvechia-Adams et al., 2001). Therefore, anatomically the NAc serves as an interface

between the limbic and motor regions to convert learned associations of motivational

significance into goal-directed behaviours (Sesack and Grace, 2010). The NAc is subdivided

into two anatomically and functionally distinct regions, the NAc shell and NAc core.

Traditionally it has been thought that the NAc shell is linked to the limbic system (Rodd-

Henricks et al., 2002; Sellings and Clarke, 2003), whereas the NAc core is part of the basal

ganglia motor circuitry (Nicola, 2007). Furthermore, the NAc shell is primarily implicated in

the regulation of reward- or drug-seeking behaviour by spatial or contextual information,

whereas the NAc core is in control over such behaviours by discrete cues (Robbins et al., 2008).

Approximately 90% of the neurons in the NAc are GABAergic medium spiny neurons

(MSNs), with the remaining neurons being cholinergic and GABAergic interneurons

(Kawaguchi et al., 1995; Meredith, 1999). In addition to the dopaminergic projections from the

VTA, the NAc also receives excitatory glutamatergic projections from the medial and lateral

PFC, the BLA, and the vSub that modulates goal-directed behaviours generated by different

sources (Ishikawa et al., 2008; Ito et al., 2008; Gruber et al., 2009). Specifically, inputs from

the PFC are thought to supply executive control, BLA to communicate information regarding

conditioned associations and affective drive, and the vSub to provide spatial and contextual

38

information (Ishikawa et al., 2008; Ito et al., 2008; Gruber et al., 2009). Interestingly, it has

been shown that the dopaminergic and glutamatergic inputs converge onto the same spines of

the MSNs in the NAc, forming a “synaptic triad” that allows dopamine to modulate the

plasticity of glutamatergic transmissions into the NAc, a process that is crucial for the

establishment of stable reward-seeking behaviours (Dehaene and Changeux, 2000; Wolf et al.,

2004; Moss and Bolam, 2008). In addition, the NAc also receives the reciprocal inhibitory

GABAergic input from the VP and modulatory dopaminergic input from the substantia nigra

pars compacta (SNc) that function to provide negative feedback control and to modulate motor

function, respectively (Brog et al., 1993; Ikemoto, 2007).

The targets of efferent MSNs from the NAc include the VTA, VP, and the SNr (Haber

et al., 1990; Heimer et al., 1991; Usuda et al., 1998). The projections from the NAc shell to the

VTA have been shown to influence the dopaminergic neurons that project back to the NAc

core, thereby creating a medial to lateral series of spirals that allow limbic associated structures

to progressively influence transmission in more motor-related parts of basal ganglia (Nauta et

al., 1978; Zahm and Heimer, 1993; Haber et al., 2000). Through this spiral of connections,

motivational limbic information can be processed into appropriate motor responses to obtain

reward. The projections from the NAc to the SNr of the basal ganglia are connected directly via

the striatonigral and indirectly via the striatopallidal pathways, where the D1R and the D2R are

respectively expressed (Gerfen and Surmeier, 2011) (see Section 1.2.2 & 1.3.2). For the NAc

core, cortical activation of the direct pathway to the EPN/SNr complex leads to the

disinhibition of appropriate action plans that facilitate reward acquisition (Redgrave et al.,

1999). On the other hand, cortical activation of the indirect pathway from the NAc core to the

GP and STN before reaching the SNr is thought to inhibit actions that are maladaptive for

39

obtaining reward, or for the purpose of avoiding punishment (Redgrave et al., 1999). In

contrast to the NAc core, the direct and indirect pathways from the NAc shell are less well-

defined since the shell subregion is a hybrid structure that is part basal ganglia and part limbic

(Heimer et al., 1991). One theory suggests that both direct and indirect pathways from the NAc

shell first project the GP, where a subset of neurons extend efferent projections to the

mediodorsal thalamus to form the direct pathway, and another subset to the STN to form the

indirect pathway (O’Donnell et al., 1997; Nicola et al., 2000). Another theory postulates that

the direct and indirect pathways from the NAc shell converge onto the VTA instead of the SNr,

with the direct pathway involving MSNs from the NAc shell to the VTA, and the indirect

pathway consisting of MSNs from the NAc shell to the ventromedial VP before reaching the

VTA (Sesack and Grace, 2010). In addition, a subset of striatal MSNs that co-express

DYN/ENK-D1R/D2R were shown to project to nuclei within both the direct and indirect

pathways, specifically to the GP, EPN, and the VP (Perreault et al., 2011) (Figure 2 in Section

1.2.2). Unpublished data from our laboratory also indicated a putative projection of the

D1R/D2R co-expressing striatal MSNs from the NAc shell to the PFC. The function of these

DYN/ENK-D1R/D2R co-expressing striatal MSNs remain to be elucidated, although it has

been postulated that they function to maintain homeostatic balance between the direct and the

indirect pathways due to their efferent projections in both pathways (Perreault et al., 2011).

1.4.1.2 The Ventral Tegmental Area

The VTA is located in the midbrain in close proximity with the SN, and is the major

source of dopaminergic input into the NAc. The VTA-NAc, mesolimbic dopamine circuitry is

40

consistently recruited by both drugs of abuse and natural rewards, making it an essential

component of the brain reward system (Wise, 2004). Approximately 60-65% of neurons in the

VTA are dopaminergic, and they are highly heterogeneous in their morphological

characteristics, afferent influences, efferent projection targets, and firing properties (Nair-

Roberts et al., 2008). The non-dopaminergic neurons in the VTA are primarily GABAergic

(30-35%), either as interneurons or as projection neurons that function to provide local

inhibition of dopaminergic neurons and long-range inhibition of projection regions (Nair-

Roberts et al., 2008). The remaining neurons in the VTA were found to be glutamatergic (2-

3%), with the ability to co-release dopamine (Yamaguchi et al., 2007).

The projection targets of the VTA vary depending on the location and the types of

neurons that originate from the VTA. Neurons from the medial VTA project to the NAc, VP,

STN, olfactory tubercle, and the amygdala (Lammel et al., 2008). The medial most rostral

linear VTA sends projections to VP, olfactory tubercle, preoptic and lateral hypothalamic areas,

Lhb, mediodorsal thalamus, and also to a lesser extent to the PFC, BLA, and dorsal raphe (Del-

Fava et al., 2007). The caudomedial and ventromedial VTA innervate the NAc, the bed nucleus

of the stria terminalis, the pallidum, the central amygdaloid nucleus (CeA), and the BLA (Del-

Fava et al., 2007). The mesostriatal (VTA-NAc) projections can be further subdivided, with the

posterior VTA projecting to the medial NAc shell, and the lateral VTA innervating the NAc

core and the lateral shell (Ikemoto, 2007). Furthermore, the dopaminergic, GABAergic and

glutamatergic neurons from the VTA each extend efferents to different destinations. The

dopaminergic neurons from the VTA project primarily to the NAc, mPFC, and the amygdala

(Ungless and Grace, 2012), while the GABAergic neurons extensively innervate the VP, lateral

hypothalamus, and the Lhb (Taylor et al., 2014). Lastly, the glutamatergic neurons from the

41

VTA were shown to project to the NAc shell, PFC, VP, amygdala, and the Lhb (Hnasko et al.,

2012). Recent studies using optogenetics to selectively stimulate specific projection neurons

from the VTA delineated that each subset of neurons influence reward-seeking behaviour in

different manners depending on the projection targets and neuronal subtypes (Reviewed in:

Lammel et al. 2014).

The VTA receives excitatory glutamatergic projections from a wide array of structures;

each provides a modest input to the VTA but also innervates other regions that also project to

the VTA. Therefore, VTA neuronal activity is modulated by an integrated network of inputs

rather than by a discrete set of brain structures (Geisler and Zahm, 2005). The major source of

excitatory cortical input originates from the PFC, involving predominantly the prelimbic and

infralimbic cortices, and to a lesser extent the cingulate and orbital divisions (Geisler and Zahm,

2005). The PFC projections synapse onto dopaminergic neurons in the VTA that project back

to the PFC to create a microcircuit that allows DA to modulate glutamatergic input from the

PFC (Carr and Sesack, 2000). Neurons from the PFC also innervate GABAergic neurons

within the VTA that in turn project to the NAc that allows the PFC to indirectly inhibit NAc

neuronal activity (Carr and Sesack, 2000). Projections from the LDTg and PPTg are another

source of excitatory input to the VTA. Stimulation of the LDTg/PPTg complex has been shown

to promote burst-firing of dopaminergic neurons to enhance dopamine release from the VTA to

the NAc lateral shell that ultimately led to reward-seeking behaviours (Lammel et al., 2012).

Other sources of excitatory input to the VTA include the lateral hypothalamus, VP, Lhb,

periaqueductal grey, and reticular formation (Geisler et al., 2007). In particular, the inputs from

the lateral hypothalamus, VP, and the Lhb to the VTA have also recently been implicated in the

42

regulation of reward-related behaviours (Aston-Jones et al., 2010; Lammel et al., 2012; Mahler

et al., 2014).

The major extrinsic inhibitory inputs into the VTA arise from the GABAergic neurons

of the NAc shell and the VP, which function as a negative feedback and a gating mechanism,

respectively, for dopamine release in response to rewarding stimuli (Heimer et al., 1991;

Geisler and Zahm, 2005). Recently, the RMTg has been identified to be another important

source of inhibitory input to the VTA. The RMTg is located just caudal to the VTA and sends

extensive GABAergic projections to the VTA (Jhou et al., 2009b). The activity of the RMTg in

turn is under the influence of excitatory glutamatergic input from the Lhb (Jhou et al., 2009b).

Optogenetic stimulation of the Lhb was found to enhance RMTg GABAergic activity that

selectively inhibits the dopaminergic neurons from the VTA to the NAc lateral shell to produce

aversive conditioning (Lammel et al., 2012). Furthermore, Lhb activity, in turn, can be

modulated by the GABAergic and glutamatergic neurons from the VTA. Selective stimulation

of VTA GABAergic neurons that project to the Lhb was shown to inhibit Lhb neuronal firing

to promote reward-seeking behaviour (Stamatakis et al., 2013), whereas the stimulation of

VTA glutamatergic neurons to the Lhb had the opposite effect (Root et al., 2014). Therefore,

the VTA is able to modulate its own neuronal activity via its efferent projections to the Lhb. In

addition to the extrinsic GABAergic projections, the effects of local GABAergic interneurons

on VTA neuronal activity have also been recently elucidated. Selective optogenetic stimulation

of GABAergic interneurons within the VTA was shown to reduce the spontaneous firing rate

of VTA dopaminergic neurons that in turn led to the production of conditioned place aversion

and the disruption of reward consumption (Tan et al., 2012; Van Zessen et al., 2012).

43

1.4.2 The Role of Mesolimbic Dopamine in Depression and Drug Addiction

Given the ability of the mesolimbic dopamine system to modulate both reward and

aversion, numerous studies have investigated the role of the mesolimbic dopamine system in

neuropsychiatric disorders that are characterized by dysfunctional brain reward function,

mainly depression and drug addiction (Nestler and Carlezon, 2006a; Koob and Volkow, 2010a;

Baik, 2013; Russo and Nestler, 2013).

One of the major symptoms of depression is the emergence of anhedonia, which is the

inability to perceive natural reward (Der-Avakian and Markou, 2012). Therefore, it has been

postulated that the mesolimbic dopamine system may be critically involved in the

pathophysiology of depression (Russo and Nestler, 2013). Studies have shown that chronic

restraint stress and repeated social defeat stress, which are rodent models for depression,

increased spontaneous and burst/phasic firing of VTA dopaminergic neurons both in vivo and

ex vivo in brain slices (Anstrom and Woodward, 2005; Chaudhury et al., 2013). More

importantly, chronic administration of the antidepressant fluoxetine was able to reverse the

enhancement of VTA dopaminergic neuron activity induced by both chronic stress models for

depression (Anstrom and Woodward, 2005; Chaudhury et al., 2013). Nevertheless, the precise

involvement of the mesolimbic dopamine system in depression is complicated by two recent

optogenetic studies that yielded opposite conclusions. In the study by Chaudhury et al (2013),

optogenetically induced burst/phasic firing of VTA dopaminergic neurons was shown to

produce a susceptible, depression-related behavioural phenotype in mice that underwent

subthreshold social defeat stress, as well as in previously resilient animals that underwent

repeated social defeat stress. They further showed that the pro-depressive effect of VTA

dopaminergic neuron burst firing was mediated by dopamine release into the NAc but not the

44

mPFC. In contrast, the study by Tye et al (2013) demonstrated that optogenetic phasic

stimulation of VTA dopaminergic neurons reversed the depressive behavioural phenotypes in

animals that were subjected to chronic mild stress, and that the antidepressant-like activity of

VTA dopaminergic stimulation was dependent on dopamine receptor activation in the NAc. In

addition to the differences in the stress models used, a possible explanation for the

discrepancies between the two studies would be that each study may have targeted different

subpopulations of VTA dopaminergic neurons that project to the NAc (Walsh and Han, 2014).

Indeed, as mentioned earlier, two distinct subpopulations of VTA dopaminergic neurons that

innervate the NAc were identified that differ drastically in their electrophysiological properties

(Hnasko et al., 2012). Therefore, it is likely that the optogenetic stimulation of each

subpopulation may result in different behavioural outcomes. Nevertheless, both studies

confirmed the potential involvement of the mesolimbic dopamine system in depression-related

behaviours despite the opposite direction of modulation.

In addition to major depression, numerous studies have conclusively demonstrated that

the mesolimbic dopamine system plays a central role in the pathogenesis of drug addiction

(Wolf et al., 2004; Thomas et al., 2008; Koob and Volkow, 2010a; Baik, 2013). Specifically, it

has been postulated that drugs of abuse hijack the brain’s reward system to produce abnormal

neuroadaptations in the mesolimbic circuitry, thus leading to heightened responses to drug-

associated stimuli and persistent drug-seeking behaviours despite negative consequences

(Lüscher and Malenka, 2011). All drugs of abuse, regardless of their respective mechanisms of

actions, ultimately result in increased dopamine release from the VTA to the NAc and an

AMPA glutamate receptor-mediated LTP of the VTA dopaminergic neurons (Lüscher and

Malenka, 2011). Moreover, although natural rewards such as food also induced LTP of VTA

45

dopaminergic neurons, its effect was transient, while the LTP caused by cocaine self-

administration was shown to last for more than 3 months (Chen et al., 2008). However, the

heterogeneity of VTA dopaminergic neurons necessitates further studies to identify the

selective neurons that are modified by drugs of abuse. To address this issue, a recent study by

Lammel et al. (2012) found that cocaine potentiated the excitatory synapse of VTA

dopaminergic neurons that project to the NAc medial and lateral shell, but not the synapse of

neurons that innervate mPFC, thus establishing the critical involvement of VTA-NAc shell

projection in mediating the behavioural effect of drugs of abuse.

1.5 The D1-D2 Heteromer as a Modulator of Brain Reward Function: Relevance to

Addiction and Depression

As described earlier in Section 1.3.6, D1-D2 heteromer stimulation modulated the

expression of several proteins in the NAc, including CaMKII, BDNF, and GAD67 (Hasbi et al.,

2009; Perreault et al., 2012a). Interestingly, each of these proteins has been previously

implicated in reward-related behaviours in the context of drug addiction and depression (Xi et

al., 2003; Fuchs et al., 2008; Russo and Nestler, 2013; Robison, 2014; Li and Wolf, 2015).

Thus, the D1-D2 heteromer may act as a modulator of brain reward function via its ability to

modulate the expression of CaMKII, BDNF, and GAD67 in the NAc.

46

1.5.1 CaMKII

CaMKII is a multifunctional Ser/Thr protein kinase that is critically involved in

learning and memory, with the α and β isoforms being the most abundant in the brain (Hudmon

and Schulman, 2002). A rise in intracellular calcium level initiates the binding of CaM to

CaMKII to relieve autoinhibition and allow the phosphorylation of its substrates following

autophosphorylation at Thr286 (Coultrap et al., 2012). Therefore, the calcium signalling

elicited by acute D1-D2 heteromer stimulation was able to increase the activation and total

expression of CaMKII in the NAc (Hasbi et al., 2009). However, repeated D1-D2 heteromer

stimulation was shown to, unexpectedly, reduce total CaMKII expression in the NAc (Perreault

et al., 2010).

Recent studies have implicated CaMKII function in the NAc in reward-related

neuropsychiatric disorders such as drug addiction and major depression due to its ability to

modulate synaptic plasticity in the mesolimbic reward system, mainly via the phosphorylation

of AMPA and NMDA glutamate receptors (Reviewed in: Robison 2014). Specifically, repeated

exposure to psychostimulants such as cocaine and amphetamine was shown to increase

CaMKII expression in the NAc (Loweth et al., 2010; Robison et al., 2013), which in turn

resulted in the accumulation of transcription factor ΔFosB to lead to increased expression of

several genes that could mediate the subsequent drug-seeking behaviours (McClung and

Nestler, 2003; Robison et al., 2013). Indeed, herpes simplex viral (HSV) vector-mediated

overexpression of CaMKII in the NAc was found to enhance amphetamine self-administration

(SA) (Loweth et al., 2010), whereas pharmacological inhibition of CaMKII in the NAc shell

was shown to attenuate the cocaine-induced reinstatement of cocaine-seeking behaviours, a

behavioural paradigm that models relapse to drug-taking in humans (Anderson et al., 2008).

47

With respect to depression, it has been shown that HSV vector-mediated inhibition of CaMKII

activity in the NAc prevented the manifestation of depressive phenotype as measured by

sucrose preference test in rats that underwent chronic social defeat stress, a commonly used

paradigm to induce a depressive state in rodents (Robison et al., 2014). Conversely, chronic

social defeat stress alone enhanced the binding of ΔFosB to CaMKII promoter to increase its

transcription, an effect that was abolished by choric treatment of antidepressant fluoxetine

(Robison et al., 2014). Collectively, these findings clearly implicated a role for CaMKII in

mediating reward-seeking and depression-related behaviours. Therefore, as acute and chronic

D1-D2 heteromer stimulation were able to differentially affect CaMKII expression in the NAc

(Hasbi et al., 2009; Perreault et al., 2010), this receptor complex may thus be potentially

involved in the etiology of drug addiction and major depression.

1.5.2 BDNF

BDNF is a neurotrophic factor that is important for neuronal cell growth and survival

(Lessmann et al., 2003). BDNF is synthesized as a pro-peptide (32kDa) in the ER and is

proteolytically processed either intracellularly or extracellularly into its mature form (14kDa).

Although bdnf mRNA is present in the NAc, BDNF in the NAc was thought to originate

predominantly from the PFC and VTA through anterograde axonal transport rather than from

local synthesis (Lessmann et al., 2003). However, we have demonstrated that D1-D2 heteromer

stimulation resulted in increased BDNF expression in striatal neurons and in the NAc (Hasbi et

al., 2009), thus providing a novel mechanism through which BDNF can be synthesized locally

in the NAc. The binding of mature BDNF to its receptor TrkB, initiates TrkB dimerization and

48

autophosphorylation of intracellular tyrosine residues, which would lead to the activation of

signalling cascades including Ras/ERK, phosphoinositide 3-kinase (PI3K)/Akt, and

phospholipase Cγ (PLC-γ) /Ca2+

pathways (Corominas et al., 2007).

Several studies have implicated BDNF in the NAc as an important mediator of drug-

seeking and depression-related behaviours. For instance, the development of stable cocaine SA

behaviour was associated with increased BDNF expression in the NAc (Grimm et al., 2003),

and the infusion of BDNF or BDNF antibody into the NAc was shown to respectively enhance

or reduce cocaine SA (Graham et al., 2007a). Similarly, increased BDNF expression was also

observed in the NAc of animals that underwent chronic social defeat stress (Berton et al., 2006).

Moreover, animals that exhibited reduced immobility in the forced swim test, another a

commonly used animal model for depressive behaviour, exhibited reduced BDNF expression

in the NAc (Sequeira-Cordero et al., 2014), whereas animals displaying anhedonic behavior in

the sucrose preference test showed increased NAc levels of BDNF mRNA (Bessa et al., 2009).

Consequently, as D1-D2 heteromer stimulation is able to modulate BDNF expression in the

NAc, it is possible that this receptor complex may similarly modulate the drug-seeking and

depression related behaviours that are, in part, mediated by BDNF in the NAc.

1.5.3 GAD67

Glutamic acid decarboxylase (GAD) is an enzyme that is responsible for the

decarboxylation of glutamate to produce GABA (Kaufman et al., 1991). Two distinct isoforms

of GAD exists, namely GAD65 and GAD67, that differ in their subcellular localizations,

functions, regulatory properties and co-factor interactions (Kaufman et al., 1991). Due to its

49

function as a major GABA producer, the level of GAD in the brain closely reflects the

inhibitory GABAergic activity in regions where it is expressed. Particularly, the expression of

GAD67 was shown to be increased in the NAc following D1-D2 heteromer stimulation

(Perreault et al., 2012a), suggesting that the D1-D2 heteromer may exert negative modulation

on the neuronal activity in this region. Indeed, further analysis showed that D1-D2 heteromer

stimulation also enhanced GABA to glutamate ratio in the NAc (Perreault et al., 2012a).

The dysregulation of GABAergic activity within the NAc has been shown to be

associated with the etiology of drug addiction and depression. For instance, the development of

cocaine SA was accompanied by a reduction in GABA level in the NAc, whereas the

extinction of such cocaine-seeking behaviour was associated with increased GABA level in the

same region (Wydra et al., 2013). The microinjection of a GABA agonist, baclofen, into the

NAc also attenuated context-induced reinstatement of cocaine SA (Fuchs et al., 2008). On the

other hand, increased GABA level was observed in the NAc of animals that underwent

psychostimulant withdrawal (Xi et al., 2003), a paradigm that also induces a depressive state in

animals (Barr and Markou, 2005a). Along the same lines, the neuronal activity of the NAc was

also found to be reduced in depressed patients (Mayberg et al., 2000). Together these findings

suggest that enhanced GABAergic activity in the NAc would not only suppress reward-seeking

behaviour, but also further promote a depressive state, likely due to reduced dopamine release

in the same region (Shirayama and Chaki, 2006). Consequently, it is possible that D1-D2

heteromer stimulation, which indirectly increased GABAergic activity in the NAc by

increasing GAD67 expression, may exert negative modulation on reward-seeking behaviours,

and if excessive, may also induce a depressive state.

50

6. Research Rationale and Objectives

Emerging evidence has suggested that abnormal dopaminergic neurotransmission in the

mesolimbic circuitry contributes to the pathogenesis of neuropsychiatric disorders involving a

dysfunctional brain reward system, such as depression and drug addiction. However, despite

our growing knowledge of the mesolimbic dopamine system, currently available interventions

that target the dopamine system show little and variable efficacies. As a result, a novel

treatment target is desirable or novel strategy is warranted to better manage the conditions.

Our laboratory has identified and thoroughly characterized a novel dopamine receptor

complex, the dopamine D1-D2 receptor heteromer. We have demonstrated that the D1-D2

heteromer is predominantly expressed in the NAc, a region that is critically involved in brain

reward function. In addition, D1-D2 heteromer stimulation by agonist SKF 83959 leads to

Gαq-mediated, PLC-dependent calcium signalling that subsequently increases the expression

of proteins that have been strongly implicated in drug addiction and depression in the NAc, and

thus it is possible that the D1-D2 heteromer may also be involved in the pathophysiology of

these neuropsychiatric disorders. Therefore, the goal of my thesis is to characterize the function

of the dopamine D1-D2 receptor heteromer in reward-related behaviours, specifically in the

context of depression and drug addiction. I investigated the effects of D1-D2 heteromer

stimulation by SKF 83959 and its inactivation by the TAT-D1 peptide on depression-related

and reward-seeking behaviours using well-established animal models for aspects of depression

and drug addiction, respectively. In addition, I also examined whether and how D1-D2

heteromer stimulation and inactivation modulate the neuronal activity of specific nuclei within

the brain reward circuitry using immunohistological staining for c-Fos, a commonly used

marker for neuronal activity.

51

OVERALL HYPOTHESIS: D1-D2 heteromer stimulation and inactivation will differentially

modulate reward-related behaviours in rodents.

The specific hypotheses were,

Hypothesis 1: D1-D2 heteromer stimulation and inactivation will bidirectionally modulate

psychostimulant-induced behaviours.

Hypothesis 2: D1-D2 heteromer stimulation will induce a depression- and anxiety-like

behavioural phenotype, while D1-D2 heteromer inactivation will exert antidepressant-like and

anxiolytic activities.

Hypothesis 3: D1-D2 heteromer stimulation and inactivation will differentially modulate the c-

Fos expression in specific nuclei of the brain reward circuitry.

52

2. The role of the dopamine D1-D2 receptor heteromer in cocaine addiction.

Abstract

The dopamine D1-D2 receptor heteromer is a novel dopamine receptor complex that is

predominantly expressed in the NAc, and its physiological function has yet to be elucidated.

Upon stimulation by agonist SKF 83959, the D1-D2 heteromer elicits a Gαq-mediated, PLC-

dependent calcium signal that subsequently increased expression of CaMKII and BDNF in the

NAc. As both of these proteins have been critically implicated in the etiology of cocaine

addiction, the aim of the current study was to investigate the effects of D1-D2 heteromer

stimulation and inactivation on cocaine-induced behaviours using animal models including the

conditioned place preference (CPP), locomotor sensitization, and intravenous self-

administration (SA) models. The results showed that D1-D2 heteromer stimulation by SKF

83959 induced conditioned place aversion (CPA), abolished the acquisition and expression of

cocaine CPP, prevented the expression of locomotor sensitization to cocaine, attenuated the

maintenance of cocaine SA, and obviated cocaine- and cue-induced reinstatement of cocaine

SA. In contrast, selective D1-D2 heteromer inactivation by the heteromer-disrupting TAT-D1

peptide consistently produced an opposite effect in all three behavioural models examined,

thereby affirming the role of the D1-D2 heteromer in these cocaine-induced behaviours. More

importantly, the sole treatment of the TAT-D1 peptide induced CPP, indicating that the D1-D2

heteromer exerts tonic inhibitory modulation on brain reward function. Collectively, these

findings indicate that the D1-D2 heteromer inhibits cocaine-induced behaviours, and thus may

potentially be utilized as a novel therapeutic target for cocaine addiction.

53

1. Introduction

Cocaine addiction is a chronic relapsing neuropsychiatric disorder that is characterized

by the persistent use of cocaine despite the adverse psychosocial and physiological

consequences. The rewarding effect of cocaine is attributed to its ability to enhance dopamine

level in the mesolimbic circuitry by blocking the reuptake of endogenously released dopamine

via the inhibition of the dopamine reuptake tranportors (DAT) (Gawin, 1991). Approximately

0.3 million individuals suffered from cocaine addiction in Canada in 2011 (Statistics Canada,

2011), and there is currently no definitive therapeutic intervention available for the treatment of

the condition.

The three most commonly used animal tests for studying aspects of cocaine addiction

include conditioned place preference (CPP), locomotor sensitization, and drug self-

administration (SA). The conditioned place preference paradigm is a commonly used animal

model that examines the rewarding or aversive effects of drugs (Tzschentke, 1998). After

repeatedly pairing the hedonic value of a drug with a specific environment, the animal’s

preference to that environment is then assessed in comparison to a saline-associated neutral

environment in a free-choice, drug-naive trial. Conditioned place preference (CPP) and

conditioned place aversion (CPA) are defined respectively by the animal spending more or less

time in the drug-paired context compared to the saline-paired context (Tzschentke, 1998).

Cocaine exhibits a classic biphasic effect in the CPP model, where low to moderate doses of

cocaine (5-20mg/kg) induce CPP and a high dose of cocaine (30mg/kg) produces CPA

(Reviewed in: Aguilar et al. 2009).

54

Locomotor sensitization to cocaine is the progressive augmentation of locomotor

responsiveness to the drug following repeated administration of the same dose in the same

environmental context (Robinson and Berridge, 1993). This phenomenon consists of two

phases: the initiation and the expression. The initiation refers to the successive enhancement of

locomotor activity following daily cocaine injections that is associated with an increase in

extracellular dopamine release in the NAc as a result of reduced D2R autoreceptor inhibition of

VTA dopaminergic neurons (Henry et al., 1989). The expression of sensitization is

characterized by the persistent hyper-responsiveness to the priming injection of cocaine

following a period of withdrawal, which is associated with long-term neuroadaptations within

the mesolimbic circuitry that may underlie the compulsive drug cravings in humans.

Amongst the three animals tests used in this study to assess aspects of cocaine addiction,

the intravenous self-administration paradigm (SA) has the best face validity since it directly

measures drug-taking behaviour as the number of active lever presses and total cocaine

infusions (Woolverton, 1992; Mello and Negus, 1996). In addition, the reinstatement phase of

cocaine SA also models the persistent relapse to drug-taking in humans that frequently occurs

even after a long period of abstinence.

Numerous studies to date have strongly implicated calcium calmodulin kinase II

(CaMKII), dopamine and cAMP-regulated protein phosphatase (DARPP-32), and brain-

derived neurotrophic factor (BDNF) in the NAc in mediating these cocaine-induced behaviours

(Svenningsson et al., 2005; Anderson et al., 2008; Li and Wolf, 2015), and the expressions of

these proteins specifically in the D1R- or the D2R-expressing MSNs were shown to have

differential behavioural effects (Bateup et al., 2008; Lobo et al., 2010a; Robison et al., 2013).

55

More recently, evidence has also emerged that indicated the existence of a MSN

population that co-expressed both the D1R and the D2R (~17% of all MSNs) in the NAc of

mice (Bertran-Gonzalez et al., 2008; Matamales et al., 2009) and rats (Perreault et al., 2010),

within which a novel dopamine receptor complex known as the dopamine D1-D2 receptor

heteromer is expressed. The potential physiological relevance of this unique subset of MSNs

with regards to cocaine addiction remains to be elucidated.

The heteromeric complex formation between the D1R and the D2R was initially

identified in HEK293 cells stably co-expressing both receptors by elucidation of a calcium

signal upon the coactivation of both receptors (Lee et al., 2004). The physical interaction

between the D1R and the D2R was later confirmed in HEK293 cells by bioluminescence

resonance energy transfer (BRET) (So et al., 2005), in primary cultured striatal neurons by

fluorescence resonance energy transfer (FRET) (Hasbi et al., 2009), and in native rat striatum

by co-immunoprecipitation and FRET (Lee et al., 2004; Perreault et al., 2010). The D1-D2

heteromer is predominantly expressed in the NAc on approximately 27% of D1R-expressing

MSNs as FRET analysis indicated that 90% of D1R/D2R co-expressing neurons in the NAc

exhibited heteromer formation (Perreault et al., 2010). The expression of the D1-D2 heteromer

was found to be much less in the CP (6-7% of D1R/D2R co-expressing neurons) (Perreault et

al., 2010). Stimulation of the D1-D2 heteromer required agonist occupancy of both constituent

receptors, and generated a Gαq-mediated, PLC-dependent, intracellular calcium release that

subsequently activated CaMKII in the NAc (Rashid et al., 2007; Hasbi et al., 2009; Ng et al.,

2010). The screening of a series of synthetic benzazepines identified a D1R-like agonist, SKF

83959, that can stimulate a calcium signal without affecting the canonical cAMP pathway

mediated by homomeric D1R and the D2R (Rashid et al., 2007). Using SKF 83959, it was

56

subsequently shown that D1-D2 heteromer stimulation modulated the expression of DARPP-32

(unpublished finding, manuscript in preparation), BDNF (Hasbi et al., 2009), and ΔFosB

(unpublished finding, manuscript in preparation) in the NAc in a manner that was opposite to

that was induced by cocaine administrations. It is likely that such opposing effects of D1-D2

heteromer stimulation on cocaine-induced actions at the molecular level may also be reflected

behaviourally in animal models for cocaine addiction.

Therefore, the purpose of the present study was to examine the physiological effects of

the D1-D2 heteromer stimulation by agonist SKF 83959 in the context of cocaine addiction

using behavioural models including CPP, locomotor sensitization, and SA paradigms. In

addition, as SKF 83959 was reported to have affinity for, or activate, a number of other

receptors, such as the dopamine D5 receptor, the α-adrenergic receptor 2C, and the serotonin

5HT-2C receptor (Sahu et al., 2009; Perreault et al., 2012b; Chun et al., 2013), we have

synthesized a peptide that can selectively disrupt and antagonize the D1-D2 heteromer by

preventing the two glutamic acid residues in the C-terminus of the D1R from electrostatically

interacting with the two arginine residues in the third intracellular loop of the D2R (O’Dowd et

al., 2012; Hasbi et al., 2014). The behavioural effect of D1-D2 heteromer inactivation by this

highly selective disrupting TAT-D1 peptide on cocaine-induced behaviours was also examined.

2. Materials and Methods

2.1 Animals

Adult male Sprague-Dawley rats (Charles River, Canada), weighing 300-350 g at the

start of the experiment, were used. Rats were housed in polyethylene cages in a temperature-

57

controlled colony room, maintained on a 12-h light-dark cycle (lights on at 0700h), with ad

libitum access to food and water. Rats were handled daily for 5 days before the start of the

experiment. All treatments were performed during the light phase of the day-night cycle.

Animals were housed and tested in compliance with the guidelines described in the Guide to

the Care and the Use of Experimental Animals (Canadian Council on Animal Care), and were

approved by the Animal Care Ethics Committee of the University of Toronto. Experimentally

naïve animals were used for each behavioural study.

2.2 Drugs

SKF 83959 hydrobromide (Tocris Bioscience) was dissolved in physiological saline

containing 5% DMSO, and was administered subcutaneously (s.c.). Amphetamine

hydrochloride (Sigma-Aldrich) was dissolved in physiological saline (0.9% NaCl), and was

administered intraperitoneally (i.p.). The TAT-D1 disrupting peptide was dissolved in saline

and administered into the intracerebroventricular (i.c.v.) space 15 minutes prior to vehicle, SKF

83959, or amphetamine injection. For non-drug injections, an equivalent volume of saline or

vehicle was administered. All systemic injections were given at a volume of 1.0 ml/kg just prior

to behavioural testing.

2.3 Surgery

Rats were anesthetized with isoflurane (5%), administered analgesic ketoprofen (5

mg/kg, s.c.) and secured in a stereotaxic frame. A cannula (22-gauge, Plastics One) was placed

58

unilaterally into the intracerebroventricular space close to the midline according to the

following stereotaxic coordinates: AP -0.8mm, ML + 1.3mm, DV – 3.7mm, and was secured by

dental cement anchored with four stainless steel screws (Plastics One) fixed on the dorsal

surface of the skull. AP and ML coordinates were taken from bregma, DV coordinate from the

dura (Paxinos and Watson, 1998). The animals were allowed to recover in their home cage for a

minimum of five days before the experiments were performed. Cannulae placement was

visually validated postmortem in brain slices. For the self-administration studies, a catheter

constructed of silastic tubing was surgically implanted in the right jugular vein. The terminal

end of the catheter consisted of a 22-gauge guide cannula (Plastics One) and was anchored

subcutaneously between the scapulae with a small piece of Marlex mesh.

2.4 Apparatus

2.4.1 Place Preference Chamber

The place preference chamber (Harvard Apparatus, UK) consisted of two equally sized

compartments (45cm in height, 34cm in width, and 40cm in length) interconnected by a

transparent rectangular corridor. The compartments were differentiated by the motifs painted

on the walls (dots or stripes) and the colour (different shade of grey tones, light or dark) and

texture (smooth or rough) of the floor.

59

2.4.2 Locomotor Activity Chamber

The locomotor activity chambers were 20cm in height, 25cm in width, and 40cm in

length. Two arrays of 16 infrared photocells were attached along the longer sides of the

polyethylene cages. The activity chambers were interfaced to a computer that provided

automated recording of horizontal locomotor activity when both top and bottom infrared

photocells were triggered. Ventilated polyethylene lids were used to cover the activity

chambers to prevent animals from escaping.

2.4.3 Operant Chamber

The operant chambers were 21cm in height, 21 cm in width, and 28 cm in length (Med.

Associates Inc., USA). Each chamber contained two response levers, with a Sonalert and a

stimulus light located above each lever. A syringe mounted on a motor driven pump (Razel)

delivered cocaine infusions to the rat’s jagular catheter via Tygon tubing attached to a swivel

located above the test chamber. Each chamber was illuminated by a house light and housed in a

sound-attenuating box equipped with a ventilating fan. All boxes were controlled by a PC

running Med-PC-IV. All programs were written in-house.

2.5 Experimental Procedures

2.5.1 Conditioned Place Preference

The animals were first habituated to the CPP chambers for 2 days, followed by the

measurement of their baseline preference for each of the two chambers. The baseline

60

preference determined which side of the chamber the drug would be paired for a balanced

experimental design. To examine the effects of the D1-D2 heteromer on cocaine CPP

acquisition, the rats underwent 6 days of conditioning sessions, during which they received 3

drug treatments (vehicle, 1.0mL/kg; SKF 83959, 1.0mg/kg, s.c.; cocaine, 10mg/kg, i.p.; TAT-

D1 peptide, 300pmol, i.c.v.; TAT-Sc peptide, 300pmol, i.c.v.) and 3 saline treatments in

alternating order. Immediately after the injection the rats were confined to the assigned

chamber for 30 minutes. The TAT-D1 peptide was given 15 minutes prior to the start of each

conditioning session. On the test day, the rats were allowed to freely explore the two chambers,

and the time the animals spent in each chamber was recorded. To examine the effects of the

D1-D2 heteromer on the expression of cocaine CPP, animals first underwent 6 days of

conditioning sessions with cocaine (10mg/kg, i.p.) and saline in alternating order. On the test

day, the animals received an injection of SKF 83959 (2.5mg/kg, s.c.) or vehicle, and the time

they spent in each chamber was recorded. Place preference or aversion was established if the

animal spent significantly more or less time in drug-paired over saline-paired chamber, and

thus each rat acted as its own control. Animals that spent more than 40% of total time in the

middle compartment were excluded.

2.5.2 Locomotor Sensitization

The animals were first habituated to the locomotor chamber for 3 days following which

they received the assigned drug treatment daily for 7 days (saline, 1.0 ml/kg, i.p.; SKF 83959,

0.4 mg/kg, s.c.; TAT-D1 peptide, 300pmol, i.c.v.; cocaine, 10 mg/kg, i.p.; cocaine + SKF

83959, cocaine + TAT-D1 peptide). The dose of SKF 83959 was previously shown to attenuate

61

amphetamine-induced locomotor sensitization without desensitizing the D1-D2 heteromer over

repeated injections (Shen et al., 2014b), while the dose of cocaine was the same as described by

Robison et al (2014). The locomotor activity of the animals was measured for a total of 60

minutes daily, 30 minutes prior and 30 minutes following the assigned treatment over 7

injection days (Referred to as the development of locomotor sensitization). All animals received

a single injection of priming cocaine (5.0 mg/kg, i.p.) on day 8 following a 24 hour withdrawal,

and their locomotor activity was again evaluated (Referred to as the expression of locomotor

sensitization).

2.5.3 Intravenous Self-Administration

Adult male Sprague-Dawley rats under a restricted diet were first trained to lever-press

for food under an FR1 schedule. Rats were allowed a total of 100 food pellets during each 30

minute training session. The rats that consumed 100 pellets for 3 consecutive days they were

considered lever-trained. Trained animals then underwent surgery for intravenous catheter

implantation to allow for IV cocaine infusion. The animals were allowed to recover from

surgery for a week, and then were first trained for cocaine SA under the FR1, FR3, and then

eventually FR5 schedules of reinforcement. Cocaine infusions were delivered only when the

left (active) lever was pressed, and the delivery was associated with a 5 second light cue

located directly above the left lever. Pressing of the right (inactive) lever had no functional

consequence. For each infusion the animals received 0.25 mg cocaine/0.1mL/5.5s. Once stable

responding was achieved, the dose response relationship of D1-D2 heteromer stimulation (SKF

83959, 0.05, 0.5, 1.5mg/kg, s.c.) and cocaine SA under the FR5 schedule was examined. Next,

62

the animals underwent extinction training, during which no cocaine was infused following

active lever presses. Extinction was achieved when responding on the active lever had reached

a stable level of less than 15 responses over 2h. Once the extinction training was completed,

the animals underwent surgery for intracerebroventricular cannula implantation for TAT-D1

peptide infusions. Following a week of recovery, the effects of D1-D2 heteromer stimulation

(SKF 83959, 0.5, 1.5mg/kg, s.c.) and inactivation (TAT-D1 peptide, 300pmol, i.c.v.) on

cocaine- (5 or 10mg/kg, i.p.) and cue-induced reinstatement of cocaine SA was then

investigated

2.6 Statistical Analysis

All data are reported as mean ± s.e.m, and was analyzed for normality prior to

ANOVAs using the Shapiro-Wilk test. Post-hoc analysis following ANOVAs was performed

using Duncan’s Multiple Range test or Student’s t-test for planned comparisons. Computations

were performed using the SPSS statistical program. Statistical criteria for significant

differences were set at p<0.05.

2.6.1 Conditioned Place Preference

The time spent in the drug-paired chamber versus the saline-paired chamber for each

drug treatment was compared using paired Student’s t-test.

63

2.6.2 Locomotor Sensitization

For the development of locomotor sensitization, the total horizontal activity was

analyzed using repeated measures of ANOVA with Injection (Injections 1 to 7) as the within

subject factor and Treatment (Veh, Cocaine, SKF 83959, TAT-D1 peptide, Cocaine+SKF

83959, Cocaine+TAT-D1 peptide) as the between subject factor. For the expression of

locomotor sensitization, the total horizontal activity was analyzed using one-way ANOVA with

Treatment (Veh, Cocaine, SKF 83959, TAT-D1 peptide, Cocaine+SKF 83959, Cocaine+TAT-

D1 peptide) as the between subject factor. Planned comparisons between groups were done

using Student’s t-test. Planned comparions were conducted in Figure I-5, Expression of cocaine

sensitization, between the following groups: Saline vs. Cocaine; Cocaine vs. SKF+Cocaine;

Cocaine vs. TAT-D1+Cocaine.

2.6.3 Intravenous Self-Administration

For the dose response study that examined the effect of D1-D2 heteromer stimulation

by SKF 83959 on the maintenance of cocaine SA under the FR5 schedule of reinforcement, the

active lever response was analyzed using repeated measures of ANOVA with Dose (0, 0.05,

0.5, 1.5mg/kg SKF 83959) as the within-subject factor. For the reinstatement studies, the active

lever response was analyzed using repeated measures of ANOVA with Treatment (Cocaine,

Cue, SKF 83959, TAT-D1 peptide, Cocaine+SKF 83959, Cue+SKF 83959, Cocaine+TAT-D1

peptide) as the within-subject factor. Planned comparisons between groups were done using

paired Student’s t-test. Planned comparions were conducted in Figure I-8C, Effect of TAT-D1

on cocaine-induced reinstatement, between the following groups: Saline vs. 10mg/kg Cocaine;

64

5mg/kg Cocaine vs. 5mg/kg Cocaine+TAT-D1; 10mg/kg Cocaine vs. 10mg/kg Cocaine+TAT-

D1.

3. Results

3.1 Conditioned Place Preference

We first investigated the effects of D1-D2 receptor heteromer activation and disruption

on basal CPP behaviour. Saline treated animals received saline in both chambers and did not

exhibit preference towards a specific chamber on average (Figure I-1A). SKF 83959 induced

place aversion as the animals spent significantly less time in the drug-paired chamber

compared to the saline-paired chamber on the test day (Figure I-1A, 432.11±80.83s vs.

1055.22±92.26s, t(9)=3.72, p<0.01). In contrast, disruption of the D1-D2 receptor heteromer

by the TAT-D1 peptide alone, but not the scrambled peptide, resulted in place preference

(Figure I-1B, 812.00±59.54s vs. 514.67±27.38s, t(9)=3.87, p<0.01). To further confirm the role

of the D1-D2 receptor heteromer in the induction of place aversion by SKF 83959 treatment,

the animals were co-treated with SKF 83959 plus the TAT-D1 peptide or the scrambled

peptide during the conditioning sessions (Figure I-1C). The place aversion induced by SKF

83959 was abolished by the TAT-D1 peptide but not the scrambled peptide, suggesting that the

aversive effect was mediated by D1-D2 receptor heteromer activation.

The effects of D1-D2 receptor heteromer activation and disruption on the acquisition

and expression of cocaine-induced CPP were next examined. Repeated cocaine treatment

successfully induced a place preference as expected, as the animals spent significantly more

A) B)

65

Figure I-1: The effects of D1-D2 heteromer stimulation and inactivation on basal

conditioned place preference. A) Vehicle-conditioned animals did not exhibit a preference

towards a particular chamber. D1-D2 heteromer stimulation by SKF 83959 induced a

conditioned place aversion as the animals spent significantly less time in the drug paired

chamber, whereas B) its inactivation by TAT-D1 resulted in a conditioned place preference as

the animals spent significantly more time in the drug paired chamber. C) The CPA induced by

SKF 83959 was abolished by the pre-treatment of the TAT-D1 but not the TAT-Sc peptide.

Data represent means ± s.e.m. of n=8-10 rats/group. (*p<0.05, **p<0.01: compared to saline-

paired).

Veh

icle

SKF 8

3959

0

250

500

750

1000

1250

Saline PairedDrug Paired

**

Tim

e (

s)

TAT-D

1-SKF

TAT-S

c-SKF

0

250

500

750

1000

1250

Saline PairedDrug Paired

*

Tim

e (

s)

TAT-D

1

TAT-S

c0

250

500

750

1000

1250

Saline PairedDrug Paired

**

Tim

e (

s)

A) B) C)

66

Figure I-2: The effects of D1-D2 heteromer stimulation and inactivation on cocaine-

induced CPP. A) Vehicle-conditioned animals did not exhibit a preference towards a

particular chamber. Cocaine-conditioned animals exhibited a conditioned place preference

(CPP) as they spent significantly more time in the drug paired chamber. The acquisition

cocaine CPP was respectively abolished and enhanced by SKF 83959 and the TAT-D1 peptide.

B) A single injection of vehicle or SKF 83959 did not affect the chamber preference of vehicle-

conditioned animals. A single injection of vehicle also did not affect the CPP of cocaine-

conditioned animals. However, a single injection of SKF 83959 abolished the expression of

cocaine CPP in cocaine-conditioned animals. Data represent means ± s.e.m. of n=8-10

rats/group. (*p<0.05, **p<0.01: compared to saline-paired).

Veh

icle

Coca

ine

SKF-C

oc

TAT-D

1-Coc

0

250

500

750

1000

1250

Saline PairedDrug Paired

* **

Tim

e (

s)

Veh

+Veh

SKF+V

eh

Coc+

Veh

Coc+

SKF

0

250

500

750

1000

1250

Saline PairedDrug Paired

**

Tim

e (

s)

A) B)

67

Figure I-3: The effect of Cdk5 inhibitor roscovitine on SKF 83959-induced CPA. Animals

conditioned with SKF 83959 exhibited conditioned place aversion (CPA) as the spent

significantly less time in the drug paired chamber. The CPA induced by D1-D2 heteromer

stimulation was abolished by roscovitine pre-treatment. Data represent means ± s.e.m. of n=8-

10 rats/group. (*p<0.05: compared to saline-paired).

Veh

icle

Rosc

o0

250

500

750

1000

1250

Saline Paired83959 Paired

*

Tim

e (

s)

68

time in the cocaine-paired chamber compared to the saline-paired chamber (Figure I-2A,

1050.11±101.53s vs. 581.67±88.54s, t(9)=2.52, p<0.05). Co-treatment of cocaine with SKF

83959 during the conditioning sessions abolished the acquisition of cocaine CPP (Figure I-2A,

613.88±87.85s vs. 815.38±122.23s, p>0.05). Co-treatment of cocaine with the TAT-D1 peptide

during the conditioning sessions had no apparent effect on cocaine CPP, but did increase in

statistical significance (Figure I-2A, 1028.14±122.06s vs. 379.86±75.76s, t(7)=3.28, p<0.01).

The expression of cocaine CPP was similarly abolished by a single injection of SKF 83959 on

the test day (2.5mg/kg, s.c.) in animals that were conditioned with cocaine (Figure I-2B,

758.44±147.53s vs. 894.89±131.51s, t(9)=3.55, p<0.01). The single injection of SKF 83959 on

the test day in saline treated animals had no effect, which was an expected outcome given the

aversive effect of D1-D2 receptor heteromer activation was not associated with a particular

environment.

Lastly, the effect of cdk5 inhibitor roscovitine on CPA induced by D1-D2 receptor

heteromer activation was investigated. (Figure I-3) Since it has previously been shown that the

activation of the D1-D2 receptor heteromer led to increased DARPP-32 Thr75 phosphorylation,

which is regulated by cdk5, it was postulated that cdk5 would be a downstream effector

following D1-D2 receptor heteromer activation. Co-treatment of SKF 83959 plus saline or

roscovitine (25-30nmol/0.5µL) during the conditioning sessions demonstrated that the

inhibition of cdk5 successfully blocked the CPA induced by SKF 83959 (696.17±68.12s vs.

667.33±87.46s, p>0.05), suggesting that cdk5 activation may be responsible for the aversive

effects of the D1-D2 receptor heteromer.

69

Figure I-4: The effects of D1-D2 heteromer stimulation and inactivation on locomotion

induced by acute and repeated cocaine treatment. A) Acute treatment of the TAT-Sc, SKF

83959, and TAT-D1 peptide did not affect basal locomotion. Acute cocaine treatment

significantly increased basal locomotion, which was further significantly enhanced by the

TAT-D1 peptide. An increase in statistical significance was found in animals treated acutely

with cocaine plus SKF 83959 compared to the saline control. B) Repeated treatment of saline,

TAT-Sc, SKF 83959, and the TAT-D1 peptide did not affect basal locomotion over the 7-day

injection period. The locomotor activity of animals treated with repeated cocaine increased

steadily over injections, indicating sensitization. Although the locomotor-stimulating effect of

cocaine was retained in animals co-treated with cocaine plus SKF 83959, the sensitization in

locomotor activity was abolished by SKF 83959 co-treatment. Animals treated with cocaine

plus TAT-D1 exhibited significantly higher locomotor activity compared to cocaine-treated

animals across injections, and the sensitization in locomotor activity occurred at an earlier time

point. Data represent means ± s.e.m. of n=8-10 rats/group. (*p<0.05, ***p<0.001: compared to

Saline; #p<0.05, ##p<0.01, ###p<0.001: compared to Cocaine).

Sal

ine

TAT-S

c

SKF 8

3959

TAT-D

1

Coca

ine

SKF+C

oc

TAT+C

oc0

1000

2000

3000

4000

5000

****

***#

Ho

rizo

nta

l A

ctiv

ity

(b

eam

bre

aks)

A)

1 2 3 4 5 6 70

1000

2000

3000

4000

5000

6000

7000

SalineTAT-ScTAT-D1

SKF+CocCocaine

***

TAT+Coc

*

##

### ### ##

#

**

** *** *SKF 83959

Injection

Ho

rizo

nta

l A

cti

vit

y

(b

ea

m b

rea

ks)

B)

70

3.2 Locomotor Sensitization

The total horizontal activity in response to the assigned treatment for 30 minutes was

reported. One-way ANOVA of the results of acute locomotor activity showed a significant

effect of Treatment {F(6,57)=19.37, p<0.001}. As shown in Figure I-4A, the animals’ basal

locomotion was significantly elevated by acute cocaine treatment (10mg/kg, i.p.), which was

not affected by SKF 83959 co-treatment (0.5mg/kg, s.c.) but was further significantly enhanced

by TAT-D1 pre-treatment (300pmol, i.c.v., p<0.05). Repeated measures of ANOVA on

locomotion across the development of sensitization phase revealed significant effects of

Injection {F(6,57)=3.08, p<0.01}, Treatment {F(6,57)=31.81, p<0.001}, and Injection x

Treatment interaction {F(36,57)=4.21, p<0.001}. Across the 7 injections, the locomotor

activity of saline-treated controls remained low, while SKF 83959, TAT-D1 and TAT-

Scrambled peptide treatment each produced a slight and non-significant increase in locomotion

compared to the saline control (Figure I-4B). In contrast, all cocaine-treated groups (cocaine

alone, cocaine plus SKF 83959, and cocaine plus TAT-D1) had significantly higher locomotor

activity compared to the saline control over the 7 injections (Figure I-4B, p<0.001 vs. Saline

group for all 3 groups). Furthermore, animals treated solely with repeated cocaine exhibited a

steady increase in locomotor activity across injections that became significant on Injection 6

and 7 (p<0.01 and p<0.001, vs. Injection 1, respectively), indicating successful development of

locomotor sensitization. SKF 83959 co-treatment abolished the development of locomotor

sensitization to cocaine but did not alter the locomotor-stimulating effect of cocaine, as the

locomotor activity for the SKF 83959 co-treated group remained elevated and unperturbed

throughout the 7 injections. In contrast, TAT-D1 peptide pre-treatment significantly augmented

cocaine-induced locomotor activity at all injections except for Injection 3. Repeated measures

71

Figure I-5: The effects of D1-D2 heteromer stimulation and inactivation on the expression

of cocaine-induced locomotor sensitization. An injection of a subthreshold dose of cocaine

did not affect the basal locomotor activity of animals previously treated with repeated saline,

TAT-Sc, and SKF 83959, but significantly increased the locomotor acitivity of cocaine-treated

animals, indicating expression of sensitization. In response to the cocaine injection at a

subthreshold dose, a sensitized locomotor phenotype was observed in animals previously

treated with repeated TAT-D1 peptide. The expression of locomotor sensitization was

respectively abolished and enhanced by SKF 83959 and the TAT-D1 peptide. Data represent

means ± s.e.m. of n=8-10 rats/group. (*p<0.05, ***p<0.001: compared to Saline; ###p<0.001:

compared to Cocaine).

Salin

e

TAT-S

c

SKF 8

3959

TAT-D

1

Coca

ine

SKF+C

oc

TAT+C

oc0

1000

2000

3000

4000

5000

****

***###

Ho

rizo

nta

l A

ctiv

ity

(b

eam

bre

aks)

5mg/kg Cocaine

Chronic

Pre-treatment

Priming

72

of ANOVA also showed that the TAT-D1 pre-treated group had significantly higher locomotor

activity than the cocaine-alone group across the 7 injections (p<0.001 vs. Cocaine group).

Moreover, the sensitization of locomotor response to cocaine was not only retained in TAT-D1

peptide pre-treated animals, but it occurred at an earlier time point compared to the cocaine

alone group (Injection 4 vs. Injection 6) and remained elevated throughout the rest of the

injection period.

24 hours following the 7th

injection, animals from all treatment groups received a

cocaine challenge (5mg/kg, i.p.) and their locomotor activity was measured (Figure I-5). One-

way ANOVA revealed a significant effect of Treatment {F(6,56)=20.92, p<0.001}. Repeated

treatment of the TAT-Scrambled peptide or SKF 83959 had no effect on the locomotor activity

induced by cocaine challenge compared to the saline-treated animals. Animals with prior

repeated exposure to the TAT-D1 peptide alone and cocaine alone both exhibited significantly

higher locomotor activity in response to cocaine challenge compared to the saline control,

which is indicative of successful expression of locomotor sensitization to cocaine (p<0.05 and

p<0.001, vs. Saline group, respectively). Interestingly, prior SKF 83959 co-treatment with

cocaine abolished, while prior TAT-D1 peptide pre-treatment with cocaine significantly

enhanced (p<0.001 vs. Cocaine group), the expression of locomotor sensitization to cocaine.

3.3 Intravenous Self-Administration

With successive cocaine SA training, the animals exhibited a steady increase in the

number of active lever presses and the number of infusions, with the latter plateaued at

approximately 25 infusions per 2 hour session (Figure I-6A). Once stable responding in the

73

Veh

icle

0.05

mg/k

g

0.5m

g/kg

1.5m

g/kg

0

50

100

150

200

0

50

100

150

200

***

**

InfusionsPresses

Activ

e L

ev

er

Pre

ss

Infu

sio

ns

Figure I-6: The effect of D1-D2 heteromer stimulation on cocaine self-administration

under the FR5 schedule. A) The animals exhibited steady cocaine self-administration

behaviour over the 2 hour session following training. B) SKF 83959 dose-dependently reduced

the number of active lever press and total cocaine infusions. Data represent means ± s.e.m. of

n=15-16 rats/group. (**p<0.01, ***p<0.001: compared to Vehicle).

1 2 3 4 5 6 7 8 90

25

50

75

100

125

150

175

200

0

10

20

30

40

50

60

70

80PressesInfusions

Training Day

Ac

tive

Le

ve

r P

ress

Infu

sio

ns

A) B)

A) B)

74

number of active lever pressed and cocaine infusions were achieved, three different doses of

SKF 83959 were tested (0.05, 0.5, and 1.5 mg/kg) to examine the dose-response relationship

between D1-D2 heteromer stimulation and cocaine self-administration behaviour under the

FR5 schedule of reinforcement (Figure I-6B). Repeated measures of ANOVA revealed a

significant effect of Dose on the number of active lever responses {F(3,16)=9.85, p<0.001} and

the number of cocaine infusions {F(3,16)=12.14, p<0.001}. Post-hoc comparisons further

showed that the pre-treatment with SKF 83959 dose-dependently reduced the number of active

lever press and the number of infusions that was significant at the 0.5mg/kg (p<0.05) and

1.5mg/kg (p<0.01) doses.

The rats then underwent extinction training, during which the number of active lever

presses steadily decreased and was eventually stabilized at 15 presses per 2 hour session

(Figure I-7). Following extinction training, two separate groups of animals were used to

examine the effect of D1-D2 stimulation on cocaine and cue-induced reinstatement using SKF

83959 at the 0.5mg/kg and 1.5mg/kg dose. A third group of animals was used to examine the

effect of D1-D2 heteromer inactivation on cocaine-induced reinstatement. Repeated measures

of ANOVA revealed a significant effect of Treatment on the number of active lever press

(0.5mg/kg SKF 83959: {F(3,8)=11.26, p<0.0001}; 1.5mg/kg SKF 83959: {F(3,10)=22.66,

p<0.0001}; TAT-D1: {F(5,15)=10.15, p<0.0001}). A priming injection of cocaine at 10mg/kg

successfully reinstated the cocaine SA behaviour as indicated by a significant increase in the

number of active lever presses in cocaine-primed animals compared to the vehicle-treated

controls (Figure 8: t(7)=3.44, p<0.05; Figure 3.4b: t(9)=5.14, p<0.01). SKF 83959 co-treatment

at 0.5mg/kg and 1.5mg/kg (Figure I-8A&B) abolished cocaine-induced reinstatement of

cocaine SA behaviour, while SKF 83959 by itself (without cocaine priming) had no effect. In

75

Figure I-7: The lever-pressing behaviour during extinction training. The number of active

lever press steadily declined over the extinction training, which was eventually maintained at

approximately 15 presses per 2 hour session.

1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 260

10

20

30

40

50

60

70

80

90

100

110

120

Ac

tiv

e L

ev

er

Pre

ss

Training Day

76

Figure I-8: The effects of D1-D2 heteromer stimulation and inactivation on cocaine-

induced reinstatement. A&B) A single injection of saline or SKF 83959 did not reinstate the

SA behaviour. A priming cocaine injection reinstated the SA behaviour, which was abolished

by SKF 83959 at the 0.5 mg/kg and 1.5 mg/kg dose. C) A single injection of TAT-Sc, TAT-D1

or cocaine at a subthreshold dose (5mg/kg) did not reinstate the SA behaviour, whereas the SA

behaviour was reinstated by a priming dose of cocaine (10mg/kg). Pre-treatment of the TAT-

D1 peptide facilitated the reinstatement of SA behaviour induced by a subthreshold dose of

cocaine and further enhanced the reinstatement induced by a priming dose of cocaine. Data

represent means ± s.e.m. of n=10-12 rats/group. (*p<0.05, **p<0.01, ***p<0.001: compared to

Veh and TAT-Sc; #p<0.05: compared to Cocaine).

Veh

1.5m

g/kg S

KF

Coc

1.5

SKF+C

oc0

25

50

75

100

125

150

**

InactiveActive

Lev

er

Pre

sses

TAT-S

c

TAT-D

1

5mg C

oc

TAT-D

1+5o

c

10m

g Coc

TAT-D

1+10

coc

0

25

50

75

100

125#

***

****#

Inactive

Active

Lev

er

Pre

sses

Veh

0.5m

g/kg S

KF

Coc

0.5

SKF+C

oc0

25

50

75

100

125

150

*

InactiveActive

Lev

er

Pre

sses

A) B)

C)

0

25

50

75

100

125

5mg C

oc

TAT-D

1+Coc

Activ

e L

ev

er

Pre

sses

0

50

100

150

200

10m

g Coc

TAT-D

1+Coc

Activ

e L

ev

er

Pre

sses

77

contrast, the co-treatment of the TAT-D1 peptide with the priming cocaine injection at a

subthreshold dose, 5mg/kg (Figure I-8C, t(14)=2.82, p<0.05), and the usual priming dose,

10mg/kg (Figure 8C, t(14)=2.26, p<0.05), further enhanced the number of active lever presses

induced by the cocaine injections (Figure I-8C, t(14)=3.54, p<0.01). Similar to cocaine priming,

the presentation of the light cue that was associated with cocaine delivery during cocaine SA

training was also able to reinstate cocaine SA behaviour (Figure I-9A: t(10)=3.29, p<0.01;

Figure I-9B: t(9)=3.70, p<0.01), an effect that was abolished by SKF 83959 treatment at the

1.5mg/kg dose (Figure I-9B) but not the 0.5mg/kg dose (Figure I-9A). The inactive lever

presses were not affected by any of the drug treatments.

It should be noted that SKF 83959-treated animals exhibited otherwise normal

behaviours such as sniffing and exploring that were comparable to vehicle-treated animals, and

thus the reduction in active lever responses following SKF 83959 administration was not due to

motor impairment. On the contrary, acute SKF 83959 treatment actually enhanced basal

locomotion (Perreault et al., 2010; Shen et al., 2014b).

4. Discussion

In the present study, we assessed the effects of D1-D2 heteromer stimulation by SKF

83959, and its selective inactivation by a disrupting TAT-D1 peptide, on cocaine conditioned

place preference (CPP), locomotor sensitization, and intravenous self-administration (SA),

three most commonly used behavioural models for cocaine addiction. The results consistently

showed that D1-D2 heteromer stimulation attenuated or abolished, whereas its inactivation

78

Figure I-9: The effect of D1-D2 heteromer stimulation on cue-induced reinstatement. The

presentation of a light cue that was previously associated with cocaine infusions was able to

reinstate the SA behaviour. Cue-induced reinstatement of SA behaviour was A) not affected by

0.5mg/kg SKF 83959, but B) was abolished by 1.5mg/kg SKF 83959. Data represent means ±

s.e.m. of n=10-12 rats/group. (**p<0.01: compared to No Cue).

No C

ue

Cue

Alo

ne

Cue+

0.5

SKF

0

25

50

75

100

125

150

****

InactiveActive

Lev

er

Pre

sses

A)

No C

ue

Cue

Alo

ne

Cue+

1.5

SKF

0

20

40

60

80

100

**

InactiveActive

Lev

er

Pre

sses

B)

79

enhanced, cocaine-induced behaviours in all three behavioural paradigms. Specific findings

from each experiment will be discussed in detail in the following sections.

4.1 The Dopamine D1-D2 Receptor Heteromer Modulates Basal and Cocaine Reward

Repeated stimulation of the D1-D2 heteromer by SKF 83959 during the conditioning

sessions produced CPA, indicating that the D1-D2 heteromer stimulation induced an aversion.

Moreover, the SKF 83959-induced CPA was abolished by TAT-D1 peptide-mediated D1-D2

heteromer inactivation, thus validating that the CPA was a D1-D2 heteromer specific effect. In

striking contrast, repeated inactivation of the D1-D2 heteromer by the TAT-D1 peptide during

the conditioning sessions resulted in CPP, indicating that the D1-D2 heteromer is tonically

active and its inhibition was rewarding. Having established that D1-D2 heteromer stimulation

and inactivation were able to bidirectionally modulate basal reward function, we next examined

the role of the D1-D2 heteromer in cocaine reward. Interestingly, stimulation of the D1-D2

heteromer simultaneously with cocaine conditioning abolished the acquisition of cocaine-

induced CPP. Moreover, acute stimulation of the D1-D2 heteromer on the test day eliminated

the expression of cocaine-induced CPP. Taken together, these data suggest that the D1-D2

heteromer possibly exerts a tonic suppression of the brain reward system. Thus, D1-D2

heteromer stimulation augments such suppression to induce a state of aversion, whereas D1-D2

heteromer inactivation removes such suppression to produce a net rewarding effect.

Furthermore, D1-D2 heteromer stimulation not only abolishes cocaine reward per se, but can

also abrogate the associative reward of a reinforcing environmental context.

80

Numerous studied have shown the critical involvement of the D1R or the D2R, as well

as D1R- or D2R-expressing MSNs, in the NAc in the modulation of cocaine CPP. For instance,

systemic or intra-NAc selective antagonism of the D1R by SCH 23390 abolished the

acquisition of cocaine CPP (Cervo and Samanin, 1995; Baker et al., 1996). In contrast,

selective D2R antagonism in the NAc failed to influence the acquisition or expression of

cocaine CPP (Cervo and Samanin, 1995; Shippenberg et al., 1996), although mice lacking pre-

synaptic D2R autoreceptors selectively were shown to exhibit enhanced expression of cocaine

CPP (Bello et al., 2011), potentially due to the lack of presynaptic inhibition of dopamine

release that in turn increased the stimulation of postsynaptic D1R. On the other hand,

optogenetic stimulation of D1R- or D2R-expressing MSNs were shown to respectively enhance

or attenuate the expression of cocaine CPP (Lobo and Nestler, 2011). Similarly, inhibition of

the D1R-expressing MSNs by tetanus toxin attenuated cocaine CPP expression whereas the

abrogation of D2R-expressing MSNs had no such effect (Hikida et al., 2010). Collectively,

these studies clearly demonstrated that cocaine CPP acquisition is critically dependent on D1R

stimulation in the NAc, and its expression is differentially modulated by D1R- and D2R-

expressing MSNs.

Nevertheless, despite the classical understanding that the expression of the D1R and the

D2R in neurons is largely segregated in the NAc (Gerfen et al., 1990; Harrison et al., 1990; Le

Moine and Bloch, 1995; Surmeier et al., 2007), recent evidence has also identified a population

of NAc MSNs in mice and rats that co-express both the D1R and the D2R (Bertran-Gonzalez et

al., 2008, 2010; Matamales et al., 2009; Perreault et al., 2010; Gangarossa et al., 2013a), within

which the D1-D2 heteromer is expressed (Lee et al., 2004; Rashid et al., 2007; Hasbi et al.,

2009, 2014; Perreault et al., 2010, 2011). This unique set of D1R/D2R co-expressing MSNs

81

was shown to have representations in regions associated with the D1R-expressing direct and

D2R-expressing indirect pathways (Deng et al., 2006; Wang et al., 2006, 2007; Perreault et al.,

2011). Therefore, it was hypothesized that this putative third neuronal pathway within the basal

ganglia circuitry may influence the activity of both the direct and indirect pathways (Perreault

et al., 2011), which suggests these neurons may have the capabilities to regulate cocaine CPP.

Indeed, this idea fits with the current findings that showed D1-D2 heteromer stimulation and

inactivation bidirectionally modulated the acquisition of cocaine CPP. Therefore, it was

demonstrated for the first time the potential physiological relevance of the D1R/D2R co-

expressing MSNs in the NAc in the negative modulation of cocaine reward.

There are a number of mechanisms by which activation of the D1-D2 heteromer in

D1R/D2R-coexpressing MSNs may have regulated cocaine CPP. These D1R/D2R co-

expressing MSNs are also unique in that they exhibit a dual GABA/glutamate co-expressing

phenotype (Perreault et al., 2012a), potentially allowing them to modulate the neuronal activity

of their efferent targets or locally in the NAc. Indeed, D1-D2 heteromer stimulation was shown

to enhance the expression of a major GABA-producing enzyme, GAD67, and to increase the

GABA to glutamate expression ratio in the NAc (Perreault et al., 2012a). A recent study has

demonstrated that a reduction in GABA-mediated inhibition of MSNs in the NAc promoted the

expression of cocaine CPP (Maguire et al., 2014), suggesting that increased GABA activity in

the NAc, as was observed following acute D1-D2 heteromer stimulation, may function to

suppress cocaine CPP, which is consistent with the current finding.

Another possibility is that the effect of the D1-D2 heteromer on cocaine CPP was

mediated, at least in part, by D1-D2 heteromer-induced changes in BDNF expression and

release (Hasbi et al., 2009). Calcium signalling elicited by D1-D2 heteromer stimulation was

82

shown to increase the phosphorylation and activation of CaMKII, which in turn enhanced

BDNF expression in the NAc (Hasbi et al., 2009), potentially via a transcriptional mechanism

(Zheng et al., 2009). Although increased BDNF in the NAc has been associated with enhanced

cocaine CPP (Bahi et al., 2008), which would potentially occlude a role for BDNF in D1-D2

heteromer-induced inhibition of cocaine CPP, a study also demonstrated that BDNF-mediated

signalling via its receptor TrkB exerted differential effects on cocaine CPP in D1R- versus

D2R-expressing MSNs (Lobo et al., 2010b). Specifically, TrkB knockout in the D1R-

expressing MSNs enhanced cocaine CPP whereas an opposite effect was observed when TrkB

was knocked-out in the D2R-expressing MSNs. This would suggest that BDNF signalling in

the D1R- and D2R-expressing MSNs may respectively function to inhibit and promote cocaine

CPP. Therefore, it is possible that D1-D2 heteromer-induced BDNF release in the NAc may

have preferentially impacted the D1R-expressing MSNs. In addition, a negative role for

mesolimbic BDNF expression in morphine reward lends further credence to the idea that

BDNF can exert positive or negative effects on reward depending on its expression in distinct

neuroanatomical subregions of the NAc (Koo et al., 2012, 2014). Furthermore, our findings

demonstrated a significant role for Cdk5 in mediating the effects of SKF 83959 on cocaine

CPP, a protein kinase that exhibits enhanced activity in the NAc following TrkB activation and

the associated PI3K/Akt signalling (Bogush et al., 2007). Interestingly, the conditional

knockout of Cdk5 in the NAc has been shown to enhance cocaine CPP (Benavides et al., 2007),

which supports our findings showing an inhibitory role for the kinase in this behaviour.

The link between D1-D2 heteromer stimulation and Cdk5 signalling additionally

suggests the potential regulation of DARPP-32 by the D1-D2 heteromer, as Cdk5 has been

shown to phosphorylate DARPP-32 at Thr75 (Bibb et al., 1999). DARPP-32, expressed in all

83

MSNs, is an important mediator of the biochemical, electrophysiological, transcriptional, and

behavioural effects of dopamine (Nishi et al., 1997; Fienberg et al., 1998; Nestler, 1999; Yan et

al., 1999; Calabresi et al., 2000; Flores-Hernandez et al., 2000; Zachariou et al., 2006). As

cocaine enhances dopaminergic signalling in the mesolimbic circuitry by inhibiting the

dopamine reuptake transporter, studies have also established a critical role of DARPP-32 in

mediating cocaine actions (Nestler, 1999; Zachariou et al., 2002, 2006; Zhang et al., 2006).

Particularly, the expression of cocaine CPP was shown to be associated with increased

DARPP-32 phosphorylation at Thr34, a D1R/cAMP/PKA-mediated effect, together with a

reduction in Thr75 phosphorylation, in the NAc (Tropea et al., 2008). In addition, mice with a

Thr34-Ala mutation that prevented phosphorylation at this site exhibited diminished cocaine

CPP expression (Zachariou et al., 2002), whereas Cdk5 conditional knockout in the NAc,

which would indirectly reduce Thr75 phosphorylation (Bibb et al., 1999), promoted the

expression of cocaine CPP (Benavides et al., 2007). Collectively, these findings indicate a

positive role for NAc DARPP-32 Thr34 phosphorylation, and a negative role for Thr75

phosphorylation, in the modulation of cocaine CPP. Interestingly, we have observed an

increase in DARPP-32 Thr75 phosphorylation in native rat NAc following acute stimulation of

the D1-D2 heteromer by SKF 83959 (Hasbi et al., unpublished finding), which occurred via a

Cdk5-dependent mechanism. Moreover, we further determined that the D1-D2 heteromer-

mediated increase in DARPP-32 Thr75 phosphorylation occurred specifically in the D1R/D2R

co-expressing MSNs in primary cultured striatal neurons and in rat NAc, with a concomitant

reduction in the Thr34 phosphorylation in the same neuronal population (Hasbi et al.,

unpublished finding). Therefore, it is possible that D1-D2 heteromer stimulation may have

abolished the acquisition and expression of cocaine CPP by enhancing DARPP-32 Thr75

84

phosphorylation in the NAc, which in turn would prevent the increase in Thr34

phosphorylation that is associated with cocaine CPP via the inhibition of PKA (Bibb et al.,

1999; Tropea et al., 2008). Taken together, D1-D2 heteromer stimulation increased BDNF

release in the NAc (Hasbi et al., 2009), which in turn could enhance Cdk5 activity via TrkB-

mediated PI3K/Akt signalling in the same region (Bogush et al., 2007), and ultimately result in

increased DARPP-32 Thr75 phosphorylation in the NAc to prevent the acquisition and

expression of cocaine CPP.

Lastly, as the acquisition of cocaine CPP is critically dependent on the stimulation of

D1R in the NAc (Cervo and Samanin, 1995; Baker et al., 1996), the possibility that SKF 83959

may antagonize the D1R (Downes and Waddington, 1993; Cools et al., 2002; Jin et al., 2003)

would confound the interpretation of the current findings. However, we argue that SKF 83959-

mediated abolishment of cocaine CPP acquisition and expression occurred through the D1-D2

heteromer as firstly, SKF 83959-induced aversion was abolished by the highly selective

antagonist TAT-D1 and secondly, place conditioning with the selective D1R antagonist SCH

23390 alone did not produce CPA (Meririnne et al., 2001, 2005) and thirdly, D1R antagonism

by SCH 23390 administered solely on the test day does not affect the expression of cocaine

CPP (Liao et al., 1998) whereas a similar administration of SKF 83959 abolished it.

4.2 The Dopamine D1-D2 Receptor Heteromer Modulates Behavioural Sensitization to

Cocaine

In addition to its inhibitory effects on the acquisition and expression of cocaine CPP, it

was further demonstrated that the D1-D2 heteromer also exerts a negative modulatory role in

85

both the development and expression of locomotor sensitization to cocaine. More importantly,

repeated D1-D2 heteromer inactivation during the initiation phase by the TAT-D1 peptide

alone was sufficient to induce a sensitized locomotor phenotype in response to cocaine priming

after a 24-hour withdrawal, indicating that the D1-D2 heteromer exerts a tonic inhibition to the

neurobiological processes underlying the development and expression of locomotor

sensitization to cocaine.

Despite its lack of clear face validity, locomotor sensitization to cocaine has been

associated with long-term neuroadaptations in the mesolimbic system that are critically

involved in the regulation of reward and motivation (Robinson and Becker, 1986; Henry and

White, 1991; Kantor et al., 1999), and which have been proposed to enhance the incentive

salience, or “wanting”, of the drug itself, as well as drug-related stimuli such as the

environmental context associated with drug use, leading to compulsive and persistent drug

seeking behaviours that are typically seen in addicted individuals (Robinson and Berridge,

2000). Supporting this notion, the progressive augmentation in locomotor activity in response

to repeated non-contingent cocaine administration mirrors the escalation of cocaine intake as

observed in animals that underwent extended access of cocaine self-administration (Ahmed and

Koob, 1998, 1999; Berridge, 2007). As both cocaine-induced locomotion and SA are critically

dependent on D1R stimulation in the NAc, it is possible that repeated exposure to cocaine, be it

contingent or non-contingent, may result in the same neurobiological alterations in the NAc

that ultimately augments the desire for cocaine. Indeed, not only have cocaine sensitized

animals been shown to exhibit enhanced cocaine CPP and SA (Lett, 1989; Vezina, 2004), but

the neurocircuitry of cocaine-induced locomotor sensitization was also shown to largely

overlap with that of the reinstatement of cocaine SA (Steketee and Kalivas, 2011), suggesting

86

that the persistent neuroadaptations associated with cocaine locomotor sensitization may also

contribute to eventual drug relapse. Thus, the ability of D1-D2 heteromer stimulation to

prevent specifically the sensitization response to cocaine, while leaving the locomotor-

stimulating effect of cocaine intact, strongly implicates this receptor complex as a negative

regulator of the physiological processes associated with cocaine addiction.

Studies to date have demonstrated that locomotor sensitization to cocaine is a complex

behaviour involving the close interplay between various neurotransmitter systems in regions of

the mesolimbic circuitry. The initiation of locomotor sensitization cocaine is thought to depend

on enhanced VTA dopaminergic neuron excitability in the form of long-term potentiation (LTP)

(Borgland et al., 2004), which occurs as a consequence of reduced D2 auto-receptor inhibition

(White and Wang, 1984) and reduced inhibitory GABAergic control in the VTA (Liu et al.,

2005) following repeated non-contingent cocaine administration. This subsequently results in

increased dopamine release in the NAc in response to cocaine priming that acts on

supersensitized D1 receptors (Henry and White, 1991), leading to the expression of cocaine-

induced locomotor sensitization. Therefore, a potential mechanism by which the D1-D2

heteromer prevents cocaine-induced locomotor sensitization may be through its ability to

enhance the expression of GAD67 in the VTA following acute stimulation (Perreault et al.,

2012a). This in turn could enhance GABAergic control in the VTA and prevent the LTP of

VTA dopaminergic neurons that are known to be associated with the initiation of cocaine-

induced locomotor sensitization.

Similar to that observed with the CPP findings, evidence suggests a potential role for

Cdk5 in the regulation of cocaine-induced locomotor sensitization by the D1-D2 heteromer. It

has been shown that rats treated with repeated intra-NAc infusions of Cdk5 inhibitor

87

roscovitine alone during the initiation phase produced a sensitized locomotor phenotype in

response to cocaine priming to a magnitude that was comparable to those exhibited by cocaine-

sensitized animals (Taylor et al., 2007). Similarly, concomitant intra-NAc infusion of

roscovitine with cocaine during the initiation phase further augmented the subsequent

expression of locomotor sensitization to cocaine (Taylor et al., 2007). These effects of Cdk5

inhibition in the NAc on cocaine-induced locomotor sensitization remarkably mirrors the

effects of selective D1-D2 heteromer inactivation as observed in the current study. Together

with the CPP data, these findings suggest that Cdk5 in the NAc may play a crucial role in the

inhibition of the development and expression of locomotor sensitization to cocaine by the D1-

D2 heteromer.

It is noteworthy that, although locomotor sensitization to cocaine was opposingly

modulated by D1-D2 heteromer stimulation and inactivation, the acute locomotor-stimulating

effect of cocaine was enhanced by both SKF 83959 and the TAT-D1 peptide, an apparent

contradictory finding as the two pharmacological agents were shown to exert opposite actions

at the D1-D2 heteromer (Rashid et al., 2007; Hasbi et al., 2014). However, the findings also

showed that while acute SKF 83959 treatment augmented basal locomotion, this effect was not

abolished by TAT-D1 peptide treatment, indicating that the locomotor stimulation induced by

SKF 83959 was not mediated through the D1-D2 heteromer (Shen et al., 2014b). Indeed,

although PLCβ-mediated signalling in the NAc has been shown to promote basal locomotion,

such an effect was dependent selectively on the D1R but not the D2R (Medvedev et al., 2013),

thus excluding the possible involvement of the D1-D2 heteromer in regulating basal

locomotion. A possible mechanism by which SKF 83959 enhanced basal locomotion may be

through the stimulation of D5R (Perreault et al., 2012b), as mice with D5R gene-deletion

88

exhibited reduced locomotor response to a D1R/D5R agonist SKF 81297 (Holmes et al., 2001),

which suggests a locomotor-stimulating action of the D5R. On the other hand, in contrast to its

stimulating effect on cocaine-induced locomotion, basal locomotion was not affected by acute

D1-D2 heteromer inactivation (Shen et al., 2014b), and thus the potentiating effect of D1-D2

heteromer inactivation on acute cocaine-induced locomotion was likely a sensitization response

to cocaine, possibly via a Cdk5-dependent mechanism as described in the previous section.

4.3 The Dopamine D1-D2 Receptor Heteromer Modulates Cocaine-seeking Behaviours

The present finding demonstrated that the D1-D2 heteromer exerts negative modulation

on cocaine CPP and locomotor sensitization, and a similar inhibitory effect was also observed

in the cocaine SA paradigm. Specifically, D1-D2 heteromer stimulation dose-dependently

reduced total cocaine intake during an SA session as reflected by reduced total number of

active lever presses and cocaine infusions. In addition, cocaine-induced reinstatement of SA

was similarly abolished by D1-D2 heteromer stimulation, whereas D1-D2 heteromer

inactivation further enhanced this behaviour. However, the reinstatement of cocaine SA by a

discriminative light cue that was previously associated with each cocaine infusion during FR5

training was only abolished by D1-D2 heteromer stimulation when a higher dose of SKF 83959

was used (1.5mg/kg but not 0.5 mg/kg), although both doses completely prevented cocaine-

induced reinstatement.

There are two possible interpretations for the reduction in cocaine SA under the FR5

schedule of reinforcement that was observed following D1-D2 heteromer stimulation. First,

such an effect may suggest that the D1-D2 heteromer enhanced the rewarding effect of each

89

cocaine infusion so that a fewer number of total infusions were required to achieve the same

euphoric state (Richardson and Roberts, 1996). Indeed, a higher dose of cocaine was shown to

result in fewer total infusions compared to a lower dose of cocaine when used in SA under the

FR5 schedule (Depoortere et al., 1993). However, as D1-D2 heteromer stimulation reduced

cocaine reward in the CPP model, it is highly unlikely that D1-D2 heteromer stimulation

enhanced cocaine reward in this paradigm. A more likely explanation is that the observed

reduction in cocaine SA reflected a loss of motivation to acquire cocaine infusions. This

interpretation is supported by the fact that D1-D2 heteromer stimulation and inactivation

respectively prevented and enhanced priming cocaine-induced reinstatement of cocaine SA

Although stimulation of the D1-D2 heteromer by SKF 83959 at the two doses examined

both prevented priming cocaine-induced reinstatement of cocaine SA, the lower dose of SKF

83959 did not affect reinstatement of SA induced by a discriminative cue. This discrepancy

may be explained by the distinct neurocircuitry involved in priming cocaine- versus cue-

induced reinstatement of SA (Shaham et al., 2003). Thus, whereas cocaine priming-induced

reinstatement was critically dependent on D1R and D2R receptor stimulation in the NAc shell

(Schmidt et al., 2006), reinstatement by discriminative cues was shown to require D1R

stimulation in the BLA but not the NAc (Ciccocioppo et al., 2001). Therefore, since it has been

shown that the expression of the D1-D2 heteromer is most prominent in the NAc shell

(Perreault et al., 2010), it is possible that D1-D2 heteromer stimulation by a lower dose of SKF

83959 will affect a behaviour that is more dependent on dopaminergic activity in the NAc shell

rather than the BLA. Furthermore, no evidence yet exists that indicate D1-D2 heteromer

expression is present in the BLA or that D1R/D2R neuron projections to the BLA exists, and

thus we cannot predict the impact of this region on our observed effects.

90

The D1-D2 heteromer may have attenuated cocaine SA through several mechanisms.

One mechanism may occur through the increase in GAD67 elicited by D1-D2 heteromer

stimulation and enhanced GABA/glutamate ratio in NAc (Perreault et al., 2012a), since

GABAergic activity in the NAc has been implicated in the attenuation of cocaine SA. The

administration of GABAB receptor agonists into NAc and VTA were shown to attenuate the

maintenance of cocaine SA under both the FR and PR schedules (Shoaib et al., 1998; Brebner

et al., 2000; Di Ciano and Everitt, 2003), as well as cocaine- and context-induced reinstatement

(Fuchs et al., 2008; Peng et al., 2008). Moreover, rats that underwent extinction of cocaine SA

were shown to have increased basal GABA levels in the NAc (Wydra et al., 2013). Together

these findings suggest that increased GABAergic activity in the NAc attenuates cocaine-

seeking behaviours, and may be an underlying neurobiological alteration that is associated with

the extinction of cocaine-taking. Therefore, the D1-D2 heteromer may have attenuated and

prevented the maintenance and reinstatement of cocaine SA respectively by enhancing

GABAergic activity in the NAc.

Another mechanism may be through the ability of the D1-D2 heteromer to regulate

DARPP-32 phosphorylation in NAc, which may also allow it to exert an inhibitory modulation

on the maintenance of cocaine SA. The establishment of stable cocaine SA was associated with

increased DARPP-32 phosphorylation at Thr34 in the NAc (Lynch et al., 2007), a site that is

phosphorylated by D1R/cAMP/PKA signalling (Hemmings et al. 1984). In addition, mice with

an inactive Thr34 phosphorylation site required a longer training period to achieve stable

cocaine SA (Zhang et al., 2006). Therefore, similar to its effects on other cocaine-mediated

behaviours (Scheggi et al., 2007; Tropea et al., 2008), DARPP-32 Thr34 phosphorylation in the

NAc seems to be an important mediator for the maintenance of cocaine SA. In contrast,

91

DARPP-32 phosphorylation at Thr75, an effect mediated by Cdk5, turns it into a potent

inhibitor of PKA, and thus the phosphorylation at this site may attenuate cocaine SA by

reducing DARPP-32 Thr34 phosphoryation. Since intra-NAc infusion of a Cdk5 inhibitor

enhanced cocaine SA under the PR schedule (Taylor et al., 2007), a negative modulatory role

for DARPP-32 Thr75 phosphorylation in cocaine SA has been demonstrated. Therefore, since

D1-D2 heteromer stimulation resulted in an increase in DARPP-32 Thr75 phosphorylation in

the NAc via a Cdk5-dependent mechanism (unpublished finding), the D1-D2 heteromer may

therefore have exerted its inhibitory modulation on the maintenance of cocaine SA by reducing

Thr34 phosphorylation in the NAc.

In addition to the mesolimbic circuit, the corticostriatal circuit (PFC-NAc) has also

been implicated in the reinstatement of cocaine SA. The PFC is anatomically divided into

various subregions, including the infralimbic cortex (IFC), prelimbic cortex (PLC), and the

orbitofrontal cortex (OFC), and each was shown to play a role in the modulation of the

reinstatement of SA behaviour induced by cocaine priming (Capriles et al., 2003; Fuchs et al.,

2004; Peters et al., 2008; Martín-García et al., 2014; Shen et al., 2014a) or cocaine-associated

cues (McLaughlin and See, 2003; Fuchs et al., 2004; Lasseter et al., 2009; LaLumiere et al.,

2012). Interestingly, acute D1-D2 heteromer inactivation by the TAT-D1 peptide was shown to

enhance neuronal activity in the IFC, PLC, and OFC as indicated by increased c-Fos

expression in these regions (Perreault et al., 2015), providing yet another potential mechanism

through which the D1-D2 heteromer could regulate cue- and cocaine-induced reinstatement of

SA behaviour. However, the use of c-Fos as the marker for neuronal activity cannot

differentiate which subpopulation of neurons in these prefrontal subregions were activated by

92

D1-D2 heteromer inactivation, and thus we cannot surmise at this point exactly how this may

contribute to the observed effects on the reinstatement of cocaine SA.

4.4 Potential Limitations

A potential limitation of the current study is the use of SKF 83959 to examine the

effects of D1-D2 heteromer stimulation on cocaine-seeking behaviours. To date, no selective

agonist for the D1-D2 heteromer has been identified, and while SKF 83959 has been shown to

definitively induce the Gq-PLC-linked calcium signalling mediated by the D1-D2 heteromer,

its pharmacology has been shown to extend to include a few other receptors including the

dopamine D5 receptor and the 5-HT2C serotonin receptor (Sahu et al., 2009; Perreault et al.,

2012b; Chun et al., 2013; Guo et al., 2013). Since the involvement of both the D5R and the 5-

HT2CR in the modulation of cocaine-seeking behaviours have been previously implicated (Filip

et al., 2000; Rocha et al., 2002; Fletcher et al., 2004; Perreault et al., 2012b), it is possible that

systemic injection of SKF 83959 may have elicited off-target effects that may also contribute

to its inhibitory modulation on cocaine-seeking behaviours as observed in the current study.

For instance, a D5R-mediated increase in BDNF expression in the mPFC has been shown and

5-HT2CR stimulation in the VTA was shown to attenuate cocaine SA (Fletcher et al., 2004;

Berglind et al., 2007; Perreault et al., 2012b). To overcome this specificity issue, we

synthesized a highly selective TAT-D1 peptide that occludes the interaction site between the

two receptors (O'Dowd et al., 2012), thus inhibiting D1-D2 heteromer expression and function

and abolishing the physiological effects of D1-D2 heteromer activation by SKF 83959 without

affecting other receptor oligomers such as D1-D1 homomers, D2-D5 heteromers, or other

93

heteromers (Hasbi et al., 2014). The use of the TAT-D1 peptide allowed us to conclusively

define the effects of D1-D2 heteromer inactivation on cocaine-seeking behaviours. However, it

is highly noteworthy that we did not directly show the abolition of SKF 83959-induced

regulation of cocaine effects by TAT-D1 in many instances as we were concerned about the

confounding effects of the repeated administration of more than two agents simultaneously to

overwhelm the in vivo system, as both SKF 83959 and cocaine affect various neurotransmitter

systems via multiple mechanisms of action (Rothman et al., 2001; Perreault et al., 2012b; Chun

et al., 2013; Guo et al., 2013). Nonetheless, the ability of SKF 83959 and the highly specific

TAT-D1 peptide to consistently exert opposite effects in all three behavioural models

examined argues for a pivotal role for the D1-D2 heteromer in mediating cocaine-seeking

behaviours.

4.5 Significance and Conclusion

In summary, I have demonstrated that D1-D2 heteromer stimulation and inactivation

consistently exerted bidirectional modulation of cocaine-induced behaviours in the three most

commonly used animal experimental paradigms that model aspects of cocaine addiction.

Specifically, D1-D2 heteromer stimulation by the agonist SKF 83959 abolished the acquisition

and expression of cocaine CPP, prevented the development and expression of cocaine-induced

locomotor sensitization, attenuated the maintenance of cocaine SA, and blocked priming

cocaine- and cue-induced reinstatement of SA. In contrast, selective D1-D2 heteromer

inactivation increased the acquisition of cocaine CPP, augmented cocaine-induced locomotor

sensitization, and enhanced priming cocaine-induced reinstatement of SA. Numerous studies to

94

date have clearly implicated the critical involvement of D1R and D2R, as well as the D1R- and

D2R-expressing MSNs, in the NAc in cocaine-induced behaviours (Thomas et al., 2008; Baik,

2013), and the present findings show for the first time a highly significant role for the

dopamine D1-D2 receptor heteromer and the unique subset of D1R/D2R co-expressing MSNs

in the NAc may also play an important role in the modulation of cocaine-induced behaviours.

Indeed, the findings of the current study indicate that the D1-D2 heteromer exerts tonic

inhibition on the neurobiological processes underlying cocaine-induced behaviours, potentially

via its signalling through Cdk5 and DARPP-32 in the NAc. Thus, the stimulation and

inactivation of this dopamine receptor complex will respectively enhance and remove such

inhibitory modulation to reduce and promote cocaine-induced behaviours.

Despite constant research efforts, no effective therapeutic intervention for cocaine

addiction has yet emerged, and a novel therapeutic target is thus warranted. Therefore, the

identification of the dopamine D1-D2 receptor heteromer as a specific negative modulator of

cocaine-induced behaviours may have great implications for the understanding and treatment

of cocaine addiction. Furthermore, given the established involvement of the D1-D2 heteromer

in amphetamine-induced locomotor sensitization, it is possible that these therapeutic effects

may be translated across other psychoactive substances.

95

3. Rapid anti-depressant and anxiolytic actions following dopamine D1-D2 receptor

heteromer inactivation

Abstract

A role for the mesolimbic dopaminergic system in the pathophysiology of depression

has become increasingly evident. Specifically, brain-derived neurotrophic factor (BDNF) has

been shown to be elevated in the nucleus accumbens of depressed patients and to positively

contribute to depression-like behaviour in rodents. The dopamine D1-D2 receptor heteromer

exhibits significant expression in NAc and has also been shown to enhance BDNF expression

and signalling in this region. We therefore examined the effects of D1-D2 heteromer

stimulation in rats by SKF 83959, or its inactivation by a selective heteromer-disrupting TAT-

D1 peptide on depression- and anxiety-like behaviours in non-stressed animals and in animals

exposed to chronic unpredictable stress. SKF 83959 treatment significantly enhanced the

latency to immobility in the forced swim test, increased the latency to drink condensed milk

and reduced total milk consumption in the novelty-induced hypophagia test, and additionally

reduced the total time spent in the open arms in the elevated plus maze test. These pro-

depressant and anxiogenic effects of SKF 83959 were consistently abolished or attenuated by

TAT-D1 peptide pre-treatment, signifying the behaviours were mediated by the D1-D2

heteromer. More importantly, in animals exposed to chronic unpredictable stress (CUS), TAT-

D1 peptide treatment alone induced significant and rapid anxiolytic and antidepressant-like

effects in two tests for CUS-induced anhedonia-like reactivity and in the novelty-suppressed

feeding test. Together these findings indicate a positive role for the D1-D2 heteromer in

mediating depression- and anxiety-like behaviours and suggest its possible value as a novel

therapeutic target.

The work on the chronic unpredictable stress in this study was done by Dr. Fancis

Bambico.

96

1. Introduction

Major depression is a chronic psychiatric disorder that is characterized by low mood,

despair, and the inability to experience pleasure from normally rewarding stimuli that severely

reduce the quality of life of patients. Roughly half of the affected individuals do not respond to

available antidepressants (Culpepper, 2010) and reports indicate that greater than 50 percent of

depressed patients also suffer from anxiety (Carter et al., 2001), which can result in increased

symptom severity and delayed recovery (Hirschfeld, 2001). Interestingly, since virtually all

antidepressants are also anxiolytic, it has been suggested that the neurobiology of anxiety and

depression may share certain common molecular substrates (Conway et al., 2006).

The hippocampus and the frontal cortex have been the most widely studied brain

regions in relation to depression, however it is unlikely that these regions account for all the

symptoms of the disorder (Russo and Nestler, 2013). The mesolimbic dopamine system for

example, comprised of the dopaminergic projection from the ventral tegmental area (VTA) to

the nucleus accumbens (NAc), plays a critical role in reward and has been most widely studied

for its role in addiction. Studies are now demonstrating, however, a major role for the

mesolimbic pathway in the regulation of depression (Nestler and Carlezon, 2006b; Krishnan et

al., 2007; Russo and Nestler, 2013; Tye et al., 2013; Friedman et al., 2014). The activity of the

NAc is reduced in depressed patients (Mayberg et al., 2000) and animal studies suggest that

this may result from increased GABAergic activity in the region leading to heightened

inhibitory tone on dopamine neurons, which could manifest as a loss of pleasurable feelings

and the development of depression-like behaviours with anhedonia (Nestler and Carlezon,

2006b; Tye et al., 2013). One key player in the mediation of depression-like responses is brain-

97

derived neurotrophic factor (BDNF) (Nestler and Carlezon, 2006b; Russo and Nestler, 2013).

Interestingly in the NAc, many of the proteins identified to be involved in depression are

downstream of BDNF signaling. These findings suggest that the mechanisms which regulate

the expression of BDNF in the NAc may represent novel therapeutic targets in depression. The

effects of BDNF appear to be region-specific, being pro-depressant in the mesolimbic system

but anti-depressant in hippocampus and frontal cortex (reviewed in: Duman, 2014).

The dopamine D1-D2 receptor heteromer is a dopamine receptor complex that is

predominantly expressed in the NAc (Perreault et al., 2010), and which has been shown to

increase the expression of a major GABA producing enzyme GAD67 in this region, thus

potentially enhancing inhibitory GABAergic tone of neurons in the NAc (Perreault et al.,

2012a). In addition, D1-D2 heteromer stimulation produced a Gq-mediated, phospholipase C

(PLC)-dependent calcium signal (Lee et al., 2004; Rashid et al., 2007; Hasbi et al., 2009, 2014;

Verma et al., 2010) that was shown to contribute to increased expression of BDNF and its

signaling in the NAc through its receptor tropomyosin receptor kinase B (TrkB) (Hasbi et al.,

2009; Perreault et al., 2012b, 2013). Given these findings, we postulated that the D1-D2

heteromer could positively contribute to the pathogenesis of depression, a notion supported by

recent studies that demonstrated the ability of D1-D2 heteromer disrupting peptides to exert

antidepressant-like activity in rats (Pei et al., 2010; Hasbi et al., 2014).

The physiological role of D1-D2 heteromer stimulation in mediating depression- and

anxiety-like behaviours in rodents can be delineated through the use of the agonist SKF 83959

(Rashid et al. 2007). However, recent reports have indicated that the pharmacology of SKF

98

83959 also involves several other receptors, such as the dopamine D5 receptor, in addition to

the D1-D2 heteromer (Sahu et al., 2009; Perreault et al., 2012b; Chun et al., 2013). Thus, we

sought to validate whether the pro-depressive effects of SKF 83959 can be attributed to the D1-

D2 heteromer using the recently developed TAT-D1 disrupting peptide (Hasbi et al., 2014).

The highly selective TAT-D1 peptide, which occludes the interaction site between the two

receptors (O’Dowd et al., 2012), disrupts D1-D2 heteromer formation and abolishes the

physiological effects of D1-D2 heteromer activation and signalling without affecting other

receptor oligomers such as D1-D1 homomers or D2-D5 heteromers (Hasbi et al., 2014).

Several lines of evidence have supported the functional existence of the D1-D2

heteromer in the brain despite the concerns raised by a recent report (Frederick et al., 2015).

We and other groups have consistently observed the co-expression of the D1R and the D2R in

a unique subset of medium spiny neurons (MSNs) in rodent striatum (Matamales et al., 2009;

Bertran-Gonzalez et al., 2010; Perreault et al., 2010; Gangarossa et al., 2013b), and that the two

receptors physically interact in the rat NAc and human striatal tissues as evidenced by in situ

confocal fluorescent resonance energy transfer (FRET) and FRET photobleaching (Hasbi et al.,

2009; Perreault et al., 2010), co-immunoprecipitation (Lee et al., 2004; Pei et al., 2010),

glutathione-S-transferase (GST) pulldown (Pei et al., 2010), and proximity ligation assays (R.

Franco, personal communication). Moreover, the calcium signalling associated with the D1-

D2 heteromer was shown to require the co-activation of both the D1R and the D2R in the

striatum (Lee et al., 2004; Rashid et al., 2007; Hasbi et al., 2014), and was dose-dependently

reduced by the TAT-D1 peptide in primary cultured striatal neurons (Hasbi et al., 2014).

Collectively, these findings clearly supported the functional existence of the D1-D2 heteromer

99

in the brain, and here we propose to further characterize the physiological role for the receptor

complex in depression and anxiety. Using a number of behavioural paradigms we were able to

demonstrate in rats that D1-D2 heteromer activation by SKF 83959 promoted both depression-

and anxiety-like behaviours. The pro-depressive and anxiogenic effects of SKF 83959 were

attenuated, or in some cases reversed, by TAT-D1 peptide pre-treatment, thereby confirming

the involvement of the D1-D2 heteromer in mediating the behavioural effects of SKF 83959. In

addition, acute TAT-D1 peptide administration also induced significant and rapid anxiolytic

and antidepressant-like properties in animals exposed to chronic unpredictable stress (CUS),

which models chronic depressive-like states and has been shown to better predict the

therapeutic onset of antidepressants (Willner et al., 1987a).

2. Materials and Methods

2.1 Animals

Adult male Sprague-Dawley and Fischer 344 rats weighing 300-400g at the beginning

of the experiments (Charles River, Canada) were used. The mesolimbic dopamine system of

Fischer 344 rats are highly responsive to CUS, which were thus used in the CUS protocol

(Ortiz et al., 1996). Animals were housed in polyethylene cages in a temperature-controlled

colony room, maintained on a 12-h light-dark cycle (lights on at 0700), with ad libitum access

to food and water. Animals were handled daily for 3 days before the start of the experiments.

All treatments were performed during the light phase of the day-night cycle. Animals were

housed and tested in compliance with the guidelines described in the Guide to the Care and the

Use of Experimental Animals (Canadian Council on Animal Care), and protocols were

approved by the Animal Care Ethics Committee of the University of Toronto.

100

2.2 Drugs

SKF 83959 hydrobromide (Tocris Bioscience) was dissolved in physiological saline

containing 5% DMSO, and was administered subcutaneously (s.c.). All systemic injections

were given at a volume of 1.0 mL/kg and administered 5 minutes prior to behavioural testing.

The TAT-D1 disrupting peptide or an inactive TAT-scrambled peptide control (TAT-Sc) was

dissolved in physiological saline containing a protease inhibitor cocktail (1:1000) and

administered into the intracerebroventricular space (300 pmol/4µL, i.c.v.) 15 minutes prior to

SKF 83959 injection. The amino acid sequence of the TAT-D1 peptide derived from the D1R

carboxyl-tail is a unique sequence not appearing in any other protein (BLAST search). For non-

drug injections, an equivalent volume of saline or vehicle was administered.

2.3 Surgery

Rats were anaesthetized with isoflurane (5%), administered analgesic ketoprofen

(5mg/kg) and secured in a stereotaxic frame. A cannula (22-gauge, Plastics One) was placed

unilaterally into the intracerebroventricular space close to the midline according to the

following stereotaxic anterior-posterior (AP), mediolateral (ML), and dorsoventral (DV)

coordinates: AP -0.8mm, ML + 1.3mm, DV – 3.7mm, and was secured by dental cement

anchored with four stainless steel screws (Plastics One) fixed on the dorsal surface of the skull.

AP and ML coordinates were taken from bregma, DV coordinate from the dura (Paxinos and

Watson, 1998). The animals were allowed to recover in their home cage for five days before

the experiments were initiated. Cannulae placement was visually verified in post-mortem brain

slices.

101

2.4 Apparatus and Procedures

All animals received a minimum of five days of handling (2 minutes/animal/day) prior

to behavioural testing. The dose of the TAT-D1 peptide (300pmol, i.c.v.) was previously

shown to disrupt D1-D2 heteromer formation in vivo as indicated by the loss of co-

immunoprecipitation of D1R with D2R in rat NAc tissue (Hasbi et al., 2014).

2.4.1 Forced Swim Test

The forced swim test (FST) was conducted as previously described (Hasbi et al., 2014)

in a non-colony room isolated from external noise. During the pre-test, animals were placed in

a glass container with room temperature water filled to a height of approximately 40 cm. The

animals remained in the water for 15 minutes after which they were towel-dried and placed in a

cage under a heat lamp until completely dry. 24 hours following the pre-test, animals were

administered their designated treatment, TAT-D1 peptide (0, 300 pmol, i.c.v; given 15 minutes

prior to SKF 83959 treatment) and SKF 83959 (0 or 2.5 mg/kg, s.c), and placed again in the

water-filled container for 5 minutes. The following parameters were measured at 5-second

intervals: immobility (passively floating in water), swimming (moving across the water

surface), or climbing (front paws touching the glass wall while attempting to jump out of

water). Behavioural testing took place 5 minutes following SKF 83959 injection. To avoid

experimenter bias, behavioural measurements were done by an individual who was blind to the

treatment assigned to each animal.

2.4.2 Elevated Plus Maze

102

Behavioural testing was conducted as previously described (Rogóż and Skuza, 2011) in

an elevated plus maze (EPM) (Harvard Apparatus) situated in a non-colony room isolated from

external noise. The EPM was constructed of black plexiglass and consisted of a central square

with two sets of opposing open and closed arms each with dimensions 50cm x 10cm. Closed

arms were enclosed by 40cm high black plexiglass walls along the longitudinal edges, with the

roof and ends open. The entire maze was suspended 50cm off the ground. Following the

assigned drug treatment, TAT-D1 (0, 300 pmol, i.c.v; given 15 minutes prior to SKF 83959

treatment) and SKF 83959 (0, 0.5, 2.5 mg/kg, s.c.), rats were placed in the center of the maze

and behaviour was recorded for 5 minutes. Behavioural scoring of the videos occurred after the

testing was complete and the following parameters were measured: the number of open arm

and closed arm entries, time spent in the open and closed arms, and time spent per entry in

open and closed arms. Entry to or exit from an arm was defined by both front paws crossing the

arm boundaries. Behavioural testing took place 5 minutes following SKF 83959 injection. To

avoid experimenter bias, behavioural measurements were done by an individual who was blind

to the treatment assigned to each animal.

2.4.3 Novelty-induced Hypophagia

Behavioural testing was conducted as previously described (Dulawa and Hen, 2005).

Animals were trained to consume sweetened condensed milk (1:3 ; milk:water) in the home

cage under dim light for 2 hours daily over 3 days to establish stable preference to milk over

water. Milk and water were presented in plastic bottles closed with rubber stoppers and

positioned through meshed cage lids. Upon the establishment of stable milk preference (day 4),

the latency to drink from milk bottle and total milk consumed in the training environment was

103

measured. The few rats that failed to establish milk preference (~2-3%) were eliminated from

the test. On the test day (day 5), after receiving the assigned drug treatment, TAT-D1 (0, 300

pmol, i.c.v; given 15 minutes prior to SKF 83959 treatment) and SKF 83959 (0, 0.5 mg/kg,

s.c.), rats were placed in novel cages without bedding and under bright light to induce anxiety.

The latency to drink from the milk bottle and total milk consumed over the 2 hour period in the

novel environment was measured. Behavioural testing began 5 minutes following SKF 83959

injection.

2.4.4 Chronic Unpredictable Stress (CUS)

The CUS protocol employed was modified after Bambico et al. (2009) and Willner et al.

(1987). All animals were handled during one week of acclimation. They were then exposed to

sucrose solution (1% w/v) for 3 days, and then allowed to discriminate between the sucrose

solution and water (sucrose preference test) for 1 hour every day (at 10:00) after overnight

water deprivation for 16-18 hours. The position of the bottles on the cage top cover was

switched midway to minimize a possible egocentric orientation bias. Sucrose preference was

calculated as the ratio between sucrose solution intake and total fluid intake. During this task,

the animals were individually caged, tested within the same housing room under normal

lighting condition. After 4 days of initial sucrose preference tests, the last sucrose preference

score was used to assign animals to 4 groups such that no significant between-groups

difference was obtained. Two CUS groups were exposed to stress for 5 weeks. Two control

groups were not exposed to any of the stressors and were housed in a different vivarium.

Sucrose preference testing was performed for all animals after each week. Overtly stable

anhedonia-like reduction of sucrose preference scores in CUS animals typically ensues after a

104

few weeks of exposure. The CUS regimen consisted of various non-debilitating, inescapable

and uncontrollable physical, psychological and circadian stressors with the following elements:

modifications of the light/dark cycle and of the lighting characteristics (light-dark cycle

reversal, intermittent on-off lighting and stroboscopic lights), in food and water availability

(12-18 hours of deprivation), housing conditions (isolation, damp bedding, novel room

environment, cage tilt), and various ecological challenges (restraint, cold room, high frequency

sound or static noise, predator odor). Stress schedules across 5 weeks were not identical.

Within each one week block, stressors could occur at any time of the day (or night), and were

each applied for a period of 15 min to 18 hours. Sequence was at random and combinations

varied, in order to be completely unpredictable to the animal. However, more severe stressors

were not combined. The combinatory and cumulative effect of these stressors but not any one

of the individual stressors is sufficient and necessary to produce depressive-like reactivity.

During the post-CUS test day, animals were administered with TAT-D1 or TAT-Sc

(300pmol, i.c.v.) and were subjected to two tests for anhedonia (Sucrose Preference Test and

Fruit Loops Test) and two tests for anxiety-related reactivity (Novelty-Suppressed Feeding test

and Elevated Plus Maze) in the following order: Sucrose Preference Test, Novelty-Suppressed

Feeding, Elevated Plus Maze, and Fruit Loops Test. Procedures for the Sucrose Preference Test

and Elevated Plus Maze were as described earlier.

2.4.5 Fruit Loop Test

5 pieces of fruit loops (sweetened fruit-flavored cereal) were placed at the center of the

home cage. 15 minutes following the treatment with the TAT-D1 or the TAT-Sc peptide

105

(300pmol, i.c.v.), the animals were placed back into the home cage, and the latency to and total

fruit loop consumption over 10 minutes were recorded. Animals were pre-exposed to this

procedure for 5 days before the CUS exposure and before the TAT-D1 or TAT-Sc treatment to

eliminate a possible effect of novelty.

2.4.6 Novelty Suppressed Feeding Test

Animals were food-deprived for 36 hours, then placed in a novel environment (80x80x15

cm Plexiglas chamber) where the latency to approach and feed from 12 food pellets placed at

the center was measured. Once the animals approached the pellets, they were immediately

removed and placed in a familiar environment (Home Cage), where the latency to approach

and feed from 12 food pellets placed at the center was again measured. The cut-off time was

600 seconds. Novelty-induced suppression of feeding typically occurs in the novel but not in

the familiar environment, indicating an anxious-depressive behavioural phenotype that is

induced by the exposure to a novel environment. This delay in feeding has been shown to be

reversed by chronic but not acute treatment with standard antidepressants (Bodnoff et al.,

1989).

2.5 Statistical Analyses

Values are reported as mean ± s.e.m. All data was analyzed for normality prior to

ANOVAs using the Shapiro-Wilk test. Means were analyzed using the appropriate ANOVA

with Treatment as the between subjects factors for the forced swim test, elevated plus maze,

and novelty-induced hypophagia tests. Additionally for the novelty-induced hypophagia test,

data comparisons between the home cage and novel cage were performed by Student’s t test.

106

Sucrose preference scores during CUS exposure were analyzed with Stress Exposure as the

between subject factor and Weeks as the within subject factor. Post-CUS behavioural tests

were analyzed with Stress Exposure and Treatment as between subject factors. Duncan’s

multiple range test was used for post-hoc comparisons following appropriate ANOVAs.

Statistical significance was set at P<0.05 and computations were performed using the

SPSS/PC+ statistical package.

3. Results

3.1 Forced Swim Test

We first examined the effects SKF 83959 on passive coping or behavioural despair in

rats in the forced swim test, and determined whether SKF 83959-induced behaviour could be

reversed by the TAT-D1 peptide. The ANOVA revealed a significant main effect of Treatment

on the latency to {F(3,48)=13.5, p<0.0001} and total duration of {F(3,48)=26.4, p<0.0001}

immobility. Post-hoc analysis showed that SKF 83959 administration significantly reduced the

latency to develop immobility compared to the vehicle control, an effect that was abolished by

TAT-D1 peptide pretreatment (Figure II-1A, p<0.001). TAT-D1 peptide treatment alone did

not affect the latency to develop immobility. SKF 83959 did not alter the total immobility time

over the 5 minute testing period, whereas TAT-D1 administration alone or with SKF 83959

produced an antidepressant-like activity as indicated by significantly reduced total immobility

time (Figure II-1B, p<0.001). Time course analysis of the effects of SKF 83959 and TAT-D1

peptide on total immobility time during each minute of testing (Figure II-1C) revealed

significant effects of Time {F(4,120)=239.1, p<0.0001}, Treatment {F(3,43)=40.4, p<0.0001}

107

Figure II-1: The effects of D1-D2 heteromer stimulation or inactivation on the latency to

and total immobility time in the forced swim test in rats. A) SKF 83959 significantly

reduced the latency to immobility, and effect that was reversed by TAT-D1 peptide pre-

treatment. B) The total immobility was not affected by SKF 83959, but was significantly

reduced by TAT-D1 peptide alone or with SKF 83959. C&D) The time course of the total

immobility for each treatment over the 5 minute testing period. Data represent means ± s.e.m.

of n=10-14 rats/group. (***p<0.001, compared to vehicle-treated rats).

0

50

100

150

200

La

ten

cy

(se

c)

***

Veh

icle

SKF

8395

9

TAT-D

1

TAT-D

1+SKF

0

50

100

150

200

250

Imm

ob

ilit

y (

se

c)

***

Veh

icle

SKF

8395

9

TAT-D

1

TAT-D

1+SKF

***

-10

0

10

20

30

40

VehicleSKF 83959

TAT-D1TAT-D1+SKF

0 15 30 45 60

***

Time (sec)

Imm

ob

ilit

y (

se

c)

10

20

30

40

50

60

VehicleSKF 83959

TAT-D1TAT-D1+SKF

1 2 3 4 5

***

Time (min)

Imm

ob

ilit

y (

se

c)

A) B)

C) D)

108

and a Time x Treatment interaction {F(12,120)=15.6, p<0.0001}. SKF 83959 induced a

significant increase in immobility within the first minute of testing compared to controls

(Figure II-1D, p<0.001), an effect that was abolished by TAT-D1 pre-treatment.

3.2 Elevated Plus Maze

To examine the role of the D1-D2 heteromer in mediating anxiety, the effects of SKF

83959 and the TAT-D1 peptide were assessed in rats in the elevated plus maze. One-way

ANOVA revealed significant effects of Drug Treatment on the total time spent in the open

arms of the maze {Treatment: F(4,42)=4.6, p<0.01}, total time spent in the closed arms

{Treatment: F(4,42)=7.0, p<0.001}, total open arm entries {Treatment: F(4,42)=4.1, p<0.01},

and time spent in open arms per entry {Treatment: F(4,42)=16.8, p<0.001}. SKF 83959

administration resulted in a dose-dependent reduction in the total time spent in the open arms

that became significant at the 2.5 mg/kg dose, an effect that was abolished by TAT-D1 peptide

pre-treatment (Figure II-2A, p<0.05). TAT-D1 alone, or in combination with SKF 83959,

resulted in a small and insignificant increase in the total time spent in open arms (Figure II-2A).

Accordingly, the total time spent in the closed arms was significantly higher in animals treated

with 2.5mg/kg of SKF 83959 (Figure II-2B, p<0.05). In addition, the number of total open arm

entries was significantly reduced by SKF 83959 at the 2.5mg/kg dose (p<0.01), as well as by

TAT-D1 treatment alone (p<0.05) or with SKF 83959 (p<0.01) (Figure II-2C). Combined with

the total time spent in the open arm, the reduced number of total open arm entries for TAT-D1

peptide-treated animals reflected the significantly more time spent in the open arms with each

entry compared to the vehicle control, (p<0.01, Figure II-2D) indicating that D1-D2 heteromer

disruption is anxiolytic in the elevated plus maze. In contrast, the time spent in the open arm

109

Figure II-2: D1-D2 heteromer stimulation or inactivation had bidirectional effects on

anxiety-like behaviours in the elevated plus maze. A) SKF 83959 dose-dependently reduced

the total time spent in the open arms and B) increased the total time spent in the closed arms C)

SKF 83959 at 2.5mg/kg or the TAT-D1 peptide treatment alone significantly reduced total

open arm entries. D) SKF 83959 at 2.5mg/kg significantly reduced the time spent in open arms

per entry, whereas TAT-D1 peptide treatment alone or with SKF 83959 had the opposite effect.

Data represent means ± s.e.m. of n=8-12 rats/group. (*p<0.05, **p<0.01, ***p<0.001,

compared to vehicle-treated rats)

0

20

40

60

To

tal

Tim

e S

pe

nt

in

Op

en

Arm

(se

c)

*

Veh

icle

0.5_

SKF

2.5_

SKF

TAT-D

1

TAT-D

1+SKF

0

2

4

6

8

10

To

tal

Op

en

Arm

En

trie

s

** ***

Veh

icle

0.5_

SKF

2.5_

SKF

TAT-D

1

TAT-D

1+SKF

0

100

200

300

To

tal

Tim

e S

pe

nt

in

Clo

se

d A

rms (

se

c)

*

Veh

icle

0.5_

SKF

2.5_

SKF

TAT-D

1

TAT-D

1+SKF

0

5

10

15

Tim

e S

pe

nt

in O

pe

n

Arm

s P

er

En

try

(se

c)

Veh

icle

0.5_

SKF

2.5_

SKF

TAT-D

1

TAT-D

1+SKF

**

***

*

A) B)

C) D)

110

with each entry was significantly reduced by SKF 83959 treatment at the 2.5mg/kg dose, an

effect that was abolished by the pre-treatment with TAT-D1 (Figure II-2D, p<0.05).

3.3 Novelty-induced Hypophagia

To further assess the effects of the D1-D2 heteromer on anxiety responses we next

utilized the novelty-induced hypophagia paradigm, a behavioural model that examines anxiety

induced by the stress of a novel environment, and which assesses an animal’s willingness to

approach and consume food or drink in a novel environment. There were significant effects of

Drug Treatment in the total sweetened milk consumed {Treatment: F(4,42)=7.612, p<0.001}

and the latency to drink milk {Treatment: F(4,42)=672.219, p<0.001}. As expected, the

anxiogenic effect of the novel testing environment prolonged the latency to drink milk in

control animals (Figure II-3A, p<0.01), whereas it did not affect the total milk consumption

(Figure II-3B). For every rat tested, SKF 83959 treatment completely prevented any drinking

of milk over the 2 hour testing period (Figure II-3A&3B), whereas TAT-D1 treatment

significantly reduced the latency to drink milk in the novel environment (Figure II-3A, p<0.05)

without affecting total milk consumption (Figure II-3B). It should be noted that the SKF

83959-treated animals were able to move around normally in the novel cage. Thus the loss of

milk consumption was due to the animals’ choice rather than locomotor impairment. TAT-D1

pre-treatment robustly attenuated the enhanced latency to drink milk induced by SKF 83959

(Figure II-3B, p<0.001) and reinstated total milk consumption (Figure II-3B, p<0.001).

Together these findings indicate that D1-D2 heteromer activation has pro-depressive and

anxiogenic effects in rodents that in many instances are tonically active, which can be obviated

by disruption of the receptor complex.

111

Figure II-3: D1-D2 heteromer stimulation or inactivation had bidirectional effects on

anxiety-like behaviours in the novelty-induced hypophagia test. A) Exposure to novel

environment and SKF 83959 treatment significantly increased the latency to drink from milk

bottle, an effect that was significantly reduced by TAT-D1. B) The total milk consumption was

not affected by exposure to novel environment or TAT-D1, but was significantly reduced by

SKF 83959, an effect normalized by TAT-D1 pre-treatment. Data represent means ± s.e.m. of

n=8-12 rats/group. (*p<0.05, **p<0.01, ***p<0.001, compared to vehicle-treated group in

novel cage; ###p<0.001, compared to SKF 83959-treated group)

10

10

20

30

To

tal

Mil

k

Co

nsu

me

d (

mL

)

Pre

-Tes

t

Veh

icle

TAT-D

1

SKF

8395

9

TAT-D

1+SKF

###

***

Pre

-Tes

t

Veh

icle

TAT-D

1

SKF 8

3959

TAT-D

1+SK

F0

50

100

150

200

2000

4000

6000

80007200 ***

###

*

**

La

ten

cy

(se

c)

A) B)

112

3.4 CUS-induced Anhedonic Response

Having shown that D1-D2 heteromer stimulation by SKF 83959 induced a pro-

depressive and anxiogenic behavioural phenotype, we next wanted to determine whether D1-

D2 heteromer inactivation by the TAT-D1 peptide could alleviate the anhedonic response

exhibited by animals that underwent 5-week exposure to CUS. Two-way repeated measures

ANOVA on sucrose preference scores over the 5-week exposure showed significant effects of

Weeks {F(4,31)=11.08, p<0.001}, Stress Exposure {F(1,31)=40.34, p<0.001} and Weeks x

Stress Exposure interactions {F(4,31)=5.99, p<0.001}. Post-hoc comparisons revealed that

sucrose preference scores of CUS-exposed animals were significantly lower than control

animals at the end of the 5-week CUS exposure (Figure II-4A, p<0.01). On the test day, both

the control and CUS-exposed animals were administered TAT-D1 peptide or its inactive

control TAT-Sc. Two-way ANOVA revealed a significant effect of Stress Exposure

{F(1,31)=7.01, p<0.05}. Thus, TAT-Sc peptide-treated CUS animals had significantly lower

sucrose preference score compared to the control animals (Figure II-4B, p<0.05). Interestingly,

a single infusion of TAT-D1 peptide (300pmol, i.c.v.) partially reversed this CUS-induced

reduction of sucrose preference scores as no significant difference between scores of the

controls and those of the TAT-D1-treated CUS animals was observed (Figure II-4B).

In the fruit loops test, two-way ANOVA on the latency to consume fruit loops yielded a

significant main effect of Stress Exposure {F(1,27)=5.831, p<0.05} and Treatment

{F(1,27)=7.646, p<0.05}, with a non-significant Stress Exposure x Treatment interaction

{F(1,27)=3.192, p=0.085}(Figure II-4C). In non-CUS exposed animals, TAT-Sc and TAT-D1

113

Figure II-4. The effects of TAT-D1 peptide treatment on anhedonia-like reactivity

induced by chronic unpredictable stress. A) In the SPT, CUS-exposed animals (black)

exhibited a progressive and significant decrease in sucrose preference scores (% of total

volume consumed) in comparison to non-stressed controls (CTR, white) across 5 weeks of

exposure. B) A single administration of the TAT-D1 peptide (300pmol, i.c.v.) partially

reversed the CUS-induced reduction in sucrose preference score. C) In the FLT, CUS exposure

significantly increased the latency to consume fruit loops, an effect that was abolished by a

single TAT-D1 peptide infusion (300pmol, i.c.v.) D) CUS exposure had no effect on total fruit

loops consumption, whereas TAT-D1 peptide treatment (300pmol, i.c.v.) significantly

increased consumption in both the CTR and CUS-exposed animals. Data represent means ±

50

60

70

80

90

100

Su

cro

se

Co

nsu

mp

tio

n

(% o

f T

ota

l V

olu

me

)

TAT-D

1

TAT-S

c

TAT-D

1

TAT-S

c

CTRCUS

*

0

75

150

225

300

375

450

525

La

ten

cy

to

Co

su

me

Fru

it L

oo

ps (

se

c)

TAT-D

1

TAT-S

c

TAT-D

1

TAT-S

c

CTRCUS

**

0

1

2

3

4

5

6

To

tal

Fru

it L

oo

ps

Co

nsu

me

d

TAT-D

1

TAT-S

c

TAT-D

1

TAT-S

c

CTRCUS

**##

20

40

50

60

70

80

90

100

CTR CUS

1 2 3 4 5

Su

cro

se

Co

nsu

mp

tio

n

(% o

f T

ota

l V

olu

me

)

***

Weeks

A) B)

C) D)

114

peptide did not affect the latency to consume fruit loops (Figure II-4C). For TAT-Sc treated

animals, CUS exposure led to a significant increase in the latency to consume fruit loops

(Figure II-4C, p<0.01). The TAT-D1 peptide completely reversed CUS-induced increase in

latency to the level of TAT-Sc-treated control animals, with no significant effect in the control

animals (Figure II-4C). On the other hand, two-way ANOVA on the total fruit loops

consumption yielded a significant main effect of Treatment {F(1,22)=13.3, p<0.01}(Figure II-

4D), with TAT-D1 significantly increasing the number of fruit loops consumed in both the

control (p<0.05) and CUS-exposed rats (p<0.01), whereas CUS exposure did not affect total

fruit loop consumption compared to the control animals (Figure II-4D).

3.5 CUS-induced Anxiety-like Behaviours

In the novelty suppressed feeding test, exposure to the novel environment consistently

prolonged the feeding latency in all four groups as expected, regardless of stress exposure or

peptide treatment (Figure II-5A). Two-way ANOVA on the feeding latency in the novel

environment resulted in a significant Stress Exposure x Treatment interaction {F(1,27)=6.729,

p<0.05}(Figure II-5A). Post-hoc comparisons revealed that CUS exposure further significantly

prolonged the feeding latency in the novel environment (p<0.01), an effect that was completely

nullified by TAT-D1 peptide treatment (Figure II-5A). However, the TAT-D1 peptide did not

significantly reduce novelty-induced increase in feeding latency in the control animals (Figure

II-5A). On the other hand, despite the robust anxiogenic effect of CUS in the novelty-

suppressed feeding test, the total time spent in the open arms in the elevated plus maze was not

affected by CUS exposure or TAT-D1 peptide treatment (Figure II-5B).

115

Figure II-5. The effects of TAT-D1 peptide treatment on anxiety-like reactivity induced

by chronic unpredictable stress. A) In the NSFT, exposure to the novel environment

significantly increased the feeding latency in control (CTR) rats, an effect that was further

enhanced by exposure to CUS. The effect of CUS on novelty-induced increase in feeding

latency was completely abrogated by the TAT-D1 peptide (300pmol, i.c.v.). B) In the EPM,

neither CUS nor TAT-D1 treatment influenced total time spent in the open arms. Data

represent means ± s.e.m. of n=7-8/group. (**p<0.01, compared to CTR/TAT-Sc; CUS:

Chronic Unpredictable Stress, NSFT: Novelty Suppressed Feeding Test, EPM: Elevated Plus

Maze)

0

50

100

150

To

tal

Tim

e S

pe

nt

in

Op

em

Arm

s (

se

c)

TAT-D

1

TAT-S

c

TAT-D

1

TAT-S

c

CTRCUS

CTR

/TAT-S

c

CTR

/TAT-D

1

CUS/T

AT-S

c

CUS/T

AT-D

10

100

200

300

400

500 HomeNovel

La

ten

cy

to

Co

nsu

me

Fo

od

Pe

lle

ts (

se

c)

**

A) B)

116

4. Discussion

In the present study we demonstrated that stimulation of the dopamine D1-D2 receptor

heteromer induced depression- and anxiety-like behaviours. In the forced swim test, used as a

measure of passive coping or behavioural despair, D1-D2 heteromer stimulation by SKF 83959

significantly reduced the latency to immobility. When anxiety responses were tested in the

elevated plus maze, D1-D2 heteromer stimulation resulted in less time spent in the open arms

of the maze. In the novelty-induced hypophagia test, which measures anxiety induced by the

stress of a novel environment, D1-D2 heteromer stimulation abolished the eagerness of trained

animals to approach and consume highly palatable sweetened milk. More importantly, these

effects were consistently attenuated, or reversed, by pre-exposure to the selective TAT-D1

disrupting peptide prior to D1-D2 heteromer stimulation by SKF 83959, thus confirming the

contribution of the D1-D2 heteromer in these depression- and anxiety-related behavioural

phenotypes.

In addition, TAT-D1 peptide administration on its own exhibited rapid antidepressant-

like and anxiolytic properties in animals exhibiting depressive and anxiety-like phenotypes. In

CUS-exposed animals, TAT-D1 peptide partially reversed anhedonia-like reductions in sucrose

preference after just a single administration, suggesting that it could have a rapid antidepressant

action. Indeed, such an effect requires at least 2 weeks of daily treatment with conventional

antidepressants, during which improvements in sucrose preference scores start to ensue

(Willner et al., 1987a). While the improvement we observed with a single dose of TAT-D1

peptide was partial, it is likely that a few additional administrations could completely

normalize sucrose preference scores. Further investigations are warranted to assess the effects

117

of long-term TAT-D1 peptide treatments. Nevertheless, in our hands, TAT-D1 peptide

treatment significantly reversed CUS-evoked novelty-induced suppression of feeding to control

levels, thereby further supporting a potentially rapid-acting antidepressant-like activity induced

by D1-D2 heteromer inactivation. Such effect typically also requires chronic daily treatment

with conventional antidepressants to become apparent (Bodnoff et al., 1989; Bessa et al., 2009).

Moreover, TAT-D1 peptide also increased both the latency to and total consumption of fruit

loops in CUS-exposed animals, which further suggests an anti-anhedonic (or hedonic) property

that is associated with D1-D2 heteromer inactivation.

However, it is worth noting that TAT-D1 treatment by itself did not influence the total

time spent in the open arms in the elevated plus maze test, which was recapitulated in the CUS

experiments where neither CUS nor TAT-D1 treatment altered total time spent in the open

arms. This seems to be at odds with the robust effects of CUS and TAT-D1 peptide treatment

on anxiety-like reactivity in the novelty-suppressed feeding test. Such discrepancy in the

effects of CUS and the TAT-D1 peptide on anxiety-related behaviours as observed in the

elevated plus maze and novelty-suppressed feeding test may be explained by different forms of

anxiety triggered by CUS and antidepressants. Indeed, it was demonstrated that depressive-like

behaviour in the forced swim test and sucrose preference test correlated with anxiety-related

behaviour in the novelty-suppressed feeding test but not with the elevated plus maze (Bessa et

al., 2009).

The specificity of the TAT-D1 peptide to the D1-D2 heteromer has been thoroughly

characterized both in vitro and in vivo (Hasbi et al., 2014). The TAT-D1 peptide was

118

comprised of the same 16 amino acid sequence as the D1R C-terminus, including the two

glutamic acid residues that the two arginine residues in the third intracellular loop of the D2R

interact with to form the D1-D2 heteromer (O’Dowd et al., 2012; Hasbi et al., 2014). The TAT-

D1 peptide was able to dose-dependently reduce the BRET and FRET signals elicited by the

D1-D2 heteromer when it was incubated with HEK293 cells stably co-expressing both

receptors and primary cultured striatal neurons, respectively (Hasbi et al., 2014). In addition,

the TAT-D1 peptide also reduced the co-immunoprecipitation of D1R and D2R in cells as well

as in native striatal tissues, and dose-dependently attenuated SKF 83959-induced calcium

mobilization in primary cultured striatal neurons (Hasbi et al., 2014). The specificity of the

TAT-D1 peptide towards the D1-D2 heteromer was further confirmed by findings showing that

the BRET signals elicited by D1R homoligomer, D2R homoligomer, or D2-D5 heteromer were

not affected by TAT-D1 administration (Hasbi et al., 2014). Furthermore, a TAT-scrambled

control peptide that contains the same amino acid residues as the TAT-D1 peptide but in

random order had no effect in the BRET, FRET, Co-IP and calcium mobilization experiments

(Hasbi et al., 2014). Collectively these findings indicate that the TAT-D1 peptide was able to

selectively disrupt the D1-D2 heteromer formation and antagonize its signalling. Therefore, the

use of the TAT-D1 peptide would allow us to exclusively determine D1-D2 heteromer-specific

physiological effects.

Physiologically, activation of the D1-D2 heteromer induces Gq-mediated and

phospholipase C-dependent intracellular calcium release that results in the phosphorylation of

calcium calmodulin kinase II and expression of BDNF in the NAc (Lee et al., 2004; Rashid et

al., 2007; Hasbi et al., 2009). In addition, expression of the major GABA-producing enzyme,

119

GAD67, was also found to be significantly increased in the NAc following D1-D2 heteromer

stimulation (Perreault et al., 2012a), thereby potentially allowing D1-D2 heteromer to modulate

the inhibitory GABAergic tone in the NAc. Interestingly, although the proteins that are

modulated by the D1-D2 heteromer signalling are known for their involvement in the

regulation of psychostimulant reward (Anderson et al., 2008; Ghitza et al., 2010), evidence also

suggests their potential roles in the development of anxiety-and depression-like behavioural

phenotypes.

BDNF is well-known for its crucial involvement in the rewarding and addictive

properties of psychostimulants (Ghitza et al., 2010). Nevertheless, several studies have found

strong associations between BDNF expression and the development of anxiety- and

depression-like behaviours that are brain region-specific. For instance, reduced BDNF

expression has been documented in the hippocampus and the PFC of animals that underwent

chronic stress, as well as in post-mortem hippocampal tissue of depressed patients (reviewed in:

Duman, 2014). In contrast, depressed patients also displayed significantly increased levels of

BDNF in postmortem NAc that was not attributed to antidepressant administration (Krishnan et

al., 2007). In addition, powerful pro-depressive effects of BDNF signaling in VTA and NAc

have also been reported in rodents, highlighting the importance of BDNF in the mesolimbic

pathways of brain (reviewed in: Nestler and Carlezon, 2006; Russo and Nestler, 2013) in

mediating the pathophysiological effects of the disorder. Although BDNF in the NAc has been

traditionally thought to be derived from sources outside NAc such as VTA or cortex, and

transported to NAc via dopaminergic or glutamatergic projections respectively, we have now

shown that MSNs that express the D1-D2 heteromer provide a novel endogenous source for

BDNF in NAc (Hasbi et al., 2009). It has been postulated that the contribution of NAc to

120

depression may reflect an enhancement of GABAergic signaling within the region manifesting

depression-like responses such as anhedonia (Nestler and Carlezon, 2006b; Tye et al., 2013),

an interesting hypothesis given the strong positive correlation between BDNF and GABAergic

signaling in a number of regions in the brain including prefrontal cortex, and more recently as

we showed in NAc upon activation of the D1-D2 heteromer (Perreault et al., 2012a)

Animals that displayed reduced immobility in the FST also exhibited reduced BDNF

expression in the NAc (Sequeira-Cordero et al., 2014) whereas animals displaying anhedonic

behavior in the sucrose preference test showed increased NAc levels of BDNF mRNA (Bessa

et al., 2013). Similarly, the depressive effects of social defeat stress in rodents are associated

with a rapid and long-lasting increase in BDNF signaling in NAc in susceptible mice that is

concomitant with elevated c-fos expression (Berton et al., 2006), an effect that is mimicked by

selective activation of the D1-D2 heteromer (Hasbi et al., 2009; Perreault et al., 2012a, 2015).

Coincident with an increase in BDNF protein in NAc, there was increased expression of the

BDNF receptor TrkB (Perreault et al., 2013), indicative of heightened BDNF signaling in this

region. Interestingly activation of the D1-D2 heteromer also resulted in weight loss

(unpublished data), a physiological response also seen in mice susceptible to social defeat

stress. In line with these findings, suppression of BDNF signaling in VTA attenuated the stress-

induced increase of BDNF in NAc and promoted resilience in social defeat tests (Berton et al.,

2006; Krishnan et al., 2007; Fanous et al., 2011; Wang et al., 2014). Similarly, blockade of

BDNF signaling in NAc, using a negative functional mutant of TrkB, induced an

antidepressant effect in the FST (Eisch et al., 2003) whereas infusion of BDNF into the NAc

121

enhanced susceptibility to develop depression-like behaviours in animals that underwent social

defeat stress (Krishnan et al., 2007).

A potential involvement for the D1-D2 heteromer in prefrontal cortex (PFC) in

depression has been previously implicated (Pei et al., 2010). In the study by Pei et al, with the

use of a synthetic TAT-D2 peptide, disruption of the D1-D2 heteromer reduced both the total

immobility in the FST and the total number of escape failures in a learned helplessness

paradigm that involved chronic foot shock (Pei et al., 2010). Although the mechanism by

which D1-D2 heteromer disruption in PFC induced antidepressant effects in rodents was not

explored, evidence suggests it did not likely involve changes in BDNF or GAD67 as systemic

activation of the D1-D2 heteromer had no effect on the expression of either of these proteins in

the PFC (Perreault et al., 2012b). In addition, although enhanced glycogen synthase kinase-3

(GSK-3) activation in PFC has been implicated in the pathology of depression (Karege et al.,

2007), D1-D2 heteromer activation phosphorylated GSK-3 in PFC to suppress its activity

(Perreault et al., 2012b), thus negating a role for this kinase in mediating the antidepressant

effects of D1-D2 heteromer disruption in PFC. One possibility is that disruption of the D1-D2

heteromer in PFC regulates the activity of glutamatergic efferents, as the D1-D2 heteromer is

likely localized to cortical pyramidal neurons as a result of the high levels of D1R and D2R

coexpression in these neurons (Zhang et al., 2010). Nonetheless, although BDNF in NAc

appears to be a likely candidate in mediating the pro-depressive effects of the D1-D2 heteromer,

the molecular substrates that contribute to the antidepressant effects of D1-D2 heteromer

disruption in PFC remain to be elucidated.

122

A potential caveat of the present study is the use of SKF 83959 to examine the effect of

D1-D2 heteromer stimulation on the expression of depression- and anxiety-like behavioural

phenotypes. Although SKF 83959 is a pharmacological agent that can stimulate the D1-D2

heteromer without concurrently activating the D1R or the D2R, this alleged dopamine agonist

has also been shown to have affinity for, or activates, other receptors including the dopamine

D5 receptor (Sahu et al., 2009; Perreault et al., 2012b), the α2C-adrenergic receptor, and the

serotonin 5HT-2C receptor (Chun et al., 2013). Therefore, the pro-depressive and anxiogenic

effect of SKF 83959 treatment as observed in the current study may possibly not be solely

mediated by the D1-D2 heteromer, but could also be partly due to the involvement of other

receptor systems as well (Holmes et al., 2001; Bagdy et al., 2002). However, the ability of the

highly selective TAT-D1 peptide to attenuate or even reverse the pro-depressive and

anxiogenic effects of SKF 83959 in all three behavioural paradigms validated the involvment

of the D1-D2 heteromer in the manifestation of depressive and anxiety-like behavioural

phenotypes, as we have shown that the TAT-D1 peptide did not affect the D5R or the 5HT-2C

receptor (Hasbi et al., 2014). In addition, SKF 83959 treatment has also been previously shown

to suppress amphetamine reward (Shen et al., 2015), thus implicating the D1-D2 heteromer in

the negative modulation of the brain reward system. This may explain why a dose of SKF

83959 that was insufficient to promote anxiety in the elevated plus maze was able to

completely prevent milk consumption in the novelty-induced hypophagia test, which may be a

result of simultaneous inhibition of milk reward and the emergence of anxiogenesis in response

to SKF 83959 treatment. However, the reduction in the latency to consume sweetened milk in

the novelty-induced hypophagia test following TAT-D1 peptide pre-treatment again verifies a

role for the D1-D2 heteromer in promoting anxiety-like behaviours.

123

In summary, using agonist stimulation and selective antagonism by the use of a

heteromer disrupting TAT-D1 peptide, we have demonstrated that the dopamine D1-D2

receptor heteromer induced robust depression- and anxiety-like behaviour in rodents upon its

stimulation, a physiological function which has not been demonstrated for any other receptor

complex. More importantly, a potentially rapid onset of antidepressant-like activity produced

by selective inactivation of the D1-D2 heteromer may have great implication in the treatment

of depression and anxiety disorders as current interventions require chronic treatment to

achieve their therapeutic effects (Culpepper, 2010). Nevertheless, future studies will be

required to elucidate the exact mechanism by which the D1-D2 heteromer inactivation in the

NAc modulates the fast-onset antidepressant-like and anxiolytic effects as observed in the

current study.

124

3. The dopamine D1-D2 receptor heteromer exerts tonic inhibitory effect on the

expression of amphetamine-induced locomotor sensitization

Abstract

A role for the dopamine D1-D2 receptor heteromer in the regulation of reward and

addiction-related processes has been previously implicated. In the present study, we examined

the effects of D1-D2 heteromer stimulation by the agonist SKF 83959 and its disruption by a

selective TAT-D1 peptide on amphetamine-induced locomotor sensitization, a behavioural

model widely used to study the neuroadaptations associated with psychostimulant addiction.

D1-D2 heteromer activation by SKF 83959 did not alter the acute locomotor effects of

amphetamine but significantly inhibited amphetamine-induced locomotor responding across

the 5 day treatment regimen. In addition, a single injection of SKF 83959 was sufficient to

abolish the expression of locomotor sensitization induced by a priming injection of

amphetamine after a 72-hour withdrawal. Conversely, inhibition of D1-D2 heteromer activity

by the TAT-D1 peptide enhanced subchronic amphetamine-induced locomotion and the

expression of amphetamine locomotor sensitization. Treatment solely with the TAT-D1

disrupting peptide during the initial 5 day treatment phase was sufficient to induce a sensitized

locomotor phenotype in response to the priming injection of amphetamine. Together these

findings demonstrate that the dopamine D1-D2 receptor heteromer exerts tonic inhibitory

control on neurobiological processes involved in sensitization to amphetamine, indicating that

the dopamine D1-D2 receptor heteromer may be a novel molecular substrate in addiction

processes involving psychostimulants.

Shen M.Y.F., Perreault M.L., Fan T., George S.R. The dopamine D1-D2 receptor heteromer

exerts a tonic inhibitory effect on the expression of amphetamine-induced locomotor

sensitization. Pharmacol Biochem Behav. 2015 Jan;128:33-40.

125

1. Introduction

The dopamine D1-D2 receptor heteromer is a G protein-coupled receptor (GPCR)

complex that couples to the Gαq protein to elicit a phospholipase C (PLC)-dependent calcium

signal upon its activation (Rashid et al., 2007). It has been reported that a significant proportion

of D1 receptor (D1R)-expressing medium spiny neurons (MSNs) in the nucleus accumbens

(NAc) co-express the dopamine D1 and D2 receptors (17-25%) (Bertran-Gonzalez et al., 2008;

Matamales et al., 2009; Perreault et al., 2010; Gangarossa et al., 2013a) and approximately 90%

of these MSNs express the D1-D2 heteromer (Hasbi et al., 2009; Perreault et al., 2010). In

contrast, only 2-6% of D1R-expressing MSNs in the caudate putamen co-express the D1R and

D2R, of which only 25% of the neurons exhibit D1-D2 heteromer formation (Perreault et al.,

2010). The D1-D2 co-expressing neurons in the NAc extend efferent projections which directly

or indirectly influence the ventral tegmental area (VTA) (Perreault et al., 2012a), a region

widely known for its role in mediating addiction-like behaviours and reward through the

regulation of mesolimbic dopamine activity (Reviewed: Chen et al., 2010; Koob & Volkow,

2010).

We have previously shown that activation of the D1-D2 receptor heteromer modulated the

expression of proteins involved in drug addiction (Hasbi et al., 2009; Ng et al., 2010; Perreault

et al., 2010, 2012a), such as brain-derived neurotrophic factor (BDNF) and calcium-calmodulin

kinase II (CaMKII) in the NAc and VTA, and D1-D2 heteromer activation in NAc shell

enhanced production of the inhibitory neurotransmitter GABA in VTA (Perreault et al., 2012a).

These findings thus suggest a potential role for the D1-D2 heteromer in the regulation of

neuronal activity in the VTA and possibly as a regulator of brain reward processes.

126

Since repeated amphetamine treatment was previously shown to enhance the functional

activity of the D1-D2 heteromer in rat striatum (Perreault et al., 2010), in this study we aimed to

further examine the potential involvement of the D1-D2 heteromer in processes linked with

addiction using the amphetamine-induced locomotor sensitization model in rats.

Psychostimulant-induced locomotor sensitization was proposed to be an animal model for drug

craving, and is characterized by a context-dependent, progressive augmentation of locomotor

responsiveness following repeated non-contingent administration of psychostimulants such as

cocaine and amphetamine (Robinson and Becker, 1986; Kalivas and Stewart, 1991; Robinson

and Berridge, 1993; Anagnostaras and Robinson, 1996). amphetamine-induced locomotor

sensitization is associated with neuroadaptations of the mesolimbic dopamine system that may

enhance the reinforcing properties of cocaine and amphetamine, as animals that were previously

sensitized with repeated amphetamine treatment showed increased acquisition of drug self-

administration (Mendrek et al., 1998; Suto et al., 2002; Vezina et al., 2002).

Once established, amphetamine locomotor sensitization has been reported to persist for

over a year (Paulson and Robinson, 1991), which may be a reflection of some of the long-term

neurobiological adaptations that accompany the persistent drug-seeking behaviours typically

seen in addicted patients (Robinson and Berridge, 2000). Similarly, animals that exhibited the

reinstatement of cocaine-seeking behaviour induced by a single exposure to amphetamine also

expressed locomotor sensitization (De Vries, 1998), suggesting that changes in the mesolimbic

dopamine system that accompany the expression of amphetamine locomotor sensitization may

also contribute to relapse of drug seeking.

The dopamine agonist SKF 83959 is a partial agonist for the D1-D2 heteromer with a

number of in vitro and in vivo studies demonstrating its ability to induce D1-D2 heteromer-

127

mediated calcium signaling (Rashid et al., 2007; Verma et al., 2010), Gq and PLC activation

(Rashid et al., 2007), CaMKII phosphorylation (Ng et al., 2010) and BDNF production (Hasbi

et al., 2009). Validation of selectivity to the D1-D2 heteromer in these studies employed D1 and

D2 antagonists and dopamine receptor knockout mice. Recent studies, however, have indicated

that SKF 83959 has affinity for, or activates, a number of other receptors, such as the dopamine

D5 receptor, the α-adrenergic receptor 2C, and the serotonin 5HT-2C receptor (Sahu et al., 2009;

Perreault et al., 2012b; Chun et al., 2013). In addition, there are conflicting reports as to

whether SKF 83959 functions as an antagonist (Downes and Waddington, 1993; Cools et al.,

2002; Jin et al., 2003), a partial agonist (Lee et al., 2014), or has no effect (Lee et al., 2004;

Rashid et al., 2007) at the D1R. To assist in elucidating the physiological role of the D1-D2

heteromer, we developed a selective D1-D2 heteromer antagonist, the TAT-D1 peptide, which

occludes the interaction site between the two receptors (O'Dowd et al., 2012), thus inhibiting

D1-D2 heteromer expression and function and abolishing the physiological effects of D1-D2

heteromer activation by SKF 83959 without affecting other receptor oligomers such as D1-D1

homomers or D2-D5 heteromers (Hasbi et al., 2014). In the present study, we assessed the

effects of SKF 83959 on the expression of amphetamine locomotor responses and sensitization.

We only attributed an effect to be D1-D2 heteromer-specific when the TAT-D1 peptide

produced opposite behavioural output compared to SKF 83959. Our findings showed a novel

role for the D1-D2 heteromer in the suppression of amphetamine-induced locomotion and

locomotor sensitization.

128

2. Materials and Methods

2.1 Animals

Ninety-six adult male Sprague-Dawley rats (Charles River, Canada), weighing 300-350 g

at the start of the experiment, were used. Rats were housed in polyethylene cages in a

temperature-controlled colony room, maintained on a 12-h light-dark cycle (lights on at 0700h),

with ad libitum access to food and water. Rats were handled daily for 5 days before the start of

the experiment. All treatments were performed during the light phase of the day-night cycle.

Animals were housed and tested in compliance with the guidelines described in the Guide to the

Care and the Use of Experimental Animals (Canadian Council on Animal Care, 1993), and

were approved by the Animal Care Ethics Committee of the University of Toronto.

2.2 Drugs

SKF 83959 hydrobromide (Tocris Bioscience) was dissolved in physiological saline

containing 5% DMSO, and was administered subcutaneously (s.c.). Amphetamine

hydrochloride (Sigma-Aldrich) was dissolved in physiological saline (0.9% NaCl), and was

administered intraperitoneally (i.p.). The TAT-D1 disrupting peptide was dissolved in saline

and administered into the intracerebroventricular (i.c.v.) space 15 minutes prior to vehicle, SKF

83959, or amphetamine injection. For non-drug injections, an equivalent volume of saline or

vehicle was administered. All systemic injections were given at a volume of 1.0 ml/kg just prior

to behavioural testing.

129

2.3 Surgery

Rats were anesthetized with isoflurane (5%), administered analgesic ketoprofen (5 mg/kg,

s.c.) and secured in a stereotaxic frame. A cannula (22-gauge, Plastics One) was placed

unilaterally into the intracerebroventricular space close to the midline according to the

following stereotaxic coordinates: AP -0.8mm, ML + 1.3mm, DV – 3.7mm, and was secured by

dental cement anchored with four stainless steel screws (Plastics One) fixed on the dorsal

surface of the skull. AP and ML coordinates were taken from bregma, DV coordinate from the

dura (Paxinos and Watson, 1998). The animals were allowed to recover in their home cage for a

minimum of five days before the experiments were performed. Cannulae placement was

visually validated postmortem in brain slices.

2.4 Locomotor Activity Apparatus

The testing environment was a non-colony room containing 8 empty activity chambers that

are 20cm in height, 25cm in width, and 40cm in length. Two arrays of 16 infrared photocells

were attached along the longer sides of the polyethylene cages. The activity chambers were

interfaced to a computer that provided automated recording of horizontal locomotor activity

when both top and bottom infrared photocells were triggered. Ventilated polyethylene lids were

used to cover the activity chambers to prevent animals from escaping.

2.5 Locomotor Sensitization Protocol

The behavioural testing for locomotor sensitization to amphetamine was conducted using a

previously described protocol (Hall et al., 2008), which consisted of 3 phases:

Phase I: Habituation.

130

Rats were first habituated to the activity chamber for 2 days for 30 minutes per day.

Phase II: The Sensitizing Regimen

In this phase, we examined the effects of D1-D2 heteromer stimulation and inactivation on

locomotion induced by acute and subchronic amphetamine treatment (1.5mg/kg, i.p.). Animals

were administered their designated drug treatment, VEH+SAL, SKF 83959+SAL, TAT-

D1+SAL, VEH+amphetamine, SKF 83959+amphetamine, TAT-D1+amphetamine, once daily

for 5 consecutive days. Immediately following each injection, horizontal locomotor activity was

monitored for 60 minutes. The dose of SKF 83959 (0.4 mg/kg, s.c.) given in this phase was

chosen based on our previous study showing repeated SKF 83959 treatment significantly

enhanced locomotor activity and grooming responses without desensitizing the D1-D2

heteromer (Perreault et al., 2010). The dose of the TAT-D1 peptide (300pmol, i.c.v.) was

previously shown to disrupt D1-D2 heteromer formation in vivo as indicated by the loss of co-

immunoprecipitation of D1R with D2R in rat NAc tissue (Hasbi et al., 2014).

Phase III: The Expression of Locomotor Sensitization

Here we examined the effect of D1-D2 heteromer stimulation/inactivation alone or with

amphetamine during the sensitizing regimen on the expression of locomotor sensitization to

amphetamine. All animals from Phase II underwent a 72 hour withdrawal following the 5th

injection, which was a time point previously shown to be sufficient to induce the expression of

locomotor sensitization in response to amphetamine challenge (Hall et al., 2008). Following

withdrawal, all animals were challenged with a priming injection of amphetamine (1.0 mg/kg,

i.p.), and their horizontal locomotor activity were measured for 60 minutes. In addition, a subset

of animals that were treated solely with amphetamine during Phase II received the co-

131

administration of SKF 83959 (2.5 mg/kg, s.c.) or TAT-D1 peptide (300 pmol, i.c.v.) with the

priming amphetamine to examine the effect of acute D1-D2 heteromer stimulation or

inactivation on established locomotor sensitization to amphetamine. We increased the dose of

SKF 83959 for acute injection in the third phase, which was a dose that we have previously

shown to induce changes in CaMKII expression in rat striatum (Ng et al., 2010).

In addition, in order to elucidate the involvement of the D1-D2 heteromer on basal

locomotion, we performed a separate experiment in which animals were treated acutely with

VEH+SAL, SKF 83959 (0.4mg/kg, s.c.)+SAL, VEH+TAT-D1 (300pmol, i.c.v.), SKF

83959+TAT-D1 without habituation to the test chamber. The horizontal locomotor activity of

these animals was measured for 60 minutes immediately following SKF 83959 or VEH

treatment.

2.6. Data Analysis

All horizontal locomotor activity values are reported as mean ± s.e.m. All data was

analyzed for normality prior to ANOVAs using the Shapiro-Wilk test. The results shown in

Figure 1 were analyzed by Two-way ANOVA with D1-D2 heteromer Treatment (VEH, SKF

83959, TAT-D1 peptide) and amphetamine Treatment (SAL, amphetamine) as the between

subjects factors. The results shown in Figure 2 were analyzed by Repeated measures of

ANOVA with Injection as the within subject factor and Drug Treatments as the between subject

factor, followed by post-hoc test (Duncan multiple range test). The results shown in Figure 4

were analyzed by One-way ANOVA with Drug Treatments as the between subjects factor.

Planned comparisons between selected groups were performed using Student’s t-test. Planned

132

comparisons were performed for Figure III-4A, Expression of amphetamine sensitization,

between the following groups: Saline vs. AMPH; AMPH vs. SKF+AMPH; AMPH vs. TAT-

D1+AMPH. Computations were performed using the SPSS statistical program. Statistical

criteria for significant differences were set at p<0.05.

3. Results

3.1 The effects of acute D1-D2 heteromer activation and disruption on basal and amphetamine-

induced locomotor activity

Using the horizontal locomotor activity data obtained from the 1st injection, we first

examined the effect of acute SKF 83959 administration or TAT-D1-induced D1-D2 heteromer

disruption on basal and amphetamine-induced locomotor activity (Figure III-1A-C). Two-way

ANOVA with D1-D2 heteromer Treatment and amphetamine Treatment as between subject

factors revealed a significant main effect of D1-D2 Treatment {F(2:82)=3.489, p<0.05} and a

significant main effect of amphetamine Treatment {F(1:82)=78.690, p<0.0001}, as well as a

significant D1-D2 Treatment X amphetamine Treatment interaction {F(2:82)=7.335, p<0.001}.

Planned comparisons between treatment groups were then made using Student’s t-test. As

shown in Figure 1A, over the 60 minute period, administration of the agonist SKF 83959 (0.4

mg/kg, s.c.) alone induced a significant increase in locomotion compared to saline-treated

controls (bar 1: 818.6 ± 70.8 vs. bar 2: 3314.4 ± 389.6, t(14)=6.30, p<0.001). Amphetamine

treatment alone induced a significant increase in locomotion (bar 1: 818.6 ± 70.8 vs. bar 4:

6081.2 ± 291.1, t(46)=8.00, p<0.001), which was not influenced by SKF 83959 co-treatment

(bar 5: 5124.9 ± 377.2). Similar to SKF 83959, acute pre-administration (15 min) of the TAT-

133

Figure III-1. The effects acute D1-D2 heteromer stimulation and inactivation on basal

and amphetamine-induced locomotor activity. A) The total horizontal activity over 60

minutes in units of beam breaks in response to acute administration of amphetamine (1.5 mg/kg,

i.p.), alone or in combination with SKF 83959 (0.4 mg/kg, s.c.) or TAT-D1 peptide (300 pmol,

i.c.v.). B) Time course of the acute locomotor response measured in 5-min intervals over 60

minutes. C) The total horizontal activity for the first 5 minutes. D) Total horizontal activity in

the absence of habituation in response to SKF 83959, TAT-D1 peptide, or SKF 83959 + TAT-

D1. Data represent means ± s.e.m. of n=7-11 rats/group with the exception of amphetamine,

n=40 rats/group, and controls n=16 rats/group. AMPH=amphetamine. (*p<0.05, ** p<0.01,

*** p<0.001 compared to Saline Control.)

0

200

400

600

800SalineSKF 83959

AMPHSKF+AMPH

TAT-D1 TAT-D1+AMPH

5 10 15 20 25 30 35 40 45 50 55 60

Time (min)

Ho

rizo

nta

l A

cti

vit

y

(b

ea

m b

rea

ks)

0

200

400

600

800

1000

***

SKF 83959AMPH

TAT-D1

_ ____

__

__

__+

+ + +

++

+

Ho

rizo

nta

l A

cti

vit

y

(be

am

bre

ak

s)

Veh

icle

Veh

icle

TAT-D

1

SKF 8

3959

TAT+S

KF

0

1000

2000

3000

4000

5000* *

Novel EnvironmentHabituated

**

Ho

rizo

nta

l A

cti

vit

y

(be

am

bre

ak

s)

+

+

++

+__

0

1000

2000

3000

4000

5000

6000

7000

8000

***

SKF 83959AMPH

TAT-D1

_ ___

___

__+

+

*** ***

******

Ho

rizo

nta

l A

cti

vit

y

(be

am

bre

ak

s)

A) B)

C) D)

134

D1 peptide (300 pmol, i.c.v.) to rats, previously shown to disrupt the D1-D2 heteromer in rat

NAc (Hasbi et al., 2014), also elevated total locomotor activity compared to controls (bar 1:

818.6 ± 70.8 vs. bar 3: 2918.1 ± 317.6, t(14)=6.45, p<0.001) with no effect on acute

amphetamine-induced locomotion (bar 6: 6420.7 ± 440.2).

Examination of the time course of locomotion revealed that the acute locomotor activating

effects of SKF 83959 and TAT-D1 peptide steadily declined across the 60 minutes (Figure III-

1B, dotted lines). Conversely, the locomotor activity of animals treated with amphetamine

alone or in combination with SKF 83959 or TAT-D1 peptide remained elevated throughout the

testing period (Figure III-1B, solid lines). It is noteworthy that in the first 5 minutes of the

testing period (Figure III-1C), the animals treated with TAT-D1 peptide alone exhibited a very

robust increase in locomotor activity (t(22)=6.15, p<0.001) that was not apparent in the other

treatment groups. {D1-D2 Treatment: F(2,82)=17.6, p<0.0001; amphetamine: F(1,82)=0.2,

p<0.68; D1-D2 Treatment X amphetamine: F(2,82)=7.7, p<0.001}.

Since D1-D2 heteromer stimulation by SKF 83959 and its inactivation by the TAT-D1

peptide both stimulated basal locomotion in habituated animals, we wanted to further resolve

the role of D1-D2 heteromer in modulating basal locomotion. Therefore, we next tested animals

in the absence of habituation, a process which may affect total horizontal locomotor activity

due to the animal’s natural tendency to explore a novel environment (Eilam and Golani, 1989).

One-way ANOVA revealed a significant main effect of Drug Treatments {F(4,56)=16.749,

p<0.0001}. As shown in Figure III-1D, Duncan’s post hoc analysis showed that vehicle-treated

rats indeed exhibited significantly higher locomotion in the absence of habituation (bar 1: 818.6

± 70.8 vs. bar 2: 2466.4 ± 193.4, p<0.01). The exposure to a novel environment also abolished

135

Figure III-2. The effects of subchronic D1-D2 heteromer stimulation and inactivation on

basal and amphetamine-induced locomotor activity. Repeated amphetamine treatment (1.5

mg/kg, i.p.) produced a slight increase in total locomotion over the 5 injections. Repeated SKF

83959 co-treatment (0.4 mg/kg, s.c.) attenuated, while repeated TAT-D1 peptide pre-treatment

(300pmol, i.c.v.) enhanced, amphetamine-induced locomotion over the 5 injections. Across

injections animals treated solely with SKF 83959 or TAT-D1 peptide did not significantly

differ from Saline controls. Data represent means ± s.e.m. of n=7-11 rats/group with the

exception of amphetamine, n=40 rats/group, and controls n=16 rats/group.

AMPH=amphetamine. (*p<0.05, compared to amphetamine group)

1 2 3 4 50

1000

2000

3000

4000

5000

6000

7000

8000

9000

10000

11000

12000

SalineSKF 83959

AMPHSKF+AMPH

TAT-D1 TAT-D1+AMPH

*

*

Injection

Ho

rizo

nta

l A

cti

vit

y

(b

ea

m b

rea

ks)

136

the locomotor-activating effect of acute TAT-D1 peptide pre-treatment (bar 2: 2466.4 ± 193.4

vs. bar 3: 2970.8 ± 280.1), while the locomotion induced by SKF 83959 was maintained (bar 2:

2466.4 ± 193.4 vs. bar 4: 4112.3 ± 350.5, p<0.05). We further showed that pre-treatment with

TAT-D1 peptide did not attenuate the locomotor-activating effect of SKF 83959 in the absence

of habituation (bar 4: 4112.3 ± 350.5 vs. bar 5: 3806.8 ± 359.0). This demonstrates that

locomotor activation induced by SKF 83959 occurs via a mechanism not involving the D1-D2

heteromer.

3.2 The effects of D1-D2 heteromer activation and disruption on locomotor activity induced by

repeated amphetamine

To determine whether the D1-D2 heteromer could influence locomotor responding

induced by repeated amphetamine treatment, rats were administered 5 daily injections of

amphetamine alone or in combination with SKF 83959 or the TAT-D1 peptide, and locomotion

was monitored for 60 minutes following each drug injection (Figure III-2). Repeated measures

of ANOVA showed a significant within subject effect of Injection for the groups that received

amphetamine alone {F(4,40)=13.092, p<0.001} and amphetamine plus TAT-D1 {F(4,8)=5.106,

p<0.01}. Duncan’s post hoc analysis using Drug Treatment as the between subject factor

following repeated measures of ANOVA {F(5,83)=35.410, p<0.0001} further revealed that,

compared to the amphetamine alone group, co-administration of SKF 83959 significantly

diminished locomotor responsiveness to repeated amphetamine treatment (p<0.05) whereas

TAT-D1 co-treated rats exhibited significantly enhanced repeated amphetamine-induced

137

locomotion (p<0.05). Across injections, animals treated solely with SKF 83959 or TAT-D1

peptide did not significantly differ from saline controls.

We also examined the change in the temporal pattern of locomotor activity across the 5

injections for each of the treatments described in Figure 2. Repeated saline treatment had no

effect on the temporal pattern of locomotor activity, which showed a steady decline over the 60

minutes (Figure III-3, upper left panel). SKF 83959 treatment at the first injection produced an

initial excitation in locomotor activity that declined after the 15 minute mark (Figure III-3,

upper middle panel). With subsequent injections, this period of initial excitation gradually

diminished and ultimately disappeared. On the other hand, TAT-D1 peptide treatment

consistently produced a robust initial increase in locomotor activity within the first 5 minutes

across the 5 injections (Figure III-3, upper right panel). In contrast to SKF 838959 treatment,

repeated injections did not alter the temporal pattern of locomotor activity produced by the

TAT-D1 peptide.

The temporal pattern of locomotor activity for the first amphetamine injection steadily

increased and peaked at approximately 40 minutes (Figure III-3, lower left panel). With

successive amphetamine injections, a period of initial excitation emerged at the 10 minute mark,

followed by a period of secondary excitation that peaked at 40 minutes, indicating that repeated

injections altered the temporal pattern of locomotor activity associated with amphetamine

treatment. SKF 83959 co-treatment with amphetamine produced a similar temporal pattern of

locomotor activity over the 5 injections as the amphetamine alone group (Figure 3, lower

middle panel), although the overall locomotor activity was consistently lower in the SKF 83959

co-treated group. In contrast, TAT-D1 peptide pre-treatment with amphetamine progressively

138

Figure III-3. The time course of basal or amphetamine-induced locomotor responses

following D1-D2 heteromer stimulation or inactivation during the course of repeated

treatment. Horizontal activity is plotted at 5-min intervals. Lines depict overall group means.

AMPH=amphetamine. (n=8-11 rats/group with the exception of amphetamine, n=40 rats/group,

and controls n=16 rats/group).

0 10 20 30 40 50 600

200

400

600

800

1000TAT-D1

Time (min)

Dis

tan

ce

(b

ea

m b

rea

ks

)

Saline

0 10 20 30 40 50 600

200

400

600

800

1000

1

2

3

4

5

Injection

Time (min)

Dis

tan

ce

(b

ea

m b

reak

s)

0 10 20 30 40 50 600

200

400

600

800

1000Saline

Time (min)

Ho

rizo

nta

l A

cti

vit

y

(be

am

bre

ak

s)

0 10 20 30 40 50 600

200

400

600

800

1000SKF 83959

Time (min)

Dis

tan

ce

(b

ea

m b

rea

ks

)

0 10 20 30 40 50 600

200

400

600

800

1000

Time (min)

Ho

rizo

nta

l A

cti

vit

y

(be

am

bre

ak

s)

AMPH

0 10 20 30 40 50 600

200

400

600

800

1000 SKF+AMPH

Time (min)

Dis

tan

ce

(b

ea

m b

rea

ks)

0 10 20 30 40 50 600

200

400

600

800

1000 TAT-D1+AMPH

Time (min)

Dis

tan

ce

(b

ea

m b

rea

ks)

139

increased the maximum locomotor activity during the secondary excitation over the 5 injections

compared to the amphetamine alone group.

3.3 The effects of D1-D2 receptor heteromer activation and disruption on the expression of

amphetamine-induced locomotor sensitization

We next wanted to determine whether the D1-D2 heteromer could influence amphetamine-

induced expression of locomotor sensitization in response to a priming amphetamine injection

following a 72 hour withdrawal from the 5th

injection (Figure III-4). One-way ANOVA

revealed a significant main effect of Drug Treatment {F(7,84)=3.950, p<0.01}. As shown in

Figure III-4A, Planned comparisons further showed that the group treated with repeated prior

amphetamine injections (1.5 mg/kg, i.p.) exhibited significantly higher locomotor activity

compared to saline pretreated controls in response to the priming injection of amphetamine (1.0

mg/kg, i.p.) administered on the test day, indicative of locomotor sensitization to amphetamine

(bar 4: 6633.3 ± 354.3 vs. bar 1: 4723.0 ± 317.7, t(22)=3.46, p<0.01). Animals previously

administered SKF 83959 alone did not differ from saline pretreated controls in locomotor

activity following the priming amphetamine injection (bar 2 vs. bar 1). Interestingly, repeated

administration of TAT-D1 peptide alone elicited a sensitized locomotor response to the priming

amphetamine that was significantly higher than saline pretreated controls (bar 3: 7337.0 ± 920.4

vs. bar 1: 4723.0 ± 317.7, t(14)=2.69, p<0.05), and similar in magnitude to that observed in the

amphetamine-sensitized group.

Animals that received SKF 83959 plus amphetamine during the 5 day injection period did

not exhibit sensitized responding to the priming amphetamine on the test day (bar 5 vs. bar 1).

140

Figure III-4. The effects of D1-D2 receptor heteromer stimulation and inactivation on the

expression of amphetamine-induced locomotor sensitization. A) Total horizontal activity in

response to a priming injection of amphetamine (1.0 mg/kg, i.p.) for animals treated daily for 5

days with vehicle, SKF 83959 (0.4 mg/kg, s.c.), TAT-D1 peptide (300 pmol, i.c.v.),

amphetamine (1.5 mg/kg, s.c.), AMPH+SKF 83959 or AMPH+TAT-D1 peptide. Total

horizontal activity in response to priming amphetamine plus SKF 83959 (2.5mg/kg, s.c.) or

TAT-D1 (300pmol, i.c.v.) for amphetamine-sensitized animals are also shown. B) Time course

of the acute locomotor response to priming amphetamine in rats that received subchronic

pretreatment with Saline, SKF 83959 or TAT-D1 peptide. C) The time course showing the

effect of subchronic pretreatment with amphetamine, SKF 83959+amphetamine or TAT-

D1+amphetamine on priming amphetamine-induced locomotor sensitization. D) Time course

of the sensitized locomotor response to priming amphetamine alone, amphetamine+SKF 83959

or amphetamine+TAT-D1 for amphetamine pre-treated rats. Horizontal activity in 5-min

intervals is shown and is plotted in units of beam breaks. Data represent means ± s.e.m. of n=8-

16 rats/group. AMPH=amphetamine. (*p<0.05, ** p<0.01 compared to SAL group, #p<0.05

compared to amphetamine group.)

0

300

600

900

Saline SKF TAT-D1

Pretreatment:

Test: AMPH

5 10 15 20 25 30 35 40 45 50 55 60

*

Time (min)

Ho

rizo

nta

l A

cti

vit

y

(b

eam

bre

ak

s)

0

300

600

900

AMPH SKF+AMPH

Test: AMPH

Pretreatment:TAT-D1+

AMPH

5 10 15 20 25 30 35 40 45 50 55 60

#

Time (min)

Ho

rizo

nta

l A

cti

vit

y

(beam

bre

ak

s)

0

300

600

900 AMPH SKF+AMPH TAT-D1

+AMPH

Test:

Pretreatment: AMPH

5 10 15 20 25 30 35 40 45 50 55 60

#

Time (min)

Ho

rizo

nta

l A

cti

vit

y

(b

eam

bre

ak

s)

SAL

SKF

TAT-D

1

AM

PH

SKF+A

MPH

TAT-D

1+AM

PH

AM

PH

AM

PH

0

2000

4000

6000

8000

10000

12000

***

**#

AMPH TAT-D1+

AMPH

SKF+AMPH

*

Test:H

ori

zo

nta

l A

cti

vit

y

(be

am

bre

ak

s)

A) B)

C) D)

141

In contrast, animals that received prior exposure of amphetamine plus TAT-D1 for 5 days not

only produced a sensitized locomotor phenotype in response to the priming amphetamine

compared to the saline pretreated controls (bar 6: 8989.3 ± 994.0 vs. bar 1: 4723.0 ± 317.7,

t(14)=4.09, p<0.01), but the sensitized responding was also further significantly enhanced

compared to amphetamine-sensitized animals (bar 6: 8989.3 ± 994.0 vs. bar 4: 6633.3 ± 354.3.

t(22)=2.76, p<0.05). Acute co-administration of SKF 83959 with the priming amphetamine on

the test day (bar 7) abolished the expression of locomotor sensitization in animals pretreated

with amphetamine whereas acute TAT-D1 peptide pre-treatment with the priming amphetamine

did not affect the expression of locomotor sensitization to amphetamine (bar 8: 7948.5 ± 1179.5

vs. bar 1: 4723.0 ± 317.7, t(14)=2.64, p<0.05).

Lastly, we also examined the time course of the locomotion induced by the priming

amphetamine on the test day for each of the drug treatments described previously in Figure III-

4A. Prior repeated treatment solely with SKF 83959 or the TAT-D1 peptide did not alter the

dynamics of horizontal locomotor activity over the 60 minutes in response to the priming

amphetamine on the test day compared to the saline pretreated group (Figure III-4B). However,

post hoc analysis following repeated measures of ANOVA using Treatment as the between

subject factor {F(2,24)=3.31, p<0.056} showed a significant increase in locomotion over the 60

minutes following amphetamine priming for animals treated solely with repeated TAT-D1

peptide during the sensitizing regimen compared to those treated with repeated saline (p<0.05).

Prior SKF 83959 co-treatment or TAT-D1 pre-treatment with amphetamine during the

sensitizing regimen also did not affect the dynamics of horizontal locomotor activity over the

60 minutes in response to the priming amphetamine on the test day compared to the

amphetamine alone group (Figure III-4C). Nevertheless, SKF 83959 co-treatment and TAT-D1

142

pre-treatment with amphetamine consistently showed an opposing effect on priming

amphetamine-induced locomotion at each 5 minute time point throughout the 60 minute period

{F(2,34)=5.28, p<0.05}, with the TAT-D1 pre-treated group being significantly higher than the

amphetamine alone group (p<0.05) and SKF 83959 co-treatment showing a trend towards a

significant decrease (p<0.07). Similarly, acute co-treatment of SKF 83959 and pre-treatment of

TAT-D1 with the priming amphetamine also showed an opposing effect on priming

amphetamine-induced locomotion at each 5 minute time point throughout the 60 minute period

in animals previously sensitized with repeated amphetamine {F(2,36)=6.14, p<0.01} (Figure

III-4D), with SKF 83959 co-treatment being significantly lower than the amphetamine alone

group (p<0.05). Moreover, SKF 83959 co-treatment with the priming amphetamine also altered

the dynamics of horizontal locomotor activity starting from the 15 minute mark compared to the

amphetamine alone group, while TAT-D1 pre-treatment had no effect.

4. Discussion

In the present study we demonstrated that the dopamine D1-D2 receptor heteromer plays a

significant regulatory role in the locomotor activating effects of amphetamine and in locomotor

sensitization to amphetamine. Specifically, we showed that SKF 83959 co-treatment reduced

the magnitude of the locomotor response induced by repeated amphetamine administration and

abolished the expression of locomotor sensitization, with even a single treatment of the agonist

SKF 83959 being sufficient to abolish the sensitized phenotype of amphetamine-treated rats.

Conversely, inactivation of the D1-D2 receptor heteromer by a selective disrupting TAT-D1

peptide enhanced locomotion induced by repeated amphetamine treatment and augmented the

143

expression of sensitized locomotor responding to amphetamine priming, findings which

strongly suggest that the D1-D2 heteromer functions to tonically suppress the neurobiological

processes involved in these responses.

Amphetamine-induced locomotor sensitization has been associated with neuroadaptations

in the mesolimbic system that are critically involved in the regulation of reward and motivation

(Robinson and Becker, 1986; Henry and White, 1991; Kantor et al., 1999), and which have

been proposed to enhance the incentive salience, or “wanting”, of the drug itself, as well as

drug-related stimuli such as the environmental context associated with drug use, leading to

compulsive drug seeking behaviors that are typically seen in addicted individuals (Robinson

and Berridge, 2000; Schmidt and Beninger, 2006). Thus, these findings provide the first

evidence of a possible role for the receptor complex as a negative regulator of the physiological

processes associated with amphetamine addiction.

Numerous studies have demonstrated that amphetamine-induced locomotor sensitization is

a complex behaviour involving the close interplay between various neurotransmitter systems in

regions of the mesocorticolimbic circuitry. The initiation of amphetamine-induced locomotor

sensitization is thought to depend on enhanced VTA dopaminergic neuron excitability in the

form of long-term potentiation (LTP), which occurs as a consequence of reduced D2 auto-

receptor inhibition and reduced inhibitory GABAergic control in the VTA, as well as increased

glutamate release from the medial prefrontal cortex, following repeated non-contingent

amphetamine administration (White and Wang, 1984; Wolf and Xue, 1998; Saal et al., 2003).

This subsequently results in increased dopamine release in the NAc in response to amphetamine

challenge that act on supersensitized D1 receptors (Robinson et al., 1988; Wolf et al., 1994),

leading to the expression of amphetamine-induced locomotor sensitization. It should be noted

144

that the increased dopamine release into the NAc was also shown to be dependent on the

enhanced activity of CaMKII in the NAc following amphetamine challenge in cocaine-

sensitized rats (Pierce and Kalivas, 1997). In addition to increased dopamine release,

amphetamine challenge also results in increased glutamate release into the NAc from the

medial PFC in amphetamine-sensitized animals, the attenuation of which has been shown to

block the expression of amphetamine-induced locomotor sensitization (Kim et al., 2005).

There are several potential mechanisms by which the D1-D2 heteromer stimulation may

counteract the neuroadaptations associated with amphetamine-induced locomotor sensitization.

We have previously demonstrated that D1-D2 heteromer stimulation by a single systemic SKF

83959 injection significantly increased the expression of a major GABA producing enzyme,

glutamic acid decarboxylase (GAD67), in the VTA (Perreault et al., 2012a), potentially

enhancing GABAergic control in the VTA and preventing LTP of VTA dopaminergic neurons

known to be associated with the initiation of amphetamine-induced locomotor sensitization. In

addition, we have also shown that repeated stimulation of the D1-D2 heteromer significantly

reduced the expression of CaMKII in the NAc (Perreault et al., 2010), which may prevent the

enhanced dopamine release in the NAc upon amphetamine challenge and thereby abolish the

expression of amphetamine-induced sensitization. Another possibility is that the D1-D2

heteromer may reduce AMPA receptor activity in the NAc, as enhanced glutamatergic activity

was shown to be required for the expression of this behaviour (Kim et al., 2005). This idea is

supported by our previous work which showed repeated administration of SKF 83959

significantly reduced phosphorylation of AMPA receptor subunit GluA1 at the Ser831 site in

the NAc (Perreault et al., 2010), a site shown to be critical for inducing a long-lasting

potentiation of acute amphetamine-induced locomotion (Loweth et al., 2010).

145

Perhaps the most intriguing finding in the present study is the fact that subchronic D1-D2

heteromer inactivation by the TAT-D1 peptide alone was able to produce a sensitized

locomotor phenotype following amphetamine priming to a magnitude comparable to that which

was observed in amphetamine-sensitized animals, suggesting that the D1-D2 heteromer exerts

tonic suppression on the neurological processes associated with the expression of amphetamine-

induced locomotor sensitization. Although the mechanism through which the D1-D2 heteromer

exerts such effect remains unclear, a study suggests the potential involvement of cyclin-

dependent kinase 5 (cdk5). In the study by Taylor et al., animals that received subchronic intra-

NAc administration of cdk5 inhibitor roscovitine also exhibited a sensitized locomotor

phenotype upon cocaine priming to a degree similar to that produced by cocaine-sensitized

animals (Taylor et al., 2007). Furthermore, co-administration of roscovitine with cocaine during

the induction phase significantly enhanced the expression of cocaine locomotor sensitization

induced by cocaine priming (Taylor et al., 2007), which remarkably mirrors the effect of TAT-

D1 peptide co-administration on the expression of amphetamine locomotor sensitization in the

current study. Taken together, it is possible that cdk5 may be a downstream effector of the D1-

D2 heteromer signaling, and thus D1-D2 heteromer inactivation by the TAT-D1 peptide had

similar effect on the expression of psychostimulant-induced locomotor sensitization as the cdk5

inhibitor roscovitine. This is an interesting hypothesis that will be addressed in future studies.

We have previously demonstrated that the D1-D2 heteromer has representations in the

mesolimbic, striatonigral and striatopallidal pathways (Perreault et al., 2010, 2011) and that

medium spiny neurons that express the D1-D2 heteromer also express both GABA and

glutamate (Perreault et al., 2012a). In addition, the acute effects of D1-D2 heteromer activation

in the NAc on the expression of proteins involved in GABA and glutamate transmission were

146

shown to be pathway specific, with enhanced GABA-related protein expression in NAc/VTA

and increased glutamate-related protein expression in the dorsal striatum (Perreault et al.,

2012a). It is therefore possible that the D1-D2 heteromer may exert dual modulation on

amphetamine-induced locomotion by differentially regulating glutamate and GABA

transmission in several nuclei within the mesocorticolimbic and the basal ganglia circuitry

(Perreault et al., 2011, 2012a). This ability of the D1-D2 heteromer to potentially regulate both

GABA and glutamate neurotransmission would allow for the “fine tuning” of neuronal activity

within the mesolimbic direct and indirect dopamine pathways (Lobb et al., 2010). This line of

reasoning is supported by our previous study which showed increased functional activity of the

D1-D2 heteromer in the striatum of amphetamine-sensitized rats (Perreault et al., 2010), which

may be a potential negative-feedback compensatory process in an attempt to normalize

disturbances in dopaminergic transmission induced by subchronic amphetamine treatment.

A potential limitation of the present study is the use of SKF 83959 to examine the

behavioural effects of D1-D2 heteromer activation. No selective agonists for the D1-D2

heteromer are available, and while SKF 83959 has been shown to definitively induce the Gq-

PLC-linked calcium signaling mediated by the D1-D2 heteromer, its pharmacology has been

shown to extend to include a few other receptors including the dopamine D5 receptor (Sahu et

al., 2009; Perreault et al., 2012b; Chun et al., 2013; Guo et al., 2013). While the use of the

highly selective disrupting TAT-D1 peptide that exclusively targets the specific interaction sites

between the D1R and the D2R allows us to conclusively isolate D1-D2 heteromer-specific

effects on specific amphetamine-induced locomotor responses, off-target effects of SKF 83959

are also clearly evident. Indeed, we showed herein that acute locomotor activation induced by

147

SKF 83959 was not regulated by the D1-D2 heteromer as the TAT-D1 peptide did not abolish

these locomotor effects in the absence of habituation to the testing chamber.

It is also important to note that SKF 83959 attenuated, but did not abolish, amphetamine-

induced locomotion during the five day sensitizing phase, indicating that other receptors also

likely contribute to locomotor activation by this psychostimulant. In addition, we did not

directly demonstrate that TAT-D1 could abolish the effects of SKF 83959 on subchronic

amphetamine-induced locomotion as we were concerned that administering all three agents

simultaneously may overwhelm the in vivo system, as both SKF 83959 and amphetamine affect

various neurotransmitter systems via multiple mechanisms of action (Fleckenstein et al., 2007;

Perreault et al., 2012b; Chun et al., 2013; Guo et al., 2013). Nonetheless, the ability of SKF

83959 and the highly specific TAT-D1 peptide to opposingly regulate amphetamine-induced

locomotor activation and sensitization argues for a pivotal role for the D1-D2 heteromer in

mediating these behaviours.

In the present study, we demonstrated that the dopamine D1-D2 receptor heteromer exerts

tonic suppression on the locomotor sensitization behaviour. Despite its lack of clear face

validity, psychostimulant-induced locomotor sensitization results in neurobiological alterations

in the mesolimbic dopamine system that have been suggested to be the underlying molecular

basis for the subsequent manifestation of drug-seeking behaviours, which makes

psychostimulant-induced locomotor sensitization an important model for the study of drug

addiction (Robinson and Berridge, 2000; Steketee and Kalivas, 2011). Indeed, not only have

behaviorally sensitized animals been shown to concomitantly exhibit conditioned place

preference and drug self-administration (Lett, 1989; Pierre and Vezina, 1997, 1998; Vezina,

2004), but the neurocircuitry of psychostimulant-induced locomotor sensitization was also

148

shown to largely overlap with that of the reinstatement model of drug self-administration

(Steketee and Kalivas, 2011), suggesting that the persistent neuroadaptations associated with

psychostimulant sensitization may also contribute to eventual drug relapse. Therefore, the

identification of the dopamine D1-D2 receptor heteromer as a negative modulator of locomotor

sensitization to amphetamine may have implications in the understanding and treatment of drug

addiction.

149

5. Regulation of c-fos expression by the dopamine D1-D2 receptor heteromer

Abstract

The dopamine D1 and D2 receptors form the D1-D2 receptor heteromer in a subset of

neurons and couple to the Gq protein to regulated intracellular calcium signaling. In the present

study the effect of D1-D2 heteromer activation and disruption on neuronal activation in rat

brain was mapped. This was accomplished using the dopamine agonist SKF 83959 to activate

the D1-D2 heteromer in combination with a TAT-D1 disrupting peptide we developed, and

which has been shown to disrupt the D1/D2 receptor interaction and antagonize D1-D2

heteromer-induced cell signaling and behaviour. Acute SKF 83959 administration to rats

induced significant c-fos expression in nucleus accumbens that was significantly inhibited by

TAT-D1 pretreatment. No effects of SKF 83959 were seen in caudate putamen. D1-D2

heteromer disruption by TAT-D1 did not have any effects in any striatal subregions, but

induced significant c-fos immunoreactivity in a number of cortical regions including the

orbitofrontal cortex, prelimbic and infralimbic cortices and the piriform cortex. The induction

of c-fos by TAT-D1 was also evident in the anterior olfactory nucleus, as well as the lateral

habenula and thalamic nuclei. These findings show for the first time that the D1-D2 heteromer

can differentially regulate c-fos expression in a region-dependent manner either through its

activation or through tonic inhibition of neuronal activity.

Perreault M.L.*, Shen M.Y.F.*, Fan T., George S.R. Regulation of c-fos expression by the

dopamine D1-D2 receptor heteromer. Neuroscience. 2015 Jan 29;285:194-203. *These

authors contributed equally to this work

The work on the immunohistostaining in this study was done by Dr. Melissa Perreault.

150

1. Introduction

The dopamine D1 and D2 receptors (D1R and D2R) can form a heteromeric receptor

complex, the D1-D2 receptor heteromer, that exhibits pharmacological and functional

properties distinct from its constituent receptors (Lee et al., 2004, Rashid et al., 2007, Hasbi et

al., 2009). The distribution of the dopamine D1-D2 receptor heteromer has only been partially

characterized thus far, with the major focus being directed to the striatal subregions. In rodents

and non-human primates an abundance of neuroanatomical evidence now suggests that the

D1R and D2R are coexpressed in a subset of striatal medium spiny neurons (Meador-Woodruff

et al., 1991, Surmeier et al., 1992, Lester et al., 1993, Surmeier et al., 1996, Aubert et al., 2000,

Lee et al., 2004, Deng et al., 2006, Bertran-Gonzalez et al., 2008, Hasbi et al., 2009, Matamales

et al., 2009, Perreault et al., 2010, Gangarossa et al., 2013), with low receptor coexpression

(~4-6% of neurons) in caudate putamen (CP) and higher coexpression levels in the nucleus

accumbens (NAc) (~17-30% of neurons) (Bertran-Gonzalez et al., 2008, Perreault et al., 2010,

Gangarossa et al., 2013). Approximately 90% of coexpressing neurons in NAc expressed the

D1-D2 heteromer with only about 25% in CP (Perreault et al., 2010). While D1-D2 heteromer

expression in other brain regions has for the most part not been determined, it is expressed in

globus pallidus (Perreault et al., 2011), in the medial prefrontal cortex (mPFC) (Pei et al.,

2010). The dopamine D1-D2 receptor heteromer has been linked to Gq-mediated

phospholipase C activation and intracellular calcium signaling (Lee et al., 2004, Rashid et al.,

2007, Hasbi et al., 2009), the activation of calcium calmodulin kinase II (Rashid et al., 2007,

Ng et al., 2010), the expression and release of brain-derived neurotrophic factor (BDNF) in

NAc (Hasbi et al., 2009, Perreault et al., 2012) and reduced activation of glycogen synthase

kinase-3β (GSK-3β) in the PFC (Perreault et al., 2013). Furthermore, activation of the D1-D2

151

heteromer in NAc shell was shown to regulate the expression of protein markers of GABA and

glutamate in ventral tegmental area and substantia nigra (Perreault et al., 2012) suggesting that

activation of the D1-D2 heteromer may exert local effects as well as have farther reaching

effects through efferent projections.

A role for the dopamine D1-D2 receptor heteromer in the regulation of neuronal

activity has not been explored but can be examined through the expression of immediate early

genes, such as c-fos, which have often been used as a measure of neuronal activation within

circuits (Perez-Cadahia et al., 2011). The dopamine agonist SKF 83959, which is a partial

agonist for the D1-D2 heteromer (Rashid et al., 2007), has been shown to induce dorsal striatal

Fos expression at high doses (Wirtshafter and Osborn, 2005). SKF 83959 has often been used

to activate the D1-D2 heteromer with dopamine receptor knockout mice (D1R-/- and D5R-/-)

used to validate selectivity as SKF 83959 also activates the PLC-coupled D5R (Sahu et al.,

2009, Perreault et al., 2013). However recent reports indicate that this compound also exhibits

affinity at other receptors such as the serotonin 5HT-2c receptor (Chun et al., 2013), may act as

an allosteric modulator at the sigma-1 receptor (Guo et al., 2013), and there are conflicting

reports as to whether SKF 83959 functions as an antagonist (Downes and Waddington, 1993,

Cools et al., 2002, Jin et al., 2003), a partial agonist (Lee et al., 2014), or has no effect (Lee et

al., 2004, Rashid et al., 2007) at the D1R To assist in elucidating the physiological role of the

D1-D2 heteromer, we developed a selective D1-D2 heteromer antagonist, the TAT-D1 peptide,

which occludes the interaction site between the two receptors (O'Dowd et al., 2012), thus

inhibiting D1-D2 heteromer expression and function and abolishing the physiological effects of

D1-D2 heteromer activation by SKF 83959 (Hasbi et al., 2014). Therefore in the present study,

152

using SKF 83959 together with TAT-D1, we sought to address the involvement of the D1-D2

heteromer in regulating neuronal activation as indexed by the induction of c-fos expression.

2. Materials and Methods

2.1 Animals

Sixty-eight adult male Sprague-Dawley rats (Charles River, Canada), weighing 300-350

g at the start of the experiment, were used. Rats were housed in polyethylene cages in a

temperature-controlled colony room, maintained on a 12-h light-dark cycle (lights on at 0700),

with ad libitum access to food and water. Rats were handled daily for 5 days before the start of

the experiment. All treatments were performed during the light phase of the day-night cycle.

Animals were housed and tested in accordance with the guidelines described in the Guide to the

Care and the Use of Experimental Animals (Canadian Council on Animal Care, 1993), and

were approved by the Animal Care Ethics Committee of the University of Toronto.

2.2 Drugs and Peptides

SKF 83959 hydrobromide (Tocris Bioscience) was dissolved in physiological saline

containing 5% DMSO, and was administered subcutaneously (0.4, 2.5 mg/kg, s.c.). Haloperidol

(0.5 mg/kg, Sigma Aldrich) was used as a positive control for c-fos immunochemistry in

striatum, dissolved in a 0.3% tartaric acid in water, and administered intraperitoneally (i.p.). For

non-drug injections, an equivalent volume of vehicle was administered and all injections were

given at a volume of 1.0 ml/kg. The TAT-D1 disrupting peptide, or TAT-scrambled peptide

153

control (TAT-Sc) (Hasbi et al., 2014), was dissolved in saline and administered into the

intracerebroventricular space (300 pmol/4µL, i.c.v.) 15 minutes prior to vehicle or SKF 83959.

2.3 Surgery

Rats were anesthetized with isoflurane, administered analgesic ketoprofen (5 mg/kg)

and secured in a stereotaxic frame. A cannula was placed unilaterally into the

intracerebroventricular space close to the midline according to the following stereotaxic

anterior-posterior (AP), mediolateral (ML), and dorsoventral (DV) coordinates: AP -0.8mm,

ML + 1.3mm, DV – 3.7mm. AP and ML coordinates were taken from bregma, DV coordinate

from skull surface (Paxinos and Watson, 1998). The animals were allowed to recover in their

home cage for a minimum of five days before the experiments were performed.

2.4 Grooming

Grooming activity was monitored for 30 minutes immediately following SKF 83959

(0.4 mg/kg) injection. Animals were placed in clear cages containing no bedding

(20x20x45cm3). The measurement of grooming behavior followed a previously described

protocol (Culver et al., 2000). The animal’s grooming was scored for 30 second intervals, for a

total of 4 minutes (2 minutes sampled from the first 15 minutes of testing and 2 minutes

sampled from the last 15 minutes of testing). Ventilated polyethylene lids were used to cover

the cages to prevent animals from escaping.

154

2.5 Immunochemistry

Ninety minutes following SKF 83959 (2.5 mg/kg) or vehicle injection, brains were

rapidly removed and frozen in isopentane (−60°C) and stored at −80°C until cryostat sectioning.

Serial sections through the prefrontal cortex (PFC, Bregma 3.2 mm), CP/NAc (Bregma

1.6 mm), ventral pallidum (Bregma 0.2), globus pallidus (Bregma -0.8 mm), lateral

hypothalamus (Bregma -1.8), habenula/hippocampus/thalamus/amygdala (Bregma -3.6),

substantia nigra/ventral tegmental area (Bregma −5.6 mm) and rostromedial tegmental nucleus

(Bregma -6.8 mm) were cut coronally at 16-μm thickness and mounted onto gelatin-coated

glass slides. Sections were air dried and stored at −35°C until use. Immunochemistry for c-fos

was performed as previously described (Sundquist and Nisenbaum, 2005) with the following

changes. Slides were brought to room temperature, immersed in 4% paraformaldehyde and

washed several times in TBS-Tween (0.05%). Slides were then placed in methanol containing

0.3% hydrogen peroxide, washed, and blocked by a 10 minute incubation (humid chamber) in a

solution of 10% goat serum in TBS-T. Tissue was incubated with anti-rabbit c-fos primary

antibody (1:250, Cell Signaling) in antibody diluent (1% BSA in TBS-T) for 2 hours in humid

chamber, washed, and then incubated with a biotinylated goat anti-rabbit secondary antibody

(Vector Laboratories) in antibody diluent for 45 mintues followed by several washes. Sections

were reacted with avidin-biotin peroxidase complex (Vectastain Elite Kit; Vector Laboratories)

for 30 minutes and washed. c-fos positive nuclei were visualized with VIP (Vector

Laboratories), washed and dehydrated, cleared in xylene and coverslipped with Vectamount

(Vector Laboratories). Images were obtained at 10X magnification using an Axioplan2

microscope (Carl Zeiss).

155

2.6 Statistical Analysis

Values are reported as mean ± s.e.m. The grooming data was measured in seconds and

the immunochemistry data was collected by cell counting of c-fos positive nuclei (4

fields/animal). Comparisons of means for time spent grooming and cell counts in NAc and CP

were performed by ANOVA, with Treatment as the between subjects factor, followed by

Bonferroni post-hoc tests. Comparisons of means in all other brain regions were performed by

Student’s t test (two-tailed, unpaired). Statistical significance set at P<0.05 and computations

were performed using the SPSS/PC+ statistical package.

3. Results

3.1 Effects of dopamine D1-D2 receptor heteromer activation on striatal c-fos expression

A role for the dopamine D1-D2 heteromer in the induction of c-fos expression in NAc

core and shell (Figure IV-1A) was first evaluated. Administration of the D1-D2 receptor

heteromer agonist SKF 83959 or the D2 receptor antagonist haloperidol resulted in a

homogenous distribution of highly labelled c-fos positive nuclei in NAc core (Figure IV-1B)

and in NAc shell. In contrast, disruption of the dopamine D1-D2 heteromer by administration

of TAT-D1 resulted in little to no c-fos expression in either NAc subregion. Quantification of

the number of immunoreactive cells (Figure IV-1C&D) showed a significant increase in highly

labeled c-fos positive cells compared to TAT-Sc-treated controls in both the NAc core (22.1 ±

2.0 vs. 0.4 ± 0.2 cells/field, p<0.0001) and NAc shell (22.6 ± 2.2 vs. 0.2 ± 0.2 cells/field,

P<0.0001) in response to SKF 83959, and these effects of SKF 83959 on c-fos expression in

156

Figure IV-1: Regulation of c-fos expression in rat nucleus accumbens (NAc) by the

dopamine D1-D2 receptor heteromer. A) Schematic showing regions of sampling for c-fos

positive nuclei. B) Representative images showing the effect of D1-D2 heteromer activation

and disruption on c-fos immunoreactivity in NAc core. Magnification of select immunopositive

nuclei (solid arrow) and nuclei considered negative (dashed arrow) are shown inset. C, D) SKF

83959 (2.5 mg/kg) administration resulted in a significant increase in the number of c-fos

positive cells in both NAc core and shell and to a magnitude similar to that observed with

haloperidol (HALO) which was used as positive control. The effects of SKF 83959 were

significantly inhibited by pretreatment with a TAT-D1 disrupting peptide. TAT-D1 alone had

no effect on c-fos levels. (Bars shown represent means ± s.e.m. ***p<0.001 compared to TAT-

Sc controls, ###

p<0.001 compared to SKF 83959-treated rats. Scale bar 100 µm).

A) B)

C) D)

157

NAc were of similar magnitude to that seen with haloperidol treatment. Pre-administration of

TAT-D1 significantly reduced SKF 83959-induced c-fos expression (NAc Core: 22.1 ± 2.0 vs.

10.1 ± 0.9; NAc Shell: 22.6 ± 2.0 vs. 13.3 ± 0.9 cells/field) indicating that the effects of SKF

83959 were mediated by the D1-D2 heteromer. Overall TAT-Sc treatment had very little effect

on c-fos expression in either NAc subregion. ANOVA revealed a significant effect of

Treatment {F(4, 83)=82.9, p<0.0001}.

SKF 83959 has been demonstrated to induce oral movements and grooming behaviour

when administered systemically (Downes and Waddington, 1993, Deveney and Waddington,

1995, Perreault et al., 2010, Perreault et al., 2012) or directly into NAc shell (Perreault et al.,

2012). To determine whether these effects were mediated by the D1-D2 heteromer, we

assessed a role for the D1-D2 heteromer on SKF 83959-induced grooming behaviour (Figure

IV-2). We showed that the increased grooming response induced by SKF 83959 was abolished

by pretreatment with TAT-D1 (87.4 ± 8.4 vs. 48.3 ± 6.6 seconds, p=0.003) whereas TAT-D1

alone did not influence the amount of time animals spent grooming compared to controls (51.3

± 6.7 vs. 46.8 ± 7.0 seconds, p=1.00) {Treatment: F(3, 44)=7.2, p<0.0001}.

We next evaluated the effects of SKF 83959 and TAT-D1 on c-fos immunoreactivity in CP

(Figure IV-3A), a region with very little dopamine D1-D2 heteromer expression. Similar to that

observed in NAc, haloperidol-treated rats exhibited homogeneous expression of c-fos-positive

nuclei in CP, whereas SKF 83959 and TAT-D1, and TAT-Sc had little to no effect on c-fos in

this region (Figure IV-3B-D). Furthermore, in contrast to haloperidol which showed robust c-

fos labelling in many of the nuclei, c-fos positive cells in CP of SKF 83959 or TAT-D1/SKF

83959-treated rats were only lightly labelled (Figure IV-3C). TAT-D1 pretreatment had no

158

Figure IV-2. Grooming induced by SKF 83959 in rats is mediated by the dopamine D1-

D2 heteromer. A single systemic injection of SKF 83959 (0.4 mg/kg) induced a significant

increase in the amount of time spent grooming compared to TAT-Sc-treated rats. TAT-D1

pretreatment abolished SKF 83959-induced grooming but did not influence grooming

behaviour when administered alone (n=12 rats/group). (Bars shown represent means ± s.e.m.

and are expressed in seconds (s). ***p<0.001 compared to TAT-Sc controls).

159

Figure IV-3: Regulation of c-fos expression in rat caudate putamen (CP) by the dopamine

D1-D2 receptor heteromer. A) Schematic showing area of sampling for c-fos positive nuclei.

B) Representative images showing the effect of D1-D2 heteromer activation and disruption on

c-fos immunoreactivity in CP. Select immunopositive nuclei chosen for magnification are in

the boxed area. C) Magnification of select c-fos positive cells showing the intensity of antibody

labeling between Treatment groups. D) SKF 83959 (2.5 mg/kg) induced a very modest

increase in the number of c-fos positive cells CP that was not abolished by pretreatment with

TAT-D1. In contrast, haloperidol (HALO) induced a robust increase in the number of c-fos

positive cells. (Bars shown represent means ± s.e.m. ***p<0.001 compared to TAT-Sc

controls. Scale bar 100 µm).

A) B)

C) D)

160

effect on the small amount of c-fos labelling induced by SKF 83959 {Treatment: F(4, 83)=82.9,

p<0.0001}.

3.2 Effects of dopamine D1-D2 heteromer disruption on c-fos expression in anterior olfactory

nucleus and cortical subregions

SKF 83959 induced c-fos expression in a number of regions of the brain that was not

attributed to the D1-D2 heteromer. This significantly confounded the findings in animals that

received both TAT-D1 and SKF 83959. We therefore chose to focus selectively on a role for

D1-D2 heteromer disruption on c-fos expression for the remainder of the experiments. The

effects of acute TAT-D1 administration on c-fos expression in were evaluated in anterior

olfactory nucleus (AOP) and a number of cortical subregions (Figure IV-4). Dopamine D1-D2

heteromer disruption resulted in increased c-fos immunoreactivity in AOP and most of the

cortical regions including orbitofrontal cortex (OFC), the infralimbic and prelimbic cortices (IL,

PL) and piriform cortex (PiC), indicating the release of a tonic inhibitory effect. In the AOP,

TAT-D1-induced c-fos expression was relatively dense, with medium to high labelling (Figure

IV-4A). Quantitative data showed an approximate 5-fold increase in the number of positive

nuclei compared to TAT-Sc-treated controls (20.6 ± 2.0 vs. 3.5 ± 1.2 cells/field; t(32)=7.9,

p<0.0001). The distribution of c-fos immunoreactivity in OFC by TAT-D1 (Figure IV-4B) was

similar to that of AOP with relatively strong and dense labelling compared to TAT-Sc controls

(20.7 ± 2.2 vs. 2.1 ± 0.6 cells/field; t(32)=9.4, p<0.0001).

Administration of TAT-D1 resulted in c-fos expression in both the IL and PL that was

significantly higher than controls (IL: 9.1 ± 1.0 vs. 2.1 ± 0.6 cells/field; t(32)=6.3, PL: 10.9 ±

161

Figure IV-4. Effects of dopamine D1-D2 heteromer disruption on c-fos immunoreactivity

in rat cortex. Representative images and quantitative data showing significantly increased c-

fos staining in A) the anterior olfactory nucleus (AOP) and B) the orbitofrontal cortex (OFC)

following administration of the TAT-D1 disrupting peptide. C) TAT-D1 treatment resulted in

intense labeling of c-fos cells in the infralimbic cortex (IL) but fewer cells were labelled than

that observed in AOP and OFC. D) c-fos immunoreactivity was increased in the prelimbic

cortical region following TAT-D1 administration. E) No effect of TAT-D1 on c-fos expression

in cingulated cortex (Cg). F) TAT-D1 increased c-fos levels in piriform cortex (PiC). G)

Schematic showing regions of sampling for c-fos positive nuclei. Select magnified neurons

displaying c-fos positive (solid arrow) and negative (dashed arrow) are shown inset.

(***p<0.001 compared to TAT-Sc controls. Scale bar 100 µm).

A)

B)

C)

D) E) F) G)

162

1.7 vs. 1.7 ± 0.5 cells/field; t(32)=6.0, p<0.0001; Figure IV-4C&D) but with a more sparse

distribution than that observed in AOP and OFC. Where c-fos immunoreactivity was present,

nuclei were highly labelled. Only modest effects of TAT-D1 peptide on c-fos expression were

seen in the cingulate cortex (Figure IV-4E). In the PiC, TAT-D1 significantly increased c-fos

expression with mild to medium intensity labelling (14.9 ± 2.1 vs. 2.2 ± 0.7 cells/field;

t(32)=6.5, p<0.0001; Figure IV-4F).

3.3 Effects of dopamine D1-D2 heteromer disruption on c-fos immunoreactivity in lateral

habenula and other regions

Disruption of the dopamine D1-D2 heteromer by TAT-D1 induced c-fos expression in

the lateral habenula (LHb), but not medial habenula (MHb), as well as the thalamic nuclei

(Figure IV-5). Medium intensity labelling occurred predominantly in the ventral LHb with

sparse labelling in the dorsal area (Figure IV-5B). Quantification of the number of

immunoreactive nuclei revealed a significantly higher number of c-fos positive cells in LHb

following TAT-D1 treatment (13.4 ± 1.8 vs. 1.7 ± 0.5 cells/field; t(26)=7.1, p<0.0001). TAT-

D1 induced a homogenous distribution of c-fos expression, of medium to high intensity, in the

mediodorsal thalamic nucleus with modest staining observed in the paraventricular thalamic

nucleus (PVP) (Figure IV-5C). The number of c-fos immunoreactive cells were significantly

higher in the thalamic regions following treatment with TAT-D1 as compared to TAT-Sc

treatment (17.8 ± 1.7 vs. 3.0 ± 0.7 cells/field; t(26)=8.8, p<0.0001).

163

Figure IV-5: Effect of dopamine D1-D2 heteromer disruption on c-fos expression in rat

lateral habenula and thalamus. A) Schematic showing area of sampling for c-fos positive

nuclei. B) Representative images and quantitative data showing significantly increased c-fos

staining in the lateral habenula (LHb) but not the medial habenula (MHb) following D1-D2

heteromer disruption by TAT-D1. C) TAT-D1 significantly increased c-fos expression in the

thalamic nuclei including the mediodorsal thalamic nucleus and paraventricular nucleus. Select

magnified neurons displaying c-fos positive (solid arrow) and negative (dashed arrow) are

shown inset. ***P<0.001 compared to TAT-Sc controls. Scale bar 100 µm.

A)

B)

C)

164

No significant effects of TAT-D1 on c-fos expression were observed in any other brain

region including the ventral pallidum, globus pallidus, lateral hypothalamus, hippocampus,

amygdala, substantia nigra, ventral tegmental area and rostromedial tegmental nucleus.

4. Discussion

In the present study we showed that the activation state of the dopamine D1-D2

receptor heteromer contributed to neuronal activation in a region-specific manner as evidenced

by increased c-fos expression. Specifically, whereas activation of the D1-D2 heteromer

induced c-fos in the NAc core and shell, its disruption resulted in significantly elevated c-fos

immunoreactivity in a number of cortical regions, as well as the LHb and thalamus. These

findings suggest that the D1-D2 heteromer plays a dual role in the regulation of neuronal

activity, increasing activity in some regions and exerting tonic suppression on activity in others.

Recent studies have indicated that SKF 83959 has affinity for, or activates, a number of

different receptors (Sahu et al., 2009, Chun et al., 2013, Perreault et al., 2013) highlighting the

critical importance of using TAT-D1 in this study. It was especially obvious in the cortical

regions whereby SKF 83959 significantly induced c-fos expression, an effect not attenuated by

pretreatment with TAT-D1 (data not shown), and likely mediated by dopamine D5R (Perreault

et al., 2013), or other receptors. This was additionally confounded by our findings which

showed that D1-D2 heteromer disruption by TAT-D1 induced c-fos expression in several

subregions of the cortex. We therefore chose to examine the effects of SKF 83959 solely in

striatal regions where administration of TAT-D1 individually had no effect. In CP very

minimal effects of the 2.5 mg/kg dose of SKF 83959 were observed, a finding consistent with

165

previous reports showing a lack of effect on c-fos expression of another dopamine agonist

linked to PLC activation, SKF 38393 (potentially induced through activity at the D1-D2

heteromer), in this region in normal rats (Robertson et al., 1991, Paul et al., 1992, LaHoste et

al., 1993). In NAc core and shell a marked increase in c-fos levels were observed following

SKF 83959 administration and this effect was significantly inhibited, but not abolished by

pretreatment with TAT-D1. These findings in NAc indicate that the observed effects on c-fos

expression were mediated in large part by the D1-D2 heteromer. The effects of D1-D2

heteromer activation in CP and NAc on c-fos expression are also consistent with the known

distribution of the receptor complex in striatum. In CP, coexpression of the D1R and D2R is

very low (~4-6% of cell bodies) (Bertran-Gonzalez et al., 2008, Perreault et al., 2010) with

only about 25% of these coexpressing neurons showing D1-D2 heteromer formation (Perreault

et al., 2010), and thus only ~1-2% of CP neurons in total expressing the D1-D2 heteromer. In

contrast, in NAc expression of the D1-D2 heteromer is much higher with ~17-30% of neurons

coexpressing the D1R and D2R (Bertran-Gonzalez et al., 2008, Perreault et al., 2010,

Gangarossa et al., 2013) and the majority of these (>90%) also expressing the D1-D2

heteromer (Perreault et al., 2010). These findings are also consistent with our previous data

showing that SKF 83959 could induce grooming behaviour upon injection into NAc shell

(Perreault et al., 2012). In the present study we were able to conclusively identify the D1-D2

heteromer as mediating SKF 83959-induced grooming but could not establish a role for the

receptor complex in basal grooming behaviour as TAT-D1 alone had no effect.

In NAc, the TAT-D1 peptide greatly reduced, but could not completely abolish SKF

83959-induced c-fos expression. There are two possible explanations. Firstly, it may be that the

dose of SKF 83959 was too high to be completely inhibited by TAT-D1. While this is certainly

166

a possibility, our previous behavioural study showed a complete loss of SKF 83959-induced

effects in the forced swim test with TAT-D1 pretreatment when using the same doses (Hasbi et

al., 2014). Another possibility is that some of the effects of SKF 83959 on c-fos expression in

NAc were mediated by a receptor other than the D1-D2 heteromer. SKF 83959 has been

shown to have affinity for the serotonin 5HT-2c receptor, as well as the α-adrenergic-2b

receptor (Chun et al., 2013). Although activation of signaling was not shown at these receptors

by SKF 83959, it cannot be ruled out as a possible contributor to the observed effects. The

dopamine D5R, which has lower expression in striatum being confined to cholinergic

interneurons, is also a potential candidate as SKF 83959 has been reported to activate D5R-

mediated PLC signaling (Sahu et al., 2009) and to induce D5R-mediated BDNF expression in

mPFC (Perreault et al., 2013). Nonetheless, our evidence clearly demonstrates a significant role

for the D1-D2 heteromer in mediating the effects of SKF 83959 on c-fos expression in NAc.

Disruption of D1-D2 heteromer activity by TAT-D1 resulted in the induction of c-fos in

numerous cortical regions including the PL and IL regions of the mPFC, OFC and PiC.

Significant coexpression of the D1R and D2R in mPFC has been reported (Zhang et al., 2010)

whereby double transgenic Drd1a-tdTomato/Drd2-EGFP mice were used to show that almost all

of the pyramidal neurons that expressed the D1R also expressed the D2R. That same year using

coimmunoprecipitation, the existence of the D1-D2 heteromer in the mPFC was confirmed (Pei

et al., 2010). The dopamine D1-D2 heteromer in PFC has now been shown to play a

significant role in depressive-like behaviour (Pei et al., 2010) and to suppress GSK-3β activity

by a mechanism likely involving activation of Akt (protein kinase B) (Perreault et al., 2013).

Increased activation of cortical GSK-3β has been linked to cognitive decline in schizophrenia

(Kozlovsky et al., 2005) and to contribute to the neurodegenerative process in Alzheimer’s

167

disease (Llorens-Maritin et al., 2014). Although expression of the D1-D2 heteromer has not yet

been established in OFC and PiC, the present findings indicate that the involvement of this

complex in these regions may worthy of further investigation, especially given their importance

in reward and decision making processes (Noonan et al., 2012) as well as social interactions

(Zenko et al., 2011). Indeed, one study reported social interaction impairment induced solely

by coactivation, not individual receptor activation, of D1R and D2R in the PiC (Zenko et al.,

2011).

Significant increases in c-fos were observed in the LHb, as well as the PVP and

mediodorsal thalamic nuclei, following acute TAT-D1 treatment. The LHb and thalamus has

received much attention in recent years due to their emerging role in drug addiction (Matzeu et

al., 2014, Velasquez et al., 2014). The acute administration of cocaine was shown to result in

increased c-fos expression in LHb neurons (Zahm et al., 2010) and notably, evidence now

suggests a relationship between c-fos levels in LHb and cocaine reinstatement. For example, in

one study a significant correlation between activation of LHb neurons and cocaine-primed

reinstatement in a conditioned place preference paradigm in mice has been identified (Brown et

al., 2010). Similarly, the propensity for cue-induced relapse in rats trained to self-administer

cocaine was shown to be related to increased c-fos expression in LHb neurons, as well as

specific nuclei of the thalamus including the PVP and mediodorsal nuclei (James et al., 2011).

Conversely, inactivation of the PVP abolished the expression of cocaine CPP in rats (Browning

et al., 2014). Together these findings suggest a positive contribution of LHb and thalamic

nuclei to cocaine-induced behaviours and thus highlight a potential contribution of the D1-D2

heteromer in the regulation of processes involved in addiction.

168

In summary, the current study demonstrates that the dopamine D1-D2 receptor

heteromer regulates neuronal activity in different regions of the brain, and does so in a highly

selective manner. Whereas activation of the D1-D2 heteromer increases c-fos

immunoreactivity in some regions such as NAc, its disruption promotes c-fos expression in

others such as in PFC, LHb and thalamus. These results indicate an important functional

duality of the D1-D2 heteromer in the regulation of neuronal output through direct activation or

tonic inhibition. Although the mechanisms by which this is achieved have not been elucidated,

these findings are indicative of the inherent inhibitory and excitatory characteristics of medium

spiny neurons that express the D1-D2 heteromer in striatum (Perreault et al., 2012). Future

studies examining more closely the neuroanatomical distribution of D1-D2 heteromer-

expressing neurons and their projections could provide important insights into how neuronal

activity is being regulated by this receptor complex.

169

6. General Discussion

The overall purpose of the present thesis was to characterize a functional role for the

D1-D2 heteromer in the brain reward system and was accomplished using a series of model

systems in rodents for addiction, depression, and anxiety. The present findings showed that D1-

D2 heteromer stimulation was able to suppress reward-seeking behaviour under conditions of

elevated dopamine release (Chapter I&III), whereas it promoted anhedonic, depression- and

anxiety-like behaviour under basal dopamine tone (Chapter II). In addition, it was

demonstrated that the D1-D2 heteromer modulated the neuronal activity of several regions

within the mesocorticolimbic system (Chapter IV), a circuitry significantly involved in the

reward pathways, and thus supporting a role for the D1-D2 heteromer in the regulation of

reward. Collectively, these findings demonstrate a novel role for the D1-D2 heteromer in the

negative regulation of brain reward function, and suggest that the dysregulation of D1-D2

heteromer activity may contribute to the pathophysiology of neuropsychiatric disorders such as

drug addiction and depression.

Given the highly comorbid nature of drug addiction and depression, it is not surprising

that studies have identified that the two neuropsychiatric disorders share a similar

neurobiological pathology, namely, a dysfunctional mesolimbic reward system (Nestler and

Carlezon, 2006b; Russo and Nestler, 2013). Drug addiction and depression may be viewed as

the two extreme ends of the hedonic spectrum, with drug addiction being the pathological

consequence of inducing reward system hyperactivity, and depression being characterized by

hypoactive brain reward function. This idea is supported by studies that examined the

functional activity of the brain reward system using intracranial self-stimulation (ICSS), which

demonstrated that drugs of abuse sensitized, whereas exposure to depression-inducing chronic

170

stressors dampened, brain reward function (West and Michael, 1986; Moody and Frank, 1990;

Vogel et al., 1990). In addition, while the mesolimbic dopaminergic activity is consistently

enhanced by drugs of abuse (Bradberry and Roth, 1989; Pellegrino and Druse, 1992; Nisell et

al., 1994; Stewart and Rajabi, 1996), recent evidence also showed that its inhibition promoted

depression-like symptoms in rats that underwent chronic stress (Tye et al., 2013), reinforcing

the notion that bidirectional modulation of the brain reward system may contribute to the

pathogenesis of drug addiction and depression. In this sense, as D1-D2 heteromer stimulation

and inactivation respectively induced a depression-like behavioural phenotype and promoted

reward-seeking behaviours, the D1-D2 heteromer may be a single molecular entity that

functions to regulate brain reward function, and thus may be a suitable therapeutic target for

drug addiction and depression though its activation or inactivation depending on the

neuropsychiatric disorder being evaluated.

6.1 The D1-D2 heteromer is a negative modulator of psychostimulant and natural reward

In Chapter I, it was shown that D1-D2 heteromer stimulation by the agonist SKF 83959

abolished cocaine reward in the CPP model, prevented cocaine-induced locomotor sensitization,

and attenuated cocaine intake in the SA model. More importantly, in all three behavioural

models examined, selective inactivation of D1-D2 heteromer consistently produced an opposite

behavioural effect compared to its stimulation, thereby providing strong support for a role for

the D1-D2 heteromer in the negative modulation of cocaine reward. Moreover, the inhibitory

effect of the D1-D2 heteromer was not limited to cocaine-mediated behaviours as D1-D2

heteromer stimulation similarly prevented the locomotor sensitization to amphetamine (Chapter

171

III, Shen et al. 2014b) and attenuated the consumption of sucrose solution in the sucrose

preference test (Hasbi et al. unpublished data), whilst selective D1-D2 heteromer inactivation

enhanced amphetamine sensitization and sucrose intake. These results indicate that the D1-D2

heteromer negatively modulated psychostimulant and natural reward, and prevented the

behavioural sensitization that are associated with repeated exposure to psychostimulants.

Despite the traditional view that the D1R- and D2R-expressing MSNs are largely

segregated in the NAc (Gerfen et al., 1990; Harrison et al., 1990; Le Moine and Bloch, 1995;

Surmeier et al., 2007), evidence has also emerged that supported the existence of D1R/D2R co-

expressing MSNs (Bertran-Gonzalez et al., 2008, 2010; Matamales et al., 2009; Perreault et al.,

2010; Gangarossa et al., 2013a), within which the D1-D2 heteromer is expressed (Perreault et

al., 2010, 2011; Hasbi et al., 2014). This specific population of D1R/D2R co-expressing MSNs

in the NAc was shown to extend efferent projections to nuclei within both the direct D1R-

expressing and the indirect D2R-expressing pathways (Deng et al., 2006; Wang et al., 2006,

2007; Perreault et al., 2011), and has thus been suggested to function as a third neuronal

pathway within the basal ganglia circuitry that maintains homeostatic balance between the

neuronal output from the direct and indirect pathways (Perreault et al., 2011). This notion is

supported by the finding showing that the D1R/D2R co-expressing MSNs also possess a

unique GABA/glutamate co-expressing phenotype (Perreault et al., 2012a), which may allow

the D1-D2 heteromer to differentially regulate the inhibitory GABAergic and excitatory

glutamatergic activities in the direct and indirect pathways.

Given the established critical involvement of D1R- and D2R-expressing MSNs in the

modulation of psychostimulant reward (Nestler and Carlezon, 2006b; Thomas et al., 2008;

Steketee and Kalivas, 2011; Baik, 2013), it would be expected that stimulation of the D1R/D2R

172

co-expressing MSNs could similarly have a modulatory influence on the rewarding effects of

psychostimulants. Coincidentally, stimulation of the D1-D2 heteromer, which results in a Gαq-

mediated and PLCβ-dependent calcium signalling (Lee et al., 2004; Rashid et al., 2007; Hasbi

et al., 2009), may be a means to activate the D1R/D2R co-expressing MSNs in the NAc since it

has been shown that PLCβ-mediated PIP2 hydrolysis in the striatum resulted in the inhibition of

KCNQ potassium channels, leading to enhanced MSN activity in the striatum (Zhang et al.,

2003; Shen et al., 2005). Supporting this notion, the DREADD technology uses engineered

muscarinic M3 receptors (hM3Dq) that signals through Gαq-mediated, PLCβ-dependent

calcium signalling to selectively stimulate neurons on which it is virally expressed (Alexander

et al., 2009). Therefore, the ability for D1-D2 heteromer stimulation to inhibit psychostimulant-

induced reward-seeking behaviours indirectly implicated a novel physiological role for the

D1R/D2R co-expressing MSNs in the negative modulation of psychostimulant reward.

Nevertheless, although D1-D2 heteromer stimulation was shown to increase c-fos induction in

the NAc, indicating neuronal activation in this region (Chapter IV, Perreault et al. 2015),

electrophysiology studies would be beneficial in determining the relationship between D1-D2

heteromer stimulation and the neuronal activity of the specific D1R/D2R co-expressing MSNs

in the NAc.

In addition, it is possible that the negative modulatory effect of the D1-D2 heteromer on

psychostimulant reward may also extend to influence natural reward, since D1-D2 heteromer

stimulation and inactivation respectively attenuated and enhanced the consumption of sucrose

solution, and repeated D1-D2 heteromer stimulation prevented animals from gaining weight

(Hasbi et al. unpublished finding). Indeed, it has been proposed that the neurobiology

underlying psychostimulant and natural reward may overlap to some degree (Cooper et al.,

173

1992; Hodge et al., 1994; Hajnal and Norgren, 2001; Gambarana et al., 2003; Halpern et al.,

2013; Cameron et al., 2014), and thus repeated exposure to natural reward stimuli may

similarly result in neuroadaptations in the mesolimbic dopamine system as those induced by

repeated psychostimulant exposure to contribute to the pathophysiology of eating disorders and

obesity (Volkow et al., 2008; Grosshans et al., 2011). Therefore, the D1-D2 heteromer may

function to inhibit the fundamental brain reward mechanism, allowing its stimulation to

simultaneously dampen both psychostimulant and natural reward. Interestingly, this notion

also raises the possibility that D1-D2 heteromer stimulation under basal dopamine level, that is,

without an external reward stimulus, may result in excessive suppression of brain reward

function, potentially leading to an aversive and depressive state. Indeed, the present findings

also showed the development of depression- and anxiety-like behaviour following D1-D2

heteromer activation under normal physiological conditions.

6.2 D1-D2 heteromer stimulation induced a depressive-like state

In Chapter II, it was shown that D1-D2 heteromer stimulation by SKF 83959 promoted

depression- and anxiety-like behavioural phenotypes in the forced swim test, novelty-induced

hypophagia test, and in the elevated plus maze, effects suppressed or abolished by the TAT-D1

peptide. In addition, the TAT-D1 peptide by itself in many instances produced anxiolytic and

antidepressant-like activity in these behavioural paradigms. More importantly, acute D1-D2

inactivation by the TAT-D1 peptide also elicited rapid antidepressant-like reactivity in animals

that underwent chronic unpredictable stress, which models chronic depressive states (Willner et

al., 1987b), as measured in the sucrose preference test, fruit loop test, and the novelty-

174

suppressed feeding test. These findings together strongly suggest that activation of the D1-D2

heteromer, under normal physiological conditions, can significantly suppress brain reward

function, leading to a depressive and anxiety-like state.

Although the majority of studies in the depression field have largely focused on

noradrenergic and serotonergic signalling in the hippocampus and the frontal cortex

(Dranovsky and Hen, 2006; Duman and Monteggia, 2006; Müller and Holsboer, 2006; Albert

et al., 2014; Mahar et al., 2014), emerging evidence has also implicated a role for the

mesolimbic dopamine system in the pathophysiology of depression (Nestler and Carlezon,

2006b; Krishnan and Nestler, 2008; Russo and Nestler, 2013). Indeed, although the

hippocampus and the frontal cortex are undoubtedly involved in the cognitive and memory

aspects of depression, it does not encompass the core symptom of depression: the reward

deficits, which likely involves the mesolimbic dopamine system due to its well-established role

in regulating brain reward function (Nestler and Carlezon, 2006b). In addition, the fact that

drug addiction and depression are highly comorbid further suggests that the disturbances in the

mesolimbic dopamine system may be a common mechanism underlying the pathophysiology

of neuropsychiatric disorders characterized by reward dysfunction (Conway et al., 2006).

Specfically, findings showing that withdrawal from psychostimulants induced a

depressive-like behavioural phenotype in both humans and rodents (Barr and Markou, 2005b;

D’Souza and Markou, 2010) effectively placed the mesolimbic dopamine system between the

two reward-related neuropsychiatric disorders. Numerous studies have suggested that

withdrawal from psychostimulants results in neuroadaptations in the brain reward circuit that

would subsequently contribute to the manifestation of depressive behaviours (Markou and

Koob, 1992; Wise and Munn, 1995; Coffey et al., 2000; Barr and Phillips, 2002; McGregor et

175

al., 2005). Indeed, common alterations in the mesolimbic dopamine system were observed in

animals with a depressive-like behavioural phenotype and those that underwent

psychostimulant withdrawal, such as reduced basal extracellular dopamine concentration in the

NAc (Paulson et al., 1991; Rossetti et al., 1992, 1993; Gambarana et al., 1999), and decreased

firing of VTA dopaminergic neurons (Henry et al., 1989; Ackerman and White, 1992; Tye et

al., 2013).

The theoretical perspective of the emergence of depressive behaviours following

psychostimulant withdrawal can be explained by the “opponent process theory” of drug

addiction. Originally proposed by Solomon and Corbit, the opponent process theory postulates

that, whenever there is a disturbance in the neutrality of the brain reward system, an “opponent

process” ensues in an attempt to maintain homeostatic balance in the brain reward system

(Solomon and Corbit, 1974). According to the theory, the “opponent process” is slow to onset,

gradually reaches an asymptote, and slow to decay. In the case of psychostimulant exposure,

the excessive stimulation of the brain reward system by psychostimulants is counteracted by

neuroadaptations that inhibit the brain reward system in order to restore homeostatic balance.

Since the opponent process is slow to decay, the compensatory inhibition of the brain reward

system persists during psychostimulant withdrawal, thus dampening basal reward function to

result in an anhedonic and depressive state, which is subsequently proposed as a negative

reinforcement that drives addicts to seek and obtain psychostimulants in order to avoid the

aversive effect (Koob and Le Moal, 2001).

In this sense, the D1-D2 heteromer functions very much like an opponent process in

that its stimulation attenuated reward-seeking behaviours induced by external stimuli such as

psychostimulants and sucrose, and induced an anhedonic and depressive state in the absence of

176

an external rewarding stimulus due to excessive inhibition of brain reward function. In support

of this idea, enhanced functional activity of the D1-D2 heteromer was found in the NAc of

amphetamine-sensitized rats (Perreault et al., 2010), which may potentially be a compensatory

opponent process activated in an attempt to normalize disturbances in dopaminergic

transmission induced by subchronic amphetamine treatment. Thus, the heightened D1-D2

heteromer function following repeated amphetamine administration may be one of the

underlying mechanisms for the depressive-like behavioural phenotypes that have been

observed during amphetamine withdrawal (Markou et al., 2005; D’Souza and Markou, 2010).

It is an intriguing idea that the expression and functional activation of the D1-D2 heteromer

may be transient, which functions to suppress or enhance the reward pathways as required to

maintain balance and avoid the pathogenic changes that contribute to addiction and depression,

an idea that fits with the previously suggested theory of a homeostatic role for D1R/D2R

coexpressing MSNs in the regulation striatonigral and striatopallidal output (Perreault et al.,

2011).

6.3 The D1-D2 heteromer modulation of reward: A potential role for BDNF and GAD67

Amongst the various proteins that are regulated by the D1-D2 heteromer signalling

pathway, BDNF and GAD67 were shown to be the common molecular substrates underlying

both drug addiction and depression (Nestler and Carlezon, 2006b; Russo and Nestler, 2013;

Sanchez-Catalan et al., 2014). Whereas systemic D1-D2 heteromer stimulation increased the

expression of BDNF in the NAc (Hasbi et al., 2009; Perreault et al., 2013), D1-D2 heteromer

177

stimulation selectively in the NAc shell enhanced the expression of the major GABA-

producing enzyme, GAD67, in the VTA (Perreault et al., 2012a).

BDNF is well known for its role in mediating the rewarding effect of cocaine (Ghitza et

al., 2010; McGinty et al., 2010). In particular, infusion of BDNF into the NAc was shown to

enhance cocaine-seeking whereas the adenovirus-mediated knockdown of endogenous BDNF

in the NAc had the opposite effect (Graham et al., 2007b). Although these findings would

appear to be in contradiction with the previous findings which showed that D1-D2 heteromer

stimulation increased BDNF levels in the NAc (Hasbi et al., 2009; Perreault et al., 2012a), the

effects of BDNF on cocaine reward was shown to be dependent upon the subset of NAc MSNs

in which it signals (Lobo et al., 2010a). We therefore posit that D1-D2 heteromer stimulation

increased BDNF levels selectively in the D1R/D2R co-expressing MSNs in the NAc that

contributed to reduced cocaine reward. As BDNF signaling through TrkB in D1R-expressing

MSNs suppresses reward (Lobo et al., 2010a), its release from D1R/D2R-expressing MSNs

would act predominantly on the D1R MSN subtype to achieve this. We further suggest that

increased BDNF signaling in D1R-expressing MSNs may subsequently lead to increased Cdk5

activity (Bogush et al., 2007), a kinase that was shown to reduce the motivation to obtain

cocaine when expressed in the NAc (Taylor et al., 2007).

In addition to its effect on cocaine-seeking, BDNF has been shown to have region-

specific effects on depression-like behaviours. Whereas BDNF was shown to exert

antidepressant-like activity when expressed in the hippocampus (Schmidt and Duman, 2010;

Kavalali and Monteggia, 2012), BDNF expression in the NAc was found to be strongly pro-

depressive in various animal models for depression (Eisch et al., 2003; Krishnan et al., 2007;

Fanous et al., 2011; Wang et al., 2014). Moreover, depressed patients also displayed

178

significantly increased levels of BDNF in postmortem NAc that was not due to antidepressant

administration (Krishnan et al., 2007). Therefore, the pro-depressive behavioural effects of D1-

D2 heteromer stimulation as observed in the current study could be attributed to the ability of

the D1-D2 heteromer to provide an endogenous source of BDNF in the NAc (Hasbi et al.,

2009). Although a role for Cdk5 activity in NAc in depression has not yet been elucidated, it is

possible that the suppression of cocaine reward, and the induction of a depression-like

phenotype, may occur through a similar mechanism.

Another potential mechanism by which the D1-D2 heteromer regulates reward is

through the regulation of GAD67 expression in VTA which could result in the suppression of

dopaminergic activity within the mesolimbic reward system by increasing the inhibitory

GABAergic modulation of VTA dopamine neurons. Several studies have shown that the

inhibition of VTA dopaminergic activity attenuated reward-seeking behaviours (Friedman et al.,

2011; Van Zessen et al., 2012; Ilango et al., 2014). For instance, increasing the excitability of

GABAergic synapses on to VTA dopaminergic neurons was shown to suppress cocaine-

seeking behaviours (Friedman et al., 2010). Similarly, optogenetic stimulation of VTA

GABAergic neurons was shown to reduce dopamine release in the NAc and subsequently

disrupted the consumption of sucrose solution (Van Zessen et al., 2012).

More recently, studies have also implicated a role for VTA GABAergic neurons in the

induction of depression-like behaviours (Lecca et al., 2014; Proulx et al., 2014). The activity of

GABAergic neurons in the VTA that synapse on NAc-projecting dopaminergic neurons was

shown to be increased in animals that exhibited behavioural models for depression (Li et al.,

2011, 2013), as well as in animals that were congenitally depressed (Yang et al., 2008),

suggesting that reducing dopaminergic activity in the mesolimbic reward system may

179

contribute to the pathogenesis of depression. Supporting this, a recent study by Tye et al (2013)

showed that the chronic unpredictable stress model of depression resulted in a reduction of

VTA-NAc dopaminergic neuron firing, and optogenetic phasic stimulation of these neurons

induced an antidepressant-like effect in the tail suspension test and sucrose preference test.

Consequently, the induction of GAD67 expression in the VTA following D1-D2 heteromer

simulation may result in an increase in GABAergic activity in the same region as GAD67 is the

predominant GABA-producing enzyme, thereby contributing to the induction of depression-

like behaviours as observed in the current study. Similarly, the rapid antidepressant-like

activity of acute D1-D2 heteromer inactivation in the chronic unpredictable stress model of

depression shown herein is in line with the findings from the Tye et al. (2013) study in that

increasing VTA dopaminergic activity, potentially as a result of reduced D1-D2 heteromer-

mediated GABAergic influence in the same region, exerted antidepressant-like behavioural

effect. Nevertheless, the exact effect of D1-D2 heteromer-mediated GAD67 induction on the

activity of specific dopaminergic neurons in the VTA (NAc- or PFC-projecting) remains to be

investigated.

Collectively, the ability of the D1-D2 heteromer to modulate the expression of proteins

that are critically involved in both reward-seeking and depression-like behaviours reinforces

the notion that this receptor complex may be a link between drug addiction and major

depression, two highly comorbid neuropsychiatric disorders (Davis et al., 2008).

180

6.4 The D1-D2 heteromer is a novel molecular substrate for aversions

Perhaps the most intriguing finding in the current study was that agonist stimulation of

the D1-D2 heteromer was aversive, whereas its selective inactivation was rewarding (Chapter

I). This suggests that the D1-D2 heteromer is constantly stimulated by the basal dopamine

outflow in the NAc to exert tonic inhibition on brain reward function, possibly to maintain a

neutral hedonic perception in the absence of an external rewarding stimulus. The identification

of a dopamine receptor complex in the NAc that induces aversion is unprecedented.

Nevertheless, the concept of the mesolimbic dopamine system to modulate both reward and

aversion was recently introduced in light of the advancements in molecular technology that

allowed for cell type-specific analysis of the brain reward circuitry.

The VTA dopaminergic neurons that project to the NAc exhibit two distinct patterns of

firings: phasic/burst firing that is responsible for stimulating the low affinity D1R, and tonic

firing that is crucial for modulating the high affinity D2R (Mirenowicz and Schultz, 1994, 1996;

Grace et al., 2007). Phasic/burst firing of the VTA-NAc dopaminergic neurons via optogenetics

produced CPP, whereas tonic firing of these neurons had no effect (Tsai et al., 2009),

indicating that D1R stimulation in the NAc is critical in generating reward. On the other hand,

aversive stimuli were shown to reduce the tonic firing of VTA-NAc dopaminergic neurons

(Mirenowicz and Schultz, 1996; Ungless et al., 2004; Brischoux et al., 2009), resulting in

blunted tonic dopamine release in the NAc that eventually led to CPA (Liu et al., 2008). Since

D2R in the NAc is more sensitive to changes in the tonic firing of VTA-NAc dopaminergic

neurons (Grace et al., 2007), such a finding indirectly implicated a role for the D2R in the NAc

in mediating aversions. In addition, although individual optogenetic stimulation of the D1R-

and D2R-expressing MSNs in the NAc was shown to be ineffective in eliciting CPP or CPA

181

(Lobo et al., 2010b), it was found that selective blockade of the GABAergic neurotransmission

mediated by the D1R- and D2R-expressing MSNs respectively prevented reward and aversive

conditioning (Hikida et al., 2010). Together these findings indicate opposing roles for the D1R

and D2R neuronal systems in the NAc in modulating hedonic perception, with the D1R

facilitating reward and the D2R promoting aversion. Herein we additionally report a novel

mechanism by which dopaminergic transmission in the mesolimbic system could mediate

aversion: via the stimulation of the D1-D2 heteromer.

As aversive stimuli were found to reduce the tonic firing of VTA-NAc dopaminergic

neurons (Mirenowicz and Schultz, 1996; Ungless et al., 2004; Brischoux et al., 2009), it raised

the possibility that the attenuation of the neuronal activity of VTA-NAc dopaminergic neurons

may be an underlying mechanism of inducing aversions. Indeed, it was found that the

rostromedial tegmental nucleus (RMTg), which sends GABAergic projections to VTA

dopaminergic neurons, showed increased phasic activation and c-fos expression following the

presentation of aversive stimuli (Jhou et al., 2009a). Furthermore, optogenetic stimulation of

GABAergic neurons in the VTA resulted in CPA (Tan et al., 2012), indicating the GABA-

mediated inhibition of VTA dopaminergic neurons is responsible for aversion. Therefore, the

CPA induced by D1-D2 heteromer stimulation may be a consequence of increased GABAergic

activity in the VTA (Perreault et al., 2012a), which would subsequently reduce the firing of

VTA-NAc dopaminergic neurons, thereby resulting in an aversion.

The major GABAergic innervations onto VTA-NAc dopaminergic neurons originate

from the RMTg and local interneurons (Nair-Roberts et al., 2008; Jhou et al., 2009a). Notably,

the activity of these VTA-projecting GABAergic neurons is in turn modulated by excitatory

glutamatergic inputs from the lateral habenula (Lhb) (Jhou et al., 2009b; Omelchenko et al.,

182

2009; Lammel et al., 2012; Stamatakis and Stuber, 2012). As a result, stimulation of the Lhb

was shown to inhibit up to 90% of VTA dopaminergic neurons (Matsumoto and Hikosaka,

2007), whereas its lesioning enhanced dopamine release in the NAc (Lecourtier et al., 2008).

The potent inhibitory influence of the Lhb on VTA dopaminergic neurons thus makes it a

critical nucleus that could mediate aversion. Indeed, Lhb neurons were strongly excited by

aversive stimuli (Matsumoto and Hikosaka, 2009), and CPA was elicited by deep brain

stimulation (DBS) or optogenetic activation of Lhb neurons (Friedman et al., 2011; Lammel et

al., 2012). Interestingly, a recent study demonstrated that the activity of Lhb is modulated by a

unique population of neurons from the VTA that co-express both GABA and glutamate (Shabel

et al., 2014). While the co-expression of GABA and glutamate in the same subset of neurons is

not yet a universally accepted phenomenon, we have previously demonstrated that D1-D2

heteromer expressing neurons in the NAc also co-express both GABA and glutamate (Perreault

et al., 2012a), thereby making the D1-D2 heteromer also a highly possible modulator of Lhb

activity. Therefore, in addition to increasing GAD67 expression in the VTA, the D1-D2

heteromer may have also induced aversive perception via the modulation of Lhb activity. This

notion is supported by the finding that selective D1-D2 heteromer inactivation regulated c-fos

induction in the Lhb (Chapter IV, Perreault et al. 2015), however additional studies will need to

provide definitive evidence for direct or indirect projections from NAc to Lhb.

6.5 Significance and Conclusion

In summary, we have identified the novel dopamine receptor complex, the D1-D2

heteromer, as a negative modulator of brain reward function. We postulate that, on a hedonic

183

spectrum (Figure 8), the D1-D2 heteromer is tonically active at basal state in the absence of an

external rewarding stimulus to exert constant suppression of brain reward function to maintain

a neutral hedonic perception. Stimulation of the D1-D2 heteromer in the presence of external

rewarding stimuli such as psychostimulants and natural rewards will attenuate, whereas its

inactivation will enhance, the rewarding effect (positive hedonic value) induced by these

stimuli. In contrast, stimulation of the D1-D2 heteromer in the absence of an external

rewarding stimulus will result in an aversive state (negative hedonic value) and a depression-

like behavioural phenotype as a consequence of excessive suppression of brain reward function.

As the D1-D2 heteromer also modulated the neuronal activity of various nuclei that are

involved in brain reward function, including the NAc, prelimbic and infralimbic regions of the

mPFC, the orbitofrontal cortex, and the Lhb, these findings together suggest the D1-D2

heteromer may be a novel molecular substrate through which the mesolimbic dopamine system

could self-regulate the reward processes via homeostatic inhibition.

The identification of the D1-D2 heteromer as a negative modulator of brain reward

processes has profound implications for the understanding and treatment of neuropsychiatric

disorders characterized by a dysfunctional reward system, such as drug addiction and

depression. Using pharmacological agents that stimulate or inactivate the D1-D2 heteromer,

brain reward function could be respectively suppressed or enhanced to treat specific

pathological conditions accordingly. For instance, D1-D2 heteromer stimulation was shown to

abolish cocaine- and cue-induced reinstatement of drug self-administration, and thus could be a

potential method to prevent relapse to drug-seeking in human addicts. On the other hand, D1-

D2 heteromer inactivation was found to elicit rapid antidepressant-like activity in animals with

chronic stress-induced depressive behavioural phenotype, thus signifying its utility in treating

184

Figure 7: The proposed role of the D1-D2 heteromer in the modulation of brain reward

function. The exposure to psychostimulants or natural reward results in a positive hedonic

value to induce a euphoric feeling, whereas depressed patients have negative hedonic value and

thus are less sensitive to reward perception. The stimulation of the D1-D2 heteromer reduces

the hedonic value to promote depression- and anxiety-like behaviour, whereas its stimulation

increases the hedonic value to promote psychostimulant-induced behaviours. The ability of the

D1-D2 heteromer to exert bidirectional modulation of the brain reward function makes it a

novel therapeutic target for the treatment of drug addiction and depression.

Positive

Negative

Hed

on

ic S

pec

tru

m Natural Reward

Psychostimulant Reward

and Behaviour

Aversion

Depression

Anxiety

Homeostatic Point

D1-D2 Heteromer

Stimulation

D1-D2 Heteromer

Inactivation

D1-D2 Heteromer

Stimulation

D1-D2 Heteromer

Inactivation

185

patients with major depression. Since currently no effective intervention is available for drug

addiction or major depression, the D1-D2 heteromer may be a potential novel therapeutic target

for the treatment of these pathological conditions.

7. Future Directions

Our study thus far has implicated the D1-D2 heteromer in the negative modulation of

brain reward function, allowing its stimulation to suppress drug-seeking behaviours and its

inactivation to exert antidepressant-like activity. Nevertheless, additional studies are required

to further characterize the physiological role of the D1-D2 heteromer in reward- and

depression-related behaviours.

Although behavioural studies have clearly implicated the D1-D2 heteromer in reward-

seeking and depression-like behaviours, the precise molecular substrates through which the

D1-D2 heteromer signalling exerted its behavioural effects remain to be elucidated. It would be

imperative to determine the exact involvement of proteins downstream from the D1-D2

heteromer signalling, such as BDNF, DARPP-32, Cdk5, and GAD67, on the behavioural

effects of the D1-D2 heteromer as described in the current study. This can be achieved using

adenovirus-mediated knock-down of these proteins in specific brain regions, namely the NAc

and the VTA, and then examine reward- and depression-related behavioural effects of D1-D2

heteromer stimulation.

To further support the notion that the D1-D2 heteromer is a novel receptor complex that

functions to modulate brain reward function, studies should be conducted that assess the effects

186

of D1-D2 heteromer stimulation and inactivation on intracranial self-stimulation behaviour, a

well-validated behavioural model that directly reflects the activity of brain reward circuitry.

In addition, given that D1R- and D2R-expressing MSNs in the NAc have differential

effects on reward-seeking behaviours, the effect of direct stimulation of D1R/D2R co-

expressing MSNs should also be examined to establish the physiological function of this

unique subset of neurons in the NAc. This question can be easily addressed with the aid of

recent advancement in molecular techniques, such as optogenetics and designer receptors

exclusively activated by designer drugs (DREADDs). On the other hand, the effect of silencing

D1R/D2R co-expressing MSNs on brain reward function can be examined by injecting a

doubled-floxed Cre-dependent adeno-associated virus type 2 (AAV2) vector that expresses

D1R siRNA under the control of D1 receptor promoter into D2R-Cre mice, and behavioural

analyses will be performed on these mice.

Lastly, the projection targets of the D1R/D2R co-expressing MSNs remain largely

unknown, and it would be important to establish the network of nuclei that are interconnected

by this unique subset of NAc MSNs in order to further deduce how the D1-D2 heteromer

exerts its behavioural effects. In this respect, anterograde or retrograde tracing experiments

could be performed to determine the potential projection targets of the D1R/D2R co-expressing

MSNs.

The completion of these studies will allow us to firmly establish the role of the D1-D2

heteromer in the modulation of brain reward function, which in turn would make this novel

receptor complex an attractive therapeutic target for neuropsychiatric disorders that are

characterized by a dysfunctional reward system, such as drug addiction and major depression.

187

Nevertheless, as currently no pharmacological agent yet exists that could selectively stimulate

the D1-D2 heteromer, it would be imperative to conduct a comprehensive screening of existing

dopamine receptor agonists or use bioinformatics and in silico modeling strategies to search for

a selective agonist for this receptor complex before its modulatory effect on brain reward

function can be utilized in clinical settings.

Furthermore, in addition to its potential therapeutic benefits for the treatment of drug

addiction and major depression, studies have additionally suggested possible physiological role

for the D1-D2 heteromer in other neuropsychiatric disorders. For instance, the ability of D1-D2

heteromer stimulation to increase the production of BDNF (Hasbi et al., 2009; Perreault et al.,

2012a), which is crucial for neuronal growth and maturation, may have great implications in

the management of neurodegenerative disorders such as Alzheimer’s disease. In addition, the

D1-D2 heteromer has also been shown to signal through GSK-3 (Perreault et al., 2014), a

important mediator of the therapeutic effect of lithium for the the treatment of schizophrenia

(Beaulieu et al., 2004). Comprehensive behavioural and biochemical studies would be required

to explore the potential therapeutic benefits of the D1-D2 heteromer beyond drug addiction and

depression as suggested by these early studies.

188

8. References

Ackerman JM, White FJ (1992) Decreased activity of rat A10 dopamine neurons following

withdrawal from repeated cocaine. Eur J Pharmacol 218:171–173.

Adachi K, Ikeda H, Hasegawa M, Nakamura S, Waddington JL, Koshikawa N (1999) SK&F

83959 and non-cyclase-coupled dopamine D1-like receptors in jaw movements via

dopamine D1-like/D2-like receptor synergism. Eur J Pharmacol 367:143–149.

Aguilar MA, Rodríguez-Arias M, Miñarro J (2009) Neurobiological mechanisms of the

reinstatement of drug-conditioned place preference. Brain Res Rev 59:253–277.

Ahmed SH, Koob GF (1998) Transition from moderate to excessive drug intake: change in

hedonic set point. Science 282:298–300.

Ahmed SH, Koob GF (1999) Long-lasting increase in the set point for cocaine self-

administration after escalation in rats. Psychopharmacology (Berl) 146:303–312.

Albert PR, Vahid-Ansari F, Luckhart C (2014) Serotonin-prefrontal cortical circuitry in anxiety

and depression phenotypes: pivotal role of pre- and post-synaptic 5-HT1A receptor

expression. Front Behav Neurosci 8:199.

Alexander GM, Rogan SC, Abbas AI, Armbruster BN, Pei Y, Allen JA, Nonneman RJ,

Hartmann J, Moy SS, Nicolelis MA, McNamara JO, Roth BL (2009) Remote control of

neuronal activity in transgenic mice expressing evolved G protein-coupled receptors.

Neuron 63:27–39.

Anagnostaras SG, Robinson TE (1996) Sensitization to the psychomotor stimulant effects of

amphetamine: modulation by associative learning. Behav Neurosci 110:1397–1414.

Anden NE, Carlsson A, Dahlstroem A, Fuxe K, Hillarp NA, Larsson K (1964) Demonstration

and mapping out of nigro-neostriatal dopamine neurons. Life Sci 3:523–530.

Andersen PH, Gingrich JA, Bates MD, Dearry A, Falardeau P, Senogles SE, Caron MG (1990)

Dopamine receptor subtypes: beyond the D1/D2 classification. Trends Pharmacol Sci

11:231–236.

Anderson SM, Famous KR, Sadri-Vakili G, Kumaresan V, Schmidt HD, Bass CE, Terwilliger

EF, Cha J-HJ, Pierce RC (2008) CaMKII: a biochemical bridge linking accumbens

dopamine and glutamate systems in cocaine seeking. Nat Neurosci 11:344–353.

Andringa G, Drukarch B, Leysen JE, Cools AR, Stoof JC (1999) The alleged dopamine D1

receptor agonist SKF 83959 is a dopamine D1 receptor antagonist in primate cells and

interacts with other receptors. Eur J Pharmacol 364:33–41.

189

Anstrom KK, Woodward DJ (2005) Restraint increases dopaminergic burst firing in awake rats.

Neuropsychopharmacology 30:1832–1840.

Aston-Jones G, Smith RJ, Sartor GC, Moorman DE, Massi L, Tahsili-Fahadan P, Richardson

KA (2010) Lateral hypothalamic orexin/hypocretin neurons: A role in reward-seeking and

addiction. Brain Res 1314:74–90.

Bagdy G, Graf M, Anheuer ZE, Modos EA, Kantor S (2002) Anxiety-like effects induced by

acute fluoxetine, sertraline or m-CPP treatment are reversed by pretreatment with the 5-

HT2C receptor antagonist SB-242084 but not the 5-HT1A receptor antagonist WAY-

100635. Int J Neuropsychopharmacol 4:399–408.

Bahi A, Boyer F, Dreyer JL (2008) Role of accumbens BDNF and TrkB in cocaine-induced

psychomotor sensitization, conditioned-place preference, and reinstatement in rats.

Psychopharmacology (Berl) 199:169–182.

Baik J-H (2013) Dopamine signaling in reward-related behaviors. Front Neural Circuits 7:152.

Baker DA, Khroyan T V, O’Dell LE, Fuchs RA, Neisewander JL (1996) Differential effects of

intra-accumbens sulpiride on cocaine-induced locomotion and conditioned place

preference. J Pharmacol Exp Ther 279:392–401.

Bambico FR, Nguyen N-T, Gobbi G (2009) Decline in serotonergic firing activity and

desensitization of 5-HT1A autoreceptors after chronic unpredictable stress. Eur

Neuropsychopharmacol 19:215–228.

Barr AM, Markou A (2005a) Psychostimulant withdrawal as an inducing condition in animal

models of depression. Neurosci Biobehav Rev 29:675–706.

Barr AM, Markou A (2005b) Psychostimulant withdrawal as an inducing condition in animal

models of depression. Neurosci Biobehav Rev 29:675–706.

Barr AM, Phillips AG (2002) Increased successive negative contrast in rats withdrawn from an

escalating-dose schedule of D-amphetamine. Pharmacol Biochem Behav 71:293–299.

Bateup HS, Svenningsson P, Kuroiwa M, Gong S, Nishi A, Heintz N, Greengard P (2008) Cell

type-specific regulation of DARPP-32 phosphorylation by psychostimulant and

antipsychotic drugs. Nat Neurosci 11:932–939.

Beaulieu J-M, Sotnikova TD, Gainetdinov RR, Caron MG (2006) Paradoxical Striatal Cellular

Signaling Responses to Psychostimulants in Hyperactive Mice. J Biol Chem 281:32072–

32080.

Beaulieu J-M, Sotnikova TD, Yao W-D, Kockeritz L, Woodgett JR, Gainetdinov RR, Caron

MG (2004) Lithium antagonizes dopamine-dependent behaviors mediated by an

190

AKT/glycogen synthase kinase 3 signaling cascade. Proc Natl Acad Sci U S A 101:5099–

5104.

Bello EP, Mateo Y, Gelman DM, Noaín D, Shin JH, Low MJ, Alvarez VA, Lovinger DM,

Rubinstein M (2011) Cocaine supersensitivity and enhanced motivation for reward in

mice lacking dopamine D2 autoreceptors. Nat Neurosci 14:1033–1038.

Benavides DR, Quinn JJ, Zhong P, Hawasli AH, DiLeone RJ, Kansy JW, Olausson P, Yan Z,

Taylor JR, Bibb JA (2007) Cdk5 modulates cocaine reward, motivation, and striatal

neuron excitability. J Neurosci 27:12967–12976.

Berglind WJ, See RE, Fuchs RA, Ghee SM, Whitfield TW, Miller SW, McGinty JF (2007) A

BDNF infusion into the medial prefrontal cortex suppresses cocaine seeking in rats. Eur J

Neurosci 26:757–766.

Berhow MT, Hiroi N, Nestler EJ (1996) Regulation of ERK (extracellular signal regulated

kinase), part of the neurotrophin signal transduction cascade, in the rat mesolimbic

dopamine system by chronic exposure to morphine or cocaine. J Neurosci 16:4707–4715.

Berridge KC (2007) The debate over dopamine’s role in reward: The case for incentive

salience. Psychopharmacology (Berl) 191:391–431.

Berton O, McClung CA, Dileone RJ, Krishnan V, Renthal W, Russo SJ, Graham D, Tsankova

NM, Bolanos CA, Rios M, Monteggia LM, Self DW, Nestler EJ (2006) Essential role of

BDNF in the mesolimbic dopamine pathway in social defeat stress. Science 311:864–868.

Bertran-Gonzalez J, Bosch C, Maroteaux M, Matamales M, Hervé D, Valjent E, Girault J-A

(2008) Opposing patterns of signaling activation in dopamine D1 and D2 receptor-

expressing striatal neurons in response to cocaine and haloperidol. J Neurosci 28:5671–

5685.

Bertran-Gonzalez J, Hervé D, Girault J-A, Valjent E (2010) What is the Degree of Segregation

between Striatonigral and Striatopallidal Projections? Front Neuroanat 4.

Bessa JM, Mesquita AR, Oliveira M, Pêgo JM, Cerqueira JJ, Palha JA, Almeida OFX, Sousa N

(2009) A trans-dimensional approach to the behavioral aspects of depression. Front Behav

Neurosci 3:1.

Bessa JM, Morais M, Marques F, Pinto L, Palha JA, Almeida OFX, Sousa N (2013) Stress-

induced anhedonia is associated with hypertrophy of medium spiny neurons of the nucleus

accumbens. Transl Psychiatry 3:e266.

Bibb JA, Snyder GL, Nishi A, Yan Z, Meijer L, Fienberg AA, Tsai LH, Kwon YT, Girault JA,

Czernik AJ, Huganir RL, Hemmings HC, Nairn AC, Greengard P (1999) Phosphorylation

of DARPP-32 by Cdk5 modulates dopamine signalling in neurons. Nature 402:669–671.

191

Bodnoff SR, Suranyi-Cadotte B, Quirion R, Meaney MJ (1989) A comparison of the effects of

diazepam versus several typical and atypical anti-depressant drugs in an animal model of

anxiety. Psychopharmacology (Berl) 97:277–279.

Bogush A, Pedrini S, Pelta-Heller J, Chan T, Yang Q, Mao Z, Sluzas E, Gieringer T, Ehrlich

ME (2007) AKT and CDK5/p35 mediate brain-derived neurotrophic factor induction of

DARPP-32 in medium size spiny neurons in vitro. J Biol Chem 282:7352–7359.

Borgland SL, Malenka RC, Bonci A (2004) Acute and chronic cocaine-induced potentiation of

synaptic strength in the ventral tegmental area: electrophysiological and behavioral

correlates in individual rats. J Neurosci 24:7482–7490.

Bradberry CW, Roth RH (1989) Cocaine increases extracellular dopamine in rat nucleus

accumbens and ventral tegmental area as shown by in vivo microdialysis. Neurosci Lett

103:97–102.

Brami-Cherrier K, Valjent E, Hervé D, Darragh J, Corvol J-C, Pages C, Arthur SJ, Simon AJ,

Girault J-A, Caboche J (2005) Parsing molecular and behavioral effects of cocaine in

mitogen- and stress-activated protein kinase-1-deficient mice. J Neurosci 25:11444–11454.

Brebner K, Phelan R, Roberts DCS (2000) Intra-VTA baclofen attenuates cocaine self-

administration on a progressive ratio schedule of reinforcement. Pharmacol Biochem

Behav 66:857–862.

Brischoux F, Chakraborty S, Brierley DI, Ungless MA (2009) Phasic excitation of dopamine

neurons in ventral VTA by noxious stimuli. Proc Natl Acad Sci U S A 106:4894–4899.

Brog JS, Salyapongse A, Deutch AY, Zahm DS (1993) The patterns of afferent innervation of

the core and shell in the “accumbens” part of the rat ventral striatum:

immunohistochemical detection of retrogradely transported fluoro-gold. J Comp Neurol

338:255–278.

Calabresi P, Gubellini P, Centonze D, Picconi B, Bernardi G, Chergui K, Svenningsson P,

Fienberg AA, Greengard P (2000) Dopamine and cAMP-regulated phosphoprotein 32

kDa controls both striatal long-term depression and long-term potentiation, opposing

forms of synaptic plasticity. J Neurosci 20:8443–8451.

Cameron CM, Wightman RM, Carelli RM (2014) Dynamics of rapid dopamine release in the

nucleus accumbens during goal-directed behaviors for cocaine versus natural rewards.

Neuropharmacology 86:319–328.

Capriles N, Rodaros D, Sorge RE, Stewart J (2003) A role for the prefrontal cortex in stress-

and cocaine-induced reinstatement of cocaine seeking in rats. Psychopharmacology (Berl)

168:66–74.

192

Carr DB, Sesack SR (2000) Projections from the rat prefrontal cortex to the ventral tegmental

area: target specificity in the synaptic associations with mesoaccumbens and mesocortical

neurons. J Neurosci 20:3864–3873.

Carter RM, Wittchen HU, Pfister H, Kessler RC (2001) One-year prevalence of subthreshold

and threshold DSM-IV generalized anxiety disorder in a nationally representative sample.

Depress Anxiety 13:78–88.

Cervo L, Samanin R (1995) Effects of dopaminergic and glutamatergic receptor antagonists on

the acquisition and expression of cocaine conditioning place preference. Brain Res

673:242–250.

Chaudhury D et al. (2013) Rapid regulation of depression-related behaviours by control of

midbrain dopamine neurons. Nature 493:532–536.

Chen BT, Bowers MS, Martin M, Hopf FW, Guillory AM, Carelli RM, Chou JK, Bonci A

(2008) Cocaine but not natural reward self-administration nor passive cocaine infusion

produces persistent LTP in the VTA. Neuron 59:288–297.

Chen BT, Hopf FW, Bonci A (2010) Synaptic plasticity in the mesolimbic system: Therapeutic

implications for substance abuse. Ann N Y Acad Sci 1187:129–139.

Chien EYT, Liu W, Zhao Q, Katritch V, Han GW, Hanson MA, Shi L, Newman AH, Javitch

JA, Cherezov V, Stevens RC (2010) Structure of the human dopamine D3 receptor in

complex with a D2/D3 selective antagonist. Science 330:1091–1095.

Chun LS, Free RB, Doyle TB, Huang X-P, Rankin ML, Sibley DR (2013) D1-D2 dopamine

receptor synergy promotes calcium signaling via multiple mechanisms. Mol Pharmacol

84:190–200.

Ciccocioppo R, Sanna PP, Weiss F (2001) Cocaine-predictive stimulus induces drug-seeking

behavior and neural activation in limbic brain regions after multiple months of abstinence:

reversal by D(1) antagonists. Proc Natl Acad Sci U S A 98:1976–1981.

Coffey SF, Dansky BS, Carrigan MH, Brady KT (2000) Acute and protracted cocaine

abstinence in an outpatient population: a prospective study of mood, sleep and withdrawal

symptoms. Drug Alcohol Depend 59:277–286.

Comps-Agrar L, Kniazeff J, Nørskov-Lauritsen L, Maurel D, Gassmann M, Gregor N, Prézeau

L, Bettler B, Durroux T, Trinquet E, Pin J-P (2011) The oligomeric state sets GABA B

receptor signalling efficacy. EMBO J 30:2336–2349.

Conway KP, Compton WM, Stinson FS, Grant BF (2006) Lifetime comorbidity of DSM-IV

mood and anxiety disorders and specific drug use disorders: Results from the National

Epidemiologic Survey on Alcohol and Related Conditions. J Clin Psychiatry 67:247–257.

193

Cools AR, Lubbers L, Van Oosten R V., Andringa G (2002) SKF 83959 is an antagonist of

dopamine D1-like receptors in the prefrontal cortex and nucleus accumbens: A key to its

antiparkinsonian effect in animals? Neuropharmacology 42:237–245.

Cooper SJ, Francis J, Al-Naser H, Barber D (1992) Evidence for dopamine D-1 receptor-

mediated facilitatory and inhibitory effects on feeding behaviour in rats. J

Psychopharmacol 6:27–33.

Corominas M, Roncero C, Ribases M, Castells X, Casas M (2007) Brain-derived neurotrophic

factor and its intracellular signaling pathways in cocaine addiction. Neuropsychobiology

55:2–13.

Coultrap SJ, Barcomb K, Bayer KU (2012) A significant but rather mild contribution of T286

autophosphorylation to Ca2+/CaM-stimulated CaMKII activity. PLoS One 7:e37176.

Culpepper L (2010) Why do you need to move beyond first-line therapy for major depression?

J Clin Psychiatry 71:4–9.

D’Souza MS, Markou A (2010) Neural substrates of psychostimulant withdrawal-induced

anhedonia. Curr Top Behav Neurosci 3:119–178.

Dahlstroem A, Fuxe K (1964) Evidence for the existence of monoaminecontaining neurons in

the central nervous system. I. Demonstration of monoamines in the cell bodies of brain

stem neurons. Acta Physiol Scand Suppl:SUPPL 232:1–55.

Dallvechia-Adams S, Smith Y, Kuhar MJ (2001) CART peptide-immunoreactive projection

from the nucleus accumbens targets substantia nigra pars reticulata neurons in the rat. J

Comp Neurol 434:29–39.

Davis L, Uezato A, Newell JM, Frazier E (2008) Major depression and comorbid substance use

disorders. Curr Opin Psychiatry 21:14–18.

De Lean A, Stadel JM, Lefkowitz RJ (1980) A ternary complex model explains the agonist-

specific binding properties of the adenylate cyclase-coupled beta-adrenergic receptor. J

Biol Chem 255:7108–7117.

De Mei C, Ramos M, Iitaka C, Borrelli E (2009) Getting specialized: presynaptic and

postsynaptic dopamine D2 receptors. Curr Opin Pharmacol 9:53–58.

De Vries TJ (1998) Drug-induced reinstatement of heroin- and cocaine-seeking behaviour

following long-term extinction is associated with expression of behavioural sensitization.

Eur J Neurosci 10:3565–3571.

Dearry A, Gingrich JA, Falardeau P, Fremeau RT, Bates MD, Caron MG (1990) Molecular

cloning and expression of the gene for a human D1 dopamine receptor. Nature 347:72–76.

194

Dehaene S, Changeux JP (2000) Reward-dependent learning in neuronal networks for planning

and decision making. Prog Brain Res 126:217–229.

Del-Fava F, Hasue RH, Ferreira JGP, Shammah-Lagnado SJ (2007) Efferent connections of the

rostral linear nucleus of the ventral tegmental area in the rat. Neuroscience 145:1059–

1076.

Deng YP, Lei WL, Reiner A (2006) Differential perikaryal localization in rats of D1 and D2

dopamine receptors on striatal projection neuron types identified by retrograde labeling. J

Chem Neuroanat 32:101–116.

Depoortere RY, Li DH, Lane JD, Emmett-Oglesby MW (1993) Parameters of self-

administration of cocaine in rats under a progressive-ratio schedule. Pharmacol Biochem

Behav 45:539–548.

Der-Avakian A, Markou A (2012) The neurobiology of anhedonia and other reward-related

deficits. Trends Neurosci 35:68–77.

Desdouits F, Siciliano JC, Greengard P, Girault JA (1995) Dopamine- and cAMP-regulated

phosphoprotein DARPP-32: phosphorylation of Ser-137 by casein kinase I inhibits

dephosphorylation of Thr-34 by calcineurin. Proc Natl Acad Sci U S A 92:2682–2685.

Deupi X, Olivella M, Govaerts C, Ballesteros JA, Campillo M, Pardo L (2004) Ser and Thr

residues modulate the conformation of pro-kinked transmembrane alpha-helices. Biophys

J 86:105–115.

Deveney AM, Waddington JL (1995) Pharmacological characterization of behavioural

responses to SK&F 83959 in relation to “D1-like” dopamine receptors not linked to

adenylyl cyclase. Br J Pharmacol 116:2120–2126.

Dhanasekaran N, Dermott JM (1996) Signaling by the G12 class of G proteins. Cell Signal

8:235–245.

Di Ciano P, Everitt BJ (2003) The GABA(B) receptor agonist baclofen attenuates cocaine- and

heroin-seeking behavior by rats. Neuropsychopharmacology 28:510–518.

Dohlman HG, Thorner J (1997) RGS proteins and signaling by heterotrimeric G proteins. J

Biol Chem 272:3871–3874.

Downes RP, Waddington JL (1993) Grooming and vacuous chewing induced by SK&F 83959,

an agonist of dopamine “D1-like” receptors that inhibits dopamine-sensitive adenylyl

cyclase. Eur J Pharmacol 234:135–136.

Dranovsky A, Hen R (2006) Hippocampal neurogenesis: regulation by stress and

antidepressants. Biol Psychiatry 59:1136–1143.

195

Dulawa SC, Hen R (2005) Recent advances in animal models of chronic antidepressant effects:

The novelty-induced hypophagia test. Neurosci Biobehav Rev 29:771–783.

Duman RS (2014) Pathophysiology of depression and innovative treatments: remodeling

glutamatergic synaptic connections. Dialogues Clin Neurosci 16:11–27.

Duman RS, Monteggia LM (2006) A neurotrophic model for stress-related mood disorders.

Biol Psychiatry 59:1116–1127.

Ehringer H, Hornykiewicz O (1960) Distribution of noradrenaline and dopamine (3-

hydroxytyramine) in the human brain and their behavior in diseases of the extrapyramidal

system. Klin Wochenschr 38:1236–1239.

Eilam D, Golani I (1989) Home base behavior of rats (Rattus norvegicus) exploring a novel

environment. Behav Brain Res 34:199–211.

Eisch AJ, Bolaños C a, de Wit J, Simonak RD, Pudiak CM, Barrot M, Verhaagen J, Nestler EJ

(2003) Brain-derived neurotrophic factor in the ventral midbrain–nucleus accumbens

pathway: a role in depression. Biol Psychiatry 54:994–1005.

El Moustaine D, Granier S, Doumazane E, Scholler P, Rahmeh R, Bron P, Mouillac B, Banères

J-L, Rondard P, Pin J-P (2012) Distinct roles of metabotropic glutamate receptor

dimerization in agonist activation and G-protein coupling. Proc Natl Acad Sci U S A

109:16342–16347.

Fanous S, Terwilliger EF, Hammer RP, Nikulina EM (2011) Viral depletion of VTA BDNF in

rats modulates social behavior, consequences of intermittent social defeat stress, and long-

term weight regulation. Neurosci Lett 502:192–196.

Farnsworth CL, Freshney NW, Rosen LB, Ghosh A, Greenberg ME, Feig LA (1995) Calcium

activation of Ras mediated by neuronal exchange factor Ras-GRF. Nature 376:524–527.

Felder CC, Jose PA, Axelrod J (1989) The dopamine-1 agonist, SKF 82526, stimulates

phospholipase-C activity independent of adenylate cyclase. J Pharmacol Exp Ther

248:171–175.

Fienberg AA et al. (1998) DARPP-32: regulator of the efficacy of dopaminergic

neurotransmission. Science 281:838–842.

Filip M, Thomas ML, Cunningham KA (2000) Dopamine D5 receptors in nucleus accumbens

contribute to the detection of cocaine in rats. J Neurosci 20:RC98.

Fleckenstein AE, Volz TJ, Riddle EL, Gibb JW, Hanson GR (2007) New insights into the

mechanism of action of amphetamines. Annu Rev Pharmacol Toxicol 47:681–698.

196

Fletcher PJ, Chintoh AF, Sinyard J, Higgins GA (2004) Injection of the 5-HT2C receptor

agonist Ro60-0175 into the ventral tegmental area reduces cocaine-induced locomotor

activity and cocaine self-administration. Neuropsychopharmacology 29:308–318.

Flores-Hernandez J, Hernandez S, Snyder GL, Yan Z, Fienberg AA, Moss SJ, Greengard P,

Surmeier DJ (2000) D(1) dopamine receptor activation reduces GABA(A) receptor

currents in neostriatal neurons through a PKA/DARPP-32/PP1 signaling cascade. J

Neurophysiol 83:2996–3004.

Ford CP (2014) The role of D2-autoreceptors in regulating dopamine neuron activity and

transmission. Neuroscience 282C:13–22.

Frederick AL, Yano H, Trifilieff P, Vishwasrao HD, Biezonski D, Mészáros J, Urizar E, Sibley

DR, Kellendonk C, Sonntag KC, Graham DL, Colbran RJ, Stanwood GD, Javitch JA

(2015) Evidence against dopamine D1/D2 receptor heteromers. Mol Psychiatry.

Freedman NJ, Lefkowitz RJ (1996) Desensitization of G protein-coupled receptors. Recent

Prog Horm Res 51:319–351; discussion 352–353.

Freire E (1998) Statistical thermodynamic linkage between conformational and binding

equilibria. Adv Protein Chem 51:255–279.

Friedman A, Lax E, Dikshtein Y, Abraham L, Flaumenhaft Y, Sudai E, Ben-Tzion M, Ami-Ad

L, Yaka R, Yadid G (2010) Electrical stimulation of the lateral habenula produces

enduring inhibitory effect on cocaine seeking behavior. Neuropharmacology 59:452–459.

Friedman A, Lax E, Dikshtein Y, Abraham L, Flaumenhaft Y, Sudai E, Ben-Tzion M, Yadid G

(2011) Electrical stimulation of the lateral habenula produces an inhibitory effect on

sucrose self-administration. Neuropharmacology 60:381–387.

Friedman AK, Walsh JJ, Juarez B, Ku SM, Chaudhury D, Wang J, Li X, Dietz DM, Pan N,

Vialou VF, Neve RL, Yue Z, Han M-H (2014) Enhancing depression mechanisms in

midbrain dopamine neurons achieves homeostatic resilience. Science 344:313–319.

Friedman E, Jin LQ, Cai GP, Hollon TR, Drago J, Sibley DR, Wang HY (1997) D1-like

dopaminergic activation of phosphoinositide hydrolysis is independent of D1A dopamine

receptors: evidence from D1A knockout mice. Mol Pharmacol 51:6–11.

Fuchs RA, Evans KA, Parker MP, See RE (2004) Differential involvement of orbitofrontal

cortex subregions in conditioned cue-induced and cocaine-primed reinstatement of

cocaine seeking in rats. J Neurosci 24:6600–6610.

Fuchs RA, Ramirez DR, Bell GH (2008) Nucleus accumbens shell and core involvement in

drug context-induced reinstatement of cocaine seeking in rats. Psychopharmacology (Berl)

200:545–556.

197

Gainetdinov RR, Bohn LM, Sotnikova TD, Cyr M, Laakso A, Macrae AD, Torres GE, Kim

KM, Lefkowitz RJ, Caron MG, Premont RT (2003) Dopaminergic supersensitivity in G

protein-coupled receptor kinase 6-deficient mice. Neuron 38:291–303.

Gambarana C, Masi F, Leggio B, Grappi S, Nanni G, Scheggi S, De Montis MG, Tagliamonte

A (2003) Acquisition of a palatable-food-sustained appetitive behavior in satiated rats is

dependent on the dopaminergic response to this food in limbic areas. Neuroscience

121:179–187.

Gambarana C, Masi F, Tagliamonte A, Scheggi S, Ghiglieri O, De Montis MG (1999) A

chronic stress that impairs reactivity in rats also decreases dopaminergic transmission in

the nucleus accumbens: a microdialysis study. J Neurochem 72:2039–2046.

Gangarossa G, Espallergues J, de Kerchove d’Exaerde A, El Mestikawy S, Gerfen CR, Hervé

D, Girault J-A, Valjent E (2013a) Distribution and compartmental organization of

GABAergic medium-sized spiny neurons in the mouse nucleus accumbens. Front Neural

Circuits 7:22.

Gangarossa G, Espallergues J, Mailly P, De Bundel D, de Kerchove d’Exaerde A, Hervé D,

Girault J-A, Valjent E, Krieger P (2013b) Spatial distribution of D1R- and D2R-

expressing medium-sized spiny neurons differs along the rostro-caudal axis of the mouse

dorsal striatum. Front Neural Circuits 7:124.

Gawin FH (1991) Cocaine addiction: psychology and neurophysiology. Science 251:1580–

1586.

Geisler S, Derst C, Veh RW, Zahm DS (2007) Glutamatergic afferents of the ventral tegmental

area in the rat. J Neurosci 27:5730–5743.

Geisler S, Zahm DS (2005) Afferents of the ventral tegmental area in the rat-anatomical

substratum for integrative functions. J Comp Neurol 490:270–294.

George SR, O’Dowd BF, Lee SP (2002) G-protein-coupled receptor oligomerization and its

potential for drug discovery. Nat Rev Drug Discov 1:808–820.

Gerfen CR, Engber TM, Mahan LC, Susel Z, Chase TN, Monsma FJ, Sibley DR (1990) D1

and D2 dopamine receptor-regulated gene expression of striatonigral and striatopallidal

neurons. Science 250:1429–1432.

Gerfen CR, Surmeier DJ (2011) Modulation of striatal projection systems by dopamine. Annu

Rev Neurosci 34:441–466.

Ghitza UE, Zhai H, Wu P, Airavaara M, Shaham Y, Lu L (2010) Role of BDNF and GDNF in

drug reward and relapse: A review. Neurosci Biobehav Rev 35:157–171.

198

Gilman AG (1987) G proteins: transducers of receptor-generated signals. Annu Rev Biochem

56:615–649.

Gingrich JA, Caron MG (1993) Recent Advances in the Molecular Biology of Dopamine

Receptors. Annu Rev Neurosci 16:299–321.

Girault JA, Hemmings HC, Williams KR, Nairn AC, Greengard P (1989) Phosphorylation of

DARPP-32, a dopamine- and cAMP-regulated phosphoprotein, by casein kinase II. J Biol

Chem 264:21748–21759.

Giros B, Martres MP, Pilon C, Sokoloff P, Schwartz JC (1991) Shorter variants of the D3

dopamine receptor produced through various patterns of alternative splicing. Biochem

Biophys Res Commun 176:1584–1592.

Gomes I, Fujita W, Chandrakala M V, Devi LA (2013) Disease-specific heteromerization of G-

protein-coupled receptors that target drugs of abuse. Prog Mol Biol Transl Sci 117:207–

265.

Gonzalez A, Cordomí A, Matsoukas M, Zachmann J, Pardo L (2014) Modeling of G protein-

coupled receptors using crystal structures: from monomers to signaling complexes. Adv

Exp Med Biol 796:15–33.

Grace AA, Floresco SB, Goto Y, Lodge DJ (2007) Regulation of firing of dopaminergic

neurons and control of goal-directed behaviors. Trends Neurosci 30:220–227.

Graham DL, Edwards S, Bachtell RK, DiLeone RJ, Rios M, Self DW (2007a) Dynamic BDNF

activity in nucleus accumbens with cocaine use increases self-administration and relapse.

Nat Neurosci 10:1029–1037.

Graham DL, Edwards S, Bachtell RK, DiLeone RJ, Rios M, Self DW (2007b) Dynamic BDNF

activity in nucleus accumbens with cocaine use increases self-administration and relapse.

Nat Neurosci 10:1029–1037.

Grimm JW, Lu L, Hayashi T, Hope BT, Su T-P, Shaham Y (2003) Time-dependent increases

in brain-derived neurotrophic factor protein levels within the mesolimbic dopamine

system after withdrawal from cocaine: implications for incubation of cocaine craving. J

Neurosci 23:742–747.

Groenewegen HJ, Vermeulen-Van der Zee E, te Kortschot A, Witter MP (1987) Organization

of the projections from the subiculum to the ventral striatum in the rat. A study using

anterograde transport of Phaseolus vulgaris leucoagglutinin. Neuroscience 23:103–120.

Groenewegen HJ, Wright CI, Beijer A V (1996) The nucleus accumbens: gateway for limbic

structures to reach the motor system? Prog Brain Res 107:485–511.

199

Grosshans M, Loeber S, Kiefer F (2011) Implications from addiction research towards the

understanding and treatment of obesity. Addict Biol 16:189–198.

Gruber AJ, Hussain RJ, O’Donnell P (2009) The nucleus accumbens: a switchboard for goal-

directed behaviors. PLoS One 4:e5062.

Guo L, Zhao J, Jin G, Zhao B, Wang G, Zhang A, Zhen X (2013) SKF83959 Is a Potent

Allosteric Modulator of Sigma-1 Receptor. Mol Pharmacol 83:577–586.

Guo W, Urizar E, Kralikova M, Mobarec JC, Shi L, Filizola M, Javitch JA (2008) Dopamine

D2 receptors form higher order oligomers at physiological expression levels. EMBO J

27:2293–2304.

Haber SN, Fudge JL, McFarland NR (2000) Striatonigrostriatal pathways in primates form an

ascending spiral from the shell to the dorsolateral striatum. J Neurosci 20:2369–2382.

Haber SN, Lynd E, Klein C, Groenewegen HJ (1990) Topographic organization of the ventral

striatal efferent projections in the rhesus monkey: an anterograde tracing study. J Comp

Neurol 293:282–298.

Hajnal A, Norgren R (2001) Accumbens dopamine mechanisms in sucrose intake. Brain Res

904:76–84.

Hall DA, Stanis JJ, Marquez Avila H, Gulley JM (2008) A comparison of amphetamine- and

methamphetamine-induced locomotor activity in rats: Evidence for qualitative differences

in behavior. Psychopharmacology (Berl) 195:469–478.

Halpern CH, Tekriwal A, Santollo J, Keating JG, Wolf JA, Daniels D, Bale TL (2013)

Amelioration of binge eating by nucleus accumbens shell deep brain stimulation in mice

involves D2 receptor modulation. J Neurosci 33:7122–7129.

Hanson MA, Roth CB, Jo E, Griffith MT, Scott FL, Reinhart G, Desale H, Clemons B,

Cahalan SM, Schuerer SC, Sanna MG, Han GW, Kuhn P, Rosen H, Stevens RC (2012)

Crystal Structure of a Lipid G Protein-Coupled Receptor. Science (80- ) 335:851–855.

Harrison MB, Wiley RG, Wooten GF (1990) Selective localization of striatal D1 receptors to

striatonigral neurons. Brain Res 528:317–322.

Hasbi A, Fan T, Alijaniaram M, Nguyen T, Perreault ML, O’Dowd BF, George SR (2009)

Calcium signaling cascade links dopamine D1-D2 receptor heteromer to striatal BDNF

production and neuronal growth. Proc Natl Acad Sci U S A 106:21377–21382.

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 10:93–99.

200

Hasbi A, Perreault ML, Shen MYF, Zhang L, To R, Fan T, Nguyen T, Ji X, O’Dowd BF,

George SR (2014) A peptide targeting an interaction interface disrupts the dopamine D1-

D2 receptor heteromer to block signaling and function in vitro and in vivo: Effective

selective antagonism. FASEB 28:4806–4820.

Heimer L, Zahm DS, Churchill L, Kalivas PW, Wohltmann C (1991) Specificity in the

projection patterns of accumbal core and shell in the rat. Neuroscience 41:89–125.

Hemmings HC, Greengard P, Tung HY, Cohen P DARPP-32, a dopamine-regulated neuronal

phosphoprotein, is a potent inhibitor of protein phosphatase-1. Nature 310:503–505.

Henry DJ, Greene MA, White FJ (1989) Electrophysiological effects of cocaine in the

mesoaccumbens dopamine system: repeated administration. J Pharmacol Exp Ther

251:833–839.

Henry DJ, White FJ (1991) Repeated cocaine administration causes persistent enhancement of

D1 dopamine receptor sensitivity within the rat nucleus accumbens. J Pharmacol Exp Ther

258:882–890.

Hikida T, Kimura K, Wada N, Funabiki K, Nakanishi Shigetada S (2010) Distinct Roles of

Synaptic Transmission in Direct and Indirect Striatal Pathways to Reward and Aversive

Behavior. Neuron 66:896–907.

Hirschfeld RMA (2001) The Comorbidity of Major Depression and Anxiety Disorders:

Recognition and Management in Primary Care. Prim Care Companion J Clin Psychiatry

3:244–254.

Hnasko TS, Hjelmstad GO, Fields HL, Edwards RH (2012) Ventral tegmental area glutamate

neurons: electrophysiological properties and projections. J Neurosci 32:15076–15085.

Hodge CW, Samson HH, Tolliver GA, Haraguchi M (1994) Effects of intraaccumbens

injections of dopamine agonists and antagonists on sucrose and sucrose-ethanol reinforced

responding. Pharmacol Biochem Behav 48:141–150.

Holmes A, Hollon TR, Gleason TC, Liu Z, Dreiling J, Sibley DR, Crawley JN (2001)

Behavioral characterization of dopamine D5 receptor null mutant mice. Behav Neurosci

115:1129–1144.

Hu XT, Wachtel SR, Galloway MP, White FJ (1990) Lesions of the nigrostriatal dopamine

projection increase the inhibitory effects of D1 and D2 dopamine agonists on caudate-

putamen neurons and relieve D2 receptors from the necessity of D1 receptor stimulation. J

Neurosci 10:2318–2329.

Huang J, Chen S, Zhang JJ, Huang X-Y (2013) Crystal structure of oligomeric β1-adrenergic G

protein-coupled receptors in ligand-free basal state. Nat Struct Mol Biol 20:419–425.

201

Hudmon A, Schulman H (2002) Structure-function of the multifunctional Ca2+/calmodulin-

dependent protein kinase II. Biochem J 364:593–611.

Ikemoto S (2007) Dopamine reward circuitry: two projection systems from the ventral

midbrain to the nucleus accumbens-olfactory tubercle complex. Brain Res Rev 56:27–78.

Ilango A, Kesner AJ, Keller KL, Stuber GD, Bonci A, Ikemoto S (2014) Similar roles of

substantia nigra and ventral tegmental dopamine neurons in reward and aversion. J

Neurosci 34:817–822.

Ishikawa A, Ambroggi F, Nicola SM, Fields HL (2008) Contributions of the amygdala and

medial prefrontal cortex to incentive cue responding. Neuroscience 155:573–584.

Ito R, Robbins TW, Pennartz CM, Everitt BJ (2008) Functional interaction between the

hippocampus and nucleus accumbens shell is necessary for the acquisition of appetitive

spatial context conditioning. J Neurosci 28:6950–6959.

Iversen SD, Iversen LL (2007) Dopamine: 50 years in perspective. Trends Neurosci 30:188–

193.

Jakel RJ, Maragos WF (2000) Neuronal cell death in Huntington’s disease: a potential role for

dopamine. Trends Neurosci 23:239–245.

Jenkins MA, Traynelis SF (2012) PKC phosphorylates GluA1-Ser831 to enhance AMPA

receptor conductance. Channels 6:60–64.

Jhou TC, Fields HL, Baxter MG, Saper CB, Holland PC (2009a) The rostromedial tegmental

nucleus (RMTg), a GABAergic afferent to midbrain dopamine neurons, encodes aversive

stimuli and inhibits motor responses. Neuron 61:786–800.

Jhou TC, Geisler S, Marinelli M, Degarmo BA, Zahm DS (2009b) The mesopontine

rostromedial tegmental nucleus: A structure targeted by the lateral habenula that projects

to the ventral tegmental area of Tsai and substantia nigra compacta. J Comp Neurol

513:566–596.

Jin L-Q, Goswami S, Cai G, Zhen X, Friedman E (2003) SKF83959 selectively regulates

phosphatidylinositol-linked D1 dopamine receptors in rat brain. J Neurochem 85:378–386.

Kalivas PW, Stewart J (1991) Dopamine transmission in the initiation and expression of drug-

and stress-induced sensitization of motor activity. Brain Res Brain Res Rev 16:223–244.

Kantor L, Hewlett GH, Gnegy ME (1999) Enhanced amphetamine- and K+-mediated

dopamine release in rat striatum after repeated amphetamine: differential requirements for

Ca2+- and calmodulin-dependent phosphorylation and synaptic vesicles. J Neurosci

19:3801–3808.

202

Karege F, Perroud N, Burkhardt S, Schwald M, Ballmann E, La Harpe R, Malafosse A (2007)

Alteration in kinase activity but not in protein levels of protein kinase B and glycogen

synthase kinase-3beta in ventral prefrontal cortex of depressed suicide victims. Biol

Psychiatry 61:240–245.

Katritch V, Cherezov V, Stevens RC (2012) Diversity and modularity of G protein-coupled

receptor structures. Trends Pharmacol Sci 33:17–27.

Kaufman DL, Houser CR, Tobin AJ (1991) Two forms of the gamma-aminobutyric acid

synthetic enzyme glutamate decarboxylase have distinct intraneuronal distributions and

cofactor interactions. J Neurochem 56:720–723.

Kaupmann K, Malitschek B, Schuler V, Heid J, Froestl W, Beck P, Mosbacher J, Bischoff S,

Kulik A, Shigemoto R, Karschin A, Bettler B (1998) GABA(B)-receptor subtypes

assemble into functional heteromeric complexes. Nature 396:683–687.

Kavalali ET, Monteggia LM (2012) Synaptic mechanisms underlying rapid antidepressant

action of ketamine. Am J Psychiatry 169:1150–1156.

Kawaguchi Y, Wilson CJ, Augood SJ, Emson PC (1995) Striatal interneurones: chemical,

physiological and morphological characterization. Trends Neurosci 18:527–535.

Kebabian JW, Calne DB (1979) Multiple receptors for dopamine. Nature 277:93–96.

Keefe KA, Gerfen CR (1995) D1-D2 dopamine receptor synergy in striatum: effects of

intrastriatal infusions of dopamine agonists and antagonists on immediate early gene

expression. Neuroscience 66:903–913.

Kelley AE, Domesick VB (1982) The distribution of the projection from the hippocampal

formation to the nucleus accumbens in the rat: an anterograde- and retrograde-horseradish

peroxidase study. Neuroscience 7:2321–2335.

Kelley AE, Domesick VB, Nauta WJ (1982) The amygdalostriatal projection in the rat--an

anatomical study by anterograde and retrograde tracing methods. Neuroscience 7:615–630.

Kim JH, Austin JD, Tanabe L, Creekmore E, Vezina P (2005) Activation of group II mGlu

receptors blocks the enhanced drug taking induced by previous exposure to amphetamine.

Eur J Neurosci 21:295–300.

Kohout TA, Lefkowitz RJ (2003) Regulation of G protein-coupled receptor kinases and

arrestins during receptor desensitization. Mol Pharmacol 63:9–18.

Kong MMC, Fan T, Varghese G, O’dowd BF, George SR (2006) Agonist-induced cell surface

trafficking of an intracellularly sequestered D1 dopamine receptor homo-oligomer. Mol

Pharmacol 70:78–89.

203

Koo JW et al. (2012) BDNF is a negative modulator of morphine action. Science 338:124–128.

Koo JW, Lobo MK, Chaudhury D, Labonté B, Friedman A, Heller E, Peña CJ, Han M-H,

Nestler EJ (2014) Loss of BDNF signaling in D1R-expressing NAc neurons enhances

morphine reward by reducing GABA inhibition. Neuropsychopharmacology 39:2646–

2653.

Koob GF, Le Moal M (2001) Drug addiction, dysregulation of reward, and allostasis.

Neuropsychopharmacology 24:97–129.

Koob GF, Volkow ND (2010a) Neurocircuitry of addiction. Neuropsychopharmacology

35:217–238.

Koob GF, Volkow ND (2010b) Neurocircuitry of addiction. Neuropsychopharmacology

35:217–238.

Krishnan V et al. (2007) Molecular adaptations underlying susceptibility and resistance to

social defeat in brain reward regions. Cell 131:391–404.

Krishnan V, Nestler EJ (2008) The molecular neurobiology of depression. Nature 455:894–902.

LaLumiere RT, Smith KC, Kalivas PW (2012) Neural circuit competition in cocaine-seeking:

roles of the infralimbic cortex and nucleus accumbens shell. Eur J Neurosci 35:614–622.

Lammel S, Hetzel A, Häckel O, Jones I, Liss B, Roeper J (2008) Unique Properties of

Mesoprefrontal Neurons within a Dual Mesocorticolimbic Dopamine System. Neuron

57:760–773.

Lammel S, Lim BK, Malenka RC (2014) Reward and aversion in a heterogeneous midbrain

dopamine system. Neuropharmacology 76 Pt B:351–359.

Lammel S, Lim BK, Ran C, Huang KW, Betley MJ, Tye KM, Deisseroth K, Malenka RC

(2012) Input-specific control of reward and aversion in the ventral tegmental area. Nature.

Lasseter HC, Ramirez DR, Xie X, Fuchs RA (2009) Involvement of the lateral orbitofrontal

cortex in drug context-induced reinstatement of cocaine-seeking behavior in rats. Eur J

Neurosci 30:1370–1381.

Latek D, Modzelewska A, Trzaskowski B, Palczewski K, Filipek S (2012) G protein-coupled

receptors--recent advances. Acta Biochim Pol 59:515–529.

Le Moine C, Bloch B (1995) D1 and D2 dopamine receptor gene expression in the rat striatum:

sensitive cRNA probes demonstrate prominent segregation of D1 and D2 mRNAs in

distinct neuronal populations of the dorsal and ventral striatum. J Comp Neurol 355:418–

426.

204

Lecca S, Meye FJ, Mameli M (2014) The lateral habenula in addiction and depression: an

anatomical, synaptic and behavioral overview. Eur J Neurosci 39:1170–1178.

Lecourtier L, Defrancesco A, Moghaddam B (2008) Differential tonic influence of lateral

habenula on prefrontal cortex and nucleus accumbens dopamine release. Eur J Neurosci

27:1755–1762.

Lee SM, Kant A, Blake D, Murthy V, Boyd K, Wyrick SJ, Mailman RB (2014) SKF-83959 is

not a highly-biased functionally selective D1 dopamine receptor ligand with activity at

phospholipase C. Neuropharmacology 86:145–154.

Lee SP, O’Dowd BF, Rajaram RD, Nguyen T, George SR (2003) D2 dopamine receptor

homodimerization is mediated by multiple sites of interaction, including an intermolecular

interaction involving transmembrane domain 4. Biochemistry 42:11023–11031.

Lee SP, So CH, Rashid AJ, Varghese G, Cheng R, Lança AJ, O’Dowd BF, George SR (2004)

Dopamine D1 and D2 receptor Co-activation generates a novel phospholipase C-mediated

calcium signal. J Biol Chem 279:35671–35678.

Leff P (1995) The two-state model of receptor activation. Trends Pharmacol Sci 16:89–97.

Lessmann V, Gottmann K, Malcangio M (2003) Neurotrophin secretion: current facts and

future prospects. Prog Neurobiol 69:341–374.

Lett BT (1989) Repeated exposures intensify rather than diminish the rewarding effects of

amphetamine, morphine, and cocaine. Psychopharmacology (Berl) 98:357–362.

Levey AI, Hersch SM, Rye DB, Sunahara RK, Niznik HB, Kitt CA, Price DL, Maggio R,

Brann MR, Ciliax BJ (1993) Localization of D1 and D2 dopamine receptors in brain with

subtype-specific antibodies. Proc Natl Acad Sci U S A 90:8861–8865.

Li B, Piriz J, Mirrione M, Chung C, Proulx CD, Schulz D, Henn F, Malinow R (2011) Synaptic

potentiation onto habenula neurons in the learned helplessness model of depression.

Nature 470:535–539.

Li K, Zhou T, Liao L, Yang Z, Wong C, Henn F, Malinow R, Yates JR, Hu H (2013) βCaMKII

in lateral habenula mediates core symptoms of depression. Science 341:1016–1020.

Li X, Wolf ME (2015) Multiple faces of BDNF in cocaine addiction. Behav Brain Res

279:240–254.

Liao R, Chang Y, Wang S (1998) Influence of SCH23390 and spiperone on the expression of

conditioned place preference induced by d-amphetamine or cocaine in the rat. Chin J

Physiol 41:85–92.

205

Liu Q, Pu L, Poo M (2005) Repeated cocaine exposure in vivo facilitates LTP induction in

midbrain dopamine neurons. Nature 437:1027–1031.

Liu Z-H, Shin R, Ikemoto S (2008) Dual role of medial A10 dopamine neurons in affective

encoding. Neuropsychopharmacology 33:3010–3020.

Lobb CJ, Wilson CJ, Paladini CA (2010) A dynamic role for GABA receptors on the firing

pattern of midbrain dopaminergic neurons. J Neurophysiol 104:403–413.

Lobo MK, Covington HE, Chaudhury D, Friedman AK, Sun H, Damez-Werno D, Dietz DM,

Zaman S, Koo JW, Kennedy PJ, Mouzon E, Mogri M, Neve RL, Deisseroth K, Han M-H,

Nestler EJ (2010a) Cell type-specific loss of BDNF signaling mimics optogenetic control

of cocaine reward. Science 330:385–390.

Lobo MK, Covington HE, Chaudhury D, Friedman AK, Sun H, Damez-Werno D, Dietz DM,

Zaman S, Koo JW, Kennedy PJ, Mouzon E, Mogri M, Neve RL, Deisseroth K, Han M-H,

Nestler EJ (2010b) Cell type-specific loss of BDNF signaling mimics optogenetic control

of cocaine reward. Science 330:385–390.

Lobo MK, Nestler EJ (2011) The striatal balancing act in drug addiction: distinct roles of direct

and indirect pathway medium spiny neurons. Front Neuroanat 5:41.

Lopez-Gimenez JF, Canals M, Pediani JD, Milligan G (2007) The 1b-Adrenoceptor Exists as a

Higher-Order Oligomer: Effective Oligomerization Is Required for Receptor Maturation,

Surface Delivery, and Function. Mol Pharmacol 71:1015–1029.

Loweth JA, Singer BF, Baker LK, Wilke G, Inamine H, Bubula N, Alexander JK, Carlezon

WA, Neve RL, Vezina P (2010) Transient overexpression of alpha-Ca2+/calmodulin-

dependent protein kinase II in the nucleus accumbens shell enhances behavioral

responding to amphetamine. J Neurosci 30:939–949.

Lüscher C, Malenka RC (2011) Drug-evoked synaptic plasticity in addiction: from molecular

changes to circuit remodeling. Neuron 69:650–663.

Lynch WJ, Kiraly DD, Caldarone BJ, Picciotto MR, Taylor JR (2007) Effect of cocaine self-

administration on striatal PKA-regulated signaling in male and female rats.

Psychopharmacology (Berl) 191:263–271.

Maeda A, Okano K, Park PS-H, Lem J, Crouch RK, Maeda T, Palczewski K (2010)

Palmitoylation stabilizes unliganded rod opsin. Proc Natl Acad Sci U S A 107:8428–8433.

Maguire EP, Macpherson T, Swinny JD, Dixon CI, Herd MB, Belelli D, Stephens DN, King

SL, Lambert JJ (2014) Tonic Inhibition of Accumbal Spiny Neurons by Extrasynaptic

α4βδ GABAA Receptors Modulates the Actions of Psychostimulants. J Neurosci 34:823–

838.

206

Mahar I, Bambico FR, Mechawar N, Nobrega JN (2014) Stress, serotonin, and hippocampal

neurogenesis in relation to depression and antidepressant effects. Neurosci Biobehav Rev

38:173–192.

Mahler S V, Vazey EM, Beckley JT, Keistler CR, McGlinchey EM, Kaufling J, Wilson SP,

Deisseroth K, Woodward JJ, Aston-Jones G (2014) Designer receptors show role for

ventral pallidum input to ventral tegmental area in cocaine seeking. Nat Neurosci 17:577–

585.

Manglik A, Kruse AC, Kobilka TS, Thian FS, Mathiesen JM, Sunahara RK, Pardo L, Weis WI,

Kobilka BK, Granier S (2012) Crystal structure of the µ-opioid receptor bound to a

morphinan antagonist. Nature 485:321–326.

Markou A, Harrison AA, Chevrette J, Hoyer D (2005) Paroxetine combined with a 5-HT(1A)

receptor antagonist reversed reward deficits observed during amphetamine withdrawal in

rats. Psychopharmacology (Berl) 178:133–142.

Markou A, Koob GF (1992) Bromocriptine reverses the elevation in intracranial self-

stimulation thresholds observed in a rat model of cocaine withdrawal.

Neuropsychopharmacology 7:213–224.

Martín-García E, Courtin J, Renault P, Fiancette J-F, Wurtz H, Simonnet A, Levet F, Herry C,

Deroche-Gamonet V (2014) Frequency of cocaine self-administration influences drug

seeking in the rat: optogenetic evidence for a role of the prelimbic cortex.

Neuropsychopharmacology 39:2317–2330.

Matamales M, Bertran-Gonzalez J, Salomon L, Degos B, Deniau JM, Valjent E, Herv?? D,

Girault JA (2009) Striatal medium-sized spiny neurons: Identification by nuclear staining

and study of neuronal subpopulations in BAC transgenic mice. PLoS One 4.

Matsumoto M, Hikosaka O (2007) Lateral habenula as a source of negative reward signals in

dopamine neurons. Nature 447:1111–1115.

Matsumoto M, Hikosaka O (2009) Representation of negative motivational value in the

primate lateral habenula. Nat Neurosci 12:77–84.

Maudsley S, Siddiqui S, Martin B (2013) Systems analysis of arrestin pathway functions. Prog

Mol Biol Transl Sci 118:431–467.

Mayberg HS, Brannan SK, Tekell JL, Silva JA, Mahurin RK, McGinnis S, Jerabek PA (2000)

Regional metabolic effects of fluoxetine in major depression: serial changes and

relationship to clinical response. Biol Psychiatry 48:830–843.

McClung CA, Nestler EJ (2003) Regulation of gene expression and cocaine reward by CREB

and DeltaFosB. Nat Neurosci 6:1208–1215.

207

McGinty JF, Whitfield TW, Berglind WJ (2010) Brain-derived neurotrophic factor and cocaine

addiction. Brain Res 1314:183–193.

McGregor C, Srisurapanont M, Jittiwutikarn J, Laobhripatr S, Wongtan T, White JM (2005)

The nature, time course and severity of methamphetamine withdrawal. Addiction

100:1320–1329.

McLaughlin J, See RE (2003) Selective inactivation of the dorsomedial prefrontal cortex and

the basolateral amygdala attenuates conditioned-cued reinstatement of extinguished

cocaine-seeking behavior in rats. Psychopharmacology (Berl) 168:57–65.

Medvedev IO, Ramsey AJ, Masoud ST, Bermejo MK, Urs N, Sotnikova TD, Beaulieu J-M,

Gainetdinov RR, Salahpour A (2013) D1 dopamine receptor coupling to PLCβ regulates

forward locomotion in mice. J Neurosci 33:18125–18133.

Mello NK, Negus SS (1996) Preclinical evaluation of pharmacotherapies for treatment of

cocaine and opioid abuse using drug self-administration procedures.

Neuropsychopharmacology 14:375–424.

Mendrek A, Blaha CD, Phillips AG (1998) Pre-exposure of rats to amphetamine sensitizes self-

administration of this drug under a progressive ratio schedule. Psychopharmacology (Berl)

135:416–422.

Meredith GE (1999) The synaptic framework for chemical signaling in nucleus accumbens.

Ann N Y Acad Sci 877:140–156.

Meririnne E, Kajos M, Kankaanpää A, Koistinen M, Kiianmaa K, Seppälä T (2005) Rewarding

properties of the stereoisomers of 4-methylaminorex: Involvement of the dopamine

system. Pharmacol Biochem Behav 81:715–724.

Meririnne E, Kankaanpää A, Seppälä T (2001) Rewarding properties of methylphenidate:

sensitization by prior exposure to the drug and effects of dopamine D1- and D2-receptor

antagonists. J Pharmacol Exp Ther 298:539–550.

Meruelo AD, Samish I, Bowie JU (2011) TMKink: a method to predict transmembrane helix

kinks. Protein Sci 20:1256–1264.

Mirenowicz J, Schultz W (1994) Importance of unpredictability for reward responses in

primate dopamine neurons. J Neurophysiol 72:1024–1027.

Mirenowicz J, Schultz W (1996) Preferential activation of midbrain dopamine neurons by

appetitive rather than aversive stimuli. Nature 379:449–451.

Missale C, Nash SR, Robinson SW, Jaber M, Caron MG (1998) Dopamine receptors: from

structure to function. Physiol Rev 78:189–225.

208

Mogenson GJ, Jones DL, Yim CY (1980) From motivation to action: functional interface

between the limbic system and the motor system. Prog Neurobiol 14:69–97.

Moody CA, Frank RA (1990) Cocaine facilitates prefrontal cortex self-stimulation. Pharmacol

Biochem Behav 35:743–746.

Moreno JL, Holloway T, González-Maeso J (2013) G protein-coupled receptor

heterocomplexes in neuropsychiatric disorders. Prog Mol Biol Transl Sci 117:187–205.

Moss J, Bolam JP (2008) A dopaminergic axon lattice in the striatum and its relationship with

cortical and thalamic terminals. J Neurosci 28:11221–11230.

Müller DJ, Wu N, Palczewski K (2008) Vertebrate membrane proteins: structure, function, and

insights from biophysical approaches. Pharmacol Rev 60:43–78.

Müller MB, Holsboer F (2006) Mice with mutations in the HPA-system as models for

symptoms of depression. Biol Psychiatry 59:1104–1115.

Nairn AC, Svenningsson P, Nishi A, Fisone G, Girault J-A, Greengard P (2004) The role of

DARPP-32 in the actions of drugs of abuse. Neuropharmacology 47 Suppl 1:14–23.

Nair-Roberts RG, Chatelain-Badie SD, Benson E, White-Cooper H, Bolam JP, Ungless MA

(2008) Stereological estimates of dopaminergic, GABAergic and glutamatergic neurons in

the ventral tegmental area, substantia nigra and retrorubral field in the rat. Neuroscience

152:1024–1031.

Nally RE, Kinsella A, Tighe O, Croke DT, Fienberg AA, Greengard P, Waddington JL (2004)

Ethologically based resolution of D2-like dopamine receptor agonist-versus antagonist-

induced behavioral topography in dopamine- and adenosine 3’,5'-monophosphate-

regulated phosphoprotein of 32 kDa “knockout” mutants congenic on the C57BL/6

genetic backgr. J Pharmacol Exp Ther 310:1281–1287.

Nauta WJ, Smith GP, Faull RL, Domesick VB (1978) Efferent connections and nigral afferents

of the nucleus accumbens septi in the rat. Neuroscience 3:385–401.

Navarro G, Ferre S, Cordomi A, Moreno E, Mallol J, Casado V, Cortes A, Hoffmann H, Ortiz J,

Canela EI, Lluis C, Pardo L, Franco R, Woods AS (2010) Interactions between

Intracellular Domains as Key Determinants of the Quaternary Structure and Function of

Receptor Heteromers. J Biol Chem 285:27346–27359.

Nestler EJ (1999) Neuronal and behavioural abnormalities in striatal function in DARPP-32-

mutant mice. Eur J Neurosci 11:1114–1118.

Nestler EJ, Carlezon W a (2006a) The mesolimbic dopamine reward circuit in depression. Biol

Psychiatry 59:1151–1159.

209

Nestler EJ, Carlezon WA (2006b) The Mesolimbic Dopamine Reward Circuit in Depression.

Biol Psychiatry 59:1151–1159.

Ng J, Rashid AJ, So CH, O’Dowd BF, George SR (2010) Activation of calcium/calmodulin-

dependent protein kinase IIalpha in the striatum by the heteromeric D1-D2 dopamine

receptor complex. Neuroscience 165:535–541.

Nicola SM (2007) The nucleus accumbens as part of a basal ganglia action selection circuit.

Psychopharmacology (Berl) 191:521–550.

Nicola SM, Surmeier J, Malenka RC (2000) Dopaminergic modulation of neuronal excitability

in the striatum and nucleus accumbens. Annu Rev Neurosci 23:185–215.

Niesen MJM, Bhattacharya S, Vaidehi N (2011) The role of conformational ensembles in

ligand recognition in G-protein coupled receptors. J Am Chem Soc 133:13197–13204.

Nisell M, Nomikos GG, Svensson TH (1994) Systemic nicotine-induced dopamine release in

the rat nucleus accumbens is regulated by nicotinic receptors in the ventral tegmental area.

Synapse 16:36–44.

Nishi A, Bibb JA, Matsuyama S, Hamada M, Higashi H, Nairn AC, Greengard P (2002)

Regulation of DARPP-32 dephosphorylation at PKA- and Cdk5-sites by NMDA and

AMPA receptors: Distinct roles of calcineurin and protein phosphatase-2A. J Neurochem

81:832–841.

Nishi A, Snyder GL, Greengard P (1997) Bidirectional regulation of DARPP-32

phosphorylation by dopamine. J Neurosci 17:8147–8155.

Nishi A, Watanabe Y, Higashi H, Tanaka M, Nairn AC, Greengard P (2005) Glutamate

regulation of DARPP-32 phosphorylation in neostriatal neurons involves activation of

multiple signaling cascades. Proc Natl Acad Sci U S A 102:1199–1204.

O’Donnell P, Lavín A, Enquist LW, Grace AA, Card JP (1997) Interconnected parallel circuits

between rat nucleus accumbens and thalamus revealed by retrograde transynaptic

transport of pseudorabies virus. J Neurosci 17:2143–2167.

O’Dowd BF, Ji X, Alijaniaram M, Rajaram RD, Kong MMC, Rashid A, Nguyen T, George SR

(2005) Dopamine receptor oligomerization visualized in living cells. J Biol Chem

280:37225–37235.

O’Dowd BF, Ji X, Nguyen T, George SR (2012) Two amino acids in each of D1 and D2

dopamine receptor cytoplasmic regions are involved in D1–D2 heteromer formation.

Biochem Biophys Res Commun 417:23–28.

O’Sullivan GJ, Kinsella A, Sibley DR, Tighe O, Croke DT, Waddington JL (2005) Ethological

resolution of behavioural topography and D1-like versus D2-like agonist responses in

210

congenic D5 dopamine receptor mutants: identification of D5:D2-like interactions.

Synapse 55:201–211.

Oda S, Funato H, Adachi-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.

Oldham WM, Van Eps N, Preininger AM, Hubbell WL, Hamm HE (2006) Mechanism of the

receptor-catalyzed activation of heterotrimeric G proteins. Nat Struct Mol Biol 13:772–

777.

Omelchenko N, Bell R, Sesack SR (2009) Lateral habenula projections to dopamine and

GABA neurons in the rat ventral tegmental area. Eur J Neurosci 30:1239–1250.

Ortiz J, Fitzgerald LW, Lane S, Terwilliger R, Nestler EJ (1996) Biochemical adaptations in

the mesolimbic dopamine system in response to repeated stress.

Neuropsychopharmacology 14:443–452.

Palczewski K, McDowell JH, Jakes S, Ingebritsen TS, Hargrave PA (1989) Regulation of

rhodopsin dephosphorylation by arrestin. J Biol Chem 264:15770–15773.

Paulson PE, Camp DM, Robinson TE (1991) Time course of transient behavioral depression

and persistent behavioral sensitization in relation to regional brain monoamine

concentrations during amphetamine withdrawal in rats. Psychopharmacology (Berl)

103:480–492.

Paulson PE, Robinson TE (1991) Sensitization to systemic amphetamine produces an enhanced

locomotor response to a subsequent intra-accumbens amphetamine challenge in rats.

Psychopharmacology (Berl) 104:140–141.

Paxinos G, Watson C (1998) The Rat Brain in Stereotaxic Coordinates Fourth Edition.

Pedersen UB, Norby B, Jensen AA, Schiødt M, Hansen A, Suhr-Jessen P, Scheideler M,

Thastrup O, Andersen PH (1994) Characteristics of stably expressed human dopamine

D1a and D1b receptors: atypical behavior of the dopamine D1b receptor. Eur J Pharmacol

267:85–93.

Pei L, Li S, Wang M, Diwan M, Anisman H, Fletcher PJ, Nobrega JN, Liu F (2010)

Uncoupling the dopamine D1-D2 receptor complex exerts antidepressant-like effects. Nat

Med 16:1393–1395.

Pellegrino SM, Druse MJ (1992) The effects of chronic ethanol consumption on the

mesolimbic and nigrostriatal dopamine systems. Alcohol Clin Exp Res 16:275–280.

211

Peng XQ, Li X, Gilbert JG, Pak AC, Ashby CR, Brodie JD, Dewey SL, Gardner EL, Xi ZX

(2008) Gamma-vinyl GABA inhibits cocaine-triggered reinstatement of drug-seeking

behavior in rats by a non-dopaminergic mechanism. Drug Alcohol Depend 97:216–225.

Perreault ML, Fan T, Alijaniaram M, O’Dowd BF, George SR (2012a) Dopamine D1-D2

receptor heteromer in dual phenotype GABA/glutamate-coexpressing striatal medium

spiny neurons: Regulation of BDNF, GAD67 and VGLUT1/2. PLoS One 7.

Perreault ML, Fan T, O’Dowd BF, George SR (2013) Enhanced brain-derived neurotrophic

factor signaling in the nucleus accumbens of juvenile rats. Dev Neurosci 35:384–395.

Perreault ML, Hasbi A, Alijaniaram M, Fan T, Varghese G, Fletcher PJ, Seeman P, O’Dowd

BF, George SR (2010) The dopamine D1-D2 receptor heteromer localizes in

dynorphin/enkephalin neurons: increased high affinity state following amphetamine and in

schizophrenia. J Biol Chem 285:36625–36634.

Perreault ML, Hasbi A, O’Dowd BF, George SR (2011) The dopamine d1-d2 receptor

heteromer in striatal medium spiny neurons: evidence for a third distinct neuronal

pathway in Basal Ganglia. Front Neuroanat 5:31.

Perreault ML, Jones-Tabah J, O’Dowd BF, George SR (2012b) A physiological role for the

dopamine D5 receptor as a regulator of BDNF and Akt signalling in rodent prefrontal

cortex. Int J Neuropsychopharmacol:1–7.

Perreault ML, O’Dowd BF, George SR (2014) Dopamine D1-D2 receptor heteromer regulates

signaling cascades involved in addiction: potential relevance to adolescent drug

susceptibility. Dev Neurosci 36:287–296.

Perreault ML, Shen MYF, Fan T, George SR (2015) Regulation of c-fos expression by the

dopamine D1-D2 receptor heteromer. Neuroscience 285:194–203.

Peters J, LaLumiere RT, Kalivas PW (2008) Infralimbic prefrontal cortex is responsible for

inhibiting cocaine seeking in extinguished rats. J Neurosci 28:6046–6053.

Pierce RC, Kalivas PW (1997) Repeated cocaine modifies the mechanism by which

amphetamine releases dopamine. J Neurosci 17:3254–3261.

Pierre PJ, Vezina P (1997) Predisposition to self-administer amphetamine: The contribution of

response to novelty and prior exposure to the drug. Psychopharmacology (Berl) 129:277–

284.

Pierre PJ, Vezina P (1998) D1 dopamine receptor blockade prevents the facilitation of

amphetamine self-administration induced by prior exposure to the drug.

Psychopharmacology (Berl) 138:159–166.

212

Pozzi L, Håkansson K, Usiello A, Borgkvist A, Lindskog M, Greengard P, Fisone G (2003)

Opposite regulation by typical and atypical anti-psychotics of ERK1/2, CREB and Elk-1

phosphorylation in mouse dorsal striatum. J Neurochem 86:451–459.

Proulx CD, Hikosaka O, Malinow R (2014) Reward processing by the lateral habenula in

normal and depressive behaviors. Nat Neurosci 17:1146–1152.

Rashid AJ, So CH, Kong MMC, Furtak T, El-Ghundi M, Cheng R, O’Dowd BF, George SR

(2007) D1-D2 dopamine receptor heterooligomers with unique pharmacology are coupled

to rapid activation of Gq/11 in the striatum. Proc Natl Acad Sci U S A 104:654–659.

Rasmussen SGF et al. (2011) Crystal structure of the β2 adrenergic receptor-Gs protein

complex. Nature 477:549–555.

Rebholz H, Zhou M, Nairn AC, Greengard P, Flajolet M (2013) Selective knockout of the

casein kinase 2 in d1 medium spiny neurons controls dopaminergic function. Biol

Psychiatry 74:113–121.

Redgrave P, Prescott TJ, Gurney K (1999) The basal ganglia: a vertebrate solution to the

selection problem? Neuroscience 89:1009–1023.

Reiter E, Lefkowitz RJ (2006) GRKs and beta-arrestins: roles in receptor silencing, trafficking

and signaling. Trends Endocrinol Metab 17:159–165.

Resh MD (1999) Fatty acylation of proteins: new insights into membrane targeting of

myristoylated and palmitoylated proteins. Biochim Biophys Acta 1451:1–16.

Richardson NR, Roberts DCS (1996) Progressive ratio schedules in drug self-administration

studies in rats: A method to evaluate reinforcing efficacy. J Neurosci Methods 66:1–11.

Rivero-Müller A, Jonas KC, Hanyaloglu AC, Huhtaniemi I (2013) Di/oligomerization of

GPCRs-mechanisms and functional significance. Prog Mol Biol Transl Sci 117:163–185.

Robbins TW, Ersche KD, Everitt BJ (2008) Drug addiction and the memory systems of the

brain. Ann N Y Acad Sci 1141:1–21.

Robinson TE, Becker JB (1986) Enduring changes in brain and behavior produced by chronic

amphetamine administration: a review and evaluation of animal models of amphetamine

psychosis. Brain Res 396:157–198.

Robinson TE, Berridge KC (1993) The neural basis of drug craving: An incentive-sensitization

theory of addiction. Brain Res Rev 18:247–291.

Robinson TE, Berridge KC (2000) The psychology and neurobiology of addiction: an

incentive-sensitization view. Addiction 95 Suppl 2:S91–S117.

213

Robinson TE, Jurson PA, Bennett JA, Bentgen KM (1988) Persistent sensitization of dopamine

neurotransmission in ventral striatum (nucleus accumbens) produced by prior experience

with (+)-amphetamine: a microdialysis study in freely moving rats. Brain Res 462:211–

222.

Robison AJ (2014) Emerging role of CaMKII in neuropsychiatric disease. Trends Neurosci.

Robison AJ, Vialou V, Mazei-Robison M, Feng J, Kourrich S, Collins M, Wee S, Koob G,

Turecki G, Neve R, Thomas M, Nestler EJ (2013) Behavioral and structural responses to

chronic cocaine require a feedforward loop involving ΔFosB and calcium/calmodulin-

dependent protein kinase II in the nucleus accumbens shell. J Neurosci 33:4295–4307.

Robison AJ, Vialou V, Sun H-S, Labonte B, Golden SA, Dias C, Turecki G, Tamminga C,

Russo S, Mazei-Robison M, Nestler EJ (2014) Fluoxetine epigenetically alters the

CaMKIIα promoter in nucleus accumbens to regulate ΔFosB binding and antidepressant

effects. Neuropsychopharmacology 39:1178–1186.

Rocha BA, Goulding EH, O’Dell LE, Mead AN, Coufal NG, Parsons LH, Tecott LH (2002)

Enhanced locomotor, reinforcing, and neurochemical effects of cocaine in serotonin 5-

hydroxytryptamine 2C receptor mutant mice. J Neurosci 22:10039–10045.

Rodd-Henricks ZA, McKinzie DL, Li T-K, Murphy JM, McBride WJ (2002) Cocaine is self-

administered into the shell but not the core of the nucleus accumbens of Wistar rats. J

Pharmacol Exp Ther 303:1216–1226.

Rogóż Z, Skuza G (2011) Anxiolytic-like effects of olanzapine, risperidone and fluoxetine in

the elevated plus-maze test in rats. Pharmacol reports PR 63:1547–1552.

Root DH, Mejias-Aponte CA, Qi J, Morales M (2014) Role of Glutamatergic Projections from

Ventral Tegmental Area to Lateral Habenula in Aversive Conditioning. J Neurosci

34:13906–13910.

Rosenbaum DM, Rasmussen SGF, Kobilka BK (2009) The structure and function of G-

protein-coupled receptors. Nature 459:356–363.

Rossetti ZL, Hmaidan Y, Gessa GL (1992) Marked inhibition of mesolimbic dopamine release:

a common feature of ethanol, morphine, cocaine and amphetamine abstinence in rats. Eur

J Pharmacol 221:227–234.

Rossetti ZL, Lai M, Hmaidan Y, Gessa GL (1993) Depletion of mesolimbic dopamine during

behavioral despair: partial reversal by chronic imipramine. Eur J Pharmacol 242:313–315.

Rothman RB, Baumann MH, Dersch CM, Romero D V, Rice KC, Carroll FI, Partilla JS (2001)

Amphetamine-type central nervous system stimulants release norepinephrine more

potently than they release dopamine and serotonin. Synapse 39:32–41.

214

Russo SJ, Nestler EJ (2013) The brain reward circuitry in mood disorders. Nat Rev Neurosci

14:609–625.

Saal D, Dong Y, Bonci A, Malenka RC (2003) Drugs of abuse and stress trigger a common

synaptic adaptation in dopamine neurons. Neuron 37:577–582.

Sahu A, Tyeryar KR, Vongtau HO, Sibley DR, Undieh AS (2009) D5 dopamine receptors are

required for dopaminergic activation of phospholipase C. Mol Pharmacol 75:447–453.

Salahpour A, Angers S, Mercier J-F, Lagace M, Marullo S, Bouvier M (2004)

Homodimerization of the 2-Adrenergic Receptor as a Prerequisite for Cell Surface

Targeting. J Biol Chem 279:33390–33397.

Sanchez-Catalan M-J, Kaufling J, Georges F, Veinante P, Barrot M (2014) The Antero-

Posterior Heterogeneity of the Ventral Tegmental Area. Neuroscience 282C:198–216.

Scheggi S, Raone A, De Montis MG, Tagliamonte A, Gambarana C (2007) Behavioral

expression of cocaine sensitization in rats is accompanied by a distinct pattern of

modifications in the PKA/DARPP-32 signaling pathway. J Neurochem 103:1168–1183.

Schmidt HD, Anderson SM, Pierce RC (2006) Stimulation of D1-like or D2 dopamine

receptors in the shell, but not the core, of the nucleus accumbens reinstates cocaine-

seeking behaviour in the rat. Eur J Neurosci 23:219–228.

Schmidt HD, Duman RS (2010) Peripheral BDNF produces antidepressant-like effects in

cellular and behavioral models. Neuropsychopharmacology 35:2378–2391.

Schmidt WJ, Beninger RJ (2006) Behavioural sensitization in addiction, schizophrenia,

Parkinson’s disease and dyskinesia. Neurotox Res 10:161–166.

Schwarze SR, Ho A, Vocero-Akbani A, Dowdy SF (1999) In vivo protein transduction:

delivery of a biologically active protein into the mouse. Science 285:1569–1572.

Seeman P (2006) Targeting the dopamine D2 receptor in schizophrenia. Expert Opin Ther

Targets 10:515–531.

Seeman P, Lee T, Chau-Wong M, Wong K (1976) Antipsychotic drug doses and

neuroleptic/dopamine receptors. Nature 261:717–719.

Sellings LHL, Clarke PBS (2003) Segregation of amphetamine reward and locomotor

stimulation between nucleus accumbens medial shell and core. J Neurosci 23:6295–6303.

Senes A, Gerstein M, Engelman DM (2000) Statistical analysis of amino acid patterns in

transmembrane helices: the GxxxG motif occurs frequently and in association with beta-

branched residues at neighboring positions. J Mol Biol 296:921–936.

215

Sequeira-Cordero A, Mora-Gallegos A, Cuenca-Berger P, Fornaguera-Trías J (2014)

Individual differences in the forced swimming test and neurochemical kinetics in the rat

brain. Physiol Behav 128:60–69.

Sesack SR, Grace AA (2010) Cortico-Basal Ganglia reward network: microcircuitry.

Neuropsychopharmacology 35:27–47.

Shabel S, Proulx C, Piritz J, Malinow R (2014) Mood regulation. GABA/glutamate co-release

controls habenula output and is modified by antidepressant treatment. Science (80- )

345:1494–1498.

Shaham Y, Shalev U, Lu L, De Wit H, Stewart J (2003) The reinstatement model of drug

relapse: History, methodology and major findings. Psychopharmacology (Berl) 168:3–20.

Shen H, Gipson CD, Huits M, Kalivas PW (2014a) Prelimbic cortex and ventral tegmental area

modulate synaptic plasticity differentially in nucleus accumbens during cocaine-reinstated

drug seeking. Neuropsychopharmacology 39:1169–1177.

Shen MYF, Perreault ML, Fan T, George SR (2014b) The dopamine D1-D2 receptor

heteromer exerts tonic inhibitory effect on the expression of amphetamine-induced

locomotor sensitization. Behav Brain Res.

Shen MYF, Perreault ML, Fan T, George SR (2015) The dopamine D1-D2 receptor heteromer

exerts a tonic inhibitory effect on the expression of amphetamine-induced locomotor

sensitization. Pharmacol Biochem Behav 128:33–40.

Shen W, Hamilton SE, Nathanson NM, Surmeier DJ (2005) Cholinergic suppression of KCNQ

channel currents enhances excitability of striatal medium spiny neurons. J Neurosci

25:7449–7458.

Shippenberg TS, Heidbreder C, Lefevour A (1996) Sensitization to the conditioned rewarding

effects of morphine: Pharmacology and temporal characteristics. Eur J Pharmacol 299:33–

39.

Shirayama Y, Chaki S (2006) Neurochemistry of the nucleus accumbens and its relevance to

depression and antidepressant action in rodents. Curr Neuropharmacol 4:277–291.

Shoaib M, Swanner LS, Beyer CE, Goldberg SR, Schindler CW (1998) The GABAB agonist

baclofen modifies cocaine self-administration in rats. Behav Pharmacol 9:195–206.

Shuen JA, Chen M, Gloss B, Calakos N (2008) Drd1a-tdTomato BAC transgenic mice for

simultaneous visualization of medium spiny neurons in the direct and indirect pathways of

the basal ganglia. J Neurosci 28:2681–2685.

Sibley DR (1999) New insights into dopaminergic receptor function using antisense and

genetically altered animals. Annu Rev Pharmacol Toxicol 39:313–341.

216

Snyder GL, Allen PB, Fienberg AA, Valle CG, Huganir RL, Nairn AC, Greengard P (2000)

Regulation of phosphorylation of the GluR1 AMPA receptor in the neostriatum by

dopamine and psychostimulants in vivo. J Neurosci 20:4480–4488.

So CH, Varghese G, Curley KJ, Kong MMC, Alijaniaram M, Ji X, Nguyen T, O’dowd BF,

George SR (2005) D1 and D2 dopamine receptors form heterooligomers and cointernalize

after selective activation of either receptor. Mol Pharmacol 68:568–578.

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 75:843–854.

So CH, Verma V, O’Dowd BF, George SR (2007) Desensitization of the dopamine D1 and D2

receptor hetero-oligomer mediated calcium signal by agonist occupancy of either receptor.

Mol Pharmacol 72:450–462.

Sokoloff P, Andrieux M, Besançon R, Pilon C, Martres MP, Giros B, Schwartz JC (1992)

Pharmacology of human dopamine D3 receptor expressed in a mammalian cell line:

comparison with D2 receptor. Eur J Pharmacol 225:331–337.

Solomon RL, Corbit JD (1974) An opponent-process theory of motivation. Temporal dynamics

of affect. Psychol Rev 81:119–145.

Spano PF, Govoni S, Trabucchi M (1978) Studies on the pharmacological properties of

dopamine receptors in various areas of the central nervous system. Adv Biochem

Psychopharmacol 19:155–165.

Stamatakis AM, Jennings JH, Ung RL, Blair GA, Weinberg RJ, Neve RL, Boyce F, Mattis J,

Ramakrishnan C, Deisseroth K, Stuber GD (2013) A unique population of ventral

tegmental area neurons inhibits the lateral habenula to promote reward. Neuron 80:1039–

1053.

Stamatakis AM, Stuber GD (2012) Activation of lateral habenula inputs to the ventral midbrain

promotes behavioral avoidance. Nat Neurosci 15:1105–1107.

Steketee JD, Kalivas PW (2011) Drug wanting: behavioral sensitization and relapse to drug-

seeking behavior. Pharmacol Rev 63:348–365.

Stewart J, Rajabi H (1996) Initial increases in extracellular dopamine in the ventral tegmental

area provide a mechanism for the development of desipramine-induced sensitization

within the midbrain dopamine system. Synapse 23:258–264.

Stipanovich A, Valjent E, Matamales M, Nishi A, Ahn J-H, Maroteaux M, Bertran-Gonzalez- J,

Brami-Cherrier K, Enslen H, Corbille A-G, Filhol O, Nairn AC, Greengard P, Herve D,

217

Girault J-A (2008) A phosphatase cascade by which natural rewards and drugs of abuse

regulate nucleosomal response in the mouse. Nature 453:879–884.

Strathmann M, Simon MI (1990) G protein diversity: a distinct class of alpha subunits is

present in vertebrates and invertebrates. Proc Natl Acad Sci U S A 87:9113–9117.

Sunahara RK, Guan HC, O’Dowd BF, Seeman P, Laurier LG, Ng G, George SR, Torchia J,

Van Tol HH, Niznik HB (1991) Cloning of the gene for a human dopamine D5 receptor

with higher affinity for dopamine than D1. Nature 350:614–619.

Surmeier DJ, Ding J, Day M, Wang Z, Shen W (2007) D1 and D2 dopamine-receptor

modulation of striatal glutamatergic signaling in striatal medium spiny neurons. Trends

Neurosci 30:228–235.

Suto N, Austin JD, Tanabe LM, Kramer MK, Wright DA, Vezina P (2002) Previous exposure

to VTA amphetamine enhances cocaine self-administration under a progressive ratio

schedule in a D1 dopamine receptor dependent manner. Neuropsychopharmacology

27:970–979.

Svenningsson P, Nairn AC, Greengard P (2005) DARPP-32 mediates the actions of multiple

drugs of abuse. AAPS J 7:E353–E360.

Svenningsson P, Nishi A, Fisone G, Girault J-A, Nairn AC, Greengard P (2004) DARPP-32:

an integrator of neurotransmission. Annu Rev Pharmacol Toxicol 44:269–296.

Tan KR, Yvon C, Turiault M, Mirzabekov JJ, Doehner J, Labouèbe G, Deisseroth K, Tye KM,

Lüscher C (2012) GABA neurons of the VTA drive conditioned place aversion. Neuron

73:1173–1183.

Taylor JR, Lynch WJ, Sanchez H, Olausson P, Nestler EJ, Bibb JA (2007) Inhibition of Cdk5

in the nucleus accumbens enhances the locomotor-activating and incentive-motivational

effects of cocaine. Proc Natl Acad Sci U S A 104:4147–4152.

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

522:3308–3334.

Thomas MJ, Kalivas PW, Shaham Y (2008) Neuroplasticity in the mesolimbic dopamine

system and cocaine addiction. Br J Pharmacol 154:327–342.

Tiberi M, Jarvie KR, Silvia C, Falardeau P, Gingrich JA, Godinot N, Bertrand L, Yang-Feng

TL, Fremeau RT, Caron MG (1991) Cloning, molecular characterization, and

chromosomal assignment of a gene encoding a second D1 dopamine receptor subtype:

differential expression pattern in rat brain compared with the D1A receptor. Proc Natl

Acad Sci U S A 88:7491–7495.

218

Tol HHM Van, Wu CM, Guan H-C, Ohara K, Bunzow JR, Civelli O, Kennedy J, Seeman P,

Niznik HB, Jovanovic V (1992) Multiple dopamine D4 receptor variants in the human

population. Nature 358:149–152.

Tomiyama K, McNamara FN, Clifford JJ, Kinsella A, Drago J, Tighe O, Croke DT,

Koshikawa N, Waddington JL (2002) Phenotypic resolution of spontaneous and D1-like

agonist-induced orofacial movement topographies in congenic dopamine D1A receptor

“knockout” mice. Neuropharmacology 42:644–652.

Tropea TF, Kosofsky BE, Rajadhyaksha AM (2008) Enhanced CREB and DARPP-32

phosphorylation in the nucleus accumbens and CREB, ERK, and GluR1 phosphorylation

in the dorsal hippocampus is associated with cocaine-conditioned place preference

behavior. J Neurochem 106:1780–1790.

Tsai H-C, Zhang F, Adamantidis A, Stuber GD, Bonci A, de Lecea L, Deisseroth K (2009)

Phasic firing in dopaminergic neurons is sufficient for behavioral conditioning. Science

324:1080–1084.

Tsuji Y, Shimada Y, Takeshita T, Kajimura N, Nomura S, Sekiyama N, Otomo J, Usukura J,

Nakanishi S, Jingami H (2000) Cryptic dimer interface and domain organization of the

extracellular region of metabotropic glutamate receptor subtype 1. J Biol Chem

275:28144–28151.

Tye KM, Mirzabekov JJ, Warden MR, Ferenczi E a, Tsai H-C, Finkelstein J, Kim S-Y,

Adhikari A, Thompson KR, Andalman AS, Gunaydin L a, Witten IB, Deisseroth K (2013)

Dopamine neurons modulate neural encoding and expression of depression-related

behaviour. Nature 493:537–541.

Tzschentke TM (1998) Measuring reward with the conditioned place preference paradigm: a

comprehensive review of drug effects, recent progress and new issues. Prog Neurobiol

56:613–672.

Undie AS, Weinstock J, Sarau HM, Friedman E (1994) Evidence for a distinct D1-like

dopamine receptor that couples to activation of phosphoinositide metabolism in brain. J

Neurochem 62:2045–2048.

Ungless MA, Grace AA (2012) Are you or aren’t you? Challenges associated with

physiologically identifying dopamine neurons. Trends Neurosci 35:422–430.

Ungless MA, Magill PJ, Bolam JP (2004) Uniform inhibition of dopamine neurons in the

ventral tegmental area by aversive stimuli. Science 303:2040–2042.

Usiello A, Baik JH, Rougé-Pont F, Picetti R, Dierich A, LeMeur M, Piazza P V, Borrelli E

(2000) Distinct functions of the two isoforms of dopamine D2 receptors. Nature 408:199–

203.

219

Usuda I, Tanaka K, Chiba T (1998) Efferent projections of the nucleus accumbens in the rat

with special reference to subdivision of the nucleus: biotinylated dextran amine study.

Brain Res 797:73–93.

Vaidehi N, Bhattacharya S (2011) Multiscale computational methods for mapping

conformational ensembles of G-protein-coupled receptors. Adv Protein Chem Struct Biol

85:253–280.

Vaidehi N, Bhattacharya S, Larsen AB (2014) Structure and dynamics of G-protein coupled

receptors. Adv Exp Med Biol 796:37–54.

Valjent E, Aubier B, Corbillé A-G, Brami-Cherrier K, Caboche J, Topilko P, Girault J-A,

Hervé D (2006a) Plasticity-associated gene Krox24/Zif268 is required for long-lasting

behavioral effects of cocaine. J Neurosci 26:4956–4960.

Valjent E, Bertran-Gonzalez J, Hervé D, Fisone G, Girault J-A (2009) Looking BAC at striatal

signaling: cell-specific analysis in new transgenic mice. Trends Neurosci 32:538–547.

Valjent E, Corvol JC, Pages C, Besson MJ, Maldonado R, Caboche J (2000) Involvement of

the extracellular signal-regulated kinase cascade for cocaine-rewarding properties. J

Neurosci 20:8701–8709.

Valjent E, Corvol J-C, Trzaskos JM, Girault J-A, Hervé D (2006b) Role of the ERK pathway

in psychostimulant-induced locomotor sensitization. BMC Neurosci 7:20.

Valjent E, Pascoli V, Svenningsson P, Paul S, Enslen H, Corvol J-C, Stipanovich A, Caboche J,

Lombroso PJ, Nairn AC, Greengard P, Hervé D, Girault J-A (2005) Regulation of a

protein phosphatase cascade allows convergent dopamine and glutamate signals to

activate ERK in the striatum. Proc Natl Acad Sci U S A 102:491–496.

Van Zessen R, Phillips JL, Budygin EA, Stuber GD (2012) Activation of VTA GABA Neurons

Disrupts Reward Consumption. Neuron 73:1184–1194.

Verma V, Hasbi A, O&apos;Dowd BF, George SR (2010) Dopamine D1-D2 receptor

heteromer-mediated calcium release is desensitized by D1 receptor occupancy with or

without signal activation: Dual functional regulation by G protein-coupled receptor kinase

2. J Biol Chem 285:35092–35103.

Vezina P (2004) Sensitization of midbrain dopamine neuron reactivity and the self-

administration of psychomotor stimulant drugs. In: Neuroscience and Biobehavioral

Reviews, pp 827–839.

Vezina P, Lorrain DS, Arnold GM, Austin JD, Suto N (2002) Sensitization of midbrain

dopamine neuron reactivity promotes the pursuit of amphetamine. J Neurosci 22:4654–

4662.

220

Vogel G, Neill D, Hagler M, Kors D, Hartley P (1990) Decreased intracranial self-stimulation

in a new animal model of endogenous depression. Neurosci Biobehav Rev 14:65–68.

Volkow ND, Wang G-J, Fowler JS, Telang F (2008) Overlapping neuronal circuits in addiction

and obesity: evidence of systems pathology. Philos Trans R Soc Lond B Biol Sci

363:3191–3200.

Von Heijne G (1991) Proline kinks in transmembrane alpha-helices. J Mol Biol 218:499–503.

Walsh JJ, Han MH (2014) The heterogeneity of ventral tegmental area neurons: Projection

functions in a mood-related context. Neuroscience 282C:101–108.

Wang HB, Deng YP, Reiner A (2007) In situ hybridization histochemical and

immunohistochemical evidence that striatal projection neurons co-containing substance P

and enkephalin are overrepresented in the striosomal compartment of striatum in rats.

Neurosci Lett 425:195–199.

Wang HB, Laverghetta A V., Foehring R, Deng YP, Sun Z, Yamamoto K, Lei WL, Jiao Y,

Reiner A (2006) Single-cell RT-PCR, in situ hybridization histochemical, and

immunohistochemical studies of substance P and enkephalin co-occurrence in striatal

projection neurons in rats. J Chem Neuroanat 31:178–199.

Wang HY, Undie AS, Friedman E (1995) Evidence for the coupling of Gq protein to D1-like

dopamine sites in rat striatum: possible role in dopamine-mediated inositol phosphate

formation. Mol Pharmacol 48:988–994.

Wang J, Bina RW, Wingard JC, Terwilliger EF, Hammer RP, Nikulina EM (2014) Knockdown

of tropomyosin-related kinase B receptor expression in the nucleus accumbens shell

prevents intermittent social defeat stress-induced cross-sensitization to amphetamine in

rats. Eur J Neurosci 39:1009–1017.

West CH, Michael RP (1986) Acquisition of intracranial self-stimulation in medial prefrontal

cortex of rats facilitated by amphetamine. Pharmacol Biochem Behav 24:1617–1622.

Wettschureck N, Offermanns S (2005) Mammalian G proteins and their cell type specific

functions. Physiol Rev 85:1159–1204.

White FJ, Bednarz LM, Wachtel SR, Hjorth S, Brooderson RJ (1988) Is stimulation of both D1

and D2 receptors necessary for the expression of dopamine-mediated behaviors?

Pharmacol Biochem Behav 30:189–193.

White FJ, Wang RY (1984) Electrophysiological evidence for A10 dopamine autoreceptor

subsensitivity following chronic D-amphetamine treatment. Brain Res 309:283–292.

221

White JF, Noinaj N, Shibata Y, Love J, Kloss B, Xu F, Gvozdenovic-Jeremic J, Shah P,

Shiloach J, Tate CG, Grisshammer R (2012) Structure of the agonist-bound neurotensin

receptor. Nature 490:508–513.

Whorton MR, Bokoch MP, Rasmussen SGF, Huang B, Zare RN, Kobilka B, Sunahara RK

(2007) A monomeric G protein-coupled receptor isolated in a high-density lipoprotein

particle efficiently activates its G protein. Proc Natl Acad Sci 104:7682–7687.

Whorton MR, Jastrzebska B, Park PS-H, Fotiadis D, Engel A, Palczewski K, Sunahara RK

(2008) Efficient coupling of transducin to monomeric rhodopsin in a phospholipid bilayer.

J Biol Chem 283:4387–4394.

Williams C, Hill SJ (2009) GPCR signaling: understanding the pathway to successful drug

discovery. Methods Mol Biol 552:39–50.

Willner P, Towell A, Sampson D, Sophokleous S, Muscat R (1987a) Reduction of sucrose

preference by chronic unpredictable mild stress, and its restoration by a tricyclic

antidepressant. Psychopharmacology (Berl) 93:358–364.

Willner P, Towell A, Sampson D, Sophokleous S, Muscat R (1987b) Reduction of sucrose

preference by chronic unpredictable mild stress, and its restoration by a tricyclic

antidepressant. Psychopharmacology (Berl) 93:358–364.

Wise RA (2004) Dopamine, learning and motivation. Nat Rev Neurosci 5:483–494.

Wise RA, Munn E (1995) Withdrawal from chronic amphetamine elevates baseline intracranial

self-stimulation thresholds. Psychopharmacology (Berl) 117:130–136.

Wolf ME, Sun X, Mangiavacchi S, Chao SZ (2004) Psychomotor stimulants and neuronal

plasticity. Neuropharmacology 47 Suppl 1:61–79.

Wolf ME, White FJ, Hu XT (1994) MK-801 prevents alterations in the mesoaccumbens

dopamine system associated with behavioral sensitization to amphetamine. J Neurosci

14:1735–1745.

Wolf ME, Xue CJ (1998) Amphetamine and D1 dopamine receptor agonists produce biphasic

effects on glutamate efflux in rat ventral tegmental area: modification by repeated

amphetamine administration. J Neurochem 70:198–209.

Woolverton WL (1992) Cocaine self-administration: pharmacology and behavior. NIDA Res

Monogr 124:189–202.

Wydra K, Golembiowska K, Zaniewska M, Kamińska K, Ferraro L, Fuxe K, Filip M (2013)

Accumbal and pallidal dopamine, glutamate and GABA overflow during cocaine self-

administration and its extinction in rats. Addict Biol 18:307–324.

222

Xi Z-X, Ramamoorthy S, Shen H, Lake R, Samuvel DJ, Kalivas PW (2003) GABA

transmission in the nucleus accumbens is altered after withdrawal from repeated cocaine. J

Neurosci 23:3498–3505.

Yamaguchi T, Sheen W, Morales M (2007) Glutamatergic neurons are present in the rat ventral

tegmental area. Eur J Neurosci 25:106–118.

Yan Z, Hsieh-Wilson L, Feng J, Tomizawa K, Allen PB, Fienberg AA, Nairn AC, Greengard P

(1999) Protein phosphatase 1 modulation of neostriatal AMPA channels: regulation by

DARPP-32 and spinophilin. Nat Neurosci 2:13–17.

Yang L-M, Hu B, Xia Y-H, Zhang B-L, Zhao H (2008) Lateral habenula lesions improve the

behavioral response in depressed rats via increasing the serotonin level in dorsal raphe

nucleus. Behav Brain Res 188:84–90.

Yger M, Girault J-A (2011) DARPP-32, Jack of All Trades… Master of Which? Front Behav

Neurosci 5:56.

Yohannan S, Faham S, Yang D, Whitelegge JP, Bowie JU (2004) The evolution of

transmembrane helix kinks and the structural diversity of G protein-coupled receptors.

Proc Natl Acad Sci U S A 101:959–963.

Zachariou V, Benoit-Marand M, Allen PB, Ingrassia P, Fienberg AA, Gonon F, Greengard P,

Picciotto MR (2002) Reduction of cocaine place preference in mice lacking the protein

phosphatase 1 inhibitors DARPP 32 or Inhibitor 1. Biol Psychiatry 51:612–620.

Zachariou V, Sgambato-Faure V, Sasaki T, Svenningsson P, Berton O, Fienberg AA, Nairn AC,

Greengard P, Nestler EJ (2006) Phosphorylation of DARPP-32 at Threonine-34 is

required for cocaine action. Neuropsychopharmacology 31:555–562.

Zahm DS, Heimer L (1993) Specificity in the efferent projections of the nucleus accumbens in

the rat: comparison of the rostral pole projection patterns with those of the core and shell.

J Comp Neurol 327:220–232.

Zalewska M, Siara M, Sajewicz W (2014) G protein-coupled receptors: abnormalities in signal

transmission, disease states and pharmacotherapy. Acta Pol Pharm 71:229–243.

Zhang H, Craciun LC, Mirshahi T, Rohács T, Lopes CMB, Jin T, Logothetis DE (2003) PIP2

activates KCNQ channels, and its hydrolysis underlies receptor-mediated inhibition of M

currents. Neuron 37:963–975.

Zhang L, Bose P, Warren RA (2014) Dopamine preferentially inhibits NMDA receptor-

mediated EPSCs by acting on presynaptic D1 receptors in nucleus accumbens during

postnatal development. PLoS One 9:e86970.

223

Zhang L, Lou D, Jiao H, Zhang D, Wang X, Xia Y, Zhang J, Xu M (2004) Cocaine-induced

intracellular signaling and gene expression are oppositely regulated by the dopamine D1

and D3 receptors. J Neurosci 24:3344–3354.

Zhang Y, Svenningsson P, Picetti R, Schlussman SD, Nairn AC, Ho A, Greengard P, Kreek MJ

(2006) Cocaine self-administration in mice is inversely related to phosphorylation at

Thr34 (protein kinase A site) and Ser130 (kinase CK1 site) of DARPP-32. J Neurosci

26:2645–2651.

Zhang Z-W, Burke MW, Calakos N, Beaulieu J-M, Vaucher E (2010) Confocal Analysis of

Cholinergic and Dopaminergic Inputs onto Pyramidal Cells in the Prefrontal Cortex of

Rodents. Front Neuroanat 4:21.

Zhen X, Goswami S, Abdali SA, Gil M, Bakshi K, Friedman E (2004) Regulation of cyclin-

dependent kinase 5 and calcium/calmodulin-dependent protein kinase II by

phosphatidylinositol-linked dopamine receptor in rat brain. Mol Pharmacol 66:1500–1507.

Zheng F, Luo Y, Wang H (2009) Regulation of brain-derived neurotrophic factor-mediated

transcription of the immediate early gene Arc by intracellular calcium and calmodulin. J

Neurosci Res 87:380–392.

Zhou Z, Hong EJ, Cohen S, Zhao W-N, Ho H-YH, Schmidt L, Chen WG, Lin Y, Savner E,

Griffith EC, Hu L, Steen JAJ, Weitz CJ, Greenberg ME (2006) Brain-specific

phosphorylation of MeCP2 regulates activity-dependent Bdnf transcription, dendritic

growth, and spine maturation. Neuron 52:255–269.