Novel Regulatory Mechanisms of D1 Dopamine Receptor Maturation and Internalization

Michael Ming Chuen Kong

A thesis submitted in conformity with the requirements

for the degree of Doctor of Philosophy

Graduate Department of Pharmacology

University of Toronto

Novel Regulatory Mechanisms of D1 Dopamine Receptor Maturation and Internalization

Michael Ming Chuen Kong Degree of Doctor of Philosophy, 2008 Department of Pharmacology University of Toronto

ABSTRACT

Dopamine is the most abundant catecholamine neurotransmitter in the mammalian

brain and controls various physiological processes. The D1 dopamine receptor (D1DR) is the

predominant dopamine receptor in the brain and traditionally couples to stimulatory G

proteins, such as Gs, to activate adenylyl cyclase and generate cAMP.

Although the trafficking itinerary of ER/Golgi maturation, agonist-induced internalization, and recycling/degradation are features common to many G protein-coupled receptors (GPCRs), the molecular regulation of these individual processes for the D1DR is not fully elucidated. Many GPCRs have been shown to form homo-oligomers; the work presented in this thesis explores how multimerization of D1DR has a role in regulating how these receptors are trafficked to the plasma membrane. In addition, the regulation of D1DR internalization is investigated in the context of emerging evidence highlighting the importance of lipid rafts.

Using strategically designed point mutations of the D1DR, specific receptor mutants were found to intracellularly sequester the wild-type receptor by oligomerization. This level of scrutiny by the quality control machinery in the cell could be circumvented by treatment with cell permeable dopaminergic agonists, but not antagonists or inverse agonists. This finding suggests that specific conformational requirements must be achieved before full

ii maturation and anterograde trafficking of the D1DR can proceed. Furthermore, it was

determined that cell surface bound D1DRs could internalize through a novel clathrin

independent pathway that required binding to the scaffolding protein, caveolin-1. This

interaction with caveolin-1 was identified in whole rat brain and was found to require a

putative caveolin binding motif in transmembrane domain 7. Palmitoylation of D1DR was

found to regulate the rate of agonist-induced caveolae mediated internalization. Finally, we determined that the integrity of caveolae was important in regulating cAMP signaling through D1DR.

These findings provide novel insight into the trafficking requirements of newly synthesized D1DRs as well as alternative mechanisms of regulation of receptors after agonist activation. The oligomerization of GPCRs and the localization of GPCRs in lipid rafts represent two emerging concepts important to many aspects of GPCR function. Future work

aimed at integrating these overlapping processes will further our understanding of this

important group of cell surface receptors.

iii ACKNOWLEDGEMENTS

The compilation of work embodied in this thesis could not have been possible without the help

of many individuals, all of whom I am indebted to.

I’d first like to thank members of the examination committee: Dr. Stephane Angers, Dr. Peter

Chidiac, and Dr. Denis Grant, for a very stimulating discussion and for providing unique

perspectives on this work. I’d also like to thank Dr. Reinhart Reithmeier and Dr. Bernard Schimmer

for all of their guidance and advice throughout my doctoral studies.

I would like to thank all members of the lab for their outstanding expertise and for providing a

congenial environment to work and learn in. In particular, I’d like to thank Theresa Fan, Ahmed

Hasbi, Tuan Nguyen, and George Varghese for their technical skills and for sometimes saving me

from the agony of troubleshooting. I’d also like to send a special thanks to Kevin Curley, Samuel

Lee, Dennis Lee, Jason Juhasz, Jennifer Ng, Ryan Rajaram, Asim Rashid, Christopher So, and

Vaneeta Verma for their banter, advice, and most importantly, their friendship, all of which has

made graduate school go by so quickly.

To my supervisors, Dr. Susan George and Dr. Brian O’Dowd, I am grateful for their outstanding

mentorship and for giving me the independence to develop as a scientist. Your confidence in my

abilities is highly valued and very much appreciated.

To my parents, Robert and Vivian Kong, who have been so supportive and patient throughout every aspect of my life and for providing an endless source of encouragement and love. This work is dedicated to you.

And to my wife and best friend, Siu Lan Lee, for not only her patience and understanding through this endeavour but for always having an open ear and an open heart.

iv PREVIOUSLY COPYRIGHTED MATERIAL

Some parts of this thesis have been reproduced in whole or in part from the following sources with permission:

Gouldson PR, Higgs C, Smith RE, Dean MK, Gkoutos GV, Reynolds CA 2000. Dimerization and domain swapping in G protein-coupled receptors: a computational study. Neuropsychopharmacology, Oct;23(4 Suppl):S60-77. Copyright © by Macmillan Publishers.

Fotiadis D, Liang Y, Filipek S, Saperstein DA, Engel A, Palczewski K. 2004. The G protein-coupled receptor rhodopsin in the native membrane. FEBS Letters, Apr 30; 564(3):281-8. Copyright © by Elsevier Limited.

Kong MM, So CH, O’Dowd BF, and George SR. The Role of Oligomerization in G Protein-Coupled Receptor Maturation, The G Protein-Coupled Receptors Handbook; Humana Press, 2005. Copyright © by Springer Science and Business Media.

So CH, Varghese G, Curley KJ, Kong MM, Alijaniaram M, Ji X, Nguyen T, O'Dowd BF, and George SR. 2005. D1 and D2 dopamine receptors form heterooligomers and cointernalize after selective activation of either receptor. Molecular Pharmacology, Sept; 68(3): 568-78. Copyright © 2005 by the American Society for Pharmacology and Experimental Therapeutics.

O'Dowd BF, Ji X, Alijaniaram M, Rajaram RD, Kong MM, Rashid A, Nguyen T, and George SR. 2005. Dopamine receptor oligomerization visualized in living cells. Journal of Biological Chemistry, Nov 4; 280 (44): 37225-35. Copyright © 2005 by the American Society for Biochemistry and Molecular Biology.

Kong MM, Fan T, Varghese G, O’Dowd BF, and George SR. 2006. Agonist-induced cell surface trafficking of an intracellularly sequestered D1 dopamine receptor homo-oligomer. Molecular Pharmacology, Jul; 70(1): 78-89. Copyright © 2005 by the American Society for Pharmacology and Experimental Therapeutics.

Rashid AJ, So CH, Kong MM, Furtak T, Cheng R, O’Dowd BF,and George SR. 2006. A D1-D2 dopamine receptor heterooligomer with unique pharmacology is coupled to rapid activation of Gq/11 in the striatum. Proceedings of the National Academy of Sciences, Jan 9; 104(2):654-9. Copyright © 2007 by the National Academy of Sciences

Kong MM, Hasbi A, Mattocks M, Fan T, O’Dowd BF, George SR. Regulation of D1 Dopamine Receptor Trafficking and Signaling by Caveolin-1. Molecular Pharmacology in press. Copyright © 2007 by the American Society for Pharmacology and Experimental Therapeutics

TABLE OF CONTENTS

ABSTRACT ii

ACKNOWLEDGEMENTS iv

PREVIOUSLY COPYRIGHTED MATERIAL v

LIST OF REFEREED AND NON-REFEREED PUBLICATIONS 1

LIST OF ABSTRACTS 2

ABBREVIATIONS 4

LIST OF FIGURES 8

LIST OF TABLES 11

1 GENERAL INTRODUCTION 14

1.1 G Protein-Coupled Receptor Overview 14 1.2 Traditional Classification of GPCRs 16 1.3 G protein and Effector Signaling 19 1.4 Biosynthesis of GPCRs 20 1.4.1 Molecular Chaperones and Anterograde Trafficking 21 1.5 Receptor Activation and Desensitization 22 1.6 Internalization of GPCRs 25 1.6.1 Post Endocytic Sorting 26 1.7 Palmitoylation of GPCRs 27 1.8 Dopamine Receptor Subfamily 29 1.8.1 D1 Dopamine Receptor Structure 30 1.8.2 Pharmacology of the D1 Dopamine Receptor 32 1.8.3 Signal Transduction of the D1 Dopamine Receptor 33 1.9 Dopamine Neurotransmission in the Brain 35 1.10 Anatomical Distribution of the D1 Dopamine Receptor 36 1.10.1 Physiological Role of D1 Dopamine Receptors 37 1.11 Introduction to G Protein-Coupled Receptor Oligomerization 39 1.12 Evidence for G Protein-Coupled Receptor Oligomerization 40 1.12.1 Receptor Complementation 40 1.12.2 Co-immunoprecipitation 41 1.12.3 Novel Pharmacology 44 1.12.4 Receptor Trafficking Studies 45 1.12.5 Resonance Energy Transfer 46 1.12.6 Atomic Force Microscopy 49 1.12.7 X-ray Crystallography 50 1.13 Structural Features of G Protein-Coupled Receptor Oligomers 50 1.13.1 Disulphide Bonds 51 1.13.2 Transmembrane Interactions 52

vii 1.13.3 Intracellular and Extracellular Domain Interactions 53 1.14 Mechanisms of Agonist-Induced Activation of G Protein-Coupled Receptor Oligomers 54 1.15 G Protein-Coupled Receptor Oligomers In Native Brain Tissue 56 1.16 Introduction to Lipid Rafts 58 1.17 Platforms for Signaling Complex Assembly 59 1.17.1 Nitric Oxide Synthase Signaling Pathway 61 1.17.2 The Ras GTPase-Raf Signaling Pathway 62 1.17.3 Heterotrimeric GTP Binding Protein Signaling 63 1.18 Transcytosis 63 1.19 Endocytosis 64 1.20 Potocytosis 66 1.21 Thesis Rationale and Research Objectives 67

2 MATERIALS AND METHODS 70

2.1 Experimental procedures for Chapter 3 70 2.1.1 DNA constructs and Site-Directed Mutagenesis 70 2.1.2 Cell culture and DNA transfection 71 2.1.3 Membrane preparation 71 2.1.4 Drugs 71 2.1.5 Radioligand Saturation Binding 72 2.1.6 Whole Cell Radioligand Binding 72 2.1.7 Cell Surface Fluorometric Analysis 73 2.1.8 Co-immunoprecipitation 73 2.1.9 Gel Electrophoresis and Immunoblotting 74 2.1.10 Cell Surface Biotinylation 74 2.1.11 Whole Cell cAMP Determination 75 2.1.12 Immunofluorescence Microscopy 76 2.1.13 Co-localization Image Analysis 76 2.1.14 Time-resolved Fluorescence Resonance Energy Transfer (trFRET) 77 2.1.15 Statistical Analysis 77 2.2 Experimental procedures for Chapter 4 and 5 77 2.2.1 Chemicals 77 2.2.2 DNA constructs and Site-Directed Mutagenesis 78 2.2.3 Cell Culture and DNA transfection 79 2.2.4 Detergent-free Sucrose Gradient Fractionation 79 2.2.5 Co-immunoprecipitation 80 2.2.6 Cell Surface Biotinylation 81 2.2.7 Membrane Preparation 82 2.2.8 Radioligand Binding 82 2.2.9 Immunocytochemistry and Confocal Microscopy 83 2.2.10 Bioluminescence Resonance Energy Transfer (BRET) 84 2.2.11 cAMP Accumulation 84 2.2.12 GTPγS Binding Assay 85 2.2.13 Statistical Analysis 85

3 ASSEMBLY OF D1 DOPAMINE RECEPTOR OLIGOMERS 86

3.1 Introduction 86 3.2 Detection of D1 Dopamine Receptor Oligomers in HEK293t and COS 7cells 86 3.3 Role of Glycosylation in D1 Receptor Oligomeric Assembly 87 3.4 Generation of D1 Receptors with Key Structural Mutations 89

viii 3.4.1 Pharmacological and Expression Analysis of D1 Receptors with Transmembrane Domain 3 Mutations (D103A, D103E, D120A, D120N) 91 3.4.2 Pharmacological and Expression Analysis of D1 Receptors with Transmembrane Domain 5 Mutations (S198A/S199A, S199A/S202A) 93 3.4.3 Adenylyl Cyclase Activity of D103A, D103E, and S198A/S199A D1DR mutants 96 3.5 Antagonist Rescue of D120A and D120N 98 3.6 Co-expression of D103A and D103E D1DR with wild-type D1DR 98 3.7 Co-expression of S198A/S199A and Apelin receptor with wild-type D1DR 104 3.8 Interaction between D103A and wild-type D1DR 107 3.9 Staggered co-expression of D03A and wild-type D1DR 110 3.10 Agonist-specific Restoration of Wild-type/D103A Oligomer 110 3.11 Cell Surface Analysis of Pharmacologically Rescued Oligomers 112 3.11.1 FRET Analysis of Rescued Oligomers 117 3.12 Discussion 119 3.13 Acknowledgements 131

4 REGULATION OF D1 DOPAMINE RECEPTORS IN LIPID RAFTS 133

4.1 Introduction 133 4.2 Differential Caveolin-1 Expression in HEK293t and COS7 cells 135 4.3 Localization of D1DR in Caveolin-enriched Fractions 137 4.3.1 Co-localization of D1DR with Caveolin-1 140 4.3.2 Interaction of D1DR with Caveolin-1 140 4.4 Agonist-induced Translocation of D1DR into Caveolae 145 4.4.1 Effect of Endocytosis Inhibitors on D1DR Internalization 147 4.4.2 Kinetics of D1DR Internalization 151 4.5 Pharmacological Characterization of D1DR Mutants Lacking an Intact Caveolin Binding Motif 154 4.5.1 Role of Caveolin Binding Motif in D1DR Internalization 157 4.6 Role of Caveolae in D1DR Signaling 159 4.7 Caveolin Binding Mutants Exhibit Constitutive Desensitization 161 4.8 Discussion 163 4.9 Acknowledgements 172

5 ROLE OF PALMITOYLATION IN D1DR INTERNALIZATION 173

5.1 Introduction 173 5.1.1 Role of Palmitoylation in Association of Proteins with Caveolae 175 5.2 Pharmacological Analysis of a Palmitoylation Deficient D1 Receptor (C347A/C351A) 176 5.3 Kinetics of C347A/C351A Internalization 176 5.4 Effect of Endocytosis Inhibitors on Internalization of C347A/C351A 178 5.5 Subcellular Localization of C347A/C351A 180 5.6 Discussion 182

6 RELATED STUDIES 187

6.1 Introduction 187 6.2 D1 and D2 Dopamine Receptors Form Hetero-oligomers and Co-internalize After Selective Activation of Either Receptor (So et al, Mol Pharmacology, 2005) 187 6.2.1 Summary 187 6.2.2 Contribution by thesis author 187 6.3 Dopamine Receptor Oligomerization Visualized in Living Cells (O’Dowd et al, J Biol Chem, 2005) 200

ix 6.3.1 Summary 200 6.3.2 Contribution by thesis author 200 6.4 D1-D2 Dopamine Receptor Hetero-oligomers With Unique Pharmacology Are Coupled to Rapid Activation of Gq/11 in the Striatum (Rashid et al, PNAS, 2006) 212 6.4.1 Summary 212 6.4.2 Contribution by thesis author 212

7 GENERAL CONCLUSIONS AND FUTURE DIRECTIONS 219

7.1 General Conclusions 219 7.1.1 Conformational Checkpoints Regulate D1 Dopamine Receptor Homo-oligomerization 220 7.1.2 D1 Dopamine Receptors are Localized in and Functionally Regulated by Caveolae 223 7.1.3 Palmitoylation of D1 Dopamine Receptors Regulates Caveolar Internalization 225 7.2 Final Thoughts 227

8 REFERENCES 228

x LIST OF REFEREED AND NON-REFEREED PUBLICATIONS

The following bibliography lists the publications that arose from work in this thesis.

So CH, Varghese G, Curley KJ, Kong MM, Alijaniaram M, Ji X, Nguyen T, O'Dowd BF, and George SR. D1 and D2 dopamine receptors form heterooligomers and cointernalize after selective activation of either receptor. Molecular Pharmacology, Sept; 68(3): 568-78. (2005)

Kong MM, Fan T, Varghese G, O’Dowd BF, and George SR. Agonist-induced cell surface trafficking of an intracellularly sequestered D1 dopamine receptor homo-oligomer. Molecular Pharmacology, Jul; 70(1): 78-89. (2006)

Rashid AJ, So CH, Kong MM, Furtak T, Cheng R, O’Dowd BF, and George SR. 2006. A D1- D2 dopamine receptor heterooligomer with unique pharmacology is coupled to rapid activation of Gq/11 in the striatum. Proceedings of the National Academy of Sciences, Jan 9; 104(2):654-9. (2007)

Kong MM, Hasbi A, Mattocks M, Fan T, O’Dowd BF, George SR. Regulation of D1 Dopamine Receptor Trafficking and Signaling by Caveolin-1. Molecular Pharmacology, Nov; 72(5): 1157- 70. (2007)

LIST OF ABSTRACTS

The following bibliography lists the presentations, in reverse chronological order, by Michael

M.C. Kong at scientific meetings and conferences.

Kong MM, Hasbi A, Mattocks M, O’Dowd BF, George SR. Functional Regulation of the D1 Dopamine Receptor by Caveolin-1. Visions in Pharmacology, Toronto, ON (2006).

Kong MM, Yassa M, Hasbi A, O’Dowd BF, George SR. Functional Regulation of the D1 Dopamine Receptor by Caveolin-1. Society for Neuroscience Annual Meeting, Washington, DC (2005).

Kong MM, Yassa M, Hasbi A, O’Dowd BF, George SR. Functional Regulation of the D1 Dopamine Receptor by Caveolin-1. Annual Great Lakes GPCR Retreat, Montebello, QC (2005).

Kong MM, Yassa M, Hasbi A, O’Dowd BF, George SR. Functional Regulation of the D1 Dopamine Receptor by Caveolin-1. Visions in Pharmacology, Toronto, ON (2005).

Kong MM, Fan T, Varghese G, O’Dowd BF, George SR. Pharmacological Rescue of an Intracellularly Sequestered D1 Dopamine Receptor Oligomer. Annual Great Lakes GPCR Retreat, Bromont, QC (2004).

Kong MM, Fan T, O’Dowd BF, George SR. Pharmacological Rescue of an Intracellularly Sequestered D1 Dopamine Receptor Oligomer. University of Toronto, Division of Endocrinology Clinical and Scientific Day, Toronto, ON (2003).

Kong MM, Fan T, O’Dowd BF, George SR. Pharmacological Rescue of an Intracellularly Sequestered D1 Dopamine Receptor Oligomer. Society for Neuroscience Annual Meeting, New Orleans, LA (2003).

Kong MM, Fan T, O’Dowd BF, George SR. Pharmacological Rescue of an Intracellularly Sequestered D1 Dopamine Receptor Oligomer. Annual Great Lakes GPCR Retreat, Honey Harbour, ON (2003).

Kong MM, Fan T, O’Dowd BF, George SR. Pharmacological Rescue of an Intracellularly Sequestered D1 Dopamine Receptor Oligomer. Visions in Pharmacology, Toronto, ON (2003).

Kong MM, Lee SP, Varghese G, O’Dowd BF, George SR. The Conformation of Dopamine D1 and D2 Receptor Homo-oligomers Is a Critical Determinant in Cell Surface Trafficking. Society for Neuroscience Annual Meeting, Orlando, FL (2002).

2 Kong MM, Lee SP, Varghese G, O’Dowd BF, George SR. The Conformation of Dopamine D1 and D2 Receptor Homo-oligomers Is a Critical Determinant in Cell Surface Trafficking. Visions in Pharmacology, Toronto, ON (2002).

Kong MM, Lee SP, O’Dowd BF, George SR. D1 Dopamine Receptor Mutants as Modulators of Wild-type Receptor Expression: Evidence for Oligomerization. Visions in Pharmacology, Toronto, ON (2001).

3 ABBREVIATIONS

2-ME β-mercaptoethanol

5-HT 5-hydroxytryptamine

AFM atomic force microscopy

ANOVA analysis of variance

ARF-6 ADP ribosylation factor-6

ATP adenosine triphosphate

BRET bioluminescence resonance energy transfer

cAMP 3’-5’-cyclic adenosine monophosphate

CBM caveolin binding motif

cDNA complementary deoxyribonucleic acid

CHAPS 3-[(3-Cholamidopropyl)dimethylammonio]-1-propanesulfonate

CHO chinese hamster ovary

CNS central nervous system

C-terminus carboxyl terminus

D1DR D1 dopamine receptor

DAG diacylglycerol

DMSO dimethyl sulfoxide

DNA deoxyribonucleic acid

DOR delta opioid receptor

DTT dithiothreitol

EBD extracellular binding domain

ECL extracellular loop, enhanced chemiluminescence

EDTA ethylenediamine tetraacetic acid

EGTA ethylene glycol tetraacetic acid

EM electron microscopy eNOS endothelial nitric oxide synthase

ER endoplasmic reticulum

FITC fluorescein isothiocyanate

FRET fluorescence resonance energy transfer

GABA γ-aminobutyric acid

GAPs GTPase-activating proteins

GDP guanosine diphosphate

GFP green fluorescent protein

GpA glycophorin A

GPCR G protein-coupled receptor

GRK G protein receptor kinase

GTP guanosine triphosphate

HA hemagglutinin

HC heavy chain

HEK human embryonic kidney

HEPES 4-(2-hydroxyethyl)-1-piperazineethanesulfonic acid

HIR human insulin receptor

HRP horseradish peroxidase

HSP heat shock protein

ICL intracellular loop

IP3 inositol triphosphate

L-Dopa L-dihydroxyphenylalanine

MAPK mitogen-activated protein kinase mβCD methyl-β-cyclodextrin mGluR metabotropic glutamate receptor mRFP monomeric red fluorescent protein mRNA messenger ribonucleic acid

N- terminus amino terminus

PAR1 protease-activated receptor-1

PBS phosphate buffered saline

PCR polymerase chain reaction

PKA protein kinase A

PKC protein kinase C

PLC phospholipase

PVDF polyvinylidene difluoride

RAMPs receptor activity modifying proteins

RGS regulators of G protein signalling

Rh 1 rhodopsin 1

Rluc Renilla luciferase

RT room temperature

SDS-PAGE sodiumdodecylsulfate-polyacrylamide gel electrophoresis

TM transmembrane domain

trFRET time resolved fluorescence resonance energy transfer

TRITC tetramethylrhodamine-5-isothiocyanate

VTA ventral tegmental area

WT wild-type D1 dopamine receptor

7 LIST OF FIGURES

Figure 1-1. The amino acid sequence and putative topology of the D1 dopamine receptor. 17

Figure 1-2. Schematic representation of the classical regulation of GPCR signalling. 23

Figure 1-3. Dopamine receptor signalling 31

Figure 1-4. Contact dimer vs. Domain swapped dimer 42

Figure 1-5. Bioluminescence Resonance Energy Transfer 48

Figure 1-6. Morphology of Caveolae 60

Figure 3-1. Immunoblot of D1DR from membrane preparations of A) HEK293t and B) COS7 cells 88

Figure 3-2. Immunoblot of N5A/N175A from membrane preparations of HEK293t cells 90

Figure 3-3. Analysis of D103A, D103E, and S198A/S199A by A) saturation binding and B) immunblotting from membrane prepartions of HEK293t cells 92

Figure 3-4. Immunoblot of D120A and D120N from membrane preparations of HEK293t cells 94

Figure 3-5. One point saturation binding of S199A/S202A from membrane preparations of HEK 293t cells 95

Figure 3-6. Basal cAMP accumulation of D103A, D103E, and S198A/S199A 97

Figure 3-7. Immunofluorescence imaging of D1DR and D103A 99

Figure 3-8. Immunoblot of D120A and D120N after antagonist treatment 100

Figure 3-9. Effect of increasing co-expressed A) D103A or B) D103E on D1DR binding 102

Figure 3-10. A) Competition binding of dopamine against [3H] SCH23390 and B) effect of increasing co-expressed D103A on cell surface D1DR binding 103

Figure 3-11. Immunoblot of cell surface biotinylated D1DR in presence of increasing co-expressed D103A D1DR and vice versa 105

Figure 3-12. Whole cell fluorometric analysis of D1DR in presence of co-expressed D103A 106

Figure 3-13. Effect of co-expressed S198A/S199A and apelin receptor on D1DR binding 108

Figure 3-14. Co-immunoprecipitation of D1DR and D103A from HEK293t cell lysates 109

Figure 3-15. Effect of staggered co-expression of D103A on D1DR binding 111

Figure 3-16. Effect of extended dopaminergic ligand treatment on A) D1DR binding and B) cell surface D1DR binding from membrane preparations of HEK293t cells co-expressing D1DR and D103A 113

Figure 3-17. Immunofluorescence microscopy of D1DR co-expressed with D103A from HEK293t cells after extended dopaminergic ligand treatment 115

Figure 3-18. Immunofluorescence microscopy of D1DR and calnexin from HEK293t cells coexpressing D1DR and D103A after extended agonist treatment 116

Figure 3-19. Time-resolved fluorescence resonance energy transfer between D1DR and D103A after extended dopaminergic ligand treatment 120

Figure 3-20. Schematic of proposed mechanism of agonist rescue of D1DR/D103A oligomeric complex 126

Figure 3-21. Schematic of GPCR intra- and inter-dimeric interfaces as predicted by atomic force microscopy studies from rhodopsin 132

Figure 4-1. Immunblot of endogenous caveolin-1 from sucrose gradient fractions of HEK293t cells after a A) short and B) long film exposure time 136

Figure 4-2. Immunoblot of endogenous caveolin-1 from sucrose gradient fractions of COS7 cells 138

+ + Figure 4-3. Immunoblot of endogenous Na -K ATPase, Gsα, ARF-6, clathrin HC, and D1DR from sucrose gradient fractions of COS7 cells 139

Figure 4-4. Confocal microscopy of D1DR co-expressed with caveolin-1 in HEK293t cells 141

Figure 4-5. Co-immunoprecipation of D1DR and caveolin-1 from A) COS7 cell lysates and B) whole rat brain lysate 143

Figure 4-6. Bioluminescence resonance energy transfer between co-expressed D1DR and caveolin-1 in HEK293t cells 144

Figure 4-7. Effect of agonist stimulation on D1DR distribution in sucrose gradient fractions of COS7 cell lysates 146

9 Figure 4-8. Effect of inhibition of clathrin or caveolar endocytic function on D1DR internalization detected by whole cell binding 148

Figure 4-9. Effect of inhibition of clathrin or caveolar endocytic function on D1DR internalization detected by cell surface biotinylation 150

Figure 4-10. Caveolar internalization kinetics of D1DR detected by whole cell binding 152

Figure 4-11. Confocal microscopy of agonist stimulated D1DR and caveolin-1 in HEK293t cells 153

Figure 4-12. Immunoblot analysis of cell surface biotinylated D1DR mutants lacking an intact caveolin binding motif 155

Figure 4-13. A) Internalization of D1DR CBM mutants detected by whole cell binding and B) BRET between D1DR and caveolin-1 in presence of D1DR CBM mutants 158

Figure 4-14. Effect of caveolar disruption on cAMP accumulation by A) pretreatment with mβCD or by B) cotransfection with cav1-P132L and C) effect of mβCD 35 on [ S] GTPγS incorporation to Gs 160

Figure 4-15 A-D) Attenuated signaling of CBM mutants due to constitutive desensitization 162

Figure 5-1. Saturation binding analysis of C347A/C351A 177

Figure 5-2. Caveolar internalization kinetics of C347A/C351A 179

Figure 5-3. Effect of inhibition of clathrin or caveolar endocytic function on C347A/C351A internalization detected by whole cell binding 181

Figure 5-4. Immunoblot of C347A/C351A from sucrose gradient fractions of COS7 cells 183

10 LIST OF TABLES

Table 1-1. List of selected therapeutic compounds targeting GPCRs. 15

Table 3-1. Co-localization of D1DR/D103A and D1DR/calnexin quantified by Mander’s correlation coefficients 118

Table 3-2. List of GPCRs that require a chaperone GPCR for functional cell surface expression 123

Table 4-1. Receptor affinities (Kd) and maximum receptor densities (Bmax) of D1DR CBM mutants 156

Table 4-2. List of selected GPCRs and GPCR signalling molecules with a caveolin binding motif 166

11 FORWARD

This thesis examines specific aspects of the trafficking itinerary of the D1 dopamine

receptor (D1DR), the most abundant dopamine receptor in the mammalian brain. Specifically, I

explore the trafficking requirements of D1DR oligomers to the plasma membrane as well as the

regulation of mature D1DRs by caveolae, a type of lipid raft. My hypothesis was that specific

conformational requirements are required for D1DR antergrade trafficking and that caveolae

modulate internalization and signaling of D1DRs.

The Introduction (Chapter 1) provides a general overview of G protein-coupled receptor biology including the biosynthetic, signalling, and post-endocytic sorting pathways that are utilized by this diverse group of cell surface receptors. I then discuss the molecular, biochemical, and pharmacological characteristics of the D1 dopamine receptor and its role in mammalian physiology. This is followed by an introduction to the field of GPCR oligomerization and an overview of the role of lipid rafts in cellular function. The Methods section (Chapter 2) details the experimental procedures used throughout the thesis.

I have divided the experimental work into three sections. The first section (Chapter 4) explores the role of receptor oligomerization in D1DR trafficking to the cell surface. Most of the results in this chapter are published in Kong et al, Molecular Pharmacology (Jul; 70(1): 78-89,

2006) but include additional unpublished data and discussion. In the next section (Chapter 5), I characterize a novel interaction between the scaffolding protein, caveolin-1, and the D1DR. Most of the results in this chapter are currently in Kong et al, Molecular Pharmacology (in press,

2007) but some supplementary data is also discussed. The last section examines the role of palmitoylation in D1DR internalization through caveolae (Chapter 5). Some additional published work related to but not directly involved with my thesis are presented in Chapter 6. Finally, an

12 overall discussion and future directions is presented in Chapter 7 followed by a list of cited

literature in Chapter 8. All of the work shown was performed by myself except where otherwise indicated in the acknowledgements.

13 1 GENERAL INTRODUCTION

1.1 G Protein-Coupled Receptor Overview

G protein-coupled receptors (GPCRs) represent the largest class of cell surface receptors that are encoded by the largest and most conserved gene family (Pierce et al., 2002). There are currently over 800 known GPCRs in the human genome with overlapping expression profiles in various tissues (Fredriksson et al., 2003). GPCRs mediate signal transduction from a diverse

array of molecules including peptides, hormones, neurotransmitters, ions, bitter and sweet taste,

and odorants, through a variety of intracellular effectors. This responsiveness to such a diverse

repertoire of extracellular inputs renders these receptors an important target of many therapeutic

compounds. It is estimated that over 30% of currently marketed drugs are modulators of GPCR

function with 25% of the top 100 selling drugs targeting GPCRs (Drews, 2000). These drugs

typically function as agonists or antagonists to GPCRs for the treatment of various ailments

including asthma, hypertension, peptic ulcers, and schizophrenia (Wise et al., 2002) (Table 1-1).

The classical model of a G protein-coupled receptor signal transduction cascade requires

the binding of an extracellular ligand to the receptor binding pocket. This triggers a

conformational change in the cytoplasmic face of the receptor that allows it to activate an

intracellular heterotrimeric G protein, which can positively or negatively regulate intracellular

effectors or enzymes which, in turn, can modulate second messenger levels. These second

messengers have important effects on various cellular events including transcription, calcium

mobilization, and neurotransmission.

The classical 2-dimensional topography of a G protein-coupled receptor reveals seven

alpha-helical transmembrane domains (TMs) connected by alternating intracellular and

14 Therapeutic product Target receptor Mechanism of Action Indication Imigran Serotonin Agonist Migraine (GlaxoSmithKline) 5HT1B/1D μ-opioid (major) Morphine δ,κ-opioid Agonist Pain (minor) Serevent β -adrenergic Agonist Asthma (GlaxoSmith Kline) 2 Claritin Histamine H1 Antagonist Rhinitis/Allergy (Schering-Plough) Singulair Cysteinyl Leukotriene Antagonist Asthma (Merck and Co.) Receptor 1 Zantac/Tagamet Histamine H2 Antagonist Peptic ulcer (GlaxoSmithKline) Serotonin 5HT / Zyprexa (Eli Lilly) 2 Antagonist Schizophrenia D2 Dopamine Cozaar (Merck) Angiotensin AT2 Antagonist Hypertension Risperdal (Johnson & Serotonin 5HT Antagonist Schizophrenia Johnson) 2

Table 1-1. Selected therapeutic compounds targeting G protein-coupled receptors (Adapted from Wise et al, 2002)

15 extracellular polypeptide loops of varying lengths (Figure 1-1). The crystal structure of rhodopsin, a prototypical Family A GPCR, has also revealed an eighth amphipathic helical domain that runs parallel to the plasma membrane (Palczewski et al., 2000). These helical domains are flanked by an extracellular amino terminus (N-terminus) and an intracellular carboxyl terminus (C-terminus). In addition to this, most GPCRs have two cysteines that form an intramolecular disulfide bridge between the first and second extracellular loops (Baldwin, 1994).

This covalent interaction enhances receptor stability and is involved in various functions including binding and activation, depending on the receptor in question. Despite their conserved

architecture, the GPCR superfamily can be grouped into three distinct families based on

sequence similarity and specific structural characteristics although some novel classification

schemes have grouped GPCRs into five main families according to their phylogenetic

relationship (Fredriksson et al., 2003).

1.2 Traditional Classification of GPCRs

The Family A class of GPCRs is the largest and most diverse subgroup, comprised of 701

receptors that respond to a myriad of ligands including odorants, catecholamines, small peptides,

and glycoprotein hormones (Fredriksson et al., 2003). There are approximately 241 non-odorant

receptors and between 322 (Glusman et al., 2001) and 900 odorant receptors (Venter et al.,

2001), the latter of which accounts for the diverse responsiveness of these receptors to many

olfactory cues. The prototypical member of this family is bovine rhodopsin for which the three-

dimensional crystal structure has been elucidated (Palczewski et al., 2000). There are several

consensus motifs unique to this class of GPCRs that contribute to a common activation

mechanism; these include an E/DRY sequence in TM3 and an NPXXY motif in TM7 (Eilers et

16 L G T G D M A S T LMT R V N I T E A L S T A N G D D V N S E C P R D D S S S Q T E G G F P S S T F C G I L P OH S D C S S R K V S F P F G OH F W S A T N P FF T N R F G Y K L C A A I I H H I D V C N F N S W V I I SH L L OHO S VII W W V S H F E F I V V H Q L A T A G F OHO I V C W A F S F A V A P F N Y A K L H D P N W I I L S S C I P F L S W III V A S L P M VI C H M I IV L V V C I I I I P N II V F S M L L I T V I V S V L G S T I A V M I A S T Y V I L T L A F L Y W L G L L A N D L S T N S T N L R K K C S V L T A A V V I V S L I L C D L I A I V S Y I A V I F VL F F F A K V R A R R N T I F K T D K R H K V E A L AR S FR P R Q T F K Y K S F W Q M T S A K L I I R L M S S P F R Y E G K R Y C R R F C L S I P S A A A E L T P E Q N S C E V P K G N G TT TQ C N K A H V A A R N A I E TV S I NNN G AAM F S S H H E PRG S I S K E CNL V Y L I P HAV GES S D L K K E E T P H Q GQN T IPQ IEK L SVD T D Y D L I VSL A P S L K ELP R A I G A A

Figure 1-1. The amino acid sequence and putative topology of the D1 dopamine receptor. Glycosylation of extracellular asparagine residues within two putative glycosylation motifs in the N-terminus and second extracellular loop is denoted by pink forks. Palmitoylation of two cysteine residues in the proximal half of the C-terminus is shown by grey palmitate moieties. Specific residues in TM 3 and 5 (highlighted in red) are required for binding to dopamine. The intramolecular disulfide bridge between cysteine residues in ECL 1 and 2 (highlighted in purple), the NPXXY motif (highlighted in orange) in TM 7, and the E/DRY motif (highlighted in blue) in ICL 2 are conserved features of most Class A G protein-coupled receptors.

17 al., 2005) (Figure 1-1). In addition to these conserved sequences, there are also distinct structural characteristics that are believed to be involved in stabilization and ligand induced conformational changes of family A GPCRs. For instance, electron microscopy (EM) studies confirmed by X- ray crystallographic methods have determined that the TM helices are bent or “kinked” at points typically positioned near proline residues or Gly-Gly sequences (Palczewski et al., 2000; Unger et al., 1997). These regions may act as hinge points for the relative movement of transmembrane domains within the helical bundle (Stenkamp et al., 2005).

In contrast to the magnitude of the Family A subgroup, the Family B class of GPCRs is relatively small with approximately 25 receptors for large peptides such as glucagon, secretin, calcitonin, and vasoactive intestinal peptide (Pierce et al., 2002). All of these receptors couple to the stimulatory G protein, Gs, and are characterized by a relatively long N-terminus which is critical for targeting the receptor to the plasma membrane. In addition to this, the N-terminus contains a putative hormone binding site that contains a conserved series of cysteines and tryptophans that are required for the interaction of the peptide with the receptor (Harmar, 2001).

Analagous to the Family A class of GPCRs, these receptors also contain a disulfide bridge linking extracellular loop (ECL) 1 and 2, although they lack a E/DRY sequence and NPXXY motif.

The Family C class of GPCRs is distinct from the other two subgroups in that members of this family possess an exceptionally large N-terminal domain that has a bi-lobed configuration. These two lobes are separated by a hinge region that resembles a “Venus flytrap” module that is crucial for ligand recognition and binding (Kunishima et al., 2000). Although these receptors do not contain any of the conserved residues found in Family A or B GPCRs, they do possess the conserved cysteines in ECL 1 and 2. Members of this family include the

18 metabotropic glutamate, γ-aminobutyric acid (GABA), calcium sensing, mammalian pheromone,

and taste receptors.

1.3 G Protein and Effector Signalling

The coupling of heterotrimeric guanine nucleotide-binding proteins (G proteins) to

heptahelical membrane spanning receptors is crucial to the initiation of intracellular signalling pathways as well as establishing the identity of the signalling pathway being activated. G proteins are signal transducers that reside at the plasma membrane where they are pre-associated with GDP and attached to the cytoplasmic face of the receptor. Each G protein is comprised of three subunits, α, β, and γ. Although they exist as heterotrimeric proteins in the basal state, G proteins functionally dissociate into monomeric Gα subunits and dimeric Gβγ subunits to

modulate distinct effectors. There are four main groups of G proteins (Gs, Gi/o, Gq/11, G12/13)

which are largely defined by the effects that they elicit on downstream effectors (Pierce et al.,

2002). The stimulatory G protein (Gs) pathway was the first G protein signalling pathway to be

described and is involved in the activation of adenylyl cyclase and subsequent production of

cAMP. In contrast, the inhibitory G protein (Gi) inhibits adenylyl cyclase, this occurring mainly

through the release of the Gαi subunit. The Gq/11 pathway is known for its ability to activate

phospholipase C-β (PLCβ) and produce the second messengers, inositol triphosphate (IP3) and diacylglycerol (DAG). The effectors that are modulated by G12/13 have not been fully

characterized though G12 has been shown to be coupled to the activation of Rho guanine-

nucleotide exchange factors (Pierce et al., 2002). Signalling through G12/13 has also been been

implicated in lysophosphatidic acid-stimulated cell migration (Bian et al., 2006) and CCK

mediated cytoskeletal remodeling (Le Page et al., 2003).

1.4 Biosynthesis of GPCRs

The current understanding of the folding and maturation process of a GPCR (or any other

alpha helical transmembrane protein) assumes an initial monomeric configuration. The formation

of a membrane spanning receptor begins in the endoplasmic reticulum (ER) and occurs in two

stages. The first stage involves the sequential pairwise insertion of transmembrane alpha-helices

into the ER membrane. Several landmark studies on single transmembrane fragments of the 7-

TM protein, opsin, were among the first to demonstrate that translocation of the nascent TMs

through the membrane requires signal sequences and stop-transfer sequences (Audigier et al.,

1987; Blobel, 1980; Friedlander and Blobel, 1985; Sabatini et al., 1982; Singer et al., 1987). The maturation of polytopic integral membrane proteins such as GPCRs begins with the insertion of two alpha helical peptide segments into the membrane as a hairpin loop. Translocation of each hairpin loop involves coincident insertion of two transmembrane domains with intrinsic alternating signal-anchor and stop-transfer sequences. Asparagine-linked (N-linked) glycosylation can occur co-translationally as the translocation mechanism proceeds. This concept of membrane insertion of integral proteins was first demonstrated in a multi- transmembrane repeat mutant of the single-membrane spanning asialoglycoprotein receptor H

(Wessels and Spiess, 1988).

The second stage of receptor formation involves assembly of the transmembrane segments into a heptahelical bundle yielding the receptor’s tertiary structure. This is driven by a number of factors including helix-helix interactions and structural constraints imposed by the connecting loops (Lemmon and Engelman, 1994). This model was first proposed in studies involving bacteriorhodopsin fragments containing multiple transmembrane domains. These

20 fragments were demonstrated to insert separately into lipid vesicles and subsequent assembly

between complementary domains was found to result in reconstitution of the native receptor

(Marti, 1998; Popot et al., 1987). This principle has been shown with other GPCRs including

rhodopsin (Ridge et al., 1996; Ridge et al., 1995) and the muscarinic receptors (Schoneberg et

al., 1995).

Following the translational assembly of GPCRs in the ER, receptors may undergo post-

translational modifications in the ER-Golgi intermediate compartment (or in an early Golgi

compartment) such as palmitoylation (Bradbury et al., 1997) or further glycosylation.

Palmitoylation involves the covalent modification of cysteines with the fatty acid palmitate,

typically at two conserved cysteines in the carboxyl tail of Class A GPCRs. This post-

translational event can have various effects on the cell surface expression and the signalling

specificity of the receptor (Qanbar and Bouvier, 2003).

1.4.1 Molecular Chaperones and Anterograde Trafficking

The processing of proteins in the ER involves rigorous quality control mechanisms to

ensure that they adopt a conformation that is compatible for proper trafficking through the distal

secretory pathway (Aridor and Balch, 1996). Given the hydrophobic nature of many nascent

proteins such as GPCRs, the cell employs ER-resident chaperone proteins that function within

the framework of a quality control mechanism, to monitor the folding of functional proteins to

ensure that they do not aggregate or misfold. One of the first accessory proteins recognized for

mediating the cell surface expression of GPCRs was Nina A (neither inactivation nor afterpotential A), which was shown to be critical for the maturation of rhodopsin 1 (Rh 1) in

Drosophila melanogaster. Mutant flies that were defective in Nina A had immaturely processed

21 Rh 1 that accumulated in the ER rendering these flies insensitive to visual light response

(Schneuwly et al., 1989; Shieh et al., 1989). Similarly, the human DnaJ protein HSJ1b, has been

shown to act as another accessory protein that modulates the trafficking of mammalian rhodopsin

in retinal cells (Chapple and Cheetham, 2003). Other examples of intracellular chaperones

include ODR-4, which is required for the cell surface transport of the olfactory receptor ODR-10

in Caenorhabditis elegans chemosensory neurons, and the RAMPs (receptor activity modifying

proteins), which facilitate the maturation of many Family B and some Family C GPCRs

(Parameswaran and Spielman, 2006).

1.5 Receptor Activation and Desensitization

The events that occur between docking of the ligand to the receptor binding pocket and

activation of a downstream effector are complex and have yet to be fully elucidated. The agonist

induced activation of Family A GPCRs is believed to involve a rearrangement of TM3 and TM6 as demonstrated by electron paramagnetic resonance spectroscopy studies in rhodopsin (Farrens et al., 1996). This notion is consistent with fluorescence spectroscopic studies of the β2- adrenergic receptor labeled with fluorescent probes demonstrating the relative movement of these same helices in response to agonist activation (Gether et al., 1997). Following acute agonist activation, conformational changes are transmitted to the cytoplasmic domains of the receptor which modulates the interaction of the receptor with a specific G protein. The activated receptor acts as a guanine nucleotide exchange factor thereby promoting the dissociation of GDP and the association of GTP with the α subunit of the G protein (Figure 1-2). In this GTP-bound state, two important events occur: 1) the receptor loses its affinity for the agonist and 2) the α subunit loses its affinity for the βγ subunit. These changes allow the activated G protein to dissociate into

22 2

GDP

GTP / PKA

Figure 1-2. Schematic representation of the classical regulation of GPCR signaling. (1) Heterotrimeric G proteins couple to specific intracellular domains of GPCRs. (2) Agonist binding of the GPCR triggers G protein activation causing the Gα subunit to dissociate from the Gβγ subunits allowing recruitment of G protein receptor kinases (GRKs) and cAMP dependent protein kinase A (PKA). These kinases phosphorylate specific intracellular serine and threonine residues to mediate homologous and heterologous desensitization, respectively. (3) GRK phosphorylation also allows binding of β-arrestin which causes internalization of the GPCR into clathrin coated pits in a dynamin- dependent manner. (4) Once the GPCR is internalized, the receptor is dephosphorylated and sorted to one of two intracellular fates. (5) The receptor may be targeted for lysosomal degradation or resensitized and recycled to the cell surface to begin another round of activation and signaling. (Reprinted with permission from Pierce et al, 2002)

23 its respective Gα and Gβγ subunits to independently activate various downstream effectors

including ion channels, kinases, and GAPs (GTPase-activating proteins). Re-association of the

heterotrimeric complex and subsequent termination of the activation cycle is dependent on 1) the

intrinsic GTPase activity of the Gα subunit and 2) the activity of a family of GAPs known as

regulators of G protein signalling (RGS) proteins. These two mechanisms of de-activation

function in concert to hydrolyse GTP back to GDP and to reduce the levels of active G proteins.

Despite this well-accepted paradigm of G protein activation, there is still much debate over

whether the G protein physically dissociates into the respective Gα and Gβγ subunits or whether

the interface between these subunits simply opens up to allow effector coupling (Digby et al.,

2006; Gales et al., 2006).

Under conditions of continued agonist exposure, the receptor must respond in a rapid

manner to dampen its own persistent activation. This mechanism of signal termination is referred

to as acute desensitization and is functionally regulated by receptor phosphorylation of key

serine and threonine residues (typically in the cytoplasmic loops and carboxyl tail) by various proteins (Figure 1-2). These proteins include second messenger protein kinases (eg: protein kinase A (PKA) or C (PKC)) and any one of seven G protein- coupled receptor kinases (GRKs)

(Pierce et al., 2002). Ultimately, the identity of the phosphorylating kinase defines the type of desensitization being initiated. Heterologous desensitization refers to kinase activation by one

receptor leading to phosphorylation and activation of another receptor. This type of

desensitization occurs through PKA or PKC and can occur independent of agonist occupancy or

G protein activation. In contrast, homologous desensitization refers to phosphorylation and

desensitization of the agonist-occupied (activated) receptor; this is typically mediated by GRKs.

Receptor phosphorylation also promotes the binding of arrestin, an important modulator of

24 GPCR function, which prevents further interactions between the receptor and the G protein thereby contributing to the desensitization response (Pierce et al., 2002).

1.6 Internalization of GPCRs

Most GPCRs exhibit a basal degree of internalization from the cell surface which is often

the result of structurally unstable receptor conformations that mimic the agonist occupied

(active) form of the receptor (constitutive activity) (Wilson et al., 2001) and result in intracellular

endocytosis of the receptor. In addition to this constitutive degree of internalization, GPCRs are

also sequestered into the cell by an agonist dependent mechanism which is essential for

desensitization, resensitization, and continued signal transduction within specific intracellular

compartments (Lefkowitz, 1998). The prototypical and best-characterized agonist-induced

internalization pathway of GPCRs is through clathrin coated pits. This pathway requires the

binding of arrestins, a family of scaffolding proteins that bind to and “arrest” GRK

phosphorylated receptors; this interaction targets the receptor to the clathrin-mediated

internalization machinery (Figure 1-2). With the co-ordinated action of the GTPase, dynamin,

the receptor embedded clathrin coated pit pinches off from the cell membrane and is routed to

the interior of the cell. There are three main types of arrestins: 1) visual arrestin, 2) cone arrestin,

and 3) β-arrestins (β-arrestin 1 and β-arrestin 2) (Ferguson, 2001). Both visual arrestin and cone

arrestin are predominantly localized in the retina whereas β-arrestin is ubiquitously expressed in

various tissue types.

An abundance of evidence from confocal microscopy studies have shown that there are

differences in the translocation kinetics of β-arrestin 1 and β-arrestin 2 to GPCRs, thus dividing

them into two classes (A and B) based on their internalization properties (Oakley et al., 1999;

25 Oakley et al., 2001; Oakley et al., 2000). In the case of Class A receptors, β-arrestin 2 translocates to the receptor more readily than β-arrestin 1. This interaction is transient as β- arrestin rapidly dissociates from the receptor and does not co-internalize with the receptor. The

β2-adrenergic receptor and the D1 dopamine receptor are examples of GPCRs that fall into this

class. In contrast, Class B receptors including the AT1a angiotensin receptor and vasopressin V2

receptor, have no preference for either β-arrestin isoform. Agonist activation of these receptors

recruits β-arrestin into a stable complex that is usually found in endosomes well after

internalization. In addition to these different affinities for β-arrestin, the resensitization and

recycling kinetics (discussed later) are much slower for Class B receptors than for Class A receptors. The carboxyl tail appears to contain the molecular determinants that define these internalization properties as demonstrated by studies using chimeric GPCRs in which this region is exchanged between Class A and B receptors (Oakley et al., 2001).

1.6.1 Post-endocytic Sorting

Following agonist-induced internalization, GPCRs are routed to one of two sorting pathways. The first pathway involves dephosphorylation, resensitization and transport of the receptor back to the plasma membrane; this is referred to as the recycling pathway (Figure 1-2).

The second pathway involves targeting of the receptor to lysosomes where they are proteolytically degraded or down-regulated; this is referred to as the degradative pathway. Not surprisingly, these functionally opposing processes are regulated by discrete cellular mechanisms

that include the recognition of specific sorting motifs. The carboxyl tail, for example, appears to

contain sorting signals that are required for both recycling and lysosomal degradation. This was

demonstrated in studies showing that the protease-activated receptor-1 (PAR1), which is sorted

26 to lysosomes, can recycle when its carboxyl tail is exchanged with that of the recycling substance

P receptor (Trejo and Coughlin, 1999). Indeed, it was later shown that the carboxyl tail of PAR1 interacts with the sorting protein, sorting nexin 1 (SNX1), to target the receptor for lysosomal proteolysis (Wang et al., 2002). This selective interaction of sorting proteins with the cytoplasmic tail of GPCRs is a recurring theme in the regulation of post-endocytic pathways. For example, interaction with the cytosolic proteins, EBP50 and GASP, are crucial for the recycling

of the β2-adrenergic receptor (Cao et al., 1999) and degradation of the δ-opioid receptor

(Whistler et al., 2002), respectively.

In addition to the binding of specific sorting proteins, the lysosomal targeting of GPCRs

can also be dictated by the posttranslational attachment of ubiquitin. Indeed, the lysosomal

degradation of the yeast α-mating factor receptor (Terrell et al., 1998) and the human CXCR4

chemokine receptor (Marchese and Benovic, 2001) both require ubiquitination for lysosomal

sorting. Interestingly, ubiquitination is also requisite for agonist induced internalization of the α-

mating factor receptor suggesting that this modification may have other cellular functions

(Terrell et al., 1998). Protein ubiquitination is carried out in a sequential process that requires the

concerted effort of three enzymes: the E1 ubiquitin-activating enzyme, the E2 ubiquitin-escorting

enzyme, and E3 ubiquitin ligase. There is also emerging evidence that the β-arrestin binding is a

requirement for ubiquitination, as shown for the V2 vasopressin receptor (Martin et al., 2003)

and the β2-adrenergic receptor (Shenoy et al., 2001). It has been suggested that β-arrestins might

serve as adapter proteins to recruit specific E3 ligases to the receptor, although this requires further investigation.

1.7 Palmitoylation of GPCRs

27 Protein-lipid interactions are an important requirement for binding of cytosolic proteins

to the plasma membrane. Lipid modifications can occur in a number of different ways although

N-myristoylation, isoprenylation, and thio(S) acylation are the most common. N-myristoylation

is a co-translational modification that involves the addition of myristate to an N-terminal glycine

residue. Isoprenylation involves the cytoplasmic addition of a farnesyl isoprenoid at a cysteine

residue that is four amino acids from the C-terminus. Thio (S) acylation is the process by which a

16 carbon lipid, palmitate, is attached post-translationally to the thiol group of cysteine residues

and is oftern sub-classified as either N-palmitoylation or S-palmitoylation, depending on how the

palmitate moiety is attached to the protein (Linder and Deschenes, 2007). It is not uncommon for

proteins to be dually lipidated through palmitoylation and myristoylation or isoprenylation as this may serve to increase the hydrophobicity of the protein rendering an enhanced membrane affinity (Escriba et al., 2006). Although many hydrophobic integral membrane proteins like

GPCRs do not require these lipid modifications for membrane targeting, they do have important

implications for maintaining the quaternary structure of GPCRs and also events such as

signalling and trafficking. Whereas, N-myristoylation and isoprenylation are extremely

uncommon modifications found in GPCRs, palmitoylation at specific C-terminal cysteines is

found in approximately 80% of GPCRs (Escriba et al., 2006). This indicates that in addition to

the dynamic lipid environment of the plasma membrane, palmitoylation represents one of the

most conserved interactions of GPCRs with fatty acids.

The constitutive (occurring after biosynthesis) and dynamic (regulation at the cell

surface) nature of GPCR palmitoylation indicates that it has important roles in receptor

trafficking and signaling, respectively. For example, the H2 histamine receptor (Fukushima et

al., 2001), CCR5 chemokine receptor (Percherancier et al., 2001), and V2 vasopressin receptor

28 (Schulein et al., 1996) all require C-terminal palmitoylation for proper cell surface targeting. In

contrast, the G protein coupling properties of other GPCRs have been shown to be modulated upon mutagenesis of the putative palmitoylation sites (Hayashi and Haga, 1997; Kennedy and

Limbird, 1993; Mouillac et al., 1992). The palmitoylation state of some of these receptors are regulated by activation through various extracellular stimuli (Hawtin et al., 2001; Ponimaskin et al., 2001). Palmitoylation by both of these mechanisms are not necessarily mutually exclusive.

Indeed, the δ-opioid receptor (DOR) has been shown to undergo palmitoylation shortly after

export from the endoplasmic reticulum; competitive blockade of palmitate incorporation was

found to inhibit cell surface expression suggesting a role for palmitoylation in DOR maturation

(Petaja-Repo et al., 2006). In the same study, an increase in palmitate turnover was found to

occur in response to agonist activation indicating that a GPCR can be regulated both

constitutively and dynamically by palmitoylation. The palmitoylation of the D1 dopamine receptor is discussed in Chapter 5.

1.8 Dopamine Receptor Subfamily

Dopamine is a key neurotransmitter in the brain that is involved in various functions

mediated by the central nervous system (CNS); these include locomotion, reward, learning, and

endocrine regulation (El-Ghundi et al., 2007). Apart from these effects, dopamine can also

modulate peripheral activities such as catecholamine release, hormone secretion, and the

maintenance of vascular tone in tissues such as the heart and kidney (Missale et al., 1998). The

cellular effects elicited by dopamine are mediated by five dopamine receptor subtypes (D1-D5),

all of which were isolated and cloned in the period beginning in 1988 (Bunzow et al., 1988) with

the cloning of the D2 dopamine receptor, and ending in 1991 with the identification of the D5

29 dopamine receptor (Sunahara et al., 1991). Despite a sudden burst of genomic information on the dopamine receptors, the presence of a “receptor” for dopamine was known since 1972 as biochemical studies during that time demonstrated the activation of adenylyl cyclase by dopamine (Kebabian et al., 1972). This entity was later identified as the D1 dopamine receptor.

All dopamine receptors belong to the Class A family of rhodopsin-like GPCRs and are typically divided into two subfamilies based on the homology between their transmembrane domains, their pharmacology, and their functional regulation of adenylyl cyclase. The D1-like subfamily of dopamine receptors, comprised of the D1 and D5 receptors, are positively associated with adenylyl cyclase through coupling to Gs (Figure 1-3A). In contrast, the D2-like subfamily of dopamine receptors, comprised of the D2, D3, and D4 receptors, all inhibit adenylyl cyclase activity through coupling to Gi. Both families of dopamine receptors can be pharmacologically identified by their ability to interact with subfamily-specific ligands.

However, the utility of such ligands has been curbed by the unavailability of subtype specific ligands thus limiting our understanding of the endogenous function of these receptors (Seeman and Van Tol, 1994).

1.8.1 D1 Dopamine Receptor Structure

The D1 dopamine receptor is the prototypical member of the D1-like subfamily of dopamine receptors. It shares 80% sequence homology with the D5 receptor in the transmembrane domains although the cytosolic loops and carboxyl tail are quite divergent

(Missale et al., 1998). As with many GPCRs, the D1 dopamine receptor possesses post- translational glycosylation sites, one in the amino terminus and the other in the second extracellular loop (Figure 1-1). Although these sites are required for D5 receptor maturation,

30 A B

D1/D5 D2/D3/D4

D1DR

Gs Gi

G G + - s olf Gq Go Adenylyl Cyclase striatum striatum hippocampus ???

nucleus accumbens nucleus accumbens amygdala + - prefrontal cortex olfactory tubercle prefrontal cortex striatum cAMP

Figure 1-3. Dopamine receptor signaling. (A) D1-like dopamine receptors (D1/D5) couple to

Gs to stimulate adenylyl cyclase mediated production of cAMP. D2-like dopamine receptors

(D2,D3,D4) couple to Gi to inhibit adenylyl cyclase mediated production of cAMP. (B) D1 dopamine receptors couple to different G proteins in distinct and overlapping regions of the brain.

31 they do not appear to be involved in translational processing and cell surface maturation of the

D1 receptor (Karpa et al., 1999). In addition to these residues, the D1 receptor carries two cysteines (C437 and C351) in the proximal half of the carboxyl tail which act as multiple sites for palmitoylation. This modification is believed to anchor the cytosolic tail to the plasma membrane but does not appear to be important for G protein coupling (Jin et al., 1997).

The dopamine binding pocket of the D1 receptor is buried within the hydrophobic transmembrane core of the receptor and agonist binding is believed to involve specific residues in TM3 and TM5, similar to many catecholamine receptors (Shin et al., 2002; Strader et al.,

1989a; Strader et al., 1988). Although the D1 receptor does not exhibit much basal activity

(constitutive activity), the D5 receptor displays a high constitutively active phenotype. This has been shown to be due to a single residue difference in ICL 3 of both receptors (Charpentier et al.,

1996). The D/ERY motif at the junction of TM3 and IC3 has also been shown to be involved in the agonist-induced activation of many GPCRs (Rovati et al., 2006), although this has not been thoroughly studied in the D1 dopamine receptor.

1.8.2 Pharmacology of the D1 Dopamine Receptor

Given the high homology between the D1 and D5 receptors in the transmembrane regions, it is not surprising that both receptors have a similar antagonist binding profile. The binding affinity for the antagonist, (+)-butaclamol, is the most differentiating with the D1 receptor having a slightly lower inhibitory constant (Ki) than the D5 receptor. In contrast, the binding affinity of dopamine for the D5 receptor is 10 times greater than for the D1 receptor

(Sunahara et al., 1991). Notably, the D5 receptor exhibits enhanced constitutive activity compared to the D1 receptor. This characteristic is defined by differences in specific residues in

32 the carboxyl terminal region of the third cytoplasmic loop of the D1 and D5 receptors

(Charpentier et al., 1996). Indeed, an increase in agonist binding affinity is a characteristic of

constitutively active receptors (Lefkowitz et al., 1993), indicating that the D5 receptor may be a

constitutively active counterpart to the D1 receptor. Nevertheless, the lack of selectivity toward

specific D1-like receptor subtypes has made it difficult to characterize the physiological nature

of these receptors in native tissue.

1.8.3 Signal Transduction of the D1 Dopamine Receptor

The coupling of Gs to the D1 dopamine receptor is the classical route of adenylyl cyclase

activation and cAMP production via dopaminergic signal transduction pathways. Several studies

have shown that the functional and physical coupling of Gs is dominant in certain regions of the

brain including the prefrontal cortex and the striatum. The D1 dopamine receptor has also been shown to activate adenylyl cyclase 5 (AC5) in regions of the brain where Gs is expressed at very low levels, such as the nucleus accumbens and olfactory tubercle (Herve et al., 2001). Of the nine membrane bound adenylyl cyclase isoforms, AC5 is the most concentrated in the striatum which is consistent with its role in dopaminergic signalling in this brain region (Glatt and

Snyder, 1993). In these brain areas, adenylyl cyclase activation is mediated by D1 receptor coupling to Golf, which is very closely related to Gs. The enzymatic activity of adenylyl cyclase

facilitates the conversion of ATP to cAMP which can act on an array of cAMP-sensitive

effectors. One of these effectors is protein kinase A (PKA), an important modulator of many

cellular effects including ion channel activity and transcriptional regulation. Some of the key proteins that are targets for PKA phosphorylation include DARPP-32 (dopamine and cAMP- regulated phosphoprotein, 32 kDa) and CREB (cAMP response element-binding protein), both

33 of which have been shown to have significant effects on behaviour and synaptic plasticity

(Blendy, 2006; Nairn et al., 2004). There has also been some evidence of MAPK activation by

PKA through D1 dopaminergic stimulation of p38 MAPK and c-jun amino-terminal kinase

(Zhen et al., 1998).

In addition to the classical coupling of D1 receptors to Gs, there is evidence supporting

the notion that these receptors may also couple to other G protein-coupled signal transduction pathways in the brain (Figure 1-3B). This initial observation came from studies showing that striatal D1 receptors could facilitate inositol phosphate production and Ca2+ mobilization in

Xenopus oocytes (Mahan et al., 1990). Subsequent studies in transgenic animals demonstrated

that mice with a null mutation of adenylyl cyclase 5 could still undergo D1 agonist-mediated locomotor activity (Lee et al., 2002b). Collectively, this implicates a D1 receptor-mediated signalling mechanism that functions through a cAMP independent Gq-PLC effector pathway.

There are currently two potential mechanisms of D1 receptor mediated signalling through Gq that have been described. The first mechanism implicates the requirement of a novel D1 receptor that directly associates with Gq while preserving its binding properties to the D1 antagonist,

SCH23390 (Undie et al., 1994; Wang et al., 1995). This Gq-coupled receptor does not react with

a D1 receptor antibody and is not derived from the drd1 gene as confirmed by phosphoinositide hydrolysis studies in D1 receptor knockout mice (Friedman et al., 1997). In addition to this, it is highly responsive to the selective agonist, SKF83959, which does not activate adenylyl cyclase through classical Gs-coupled D1 receptors (Jin et al., 2003). The second mechanism of Gq coupling has been described by our laboratory and involves the requirement of an intact D1-D2 dopamine receptor hetero-oligomer for coupling to Gq (Lee et al., 2004; Rashid et al., 2007). In

34 this situation, when D1 and D2 dopamine receptors are co-expressed, co-activation of the two

receptors triggers Ca2+ mobilization in a pertussis toxin insensitive manner.

1.9 Dopamine Neurotransmission In The Brain

Dopamine is a catecholaminergic neurotransmitter found predominantly in the brain but

also in peripheral sites such as the kidney and adrenal cortex. In the central nervous system,

dopamine is synthesized in the presynaptic nerve terminal where the precursor amino acid, tyrosine, is first converted to L-dihydroxyphenylalanine (L-Dopa) by the enzymatic activity of

tyrosine hydroxylase. This chemical intermediate, L-Dopa, is then converted to dopamine by

Dopa decarboxylase. Dopamine is subsequently packaged into storage vesicles by vesicular monoamine transporter 2 where it awaits an action potential or pharmacological intervention for release into the synaptic cleft (Iversen, 2006). In certain neurons, dopamine can be further converted to norepinephrine by the enzyme dopamine-β-hydroxylase. At sufficient concentrations, synaptic dopamine binds to and activates pre- and post-synaptic dopamine receptors. Post-synaptic dopamine receptors propagate signals to downstream effectors to elicit specific cellular effects. Activation of pre-synaptic dopamine receptors (autoreceptors) triggers a negative feedback loop where vesicular dopamine release is attenuated directly or by inhibition of tyrosine hydroxylase activity. The termination of dopamine signalling largely occurs by re- uptake into the presynaptic terminal by the dopamine transporter where it is metabolically degraded by the enzyme, monoamine oxidase into 3,4 – dihydroxyphenylacetic acid (DOPAC) and 3,4 – dihydroxyphenyl-glycol (DHPG) (Youdim et al., 2006). Alternatively, extracellular dopamine that remains in the synapse can also be metabolized by catechol-O-methyl transferase

into the metabolites, homovanillic acid (HVA) and 3-methoxy-4-hydroxyphenylglycol (MHPG).

35 The dopaminergic circuits in the brain give rise to three main pathways which are known

as the mesocorticolimbic, nigrostriatal, and tuberoinfundibular pathways. The mesocorticolimbic

pathway consists of two systems. The first originates in the A10 nucleus of the ventral tegmental

area (VTA) and sends projections to ventral regions of the striatum including the nucleus

accumbens and olfactory tubercle; this is referred to as the mesolimbic dopamine system.

Dopaminergic neurons in the second system begin in the medial VTA and project to cortical

areas including the prefrontal, perirhinal, and anterior cingulate cortices; this is referred to as the

mesocortical dopamine system (Bressan and Crippa, 2005). These pathways are collectively known to be involved in motivational and reward behaviour and are therefore sometimes referred

to as the “brain reward circuit”. The nigrostriatal pathway originates in the zona compacta of the

substantia nigra and projects to the caudate putamen. This system is most strongly implicated in

motor function. The tuberoinfundibular dopaminergic neurons have short axons that originate in

the A12 region of the hypothalamus and terminate in the median eminence. This circuit is part of

a feedback loop that is involved in regulating prolactin secretion (Tuomisto and Mannisto, 1985).

1.10 Anatomical Distribution of the D1 Dopamine Receptor

The D1 dopamine receptor is the most abundantly expressed dopamine receptor in the rat

brain. D1 receptor mRNA and protein have been found to localize in various regions including

the striatum, the nucleus accumbens, and the olfactory tubercle (Weiner et al., 1991). D1

receptors have also been shown to be expressed in the entopeduncular nucleus and in the pars

reticulata of the substantia nigra, although no receptor mRNA has been shown in these regions.

In contrast, the D5 receptor is poorly expressed in the rat brain with an mRNA distribution that is

limited to distinct layers of the hippocampus, the mammillary nuclei, and the anterior pretectal

36 nuclei (Tiberi et al., 1991); these regions are not particularly rich in D1 dopamine receptor

mRNA. Immunolabelling techniques combined with light and electron microscopy have allowed

for the cellular and subcellular detection of D1 receptors in various regions of the primate brain

(Bergson et al., 1995; Smiley et al., 1994). These studies have demonstrated the pre- , and more

predominantly, postsynaptic distribution of D1 dopamine receptors in the pyramidal neurons of

the prefrontal cortex and the hippocampus (Bergson et al., 1995; Smiley et al., 1994). The subcellular localization of D1 receptors within individual pyramidal neurons suggests the

targeting of these proteins to dendritic spines and shafts.

In addition to the well known expression pattern of D1 dopamine receptors in the brain,

there are other regions in the periphery that abundantly express these receptors. The proximal

tubules of the rat and opossum kidney have been shown to express the D1 dopamine receptor

(Nash et al., 1993; Yamaguchi et al., 1993). In addition to this, the vascular system has been shown to express D1 receptors in the medial layer of blood vessels (Missale et al., 1988). In general, these lesser known regions of D1 dopamine receptor expression have poorly understood effects.

1.10.1 Physiological Role of D1 Dopamine Receptors

The selective inhibition of specific dopamine receptor subtypes is limited by the lack of specificity of the currently available dopamine receptor antagonists. As a consequence, gene deletion and overexpression studies in animals have proven to be a useful tool in determining the role of dopamine receptors in the brain. Despite the utility of gene knockout animal models, there are several caveats that must be considered when using such models. For instance, any alterations in behaviour must rule out the compensatory upregulation of other proteins as a result

37 of gene deletion. Similarly, in the case of loss of function mutations in which only a portion of the gene is deleted, further studies must ensure that any behavioural changes are not due to atypical effects that do not depend on the functionality of the fully intact receptor (Sealfon and

Olanow, 2000). In transgenic animals where the receptor of interest is overexpressed, transgene incorporation takes place at a random site within the genome. Therefore, for any phenotype rendered, it is important to show that this is not due to inadvertent inactivation of another gene by random incorporation of the transgene.

Both genetic null mutations and genetic disrupted mutations of the D1 dopamine receptor have been generated in mice. Both mutants exhibit diminished locomotor activity in response to psychostimulants such as cocaine or amphetamine (Crawford et al., 2000; Drago et al., 1996; Xu et al., 1994). This implicates a role for the D1 receptor in psychostimulant-induced locomotion.

We have previously shown that D1DR gene-deleted mice exhibit reduced operant responding on sucrose reinforced levers suggesting a deficit in reinforcement learning in these mice (El-Ghundi et al., 2003). Other studies have demonstrated that D1DR blockade with receptor specific antagonists abolishes the discriminative-stimulus effects of cocaine, manifested as an inability to respond to the appropriately associated lever for cocaine and saline (Elliot et al., 2003).

Similarly, pharmacological blockade of D1DR blocks the conditioned place preference induced by many drugs of abuse implicating a role for D1DR in discriminating reward-related behaviour induced by these compounds (Baker et al., 1998; Beninger et al., 1989; Liao et al., 1998; Pruitt et al., 1995). However, some of these findings could not be corroborated in mice genetically lacking D1 receptors as associative learning behaviour was preserved in these animals

(Karasinska et al., 2005; Miner et al., 1995). In addition to the role of D1DR in reward and reinforcement, studies from our lab have also implicated a role for the D1 receptor in spatial

38 learning (El-Ghundi et al., 1999), alcohol-seeking behaviour (El-Ghundi et al., 1998) as well as

conditioned fear responses (El-Ghundi et al., 2001). In contrast, in transgenic mice

overexpressing the D1 receptor, agonist stimulation suppressed locomotion (Dracheva et al.,

1999). This is contrary to the data from the gene knockout studies and implicates both a

stimulatory and inhibitory role for the D1 receptor in locomotion, possibly in different regions of

the brain.

In the cardiovascular system, a combination of radioligand and autoradiographic studies

have shown the expression of postjunctional D1 receptors in the walls of systemic arteries

(Amenta et al., 1993). Stimulation of these receptors leads to arterial vasodilatation. In the

kidney, D1 receptors are localized in the proximal tubules. Stimulation of these receptors

decreases renal sodium reabsorption and stimulates rennin release as a result of Na+/K+-ATPase inhibition (Satoh et al., 1993; Yamaguchi et al., 1997).

1.11 Introduction to G Protein-Coupled Receptor Oligomerization

It has been traditionally thought that the mechanisms of GPCR ligand binding and signal transduction should be modeled after a single receptor coupled to its ligand and G protein in a

1:1:1 stoichiometry. Despite the fact that many other classes of cell surface receptors such as the protein-tyrosine kinase receptors and cytokine receptors function through either a ligand dependent or constitutive dimerization mechanism (Heldin, 1995), the monomeric model for

GPCRs prevailed for quite some time. However, increasing evidence from a diverse number of methodological approaches has successively shown that oligomerization appears to also be a fundamental characteristic of GPCR biology, as will be discussed later (Bouvier, 2001). Many of these approaches confirm earlier observations from radiation inactivation, chemical crosslinking,

39 and radioligand binding studies that GPCRs might exist as multimeric entities (Bouvier, 2001).

The functional implications of receptor oligomerization are tantalizing as this phenomenon may

implicate cross- talk between directly associated receptors resulting in alterations in effector

signalling. Homo-oligomerization (oligomerization between the same receptor subtype) may be a

cell surface mechanism of intracellular signal amplification (George et al., 2002). In the case of hetero-oligomers, the generation of novel binding pockets between different receptor subtypes presents exciting opportunities for identifying previously unrecognized drug targets (George et al., 2002). The following section will describe the current evidence for GPCR oligomerization.

1.12 Evidence for GPCR Oligomerization

1.12.1 Receptor Complementation

One of the first studies to bring the idea of receptor oligomerization into the spotlight originated from work with α2C-adrenergic/M3 muscarinic receptor chimeras by Maggio and

colleagues (Maggio et al., 1993). In this report, they described the generation of receptor

chimeras in which the C-terminal halves were exchanged between the α2C-adrenergic receptor

and the M3 muscarinic receptor yielding α2C/M3 and M3/α2C fusion proteins. Individual

expression of either of these chimeric constructs did not result in any detectable binding to either

muscarinic or adrenergic ligands. However, co-expression of the chimeras resulted in binding to

both muscarinic and adrenergic radioligands with binding properties comparable to the wild-type

receptors. Furthermore, functional activity was restored upon co-expression as

phosphatidylinositol hydrolysis was detected following stimulation by a muscarinic agonist. This

40 data infers a mechanism whereby receptors can not only simply interact but also engage in a reciprocal exchange of receptor domains (Figure 1-4).

This notion of functional complementation between receptor mutants, described as the domain swapping hypothesis of GPCR oligomerization (Gouldson et al., 2000; Gouldson et al.,

1998) has also been reported in homo-oligomerization studies using mutants of the angiotensin II type 1 receptor (Monnot et al., 1996) and the H1 histamine receptor (Bakker et al., 2004).

Although conceptually, domain swapping is a plausible explanation for describing functional rescue of such receptor mutants, there is evidence that calls into question the validity of this theory for other receptors (Schulz et al., 2000). In a study using vasopressin receptor mutants, the authors determined that reconstitution of a functional wild-type V2 vasopressin receptor could not be achieved by co-expressing an N-terminal receptor fragment and a non-functional full- length vasopressin receptor, as would be predicted by the domain swapping model. Despite this, the two mutants could still interact which is consistent with a model in which GPCRs simply form contact dimers. The contact dimer hypothesis, proposes that two binding pockets are formed from regions donated within each monomer; it requires that receptors interact laterally to form dimers.

1.12.2 Co-immunoprecipitation

The use of differentially tagged proteins has been a popular biochemical tool employed to demonstrate protein-protein interactions. It has also commonly been used to provide evidence for the existence of homo- and hetero-oligomers of GPCRs, including those that are involved in brain reward pathways (George et al., 2000; Gomes et al., 2000; Jones et al., 1998; Jordan and

Devi, 1999; Kaupmann et al., 1998; Lee et al., 2004; Scarselli et al., 2001; White et al., 1998;

41 A

4 5 4 3 3 5 6 2 2 6 7 1 1 7

B 2 1 3 7 4 5 6 6 5 7 3 4

1 2

Figure 1-4. (A) Schematic of a contact dimer involving TM 1 and 2. (B) Schematic of a domain swapped dimer in which TM 6 and 7 are mutually exchanged between the receptor monomers. The purple and green arrows indicate each of the chimeric monomers within the dimer. (Reprinted with permission from Gouldson et al, 1998)

42 Xie et al., 1999). Using this method, the two receptors of interest are usually modified by the

addition of a distinct epitope or “tag”, typically at the N-terminus of each receptor. When these

receptors are co-expressed in cells, an antibody directed to one of the epitopes is used to immunoprecipitate the receptor. The receptor is then released from the antibody-antigen complex

and separated on sodium-dodecyl-sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) and

then immunoblotted for the epitope on the second receptor. A direct interaction between the

receptors is likely if both receptors can be co-precipitated with each other. This conclusion, however, must be interpreted with caution as adapter proteins could be involved in linking the two receptors together. In addition, artefactual aggregation between receptors is a potential problem that occurs when receptors are solubilized by detergents. When removed out of the lipid environment of the plasma membrane, the hydrophobic nature of GPCRs endows them with a

propensity to nonspecifically aggregate possibly leading to erroneous conclusions. This can be

carefully controlled for by mixing cells that independently express each of the receptors.

In heterologous cell lines, non-specific interactions may also occur in the cell as a result

of overexpression of the two receptors of interest (Ramsay et al., 2002; Salim et al., 2002). As a

result, it is more reliable, albeit more difficult, to co-immunoprecipitate receptors from native

tissues where they are expressed endogenously at lower physiological levels.

Ultimately, the conclusions that can be drawn from co-immunoprecipitation experiments

are limited. One can be confident that two receptors that co-immunoprecipitate with each other

exist within a complex; however, to demonstrate a direct interaction would require further analysis using alternative techniques.

1.12.3 Novel Pharmacology

Some of the earlier radioligand binding studies performed on GPCRs demonstrated that

binding of a ligand to one site could increase or decrease the binding affinity of ligands to other

sites (Limbird et al., 1975; Mattera et al., 1985; Wreggett and Wells, 1995) . This phenomenon,

respectively termed positive or negative co-operativity, has often been used to suggest the

presence of GPCR oligomers. This work has recently been expanded to show that oligomers may

not only modulate the pharmacological properties of individual receptors, but also alter their G

protein-coupling specificity (Levac et al., 2002). Some of the first evidence to show distinct pharmacological changes in hetero-oligomers were from studies done with δ and κ opioid receptors (Jordan and Devi, 1999). In these studies, co-expression of these two opioid receptor subtypes led to a reduction in binding affinity to individual δ and κ receptor selective ligands; however simultaneous treatment with δ and κ-selective ligands resulted in a synergistic increase in high affinity binding to the δ/κ hetero-oligomer. This observation was associated with enhanced effector activation as measured by cAMP inhibition and MAPK phosphorylation

(Jordan and Devi, 1999), suggesting functional modulation by the hetero-oligomer. In a separate study, hetero-oligomers of the μ and δ receptor subtypes were shown to exhibit ligand affinity and potency properties distinct from each of the individually expressed receptor subtypes

(George et al., 2000). Furthermore, the hetero-oligomer had novel G protein coupling features indicative of the generation of a novel signalling unit; this will be discussed in greater detail later. The inference from these pharmacological data that hetero-oligomers exhibit novel binding properties unique from the receptor homo-oligomers has also been demonstrated for other

44 GPCRs (Armstrong and Strange, 2001; Galvez et al., 2001; Gines et al., 2000; Gomes et al.,

2000; Maggio et al., 1999; Rocheville et al., 2000a)

1.12.4 Receptor Trafficking

GPCR oligomerization has often been shown to be a prerequisite for cell surface

trafficking of many receptors as illustrated by numerous examples (Balasubramanian et al., 2004;

Hague et al., 2004a; Hague et al., 2004b; Jones et al., 1998; Kaupmann et al., 1998; Nelson et al.,

2001; Uberti et al., 2005; Uberti et al., 2003; White et al., 1998; Zhao et al., 2003). When

expressed alone in heterologous cell lines, these receptors are intracellularly retained; however, when co-expressed with another receptor subtype, cell surface expression is restored. This effect can be mediated by a number of different mechanisms including the masking of endoplasmic reticulum (ER) retention motifs on the intracellularly localized receptor or the dissociation of ER retention proteins (Hurtley et al., 1989; Kapoor et al., 2003; Zhang et al., 1997). One of the most widely accepted examples demonstrating this effect is the GABAB receptor in which there is an

obligate requirement for hetero-oligomerization between the GABABR1 and GABABR2

subtypes for functional plasma membrane localization. As will be further discussed later, a

GABABR2 interaction with GABABR1 is required to mask an ER retention signal on GABABR1

to allow cell surface trafficking (Jones et al., 1998; Kaupmann et al., 1998; White et al., 1998). In

contrast to this, specific receptor variants have been shown to exert trafficking inhibition of the

cognate wild-type receptor (Benkirane et al., 1997; Elmhurst et al., 2000; Grosse et al., 1997;

Karpa et al., 2000; Le Gouill et al., 1999; Lee et al., 2000a; Zhu and Wess, 1998). These mutants

can be naturally occurring splice variants or genetically modified receptors that are typically

non-functional. Both of these opposing phenomena support the concept that oligomerization

45 occurs during transport through the distal secretory pathway and possibly as early as in the ER where biosynthesis occurs. It also suggests that there are significant functional consequences that occur through these intermolecular interactions. For example, our lab has shown that receptor variants of the D2 dopamine receptor generated by mutagenesis of specific transmembrane residues, inhibit cell surface expression of the wild-type D2 receptor when co-expressed (Lee et al., 2000a). This suggests that certain oligomeric configurations are not compatible for transport to the plasma membrane.

The naturally occurring variant of the D3 dopamine receptor, D3nf, is an example of how oligomerization may have a physiological role in modulating receptor trafficking (Liu et al.,

1994). In certain cell types, this truncated variant is retained intracellularly and exhibits very little cell surface expression (Karpa et al., 2000). When co-expressed with the wild-type D3 receptor, it sequestered the wild-type receptor in the cell thus attenuating its function (Elmhurst et al., 2000; Karpa et al., 2000). This has been suggested to occur by a direct receptor interaction and provides further evidence for constitutive GPCR homo-oligomerization.

Our laboratory has also recently shown that genetically modified D1 dopamine receptors could “drag” wild-type receptors into specific cellular compartments (O'Dowd et al., 2005).

These modified receptors incorporated a nuclear localization sequence that was able to send differentially tagged wild-type receptors to the nucleus, necessitating the requirement for oligomerization.

1.12.5 Resonance Energy Transfer

One of the major shortcomings of studying GPCR oligomerization through the aforementioned biochemical and pharmacological approaches is the inability to detect oligomers

46 physiologically in live cells. However, an emerging number of investigators are turning toward biophysical methods to demonstrate that specific GPCRs are in close proximity to one another

(Angers et al., 2000; Canals et al., 2003; Cheng and Miller, 2001; Jensen et al., 2002; Kroeger et al., 2001; McVey et al., 2001; Overton and Blumer, 2002). Two of these approaches, fluorescence resonance energy transfer (FRET) and bioluminescence resonance energy transfer

(BRET), are based on the non-radiative transfer of energy between closely associated proteins

(Bouvier, 2001; Eidne et al., 2002). In FRET, the protein that transmits the signal and the protein that receives the signal are attached to donor and acceptor fluorophores, respectively. In BRET, the donor protein is fused to an enzyme called luciferase, derived from the sea pansy Renilla reniformis; the acceptor protein is attached to a variant of the green fluorescent protein (GFP) from the jellyfish Aequorea Victoria (Figure 1-5). Luciferase catalytically degrades the substrate, coelenterazine, which gives off a bioluminescent signal that is transferred to the GFP acceptor

(Devi, 2001). Energy transfer between donor and acceptor in FRET and BRET is critically dependent on the proximity between the proteins involved and the orientation of their dipoles.

Both methods require that proteins be within 100 Å (10 nm) of each other, which is a distance that would implicate a direct protein-protein interaction. In addition, both approaches rely on sufficient overlap between the emission and excitation spectra of the donor and acceptor fluorophores, respectively. Although these methods allow non-invasive detection of receptor oligomerization in live cells, their utility has two limitations. First, an increase in energy transfer efficiency, as may occur upon ligand induction, can either be interpreted as enhanced oligomerization or simply a conformational change within the oligomer that may only bring the fluorophores in close proximity to each other. The current challenge is differentiating between these two possibilities. Second, it is difficult to identify the precise cellular location in which

47 Figure 1-5. Bioluminescence Resonance Energy Transfer. The overlapping emission spectra of Renilla luciferase (R.luc) and excitation spectra of GFP make them a suitable BRET pair to test for protein interactions. Addition of the R.luc substrate, coelenterazine, results in emission at 470 nm. Energy transfer occurs if the two fusion proteins are within 100 Å of each other (implicating a direct interaction) resulting in GFP emission at 520 nm. (Reprinted with permission from Bouvier, 2001)

48 oligomers are detected, particularly using BRET. Fortunately, some progress has been made in this regard using a variation of FRET, time-resolved FRET, that substitutes the use of fluorescent fusion proteins with antibodies conjugated to fluorescent molecules (Gazi et al., 2003; McVey et

al., 2001). Because antibodies are unable to permeate the plasma membrane, any energy transfer

signal that is detected is predicted to originate only from receptors at the cell surface. This

advancement can be used to detect the presence of GPCR oligomers or any ligand induced

changes in the conformation of oligomers, specifically at the plasma membrane.

In addition to measuring the proximity of receptors to each other, BRET has also been

used to determine the affinity between receptors within a homo-oligomer or hetero-oligomer.

Homo-oligomers of either the β2-adrenergic receptor or κ opioid receptor formed with high

affinity as increased expression of receptor acceptor levels caused a linear increase in energy

transfer (Ramsay et al., 2002). However, the β2-adrenergic receptor and κ opioid receptor were

found to form hetero-oligomers less efficiently as overexpression of both receptors to non-

physiological levels was required to achieve a similar energy transfer efficiency as the homo-

oligomers. This indicates that there is specificity and preferred interactions between receptors and that oligomerization is not a result of random collisions or non-specific interactions (as a result of overexpression) in the cell.

1.12.6 Atomic Force Microscopy

Perhaps the most direct and convincing evidence to date for the existence of GPCR oligomers comes from their direct visualization by atomic force microscopy (AFM) studies by

Liang et al (Liang et al., 2003). AFM is a technique employed to study surface morphology and the properties of molecules, often at atomic resolution. The microscope used is equipped with a

49 cantilever attached to a tiny tip or probe that scans the surface of a specimen. In conjunction with

a laser beam that reflects off the cantilever to a detector, the cantilever is raster scanned to

produce an image (by a linked computer via the detector) that represents the surface topography

of the specimen. In this study, the authors used AFM to reveal that rhodopsin molecules are

organized as rows of dimers from native rod outer segment disk membranes. A model of

rhodopsin oligomers was produced that accounted for constraints imposed by the AFM data as

well as the crystallographic data of rhodopsin. This model revealed potential contact points

within and between dimers and also predicted the strength of these interactions based on these empirical data. It will be interesting to determine whether this model applies to other receptors as well, including those involved in reward pathways; this still needs to be proven by similar studies in brain.

1.12.7 X-ray Crystallography

The most thoroughly characterized GPCR structure to date is rhodopsin, due in large part to the crystallization of rhodopsin in its ground state (Palczewski et al., 2000). Subsequent studies have demonstrated higher resolution structures of rhodopsin (Okada et al., 2002; Okada et al., 2004; Teller et al., 2001) as well as the elucidation of metarhodopsin I, a conformational intermediate in which all-trans retinal is still bound through opsin, through electron- crystallography (Ruprecht et al., 2004). Recently, two new crystal forms of rhodopsin in the photo-activated state were determined (Salom et al., 2006a; Salom et al., 2006b). Consistent with the crystal structure of metarhodopsin I, these two crystal forms both suggested that rhodopsin forms dimers with contacts involving helices 1 and 8. However, the orientation of the subunits

50 relative to each other as well the interacting residues on the membrane surface were found to be

significantly different from the metarhodopsin model.

1.13 Structural Features Of GPCR Oligomers

Although the crystallization of bovine rhodopsin has provided detailed insight into the

intramolecular intricacies of GPCR monomers (Palczewski et al., 2000), the configuration of

GPCR oligomers and the intermolecular contacts that are involved remain to be fully elucidated.

The above AFM study represents the first empirical evidence that class A GPCRs are organized as dimeric rows, it remains to be determined whether GPCRs in other families have a similar organization. There is increasing evidence that receptors from different families utilize different mechanisms of oligomerization due to inherent structural differences. However, there also appears to be specific fundamental mechanisms that are shared between families. Some of these are described below.

1.13.1 Disulphide bonds

The presence of disulphide covalent linkages between cysteine groups is important to the intramolecular stability of a receptor. Most rhodopsin-like GPCRs contain a conserved disulfide bond that links the first and second extracellular loops (Strader et al., 1994). However, it has been suggested that these cysteines may also participate in oligomer formation. Site-directed mutagenesis of these specific cysteines in the M3 muscarinic-acetylcholine receptor has been shown to yield only monomeric receptors (Zeng and Wess, 1999). In addition, oligomers of the prostacyclin (Giguere et al., 2004), D1 dopamine (Lee et al., 2000b), 5-hydroxytryptamine 1B

and 1D serotonin receptors (Lee et al., 2000b) dissociate into monomers when exposed to

51 reducing agents such as beta-mercaptoethanol (2-ME) or dithiothreitol (DTT). It remains unclear

whether this bridge participates in oligomer formation through a direct or indirect interaction. It is possible that this intramolecular disulfide bridge is not directly involved in maintaining interactions between receptors; instead, disruption may cause structural instability within the receptor that may hinder its ability to form oligomers.

A direct role of disulfide bonds in dimerization has been established in the metabotropic glutamate receptor (mGluR1); this was revealed by the crystal structure of its extracellular- binding domain (EBD) (Kunishima et al., 2000). This receptor was found to exist as a dimer in

its resting and ligand occupied state. Furthermore, a single disulphide bridge was shown to connect Cys140 of each monomer within the dimeric complex. This residue cannot act as a structural scaffold due to its position within a disordered segment of the EBD suggesting that other domains are critical for dimer assembly. Nevertheless, the role of disulphide bonds in the homo-dimerization of family 3 GPCRs within the EBD appears to be well-conserved as mGluR5

(Romano et al., 1996) and the calcium-sensing receptor (Bai et al., 1999) also form disulfide linked dimers. These can be dissociated into their respective monomers by treatment with reducing agents.

1.13.2 Transmembrane interactions

The oligomerization of many single-membrane spanning proteins has been documented to involve their hydrophobic transmembrane domains with both structural and functional consequences (Cosson et al., 1991; Furthmayr and Marchesi, 1976; Manolios et al., 1990). A notable example is the ligand-dependent dimerization of the epidermal growth factor receptor, which depends on these alpha-helical interactions for signal transduction (Ullrich and

52 Schlessinger, 1990). It has also been shown that the single membrane spanning human

glycophorin A (GpA) forms SDS-resistant dimers that appears to depend on a specific amino

acid motif (Lemmon et al., 1992). The evidence implicating the GpA motif in dimerization has been shown using the alpha-factor receptor from the yeast Saccharomyces cerevisiae (Overton and Blumer, 2002; Overton et al., 2003). More recently, TM4 was found to form part of the dimerization interface of D2 dopamine receptors (Guo et al., 2003; Lee et al., 2003b). However, disruption of this domain in the full-length D2 receptor did not affect the dimerization interface as extensively as in a similarly disrupted truncated version of the receptor (Lee et al., 2003b).

Thus, it is likely that oligomerization of the D2 receptor involves multiple sites of interaction.

Packing models derived from the atomic force microscopy studies (described earlier) on native

rhodopsin predicted that the intradimeric interaction sites involved TM4 and TM5 and the

interdimeric interaction sites involved TM1 and TM2 and the third cytoplasmic loop (Carrillo et

al., 2004; Liang et al., 2003). This supports the model that multiple points of contact, particularly

in the transmembrane regions, are involved in holding receptors in an oligomeric complex.

1.13.3 Intracellular and extracellular-domain interactions

Although the prevalent mode of oligomerization in some receptors belonging to Family C

GPCRs implicates disulphide bonding at the amino-terminus (Tsuji et al., 2000), it is unknown

whether this region is equally important in other GPCR families. The amino terminus has been

implicated in oligomerization (as a result of agonist induction) of the B2 bradykinin receptor, a

family 1 GPCR, despite the absence of a large extracellular binding domain (AbdAlla et al.,

1999).

53 The carboxyl terminus has been reported as a site of intermolecular interaction in the

GABAB receptor, a member of the family 3 GPCRs (Jones et al., 1998; Kaupmann et al., 1998;

White et al., 1998). The mechanisms underlying GABAB receptor oligomerization are

particularly interesting, as functional activity requires the interaction between two GABAB receptor subtypes, GABABR1 and GABABR2. Prior to the cloning of the GABABR2 subtype it

was known that GABABR1 failed to exhibit any cell surface expression in heterologous cell lines

(Couve et al., 1998). The expression of the GABABR2 recombinant protein, however, resulted in

normal cell surface targeting with ligand binding deficits, rendering it non-functional. Co- expression of both receptor subtypes, however, restored functional activity that was comparable to the natively expressed GABAB receptor. Subsequent studies have shown that specific regions

of each receptor subtype contribute to the overall function of the hetero-oligomer and that hetero-

oligomerization is, in part, facilitated by a coiled-coil interaction involving the carboxyl tails of

GABABR2 and GABABR1 (Calver et al., 2001; Carrillo et al., 2003; Galvez et al., 2001;

Margeta-Mitrovic et al., 2000; Robbins et al., 2001). The coiled-coil interaction allows

GABABR2 to mask an ER retention motif on GABABR1, thereby allowing the complex to traffic

to the cell surface (Margeta-Mitrovic et al., 2000)

1.14 Mechanism Of Agonist-induced Activation Of GPCR Oligomers

The hetero-dimeric nature of the GABAB receptor is a prerequisite for facilitating

GABA-mediated activation. The initial observation that the GABABR1 subunit was required for

ligand binding and that the GABABR2 subunit was required for cell surface expression suggested

that each of these protomers had distinct non-redundant roles in the dimer. Indeed, it was shown

that the GABABR2 subunit is sufficient for G-protein coupling (Duthey et al., 2002; Galvez et

54 al., 2001; Margeta-Mitrovic et al., 2001; Robbins et al., 2001) and that activation of the G protein

occurs in trans through binding of the GABABR1 subunit (Galvez et al., 2001). The

effectiveness of each of these functional roles, however, is enhanced by its interaction with the

other subunit. Hence, GABABR1 improves coupling efficiency of GABABR2 and GABABR2 improves the binding affinity of GABABR1 suggesting allosteric modulation between subunits

within the dimer (Galvez et al., 2001). In agreement with these findings, it was also shown that

strategically positioned disulfide bonds within the EBD of GABABR1 could generate a fully active GABAB receptor with a response magnitude comparable to the agonist-occupied receptor

(Kniazeff et al., 2004b). Similarly positioned disulfide bonds in GABABR2 did not have an

effect indicating that the closed “active” conformation of the EBD of GABABR1 is sufficient to induce agonist independent activation of the GABAB receptor. Thus, within a GABAB receptor, binding of the agonist to one protomer triggers closing of the large EBD; the resulting conformational change may alter the orientation of the G protein-coupled protomer resulting in activation and G protein dissociation (Pin et al., 2004).

In contrast to this, ligand binding to one protomer of the homo-dimeric metabotropic glutamate receptor results in only partial activation of the receptor whereas binding to both protomers results in full activation (Kniazeff et al., 2004a). This coincides with the requirement of both EBDs to be in the closed “active” state for complete activity. In addition, whereas the

GABAB receptor depends primarily on the GABABR2 protomer for activation, the mGluR5

receptor equally depends on both protomers for activation. This may imply that G protein-

coupling to both protomers in the mGluR5 receptor, in contrast to binding to the single protomer

in the GABAB receptor, is necessary for activation.

55 The scenario in which a single dimer couples to a single G protein would agree with the predicted stoichiometry of a rhodopsin dimer interacting with Gt (transducin). In the resting state

of rhodopsin, both Gtα and Gtγ interact with helix 8 of rhodopsin (Ernst et al., 2000; Marin et al.,

2000). However the crystal structure of transducin shows that these G protein subunits are a

relatively distant 42 Å apart from each other (Hamm, 2001). These contact requirements would

be satisfied by a dimeric model of rhodopsin which has been confirmed by rhodopsin AFM

studies, as described previously. Furthermore, this 1:1 stoichiometry of the receptor dimer to G

protein heterotrimer has been demonstrated in mass spectrometry studies of the leukotriene B4

receptor, a class A rhodopsin-like GPCR (Baneres and Parello, 2003). Activation of a single

protomer within this functional pentameric complex has been shown to result in allosteric

modulation of the partner protomer, similar to that observed with the Family C GPCRs (Mesnier

and Baneres, 2004). It will be of future interest to determine whether this receptor-G protein

arrangement and activation mechanism is conserved across the entire GPCR family.

1.15 GPCR Oligomers In Native Brain Tissue

One of the essential steps that must be taken when studying the oligomerization of

GPCRs is establishing their physiological relevance in native tissues. It is important to realize

that receptors that interact in heterologous cell lines may not do so in vivo because they normally

do not co-localize in specific regions of the brain and thus have different expression profiles. In

addition, there is mounting evidence suggesting that a certain level of specificity exists within

hetero-oligomers and that receptors, especially those from different families, do not interact in a

promiscuous manner (Salim et al., 2002). For example, we have reported that D1 and D2

dopamine receptors can form a mixed population of oligomers (Lee et al., 2004). The precise

56 ratio of homo- to hetero-oligomers may be governed by regulatory mechanisms in the neuron that may not exist in heterologous cell lines they normally would not be found in.

The current methods of oligomer detection in brain are typically performed using immunological techniques such as western blotting and immunoprecipitation. The detection of

A1 adenosine receptors (A1R) in pig brain cortex was among the first of the reports showing

GPCR dimers in brain. In this particular study, putative monomers and dimers were found to coexist (Ciruela et al., 1995). These studies were subsequently expanded to show through co- immunoprecipitation methods that A1R and metabotropic glutamate 1α receptors formed heterooligomers in rat cerebellum (Ciruela et al., 2001). This data complements immunohistochemical studies in cerebellum that showed that both these receptor subtypes co-localized in the dendritic tree of Purkinje and basket cells as well as large pyramidal cells of the human cerebral cortex

(Ciruela et al., 2001). Several of the dopamine receptor subtypes have also been shown to exist as dimers and higher order oligomers in brain. The D3 dopamine receptor and its naturally occurring truncation variant, D3nf, were shown to form homo-oligomers and hetero-oligomers in human brain (Nimchinsky et al., 1997). In a separate study, photoaffinity labeling of D2 dopamine receptors derived from human caudate nucleus revealed the presence of monomeric and dimeric D2 receptors. Through similar methods of detection, D2 dimers and higher order oligomers were also found in rat striatum tissue (Zawarynski et al., 1998). Some of these receptor – receptor interactions are sensitive to reducing agents suggesting the involvement of disulfide bonds (Gama et al., 2001; Zeng and Wess, 1999) although non-covalent interactions between transmembrane domains may be the dominant interaction that holds the oligomer together. Dimers of the m3 muscarinic receptor (Zeng and Wess, 1999) and hetero-dimers of the calcium sensing receptor and metabotropic glutamate 1α receptor (Gama et al., 2001) from rat

57 brain and bovine brain, respectively, have been shown to dissociate into monomeric subunits

after DTT or β-mercaptoethanol treatment. Finally, the two subunits (GABABR1 and

GABABR2) of the heteromeric GABAB receptor, have been shown to co-immunoprecipitate in cortical membrane preparations (Kaupmann et al., 1998), implicating the physiological

requirement of oligomeric assembly in pre- and postsynaptic function.

1.16 Introduction to Lipid Rafts

The fluid-mosaic model of the plasma membrane, as proposed by Singer and Nicolson,

has long been the accepted model for describing the biochemical nature of biological membranes

(Singer and Nicolson, 1972). This model predicts the plasma membrane to be a highly dynamic

and fluid environment that allows the random dispersion of protein and lipids within the lipid

bilayer. It is now known that the plasma membrane is not as fluid and mobile as once thought

and that protein movement is restricted, in part, by membrane domains called lipid rafts, thus

limiting their random diffusion within the cell membrane. Most of the evidence for this ordered

composition of the cell membrane came from an array of biophysical and molecular data from

fluorescence-recovery after photobleaching (Edidin et al., 1994), FRET (Damjanovich et al.,

1995), EM (Jenei et al., 1997; Vereb et al., 2000), and scanning force microscopy (Jenei et al.,

1997) studies. Lipid rafts are defined as plasmalemmal planar domains that are enriched in a

complement of proteins and lipids. They exhibit a high glycosphingolipid and cholesterol content

which confers a gel-like environment within the outer leaflet of the lipid bilayer (Rajendran and

Simons, 2005). This “ordered” phase can be attributed to sphingolipids, which are comprised of

saturated hydrocarbon chains that facilitate tighter packing and therefore minimal diffusion.

58 Caveolae represent a subtype of lipid rafts that exist as morphologically distinct

invaginations at the plasma membrane (Parton and Simons, 2007) (Figure 1-6). These lipid enriched entities move laterally on the cell surface while allowing the exchange of proteins and lipids between the raft domain and the surrounding liquid disordered phospholipid environment

(Rajendran and Simons, 2005). This dynamic regulation is believed to facilitate the formation of cell surface signalling platforms for the integration of various signalling molecules, thus ensuring specificity and efficiency in signal transduction processes. The caveolin proteins are unique to caveolae and are localized to the inner leaflet of the lipid bilayer (Anderson, 1998). They serve a dual role in maintaining the structural integrity of caveolae and by acting as a scaffolding protein that binds to a battery of receptors, signalling molecules, and adapter proteins (Williams and

Lisanti, 2004). There are three caveolin isoforms, each of which can serve as biochemical markers for the identification of caveolae; caveolin-1 is the most ubiquitously expressed as it is found in tissues including the lung, heart, and brain while caveolin-3 is specific to muscle. The expression of caveolin-1 in brain has been shown in neuronal cells such as hippocampal and dorsal root ganglion neurons (Bu et al., 2003; Galbiati et al., 1998) as well as glial cells such as astrocytes and oligodendrocytes (Cameron et al., 1997)

1.17 Platforms for Signalling Complex Assembly

As previously mentioned, caveolae appear to have an important role in maintaining signal

transduction fidelity by concentrating specific signalling molecules within a locally regulated

compartment. There are numerous examples of signalling cascades which are differentially

regulated either by caveolae or non-caveolae related lipid rafts. Some of these are described

59 A B

Figure 1-6. Morphology of Caveolae. (A) Electron micrograph of caveolae from adipo- cytes labelled with electron-dense marker. (B) Schematic of caveolae highlighting oligomers of caveolin protein inserted into the caveolar membrane. The amino and carboxyl terminus of caveolin are cytoplasmic and are connected by a hairpin loop that is embedded in the membrane. (Reprinted with permission from Parton and Simons, 2007)

60 below.

1.17.1 Nitric Oxide Synthase Signalling Pathway

The activity of endothelial nitric oxide synthase (eNOS) has been demonstrated to be

negatively regulated by plasmalemmal caveolae of endothelial cells. eNOS generates nitric

oxide, an important second messenger in vascular function including vasorelaxation and platelet

aggregation (Arnal et al., 1999). Plasma membrane eNOS has been shown to be targeted to

caveolae by N-myristoylation and palmitoylation (Shaul et al., 1996); this association is further

enhanced by the direct interaction of caveolin-1 (via its caveolin scaffolding domain) with eNOS

(Garcia-Cardena et al., 1997). Upon activation by agonists specific for bradykinin, acetylcholine,

and serotonin receptors, this interaction is antagonized by the binding of eNOS to calmodulin.

This results in eNOS dissociation from caveolin thereby relieving it from caveolin’s inhibitory

effect and allowing it to generate nitric oxide. Mutation of a putative caveolin binding motif

(Couet et al., 1997) in eNOS also blocks the ability of caveolin-1 to suppress nitric oxide release

(Garcia-Cardena et al., 1997). Furthermore, caveolin-1 knockout mice exhibit enhanced

constitutive and agonist-mediated eNOS activity (Drab et al., 2001; Razani et al., 2001). This was manifested as an inability to maintain arterial contractile tone confirming that vascular

physiology is profoundly regulated by caveolar modulation of eNOS activity.

Similarly, neuronal nitric oxide synthase (nNOS) has been shown to interact with

caveolin-1 as well as caveolin-3 in skeletal muscle to functionally inhibit the catalytic activity of

nNOS (Garcia-Cardena et al., 1997; Venema et al., 1997). Although, the functional aspects of

this interaction has not been as well characterized as for eNOS, the interaction between caveolin-

3 and nNOS appears to involve two distinct caveolin scaffolding domains (Venema et al., 1997).

1.17.2 The Ras GTPase-Raf Signalling Pathway

The Ras-Raf signalling pathway and its associated receptors have also been shown to be regulated by localization in lipid rafts/caveolar microdomains. The epidermal growth factor receptor, for instance, is tightly linked to this signalling cascade and is targeted to cholesterol- dependent rafts by a cysteine-rich extracellular sequence (Yamabhai and Anderson, 2002); however, there is still ongoing debate regarding whether these represent caveolae or other lipid raft subypes (Ringerike et al., 2002). It has been shown that Raf activation through H-Ras (but not K-ras) is attenuated by cholesterol depletion with methyl-β-cyclodextrin (mβCD) or by co- expression with a dominant negative mutant of caveolin-1 (Roy et al., 1999). This differential regulation of H-Ras and K-Ras mediated signalling appears to be due to the selective

palmitoylation of H-ras which targets it to caveolae (Hancock et al., 1990). Therefore, although

H-ras and K-ras were also both shown to be localized in purified lipid raft fractions (Roy et al.,

1999), this implicates that signalling through different Ras isoforms requires segregation to different types of lipid rafts. Interestingly, GTP loading redistributes H-ras from caveolae to nonraft microdomains where activated Raf localizes (Prior et al., 2001; Prior et al., 2003). These non-raft regions do not overlap with similar cholesterol-independent domains that activated K-

Raf is targeted (Niv et al., 2002; Prior et al., 2003). Although this may appear contradictory with the aforementioned studies, it suggests that H-ras mobility is highly dynamic and that while Raf activation may initially occur in caveolae, the final site of activated Raf may reside outside of these cholesterol-dependent compartments (Prior et al., 2001).

62 1.17.3 Heterotrimeric GTP Binding Protein Signalling

A substantial number of studies have implicated the involvement of caveolae in

heterotrimeric G protein signalling. Up until recently, most of this evidence was derived from the

localization of G proteins in purified detergent insoluble caveolar fractions (Chang et al., 1994;

Li et al., 1995; Lisanti et al., 1994). Some of these studies reported a significant enrichment of G proteins (eg: Gsα, Gi3α) in caveolae with concentrations that were increased by 8-fold relative to

total plasma membrane (Chang et al., 1994). Interestingly, activated Gsα does not appear to

localize in caveolae since mutationally or pharmacologically activated Gsα was absent from

detergent insoluble caveolar fractions and unable to interact with recombinant caveolin (Li et al.,

1995). This implies that the inactive GDP bound state of Gsα preferentially interacts with

caveolin. Indeed, it was demonstrated that peptides derived from the caveolin scaffolding domain

of caveolin-1 and -3 were able to inhibit GDP/GTP exchange and Gsα basal activity suggesting

that caveolin may serve as a negative regulator of Gsα activity.

1.18 Transcytosis

Transcytosis is a cellular mechanism employed by a variety of cells to transport

macromolecular (eg: immunoglubulins, lipoproteins, albumin) and smaller cargo (ions, vitamins)

from one side of a cell to the other via a membrane-bound carrier protein. This process allows for

the exchange of molecules across cellular barriers such as that which divides the lumen and the

opposite cytoplasmic face. This transport system has been well characterized in endothelial cells

although it has also been demonstrated to occur in other cell types including osteoclasts

(Morimoto et al., 2006) and neurons (Hemar et al., 1997). In polarized cell types such as in the

63 endothelium, transcytosis can be bi-directional with movement of molecules from the apical to

the basolateral side and vice versa.

The supposition that transcytosis is mediated by caveolae is supported by several

independent findings. A number of in vitro studies have demonstrated the transport of protein

molecules > 20 Å in diameter across the microvascular endothelia through plasmalemmal

vesicles (Predescu et al., 1998; Predescu et al., 1997). Some of these protein molecules were

modified to function as tracers for detection by EM and were found to localize in these vesicles

(Bruns and Palade, 1968; Predescu et al., 1988). The transport of these molecules across cellular monolayers was inhibited by N-ethylmaleimide (Predescu et al., 1994), a compound known to interfere with exocytic and endocytic vesicular transport, and filipin (Schnitzer et al., 1994), a sterol binding agent that disrupts caveolae. Furthermore, endothelial caveolae were demonstrated to pinch off from the plasma membrane in a dynamin dependent manner involving GTPase activity (Schnitzer et al., 1996) confirming that caveolae contain dynamin, which appears to be involved in vesicular scission from the plasma membrane (Henley et al., 1998; Oh et al., 1998).

Several studies from caveolin-1 knockout animals have also implicated a role for caveolae in transcytosis. In these transgenic animals, it was shown that radioiodinated albumin delivery from the endothelial lumen to the interstitial space was completely abolished (Schubert et al., 2001).

Collectively, these observations provide an alternative mechanism to the hypothesis that transcytosis is mediated by pores that form cylindrical channels through the cell allowing the passage of solutes and molecules of various sizes (Michel and Curry, 1999).

1.19 Endocytosis

Although the internalization of many different membrane proteins, including GPCRs, has

64 been shown to occur through clathrin coated pits, the utilization of specific inhibitors of this

pathway has led to the identification of alternative endocytic pathways that are clathrin

independent. Indeed, the evidence that cholesterol depletion from cell membranes not only

renders certain proteins and lipids detergent insoluble, but also inhibits their uptake into

intracellular compartments is suggestive of an endocytic process likely to involve lipid rafts (Di

Guglielmo et al., 2003; Nichols et al., 2001; Puri et al., 2001). The involvement of caveolae in

this process has been shown through budding of caveolae (as detected by caveolin-1 GFP) and

its resident proteins into the cell (Thomsen et al., 2002) and the localization of dynamin at the

necks of plamalemmal caveolae (as discussed in the Transcytosis section). Although it is not entirely clear exactly how caveolae bud from the plasma membrane, caveolar endocytosis does not appear to be a constitutive process (Thomsen et al., 2002) and likely involves some degree of external regulation. Phosphorylation by specific kinases such as PKC or by certain non-receptor tyrosine kinases such as Src appear to regulate caveolar internalization since inhibitors to either of these appear to abolish internalization of various molecules via caveolae (Dangoria et al.,

1996; Parton et al., 1994). This is supported by evidence that phosphatase inhibitors, such as

okadaic acid, trigger enhanced internalization of caveolae (Thomsen et al., 2002) Indeed,

caveolin-1 phosphorylation of a tyrosine at position 14 (Y14) has been shown to be a substrate of

kinase activity in response to a variety of hormonal and stressful stimuli (Aoki et al., 1999;

Mastick et al., 1995; Sotgia et al., 2000; Volonte et al., 2001). Interestingly, albumin binding to

GP60, a caveolae localized protein, appears to stimulate tyrosine phosphorylation of caveolin-1

(Tiruppathi et al., 1997). The addition of genistein, a tyrosine kinase inhibitor, blocks the uptake

and subsequent transcytosis of albumin through caveolae implicating a causal link between these

two processes. Therefore, it appears that tyrosine phosphorylation of caveolin-1 is required for

65 caveolar internalization of growth factors, viruses, and other molecules and that this is dependent

on the activation of specific signalling cascades by these same ligands.

The delivery of molecules into early caveolar endosomes are distinct from those

transferring labeled endosomes that appear shortly after clathrin-mediated internalization. These

early caveolar endosomes are termed caveosomes (due to the presence of caveolin-1) and

represent intermediate “holding stations” that have distinct morphological and functional

characteristics. Caveosomes are not continuous with the plasma membrane and can exist as

clusters near the plasma membrane in a ligand independent manner (Nichols and Lippincott-

Schwartz, 2001). Furthermore, they have a neutral luminal pH unlike classical early endosomes

that have an acidic pH that facilitates ligand detachment. In addition, these organelles are quite stable and can hold their cargo for several hours before depositing their contents into tubular, caveolin-free membrane vesicles (Pelkmans et al., 2001) that can fuse with the ER. Hence, the trafficking itinerary of molecules internalized through caveolae can also be distinct from classical endosomal internalization. Indeed, the transcytosis of albumin (Tiruppathi et al., 1997)

and other plasma proteins (Predescu et al., 1998), as discussed earlier, represents one trafficking route that is unique to molecules that are internalized through caveolae.

1.20 Potocytosis

Another cellular process that has been demonstrated to involve caveolae is potocytosis.

Similar to endocytosis, potocytosis represents a mechanism used for the uptake of small molecules into the cell. However, it is distinct from endocytosis in that there are no endosomes, caveosomes, or other related vesicular structures involved. Instead, molecules bind to receptors residing in caveolae which in turn close to form a sealed microenvironment that is continuous

66 with the plasma membrane (Smart et al., 1999). These molecules subsequently gain intracellular access by direct diffusion without vesicular budding. After its contents have been emptied, the caveolar pit opens up to begin another cycle. The uptake of small molecules including vitamins such as folate (Kamen et al., 1988) and retinol (vitamin A) (Malaba et al., 1995) has been suggested to occur by this process although the precise mechanism remains unclear. In the case of folate, the acidic pH of the sealed caveolar compartment favours dissociation of the ligand from the receptor (Kamen et al., 1988). The generation of a concentration gradient in the

caveolar space may be required to facilitate the movement of folate by an anionic carrier (Smart

et al., 1999).

1.21 Thesis Rationale and Research Objectives

The concept that G protein coupled receptors are assembled as oligomeric complexes has

gained increased acceptance during the course of research compiled in this thesis and is now

recognized as a fundamental process in GPCR maturation. Before the course of work in this

thesis had begun, our laboratory had shown that the D1 and D2 dopamine receptors as well as

5HT1B and 5HT1D receptors formed dimers and higher order oligomers (George et al., 1998; Ng

et al., 1994; Xie et al., 1999). However, there was no consensus on whether GPCR oligomers

were assembled in the ER (Overton and Blumer, 2000; Zeng and Wess, 1999) or whether

receptor monomers were induced to oligomerize upon agonist stimulation (Cornea et al., 2001;

Rocheville et al., 2000a). In some cases, oligomeric trafficking could be inhibited by changing

the function of a protomer within the oligomer such that it conferred a dominant negative effect on the wild-type receptor protomer, suggesting that GPCR oligomers were assembled shortly after biosynthesis in the ER (Coge et al., 1999; Grosse et al., 1997; Zhu and Wess, 1998).

67 Similarly, our laboratory had previously demonstrated that a specific D2 dopamine receptor

mutant could “hetero-oligomerize” with and inhibit the cell surface trafficking of the wild-type

D2 receptor by oligomerization. Curiously, this mutant was not intracellularly retained and could independently traffic to the plasma membrane normally as homo-multimeric species (Lee et al.,

2000a).

Hypothesis 1: D1 dopamine receptor oligomers are subject to conformational scrutiny prior to exit from the cellular ER quality control

The first objective, in this thesis, was to determine whether D1 dopamine receptors formed ER-derived oligomeric complexes. We hypothesized that D1 dopamine receptors undergo constitutive oligomerization. Furthermore, we were interested in using specific D1 receptor mutants as a tool to explore what conformational requirements enabled or prevented a receptor oligomer from exiting the ER and being trafficked to the cell surface. Based on some of the data from the dominant negative D2 receptor mutants, we hypothesized that certain regulatory mechanisms in the cell monitor the fidelity of GPCR oligomer formation. In addition we hypothesized that the conformation of an oligomer is subject to stringent quality control in the ER before cell surface transport.

Hypothesis 2: D1 dopamine receptors are localized in and functionally regulated by caveolae-related rafts

Given the localization of various GPCR signalling components in lipid rafts and caveolae, our second objective was to determine whether mature cell surface bound D1 dopamine receptors were localized in and functionally regulated by these microdomains. We

68 hypothesized that lipid rafts act as docking stations for the integration of receptors, G proteins, and effectors in order to facilitate signalling and to enable the delivery of D1 receptors into the cell. We also explored the possibility that caveolae might be an alternative route of agonist- induced endocytosis for D1 receptors and whether this might occur in brain.

Hypothesis 3: Palmitoylation of D1 dopamine receptors is important for agonist-induced internalization

The third objective of this thesis was to explore the potential role that palmitoylation might have on D1 receptor internalization. For some GPCRs, receptor palmitoylation has been shown to regulate access to phosphorylation sites in the receptor by various kinases. Since phosphorylation is a requirement for classical arrestin dependent clathrin mediated internalization, we predicted that the manipulation of palmitoylation sites would modulate the kinetics and/or extent of agonist mediated receptor endocytosis.

Hence, the overall objectives of this thesis were to investigate the post-translational processing and cell surface distribution of D1 dopamine receptor oligomers, two important events in the life cycle of this important GPCR.

69 2 MATERIALS AND METHODS

2.1 Materials and Methods for Chapter 3

2.1.1 DNA constructs and Site-Directed Mutagenesis - Full length human D1 receptor cDNA was used for site-directed mutagenesis studies. The Asp residue in TM3 (position 103) was mutated to Ala and Glu and designated D103A and D103E, respectively. Two Ser residues

(positions 198 and 199) in TM5 were mutated to Ala and the mutant receptor was denoted as

S198A/S199A. The Asp residue at the junction of TM3 and ICL2 (position 120) was mutated to

Ala and Asn and designated D120A and D120N, respectively. All receptor mutations were generated using the Quickchange Site-directed Mutagenesis Kit (Stratagene, La Jolla, CA) as suggested by the manufacturer and using the following pairs of oligonucleotides (sense and antisense): D103A – GGGTGGCCTTTGCCATCATGTGCTCC and

GGAGCACATGATGGCAAAGGCCACCC; D103E – GGGTGGCCTTTGAAATCATGTGC

TCC and GGAGCACATGATTTCAAAGGCC ACCC; S198A/S199A - GGACATATGCCAT

CTCAGCCGCCGTAATAAGCTTTTACATCCC and GGGATGTAAAAGCTTATTACGGC

GGCTGAGATGGCATATGTCC; D120A –

CCTCTGTGTGATCAGCGTGGCCAGGTATTGGGCTATCTCCAAGCCC and

GGGCTGGAGATAGCCCAATACCTGGCCACGCTGATCACACAGAGG; D120N –

CCTCTGTGTGATCAGCGTGAATAGGTATTGGGCTATCTCCAGCCC and

GGGCTGGAGATAGCCCAATACCTATTCACGCTGATCACACAGAGG. The D103A,

D103E, and S198A/S199A receptors had a cMyc epitope subsequently inserted at the N- terminus of the receptor with an appropriate Kozak sequence through a two step PCR protocol utilizing Pfu enzyme. Wild-type D1 cDNA also had a HA epitope inserted at the N-terminus in a

70 similar fashion. All receptor isoforms were prepared in the pcDNA3 mammalian expression vector and verified by DNA sequencing (The Centre for Applied Genomics, Hospital for Sick

Children, Toronto, ON).

2.1.2 Cell Culture and DNA Transfection - HEK293t and COS7 cells were maintained as monolayer cultures at 37°C in minimum essential medium (Invitrogen, Carlsbad, CA) and alpha minimum essential medium (University of Toronto), respectively, supplemented with 10% fetal bovine serum, antimycotic and antibiotic. HEK293t cells were grown to 80% confluence before being transiently transfected using Lipofectamine (Invitrogen). For co-expression experiments in which the amount of one construct used was varied, the total amount of cDNA transfected was kept constant by the addition of a compensating amount of pcDNA3. Transfected cells were grown for 48 hrs before harvesting for functional assays.

2.1.3 Membrane Preparation - Cells were washed in PBS and centrifuged at 1500 g to obtain a pellet. Cell lysates were prepared by polytron disruption in ice-cold 5 mM Tris-HCl, 2 mM

EDTA buffer, containing protease inhibitors (5 μg/ml leupeptin, 10 μg/ml benzamidine, and 5

μg/ml soybean trypsin inhibitor). Lysates were centrifuged at 1000 g for 10 minutes to separate nuclei and unbroken cells. Crude membrane fractions were prepared by centrifuging the supernatant at 20,000 g for 25 min. Membrane protein was determined by the Bradford assay according to the manufacturer’s instructions (BioRad, Hercules, CA).

2.1.4 Drugs - The following dopaminergic ligands (Sigma, St.Louis, MI) were used in 10 μM concentrations and dissolved in either DMSO, ethanol, or water as indicated by the

71 manufacturer. Ascorbic acid 0.1% was added when dopamine was used. Cells were treated with

one of the following agonists (pergolide, SKF 81297, SKF 38393, apomorphine, or dopamine),

antagonists (SCH 23390), or inverse agonists ((+) butaclamol, flupenthixol, or fluphenazine).

2.1.5 Radioligand Saturation Binding - Near maximal saturation receptor binding was

determined with 15 - 25 μg of membrane protein with 1 nM [3H]-SCH 23390 to obtain an

estimate of receptor densities. 1 μM (+) butaclamol was used to determine non-specific binding.

Bound ligand was isolated by rapid filtration through a 48-well cell harvester (Brandel,

Gaithersburg, MD), using GF/C filters.

2.1.6 Whole Cell Radioligand Binding – Specific dopaminergic surface binding was

determined by subtracting dopamine displaced specific [3H] SCH23390 binding from specific

[3H] SCH23390 binding and taken to reflect D1 receptor cell surface density. Specific binding to

surface receptors with [3H] SCH23390 was first determined in the following manner.

Transfected cells were trypsinized and incubated with 1 nM [3H] SCH 23390 to determine total

binding. Non-specific binding was determined by incubating cells at 4°C with 1 nM [3H] SCH

23390 and an excess of unlabeled close congener, (+) butaclamol (1 μM). Specific binding was subsequently determined by total binding – non-specific binding. Specific dopamine displaced specific [3H] SCH23390 binding was then determined in the following manner. Transfected cells

were trypsinized and incubated with 1 nM [3H] SCH 23390 and an excess of 100 μM dopamine

(in 0.1% ascorbic acid) to compete off total surface bound [3H] SCH 23390; the amount of

radioligand displaced was taken to represent total binding. Non-specific binding was determined

by incubating cells with 1 nM [3H] SCH 23390, 100 μM dopamine, and 1 μM unlabeled (+)

72 butaclamol. Specific dopamine displaced [3H] SCH 23390 binding was determined by total

dopamine displaced [3H] SCH 23390 binding – non-specific dopamine displaced [3H] SCH

23390 binding. All reactions were mixed gently and incubated for 3 hrs on ice to prevent receptor internalization. Non-specific binding accounted for approximately 10-15 % of total bound ligand. Bound ligand was isolated by rapid filtration as mentioned previously.

2.1.7 Cell Surface Fluorometric Analysis – 96-well plates were coated with poly-L-ornithine before being plated with 50,000 cells/well. Cells were transfected 24 hrs later with HA-D1 and cMyc-D103A to yield expression in a 1:1 ratio. Cells were then fixed with 4% paraformaldehyde after 48 hours and subsequently washed with PBS. Blocking was performed with 4% BSA at room temperature for 30 min and cells were incubated with rat anti-HA (Roche, Penzburg,

Germany) for 1.5 hrs. After primary antibody treatment, cells were washed with 4% BSA and incubated for an additional 1.5 hrs with FITC conjugated anti-rat IgG (Sigma). Samples were excited at 485 nm and emission was measured at 530 nm using the Cytofluor 4000 microplate reader (Applied Biosystems, Foster City, CA).

2.1.8 Co-immunoprecipitation – Transfected HEK293t cells were washed in PBS and scraped with homogenization buffer consisting of 20 mM HEPES, 100 mM NaCl, 1 mM EDTA, and protease inhibitors. Cells were harvested by polytron disruption and solubilized by addition of

NP-40 to a final concentration of 1% (v/v) for 2 hours at 4°C. Samples were centrifuged at

15,000 g for 25 min and the supernatant was collected for protein determination by the Bradford reaction; 2 mg of total cell lysate was used for immunoprecipitation. After pre-clearing for 30 min with 10 μl protein G agarose beads (Sigma), the lysates were incubated overnight with 3 μg

73 of rat monoclonal anti-HA (Roche). Immuno-complexes were subsequently incubated overnight

with 50 μl of protein G agarose beads and washed 5 times with ice-cold wash buffer (20 mM

HEPES, 100 mM NaCl, 1 mM EDTA, 1 % NP-40, and protease inhibitors). The proteins were

eluted in 80 μl Laemmli buffer (63 mM Tris-HCl, 10% glycerol, 2% SDS, 0.0025% bromophenol blue) for 30 min at 37°C before immunoblotting.

2.1.9 Gel Electrophoresis and Immunoblotting - Membrane preparations were solubilized in

Laemmli buffer with 5% β-mercaptoethanol for 20 min at RT. Samples were separated on pre- cast 10% polyacrylamide gels (Invitrogen) for 2 hours. Proteins were transferred to polyvinylidene fluoride (PVDF) membranes (Amersham Biosciences, Piscataway, NJ) at 35 V for 3 hours and blocked overnight in 10% skim milk powder and then incubated with either rat monoclonal anti-HA (Roche) or mouse monoclonal anti-cMyc (Upstate Biotechnology, Lake

Placid, NY) for 3 hours. Blots were rinsed with TBS and incubated with horseradish peroxidase- conjugated goat anti-rat, or mouse secondary antibodies (Santa Cruz Biotechnology, Santa Cruz,

CA; Bio-Rad) for 1.5 hours. Immunoreactivity was detected by enhanced chemiluminescence using an ECL Plus Kit (Amersham Biosciences).

2.1.10 Cell Surface Biotinylation – Biotinylation was carried out as described by Cao et al (Cao et al., 1998) with minor modifications. Briefly, transiently transfected COS7 cells were biotinylated with 0.5 mg/ml sulfo-NHS-biotin (Pierce, Rockford, IL) for 30 min at RT.

Unreacted biotin was quenched with Tris-buffered saline three times for 5 min each. Cells were solubilized in extraction buffer (0.5% Triton X-100, 10 mM Tris-Cl, pH 7.5, 120 mM NaCl, 25 mM KCl, protease inhibitors) and rotated for 3 hrs at 4°C. The lysate was clarified by

74 centrifugation at 10,000 rpm for 20 min. The supernatant was collected and

immunoprecipitations were carried out with 8 μg of mouse anti-cMyc (Upstate Biotechnology)

or rabbit anti-HA (BD Biosciences, San Jose, CA) O/N at 4°C. Samples were incubated with 50

μl protein G agarose PLUS (Pierce) for 3 hrs. Antigen complexes were subsequently washed

sequentially in high salt wash buffer (HSB; 20 mM Tris-HCl, 7.5, 120 mM NaCl, 25 mM KCl,

0.1% SDS, 0.5% Triton X-100), 1 M NaCl in HSB (pH 7.0), and low salt wash buffer (10 mM

Tris-HCl, 7.5). Biotinylated proteins were eluted by incubation of beads in SDS-PAGE sample

buffer at 37°C for 15 min and resolved by 12% polyacrylamide gel electrophoresis. Proteins

were transferred to PVDF and blocked overnight in 10% skim milk powder. Biotinylated proteins were then complexed with horseradish peroxidase by incubating membranes with

Vectastain ABC Elite detection system (Vector Laboratories, Burlingame, CA). Biotinylated

proteins were subsequently detected with ECL. The protein bands were scanned by densitometric

analysis and relative intensities were quantified using NIH image software version 1.33.

2.1.11 Whole Cell cAMP Determination - Basal cAMP levels were measured from HEK293t

cells transiently expressing D1 or the mutated D1 receptors. Cells expressing D1 were incubated

with 10 μM dopamine for 15 min in the presence of 100 μM IBMX (Sigma). Cells expressing

D103A were incubated with 10 μM of either dopamine, pergolide, SKF 38393, SKF 81297,

fluphenazine, flupenthixol, or (+) butaclamol for 15 min before cAMP accumulation was

evaluated. Cells were lysed in 0.1 N HCl for 20 min and the supernatant was assayed for cAMP

accumulation using a highly sensitive enzyme-linked immunoassay kit (Cayman Chemical, Ann

Arbor, MI) according to the manufacturer’s instructions.

75 2.1.12 Immunofluorescence Microscopy - HEK293t cells were transiently transfected with

either HA-D1/pcDNA3, cMyc-D103A/pcDNA3, or HA-D1/cMyc-D103A on glass coverslips

that were pre-treated with poly-L-ornithine (Sigma). Forty-eight hours after transfection, cells were washed twice with PBS and incubated for 2 hours in the absence or presence of 10 μM pergolide (Sigma) or 10 μM SCH 23390 (Sigma) at 37 °C. After incubation, cells were washed twice with PBS and fixed in 4% paraformaldehyde at 4°C for 30 min. To detect the subcellular

localization of HA-D1 and cMyc-D103A, cells were blocked and, where indicated,

permeabilized with a buffer containing 0.1% Triton and 5% normal goat serum in PBS for 30

min at room temperature (RT). The cells were then incubated with rat anti-HA monoclonal antibody (Roche) at a 1:200 dilution and mouse anti-cMyc monoclonal antibody (Santa Cruz

Biotechnology) at a 1:150 dilution for 1 hour at RT. After one wash with PBS for 30 min, the cells were incubated with FITC-conjugated anti-rat IgG (Sigma) at a 1:32 dilution and TRITC-

conjugated anti-mouse IgG (Sigma) at a 1:64 dilution for 1 hour at RT. After one wash with PBS

for 30 min, coverslips were mounted, and FITC-labeled HA-D1 and TRITC-labeled cMyc-

D103A were visualized with a Zeiss LSM-510 laser confocal microscope. For detection of calnexin, cells were incubated with rabbit anti-calnexin polyclonal antibody (Stressgen

Biotechnology, Victoria, BC) at a 1:200 dilution for 1 hour at RT. Cells were subsequently incubated with Alexa-546-conjugated anti-rabbit IgG (Molecular Probes Inc, Eugene, OR) at a 1:

500 dilution for 1 hour at RT.

2.1.13 Co-localization Image Analysis – Dual channel co-localization analysis was performed on randomly selected images using the NIH Image J software with the co-localization analyses plugin. The Mander’s coefficients, Mred and Mgreen, represent the number of co-localized pixels

76 expressed as a fraction of the total number of pixels within the red and green channel,

respectively (Manders et al., 1993). The Mander’s coefficients range from 0 (no colocalization) to 1 (complete co-localization) and are independent of the pixel intensities within each respective channel. Pixels with a value of zero in both channels were ignored as background.

2.1.14 Time-resolved Fluorescence Resonance Energy Transfer (trFRET) – Cells co-

expressing D1 and D103A (in the presence or absence of ligand) were trypsinized 48 hrs after

transfection and counted. The FRET protocol was performed as described by McVey et al

(McVey et al., 2001) using 15 nM Eu3+-labeled anti-HA (donor) and 5 nM APC-labeled anti-

cMyc (acceptor) (Perkin Elmer, Norwalk, CT). Time-resolved FRET was measured using a

Victor3 microplate reader (Perkin Elmer) with excitation of Eu3+ at 615 nm and emission of APC detected at 665 nm.

2.1.15 Statistical Analysis – All pharmacological data were analyzed using Prism (GraphPad

Software, San Diego, CA). Saturation binding curves were generated using non-linear least squares regression curve fitting. One tailed, unpaired Student’s t-Tests were performed to compare groups of data sets. Statistical significance at the p < 0.05 level is denoted with ∗.

2.2 Materials and Methods for Chapters 4 and 5

2.2.1 Chemicals - Concanavalin A was purchased from Calbiochem (La Jolla, CA). Filipin,

methyl-β-cyclodexytrin, and H89 were purchased from Sigma (St. Louis, MO). [3H]SCH23390

(85 Ci/mmol) and [35S]GTPγS (1250 Ci/mmol) were purchased from Perkin Elmer Life Sciences

(Boston, MA).

77 2.2.2 DNA constructs and site-directed mutagenesis - We used the full length N-terminal HA-

tagged D1 dopamine receptor cDNA that was cloned into the mammalian expression vector,

pcDNA3.1 (Invitrogen, Carlsbad, CA) as a template for site-directed mutagenesis studies. The

mutant dopamine D1 constructs, F313A, W318A, W321A, FWW/A, C347A/C351A and mutant

caveolin-1, P132L, were generated using the Quickchange site-directed mutagenesis kit

(Stratagene, La Jolla, CA) as suggested by the manufacturer and using the following pairs of

oligonucleotides (sense and antisense): F313A –

CATTGATTCCAACACCGCCGACGTGTTTGTGTGG and

CCACACAAACACGTCGGCGGTGTTGGAATCAATG; W318A –

CTTTGACGTGTTTGTGGCCTTTGGGTGGGC and

GCCCACCCAAAGGCCACAAACACGTCAAG; W321A –

GTTTGTGTGGTTTGGGGCCGCTAATTCATCCTTG and

CAAGGATGAATTAGCGGCCCCAAACCACACAAAC; FWW/A –

GATTCCAACACCGCTGACGTGTTTGTGGCGTTTGGGGCGGCTAATTCATCC and

GGATGAATTAGCCGCCCCAAACGCCACAAACACGTCAGCGGTGTTGGAATC;

C347A/C351A – CAACCCTCTTAGGAGCCTACAGACTTGCCCCTGCGACGAATAATGC

and GCATTATTCGTCGCAGGGGCAAGTCTGTAGGCTCCTAAGAGGGTTG; caveolin-1

P132L – CTGGGCAGTTGTGCTGTGCATTAAGAGTTTCCTG and

CAGGAAACTCTTAATGCACAGCACAACTGCCCAG. The wild-type cMyc caveolin-1 was a kind gift provided by Dr. Bryan Roth (University of North Carolina). The dominant negative

GRK2-K220R was a kind gift from Dr. Jeffrey Benovic (Thomas Jefferson University). The

caveolin-1-GFP and D1R-GFP constructs were designed by cloning caveolin-1 and D1R,

respectively, into pEGFP-N1 (BD Biosciences, San Jose, CA) in frame with GFP at the carboxyl

78 tail. The D1R-mRFP construct was generated in a similar fashion by isolating mRFP (in pRSETb

vector) by PCR and replacing GFP from the pEGFP-N1 vector with mRFP. The D1R-Rluc

construct was created by replacing the eGFP from D1R-GFP with Rluc.

2.2.3 Cell culture and DNA transfection - COS7 and HEK293t cells were maintained as

monolayer cultures at 37 °C in alpha minimum essential medium (MEM) (University of

Toronto) or advanced minimum essential medium (Invitrogen), respectively, supplemented with

10% fetal bovine serum, antimycotic and antibiotic. Cells were grown to 80% confluence before

being transfected using Lipofectamine 2000 (Invitrogen). For co-expression experiments, the total amount of cDNA transfected under control conditions was kept constant by the addition of

a compensating amount of pcDNA3. Transfected cells were grown for 48 hrs before harvesting

for all functional assays.

2.2.4 Detergent-free sucrose gradient fractionation - Caveolae-enriched fractions were

prepared by separating whole cell lysates on a discontinuous sucrose gradient column by

ultracentrifugation, as described previously (Song et al., 1996). Each gradient column was

prepared with transfected COS-7 cells from three 100 mm dishes or from 3 mg of whole rat brain

lysate. Lysates were scraped into 2 ml of buffer containing 500 mM sodium carbonate (pH 11)

and sonicated before bringing to 45% sucrose (w/v) by adding 2 ml 90% sucrose, prepared in

MBS (25 mM 2-[N-Morpholino] ethanesulfonic acid and 150 mM NaCl). The resulting cell

suspension was placed at the bottom of a 12-ml ultracentrifuge tube. A discontinuous gradient was prepared by sequentially layering 4 ml of 35 % sucrose and 5% sucrose (prepared in MBS and carbonate buffer) on top of the 45% sucrose bed. After centrifugation at 35,000 rpm for 20

79 hrs at 4 °C, 1 ml fractions were collected from the top of each gradient and subjected to 20%

trichloroacetic acid precipitation. Protein pellets were subsequently washed with acetone and

resuspended in a 1:1 solution of 5:2 lysis buffer (5 mM Tris-HCl, 2 mM EDTA) and Laemmli

buffer. An equal volume of each fraction was separated on SDS-PAGE and immunoblotted with

the antibodies indicated. The antibodies used were anti-Na+/K+ ATPase (Developmental Studies

Hybridoma Bank at the University of Iowa, Iowa City, IA), mouse anti-caveolin-1 (BD

Biosciences), anti-HA-HRP (horseradish peroxidase) (Roche, Penzburg, Germany), rabbit anti-

Gsα (Santa Cruz Biotechnology, Santa Cruz, CA), goat anti-clathrin HC (Santa Cruz

Biotechnology), mouse anti-ARF6 (Santa Cruz Biotechnology), and rat anti-D1DR (Sigma). The

protein bands were scanned by densitometric analysis and relative intensities were quantified

using NIH image software version 1.33.

2.2.5 Co-immunoprecipitation - Transfected COS7 cells were washed in PBS and scraped into

homogenization buffer consisting of 50 mM Tris-HCl, 150 mM NaCl, 1 mM EDTA, 2.5 mM

MgCl2, and protease inhibitors. Cells were harvested by polytron disruption and centrifuged at

1,000 g for 10 min. 2 mg of supernatant was collected and solubilized by addition of NP-40 to a

final concentration of 1% (v/v) for 2 hours at 4°C. Samples were centrifuged at 20,000 g for 20

min and the supernatant was collected and re-centrifuged again. After pre-clearing for 20 min

with 20 μl protein G agarose beads (Sigma), the lysates were incubated overnight with 5 μg of

rabbit polyclonal anti-HA (BD Biosciences) or mouse monoclonal anti-cMyc (Upstate

Biotechnology, Lake Placid, NY) antibodies. Immuno-complexes were subsequently incubated

overnight with 60 μl of protein G agarose beads and washed 4 times with ice-cold wash buffer

(20 mM HEPES, 100 mM NaCl, 1 mM EDTA, 1 % NP-40, and protease inhibitors). The

80 proteins were eluted in 70 μl Laemmli buffer by boiling for 5 min before immunoblotting.

Immunoprecipitation from rat brain was performed similarly with the exception that 1 mg of rat brain membranes was solubilised in CHAPS buffer (50 mM Tris-HCl, 125 mM NaCl, 0.1 mM

EGTA, 20 mM CHAPS, 2mM DTT, and protease inhibitors) before immunoprecipitation with goat anti-D1 (Chemicon, Temecula, CA) and immunoblotting with mouse anti-caveolin-1.

Immunoprecipitated proteins were resolved by 12% SDS-PAGE under reducing conditions.

Proteins were transferred to PVDF before blocking in 5% skim milk for 1 hr. Blots were incubated with the appropriate primary antibodies overnight (O/N) in 1% skim milk before incubation with HRP-conjugated secondary antibodies (BioRad, Hercules, CA) for 1.5 hrs.

Immunoreactivity was detected by enhanced chemiluminescence using an ECL Plus Kit (GE

Healthcare, Waukesha, WI).

2.2.6 Cell surface biotinylation - Biochemical analysis of receptor endocytosis using cleavable biotin was performed as described previously with some modifications. Briefly, transfected cells were incubated with 0.5 mg/ml sulfo-NHS-S-S-biotin (Pierce, Rockford IL) for 40 min at RT.

Cells were then rinsed 3 X 5 min with Tris-buffered saline to quench the biotinylation reaction.

Biotinylated cells were then treated with 10 μM SKF81297 for 30 min at 37 °C to induce endocytosis. The remaining cell surface biotin was cleaved by washing cells with glutathione cleavage buffer (50 mM glutathione, 75 mM NaCl, 75 mM NaOH, 10% FBS) for 2 X 15 min at

4 °C. Unreacted glutathione was subsequently quenched with 50 mM iodoacetamide (in PBS) for

3 X 5 min at 4 °C. Cells were extracted in buffer containing 0.5% Triton-X100 (v/v), 10 mM

Tris-Cl, pH 7.5, 120 mM NaCl, 25 mM KCl, and a protease inhibitor cocktail (Sigma) by rocking at 4 °C for 3 hrs. The extracts were cleared of insoluble debris by centrifuging at 15,000

81 g for 20 min. Receptors were immunoprecipitated from the clarified supernatant by incubating

the lysate with 5 μg anti-HA polyclonal antibody (BD Biosciences) O/N, and then with 50 ul

protein G agarose PLUS beads (Pierce) for 3 hrs. Immunoprecipitations were washed sequentially with high salt buffer (HSB - 0.1% SDS, 0.5% Triton X-100, 20 mM Tris-HCl,

7.5, 120 mM NaCl, 25 mM KCl), 1 M NaCl in HSB, and low salt wash buffer (10 mM Tris-HCl,

7.5) for 20 sec/wash. Immunoprecipitated proteins were eluted by incubating in Laemmli buffer

for 25 min at 37 °C and resolved by 12% SDS-PAGE under non-reducing conditions. Proteins

were transferred to PVDF before blocking in 5% skim milk O/N. Biotinylated receptors were

detected by incubating membranes with Vectastain Elite ABC reagent (Vector Labs,

Burlingame, CA) for 30 min followed by HRP detection with ECL.

2.2.7 Membrane Preparation - Transfected cells were washed extensively in PBS and

centrifuged at 1,500 g to obtain a pellet. Cell lysates were prepared by polytron disruption in ice-

cold 5:2 buffer, containing protease inhibitors. Lysates were centrifuged at 1,000 g for 10

minutes to separate nuclei and unbroken cells. Crude membrane fractions were prepared by

centrifuging the supernatant at 20,000 g for 25 min. Membrane protein was determined by the

Bradford assay according to the manufacturer’s instructions (BioRad).

2.2.8 Radioligand binding - For whole cell binding experiments, transfected COS7 cells were

seeded onto 24-well plates (pre-treated with poly-L-ornithine) at a density of 1.75 X 105 cells/well. Cells were pre-incubated with the appropriate treatments before exposure to agonist.

Cells were washed with ice-cold buffer containing 1 mM EDTA and 50 mM Tris-HCl for 2 min to dissociate agonist before rinsing with 50 mM Tris-HCl for 1 min. Total binding was

82 determined by incubating cells with 2 nM of the D1R antagonist, [3H]SCH23390 (prepared in

antagonist binding buffer) on ice for 3 hrs. Non-specific binding was defined by [3H]SCH23390

binding in the presence of 1 μM (+) butaclamol. Cells were subsequently washed with ice-cold

wash buffer (50 mM Tris-HCl) for 3 X 1 min before lysing with 0.2 N NaOH for 20 min.

Lysates were resuspended in scintillation fluid and radioactivity was detected by a Beckman LS

6500 scintillation counter. For saturation binding analysis, membrane preparations (as described

above) were used for radioligand binding at [3H]SCH23390 concentrations of 4 nM, 1 nM, 500

pM, 100 pM, and 10 pM. Binding was performed at RT for 1.5 hrs before bound ligand was isolated by rapid filtration through a 48-well cell harvester (Brandel, Gaithersburg, MD), using

GF/C filters.

2.2.9 Immunocytochemistry and Confocal Microscopy - HEK293t cells co-transfected with

D1DR-GFP and cMyc-caveolin-1 were grown on glass coverslips in 6-well plates until 20-40% confluence. 48 hrs after transfection, they were washed 3 X with PBS/0.2% BSA (buffer B) and

fixed with freshly prepared 4% paraformaldehyde. Cells were washed with 0.02 M glycine to

quench remaining reactive aldehyde groups. The permeablization of cells was carried out in the

presence of 0.1% saponin in PBS for 5 min. After blocking in buffer B for 1 hr, fixed cells were

washed with PBS and incubated with an anti-cMyc monoclonal antibody (1:1000) (Upstate) for

2 hrs at RT. Cells were then washed 2 X in buffer B and once in PBS before incubating with

TRITC-conjugated anti-mouse secondary antibody (Sigma) at 1:1000 dilution for 2 hrs at RT.

Cells were then washed 2 X in buffer B and once in PBS for 5 min before glass coverslips were

mounted. Images were acquired with an X63 lens on a Zeiss LSM510 confocal microscope. For

the live cell monitoring of D1DR and caveolin-1, HEK-293t cells were co-transfected with

83 cDNA encoding D1DR-mRFP and caveolin-GFP for 48 hrs. To activate D1R, 10 μM SKF81297 was administered to these cells, and confocal images were acquired every 2-5 minutes over an 80 minute period using an X63 deep lens equipped on a Zeiss LSM510 confocal microscope.

2.2.10 Bioluminescence Resonance Energy Transfer (BRET) - To detect interactions between

D1R and caveolin-1, BRET studies were performed in which the D1R-Rluc fusion construct was co-transfected with increasing molar concentrations of caveolin-1-eGFP or empty vector eGFP

(up to a 10-fold difference) in HEK293t cells. Cells were transfected in 6-well plates and 24 hrs later, were seeded into 96-well plates at a density of 5 X 104 cells/well. The following day, 5 μM

of the substrate, coelenterazine H (Sigma), was added to each well to allow catalytic degradation

by Rluc and subsequent light emission. Luminescence and fluorescence were measured

separately at 480 nm and 535 nm, which corresponds to the Rluc and GFP maxima of the

emission spectra, respectively, on a Victor3 microplate reader (Perkin Elmer) equipped with

filters of the appropriate bandpass (Chroma, Rockingham, VT). The BRET ratio was defined as

[(emission at 510-590 nm) / (emission 440-500)] – Cf where Cf corresponds to (emission at

510–590)/(emission at 440–500) for the Rluc construct expressed alone in the same experiment.

2.2.11 cAMP accumulation - Basal and agonist-induced levels of cAMP (in the presence of 100

μM IBMX) were measured from COS7 cells transfected with D1DR. Cells were lysed in 0.1 N

HCl for 20 min and the supernatant was assayed for cAMP accumulation using an enzyme-

linked immunoassay kit (Cayman Chemical, Ann Arbor, MI) according to the manufacturer’s

instructions.

84 2.2.12 GTPγS Binding Assays - To quantify GTPγS binding to Gsα, membrane preparations

(100 μg/40 μl) of D1DR transfected COS7 cells were incubated with a reaction mixture containing 2 nM [35S]GTPγS,10 μM GDP, in the absence (basal) or presence of 10 μM

SKF81297. The reaction was incubated at 30°C for 1 min and terminated with 1 ml ice cold assay buffer (10 mM HEPES pH 7.4, 100 mM NaCl, 10 mM MgCl2, protease inhibitors). The samples were centrifuged at 20,000 g for 6 min at 4 °C. The resulting pellets were solubilized in

100 μl ice cold solubilization buffer (100 mM Tris pH 7.4, 200 mM NaCl, 1 mM EDTA, 1.25%

Igepal CA630, protease inhibitors) and 0.18% SDS for 1 hr at 4 °C. Unsolubilized cellular debris was pelleted at 20,000 g for 20 min. The supernatant was subsequently incubated with 5 μg of

Gsα antibody O/N at 4°C, followed by addition of 70 μl protein G agarose (Sigma) and mixing on a rotator for 4 hrs. Agarose beads were spun down at 2,500 rpm for 3 min and washed 4 X with solubilization buffer. Beads were suspended in scintillation fluid and radioactivity was determined.

2.2.13 Statistical analysis - All pharmacological data were analyzed using Prism (GraphPad

Software, San Diego, CA). Saturation binding curves were generated using non-linear least squares regression curve fitting. Statistical analysis between group means was performed using one way ANOVA followed by Tukey’s post hoc test.

3 ASSEMBLY OF D1 DOPAMINE RECEPTOR OLIGOMERS

3.1 Introduction

In this study, the cellular mechanism of D1 dopamine receptor oligomeric assembly was investigated. Although our laboratory was the first to show that D1 dopamine receptors form dimers and higher order oligomeric species, it was unknown whether these receptors were formed in the endoplasmic reticulum or at the plasma membrane. Previous studies had shown that misfolded receptors could intracellularly sequester the cognate wild-type receptor, thus rendering the latter non-functional (Coge et al., 1999; Grosse et al., 1997; Zhu and Wess, 1998).

This was interpreted as evidence for the constitutive assembly of GPCR oligomers early during biosynthesis. Similarly, we demonstrated that specific D2 dopamine receptor mutants could exert a dominant negative effect on the cell surface expression of co-expressed wild-type D2 receptors (Lee et al., 2000a). These receptor mutants were not misfolded, however, as indicated by their ability to traffic normally to the plasma membrane. We investigated the homooligomeric assembly of D1 receptors using strategically positioned point mutations within the D1 receptor to yield receptor mutants that could be used as tools to study the assembly of D1 receptor oligomers.

3.2 Detection of D1 Dopamine Receptor Oligomers in HEK293t and COS7 cells

We had previously shown that D1 receptors could form dimers in SF9 cells under SDS

(sodium dodecyl sulfate)-denaturing conditions (George et al., 1998). These receptor dimers were insensitive to co-incubation with a peptide derived from TM6 indicating that this region

86 was not critical for dimer formation. We used mammalian cell lines (HEK293t and COS7 cells) for receptor expression that have a good complement of signalling and trafficking molecules required for GPCR function. In HEK293t cells, the D1 dopamine receptor migrated on SDS-

PAGE as 64 kDa and 130 kDa – sized bands, corresponding to the monomeric and dimeric species, respectively (Figure 3-1). In contrast, in COS7 cells, the D1DR migrated as 52 kDa and

105 kDa – sized bands; the differences in size likely reflects variations in glycosylation between the two cell lines. The monomeric species observed is likely a result of the denaturing conditions employed by SDS solubilization and was also observed previously in SF9 cells. Sodium dodecyl sulfate has previously been shown to dissociate oligomeric subunits of GPCRs, including the D2 dopamine receptor (Lee et al., 2003a) and 5HT1B and 5HT1D receptors (Xie et al., 1999) into monomeric subunits. This effect has also been reported for non-GPCR multimeric membrane proteins (Casey and Reithmeier, 1993). It was demonstrated that utilization of a non-denaturing gel electrophoresis method employing perfluoro-octanoic acid could preserve D2 dopamine receptor oligomeric species and prevent its dissociation into monomeric subunits (Lee et al.,

2003a).

3.3 Role of Glycosylation in D1 Receptor Oligomeric Assembly

Asparagine-linked glycosylation is a post-translational modification that is required for maturation and, in some instances, proper functioning of receptor proteins. For the D1 and D5 dopamine receptors, glycosylation has distinct effects on cell surface trafficking. The mutation of three putative glycosylation sites in the D5 receptor was shown to reduce plasma membrane localization whereas mutation of two putative glycosylation sites in the D1 receptor had no effect

(Karpa et al., 1999). We were interested in determining whether glycosylation altered the ability

87 AB

MW (kDa) MW (kDa) HA-D1 (HEK293t) HA-D1 (COS7) 184 dimer 121 98 86 64

60 50 monomer

50 36

Figure 3-1. Immunoblot of membrane preparations from HEK293t (A) and COS7 (B) cells expressing wild-type D1DR (HA-D1). Membranes were prepared 48 h post-transfection and 25 μg of protein sample was used. Receptors were immunodetected with the anti- HA-HRP antibody. Arrows indicate the monomeric and dimeric species of HA-D1.

88 of D1 receptors to form oligomers. To address this, we transiently transfected the glycosylation deficient D1 receptor mutant (N5A/N175A) in HEK293t cells and subjected crude membrane preparations to SDS-PAGE. The lack of glycosylation did not appear to have any effect on receptor membrane expression nor did it change the proportion of dimers observed when compared to the wild-type dopamine receptor (Figure 3-2). This suggests that glycosylation is not directly involved in D1 dopamine receptor oligomeric assembly.

3.4 Generation of D1 Receptors with Key Structural Mutations

We had previously shown that mutation of the aspartic acid residue (D114) in TM3 of the

D2 dopamine receptor to asparagine, yielded a binding deficient but trafficking competent mutant receptor (Lee et al., 2000a). Co-expression of the D114N mutant with the wild-type receptor was sufficient to inhibit cell surface trafficking of the latter. This amino acid is a highly conserved residue in catecholaminergic receptors and in the D1 dopamine receptor, serves to interact with the protonated amine moiety of dopamine (Strader et al., 1989b). In addition, numerous studies have shown the equivalent residue to be integral to the activation mechanism of certain GPCRs (Befort et al., 1999; Porter et al., 1996; Robinson et al., 1992). Thus, we wished to investigate whether such a key mutation in catecholamine receptors could also generate a D1 dopamine mutant receptor with a similar dominant negative effect as the D114N mutation, and if so, what the mechanism of action might be. Thus, we targeted the aspartic acid residue in TM3 (D103) of the D1 receptor for site-directed mutagenesis. In addition to this, we also constructed additional mutations at residues that were predicted to share similar roles in receptor function. First, we targeted three serines (S198, S199, and S202) in TM5 which also have a crucial role in ligand interactions. Agonist binding to the D1 receptor requires the co-

89 MW (kDa) HA-D1 HA-N5A/N175A

184 121 } dimer 86

60 monomer 50 }

Figure 3-2. Immunoblot of membrane preparations from HEK293t cells expressing wild-type D1DR or glycosylation-deficient D1DR (N5A/N175A). Membranes were prepared 48 h post-transfection and 25 μg of protein sample was used. Receptors were immunodetected with the anti-HA-HRP antibody. Arrowheads and arrows indicate the monomeric and dimeric species of HA-D1 and N5A/N175A, respectively.

90 ordinated interaction of these three serines in TM5 with the two hydroxyl moieties in dopamine

(Pollock et al., 1992). Second, we mutated the aspartic acid residue (D120) in the highly

conserved D/ERY motif at the junction of TM3 and IC2 since this amino acid has also been

implicated in the activation mechanism of many GPCRs (Rovati et al., 2006).

3.4.1 Pharmacological and Expression Analysis of D1 Receptors with Transmembrane

Domain 3 Mutations (D103A, D103E, D120A, D120N)

Through PCR-based site-directed mutagenesis methods, we first mutated the D103

residue to alanine and performed saturation binding analysis on membrane preparations from

cells expressing this receptor mutant. We did not detect any measurable binding to the D1-like

receptor antagonist, [3H] SCH23390, compared to the wild-type D1 receptor (WT) (Figure 3-3).

To determine whether this was due to a reduction in receptor expression, we added a cMyc

epitope to the receptor and performed immunoblot analysis on membranes from cells expressing

these modified receptors. The cMyc-D103A receptor expressed robustly as dimeric and

dissociated monomeric subunits comparable to the cMyc tagged wild-type receptor (Figure 3-3).

This indicates that the lack of binding of the D103A receptor was due to a disruption in the ligand binding pocket although its ability to oligomerize was not affected. To determine whether the binding properties of this mutant could be restored by substitution with a similar negatively charged residue, we generated a mutant in which the aspartic acid was replaced with glutamic acid. Surprisingly, this mutant (D103E) bound to the antagonist as poorly as the D103A receptor

despite the fact that it too, expressed normally at the membrane (Figure 3-3). This indicates that

the aspartic acid residue has a very precise role in the ligand binding pocket.

91 A 5

1 H]SCH23390 specific binding (pmol/mg membrane protein) membrane (pmol/mg 3 [

0 0.0 0.5 1.0 1.5 2.0 2.5 [3H-SCH23390] (pM)

D1DR D103A D103E

BMAX 5.099 0.2158 0.1430

KD 0.5466 n/a n/a

MW (kDa) cMyc-D1cMyc-D103AcMyc-D103EcMyc-S198A/S199A

148 - dimer

monomer 64 -

92 To characterize the D120A and D120N mutant receptors, we performed binding with 1

nM [3H] SCH 23390 and found that membranes expressing these receptors did not bind

antagonist (data not shown). To evaluate whether this was due to a lack of expression, we

performed immunoblot analyses on these mutants and found that neither receptor exhibited any

ER or plasma membrane expression. This suggests that these mutations render misfolded

proteins that get targeted for proteasomal degradation (Figure 3-4).

3.4.2 Pharmacological and Expression Analysis of D1 Receptors with Transmembrane

Domain 5 Mutations (S198A/S199A, S199A/S202A)

In order to generate an adequate control for testing the dominant negative effects of

D103A on WT expression, we generated two additional mutants that were predicted to be

binding deficient as well. Previous studies have shown that the serine at position 198 is the most

critical in recognizing dopaminergic ligands that are structurally similar and dissimilar to

dopamine (Pollock et al., 1992). The serine at position 199 is required to maintain high affinity

binding and the serine at position 202 does not appear to be involved in binding to these

dopaminergic ligands but is important in binding to dopamine, itself. To this effect, we

constructed a mutant in which both serines at positions 198 and 199 were substituted with

alanine, S198A/S199A. One point binding analysis with 1 nM [3H] SCH 23390 indicated that

membranes expressing this receptor exhibited negligible binding despite its robust membrane

expression as dimeric and dissociated monomeric subunits (Figure 3-3, 3-5). In contrast, a

receptor with serine mutations at positions 199 and 202 maximally bound the same amount of

radioligand than wild-type receptor and was also expressed to a similar extent (Figure 3-5).

Binding levels were normalized against WT membrane expression (data not shown). This is

93 D120N D1 D120A

148

Figure 3-4. Immunoblot of membrane preparations from HEK293t cells expressing wild-type D1DR, D120A, or D120N. Membranes were prepared 48 h post-transfection and 25 μg of protein sample was used. Receptors were immunodetected with the AL25 antibody.

94 1.5

1.0

0.5 H]SCH23390 specific binding (pmol/mg membrane protein) membrane (pmol/mg 3 [ 0.0 R A A D 2 9 1 D 20 19 /S /S 9A 8A 19 19 S S

Figure 3-5. Observed binding of S199A/S202A from HEK293t membranes to 1 nM [3H] SCH23390. Results shown are the mean of 3 independent experiments.

95 consistent with the aforementioned report indicating that S198 is indispensable for ligand

binding.

3.4.3 Adenylyl Cyclase Activity of D103A, D103E, and S198A/S199A Mutant Receptors

Since a number of reports have shown that the analogous TM3 aspartic acid is involved

in the activation mechanism of various GPCRs (Befort et al., 1999; Porter et al., 1996; Robinson

et al., 1992), we hypothesized that mutation of this residue might confer an increase in the basal

level of adenylyl cyclase activity. One of the hallmarks of a constitutively activated Gs-coupled

GPCR is an increase in basal endogenous cAMP levels resulting from the enhanced catalytic conversion of ATP to cAMP by adenylyl cyclase. Using an ELISA-based method of cAMP determination, we assessed cAMP levels of the D103A, D103E, and S198A/S199A receptors expressed in HEK293t cells. The D103A and the D103E receptors had approximately 24 ± 8.2 % and 21.5 ± 2.2 % higher basal cAMP levels, respectively, than unstimulated WT (Figure 3-6) after normalizing to wild-type cell surface expression by densitometric analysis of biotinylated cell surface receptors (Figure 3-6, inset). This represented approximately 45 % of the cAMP response of wild-type receptor when maximally stimulated with 10 μM dopamine (51.3 ± 18.3 % above basal), indicating that this specific TM3 mutation had resulted in a receptor that was in a partially activated but not a fully activated conformation. In addition, no change in constitutive cAMP production was observed upon incubation of D103A with agonists (dopamine, pergolide,

SKF 38393, SKF 81297) or inverse agonists (fluphenazine, flupenthixol, (+) butaclamol) at 10

μM concentration, indicating that removal of the aspartic acid residue also abolished binding to other dopaminergic ligands (data not shown). This is consistent with confocal microscopy studies showing that pergolide, a mixed D1/D2 agonist, could not induce any detectable

96 untrans- wt S198A/ D103E D103A fected D1 S199A

- 50 kDa

200 * 175 γ γ 150 * * basal wt D1 125 of 100

0 % change in cAMP S198A/ wt D1 untransfected D1 + DA D103E D103A S199A

Figure 3-6. Constitutive Activity of D1DR and indicated mutant receptors. Basal adenylyl cyclase activity was determined by quantifying cAMP accumulation and reported as a % above basal D1DR cAMP levels (wt D1). The inset compares the cell surface expression of biotinylated receptors. All values are reported after normalization of receptor expression to wild-type D1DR by densitometric analysis. Data sets were normalized against the basal cAMP response in wild-type D1DR transfected cells and expressed as a % change of this. Significance at p<0.05 versus basal cAMP levels of cells expressing wild-type D1DR is denoted by * . Significance at p<0.05 versus dopamine stimulated D1 (D1+DA) cAMP levels is denoted with γ.

97 internalization of D103A and indicates that although it expresses normally, it does not functionally respond to agonist stimulation (Figure 3-7). The S198A/S199A receptor did not exhibit a change in basal activity that was different from wild-type D1 receptors. This data demonstrated that the aspartic acid residue in TM3 participated in keeping the D1 receptor in an inactive conformation.

3.5 Antagonist Rescue of D120A and D120N

Given that the D120A and D120N mutant receptors did not express to any appreciable degree, we hypothesized that these receptors were misfolded and subsequently degraded.

Previous studies have implicated the conformational stabilization of such misfolded receptors by so-called pharmacological chaperones (Chaipatikul et al., 2003; Janovick et al., 2003; Morello et al., 2000; Petaja-Repo et al., 2002). These compounds are typically receptor specific hydrophobic ligands that act intracellularly to promote proper folding of the receptor. To evaluate this effect, cells transfected with either D120A or D120N were incubated with the D1 receptor antagonist, SCH23390, and receptor expression from membrane fractions were assessed by immunoblot analysis. Antagonist treatment for 24 h restored the membrane expression of the immature, non-glycosylated forms of both the D120A and D120N receptors (54 kDa) (Figure 3-

8; lanes 4,5).

3.6 Co-expression of D103A and D103E with Wild-type D1DR

When the wild-type D1 receptor was co-expressed with increasing amounts of cDNA encoding D103A, there was a dose-dependent reduction of [3H]-SCH 23390 binding to cell membranes (Figure 3-9). When expressed singly, the wild-type receptor density was determined

98 No treatment 10 μM Pergolide

HA-D1

10 μm

cMyc-D103A

Figure 3-7. Immunofluorescence imaging of HA-tagged D1DR visualized by FITC (green) and cMyc-tagged D103A visualized by TRITC (red) showing basal cell surface localization and agonist-induced internalization of D1DR but no effect on the mutant D103A receptor. HEK293t cells expressing either HA-D1 (top row) or cMyc-D103A (bottom row) receptors were permeabilized and exposed to 10 μM pergolide for 20 min.

99 24 h SCH23390 treatment

D1DR D120A D120N D120A D120N

MW (kDa)

148

Figure 3-8. Restored membrane expression of D120A and D120N receptor mutants. HEK293t cells expressing indicated receptors were incubated with 10 μM SCH23390 for 24 h before immunblot analysis from membrane fractions. The arrowead indicates the mature, glycosylated wild-type D1DR species. The arrow indicates the immature, unglycosylated D120A and D120N receptor mutants.

100 to be 6.3 ± 0.1 pmol/mg. There was a 29% drop in the receptor density at a 1:0.5 DNA

transfection ratio of wild-type:D103A, and a 48% reduction at a 1:1 ratio, yielding wild-type expression levels of 4.5 ± 0.3 pmol/mg and 3.3 ± 0.2 pmol/mg respectively. Receptor density decreased by approximately 63% to 2.3 ± 0.1 pmol/mg when maximal reduction was reached at a 1:1.5 transfection ratio. Further increments of D103A expression did not alter wild-type receptor density. Co-expression of wild-type D1 with the D103E receptor yielded similar results to that of the D103A receptor, demonstrating an inverse relationship between constitutively active receptor expression and wild-type expression (Figure 3-9).

To evaluate whether these changes in D1DR expression were reflective of changes in cell surface receptor expression, three additional strategies were employed: 1) whole cell binding to dopamine 2) biotinylation of cell surface receptors; and 3) intact cell fluorometric analysis. For whole cell binding experiments, we sought to use a cell impermeable ligand to quantify cell surface expression. Whereas SCH23390 is a cell permeable antagonist that can label surface and intracellular sites, dopamine is a highly polar compound rendering it highly hydrophilic and unable to cross the lipid bilayer (Barbier et al., 1997). Hence, cell surface receptor expression was estimated by determining the difference between specific [3H]-SCH 23390 binding and the

binding displaced by an excess of dopamine as described in Materials and Methods. We

determined that 100 μM dopamine was sufficient to completely displace 1 nM [3H]- SCH 23390

from both high and low affinity sites of the wild-type D1 receptor in our competition binding

studies from membranes (Figure 3-10) and thus used these conditions for the whole cell binding

assays. Non-specific binding was defined by [3H]-SCH 23390 binding in the presence of 1 μM

(+) butaclamol. We found that there was a 7.2 ± 4.5 % decrease in wild-type cell surface

expression at a wild-type:D103A transfection ratio of 1:0.5 compared to expression at a 1:0 ratio

101 A

7.5

5.0 * * * 2.5 * H] SCH 23390 specific binding specific 23390 H] SCH (pmol/mg membrane protein) 3 [ 0.0 1:0 1:0.5 1:1 1:1.5 1:2

D1 : D103A transfection ratio

B 7.5

5.0 α α α 2.5 α H] SCH 23390 specific binding specific 23390 H] SCH (pmol/mg membrane protein) 3 [ 0.0 1:0 1:0.5 1:1 1:1.5 1:2

D1 : D103E transfection ratio

Figure 3-9. Wild-type D1DR density determined by binding of membranes to 1 nM [3H] SCH23390 from HEK293t cells co-transfected with increasing concentration of D103A (A) or D103E (B). Wild-type D1DR density was maintained constant (cDNA, 5 mg) while D103A cDNA was co-transfected in varying amounts yielding transfection ratios of 1:0, 1:0.5, 1:1, 1:1.5, and 1:2, respectively. Wild-type D1DR binding in the absence of D103A was 6.3 ± 0.9 pmol/mg (1:0) and gradually decreased with increasing expression of D103A or D103E. The data shown are the means ± S.E. of 3-5 experiments. Significance at p<0.05 versus control D1DR binding (at 1:0 transfection ratio) is denoted by * for D103A and α for D103E.

102 A

125

100

75 H] SCH 23390 bound 23390 SCH H]

3 50

25 % specific [ 0 -10 -9 -8 -7 -6 -5 -4 -3 -2 log [dopamine] (M)

100

25 % change dopamine displaced change % H] SCH 23390 specific binding specific 23390 SCH H] 3

[ 0 1:0 1:0.5 1:1 1:1.5 1:2

D1 : D103A transfection ratio

Figure 3-10. (A) Competition of various concentrations of dopamine for 1 nM [3H] SCH23390 binding to wild-type D1DR from membrane preparations. (B) Whole cell binding of dopamine displaced [3H] SCH23390 to wild-type D1DR in the presence of varying expression levels of D103A. Wild-type:D103A transfection ratios of 1:0.5, 1:1, 1:1.5, and 1:2 yielded a 7.2 ± 4.5, 48.7 ± 4.1, 57.9 ± 4.5, and 66.8 ± 0.9 % reduction in wild-type D1DR cell surface density, respectively. The data shown are presented as the means ± S.E. of 3 experiments.

103 (Figure 3-10). Further increases in D103A expression yielded 48.7 ± 4.1 %, 57.9 ± 4.5 %, and

66.8 ± 0.9 % reductions in wild-type cell surface receptor density at 1:1, 1:1.5, and 1:2 transfection ratios, respectively. These results are comparable to the data obtained through saturation binding of membrane preparations (Figure 3-9).

Cell surface expression of D103A was also assessed by determining the proportion of cell

surface biotinylated D103A receptors in response to increasing wild-type receptor expression in

COS7 cells. This cell line is more adherent than HEK 293t cells rendering it more suitable for the

cell surface biotinylation experiments. Similar to what was found through the radioligand

binding studies, the proportion of biotinylated D1DR decreased in the presence of increasing

levels of D103A expression (Figure 3-11). The converse experiment revealed a similar pattern of

reduced biotinylated D103A expression with increasing levels of D1DR expression (Fig. 3-11) consistent with our whole cell binding data.

Finally, we performed cell surface fluorescence analysis on non-permeabilized whole cells co-expressing amino terminus tagged HA-D1DR and cMyc-D103A. Wild-type receptor cell surface labeling in the presence of D103A (1:1 expression ratio) was reduced to 58.3 ± 8.3 % of wild-type levels when expressed alone (Figure 3-12). This is consistent with the data obtained through the methods described above suggesting that the wild-type/D103A oligomer is hindered from trafficking to the cell surface.

3.7 Co-expression of S198A/S199A and the Apelin Receptor with Wild-type D1DR

The co-expression of D103A with D1DR was observed to result in a reduction in D1DR cell surface expression, presumably by oligomerization and subsequent intracellular retention. To determine whether oligomerization with another mutant receptor could also inhibit expression of

104 HA-D1 : cMyc- D103A cMyc- D103A : HA-D1

1:0 1:1 1:2 1:0 1:1 1:2

MW (kDa) MW (kDa)

50 - 50 -

lane: 1 2 3 4 5 6 IP: HA IP: cMyc

Figure 3-11. Cell surface biotinylation of HA-D1DR in the presence of cMyc-D103A and vice versa. Biotinylated cell surface HA-D1DR receptor expression was determined in the presence of increasing expression levels cMyc-D103A (lanes 1-3). The converse experiment was performed with cMyc-D103A (lanes 4-6).

105 100 * 75

50 % change in AFU 25

1:0 1:1

D1 : D103A transfection ratio

Figure 3-12. Cells co-expressing cMyc-D103A and HA-D1DR (1:1 ratio) were incubated with FITC-conjugated anti-HA and cell surface D1DR expression was detected by whole-cell fluorometry. Co-expression with cMyc-D103A (1:1) resulted in a decrease in HA-D1DR expression (as measured in arbitrary fluorescence units, AFU) to 58.3 ± 8.3 % of individually expressed D1DR (1:0). Significance at p<0.05 versus control D1DR binding (at 1:0 transfection ratio) is denoted by *.

106 the wild-type receptor, we co-expressed the ligand non-binding D1DR mutant, S198A/S199A,

with WT in a 1:1 ratio and performed whole cell analysis of dopamine displaced [3H]-SCH

23390 binding as described above. As previously indicated, this mutant did not exhibit enhanced constitutive activity (Figure 3-7) and expressed comparably to D1DR as multimeric species when transfected alone (Figure 3-4). There was negligible inhibition of cell surface expression of

WT when co-expressed with S198A/S199A (Figure 3-13), indicating that a) oligomerization of

D1DR with this mutant receptor could not sequester the wild-type receptor and b) the translational efficiency of the wild-type receptor was not compromised by the co-translation of the mutated D103A receptor. To further validate this latter point, we also co-transfected a non- related peptidergic receptor, the apelin receptor, that was not predicted to oligomerize with

D1DR. Indeed, similar results were obtained with S198A/S199A indicating that co-transfection with a second receptor does not compromise the cellular biosynthetic machinery (Figure 3-13).

3.8 Interaction Between D103A and Wild-type D1DR

To confirm whether D103A existed within an oligomeric complex with the wild-type receptor, cMyc-D103A and HA-D1DR were co-expressed in a 1:1 ratio and the total HEK293t cell lysates were immunoprecipitated with either antibodies directed to cMyc or HA.

Immunoprecipitates were subsequently analyzed by immunoblot for HA immunoreactivity. As shown in Figure 3-14, cMyc-D103A and HA-D1 were co-immunoprecipitated with the other, each visualized as a 70 kDa species, indicating that the receptors were within an oligomeric complex containing the D103A and wild-type receptors.

107 125

100

25 % change dopamine displaced change % H] SCH 23390 specific binding specific 23390 SCH H] 3 [ 0 D1 / pcDNA3 D1 / S198A/S199A D1 / apelin receptor

Figure 3-13. Cell surface D1DR expression is not compromised by co- expression with S198A/S199A or the apelin receptor. Whole cell binding of cells co-transfecting wild-type and S198A/S199A or the apelin receptor in a 1:1 ratio yielded cell surface D1DR expression levels that were reduced by 9 ± 7 and 12 ± 8 %, respectively, compared with untreated cells.

108 MW IP: cMyc HA mock (kDa) IB: HA HA HA 98 -

64 -

50 -

36 - IgG

Figure 3-14. HEK293t cells were co-transfected with cMyc-D103A and HA-D1DR receptors for 48 h, and whole cell lysates were subjected to co- immunoprecipitation. Immunoprecipitation with either anti-cMyc or anti-HA precipitated HA-D1DR, detected at 70 kDa. Mock co-immunoprecipitations were done on transfected lysates in the absence of immunoprecipitating antibody.

109 3.9 Staggered Co-expression of D103A with Wild-type D1DR

To further test the hypothesis that the inhibition of WT surface expression was dependent

on its close association with the D103A receptor occurring during translation, expression of both

receptors were staggered by transfecting them 24 h apart. We have independently confirmed that

85-90% of D1 receptor protein is synthesized 24 hours post-transfection, thus achieving near

maximal expression at this time point (data not shown). Wild-type receptor co-transfected with

D103A in a 1:1 ratio resulted in a significant drop in Bmax by 35.3 ± 3.2% (Figure 3-15).

However, no significant decrease in Bmax was observed when wild-type receptor was expressed

independently compared to when it was co-expressed with D103A in a 1:1 ratio, staggered 24

hours apart (reduction in Bmax by 1.4 ± 4.4%). This suggests that the two receptors trafficked

independently when biosynthesis of both receptors was not co-translational and that the

dominant negative effect of the D103A occurred only under co-translational conditions.

3.10 Agonist-specific Restoration of Wild-type/D103A Oligomer

Despite the ability of the WT and D103A receptors to traffic as oligomers independently,

we hypothesized that the wild-type/D103A receptor oligomer was sequestered by the quality

control machinery in the cell due to an oligomeric configuration that was recognized by the cell as incompatible with cell surface trafficking. We predicted that various dopaminergic ligands

might act as pharmacological chaperones and stabilize the sequestered oligomer into a

conformation permissive for trafficking. Thus, we evaluated the potential of a series of cell-

permeable dopaminergic agonists, antagonists, and inverse agonists to reverse the cell surface

inhibition of the wild-type receptor imposed by oligomerization to the D103A mutant. Wild-type

and D103A receptors were co-transfected in a 1:1 ratio and cells expressing these receptors were

110 125

100

75 *

50 D1 D1 + D103A (1:1) 25

(% of D1 expressed alone) expressed D1 of (% D1 + D103A (1:1) H] SCH 23390 specific binding specific 23390 SCH H]

3 (staggered) [

0 0 1000 2000 3000 4000 5000 [3H-SCH 23390] pM

Figure 3-15. Saturation isotherm of wild-type D1DR co-expressed with D103A in a 1:0 (solid line) and 1:1 (dashed line) ratio. The Bmax dropped by 35.3 ± 3.2 % at a 1:1 transfection ratio. The total amount of cDNA transfected was maintained constant by addition of pcDNA3 vector. The data shown are expressed as a percentage of D1DR binding (when expressed alone) and presented as the mean ± S.E. of 3 experiments. Signficance at p < 0.05 versus Bmax of cells co-expressing D1DR and D103A is denoted by *.

111 incubated for 2 hours with an array of cell permeable D1 ligands (pergolide, SKF 81297, SKF

38393, apomorphine, SCH 23390, (+)-butaclamol, fluphenazine, and flupenthixol) and the non- permeable agonist, dopamine, and washed thoroughly before membrane receptor density was evaluated. The cell permeable compounds indicated have previously been reported to cross the plasma membrane (Barbier et al., 1997). over the time period tested. As shown in Figure 3-16, the full D1 agonists, pergolide and SKF 81297, induced 66.2 ± 26.7 % and 71.7 ± 19.4 % increases, respectively, in the amount of WT expression detected. The partial agonist, SKF

38393, induced a much greater increase in WT expression by approximately 165 ± 32.5 %.

Apomorphine and dopamine exerted little change on WT density. In contrast, the D1DR antagonist, SCH 23390, had no significant effect on WT expression. A similar non-significant change in WT density was also observed for the inverse agonist (+)-butaclamol, and little change in WT expression was found with two other inverse agonists, fluphenazine and flupenthixol

(Figure 3-16A). Thus, the ability to rescue the intracellularly retained wild-type D1 receptors was specific only to the ligands which were cell permeable agonists.

3.11 Cell Surface Analysis of Pharmacologically Rescued WT/D103A Oligomer

To determine whether this ligand induced rescue of receptors was reflective of wild-type

D1 expression at the cell surface, whole cell binding was performed on cells co-expressing WT and D103A. As shown in Figure 3-16B, pre-treatment of cells with pergolide caused a significant

46.4 ± 6.5 % increase in wild-type receptor density. In contrast, pre-incubation with SCH 23390 had only a negligible effect on cell surface expression of the wild-type D1 receptor.

The cellular distribution of co-expressed (in a 1:1 ratio) HA-D1 and cMyc-D103A receptors were visualized by immunofluorescence microscopy. In non-permeabilized cells, both

112 cell cell cell cell permeable impermeable permeable permeable agonists agonist inverse agonists antagonist A

200 * 175 150 125 100 ** 75 50

binding 25 0 -25 -50 Control Perg SKF SKF Apo DA Flupe Fluph BTC SCH

% change% specific D1 wild-type of -75 81297 38393 23390 -100

B 60 *

H] SCH 23390 specific binding specific 23390 SCH H] 0 3 % change dopamine displaced dopamine change % [ -10 Control Pergolide SCH 23390

Figure 3-16. Agonist-specific rescue of the intracellularly sequestered WT/D103A oligomer. (A) HEK293t cells co-transfected with WT and D103A in equal amounts were treated with 10 μM of the dopaminergic ligands indicated for 2 h at 37 °C. The ligands used were pergolide (Perg), SKF 81297, SKF 38393, apomorphine (Apo), dopamine (DA), flupenthixol (Flupe), fluphenazine (Fluph), (+)-butaclamol (BTC), and SCH23390. Plasma membrane fractions were prepared and D1DR density was assessed by saturation binding with 1 nM [3H] SCH23390. Data sets were normalized against D1 binding in cells co-expressing WT and D103A (control) and expressed as a % change of this. Pergolide, SKF 81297, and SKF 38393 induced a 66.2 ± 26.7, 71.7 ± 19.4, and 165 ± 32.5 % increase in WT receptor density, respectively. Results are presented as the means ± S.E. of 3-5 independent experiments. Significance at p<0.05 versus control binding is denoted with *. (B) Co-transfected HEK293t cells were pre-treated with the indicated ligands for 2 h, and whole-cell dopamine displacement of [3H] SCH23390 binding was performed as described under Materials and Methods. Data sets were normalized against D1 binding in cells co-expressing D1 and D103A (control) and expressed as a % change of this. Pergolide treatment caused a 46.4 ± 6.5 % increase in WT cell surface density. Pre-treatment with SCH23390 had no significant effect on WT113 density. Results are presented as the mean ± S.E. of 3 independent experiments. Significance at p<0.05 versus control is denoted with *. receptors were localized at the cell surface under basal conditions (Figure 3-17, row 1). In

permeabilized cells, both receptors were observed to have pronounced intracellular localization within the cell, likely within the ER and/or Golgi (Figure 3-17, row 2). This distribution was in contrast to the cell surface distribution that was observed when each of these receptors were expressed individually (Figure 3-7) and confirms that intracellular retention of both receptors occurred. Pergolide treatment caused a change in the intracellular distribution of both receptors

to a predominantly cell surface distribution (Figure 3-17, row 3). This is consistent with the data

obtained through quantitative ligand binding. Treatment with SCH 23390 did not appear to alter

the intracellular distribution of receptors (Figure 3-17 row 4); this was confirmed by the absence of any changes in cell surface expression as demonstrated by the whole cell binding experiments

(Figure 3-16B). Incubation with dopamine also had no effect on the cellular distribution of either receptor (data not shown). To identify the intracellular site of WT/D103A oligomeric retention, we used antibodies directed to both the wild-type D1 receptor and the ER-resident transmembrane protein, calnexin, in permeabilized cells co-expressing wild-type receptor and

D103A receptor. As shown in Figure 3-18 (row 1), under basal conditions, WT and calnexin exhibited extensive co-localization in the ER in the presence of D103A. In contrast, pre- treatment of cells with 10 μM pergolide, triggered a clear release of WT from the ER compartment, as evidenced by a lack of co-localization with calnexin, and a subsequent redistribution to the cell surface (Figure 3-18, row 2). Wild-type D1 receptors expressed in the absence of D103A exhibited a robust cell surface distribution without co-localization with calnexin (data not shown). A random selection of cell images were quantified in Table 3-1 by determining Mander’s colocalization coefficients for each channel (as described in Materials and

Methods). A significant portion (Mgreen > 0.75) of WT (green) was found co-localized with

114 HA-D1 cMyc-D103A merge (FITC) (TRITC)

No treatment (non-permeabilized) 1010 μM

No treatment (permeabilized)

1010 μM

10 μM Pergolide (permeabilized)

10 μM SCH 23390 (permeabilized)

Figure 3-17. The subcellular distribution of wild-type HA-D1 and cMyc-D103A coexpressed in a 1:1 ratio was revealed by fluorescence microscopy in HEK293t cells. Immunostaining with anti-HA and anti-cMyc antibodies visualized by FITC- (green) and TRITC- (red) conjugated antibodies. respectively, was performed under non- permeabilized conditions (row 1) and permeabilized conditions (rows 2-4). Ligand treatments with 10 μM and 10 μM SCH23390 werre performed for 2 h at 37 °C before fixing.

115 HA-D1 calnexin (FITC) (Alexa-546) merge

WT + D103A No treatment (permeabilized)

WT + D103A 10 μM Pergolide (permeabilized)

Figure 3-18. HEK293t cells were co-transfected with HA-D1 and cMyc- D103A were immunostained with anti-HA and anti-calnexin antibodies visualized by FITC (green) and Alexa-546 (red), respectively, before (row 1) and after 10 μM pergolide treatment (row 2), as described previously.

116 D103A (red) under all conditions when the antigen pair, D1/D103A, was assessed (Table 3-1). In

contrast, colocalization of WT with calnexin was much reduced when WT/D103A expressing

cells were treated with pergolide as shown for the antigen pair, D1/calnexin, (Mgreen, 0.99 without pergolide versus Mgreen, 0.15 following pergolide treatment).

Although not shown, we also tested the possibility that differential phosphorylation of

WT and D103A by PKA might be the regulatory step in the ER quality control that is scrutinized

for oligomer trafficking. Pre-treatment of cells with the PKA inhibitor, H-89, did not affect the

extent of pergolide-induced WT/D103A rescue, suggesting a PKA-mediated phosphorylation-

independent mechanism of oligomer restoration to the cell surface.

3.11.1 FRET Analysis of Rescued Oligomers

In order to test the possibility that prolonged agonist treatment may have triggered

conformational changes in the sequestered WT that would cause it to dissociate from the

WT/D103A oligomer and allow independent trafficking to the surface, analysis of the

association between the rescued wild-type and D103A receptors was performed at the cell

surface. To evaluate restored trafficking of the WT/D103A oligomer to the plasma membrane,

cell surface time-resolved fluorescence resonance energy transfer (trFRET) between the two

receptors was assessed in the absence and presence of agonist. Time-resolved fluorescence

resonance energy transfer is a biophysical approach used to determine the proximity of proteins

to each other. Energy transfer between two proteins implicates an intermolecular distance of 50-

100 Å which is indicative of a direct protein-protein interaction. Energy measurements are “time-

resolved” because the readings are taken after a 50 μs delay to allow for the decay of endogenous

fluorescence signals. We used Eu3+ and APC labeled antibodies to label the corresponding

117 Mgreen/Mred - Mander’s coefficient for green or red channel (represents the number of co-localized pixels in the green or red channel expressed as a fraction of the total number of non-zero pixels in the respective channel.

Antigen pair M M (green/red) green red

No treatment 0.86 0.70

D1 / D103A 10 μM pergolide 0.94 0.96

10 μM SCH23390 0.77 0.53

No treatment 0.99 0.97

D1 / calnexin

10 μM pergolide 0.15 0.19

Table 3-1. Colocalization analysis of different antigen pairs in permeabilized cells co-expressing D1DR and D103A.

118 receptors that were specifically at the plasma membrane. Cells co-expressing HA-tagged wild-

type and cMyc-tagged D103A receptors were incubated for 2 hours in the presence of either

pergolide, SKF 38393, or SCH 23390. Intact cells were incubated with Eu3+-labeled anti-HA

(donor) and APC-labeled anti-cMyc (acceptor). trFRET was determined by monitoring light

emission at 665 nm from APC. Under basal conditions, no FRET was detected in whole cells co- expressing both receptors indicating that only non-interacting receptor species were present at

the cell surface (Figure 3-19). This suggests that the receptors at the cell surface under non-

permeabilized conditions do not represent WT/D103A hetero-oligomers (Fig 3-17, row 1) but

rather, separate WT and D103A receptor homo-oligomers. Treatment of cells co-expressing

both receptors with pergolide or SKF 38393 yielded a 9.1- and 8.4- fold increase in energy

transfer, respectively, indicating that the WT/D103A mixed oligomer was now present on the

cell surface. Pre-incubation of cells with SCH 23390 or mixing of cells independently expressing

wild-type or D103A receptors yielded a negligible FRET signal. Therefore, we conclude that the

rescued wild-type D1 receptor was part of a mixed oligomeric complex with D103A that

remained intact on the cell surface after agonist rescue.

3.12 Discussion

In this section, using specific conformationally distinct mutants of the D1 dopamine

receptor, it was shown that wild-type receptor cell surface expression could be inhibited in a

dominant negative manner. It was found that a receptor with a mutation of the highly conserved

aspartic acid in transmembrane domain 3, D103A, could exert this effect despite the fact that this

mutant, itself, was expressed normally as a constitutively active homo-oligomer at the plasma

membrane. Through confocal microscopy and biochemical methods, we determined that this

119 15 * pergolide * SKF 38393 10 SCH 23390 untreated mixed 5

fold-increae cell surface FRET 0

Figure 3-19. Cell surface time-resolved fluorescence resonance energy transfer after agonist treatment. HEK293t cells co-expressing HA-D1DR and cMyc-D103A (1:1 ratio) were incubated in the absence or presence of either 10 μM pergolide, SKF38393 or SCH23390 for 2 h. The ligand was washed off and cells were then counted and incubated with Eu3+-labeled anti- HA (donor) and APC-labeled anti-cMyc (acceptor) antibodies for 2 h at 37 °C and trFRET was determined by measuring APC emission after Eu3+ excitation. Separate pools of independently transfected cells were mixed to measure non-specific FRET. Significance at p<0.05 versus control trFRET in the absence of ligand (untreated), is denoted with *.

120 inhibitory effect was found to occur in the endoplasmic reticulum as a result of an oligomeric

interaction between the wild-type receptor and D103A. It was rationalized that this mutant did

not physically sequester the wild-type receptor as demonstrated for intracellularly retained

mutants of other GPCRs since the D103A mutant, expressed alone, was not misfolded and intracellularly sequestered but was in fact, constitutively activated. Hence, we hypothesized that this wild-type/D103A oligomer was simply not conformationally compatible for ER export

through the distal secretory pathway. However, we found that this mixed oligomer could be

selectively rescued to the plasma membrane by cell-permeable dopaminergic agonist incubation

but not by antagonists or inverse agonists. Collectively, this suggests that specific post-

translational checkpoints in the cell can be bypassed by pharmacological intervention and that

constitutively synthesized D1 dopamine receptor oligomers are subject to conformational

scrutiny by stringent quality control mechanisms in the ER.

The evidence that D1 dopamine receptor oligomers are synthesized in the endoplasmic

reticulum prior to cell surface localization is derived from the fact that both the D103A and

D103E mutant receptors could associate with and inhibit the trafficking of the wild-type

receptor. In the absence of any oligomeric assembly in the ER, one would expect the D103A

mutant to have no effect on the trafficking of the wild-type receptor, even when co-expressed, as

each receptor would behave as an independently processed protein. This was not observed,

however, and is suggestive of the fact that D1 dopamine receptors function as constitutive

oligomers. Our staggered expression experiments further suggest that constitutive oligomeric

assembly might be co-translational since the temporal dissociation of synthesis of D103A and

WT within the same cell did not impair WT cell surface expression (Figure 3-15). This is

consistent with other reports demonstrating that GPCR oligomers are formed in the ER and not

121 in response to agonist induction at the cell surface. Many of such reports utilize energy transfer

techniques as evidence in support of either oligomeric model. For example, through BRET

methods, oxytocin and vasopressin receptor oligomers have been shown to constitutively reside

in ER-enriched subcellular fractions (Terrillon et al., 2003). In contrast, gonadotropin- releasing

hormone receptors (Cornea et al., 2001) as well as somatostatin SSTR5 and D2 dopamine

receptors (Rocheville et al., 2000a) have been shown to form homo-oligomers and heterooligomers, respectively, as a result of agonist binding and activation. However, in the latter

model, it was determined that the energy transfer observed upon agonist activation was likely a

result of conformational changes within pre-formed oligomers and not due to increased oligomer

assembly, per se (Ayoub et al., 2002). This notion was strengthened by the fact that neutral

antagonists and inverse agonists could also increase energy transfer independent of receptor

activation. it seems likely that oligomerization is a requirement for the trafficking and proper

assembly of GPCRs. Indeed, hetero-oligomerization between receptor subtypes has been shown

to be a prerequisite for both of these cellular processes. An excellent example is the heteromeric

GABAB receptor which requires an interaction between the GABABR1 and GABABR2 receptors

for full functional activity (Jones et al., 1998; Kaupmann et al., 1998; White et al., 1998). The

individual expression of the GABABR1 subtype results in an intracellularly sequestered receptor

whereas expression of the GABABR2 receptor results in a mature but non-functional receptor.

Similarly, in Family A GPCRs, the α1D-adrenergic receptor requires an oligomeric interaction

with the α1B-adrenergic receptor for proper plasma membrane localization (Hague et al., 2004b).

Thus, it appears that GPCRs can act as trafficking chaperones by oligomerizing to selectively

retained GPCRs; these examples are summarized in Table 3-2. In addition to this, specific non-

GPCR proteins have been shown to be required for functional plasma membrane expression of a

122

Intracellularly Chaperone Partner Functional cell surface Reference sequestered GPCR localized heteromeric receptor α1D-adrenergic α1B-adrenergic α1D / α1B heterodimer Hague et al, receptor receptor 2004b M71 olfactory β2-adrenergic M71 / β2 heterodimer Hague et al, receptor receptor 2004a α1D-adrenergic β2-adrenergic α1D / β2 heterodimer Uberti et al, receptor receptor 2005 GABABR1 GABABR2 GABAB receptor Kaupmann et al, 1998; White et al, 1998; Jones et al, 1998 T1R1 T1R3 Amino acid umami Zhao et al, receptor 2003 T1R2 T1R3 Sweet taste receptor Nelson et al, 2001

Table 3-2. List of GPCRs that require interaction with a chaperone GPCR for functional cell surface expression.

123 number of GPCRs. For example, the calcitonin receptor like receptor (CRLR) requires an

interaction with the single transmembrane proteins, RAMP1 or RAMP2, to form mature and

functional CGRP and adrenomedullin receptors, respectively (McLatchie et al., 1998). The

functional expression of the calcium sensing receptor has also been shown to be dependent on its

interaction with RAMP1 and RAMP3 (Bouschet et al., 2005). Other examples of chaperones

required for GPCR trafficking include NinaA and ODR-4 for expression of rhodopsin and

olfactory receptors (in C.elegans), respectively (Dong et al., 2007).

Although constitutive hetero-oligomerization plays a role in cell surface trafficking, it is

still not apparent whether constitutive homo-oligomerization is also a prerequisite for ER export

and plasma membrane localization. Several studies have shown that certain dimerization- deficient GPCRs are trapped intracellularly (Overton et al., 2003; Salahpour et al., 2004)

implicating such a role; however, it is not completely clear whether this is simply due to

misfolding of the protein. Other studies using somatostatin receptors have implicated a role for

homo-oligomerization in signal amplification (Rocheville et al., 2000b) although this has yet to be validated for other GPCRs. The results in this chapter, however, go further to indicate that homo-oligomerization quality control may function as a self-regulatory mechanism that ensures that only properly formed receptors (as oligomeric complexes) are allowed to reach the cell surface (Bulenger et al., 2005). Indeed, the intracellular retention of the WT/D103A oligomer is

suggestive of perhaps an oligomeric conformation that does not meet with stringent requirements of the cellular quality control mechanisms that govern ER export and anterograde trafficking.

This scrutiny allowed WT/WT or D103A/D103A oligomers to traffic as both species exhibit normal expression in plasma membrane enriched P2 fractions (Figure 3-3) and in live cells

(Figure 3-7). This further implicates an intrinsic aberration in the structure of the WT/D103A

124 oligomer that prevents its escape from the ER. In our attempt to rescue this oligomer to the cell

surface with hydrophobic ligands, we found that only cell permeable agonists, and not

antagonists or inverse agonists, could restore this “mixed” oligomer to the plasma membrane.

This suggests that this mechanism of rescue may be distinct from that described for restoring

misfolded and sequestered receptors with pharmacological chaperones, which tend to be

receptor-specific but non discriminatory among the different classes of ligands (Bernier et al.,

2004). This pharmacological restoration of misfolded receptors was demonstrated by D1DR

antagonist (SCH23390) rescue of the presumably misfolded D120A and D120N mutant

receptors (Figure 3-8).

One explanation which we have proposed to mechanistically describe this effect is

discussed. Based on our results, we reasoned that occupancy by D1DR-selective agonists may

have intracellularly “switched” the WT receptor to an activated conformation, matching that of

the constitutively active D103A receptor, thus yielding a uniform oligomeric configuration that

restored trafficking of the sequestered WT/D103A oligomer to the plasma membrane. This

concept is depicted schematically in Figure 3-20. Ligand treatment revealed that the effect was

selective and limited only to those agonists that were cell permeable, indicating an intracellular

locus of action. This was substantiated by the fact that dopamine, a permanently charged

molecule incapable of penetrating the lipid bilayer, was found to not elicit a rescue effect. We rationalized that the binding pocket of the WT protomer was the target of hydrophobic agonist treatment. Two lines of evidence exclude the D103A binding pocket as a site of action: 1) there

was no change in the cellular distribution of D103A in response to pergolide treatment, 2) there

were no changes in the basal cAMP accumulation of D103A in response to agonists (pergolide,

SKF 81297, SKF 38393). Nevertheless, although we could exclude this site, it is possible that

125 permeable agonist

A D103A dimer WT dimer

extracellular

intracellular

uniform WT/D103A dimer

mixed WT/D103A dimer

Figure 3-20. Schematic of a hypothetical mechanism of intracellular rescue of WT/D103A oligomers. The proposed model shows that the configuration of a D1 receptor oligomer is a critical determinant in its ability to traffic to the cell surface. The wild-type D1 oligomer trafficked normally (A) as did the non-binding, constitutively active D103A oligomer (B). Co-expression of wild-type D1DR with D103A results in a mixed configuration of the oligomeric complex that had an impaired ability to traffic (C). An activated conformation of the wild-type D1 receptor induced by a cell permeable agonist may alter its conformation to resemble that of the D103A receptor (D), resulting in a uniform oligomeric configuration (similar to B), resulting in restored trafficking.

126 agonist binding could have occurred on an allosteric site in either receptor protomer. This

requires further investigation. The dopaminergic agonists, pergolide, SKF 81297, and the partial

agonist, SKF 38393 all induced a significant increase in WT/D103A oligomer density in plasma

membrane preparations with varying degrees of rescue of the retained oligomeric complex.

Extended treatment with each of these agonists (2 hours) would not only achieve cell surface

expression but would also trigger varying degrees of internalization of the wild-type receptor.

This would lead to a loss in WT density from the cell surface that may, in part, be compensated

for by the resensitization and recycling mechanism. All of the full agonists used induce

internalization and recycling of the surface WT. This process, for the D1 receptor, occurs in less

than 30 minutes (Ng et al., 1995; Vickery and von Zastrow, 1999) and may perpetuate over a 2

hour period with persistent agonist stimulation. This may partially explain why full agonists such

as SKF 81297 and pergolide did not appear to promote WT cell surface expression as

dramatically as SKF 38393. We hypothesized that restored cell surface trafficking due to

intracellular adoption of an activated conformation by the WT receptor assumes that the fully

activated conformation of WT and the D103A conformation are similar. However, SKF 38393

pre-treatment yielded the highest recovery of WT cell surface receptor density. This may have

resulted from its induction of an activated conformation in the WT which most closely mimicked that of D103A. Indeed, SKF 38393 is a partial agonist implicating its ability to put the wild-type receptor into a partially activated conformation. Interestingly, D103A was found to be only partially activated compared to the maximal stimulation of WT by dopamine as indicated by its modest degree of constitutive cAMP generation and suggests that the SKF 38393-induced WT conformation may have most closely matched the conformation of the D103A receptor. These data also indicate that multiple conformational states of the receptor may be achieved in response

127 to occupancy by different agonists as previously suggested (Gether and Kobilka, 1998; Ghanouni

et al., 2001; Perez et al., 1996). Although not consistent with our mechanistic model of agonist

induced rescue, the apparent lack of efficacy of the agonist, apomorphine, may be attributed to

its structural characteristics. When compared to the SKF series of compounds, apomorphine has

a very rigid structure and limited conformational mobility (Ryman-Rasmussen et al., 2005); this

may limit its ability to trigger the appropriate oligomeric conformational requirements necessary

for cell surface trafficking. An alternative hypothesis that was tested was whether cell

permeable agonist treatment triggered intracellular dissociation (Cheng and Miller, 2001) of the

WT/D103A oligomer resulting in separated WT and D103A receptors that could traffic to the

cell surface independently, as shown when expressed individually. However, the trFRET data

showing a robust energy transfer between WT and D103A suggested that WT/D103A oligomer

dissociation did not occur by agonist treatment since FRET was detected on the cell surface only

following agonist treatment (Figure 3-19). The absence of cell surface FRET in cells co-

transfected with WT and D103A indicates that no WT/D103A mixed oligomers were present on the cell surface under basal conditions and that the receptors present on the cell surface were only the non-interacting WT and D103A receptor homo-oligomers. Thus, these data confirm that the sequestered oligomer was trafficked intact as a complex from an intracellular compartment, and not as dissociated receptors. This intracellular compartment likely corresponds to the ER as suggested by co-localization of WT with calnexin in the presence of D103A.

Although the mechanism of agonist-rescue described in Figure 3-20 is consistent with the data that was obtained experimentally, there are other plausible mechanisms that may describe this behaviour. Some of these can be drawn from the study of molecular chaperones and other

transmembrane receptors that oligomerize. For example, the human insulin receptor (HIR), a

128 member of the receptor tyrosine kinase family, is expressed at the cell surface as an ER-derived homodimer (Lu and Guidotti, 1996). HIR maturation involves the co-translational trimming of three glucose residues by glucosidase I and II to a single, terminal glucose on high-mannose type oligosaccharides. The resulting monoglucosylated core glycans serves as a substrate for binding to calnexin and calreticulin which is required for proper folding and dimerization of nascent receptor monomers. The addition of glucose trimming inhibitors, such as castanospermine, prevents the binding of these chaperones to the HIR resulting in premature processing manifested as accelerated dimerization and misfolded oligomeric assembly (Bass et al., 1998).

Thus, the HIR requires ER chaperone association to maintain oligomer fidelity, possibly by sterically masking hydrophobic interfaces that would otherwise cause aggregation of the nascent monomeric protein. This requirement for chaperone interaction has also been shown to be necessary for the maturation of other membrane bound dimeric proteins (Boyd et al., 2002;

Chang et al., 1997). It is possible then, that biosynthesis of the WT/D103A oligomer may not result in a conformation that is suitable for interaction with certain molecular chaperones.

Indeed, a previous report has shown that D1DR interacts with calnexin and that this interaction is tightly controlled, as increasing (through calnexin over-expression) or decreasing (through receptor mutagenesis and pharmacological inhibition of calnexin) D1DR-calnexin interactions causes intracellular retention of D1DR (Free et al., 2007). This study demonstrates the importance of glycosylation for D1DR trafficking, although other studies have shown that unglycosylated D1DR can traffic to the plasma membrane (Karpa et al., 1999). Alternatively, it is possible that the interaction of BiP, a prototypical HSP70 that binds to hydrophobic regions of misfolded proteins, may be required for D1DR oligomeric maturation. Alternatively, any member of the dimeric family of 14-3-3 chaperones (Fu et al., 2000) may also be involved as

129 these have been shown to promote proper assembly of multimeric potassium channels (Yuan et al., 2003). The addition of cell permeable agonists may put the WT protomer within the

WT/D103A oligomer in a conformation suitable for chaperone interaction allowing correct oligomeric assembly. These chaperones may mask ER retention motifs or expose ER export motifs, both of which can result in the release of the protein from the biosynthetic compartment.

Given this possibility, it would be interesting to determine the roles of various proteins directly implicated in D1DR export, such as DRiP78 (Bermak et al., 2001) and neurofilament-M (Kim et al., 2002), in oligomer maturation.

One major hurdle in defining the precise role of homo-oligomerization in GPCR function has been the problem of identifying a consensus dimeric interface. Such information might be useful in “monomerizing” a GPCR and studying the impact of its function in relation to the native dimeric/oligomeric state. Studies from our laboratory and others have implicated a role for

TM4 in forming a symmetrical dimeric interface in the D2 dopamine receptor (Guo et al., 2003;

Lee et al., 2003b). In contrast, such studies in the α1b-adrenergic receptor and the yeast α factor receptor have suggested that TM1 is also critical for receptor dimerization (Carrillo et al., 2004;

Carrillo et al., 2003; Overton and Blumer, 2002). Interestingly, using a disulfide trapping approach involving strategically inserted cysteines, the symmetric dimer interface in the complement 5a receptor was predicted to involve either TM1 and TM2 together or TM4 (Klco et al., 2003). The involvement of both these regions is not necessarily mutually exclusive as these helices may also participate in the formation of higher order oligomeric species (Lopez-Gimenez et al., 2007). Collectively, these studies appear consistent with computational models of certain opioid receptor homo-oligomers implicating the involvement of these helices in the formation of the dimeric interface (Filizola et al., 2002; Filizola and Weinstein, 2002; Filizola and Weinstein,

130 2005). The analysis of rhodopsin in rod outer segments of mouse retina by atomic force microscopy was the first to qualitatively show the organization of rhodopsin molecules as orderly rows of dimers (Fotiadis et al., 2003; Fotiadis et al., 2004; Liang et al., 2003). By analyzing the dimensions of rhodopsin paracrystals and their contact restrictions, these authors were able to generate a molecular model predicting the interaction points within and between these dimeric rows (Figure 3-21). It was proposed that TM4 and TM5 formed a symmetric intradimeric interface between individual monomers. The contact point between dimers was predicted to involve an interaction between ICL1 of a monomer and ICL3 of a monomer in the partnered dimer. Finally, hydrophobic residues on the extracellular face of TM1 were calculated to be involved in forming a symmetric interface between the dimeric rows. This supports the aforementioned biochemical and molecular studies of certain GPCRs predicting that TM4 and

TM1 are involved in forming the intradimeric and interdimeric interface, respectively. It should be noted, however, that this empirical data obtained through AFM is still controversial (Chabre et al., 2003). Furthermore, it remains to be determined whether these predicted interfaces are a conserved feature among GPCRs, a finding that if proven true, could shed more light on the functional significance of homo-oligomerization.

3.13 Acknowledgements

I would like to acknowledge Theresa Fan for performing most of the confocal microscopy experiments as well as George Varghese for helping with the staggered co- expression experiment.

131 B

μ A δ ε

Figure 3-21. (A) Topograph showing paracrystalline arrangement of rhodopsin dimers from native disc membranes using atomic force microscopy (Reprinted with permission from Fotiadis et al, 2004). The occasional dimer is separated from a neighbouring row (circled by ellipse) while some rhodopsin is shown as a monomeric unit (indicated by arrow). (B) Dimer contact interfaces as predicted by molecular modelling of rhodopsin.Top schematic shows the cytoplasmic side of rhodopsin oligomers; bottom schematic shows the extracellular side. The contact points between individual monomers is mediated by TM 4 and 5 (indicated by μ). Most contacts between dimers are mediated by the intracellular loop linking TM 5 and 6 of one monomer and the intracellular loop linking TM 1 and 2 of the monomer in the partnered dimer (indicated by ε). Hydrophobic residues on the extracellular face of TM 1 mediate forms the contact between rows of dimers (indicated by δ).

132

4 REGULATION OF D1 DOPAMINE RECEPTORS IN LIPID RAFTS

4.1 Introduction

Caveolae have a well defined role in mediating the activity of a number of signal transduction pathways including those involving receptor tyrosine kinases (Liu et al., 1996;

Mineo et al., 1996; Yamamoto et al., 1998) and multichain immune recognition receptors

(Dykstra et al., 2001). There is emerging evidence that G protein coupled receptor (GPCR) function is also modulated by localization in caveolae as a wide array of GPCR signalling molecules including G proteins (Allen et al., 2005; Oh and Schnitzer, 2001), RGS proteins (Hiol et al., 2003), and protein kinases (Mineo et al., 1998; Razani et al., 1999; Rybin et al., 2000) have been reported to compartmentalize in these microdomains. The caveolar localization of

GPCRs has been reported to have various functional consequences with roles in agonist-induced signalling, internalization, and the activation of various effector pathways (Bhatnagar et al.,

2004; Guzzi et al., 2002; Igarashi and Michel, 2000; Lamb et al., 2002; Roettger et al., 1995;

Rybin et al., 2000; Yamaguchi et al., 2003).

Caveolae represent a subtype of lipid rafts that exist as morphologically distinct invaginations at the plasma membrane. These lipid enriched entities move laterally on the cell surface while allowing the exchange of proteins and lipids between the raft domain and the surrounding liquid disordered phospholipid environment (Brown and London, 1998; Rajendran and Simons, 2005). This dynamic regulation is believed to facilitate the formation of cell surface signalling platforms for the integration of various signalling molecules, thus ensuring specificity and efficiency in signal transduction processes. The caveolin proteins are unique to caveolae and

133 serve a dual role in maintaining the structural integrity of caveolae and by acting as a scaffolding

protein that binds to a battery of receptors, signalling molecules, and adapter proteins (Williams

and Lisanti, 2004). There are three caveolin isoforms, each of which can serve as biochemical

markers for the identification of caveolae; caveolin-1 is the most ubiquitously expressed as it is

found in tissues including the lung, heart, and brain while caveolin-2 colocalizes with and

requires caveolin-1 for proper membrane targeting. Caveolin-3 has greater sequence similarity to

caveolin-1 than caveolin-2 but is expressed mainly in skeletal, smooth, and cardiac muscle. The

expression of caveolin-1 in brain has been shown in neuronal cells such as hippocampal and

dorsal root ganglion neurons (Braun and Madison, 2000; Bu et al., 2003; Galbiati et al., 1998;

Gaudreault et al., 2005) as well as glial cells such as astrocytes and oligodendrocytes (Cameron et al., 1997; Ikezu et al., 1998).

Despite the presence of caveolin-1 in brain, there is little known about how this ubiquitous protein modulates the function of those GPCRs that are involved in critical aspects of brain function. The D1DR is the most abundant dopamine receptor subtype in the brain with an expression profile that covers various regions including the striatum, nucleus accumbens, and hippocampus. There is also evidence for its expression in glial cells, particularly astrocytes, from striatum and basal ganglia (Miyazaki et al., 2004; Zanassi et al., 1999). In the brain, D1R participates in the modulation of various neural processes including learning, memory, reward, and locomotor activity. The D1DR couples to Gs/Golf to activate the adenylyl cyclase effector

pathway which, in turn, modulates intracellular levels of cAMP. Many components of this

signalling pathway, such as Gsα and specific adenylyl cyclase isoforms, compartmentalize in

caveolae (Schwencke et al., 1999) suggesting that receptors such as D1DR that are associated with this signalling cascade might also localize in these microdomains. The ability of such

134 signalling molecules to localize in caveolae has been proposed to be mediated by a direct

interaction between the scaffolding domain of caveolin-1 and a putative caveolin binding motif

(CBM) found in most caveolae associated proteins (Couet et al., 1997). This binding motif is

characterized by the amino acid sequence φXφXXXXφ, φXXXXφXXφ, or φXφXXXXφXXφ

(where X is any amino acid and φ is any one of the aromatic amino acids Trp, Phe, or Tyr). We

found that D1DR contains a caveolin binding motif in the proximal region of TM7 domain thus

implicating a role for caveolin-1 in D1DR function.

4.2 Differential Caveolin-1 Expression in HEK293t and COS7 cells

In order to study D1DR function under physiological caveolin-1 expression levels, we were interested in determining which cell line had an adequate caveolin-1 expression profile.

HEK293t cells have an integrated SV40 large T antigen and, like many oncogenically transformed cells, express less caveolin than the parental wild-type cells (in this case, HEK293 cells) (Koleske et al., 1995). This is believed to be linked to the role of caveolin-1 as a tumour suppressor gene (Williams and Lisanti, 2005). To isolate subcellular fractions containing caveolae related lipid raft domains, caveolin-enriched fractions were purified from both

HEK293t and COS-7 cell lysates using a discontinuous sucrose density gradient centrifugation.

This fractionation scheme utilizes sodium carbonate to separate caveolin-enriched microdomains based on the high cholesterol and sphingolipid content that renders these fractions carbonate insoluble with a low buoyant density (Song et al., 1996). All of the endogenously expressed caveolin-1 in HEK293t cells was recovered in fraction 5 (Figure 4-1B), which corresponds to the

“light” vesicle or caveolae-enriched fractions (Igarashi and Michel, 2000). As shown, we were unable to detect any obvious levels of caveolin-1 expression under short film exposure times

135 A B fraction fraction

1 2 3 4 5 6 7 8 9 10 11 12 1 2 3 4 5 6 7 8 9 10 11 250 _ 250 _ 148 _ 148 _ _ 98 98 _ 64 _ 64 _

MW _ MW _ ( kDa ) 50 ( kDa ) 50

36 _ 36 _

22 _ 22 _

anti caveolin-1 (24 kDa) anti caveolin-1 (24 kDa)

Figure 4-1. Endogenous expression of caveolin-1 (24 kDa) from sucrose gradient fractions prepared from HEK293t cells. Blots were exposed for (A) 1 min and (B) 5 min

136 (Figure 4-1A) . A substantially longer exposure time (5 min) was required to obtain a weak band

at 24 kDa corresponding to the molecular mass of caveolin-1 (Figure 4-1B). In contrast,

caveolin-1 was robustly expressed in COS7 cells with the corresponding band detected within

less than 2 minutes of film exposure in fractions 4 and 5 (Figure 4-2). This indicates that COS7

cells have abundant caveolin-1 expression implicating the presence of functional plasmalemmal

caveolae in this cell line.

4.3 Localization of D1DR in Caveolin-enriched Fractions

In order to determine whether D1DR was closely associated with caveolae, we transiently

transfected COS7 cells with HA-tagged D1DR and prepared sucrose gradient fractions as

described in the previous section. When these fractions were analyzed for the subcellular

distribution of D1DR, a substantial fraction of D1DR monomers (52 kDa) and dimers (105 kDa)

was found in the caveolin-enriched fractions with some D1DR recovery in non-caveolin-

enriched fractions (Figure 4-3). A similar distribution of endogenous D1DR was obtained when

whole rat brain lysates were subjected to sucrose density gradient centrifugation (data not

shown). This suggests that under basal conditions, a significant proportion of D1DRs were localized at the plasma membrane in caveolae microdomains. We confirmed the quality of our

+ + gradient preparations by probing for endogenously expressed Na /K -ATPase, Gsα, and clathrin.

We found that Na+/K+-ATPase co-fractionated with caveolin-1 and that clathrin fractionated in the higher density membrane fractions (fractions 9-12). Na+/K+-ATPase typically resides at the

plasma membrane and its function is dependent on the structural integrity of caveolae (Wang et

al., 2004); it has also previously been shown to bind to caveolin-1 indicating that fractions 4 and

5 likely represent plasma membrane caveolar domains. ARF-6, a GTP binding protein involved

137 fraction

1 2 3 4 5 6 7 8 9 10 11 12 kDa

64 -

MW 50 - (kDa) 36 -

16 -

anti caveolin-1 (24 kDa)

Figure 4-2. Endogenous expression of caveolin-1 (24 kDa) from sucrose gradient fractions prepared from COS7 cells with a short exposure time.

138 5 % 35 % 45 % sucrose (w/v)

fraction 1 2 3 4 5 6 7 8 9 10 11 12

Na+/K+ ATPase 105 kDa

caveolin-1 24 kDa

105 kDa HA (D1R) 52 kDa

52 kDa Gsα 45 kDa

clathrin HC 192 kDa

ARF6 22 kDa

Figure 4-3. Localization of D1DR in caveolin-1 enriched fractions. Detergent-free sucrose gradient fractions were prepared from D1DR-transfected COS7 cells, separated on SDS- PAGE, and probed with antibodies against the indicated proteins. D1DR was found as monomers and dimers (as indicated by HA immunoreactivity) in both non caveolin-1 enriched fractions and caveolin-1 enriched fractions (as indicated by the presence of caveolin-1 in fraction 5). Similar results were obtained when sucrose density gradient fractions from whole rat brain lysates were probed for endogenous D1DR. Both Gsα and Na+/K+ ATPase were also found in caveolin-1 enriched fractions whereas clathrin heavy chain and ARF-6 was found exclusively in the high density caveolin-1 free fractions.

139 in vesicular trafficking, which has previously been implicated in caveolar internalization of

certain GPCRs (Houndolo et al., 2005) was found exclusively in the higher density fractions. Gsα

was localized to both the caveolin-enriched fractions (fraction 4) as well as the high density

sucrose fractions, as expected (Li et al., 1995).

4.3.1 Co-localization of D1DR with Caveolin-1

The co-fractionation of D1DR with caveolin-1 suggested that both of these proteins were

co-localized, presumably in caveolae, under basal conditions. To provide qualitative evidence for this, we performed immunofluorescence microscopy on HEK293t cells co-expressing HA-tagged

D1DR and cMyc-tagged caveolin-1 (Figure 4-4). In the absence of ligand, D1DR was expressed

uniformly at the cell surface. In contrast, caveolin-1 exhibited a reticular cytosolic distribution

with some localization at the cell surface. This intracellular pool of caveolin is consistent with

the role of caveolin-1 in macromolecular transcytosis (Bendayan and Rasio, 1996; Ghitescu et

al., 1986), membrane trafficking (Schnitzer et al., 1996), as well as the transport of cholesterol

and lipids between the ER and the plasma membrane (Smart et al., 1996; Smart et al., 1994;

Uittenbogaard et al., 1998). The plasma membrane resident caveolin-1 also displayed a punctate

distribution with regions of co-localization with D1DR (arrows, merged).

4.3.2 Interaction of D1DR with Caveolin-1

To assess whether localization of D1DR in caveolin-enriched microdomains was due to a

physical association between caveolin-1 and D1R, we used a two-pronged approach utilizing co-

immunoprecipitation and bioluminescence resonance energy transfer (BRET) assays. Co-

immunoprecipitation experiments were done in COS-7 cell lysates transiently co-expressing HA-

140 D1 (GFP) caveolin-1 (TRITC) merged

Figure 4-4. HEK293t cells were co-transfected with D1DR-GFP and cMyc-tagged caveolin-1 which was detected using a TRITC-conjugated secondary antibody. Cell surface localization of D1DR-GFP and caveolin-1 is shown in the merged image (arrows).

141 tagged D1DR and cMyc-tagged caveolin-1. Antibodies directed against HA co-

immunoprecipitated cMyc-caveolin-1 (Figure 4-5A, lane 3). Western analysis of recombinant caveolin-1 expression yielded a doublet of 24 kDa and 21 kDa which corresponds to the α and β isoforms of caveolin-1, respectively (lane 1). The specificity of these interactions was tested by immunoprecipitating with another IgG under identical conditions (mock, Figure 4-5A); this did not yield a co-immunoprecipitate. It is not known why COS7 cells endogenously express only the α isoform of caveolin-1 although this has been observed in other studies (Igarashi and

Michel, 2000; Schwencke et al., 1999). Since the D1 receptor and caveolin-1 are prevalent in brain, we wanted to confirm the physiological nature of this interaction in whole rat brain tissue lysates. Immunoprecipitation with anti-D1DR could only co-precipitate the α isoform of caveolin-1 (Figure 4-5B, lane 2) despite the presence of both isoforms in brain (Figure 4-5B, lane 1).

To evaluate the relative proximity of caveolin-1 and D1DR, we used a BRET assay to show an interaction between these two proteins. BRET is based on the theory that bioluminescent energy transfer from the catalytic degradation of the substrate, coelenterazine, by a luciferase donor to a fluorescent acceptor requires an intermolecular distance of less than 50 Å; this distance posits a direct interaction between the proteins being investigated. Energy transfer is typically quantified by an increase in the ratio of acceptor emission (GFP) to luciferase emission

(Rluc). In this study, we used HEK293t cells (which are largely devoid of caveolin expression in order to avoid potentially confounding results from endogenous caveolin-1 interactions with

D1DR. The basal BRET was determined from HEK293t cells co-expressing constant levels of

D1DR-Rluc and increasing levels of caveolin-1-GFP (cav1-GFP) fusion protein (Figure 4-6).

The BRET ratio increased as a hyperbolic function of the concentration of GFP fusion construct

142 A B

IP: cMyc HA mock IP: D1 kDa kDa 1α 22 22 1β

IB: cMyc IB: caveolin-1

Figure 4-5. (A) Co-immunoprecipitation of D1DR and caveolin-1 from lysates of COS7 cells transfected with HA-tagged D1DR and cMyc-tagged caveolin-1. Immunoprecipitations were performed with either monoclonal anti-cMyc, polyclonal anti-HA, or irrelevant IgG (mock), and subsequently immunoblotted with a cMyc specific monoclonal antibody. Immunoblot analysis in the absence of immunoprecipitating antibody is also shown (lane 1). (B) Co-immunoprecipitation of D1DR and caveolin-1α from whole rat brain lysate. Immunoprecipitation was performed with monoclonal anti-D1DR and samples were subjected to immunoblots using monoclonal anti-caveolin-1. Both caveolin-1α and 1β are expressed in whole rat brain homogenate (lane 1).

143 0.125

0.100

0.075 caveolin-1-GFP/D1-Rluc 0.050 pEGFP/D1-Rluc BRET ratio BRET

0.025

0.000 0.0 2.5 5.0 7.5 10.0 12.5

protein-GFP/D1-Rluc DNA ratio

Figure 4-6. BRET saturation curves were generated from HEK293t cells co-transfected with a constant cDNA concentration of D1-Rluc and increasing concentrations of caveolin-1-GFP (■) or soluble GFP encoded by pEGFP vector (▼). The BRET, total fluorescence, and total luminescence were determined 48 h after transfection. The BRET levels were normalized against luminescent emission levels at 480 nm in the absence of GFP (background). Non- linear regression analysis was performed to obtain a Bmax of 0.122 in cells co-expressing D1DR-Rluc and caveolin-1-GFP. The results are expressed as the mean BRET ratio ± S.E.M. of 4-9 independent experiments carried out in replicates of six.

144 added and reached an asymptote that corresponded to a BRETmax value of 0.122. The BRETmax indicates the acceptor concentration required to attain maximum D1R/caveolin-1 coupling. This saturable characteristic indicates that the interaction between D1R and caveolin-1 was specific and not a result of random collisions that would otherwise yield a linear “bystander” BRET curve (Mercier et al., 2002). In contrast to this, co-expression of D1R-Rluc with soluble GFP resulted in a negligible BRET signal that increased linearly with increasing expression levels of pEGFP vector.

4.4 Agonist-induced Translocation of D1DR into Caveolae

To test whether stimulation of the D1 receptor with agonist altered its localization in

caveolae, HA-D1R-transfected COS7 cells were treated with the D1R agonist, SKF81297, for

various time points and the distribution of D1R was analyzed from subcellular fractions, as

described above. As shown in Figure 4-7, D1R localized to caveolin enriched (fractions 4 and 5)

and non-caveolin enriched (fraction 8) fractions under basal conditions. Treatment of the D1

receptor with 10 μM SKF81297 for 5 min increased the proportion of D1 receptors in caveolar

fractions with a concomitant decrease in D1 receptor recovery from non-caveolar fractions.

Within 20 min of SKF81297 stimulation, the majority of receptor protein (~ 70 %) was localized

in the caveolar fractions with a minor proportion of receptor in the non-caveolar fractions.

Densitometric analysis of the subcellular distribution of D1 receptors showed that there was no

change in the total amount of protein recovered following agonist stimulation thus indicating that

the D1 receptor translocated from the non-caveolar fractions to the caveolar fractions. Similar

findings were obtained when dopamine was used to stimulate the receptor (data not shown). The

localization of endogenously expressed caveolin-1 did not change in response to SKF81297

145 0 min 5 min

1 2 3 4 5 6 7 8 9 10 11 12 kDa 1 2 3 4 5 6 7 8 9 10 11 12 kDa

50 - D1R 50 - D1R

24 - cav-1 24 - cav-1

60 60

50 50

40 40

30 30

20 20 % total receptor density 10 % total receptor density 10

0 0 0123456789101112 0123456789101112 fraction fraction

10 min 20 min

kDa 1 2 3 4 5 6 7 8 9 10 11 12 kDa 1 2 3 4 5 6 7 8 9 10 11 12

50 - D1R 50 - D1R

24 - cav-1 24 - cav-1

60 60

50 50

40 40

30 30

20 20

% total receptor density 10 % total receptor density 10

0 0 0123456789101112 0123456789101112 fraction fraction

Figure 4-7. Agonist-induced translocation of D1DR into caveolin-1 enriched fractions. Detergent- free sucrose gradient subcellular fractions were prepared from HA-D1DR-transfected COS7 cells that were treated for various times (0, 5, 10, 20 min) with the D1DR agonist, SKF81297 (10 μM). An equal volume of each fraction was separated on SDS-PAGE and analyzed by immunoblotting with antibodies directed against the HA epitope (top panel) or caveolin-1 (middle panel), as shown. No change in the distribution of caveolin-1 (fractions 4 and 5) was observed with agonist treatment. Semi-quantitative analysis of D1DR expression in each fraction was performed using densitometry as shown in the bottom panel.

146 treatment.

4.4.1 Effect of Endocytosis Inhibitors on D1DR Internalization

Based on the observation that the D1 receptor translocates to caveolin enriched fractions

following SKF81297 stimulation, we wished to determine whether this might be attributed to

agonist-dependent D1DR endocytosis. Although clathrin-dependent internalization has been shown to be the major endocytic pathway for many GPCRs, including the D1 receptor in

HEK293t cells (Vickery and von Zastrow, 1999), endocytosis through caveolae has also been reported to be an alternative route for the cellular entry of certain GPCRs. While both processes appear to be dynamin dependent (Henley et al., 1998; Robinson, 1994), it has been suggested that while clathrin dependent endocytosis is dependent on phosphorylation by G protein receptor kinase (GRK) and arrestin binding, caveolar endocytosis depends on phosphorylation by protein kinase A (PKA) (Rapacciuolo et al., 2003). To address this, we performed whole cell radioligand binding assays and cell surface biotinylation studies to quantify the degree of D1 receptor internalization in the presence of various inhibitors of caveolar and clathrin mediated endocytic pathways. In whole cell binding assays, D1DR transfected COS7 cells were pre-incubated with hypertonic sucrose or concanavalin A, both of which are known inhibitors of clathrin-mediated endocytosis (Heuser and Anderson, 1989; Waldo et al., 1983), or methyl-β-cyclodextrin

(mβCD), a cholesterol depleter known to disrupt caveolae structure and function (Pike and

Miller, 1998). Whole cell surface binding of [3H]SCH23390, a selective D1DR antagonist, to D1

receptors was compared before and after 30 minute incubation with 10 μM SKF81297 in the

presence of these compounds. Pre-treatment of cells with sucrose or concanavalin A did not

significantly alter the degree of D1 receptor internalization (Figure 4-8). Similarly, transfection

147 70

40 *

% D1DR internalization 10 *

0 l ) ) e ) ro 1 :2 A D :1 9 t 1: os in C 8 ( (1 cr β (1 H R u m L Con 0R s aval 2 M 20 % μ 2 22 13 0 +2 -P 3 45 M ncan 1 + 2-K 2-K . o v K C a R +0 + c G GRK

Figure 4-8. HA-D1DR expressing cells were co-expressed with GRK2-K220R at a 1:1 and 1:2 transfection ratio or caveolin-1 P132L (cav1-P132L), or pre-treated with 0.45 M sucrose, 0.25 mg/ml concanavalin A, 2% methyl-β-cyclodextrin (mβCD), or 30 μM H89 for 30 min prior to agonist stimulation with SKF81297 (10 μM) for an additional 30 min. Receptor density was estimated by whole cell radioligand binding analysis with 2 nM [3H] SCH23390. The results are expressed as the mean % internalization ± S.E.M. of 3-5 independent experiments. Significance at p<0.05 versus % internalization under control conditions is denoted by *.

148 of the D1 receptor with a dominant negative mutant of GRK2, K220R, did not change the extent

of internalization even when over-expressed. This suggests that clathrin-mediated internalization

is not the only endocytic route used by the D1 receptor. In contrast to this, pre-treatment with 2%

mβCD almost completely abolished agonist-induced internalization. Consistent with this, co-

transfection of the D1 receptor with a dominant negative mutant of caveolin-1, P132L (cav1-

P132L) (Lee et al., 2002a), significantly attenuated D1DR internalization by approximately 38

%. To determine whether D1R internalization was PKA-dependent, we tested the effects of H89,

a PKA selective inhibitor, on receptor sequestration. The inhibition of PKA function yielded a

minor, but insignificant, decrease in agonist-mediated internalization.

To further evaluate whether D1DR internalized through a caveolar pathway, we used

glutathione-cleavable biotin to assess the amount of internalized receptor following caveolae

disruption and agonist stimulation (Figure 4-9). The radioligand binding experiments described

above were used to show changes in cell surface binding in response to caveolae disruption and

agonist stimulation. Hence, we conducted cell surface biotinylation assays to show that these

changes were actually due to differences in the amount of receptor internalized. Briefly, D1R

transfected cells were pretreated with cell impermeable cleavable biotin before stimulation with

agonist. All cell surface bound biotin was then stripped with glutathione leaving only

internalized biotinylated receptors protected from glutathione cleavage. These biotinylated

internalized receptors were then detected by immunoprecipitation and western blot analysis. As

shown in Figure 4-9, D1R stimulation with 10 μM SKF81297 triggered an increase in internalized receptor (lane 3). In contrast, pre-treatment of cells with mβCD or filipin, a sterol binding agent that inhibits caveolae formation (Orlandi and Fishman, 1998), attenuated the amount of internalized biotinylated receptor detected. Similar results were obtained when cells

149 CD β

- SKF 81297 P132L-caveolin-1control (+ SKF+ 81297) m + filipin kDa

50 -

+ SKF81297

Figure 4-9. HA-D1DR expressing cells were either co-expressed with caveolin-1 P132L, or pre-treated with 2% mβCD or 1 μg/ml filipin before cell surface biotinylation was performed. Cells were stripped of biotin after stimulation with agonist for 30 min (and in the absence of agonist, lane 1) and internalized receptors were immunoprecipitated with polyclonal anti-HA before analysis on SDS-PAGE.

150 were co-transfected with cav1-P132L, indicating that this attenuation was specifically a result of caveolae disruption. Since caveolar translocation occurs within twenty minutes of agonist induction (Figure 4-7), these results are consistent with a mechanism where D1R translocates to caveolae upon binding to agonist before undergoing caveolar endocytosis.

4.4.2 Kinetics of D1DR Internalization

Since D1DR internalization in COS7 cells was insensitive to inhibitors of clathrin mediated endocytosis but attenuated by known disrupters of caveolar function, we concluded that the D1 receptor likely undergoes caveolar endocytosis in this cell line. In order to quantify the internalization kinetics of the D1 receptor through caveolae, we measured the extent of receptor internalization over a fixed time period using binding assays. Following 5 minutes of receptor stimulation with 10 μM SKF81297, approximately 20% of the cell surface population of

D1 receptors was internalized (Figure 4-10). Maximum internalization was achieved at 45 minutes where approximately 55% of receptors were internalized. This indicates that caveolar endocytosis of D1DR is a kinetically slower process than clathrin-mediated endocytosis of

D1DR, the latter of which occurs more rapidly with approximately 65 % receptor internalization occurring within 5 minutes of agonist stimulation (Vickery and von Zastrow, 1999).

To monitor the intracellular distribution of the D1 receptor in caveolae, we used real-time live cell imaging D1DR-mRFP and caveolin-1-GFP. In the absence of agonist, D1DR-mRFP exhibited a predominantly cell surface distribution whereas caveolin-1-GFP was localized to both cell surface and perinuclear regions (Figure 4-11, top row). Within 12 minutes of incubation with agonist at 37°C, the appearance of distinct vesicles containing caveolin-1-GFP and D1DR- mRFP originating from the cell surface could be observed (Figure 4-11, second row). These

151 80

H] 70 3

SCH23390 binding) 20 % D1R internalization % D1R

(asdetermined [ by 2 nM 10

0 0 10 20 30 40 50 60 70 80 90 Time (minutes)

Figure 4-10. Whole cell radioligand binding analysis was performed with 2 nM [3H] SCH23390 on D1DR-expressing cells exposed to 10 μM SKF81297 for the indicated time periods. Results are expressed as the mean % internalization ± S.E.M. of 3 independent experiments.

152 D1R (mRFP) caveolin-1 (GFP) merged

basal

20 min

25 min

40 min

60 min

Figure 4-11. HEK293t cells were co-transfected with D1DR-mRFP and caveolin-1-GFP. Shown are live cell confocal microscopy images obtained over a 60 minute time period of SKF81297 treatment (10 μM) in serum-free MEM. Co-localization of D1DR-mRFP and caveolin-GFP is shown in the merged image (yellow) and maintained throughout the duration of agonist treatement (arrowhead)

153 vesicles, presumably caveolae, were found to redistribute to intracellular regions upon continuous agonist exposure with the colocalization of caveolin-1 and D1DR maintained throughout the endocytic process. These vesicles could be seen trafficking back to the cell surface within 40 minutes, ultimately returning to the plasma membrane within 60 minutes of initial agonist stimulation. This may implicate an additional role for caveolae in D1DR recycling.

4.5 Pharmacological Characterization of D1DR Mutants Lacking an Intact Caveolin

Binding Motif

Since caveolae-associated proteins require an intact caveolin binding motif to interact with caveolin-1 (Couet et al., 1997), we designed several D1DR mutants by site-directed mutagenesis in which the critical amino acids within the CBM in the proximal half of the seventh

TM domain were disrupted. This motif lies just upstream of the highly conserved NPXXY motif.

The F313A, W318A, and W321A mutants disrupted the CBM at the proximal, medial, and distal aromatic residues, respectively, whereas the FWW/A mutation had all three aromatic residues substituted for alanine. Plasma membrane expression of HA-tagged CBM mutants was assessed through cell surface biotinylation (Figure 4-12). The F313A and W318A mutants were found to have slightly higher and lower plasma membrane expression than wild-type D1DR, respectively.

In contrast, cell surface expression of the W321A and FWW/A mutants was strongly attenuated compared to wild-type D1DR possibly due to protein misfolding. To assess whether the pharmacological properties of the CBM mutants were altered, saturation binding analysis was performed on F313A and W318A. These single point mutants displayed high affinity binding for

3 [ H]SCH23390 with Kd values of 0.57 nM and 0.61 nM, respectively, which is comparable with that of wild-type D1DR (Table 4-1) The Bmax values were 1.8 and 0.86 pmol/mg membrane

154 D1 F313A W318A W321A FWW/A kDa

50 -

Figure 4-12. Plasma membrane expression of HA-tagged D1DR mutant receptors with a disrupted caveolin binding motif was assessed by cell surface biotinylation and immunoprecipitation with anti-HA specific antibodies. The result is representative of 3 independent experiments.

155

Bmax (pmol/mg membrane protein) Kd (nM)

Wild-type 1.22 ± 0.1 0.26 ± 0.02

F313A 1.80 ± 0.1 0.57 ± 0.05

W318A 0.86 ± 0.1 0.61 ± 0.03

W321A n/d n/d

FWW/A n/d n/d

Wild-type 1.19 ± 0.1 0.37 ± 0.08 + 2% mβCD

Table 4-1. [3H] SCH23390 binding parameters ± S.E.M. for wild-type D1DR and mutant receptors lacking an intact caveolin binding motif (n/d - not detected).

156 protein, respectively, with relative receptor densities that correlated with plasma membrane

receptor expression as determined by cell surface biotinylation. The triple substitution mutant,

FWW/A, exhibited negligible binding which was consistent with the marked decrease in cell

surface expression. This failure to translocate to the plasma membrane was attributable to the

distal aromatic residue, W321, since this mutant similarly exhibited a poor binding and

expression profile.

4.5.1 Role of Caveolin Binding Motif in D1DR Internalization

To further define the role of the caveolin binding motif in the D1DR, we performed

internalization studies using whole cell binding assays on the various CBM mutants (Figure 4-

13A). Pre-incubation of cells expressing the wild-type D1 receptor with 10 μM SKF 81297 for

30 minutes resulted in a loss of 57.7 ± 3.3 % of cell surface receptors. In contrast, the extent of receptor internalization for F313A and W318A was significantly reduced to 15.4 ± 4 % and 4.5 ±

6.7 %, respectively. We could not accurately determine the extent of internalization for W321A

due to its poor expression. To determine whether F313A and W318A could interact with

caveolin-1, we performed BRET competition experiments in which untagged F313A and

W318A were individually tested for their ability to reduce the BRET signal generated between

D1DR-Rluc and cav1-GFP. The selected CBM mutants were co-transfected with an amount of

D1DR-Rluc and cav1-GFP found to generate a near maximal BRET signal (Figure 4-6). We

transfected a specific amount of cDNA that was optimized for each mutant receptor and not

found to compromise cav1-GFP or D1DR-Rluc expression. Expression of untagged wild-type

D1DR markedly reduced the BRET signal generated by D1R-Rluc and cav1-GFP (Figure 4-

13B). In contrast, expression of either untagged F313A or W318A did not significantly affect the

157 A

30 * 20 * % D1R internalization 10

0 D1 F313A W318A

0.100

0.075

* 0.050

0.025 D1-Rluc:cav1-GFP BRET ratio BRET D1-Rluc:cav1-GFP 0.000

D1 - + - - competing F313A - - + - receptor { W318A - - - +

Figure 4-13. Caveolar internalization of D1R requires an intact caveolin binding motif. (A) COS7 cells were transfected with wild-tye D1DR, F313A, or W318A and treated with SKF81297 (10 μM) for 30 min before receptor density was estimated by whole cell radioligand binding analysis with 2 nM [3H] SCH23390. Results are expressed as the mean % internalization ± S.E.M. of at least 3 independent experiments. Significance at p

158 BRET signal between the two fusion proteins. Equivalent expression of transfected F313A and

W318A was confirmed by radioligand binding assays (data not shown). These data suggest that

the D1DR interacts with caveolin-1 through a series of aromatic amino acids defined by a putative caveolin binding motif and that the integrity of this motif is critical in mediating

caveolar endocytosis.

4.6 Role of Caveolae in D1DR Signalling

Based on the observation that Gsα and adenylyl cyclase (Head et al., 2006) co-localized with the D1 receptor in caveolin-enriched sucrose gradient fractions (Figure 4-3), we sought to investigate the role of caveolin-1 on D1R-mediated cAMP signaling. Using an ELISA assay, we wanted to determine what effect caveolar disruption would have on the ability of the D1R to activate adenylyl cyclase and enhance cAMP production. Cells expressing D1R were pre-treated with 2% mβCD before stimulating with various concentrations of SKF81297 for 20 minutes.

There was no significant difference between the EC50 of the dose-response curves for cAMP

generation (Figure 4-14A) nor were there significant differences between the Kd for SCH23390

and Bmax of D1R in the presence or the absence of mβCD (Table 4-1). However, we found that

there was an approximately 35 % increase in cAMP at the highest concentration of SKF81297

used (corresponding to the Vmax) in the presence of mβCD (Figure 4-14A). This enhancement

was reversed upon cholesterol replenishment as described in Materials and Methods (data not

shown). Similarly, co-transfection of cav1-P132L at a 4-fold cDNA concentration over that of

D1R was also found to enhance the Vmax of cAMP accumulation by approximately 43 % without

significantly altering the EC50 (Figure 4-14B).

To test the hypothesis that the enhancement in cAMP production was Gs-mediated and

159 A B

150 175 - mβCD 125 150 + mβCD 125 100 100 75 75 - P132L 50 (of WT receptor) (of WT receptor) 50 P132L (1:1) P132L (1:2) 25 25

% maximum cAMPproduced P132L (1:4) % maximum cAMPproduced

0 0 0 -8-7-6-5-4-3 0 -8-7-6-5-4-3 log [SKF 81297] (M) log [SKF 81297] (M)

300 **

200 CD basal levels) β S incorporation γ γ 100

% GTP 0

(relative to (relative -m - m βCD + mβCD

7 7 7 7 9 9 9 9 12 12 12 12 8 8 F8 F8 F F K K S SK S SK - + - +

Figure 4-14. Caveolar disruption enhances D1DR mediated cAMP production. (A) D1DR transfected COS7 cells were pre-treated with vehicle (-mβCD) or with 2% mβCD for 30 min before cells were challenged with 0, 10 nM, 100 nM, 500 nM, 10 μM, or 100 μM SKF81297 for 20 min. The results are expressed as the mean % cAMP accumulation ± S.E.M. relative to maximum cAMP levels of wild-type D1DR in the absence of mβCD. (B) D1DR was co-expressed with increasing cDNA concentrations of cav1-P132L at D1DR:cav1-P132L transfection ratios of 1:1, 1:2, and 1:4 before cells were challenged with increasing concentrations of SKF81297 (10-8 to 10-4). The results are expressed as the mean % cAMP accumulation ± S.E.M. relative to maximum cAMP levels of wild-type D1DR in the absence of cav1-P132L (-P132L) (C) Membranes from D1DR transfected 35 cells were prepared and vehicle or SKF81297-stimulated [ S] GTPγS binding to Gs was determined. The results are expressed as the mean % [35S] GTPγS binding relative to basal binding levels in the absence of mβCD. Significance at p<0.05 versus % SKF81297-induced [35S] GTPγS binding in the absence of mβCD is denoted by *. 160 not due to a global dysregulation of adenylyl cyclase, [35S]GTPγS binding assays, which serve

as a measure of receptor mediated G protein activation, were performed in D1R-transfected and untransfected cells in the presence and absence of 2% mβCD. Stimulation of D1R-transfected

35 cells with 10 μM SKF81297 for 1 minute increased [ S]GTPγS binding to Gs by 47 ± 5.9 %

35 compared to basal levels (Figure 4-14C). In contrast, basal [ S]GTPγS binding to Gs from mβCD-treated D1R-transfected cells was over two-fold higher (114 ± 52.4 %) than in untreated cells. When membranes from these cholesterol depleted cells were incubated with 10 μM

SKF81297, no further increase in [35S]GTPγS binding was observed (basal, 114 ± 52.4 % over

mβCD untreated vs. + SKF81297, 113 ± 41.5 % over mβCD untreated). No changes in

[35S]GTPγS binding were observed in untransfected cells in the presence of mβCD (data not

shown). Collectively, these data suggest that caveolar disruption activates Gs maximally without

altering cAMP levels. This constitutive activation of Gs occurred independent of agonist binding

which was required for cAMP production.

4.7 Caveolin binding mutants exhibit constitutive desensitization

To further characterize the effect of disrupting the association of D1R with caveolae, we wished to determine whether the interaction with the caveolin-1 protein, itself, was essential for cAMP signaling. Contrary to the effects of cholesterol depletion on cAMP production, both the

F313A and W318A mutants displayed an equally small, but significant, increase in the EC50 for

cAMP generation compared to wild-type D1R (Figure 4-15A). The EC50 for cAMP generation

was determined to be 4.8 X 10-7 M, 1.8 X 10-6 M, and 2.3 X 10-6 M for wild-type D1R, F313A,

and W318A, respectively. Furthermore, the F313A mutant displayed an approximately 25 %

decrease in Vmax compared to wild-type D1R when values were corrected against the cell surface

161 D1 F313A D1 W318A D1 (agonist pre-treated)

125 125

100 100

75 75

50 50 (of WT receptor) (of WT receptor) 25 25 % maximum cAMPproduced % maximum cAMPproduced

0 0 -9 -8 -7 -6 -5 -4 -3 -9 -8 -7 -6 -5 -4 -3 log [SKF 81297] (M) log [SKF 81297] (M)

F313A W318A F313A (agonist pre-treated) W318A (agonist pre-treated)

125 125

100 100

75 75

50 50 (of F313A receptor) 25 receptor) (of W318A 25 % maximum cAMPproduced % maximum cAMPproduced

0 0 -9 -8 -7 -6 -5 -4 -3 -9 -8 -7 -6 -5 -4 -3 log [SKF 81297] (M) log [SKF 81297] (M)

Figure 4-15. (A) D1DR, F313A, or W318A transfected COS7 cells were challenged with 0, 10 nM, 100 nM, 500 nM, 10 μM, or 100 μM SKF81297 for 20 min to stimulate cAMP production. The results are expressed as the mean % cAMP accumulation ± S.E.M. relative to maximum cAMP levels of wild-type D1DR. Arrows indicate EC50. Significance at p<0.05 versus EC50 of D1DR is denoted by *. (B) Cells transfected with D1DR were pre-treated with 10 μM SKF81297 for 20 min before being challenged with increasing concentrations of SKF81297 (10-8 to 10-4). The desensitization profiles of F313A (C) and W318A (D) were also determined in a similar manner. The data are presented as the mean % cAMP accumulation ± S.E.M. relative to maximum cAMP levels of the receptor indicated in the absence of agonist pre-treatment. The results are expressed as the mean of 3 independent experiments. 162 receptor density. The W318A mutant did not show any difference in its capacity to maximally

stimulate cAMP. We tested the hypothesis that these receptor mutants might be constitutively

desensitized as a result of their diminished ability to efficiently produce cAMP. Stimulation of cells (pre-treated with 10 μM SKF 81297 for 20 min) expressing wild-type D1R shifted the EC50

to the right by over half a log unit while reducing the Vmax by 33 % (Figure 4-15B). However,

stimulation of agonist pre-treated cells independently expressing either F313A (Figure 4-15C) or

-6 W318A (Figure 4-15D) did not result in any significant changes in the EC50 (F313A, 1.29 X 10

-6 M; W318A, 1.65 X 10 M). Furthermore, while the Vmax of agonist pre-treated F313A was not

altered, the Vmax of pre-treated W318A was significantly attenuated by approximately 26 %. This

suggests that F313A was constitutively desensitized while W318A exhibited only partial

constitutive desensitization. Taken together, these results suggest that caveolin-1 binding to D1R

is required to inhibit constitutive desensitization, possibly by regulating auto-phosphorylation.

4.8 Discussion

In this section, we report that the D1 dopamine receptor interacts with caveolin-1 in

COS7 cells and in rat brain and undergoes agonist-induced endocytosis through caveolae, which

is dependent on the direct interaction of caveolin-1 with the D1 receptor. The stimulation of the

D1 receptor by the full agonist, SKF81297, triggered a robust translocation of receptor to

caveolin-enriched fractions and a concomitant decrease in expression in non-caveolin-enriched

fractions without any change in the distribution of caveolin-1 following agonist treatment. We

identified a caveolin binding motif in the proximal half of transmembrane domain 7 of the D1

receptor, which was shown to be critical for the interaction between the receptor and caveolin-1.

The disruption of this motif at specific amino acid residues was found to abrogate agonist-

163 induced receptor internalization and receptor maturation. We also demonstrate that the structural

integrity of caveolae appears to not only be important for modulating receptor turnover but also

for negatively constraining the D1 receptor’s basal and agonist mediated production of cAMP.

The purification of caveolae-enriched fractions from sodium carbonate based sucrose

gradients is largely based on the unique buoyant density of these lipid enriched domains.

Caveolae contain lipids that are packed tightly thus rendering them insoluble by non-ionic

detergents such as Triton X-100 at low temperatures. We used sodium carbonate as a detergent- free alternative to Triton in our preparations. This substitution confers an advantage over conventional detergents since caveolins are not soluble in sodium carbonate and unlike detergents, do not appear to result in the dissociation and loss of certain caveolin-associated

proteins (Chang et al., 1994; Song et al., 1996). Despite these advantages, this technique has

been limited by its inability to distinguish between caveolae and non-caveolar lipid rafts since

purification is based on enriched lipid content which is common to both domains. Hence, it has

been difficult to reconcile whether signalling molecules that co-localize in caveolin-positive

fractions actually reside in caveolae or other morphologically distinct lipid rafts. This

shortcoming has recently been addressed in studies from neonatal cardiomyocytes where

caveolar fractions were separated from non-caveolar buoyant fractions using a caveolin-3

specific antibody (Morris et al., 2006). It was found that these caveolar fractions contained

approximately five-fold greater expression of Gq and relevant PLCβ isoforms than non-caveolar

fractions. This supports the notion that most signalling molecules found in lipid-enriched

fractions are, indeed, concentrated in caveolae.

In order to establish a direct link between the D1 receptor and caveolin-1, we used co-

immunoprecipitation and BRET saturation assays to show that these two proteins exist in a

164 functional complex. We adapted the BRET assay for use in live cells to show that caveolin-1 and

the D1 receptor interacted in a specific and saturable manner. In whole rat brain, we detected a

selective co-precipitation of the α isoform of caveolin-1 by the D1 receptor implicating a physiological preference for the α isoform over the β isoform, which was not observed in COS7 cells (Figure 4-5). The α and β isoforms of caveolin-1 are derived from the same cDNA, though from different methionine start sites, thus rendering a full-length caveolin-1 isoform (α) and a truncated isoform (β) lacking the first 31 N-terminal amino acid residues. It is unlikely that these

N-terminal amino acids are critical for the interaction of caveolin-1 with the D1 receptor since both isoforms were co-precipitated in our heterologous cell expression assays. Instead, despite both isoforms being present in brain, it is possible that the α isoform is the only caveolin-1 isoform that is expressed in regions of the brain where the D1 receptor is expressed.

Alternatively, D1 receptors may selectively co-segregate with caveolin-1α if caveolin-1α

containing caveolae favour the sequestration of signalling components involved in D1 receptor

function. Indeed, there is evidence indicating that there are morphological differences between

caveolae populations enriched in caveolin-1α and caveolin-1β that may affect the targeting of

various proteins (Fujimoto et al., 2000; Nohe et al., 2005).

To further study the interaction with caveolin-1, we generated D1 receptor mutants in

which the putative CBM in TM7 was disrupted. This motif has been shown to be present in

most, if not all, proteins localized in caveolae (Couet et al., 1997). These include a large number

of molecules and membrane proteins associated with GPCR signalling pathways (Table 4-2). We

found that mutation of the proximal and medial aromatic acids in the CBM did not significantly

alter the pharmacological properties of the point mutants, F313A and W318A. However, there

was a significant reduction in the cell surface expression of the mutant, W321A, that did not

165 caveolin-1 binding motif

Φ X Φ XXXXΦ XXΦ

AT1R Y G F LGKKF GGY D1DR F DVFVW FGW

G protein-coupled β1AR F VFFNW LGY receptors M2 mAchR W TIGYW LCY Endothelin W P F DHNDF GVF Gα subunits F T F KDLHF KMF MAP kinase Y IVGFY GAF GRK-2 W QRRYF YQF PKC-α FSY VNPQF

Table 4-2. Presence of a putative caveolin-1 binding motif (φXφXXXXφXXφ) in a selection of G protein-coupled receptors (including the D1 dopamine receptor, D1DR) and associated signaling molecules. φ represents any aromatic amino acid whereas X represents any amino acid.

166 permit pharmacological characterization. Similarly, the FWW/A triple mutant was poorly

expressed; thus, we determined that in the D1 receptor, the distal aromatic amino acid of the

caveolin binding motif, W321, is likely a critical residue in maintaining proper receptor

expression since this mutation yielded a comparably low expression profile relative to FWW/A.

Interestingly, similar mutations in the CBM of the angiotensin II type 1 and glucagon-like

peptide 1 receptors also abrogated receptor plasma membrane expression suggesting that

caveolin-1 may act as a chaperone for exocytic transport of these receptors to the cell surface

(Leclerc et al., 2002; Syme et al., 2006; Wyse et al., 2003). Nevertheless, this is unlikely to be

the case for the D1 receptor as both F313A and W318A were unable to interact with caveolin-1,

as demonstrated by their inability to competitively disrupt the BRET signal between D1-Rluc

and cav1-GFP (Figure 4-13B), yet each expressed robustly at the plasma membrane (Figure 4-

12). However, consistent with the requirement for a physical association with caveolin, these

receptors were not able to internalize as efficiently as the wild-type D1 receptor, as determined

by whole cell radioligand binding analysis. This suggests that caveolar internalization of D1

receptors may require an interaction with caveolin-1.

Besides the classical endocytic pathway involving clathrin coated pits, other ligand

induced internalization pathways have been described that involve caveolae (Chini and Parenti,

2004) and other clathrin/caveolae independent mechanisms (Roseberry and Hosey, 2001). To

determine whether the agonist-induced targeting of the D1 receptor to caveolin-enriched sucrose

fractions could be attributed to caveolar endocytosis, D1 receptor-expressing cells were

challenged with various known inhibitors of caveolae function before stimulation with

SKF81297. Quantification of cell surface D1 receptors by radioligand binding analysis showed

that agonist-induced receptor internalization was significantly inhibited by pre-treatment with

167 methyl-β-cyclodextrin. The cholesterol-depleting effects of this treatment were specific to caveolae, since co-expression with dominant negative caveolin-1 P132L yielded a similar effect.

These results were strengthened by reversible cell-surface biotinylation studies in which filipin was also shown to suppress receptor internalization. The inhibition of clathrin-dependent endocytosis by hypertonic sucrose, concanavalin A, or the dominant negative G protein-coupled

receptor kinase 2 mutant, K220R, did not affect the extent of internalization suggesting that in

COS7 cells, this may not be the dominant mode of D1 receptor internalization. Although these

cells express low levels of arrestin and GRK2, the endogenous expression levels of these

proteins in COS7 cells are sufficient to facilitate arrestin and clathrin-mediated internalization of

various other GPCRs (Gaborik et al., 2001; Vrecl et al., 1998) indicating that both internalization

pathways are functional in this cell line. Therefore, the D1 dopamine receptor is fully capable of

internalizing through the clathrin coated pit pathway (Vickery and von Zastrow, 1999), as well as

through caveolae. The molecular determinants that control which endocytic route (clathrin vs.

caveolae) is taken, however, remain unclear. A previous study with the β1-adrenergic receptor

(β1AR), reported that clathrin mediated endocytosis was mediated by GRK phosphorylation

while caveolae-dependent endocytosis was controlled by PKA phosphorylation (Rapacciuolo et

al., 2003); both GRK- (Lamey et al., 2002) and PKA- (Mason et al., 2002) induced

phosphorylation have been shown to occur in the D1 receptor following agonist activation.

Furthermore, the ablation of putative PKA phosphorylation sites did not affect D1 receptor internalization in C6 glioma or NS20Y neuroblastoma cells (Jiang and Sibley, 1999; Mason et

al., 2002), the former of which has been a model cell line to study caveolae in glial cells

(Bhatnagar et al., 2004; Silva et al., 1999; Silva et al., 2005; Toki et al., 1999). These results are consistent with our data showing that the extent of D1 receptor internalization through caveolae

168 was not significantly affected by PKA inhibition (Figure 4-8). Hence, this suggests that other

PKA-independent mechanisms might dictate caveolae-mediated endocytosis. For instance, it has

been demonstrated that cholesterol oxidation can also act as an internalization switch between caveolae and clathrin coated pits (Okamoto et al., 2000).

The current notion that caveolae and other lipid raft microdomains function to organize

and integrate signalling complexes to facilitate signalling efficiency would predict that caveolae disassembly would impair cAMP signalling through activation of adenylyl cyclase. In a previous study, caveolin-2 was shown to be a requirement for agonist mediated cAMP production and that agonist stimulation promoted the preferential association of the D1 receptor with caveolin-2β

(Yu et al., 2004). In addition, in a separate study, dopamine was shown to recruit D1 receptors to caveolin-2 enriched plasma membrane fractions (Trivedi et al., 2004). These studies were performed in rat renal proximal tubule cells, which do not express caveolin-1, and therefore do not form functional caveolae, suggesting that a distinct subset of lipid rafts mediate these effects.

These findings demonstrate that there are unique roles ascribed to the different caveolin subtypes in modulating D1 receptor signalling. In our studies, we showed that while caveolae disruption by cholesterol depletion or over-expression with cav1-P132L did not alter the receptor’s affinity or the EC50 for cAMP generation, the Vmax was significantly enhanced. This was not consistent

with the dose-response profile exhibited by F313A and W318A, the former of which showed an

attenuated Vmax and both of which showed a decrease in the EC50 for agonist mediated cAMP

production. Although it is difficult to reconcile these differences, the effects of disrupting

caveolae may not be equivalent to disrupting caveolin interactions when characterizing the Gs- cAMP signaling pathway. Indeed, our results suggest that these mutants’ inabilities to interact with caveolin-1 reflects a constitutively desensitized state that may be a result of constitutive

169 phosphorylation of D1R by specific GRKs. The D1R has previously been shown to be

constitutively phosphorylated and desensitized by over-expression of GRK-4 (Rankin et al.,

2006) and caveolin has been shown to inhibit GRK activity (Carman et al., 1999). The disruption of these caveolin interactions at the level of the receptor or the kinase may release this tonic level of inhibition that results in constitutive phosphorylation and desensitization. This issue requires further study, however, as these caveolin binding mutants may simply be impaired in their ability to desensitize; basal phosphorylation assays may aid in clarifying this. Nevertheless, previous

reports have shown that agonist dependent cAMP production by the β2-adrenergic receptor in a sphingolipid-deficient (and hence, caveolae deficient) cell line is enhanced only at higher agonist concentrations (Miura et al., 2001), consistent with the effects we observed with mβCD and

over-expressed caveolin-1 P132L. This suggests that the lipid-enriched environment of

morphological caveolae has additional effects on D1R mediated signaling and that caveolae-

dependent signaling is not strictly defined by an interaction with caveolin-1, per se.

Unexpectedly, we determined that pre-treatment with mβCD was found to enhance basal GTPγS

binding to Gs without a parallel basal increase in cAMP production. Furthermore, while agonist

stimulation did not change GTPγS binding levels, cAMP production was significantly enhanced.

The increase in basal GTPγS binding upon caveolae disruption suggested that the inactive state

of the receptor/G-protein complex was constrained by its localization in caveolar domains.

Indeed, caveolin-1 has been shown to have a high affinity for the inactive GDP bound state of

Gsα and can inhibit its function by suppressing the rate of basal GDP/GTP exchange (Li et al.,

1995). As a result, methyl-β-cyclodextrin mediated disruption of caveolae may release the

inhibitory effect of caveolin-1 on Gsα, thus facilitating GTP binding. The agonist-induced

increase in cAMP, under caveolae-disrupting conditions, may be a result of an agonist-induced

170 conformational reorganization of a pre-existing D1R-Gsα complex (Dupre and Hebert, 2006;

Gales et al., 2006) that may facilitate opening of the Gsα interface to more efficiently interact

with adenylyl cyclase. Therefore, the G protein activation independent increase in cAMP may be

a consequence of this agonist dependent conformational change acting in concert with the

maximally activated G protein to further enhance cAMP. These data support

the notion that caveolae have an inhibitory role on G protein activation and effector signalling by

GPCRs specifically coupled to Gs or Gi/o (Bari et al., 2005; Rybin et al., 2000; Xiang et al., 2002;

Xu et al., 2006). Interestingly, β-adrenergic receptor (βAR) knockout studies from mouse

neonatal cardiomyocytes support this inhibitory role for caveolae in Gs-mediated signalling and also shed light on the physiological consequences of the differential localization of β1ARs and

β2ARs in the sarcolemma (Xiang et al., 2002). In brief, these studies showed that filipin treated

β1AR knockout myocytes exhibited enhanced Gs signalling and an increase in the agonist- induced rate of contraction whereas caveolar disruption by filipin did not alter the rate of contraction of β2AR knockout myocytes. This is consistent with the notion that β2ARs are localized in caveolae whereas β1ARs are spatially distributed in non-caveolar domains (Rybin et

al., 2000) thus owing to the differential regulation of these two receptors. Nevertheless, this

conclusion still remains a controversial topic. Other laboratories have shown that in cardiac

myocytes, the β1 adrenergic receptor (β1AR) resides in both caveolar and non-caveolar fractions

while the β2 adrenergic receptor (β2AR) resides exclusively in caveolar fractions under basal

conditions (Ostrom et al., 2001; Rybin et al., 2000). Following agonist stimulation, β2ARs

appear to activate adenylyl cyclase less efficiently than β1ARs. This was suggested to be due to

translocation of β2ARs out of caveolae and into clathrin coated pits as opposed to β1ARs which do not translocate and remain localized in caveolae where they can remain associated with

171 adenylyl cyclase (Ostrom et al., 2001). The notion that

caveolin proteins act as a scaffold to interact with signalling molecules to integrate signalling

platforms is a contradiction in its ability to dampen Gs activity (via stabilization of the GDP-

bound state) and inhibit the enzymatic activity of various proteins involved in this signaling

cascade (Carman et al., 1999; Razani et al., 1999; Toya et al., 1998). Despite this apparent

paradox, there is still a wealth of evidence indicating that caveolae function to enhance the signal

transduction efficiency of GPCRs traditionally known to be coupled to the Gq/11-phospholipase C

signaling pathway (Bhatnagar et al., 2004; de Weerd and Leeb-Lundberg, 1997; Harikumar et

al., 2005; Kifor et al., 1998; Navratil et al., 2003). Of note, the coupling of bradykinin B1 and B2 receptors (B1R and B2R) to PLC requires agonist-induced translocation from non-caveolar

regions of the plasma membrane to caveolin-rich domains, where both PLC and the substrate,

phosphatidylinositol are enriched (Pike and Casey, 1996; Pike and Miller, 1998). Subsequent to

2+ Ca mobilization and MAPK activation, B2R undergoes endocytosis to terminate signaling

whereas B1R signalling remains intact due to continued localization in caveolar membranes

(Sabourin et al., 2002). This differential regulation of GPCR subtypes by caveolin is similar to

the β1AR and β2AR in cardiomyocytes and underscores the functional importance of the spatial

organization of GPCR signalling complexes at the plasma membrane.

4.9 Acknowledgements

I would like to acknowledge Dr. Ahmed Hasbi for assistance with optimization of the

BRET assay as well as for obtaining the confocal images presented in Figure 4-4 and Figure 4-11

would also like to thank Michael Mattocks for assistance with the whole cell internalization

assays and Theresa Fan for performing the GTPγS binding assays presented in Figure 4-14C.

172 5 ROLE OF PALMITOYLATION IN D1 DOPAMINE RECEPTOR

INTERNALIZATION

5.1 Introduction

Protein-lipid interactions are clearly important for the lateral organization of integral

membrane proteins and for their membrane stability as has been described for the role of lipid rafts on GPCR function. While transmembrane interactions of such hydrophobic proteins within

the lipid bilayer depend on characteristics of the phospholipid environment, peripheral proteins

such as G proteins require other interactions to maintain their proximity to the plasma

membrane. Such cytosolic proteins depend on acylation by long-chain saturated fatty acids to enhance their association with the lipid bilayer. Acylation is a post-translational process that can

occur through one of three mechanisms: 1) N-myristoylation, 2) isoprenylation, or 3)

palmitoylation. Whereas myristoylation and isoprenylation are stable and permanent lipid

modifications, palmitoylation is a labile and reversible process which occurs through the

attachment of palmitate to cysteine via a thioester bond (el-Husseini Ael and Bredt, 2002). Most

GPCRs have evolved to undergo palmitoylation at one or more cysteine residues in the carboxyl

tail near the seventh transmembrane domain (Escriba et al., 2006). Our laboratory has previously shown that palmitoylation of the D1 dopamine receptor occurs at two cysteines at positions 347 and 351 (Jin et al., 1999). This is believed to have a structural role in anchoring the carboxyl tail to the lipid bilayer. In contrast, to date, myristoylation has not been reported in any GPCR whereas isoprenylation has only been shown to occur in the prostacyclin receptor (Miggin et al.,

2002).

The reversible nature of palmitoylation suggests that this modification acts as a

173 regulatory mechanism, similar to phosphorylation, in controlling receptor function. In some

GPCRs, agonist stimulation has been shown to affect palmitate turnover manifested as an

increase (Hayashi and Haga, 1997; Kennedy and Limbird, 1994; Mouillac et al., 1992;

Ponimaskin et al., 2001) or decrease (Sadeghi et al., 1997) in the degree of [3H] palmitate incorporation. There is no general consensus on the effect that this has on receptor activity as each GPCR appears to be distinctly affected by palmitoylation. However, it does implicate a requirement for cycles of palmitoylation/depalmitoylation to maintain a specific function. Site- directed mutagenesis of the palmitoylated cysteines in many GPCRs has been a widely employed strategy used to demonstrate a role for this modification in various receptor functions. For example, some studies have shown that permanently depalmitoylated receptors lose their ability to couple to their respective G protein and subsequent downstream effectors (Hayashi and Haga,

1997; Hukovic et al., 1998; O'Dowd et al., 1989). In some receptors, such as the β2-adrenergic receptor, this has been suggested to be due to enhanced phosphorylation of the mutant receptor, which enhances decoupling of the G protein (Moffett et al., 1993). Interestingly, in the β2AR, a putative PKA phosphorylation site has been shown to lie in close proximity to the palmitoylated cysteines (Moffett et al., 1996). This suggests that palmitoylation obstructs the access of kinases to target residues involved in desensitization and that agonist induced depalmitoylation (as reflected by palmitate turnover) may act as a switch for specific phosphorylation events (Moffett et al., 1996; Moffett et al., 1993). This type of regulation is not consistent with all GPCRs, however, and likely depends on the specific conformational requirements, unique to each receptor, controlling depalmitoylation and phosphorylation (Hawtin et al., 2001; Kraft et al.,

2001). Indeed, some GPCRs require palmitoylation to maintain an inactive conformation (in the absence of agonist) but do not need it for G protein coupling (Du et al., 2005; Ponimaskin et al.,

174 2002).

Although palmitoylation can be dynamic in nature, most GPCRs undergo constitutive

palmitoylation in the Golgi. Hence, it is the mature, palmitoylated form of the receptor that

undergoes dynamic regulation. This early palmitoylation event, shortly after biosynthesis, has

important implications for the cell surface trafficking of some GPCRs. Indeed, ablation of the

putative palmitoylaton sites renders some receptors trafficking deficient and intracellularly retained (Fukushima et al., 2001; Karnik et al., 1993; Percherancier et al., 2001; Zhu et al.,

1995). Nevertheless, it is unclear whether the retention of a depalmitoylated receptor is simply

due to misfolding due to the lack of cysteine mediated anchoring during the maturation process.

Hence, it is necessary to use enzymatic approaches of depalmitoylation to validate whether

palmitoylation is functionally important for trafficking.

5.1.1 Role of Palmitoylation in Association of Proteins with Caveolae

There is accumulating evidence that acylation by the attachment of myristic and palmitic

fatty acids is a targeting signal for proteins into lipid-enriched and detergent insoluble cellular

fractions (Chini and Parenti, 2004). The targeting of various proteins into these detergent

resistant membranes by these two modifications are defined by targeting signals that can involve

either tandem myristoylation and palmitoylation (Moffett et al., 2000; Shaul et al., 1996;

Shenoy-Scaria et al., 1994) or dual palmitoylation (Arni et al., 1998; Zhang et al., 1998). This

notion is strengthened by the fact that fusion of the cytosolic protein, GFP, with an acylation

consensus sequence was sufficient to target GFP to caveolin-enriched plasma membrane

domains (Galbiati et al., 1999). The mutation of both myristoylation and palmitoylation sites

caused a complete dissociation of GFP from these domains. Indeed, these studies were validated

175 by FRET showing that GFP-fused acylation consensus sequences were clustered with caveolin-1

at the plasma membrane (Zacharias et al., 2002). Although acylation events, such as

palmitoylation, may be required for lipid raft association for peripheral proteins, it is not clear

whether these requirements are conserved for integral membrane proteins. Hence, we were

interested in testing whether palmitoylation was required for caveolar localization of the D1

dopamine receptor.

5.2 Pharmacological Analysis of a Palmitoylation Deficient D1 Receptor

(C347A/C351A)

Through PCR-based site-directed mutagenesis methods, we mutated both palmitoylated

cysteines (C347, C351) in the carboxyl tail of the D1 receptor to alanine. To verify that this

receptor mutant retained a similar pharmacological profile to the wild-type receptor, we

performed saturation binding analysis. We found no significant differences in the binding

affinity (Kd) between the two receptors (Fig 5-1), consistent with previous studies (Jin et al.,

1999). However, the total expression level of the palmitoylation deficient mutant was approximately half that of the wild-type D1DR.

5.3 Kinetics of C347A/C351A Internalization

To functionally characterize the palmitoylation-deficient mutant receptor, we wanted to

determine whether there was a difference in the rate of internalization compared to wild-type

D1DR. Through whole cell binding, we measured the extent of internalization of C347A/C351A

over a 50 minute time period. Within 1.7 minutes of receptor stimulation with 10 μM SKF81297,

approximately 50 % of the cell surface population of C347A/C351A was internalized (Figure 5-

176 3

1 H]SCH23390 specific binding (pmol/mg membrane protein) [3 0 0 1 2 3 4 5 [3H - SCH23390] (pM)

HA-D1DR HA-C347A/C351A

BMAX 2.597 ± 0.07 1.267 ± 0.05

KD 0.4272 ± 0.05 0.3060 ± 0.05

Figure 5-1. [3H] SCH23390 saturation binding analysis of membrane preparations from COS7 cells expressing HA- tagged wild-type D1DR (■), or C347A/C351A (▲). Results shown are the mean of 3 independent experiments. The maximum receptor density (Bmax) and the dissociation constant (Kd) ± S.E.M. are shown in the inset table.

177 2). Near-maximum internalization was reached at 35 min where approximately 70 % of receptors

were internalized; the t1/2 for C347A/C351A internalization was 1.7 minutes. In contrast, the rate

of internalization (t1/2 – 12.9 ± 3.5 minutes) of wild-type D1DR was significantly slower requiring almost 40 minutes of continuous agonist stimulation for near-maximum receptor internalization. The magnitude of maximum internalization of D1DR was also less than

C347A/C351A (49.4 ± 12 %, D1DR vs. 75.2 ± 6.7 %, C347A/C351A) suggesting that palmitoylation negatively modulates the ability of D1DR to undergo agonist induced caveolar internalization.

5.4 Effect of Endocytosis Inhibitors on Internalization of C347A/C351A

The internalization kinetic data described in Section 5.3 suggests that the lack of palmitoylation facilitates the achievement of a specific receptor conformation that is more conducive to caveolar internalization upon agonist stimulation. Alternatively, we hypothesized that palmitoylation may have a role in acting as a switch between clathrin-dependent internalization and caveolin-dependent internalization. This supposition was based on studies with the endothelin receptor type A demonstrating that cholesterol oxidation switches the internalization pathway of this receptor from caveolae to clathrin (Okamoto et al., 2000). Protein palmitoylation has been shown to be a prerequisite for protein targeting to lipid rafts and cholesterol is an integral component of lipid rafts and caveolae (Chini and Parenti, 2004). Thus, we reasoned that cholesterol disruption (by oxidation) would prevent palmitoylated proteins from associating with cell surface raft-like domains, thus yielding the reported phenotype, and that protein de-palmitoylation would likely have a similar effect. Thus, we tested the hypothesis that de-palmitoylation of the D1 receptor might act as a regulatory switch between the clathrin-

178 internalization1/2 internalizationmax D1DR 12.9 ± 3.5 min 49.4 ± 12 %

C347A/C351A 1.7 ± 0.4 min 75.2 ± 6.7 %

80 HA-C347A/C351A 70

50 HA-D1DR

20 % receptor internalization % receptor 10

0 0 10 20 30 40 50 60 Time (minutes)

Figure 5-2. Whole cell radioligand binding analysis was performed with 2 nM [3H] SCH23390 on C347A/C351A-expressing cells exposed to 10 μM SKF81297 for the indicated time periods.

Non-linear regression analysis was performed and the half-time of internalization (t½) and the maximum internalization is indicated for C347A/C351A and compared with wild-type D1DR. Results are expressed as the mean % internalization ± S.E.M. of 3 independent experiments.

179 and caveolin-dependent internalization pathways. As described in the previous chapter, we used specific inhibitors of both clathrin and caveolin-dependent internalization pathways to explore this possibility. In whole cell binding assays, C347A/C351A transfected COS7 cells were pre- incubated with hypertonic sucrose, concanavalin A, or methyl-β−cyclodextrin. Whole cell surface binding of [3H] SCH23390 was compared before and after 30 minute incubation with 10

μM SKF81297 in the presence of these compounds. Notably, the magnitude of internalization at this time point (~63 %) was significantly higher than that observed with the wild-type D1 receptor (~52 %). Pre-treatment of cells with sucrose, concanavalin A, and H89 significantly attenuated the degree of C347A/C351A internalization (Figure 5-3) although these treatements did not completely abolish endocytosis. Unexpectedly, internalization of C347A/C351A was more resistant to cholesterol depletion with mβCD than wild-type D1DR. Indeed, caveolae disruption almost completely abolished D1DR endocytosis (~3 % internalization) (Figure 4-8) whereas C347A/C351A endocytosis was only significantly attenuated (~23 % internalization).

Taken together, this suggests that palmitoylation may play a role in switching the endocytic route of D1DR between the clathrin- and caveolin-dependent internalization pathways.

5.5 Subcellular Localization of C347A/C351A

Since palmitoylation has been shown to be involved in targeting various proteins to detergent resistant membrane fractions, we wanted to test whether this was an important determinant for caveolar localization of the D1 dopamine receptor. Similar to the wild-type

D1DR, we transiently transfected COS7 cells with the HA-tagged C347A/C351A mutant and prepared sodium carbonate-based sucrose gradient fractions from the whole cell lysates. When these fractions were analyzed for the subcellular distribution of C347A/C351A, a substantial

180 70 ψϕ 60 * lization 50

30 *

10 %interna C347A/C351A 0

CD H89 β m sucrose

C347A/C351A Concanavalin A

Figure 5-3. HA-C347A/C351A expressing cells were pre-treated with 0.45 M sucrose, 0.25 mg/ml concanavalin A, 2% methyl-β-cyclodextrin, or 30 μM H89 for 30 min prior to agonist stimulation with SKF81297 (10 μM) for an additional 30 min. Receptor density was estimated by whole cell radioligand binding analysis with 2 nM [3H] SCH23390. The results are expressed as the mean % internalization ± S.E.M. of 3 independent experiments. Significance at p<0.05, p<0.01, and p<0.001 versus % internalization under control conditions is denoted by φ, ψ, and * , respectively.

181 fraction of C347A/C351A was recovered in the caveolin-enriched fractions (fraction 5) with

some recovery in non-caveolin-enriched fractions (Figure 5-4). This was very similar to the

pattern of localization for wild type D1DR (Figure 4-3) and suggests that palmitoylation does not

have a role in the propensity of D1DR to localize in caveolin-enriched domains under basal

conditions.

5.6 Discussion

The results from this section do not implicate a role for palmitoylation in the targeting of

the D1 dopamine receptor to caveolar related lipid raft domains. No significant differences in the

degree of basal localization to caveolin-enriched sucrose gradient fractions were found between

the wild-type D1DR and C347A/C351A (Fig 5-4). This is in contrast to the requirement for

palmitoylation in the targeting of peripheral proteins and even some GPCRs (Lei et al., 2005) to

caveolae or lipid rafts. It seems likely that the hydrophobic nature of multi-helix membrane

spanning proteins, such as GPCRs, overcome any role that palmitoyl moieties may play in

targeting such proteins to lipid domains. Indeed, it has been shown that covalently linked palmitate chains to synthetic transmembrane peptides do not improve the affinity of these peptides for detergent insoluble lipid raft domains (van Duyl et al., 2002). Similarly, the caveolin-1 protein itself, which forms a hairpin loop in the lipid bilayer, can partition into detergent insoluble fractions in the absence of all three of its palmitoylatable cysteines (Dietzen et al., 1995). This strongly suggests that the mechanisms by which peripheral and integral membrane proteins attach to lipid rafts are distinct and that GPCRs, in general, do not utilize these highly conserved palmitoylated cysteines for raft association.

Although the D1DR does not require palmitoylation for agonist-independent localization

182 5 % 35 % 45 % sucrose (w/v)

fraction 1 2 3 4 5 6 7 8 9 10 11 12

HA (C347A/C351A) 52 kDa _

Figure 5-4. Localization of C347A/C351A in caveolin-1 enriched fractions. Detergent- free sucrose gradient fractions were prepared from C347A/C351A-transfected COS7 cells, separated on SDS-PAGE, and probed with an anti-HA antibody.

183 to caveolae, the results do indicate that caveolar internalization is modulated by this

modification. Based on the accelerated internalization kinetic data of C347A/C351A, we had

hypothesized that in the absence of palmitoylation, the D1DR would not be able to associate with

the lipid raft domains and hence, switch over to the clathrin-dependent endocytic pathway since

it was unable to internalize through the caveolin-dependent pathway. At first glance, this does

not appear to be the case as C347A/C351A was still targeted to the detergent insoluble fractions

and agonist-dependent internalization remained sensitive to cholesterol depletion (Fig 5-3).

However, the classical inhibition of the clathrin-mediated endocytic pathway, with hypertonic sucrose or concanavalin A, did significantly attenuate the extent of C347A/C351A internalization. Furthermore, C347A/C351A internalization appeared more resistant to caveolae

disruption (compared to wild-type D1DR, see Fig 4-8) as a larger proportion of these receptors

were sequestered in response to SKF8197 treatment. It is conceivable that this apparent

“resistance” was actually due to a switching of the palmitoylation-deficient mutant to a

cholesterol-insensitive clathrin-dependent internalization pathway. However, if this is the case, it

is unclear why the inhibitory effects of sucrose and concanvalin A were nor more marked

(though significant) and further, why mβCD still attenuated C347A/C351A internalization. One

possibility is that the components required for clathrin mediated internalization (eg: GRKs,

arrestins) are not abundant and therefore the extent to which proteins are sequestered through this

pathway is minimal. Therefore, any switching to a clathrin-dependent internalization pathway

would be limited and the majority of receptors may still, by default, undergo caveolar

internalization. In such an instance, the effects of clathrin inhibition would not appear to be

robust. Consistent with this notion, previous studies have shown that COS7 cells endogenously

express lower levels of specific kinase and arrestin subtypes than other cell lines such as HEK

184 and CHO cells (Menard et al., 1997) even though clathrin mediated internalization remains intact

(Gaborik et al., 2001; Vrecl et al., 1998). An alternative explanation to these observations would

be that palmitoylation simply has a minor role in determining which of these two internalization

pathways is utilized. Further studies will need to be carried out to define the precise role of

palmitoylation, if any, in determining the endocytic fate of a receptor.

We determined in the previous section that caveolar-mediated receptor internalization is a

slow process compared to clathrin-mediated internalization. We found that the C347A/C351A

mutant receptor exhibited a significantly greater rate of internalization than wild-type D1DR

with a half time of internalization that had an almost 3-fold difference (Figure 5-2). We

surmised that this might reflect a clear switch in endocytosis pathways as this accelerated rate of

internalization was very similar to that observed for clathrin-dependent D1DR internalization

(Vickery and von Zastrow, 1999). However, this was likely not the case here as receptor

internalization was independent of clathrin inhibition and still dependent on caveolar integrity. It

appears more likely that palmitoylation is required to maintain a specific conformation that

prevents the receptor from undergoing dysregulated internalization. For instance, the lack of

palmitate moieties in C347A/C351A might facilitate the exposure of residues to certain kinases

that render the receptor more amenable to hyperphosphorylation and internalization. This was

shown with the β2-adrenergic receptor in which elimination of a single palmitoylatable cysteine

in the carboxyl tail exposed nearby serine residues that were better substrates for PKA phosphorylation (Moffett et al., 1996; Moffett et al., 1993). This notion has also been suggested for a number of G protein-coupled receptors with similar mutations of the same conserved palmitoylatable cysteines (Munshi et al., 2005; Ponimaskin et al., 2005). Although there is evidence that PKA phosphorylation is required for caveolar internalization of GPCRs

185 (Rapacciuolo et al., 2003), it is not clear whether this regulates the rate of D1 dopamine receptor

internalization through caveolae. In contrast to the β2-adrenergic receptor, the D1DR does not

have PKA consensus sites nearby the palmitoylated cysteines although the introduction of such

sites in a C347A/C351A background does confer constitutive receptor desensitization (Jin et al.,

1999).

Alternatively, another site of phosphorylation that might be exposed due to the lack of

receptor palmitate moieties is on caveolin-1 itself. A number of kinases including PKC and

certain non-receptor tyrosine kinases such as Src have been suggested to regulate caveolar

internalization of various molecules (Dangoria et al., 1996; Parton et al., 1994). Both

palmitoylation sites are in close proximity to the caveolin-1 binding domain in transmembrane

domain 7 of D1DR. Hence, it is conceivable that palmitoylation may sterically hinder the ability

of certain kinases to phosphorylate caveolin-1 and that de-palmitoylation permits phosphorylation by these kinases and subsequent internalization.

Taken together, it is suggested that constitutive palmitoylation (in addition to transmembrane anchoring) may serve to stabilize D1DR during agonist-dependent caveolar internalization. The absence of palmitate moieties may allow the receptor to adopt an agonist- induced conformational state that permits access of specific kinases to target sites. Hence, further studies are required to determine whether a) phosphorylation has any role in mediating caveolar internalization of D1DR , and if so, b) to identify these sites.

186 6 RELATED STUDIES

6.1 Introduction

This section consists primarily of publications that are not directly related to the research

objectives of the thesis but do contain studies that were contributed by the thesis author. The

specific contributions by the thesis author to each study are provided.

6.2 D1 And D2 Dopamine Receptors Form Hetero-oligomers And Co-internalize After

Selective Activation Of Either Receptor

6.2.1 Summary

This report builds on a prior study in our laboratory that showed that D1 and D2

dopamine receptors, in addition to forming individual receptor homo-oligomers, also form

hetero-oligomers with novel signalling properties. This paper shows that selective activation of

either receptor could cross-phosphorylate the other and cause internalization of the heterooligomer. Furthermore, the interaction of these two receptors was shown to alter steady state levels of each receptor subtype suggesting that ER-derived assembly of these hetero-oligomers could change the trafficking properties of each receptor.

6.2.2 Contribution by thesis author

The author contributed significantly to the design and optimization of the fluorescence resonance energy transfer experiments in this publication. The data shown in Figure 1A as well

187 as the generation of the D1 receptor expressing stable cell line was directly performed by the thesis author.

188 0026-895X/05/6803-568–578$20.00 MOLECULAR PHARMACOLOGY Vol. 68, No. 3 Copyright © 2005 The American Society for Pharmacology and Experimental Therapeutics 12229/3046149 Mol Pharmacol 68:568–578, 2005 Printed in U.S.A.

D1 and D2 Dopamine Receptors Form Heterooligomers and Cointernalize after Selective Activation of Either Receptor

Christopher H. So, George Varghese, Kevin J. Curley, Michael M. C. Kong, Mohammed Alijaniaram, Xiaodong Ji, Tuan Nguyen, Brian F. O’Dowd, and Susan R. George From the Departments of Pharmacology and Medicine, University of Toronto, Toronto, Ontario, Canada, and the Centre for Addiction and Mental Health, Toronto, Ontario, Canada Received February 24, 2005; accepted May 27, 2005

ABSTRACT We provided evidence for the formation of a novel phospho- drobromide (SKF 81297) or the D2-selective agonist quinpirole. lipase C-mediated calcium signal arising from coactivation of The D2 receptor expressed alone did not internalize after acti- D1 and D2 dopamine receptors. In the present study, robust vation by quinpirole except when coexpressed with the D1 fluorescence resonance energy transfer showed that these re- receptor. D1-D2 receptor heterooligomerization resulted in an ceptors exist in close proximity indicative of D1-D2 receptor altered level of steady-state cell surface expression compared heterooligomerization. The close proximity of these receptors with D1 and D2 homooligomers, with increased D2 and de- within the heterooligomer allowed for cross-phosphorylation of creased D1 receptor cell surface density. Together, these the D2 receptor by selective activation of the D1 receptor. results demonstrated that D1 and D2 receptors formed het- D1-D2 receptor heterooligomers were internalized when the erooligomeric units with unique cell surface localization, inter- receptors were coactivated by dopamine or either receptor was nalization, and transactivation properties that are distinct from singly activated by the D1-selective agonist (Ϯ)-6-chloro-7,8- that of D1 and D2 receptor homooligomers. dihydroxy-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine hy-

The dopamine receptor family is subdivided into two dis- the same neurons through convergent postreceptor mecha- tinct subclasses, based on structural similarities, pharmaco- nisms. The latter mechanism of D1-D2 receptor synergism is logical profiles, and signal transduction mechanisms, into possible because dopamine receptor subclasses are colocal- D1- and D2-like receptors. Although D1 and D2 receptor ized in rat brain, with colocalization of D1- and D2-like re- subclasses are biochemically distinct in that D1 receptors ceptors in virtually every neuron in the neonatal striatum couple positively and the D2 receptors couple negatively to (Aizman et al., 2000). Furthermore, our own studies have adenylyl cyclase, many physiological functions are known to demonstrated robust colocalization of D1 and D2 receptors in be mediated by the coactivation of both receptors. For exam- a subset of neurons in human caudate nucleus, rat striatum, ple, the augmentative effect of cocaine on locomotion and and cortex (Lee et al., 2004). These data suggest, therefore, intracranial self-stimulation is mediated by the activation of that functional synergism could occur within individual neu- both D1 and D2 receptors (Kita et al., 1999). Dopamine rons. In fact, the coactivation of both D1 and D2 receptors has receptor synergism could occur either at the level of neuronal been shown to result in a significant increase in action po- networks through D1- and D2-like receptors expressed in tential frequency in neurons of the substantia nigra pars separate neuronal populations or, on the other hand, within reticulata (Waszczak et al., 2002) and a potentiation of arachidonic acid release in Chinese hamster ovary cells coexpressing both receptors (Piomelli et al., 1991). Our recent This work was supported by grants from the National Institute on Drug Abuse and the Canadian Institutes of Health Research. S.G. is the holder of a discovery of a common functional output generated by the Canada Research Chair in Molecular Neuroscience. concurrent coactivation of D1 and D2 receptors within the Article, publication date, and citation information can be found at http://molpharm.aspetjournals.org. same cells resulting in the activation of a novel phospho- doi:10.1124/mol.105.012229. lipase C-dependent calcium signaling pathway (Lee et al.,

ABBREVIATIONS: HEK, human embryonic kidney; HA, hemagglutinin; APC, allophycocyanin ; PBS, phosphate-buffered saline; SCH 23390, R-(ϩ)-7-chloro-8-hydroxy-3-methyl-1-phenyl-2,3,4,5-tetrahydro-1H-3-benzazepine; SKF 81297, (Ϯ)-6-chloro-7,8-dihydroxy-1-phenyl-2,3,4,5- tetrahydro-1H-3-benzazepine hydrobromide; AFU, absolute fluorescence unit; GRK2, G-protein-coupled receptor kinase; PKA, protein kinase A; GFP, green fluorescent protein; H89, N-[2-(4-bromocinnamylamino)ethyl]-5-isoquinoline; trFRET, time- resolved fluorescence resonance energy transfer; mRFP, monomeric red fluorescent protein; ER, endoplasmic reticulum. 568 189 Novel Trafficking of D1-D2 Dopamine Receptor Heterooligomers 569

2004), provides the long awaited putative biochemical mech- Transient Transfections. Cells were transiently transfected us- anism for dopamine receptor synergism (Pollack, 2004). ing Lipofectamine reagent (Invitrogen), and membranes were pre- These data, together with the demonstration that the recep- pared 48 h after transfection. To obtain equivalent cell surface ex- tors could be coimmunoprecipitated from cells and from brain pression levels of D1 and D2 receptors, the amount of D1 receptor (Lee et al., 2004), indicate that D1 and D2 receptor syner- cDNA was transfected in a 1:2 ratio in relation to the amount of D2 receptor cDNA. pcDNA3 vector was used to keep the total amount of gism, rather than resulting from a convergence in the signal- cDNA in each transient transfection constant. ing pathways distal to the receptors, may occur directly at Stable Transfections. HEK 293T cells were transfected as de- the receptor level. scribed above, and selection was performed in the presence of 200 ␮g G-protein coupled receptors have been demonstrated to of Zeocin (Invitrogen). Between 10 and 20 clones expressing varying exist as both homooligomers, forming complexes with identi- receptor densities were screened to select those with comparable cal receptors, and heterooligomers, forming complexes with expression levels. other types of receptors (George et al., 2002). Receptor hete- Time-Resolved Fluorescence Resonance Energy Transfer. rooligomerization is believed to account for the observed syn- This protocol was similar to that of McVey et al. (2001) and col- ergism between adenosine and dopamine receptors (Franco leagues with minor modifications. In brief, the donor and acceptor et al., 2000), as well as the ␮ and ␦ opioid receptors (George fluorophores were Europium chelate and allophycocyanin (APC), respectively (PerkinElmer Life and Analytical Sciences, Boston, et al., 2000), among others (George et al., 2002). Because MA). Europium chelate was conjugated to the anti-FLAG antibody. dopamine D1 and D2 receptors have been shown to form APC was conjugated to the anti-HA antibody. Each antibody was homooligomers (George et al., 1998; Lee et al., 2000), it is diluted in a solution of 50% phosphate-buffered saline (PBS)/50% possible that some of the functional synergism observed fetal bovine serum. Cells expressing HA-D1 and FLAG-D2 receptors within the dopamine receptor subclasses is mediated by a together or separately and mixed were incubated with these anti- physical interaction between receptors expressed in the same bodies for2hat37°C on a rotating wheel. After incubation, samples neuron, forming heterooligomers. Our observations of D1 and were pelleted at 5000 rpm, washed twice with PBS, and then resus- D2 receptors within the same signaling complexes by coim- pended in a final volume of 300 ␮l of PBS. One hundred microliters munoprecipitation from rat brain and heterologous cells co- of each sample was then divided into aliquots on a 96-well plate in triplicate. Fluorescence analysis was performed on a Victor2 multi- expressing both receptors (Lee et al., 2004) suggested the label plate reader (PerkinElmer Life and Analytical Sciences) with receptors may heterooligomerize. In this report, we demon- excitation at 340 nm and emission measured, after a 400-␮s delay, at strate definitively that D1 and D2 receptors existed as 615 nm (Eu3ϩ signal) and 665 nm (trFRET signal). Energy transfer heterooligomers at the cell surface. D1 and D2 receptor ϭ Ϫ (E) was calculated as E (EAD665/EAD615) (ED665/ED615), where conformations within the heterooligomer permitted cross- EAD665 and EAD615 represent emission at 665 and 615 nm, respec- phosphorylation of the D2 receptor by D1 receptor activation. tively, in the presence of both the donor and acceptor fluorophores

Furthermore, D1 and D2 receptor heterooligomerization re- and ED665 and ED615 represent the emission at 665 and 615 nm from sulted in D1 and D2 receptor cointernalization by selective samples containing the donor only. activation of either the D1 or D2 receptor within the hete- Confocal Microscopy. HEK-293T cells were transfected with rooligomeric complex and altered steady-state cellular distri- DNA encoding D1-GFP with or without D2-mRFP for 48 h. Live-cell confocal microscopy was performed with a microscope (LSM 510; bution of both receptors, with overall enhanced D2 and de- Carl Zeiss Inc., Thornwood, NY). Fifteen to twenty optimal sections creased D1 receptor cell surface expression. along the z-axis were acquired in increments of 0.1 ␮m. Figures showed the central image or, where indicated, z sections from the adherent surface through the nucleus. Images were acquired and Materials and Methods processed with Zeiss LSM Image Browser software. For immunocytochemistry, HEK 293T stable cell lines expressing D1-HA, D2- Cell Culture. All cell culture reagents, media, antibiotics, mam- FLAG, or coexpressing both receptors were fixed by 4% paraformal- malian expression vectors, and transfection reagents were obtained dehyde, permeabilized with 0.1% Triton X-100, and incubated with from Invitrogen (Carlsbad, CA). COS-7 and human embryonic kid- 3F10 anti-HA antibody (Roche Diagnostics Canada, Laval, QC, Can- ney (HEK) 293T cells (American Tissue Culture Collection, Manas- ada) and anti-FLAG antibody (Sigma-Aldrich Canada Ltd, Oakville, sas, VA) were maintained as monolayer cultures at 37°C. COS-7 cells ON, Canada). The antibody-labeled receptors were visualized by were maintained in ␣-minimum essential medium supplemented incubating permeabilized specimens in the presence of fluorescein with 10% fetal bovine serum. HEK 293T cells were maintained in isothiocyanate-conjugated anti-rat IgG or tetramethylrhodamine advanced minimum essential medium supplemented with antibiotic/ isothiocyanate-conjugated anti-rabbit IgG (Sigma). antimycotic (Invitrogen) and 6% fetal bovine serum. Membrane Preparations. Cells transiently or stably expressing Dopamine Receptor Constructs. For transient transfections, dopamine receptors were washed with PBS, resuspended in hypo- DNA encoding the human D1 receptor, the long and short isoforms of tonic lysis buffer with protease inhibitors (5 mM Tris-HCl, 2 mM the human D2 receptor, and the human D5 receptor were each EDTA, 5 ␮g/ml leupeptin, 10 ␮g/ml benzamide, and 5 ␮g/ml soybean inserted into the mammalian expression vector pcDNA3.1 (Invitro- trypsin inhibitor, pH 7.4), and homogenized with a Polytron appa- gen). For confocal microscopy, D1 and D2 receptors were cloned ratus (Brinkmann Instruments, Westbury, NY). The homogenate in-frame into the pEGFP-N1 vector (BD Biosciences Clontech, Palo was centrifuged to pellet unbroken cells and nuclei, and the super- Alto, CA) or the mRFP-1 vector. The cDNA encoding monomeric red natant was collected. The supernatant was centrifuged at 40,000g for fluorescent protein mRFP1 was a gift from Dr. Roger Tsien (Univer- 30 min to isolate a membrane fraction enriched in plasma mem- sity of California, San Diego, CA). brane, and the resulting pellet (P2 membranes) was washed and For generation of D1, D2 and D1-D2 receptor stable cell lines, resuspended in lysis buffer. hemagglutinin (HA) epitope-tagged D1 receptor cDNA and FLAG For subcellular fractionation studies, S1 supernatant was layered on top epitope-tagged D2 receptor cDNA were introduced alone or together of a sucrose density column, which then was subjected to centrifugation at into the pBUDCE4.1 expression vector (Invitrogen). All experiments 150,000g for 90 min at 4°C. Precipitates were collected at the 15%/30% were performed with the long isoform of the D2 receptor, except interface, consisting of the light vesicular membrane fractions, and at the where indicated. 30%/60% interface, consisting of mostly surface membrane (Toews, 2000). 190 570 So et al.

The receptor density in each fraction was quantified by radioligand binding that these receptors were assembled as heterooligomers in of1nM[3H]SCH 23390 for the D1 and 1 nM [3H]raclopride for the D2 intracellular compartments before being presented on the receptor and expressed as a percentage of the total number of receptors in cell surface. To verify the purity of the fractions, subcellular the entire cell, which was determined by adding the radioligand binding in markers were used. Although the ER marker calnexin and light and heavy membrane fractions of each individual experiment. the recycling endosomal marker Rab 11 were observed within Protein content was determined by the Bradford method (Bio-Rad, ϩ ϩ both fractions, the plasma membrane marker Na /K ATP Hercules, CA) with bovine serum albumin as the standard. SDS-Polyacrylamide Gel Electrophoresis and Immunoblotting. ase was exclusively localized in the heavy fraction (Fig. 1C). The procedures used for protein gel electrophoresis and immunoblotting To determine whether the formation of D1 and D2 receptor were identical to those described previously (Lee et al., 2000). heterooligomers resulted in a change in the conformation of Radioligand Binding. Saturation binding experiments were per- the binding pocket of either receptor, ligand-binding studies formed with 15 to 30 ␮g of membrane protein with increasing concentra- were performed using the ligands that were used previously tions of [3H]raclopride (final concentration, 250–10,000 pM) for estimation to demonstrate the novel calcium signal—dopamine, the D1- of the D2 receptor density or [3H]SCH 23390 (final concentration, 100-4000 specific agonist SKF 81297, and the D2-specific agonist quin- pM) for estimation of the D1 receptor density. Nonspecific binding was pirole (Lee et al., 2004). Competition of the D1-specific an- ␮ ϩ determined by 10 M( )-butaclamol. tagonist [3H]SCH 23390 binding with dopamine (D1R alone, Competition experiments were done in duplicate with increasing ϭ Ϯ Ϫ11 Ϫ3 Khigh 116 14 nM, fraction of receptors in high-affinity concentrations (10 –10 M) of unlabeled agonists. The concentra- ϭ Ϯ ϭ Ϯ tion of [3H]SCH 23390 used in the competition assays for the D1 state 0.15 0.04, Klow 4251 691 nM; D1R coexpressed ϭ Ϯ with D2R, Khigh 118.50 5.50 nM, fraction of receptors in receptor was approximately equivalent to its Kd (1 nM). The final concentration of [3H]raclopride used in the competition assays for the D2 receptors was 2 nM. Bound ligand was isolated by rapid filtration through a 48-well cell harvester (Brandel, Montreal, QC, Canada) using Whatman GF/C filters (Whatman, Clifton, NJ). Data were analyzed by nonlinear least-squares regression using Graph- Pad Prism. Data from multiple experiments were averaged and expressed as the means Ϯ S.E.M. The results were considered significantly different when the probability of randomly obtaining a mean difference was p Ͻ 0.05 using the paired Student’s t test. Cell Surface Immunofluorometry. HEK 293T and COS-7 cells were transfected and maintained for 48 h after transfection. Cells were plated in 96-well clear-bottomed plates (Corning Glassworks, Corning, NY) at a confluence of 50,000 cells per well. Cells were then fixed by 4% paraformaldehyde, blocked with 4% bovine serum albumin in PBS, incubated with 1:200 dilution of anti-HA (Roche) or 1:1000 dilution of anti-FLAG (Invitrogen) antibodies for 1 h, washed with PBS, and then labeled with a secondary antibody conjugated with fluorescein isothiocyanate (Invitrogen). Fluorescence was detected with a spectrophotometer (Cytofluor 4400; Applied Biosys- tems, Foster City, CA). Internalization studies on HEK 293T cells were used for internalization studies. Cells were treated 30 min with agonist before fixation and processed as described above. Agonist-Induced Phosphorylation. Phosphorylation assays were performed identically to those described previously (Lamey et al., 2002).

Results To gain some insight into the cell surface arrangement of D1 and D2 receptors within the calcium signaling complexes, trFRET was performed on whole cells, which coexpressed HA-D1 and FLAG-D2 receptors. Robust energy transfer between the antibodies bound to the N termini of D1 and D2 receptors was observed in cells cotransfected with both receptors (Fig. 1A, bar 1). Cells expressing D1 and D2 receptors singly and mixed together showed no trFRET (Fig. 1A, bar 2). Fig. 1. A close proximity of coexpressed D1 and D2 receptors on the cell Only cell surface receptors were detected by this method, surface was detected using time-resolved FRET with Europium chelate as the donor and APC as the acceptor (n ϭ 3 independent cotransfec- which suggested that the D1 and D2 receptors were within tions). A, Energy transfer on the cell surface was detected in cells co- proximity of 50 to 100 Å on the cell surface and most likely transfected with the amino-terminally tagged HA-D1 receptors (D1R) existed within the same oligomeric complex. To determine and amino-terminally tagged FLAG-D2 receptors (D2R) (bar 1) but not when these receptors were transfected separately and mixed (bar 2) (*, whether D1 and D2 receptor heterooligomers were preformed p Ͻ 0.05). B, energy transfer was detected in both the light and heavy within intracellular compartments, trFRET was performed membrane fractions obtained from cells cotransfected with HA-D1 and on light and heavy sucrose gradient fractions enriched in FLAG-D2 receptors. C, the plasma membrane marker Naϩ/Kϩ ATPase endoplasmic reticulum and plasma membranes, respectively. was detected in the heavy membrane fraction (lane 2) but not in the light membrane fraction (lane 1), whereas the endoplasmic reticulum marker Equivalent robust energy transfer was detected in both the calnexin and the recycling endosomal marker Rab11 was found in both heavy and light membrane fractions (Fig. 1B), suggesting fractions (lanes 1 and 2). 191 Novel Trafficking of D1-D2 Dopamine Receptor Heterooligomers 571

ϭ Ϯ ϭ Ϯ 3 ϭ Ϯ high-affinity state 0.17 0.06, Klow 3950 848 nM) or [ H]raclopride with dopamine (D2R alone, Khigh 10.30 ϭ Ϯ ϭ Ϯ SKF 81297 (D1R alone, Khigh 31.78 7.0 nM, fraction of 5.50 nM, fraction of receptors in high affinity state 0.10 ϭ Ϯ ϭ Ϯ ϭ Ϯ receptors in high-affinity state 0.89 0.06, Klow 2156 0.03, Klow 2478 639 nM; D2R coexpressed with D1R, ϭ Ϯ ϭ Ϯ 910 nM; D1R coexpressed with D2R, Khigh 35.30 8.4 nM, Khigh 12.53 6.13 nM, the fraction of receptors in high ϭ Ϯ ϭ Ϯ ϭ Ϯ fraction of receptors in high-affinity state 0.90 0.04, Klow affinity state 0.11 0.02, Klow 2112 818 nM) or ϭ Ϯ ϭ Ϯ 2097 926 nM) to membranes of COS-7 cells expressing quinpirole (D2R alone, Khigh 149 45 nM, fraction of ϭ Ϯ ϭ Ϯ the D1 receptor alone or coexpressing D1 and D2 receptors, receptors in high affinity state 0.14 0.02, Klow 3450 ϭ Ϯ demonstrated no significant change in the agonist affinities 766 nM; D2R coexpressed with D1R, Khigh 129 63 nM, ϭ Ϯ or the proportion of receptors detected in the agonist-de- fraction of receptors in high affinity state 0.19 0.03, Klow tected high- and low-affinity states (Fig. 2, A and B; p Ͼ ϭ 3691 Ϯ 740 nM) to membranes from COS-7 cells tran- 0.05). Saturation binding isotherm with [3H]SCH 23390 in- siently expressing the D2 receptor alone or coexpressing D1 dicated no significant change when D1 and D2 receptors were and D2 receptors also indicated no significant change in ϭ Ϯ coexpressed (D1R alone, Kd 0.64 0.12 nM; D1R coex- agonist affinities or the proportion of receptors detected in pressed with D2R, kd ϭ 0.79 Ϯ 0.15 nM, p Ͼ 0.05). Likewise, the high affinity state (Fig. 2, C and D, p Ͼ 0.05). Saturation competition of the binding of the D2-specific antagonist binding isotherm with [3H]raclopride indicated no significant

Fig. 2. Coexpression of D1 and D2 receptors does not alter agonist binding affinities and agonist-detected high- and low-affinity states (n ϭ 4–5). A, competition of [3H]SCH 23390 binding by dopamine in cell membranes prepared from cells expressing D1R (f) or coexpressing D1R and D2R (ƒ). B, competition of [3H]SCH 23390 binding by the D1-specific agonist SKF 81297 in cell membranes prepared from cells expressing D1R (f) or coexpressing D1R and D2R (ƒ). C, competition of [3H]raclopride binding by dopamine in cell membranes prepared from cells expressing D2R (F) or coexpressing D1R and D2R (‚). D, competition of [3H]raclopride binding by the D2 selective agonist quinpirole in cell membranes prepared from cells expressing D2R (F) or coexpressing D1R and D2R (‚). Black arrows indicate high- and low-affinity sites for receptor expressed alone; gray arrows indicate high- and low-affinity sites for receptors coexpressed. 192 572 So et al. change when D1 and D2 receptors were coexpressed (D2R protein kinase A mediated by the occupancy of the D1 recep- alone, kd ϭ 0.87 Ϯ 0.06 nM, D2R coexpressed with D1R, kd ϭ tor (data not shown). 0.88 Ϯ 0.06 nM, p Ͼ 0.05). Because D1 and D2 receptors formed heterooligomers at Because D1 and D2 receptors form heterooligomers, we the cell surface, we investigated if the close interaction was queried if the configuration of the receptors within the com- maintained during agonist-mediated receptor internaliza- plex would allow for cross-modulation of receptor function, tion. To determine whether the D1-D2 receptor complex was such that selective activation of the D1 receptor could result internalized upon coactivation of both receptors, HEK 293T in the phosphorylation of the D2 receptor in the heterooligo- cells transiently expressing HA-D1, FLAG-D2 or both recep- meric complex. We have shown previously that the mutation tors were subjected to cell surface immunofluorescence as- of a glutamic acid residue at position 359 to alanine in a says (Fig. 4). After 10 ␮M dopamine treatment for 30 min, putative G-protein coupled receptor 2 kinase (GRK2) recog- internalization of the D1 receptor was observed when ex- nition site on the carboxyl tail of the D1 dopamine receptor pressed alone or coexpressed (16.7 Ϯ 8.7%, n ϭ 4 for D1R; (D1-E359A) completely attenuated the rapid agonist-induced 17.4 Ϯ 4.0%, n ϭ 4 for D1R with D2R, p Ͼ 0.05,) (Fig. 4a). The phosphorylation of the D1 receptor and the ability of the D2 receptor did not internalize on exposure to dopamine receptor to undergo homologous desensitization, without ef- when expressed alone but did internalize when coexpressed fect on adenylyl cyclase activation (Lamey et al., 2002) (Fig. with the D1 receptor (1.1 Ϯ 2.4%, n ϭ 4 for D2R alone; 15.1 Ϯ 3A, lane 1 and 2). When the D2 receptor was expressed alone 1.8%, n ϭ 4 for D2R with D1R, p Ͻ 0.05) (Fig. 4b). To confirm and treated with 1 nM SKF 81297, a concentration of the D1 our results obtained by transient transfection, HEK 293T selective agonist which did not activate the D2 receptor and cells stably expressing HA-D1 (1559 Ϯ 446 fmol/mg), has been shown previously to not activate the robust calcium FLAG-D2 (748 Ϯ 199 fmol/mg) and coexpressing HA-D1 and signal (Lee et al., 2004), no significant agonist-induced phos- FLAG-D2 (D1-HA ϭ 1762 Ϯ 361 fmol/mg, D2-FLAG ϭ phorylation of the receptor was observed (Fig. 3A, lane 4). 1877 Ϯ 434 fmol/mg) receptors were treated with 10 ␮M When the D2 receptor was coexpressed with D1-E359A and dopamine for 30 min after which the number of the receptor treated with 1 nM SKF 81297, phosphorylation of the D2 binding sites in P2 membrane preparations were quantified receptor was observed (Fig. 3A, lane 6). Figure 3B demon- by radioligand binding and the results observed was similar strates that the amount of receptor protein in each lane was to that observed in cells transiently expressing the receptors approximately equivalent, demonstrating that the increase (17 Ϯ 7.5%, n ϭ 4 for D1R alone; 16 Ϯ 4.0%, n ϭ 4 for D1R in phosphate accumulation did not result from changes in D2 with D2R, p Ͼ 0.05; for the D2R, Ϫ2.2 Ϯ 1.7%, n ϭ 4 for D2R receptor expression. Phosphorylation of the D2 receptor by alone; 21 Ϯ 4.0%, n ϭ 4 for D2R with D1R, p Ͻ 0.05). D1 receptor activation was also observed in the presence of Because D1 receptor activation within the heterooligomer the PKA inhibitor H89, suggesting that the mechanism of D2 allowed for cross-modulation of D2 receptor function, the phosphorylation was not dependent upon the activation of effect of selective agonists on the internalization of the hete-

Fig. 3. Phosphorylation of the D2 receptor by selective activation of the D1 receptor (representative of n ϭ 3). A, cells expressing either HA- E359A-D1 (lanes 1 and 2) or c-myc-D2 receptor alone (lanes 3 and 4) or coexpressing both receptors (lanes 5 and 6) were treated with 1 nM SKF 81297. 32P incorporation was detected after immunoprecipitation by an- Fig. 4. Cointernalization of D1 and D2 receptors by 10 ␮M dopamine in ti-HA (lanes 1 and 2) or anti-c-myc (lanes 3–6) antibodies. Arrows indi- HEK 293T cells (n ϭ 4). A, internalization of D1 receptor in cells express- cate phosphorylated receptor monomers and oligomers. B, immunoblot ing the D1 receptor alone or coexpressed with the D2 receptor. B, inter- detection of E359A-HA-D1 (lanes 1 and 2) and c-myc-D2 (lanes 3–6) nalization of the D2 receptor in cells expressing the D2 receptor alone or receptors by anti-HA or anti-c-myc antibodies. The immunoprecipitated coexpressed with the D1 receptor (*, p Ͻ 0.05). Percentage internalization samples were subjected to immunoblotting to demonstrate that the represents the loss of fluorescence from the cell surface of the whole cell amount of receptor in each lane was approximately equivalent. after agonist treatment compared with the vehicle-treated control. 193 Novel Trafficking of D1-D2 Dopamine Receptor Heterooligomers 573 rooligomer was tested. To quantify this effect, HEK 293T vated D1 receptor. Concentration-dependent internalization cells expressing either one or both HA-D1 and FLAG-D2 of the D1 receptor in response to the D2 selective agonist receptors were treated for 30 min with varying concentra- quinpirole was observed upon coexpression with the D2 re- Ϯ ϭ Ϯ ϭ tions of SKF 81297 or quinpirole and cell surface expression ceptor (31 7.7% internalized; EC50 13.2 5.4 nM, n 4 of each receptor was quantified by cell surface immunofluo- for D1R with D2R) (Fig. 5D). No internalization of the D1 rescence assays. Dose-dependent internalization of the D1 receptor was observed in response to quinpirole when ex- receptor was detected after 30-min treatment with increas- pressed alone (Fig. 5D). ing concentrations of SKF 81297 when the D1 receptor was Because the D1-D2 receptor heterooligomers function as a expressed alone or coexpressed with the D2 receptor (19 Ϯ unit at the cell surface and internalize as such, the steady- 5.8%, n ϭ 4 for the D1R alone; 23 Ϯ 8.1%, n ϭ 4 for the D1R state cellular distribution of the receptors was investigated. with D2R, p Ͼ 0.05) with a significant decrease in the po- The subcellular localization of the transiently transfected tency of the D1 agonist to result in the internalization of the D1-GFP and D2-mRFP receptors in cells expressing one or ϭ Ϯ ϭ D1-D2 receptor heterooligomer (EC50 13 7.7 nM, n 4 both receptors was examined by confocal microscopy, with ϭ Ϯ ϭ for the D1R alone; EC50 45 4.0 nM, n 4 for D1R with receptor distribution displayed by serial z-sections through D2R, p Ͻ 0.05) (Fig. 5A). To observe if the D2 receptor HEK 293T cells (Fig. 6). When expressed alone, D1 receptors cointernalized with the D1 receptor upon selective D1 recep- were found predominantly on the cell surface (Fig. 6A), tor activation, the concentration-dependent response to the whereas D2 receptors were localized at the cell surface with D1 selective agonist SKF 81297 was analyzed. Whereas no a significant proportion present intracellularly (Fig. 6B). internalization of the D2 receptor was observed in response When coexpressed, D1 and D2 receptors were observed to to SKF 81297 when expressed alone, internalization of the have a similar distribution on the plasma membrane as well D2 receptor was observed when coexpressed with the D1 as in intracellular compartments (Fig. 6C) suggesting a com- Ϯ ϭ Ϯ receptor (Fig. 5B) (13 3.9% internalized; EC50 12 6.0 mon localization of D1 and D2 receptors, probably stemming nM, n ϭ 3 for D2R with D1R). from the formation of D1-D2 heterooligomers. This distribu- Dose-dependent internalization of the D2 receptor was not tion was also observed when receptors were coexpressed in detected after 30-min treatment with increasing concentra- COS-7 cells and in the neuroblastoma cell line N1E 115 (data tions of quinpirole when D2 receptor was expressed alone. not shown). Similar localization of D1 and D2 receptors was However, on coexpression with the D1 receptor, internaliza- also observed in stable cell lines expressing the receptors tion of the D2 receptor was observed in response to quinpirole alone or coexpressing both receptors (Fig. 6, D–F), indicating Ϯ ϭ Ϯ ϭ (Fig. 5C) (14 4.8% internalized; EC50 24.5 5.7 nM, n that these observations were due to an alteration in the 4 for D2R with D1R). Because D1 and D2 receptors physically steady-state distribution of the receptors upon coexpression interact, it may be possible that the D1 receptor would coin- and not due to transient transfection. ternalize with the D2 receptor, similar to what was observed To further assess the cellular redistribution of the receptors for the D2 receptor when coexpressed with the agonist acti- upon coexpression, radioligand binding was performed on heavy

Fig. 5. Cointernalization of D1 and D2 receptors by activation of either receptor was detected by cell surface immunofluorometry in HEK 293T cells (n ϭ 3–4). A, internalization of the D1 receptor by SKF 81297 in cells expressing the HA-D1 receptor alone or coexpressed with the FLAG-D2 receptor. B, internalization of the D2 receptor by SKF 81297 in cells expressing the FLAG-D2 receptor alone or coexpressed with the HA-D1 receptor. C, internalization of the D2 receptor by quinpirole in cells expressing FLAG-D2 receptor alone or coexpressed with HA-D1 receptor. D, internalization of D1 receptor by quinpirole in cells expressing HA-D1 receptor or coexpressed with the FLAG-D2 receptor. Cells were treated for 30 min with variable concentrations of agonists. Ar-

rows indicate EC50. Percentage internalization represents the loss of fluorescence from the cell surface of the whole cell after agonist treatment compared with the vehicle- treated control.

194 574 So et al. and light subcellular membrane fractions obtained using a discon- increase in the proportion of the D2 receptors in the heavy mem- tinuous sucrose gradient, which contained distinct cellular mark- brane fraction was observed (47.8 Ϯ 3.2% for D2R alone; 66 Ϯ 5.3% ers (Fig. 1C). Expressed alone, the D1 receptor was detected pre- for D2R with D1R, p Ͻ 0.05) with a decrease observed in the light dominantly in the heavy membrane fraction. Upon coexpression fraction (51.4 Ϯ 3.4% for D2R alone; 32.6 Ϯ 5.4% for D2R with with the D2 receptor, an increased proportion of the D1 receptor D1R, p Ͻ 0.05) (Fig. 7B). These results supported the observations was detected in the light membrane fraction (Fig. 7A) (40 Ϯ 4% for from confocal microscopy indicating changes in receptor localiza- D1R alone; 64 Ϯ 5.9% for D1R with D2R, p Ͻ 0.05) with a decrease tion upon coexpression. It is interesting that the redistribution of observed in the heavy membrane fraction (68 Ϯ 4.6% for D1R the receptors in cotransfected cells changed the overall receptor alone; 34 Ϯ 5.9% for D1R with D2R, p Ͻ 0.05). When the D2 density of both receptors. When total receptor expression was receptor was expressed alone, the receptor was detected to be quantified, both by radioligand binding on whole-cell lysates and approximately equally distributed in both the heavy and light from the addition of the number of binding sites in heavy and light membrane fractions. Upon coexpression with the D1 receptor, an sucrose gradient fractions, the number of D2 receptor binding sites

Fig. 6. Confocal Microscopy of cells expressing D1 and D2 receptors or coexpressing both receptors. Shown are confocal images of HEK 293T cells transfected with D1-GFP alone (A), D2-mRFP (B), or coexpressing both receptors (C). Serial z-sections from the cell surface through the nucleus are shown for the region indicated by the dashed white square. C, D1-GFP corresponds to top and D2-mRFP corresponds to bottom. D–F, confocal image of HEK 293T stable cell lines expressing D1-HA alone (D), D2-FLAG alone (E) or coexpressing both receptors (F) in HEK 293T stable cell lines. 195 Novel Trafficking of D1-D2 Dopamine Receptor Heterooligomers 575 was increased when coexpressed with the D1 receptor, and the which lacks 26 amino acids in the third intracellular loop, the number of D1 receptor binding sites was decreased when coex- D2(short) receptor was coexpressed with the D1 receptor in COS-7 pressed with D2 receptor (data not shown). cells. A significant 50% (P Ͻ 0.05) increase in the number of To quantify changes at the cell surface, COS-7 cells were co- D2(short) receptor binding sites was observed when coexpressed transfected with D1 and D2 receptors, and the surface expression with the D1 receptor compared with when it was expressed alone ϭ Ϯ ϭ ϭ of the receptors was estimated by cell surface immunofluorescence (Bmax 1412 237 fmol/mg, n 5 for D2R(short) alone; Bmax assays. A 51% increase in cell surface expression of the D2 receptor 2578 Ϯ 208 fmol/mg, n ϭ 3 for D2R(short) with D1R, p Ͻ 0.05) 3 was documented upon coexpression with the D1 receptor with no change in the affinity of the receptor for [ H]raclopride (Kd Ϯ Ϯ ϭ Ϯ ϭ ϭ Ϯ (11,416 525 AFU for D2R, 22,009 3245 AFU for D2R with 1.2 0.25 nM, n 5 for D2R(short) alone; Kd 1 0.12 nM, D1R, p Ͻ 0.05). (Fig. 8A). Binding of [3H]raclopride in the plasma n ϭ 3 for D2R(short) with D1R, p Ͼ 0.05) (Fig. 8C). This increase membrane enriched P2 fraction also showed a significant 50% (p Ͻ in D2 receptor-binding sites also correlated with the increase in 0.05) increase in the number of D2 receptor binding sites when receptor protein species visualized by Western blot (Fig. 8C). No coexpressed with the D1 receptor compared with when the D2 enhancement of the D2 receptor was observed when coexpressed ϭ Ϯ ϭ receptor was expressed alone (Bmax 876 112 fmol/mg, n 5 with the D5 receptor (data not shown). ϭ Ϯ ϭ for D2R alone; Bmax 1600 164 fmol/mg, n 5 for D2R with To examine whether cell surface expression of the D1 receptor D1R, p Ͻ 0.05) (Fig. 8B). This increase in the number of D2 was altered in parallel with the expression of the D2 receptor when receptor-binding sites also correlated with the increase in receptor coexpressed, the number of D1 receptors was quantified by cell protein species as observed by Western blot (Fig. 8B). To deter- surface immunofluorescence assays. Decreased cell surface ex- mine whether the enhancement upon coexpression with the D1 pression of the D1 receptor when coexpressed with the D2 receptor receptor was also observed for the short isoform of the D2 receptor, was quantified (29,014 Ϯ 4995 AFU for D1R alone; 14,653 Ϯ 2464 AFU for D1R with D2R, p Ͻ 0.05). By radioligand binding studies, the number of D1 receptor binding sites within the plasma-enriched P2 membrane fractions was decreased by 55% (p Ͻ 0.05) ϭ Ϯ when coexpressed with the D2 receptor (Bmax 3156 415 ϭ ϭ Ϯ ϭ fmol/mg, n 10 for D1R alone; Bmax 1447 375 fmol/mg, n 10 for D1R with D2R, p Ͻ 0.05) (Fig. 8E). This decrease in D1 receptor binding sites corresponded to a decrease in the amount of receptor in the P2 fraction as verified by Western blot (Fig. 8E).

Discussion Our discovery of the phospholipase C-mediated calcium signal generated by the coactivation of D1 and D2 dopamine receptors (Lee et al., 2004) has provided a novel insight into the unique ability of these coactivated receptors to access a completely new signaling pathway. In this report, we definitively demonstrate that D1 and D2 receptors heterooligomerize within these signaling complexes. FRET analysis demonstrated that within these complexes, D1 and D2 receptors exist as heterooligomers on the cell surface and within intracellular compartments. As a result of heterooligomerization, these receptors displayed altered steady state cellular distribution and unique internalization kinetics compared with D1 and D2 receptor homooligomers. A unique characteristic of the heterooligomer was that it enabled robust internalization of both receptors in response to a D2-specific agonist, whereas the D2 receptor expressed alone did not internalize in response to the same agonist. The close proximity of the receptors within the heterooligomer enabled cross-phosphorylation of the D2 receptor by selective activation of the D1 receptor. FRET analysis demonstrated that the D1 and D2 receptors Fig. 7. The distribution of the D1 and D2 receptors in heavy (plasma were in proximity of 50–100 Å of each other at the cell membrane) and light (vesicular intracellular) membrane fractions was changed upon coexpression in COS-7 cells. A, distribution of D1 receptor surface and within intracellular compartments. This obser- in heavy and light fractions in cells expressing D1 receptor alone or vation indicated that the D1 and D2 receptors existed as coexpressed with D2 receptor determined by specific binding of 1 nM heterooligomers, which may have been formed within the ER [3H]SCH 23390 (n ϭ 5–7). B, distribution of D2 receptor in heavy and light membrane fractions in cells expressing D2 receptor alone or coex- before being trafficked to the cell surface. This has been pressed with the D1 receptor determined by specific binding of 1 nM suggested previously by other studies on G-protein-coupled [3H]raclopride (n ϭ 4). Significant differences between the D1 and D2 receptor homo- and heterooligomerization (Canals et al., receptor distribution in singly and doubly transfected cells was denoted 2003; Terrillon et al., 2003). However, based on the presence by * for the light vesicular membrane fractions and # for heavy membrane fractions (p Ͻ 0.05). The data were represented as the average of the recycling endosomal marker Rab 11 in the light mem- percentage expression of the total radioligand binding Ϯ S.E.M. brane fraction, it may be possible that heterooligomers in the 196 576 So et al. light fraction will also include those within recycling endo- the phosphorylation of the D2 receptor was not dependent somes. The formation of heterooligomers, however, did not upon PKA activation. change the affinity of the D1 and D2 receptors to the selected The conformation of D1-D2 receptor heterooligomers may prototypical agonists and antagonists used previously to gen- also have allowed for novel internalization characteristics in erate the novel calcium signal, indicating that the ligand response to agonist treatment. The internalization studies re- binding pocket for these drugs was unchanged by heterooli- vealed that the D1-D2 receptor heterooligomer was internalized gomer formation. by coactivation of both receptors by dopamine or by the single Despite no observable changes in the affinity of selected activation of one of the receptors within the heterooligomer. agonists and antagonists to the ligand-binding pocket, hete- D1-D2 receptor heterooligomers internalized when the D1 re- rooligomer formation enabled cross-modulation of receptor ceptor was activated by SKF 81297 or the D2 receptor was activity by phosphorylation. The activation of the D1 receptor activated by quinpirole. These results suggested that the inter- by SKF 81297 resulted in cross-phosphorylation of the D2 actions between the receptors within the heterooligomer were receptor. The phosphorylation of the D2 receptor probably maintained upon agonist exposure, unlike other examples in occurred because in the heterooligomeric conformation, as which one receptor within the heterooligomer internalizes sep- suggested by the generation of the novel calcium signal upon arately upon agonist activation (Xu et al., 2003). coactivation of both receptors and the novel pattern of inter- The D2 selective agonist quinpirole, which failed to internalization of the D2 receptor, putative phosphorylation sites nalize the D2 receptor when expressed alone, internalized located in the third intracellular loop of the D2 receptor may the D1-D2 receptor heterooligomer. This is a remarkable have been located in close enough proximity to the D1 recep- finding considering that the characteristics of D2 receptor tor to allow for phosphorylation by kinases such as GRKs. internalization depends on the cell type and may be related The PKA inhibitor H89 had no effect on the D1-receptor– to the variable expression of endogenous GRKs and ␤-ar- mediated D2 receptor phosphorylation, thus suggesting that restins (Zhang et al., 1994; Ito et al., 1999; Kim et al., 2004)

Fig. 8. Membrane expression of D2 receptor was enhanced by coexpression with the D1 receptor in COS-7 cells (n ϭ 3–5). A, cell surface expression of the D2 receptor expressed alone or coexpressed with the D1 receptor as quantified by cell surface immunofluorescence (*, p Ͻ 0.05). B, saturation binding of [3H]raclopride to D2 receptor expressed alone or coexpressed with D1 receptor. Immunoblot detecting FLAG-D2 receptor in P2 membranes from cells expressing the D2 receptor alone or coexpressed. C, saturation binding of [3H]raclopride to D2 receptor (short) expressed alone or coexpressed with D1 receptor. Immunoblot detecting FLAG-D2 receptor (short) expressed in P2 membranes in cells expressing D2 receptor (short) alone or coexpressed. D, cell surface expression of the D1 receptor expressed alone or coexpressed with the D2 receptor as quantified by cell surface immunofluorescence (*, p Ͻ 0.05). E, saturation binding of [3H]SCH 23390 to D1 receptor expressed alone or coexpressed with D2 receptor. Immunoblot detection of the HA-D1 receptor expressed on P2 membranes in cells expressing HA-D1 receptor alone or coexpressed. For immunoblots, arrows indicate monomer, dimer, and higher order of oligomeric species. 197 Novel Trafficking of D1-D2 Dopamine Receptor Heterooligomers 577 and has also been reported to up-regulate on the cell surface signal and the novel pattern of agonist-induced D2 receptor after long-term agonist exposure (Zhang et al., 1994; Ng et internalization. Furthermore, as reported within our previ- al., 1997). In vivo animal experiments, however, demon- ous article (Lee et al., 2004), the efficiency of coimmunopre- strated that the D2 receptor readily internalized in brain in cipitation of D1 and D2 receptors in cell lines and rat brains response to endogenous dopamine (Chugani et al., 1988; Sun by selective antibodies was high, indicating that a significant et al., 2003). Because D1 and D2 receptors are colocalized in proportion of receptors existed as hetero-oligomers. The re- many neurons in human and rat brain (Aizman et al., 2000; ceptors within the heterooligomeric complex have attained Lee et al., 2004), it is possible that the internalized D2 novel properties distinct from the individual D1 and D2 ho- receptor detected was colocalized with the D1 receptor. Oli- mooligomers, possibly stemming from novel receptor confor- gomerization with the D1 receptor may either put the D2 mations within these heterooligomers. The many lines of receptor into a conformation that enabled the D2 receptor to evidence we have generated all point to the creation of a recruit endocytic machinery on its own or enable the D2 novel D1-D2 signaling unit, with ability to internalize after receptor to access the endocytic processes linked to the D1 selective activation of either D1 or D2 receptors but requiring receptor. The latter is a possibility because activation of D1 concurrent activation of both receptors for generation of the receptors resulted in the phosphorylation of the D2 receptor. calcium signal. The enhanced ability of the D2 agonist to Another possibility is that the novel entity of the D1 and D2 internalize the D1-D2 heteromeric complex also signifies im- receptor heterooligomer, because of the new receptor confor- portant ramifications for the understanding of the action of mations achieved, now has different internalization charac- these clinically relevant drugs in brain. It has been proposed teristics compared with that of the D1 and D2 receptor ho- that abnormal calcium signaling may constitute the central mooligomers. This possibility is indicated by the reduced dysfunction in schizophrenia (Lidow, 2003). At the same potency of SKF 81297-mediated heterooligomer internaliza- time, the leading mainstay of schizophrenia therapeutics is tion and enhanced potency of quinpirole to induce internal- based on the transmitter dopamine. Because neither the D1 ization of the D1 and D2 receptor heteromer. nor the D2 dopamine receptor has been consistently shown in Heterooligomerization resulted in altered steady-state cel- different cellular models to directly alter calcium signaling, it lular distribution of D1 and D2 receptors within cells that was difficult to reconcile these two streams of evidence. Our was distinct from that of D1 and D2 receptor homooligomers. findings related to the D1 and D2 receptor heterooligomers This situation involving D1 and D2 receptors is unique in may provide a novel insight into how these receptors function that the cell surface expression of two receptors, which inde- synergistically and eventually lead to mechanisms of poten- pendently traffic to the cell surface as homooligomers, were tial dysfunction. The D1-D2 heterooligomers, therefore, may both altered by heterooligomer formation. D1 and D2 recep- represent an important and compelling drug target for dis- tor homooligomers differed in their cellular distribution in eases related to the dopaminergic system. that D1 receptor homooligomers were predominantly expressed on the cell surface with little intracellular localiza- Acknowledgments tion, whereas D2 receptor homooligomers displayed both We thank Theresa Fan for her excellent technical assistance. prominent cell surface and intracellular localization. Intra- cellular localization of the D2 receptor was not a product of References transient transfection because a similar result was observed Aizman O, Brismar H, Uhlen P, Zettergren E, Levey AI, Forssberg H, Greengard P, and Aperia A (2000) Anatomical and physiological evidence for D1 and D2 dopa- in stable cell lines expressing the D2 receptor alone as we mine receptor colocalization in neostriatal neurons. Nat Neurosci 3:226–230. have shown. The intracellular localization of the D2 receptor Canals M, Marcellino D, Fanelli F, Ciruela F, de Benedetti P, Goldberg SR, Neve K, Fuxe K, Agnati LF, Woods AS, et al. (2003) Adenosine A2A-dopamine D2 receptor- has been suggested previously to stem from a relatively in- receptor heteromerization: qualitative and quantitative assessment by fluores- efficient processing of the D2 receptor protein (Fishburn et cence and bioluminescence energy transfer. J Biol Chem 278:46741–46749. Chugani DC, Ackermann RF, and Phelps ME (1988) In vivo [3H]spiperone binding: al., 1995; Prou et al., 2001) or because the receptor partici- evidence for accumulation in corpus striatum by agonist-mediated receptor inter- pated in an agonist-independent constitutive internalization nalization. J Cereb Blood Flow Metab 8:291–303. Fishburn CS, Elazar Z, and Fuchs S (1995) Differential glycosylation and intracel- process (Vickery and von Zastrow, 1999). Changes in the lular trafficking for the long and short isoforms of the D2 dopamine receptor. J Biol configuration of the intracellular domains of both D1 and D2 Chem 270:29819–29824. Franco R, Ferre S, Agnati L, Torvinen M, Gines S, Hillion J, Casado V, Lledo P, Zoli receptors within the heterooligomer may have resulted in M, Lluis C, et al. Evidence for adenosine/dopamine receptor interactions: indica- changes in receptor processing or cell surface turnover and tions for heteromerization. Neuropsychopharmacology 23:S50–9, 2000. thus altered the number of receptors available for radioli- George SR, Fan T, Xie Z, Tse R, Tam V, Varghese G, and O’Dowd BF (2000) Oligomerization of ␮- and ␦-opioid receptors. Generation of novel functional prop- gand binding. These changes may have resulted in the expo- erties. J Biol Chem 275:26128–26135. sure or masking of consensus sequences such as ER retention George SR, Lee SP, Varghese G, Zeman PR, Seeman P, Ng GY, and O’Dowd BF (1998) A transmembrane domain-derived peptide inhibits D1 dopamine receptor motifs (Zerangue et al., 1999) or ER export motifs (Ma et al., function without affecting receptor oligomerization. J Biol Chem 273:30244– 2001) and may allow for novel interactions with adaptor 30248. George SR, O’Dowd BF, and Lee SP (2002) G-protein-coupled receptor oligomeriza- proteins involved in receptor trafficking. tion and its potential for drug discovery. Nat Rev Drug Discov 1:808–820. In summary, the results presented in this study provide Ito K, Haga T, Lameh J, and Sadee W (1999) Sequestration of dopamine D2 receptors depends on coexpression of G-protein-coupled receptor kinases 2 or 5. Eur J Bio- cogent evidence that heterooligomerization occurs between chem 260:112–119. D1 and D2 receptors. Together with our report of the robust Kim SJ, Kim MY, Lee EJ, Ahn YS, and Baik JH (2004) Distinct regulation of internalization and mitogen-activated protein kinase activation by two isoforms of D1-D2 receptor mediated calcium signal (Lee et al., 2004), the dopamine D2 receptor. Mol Endocrinol 18:640–652. these results provide a molecular mechanism for how D1-D2 Kita K, Shiratani T, Takenouchi K, Fukuzako H, and Takigawa M (1999) Effects of D1 and D2 dopamine receptor antagonists on cocaine-induced self-stimulation and receptor synergism may occur in cells in which these recep- locomotor activity in rats. Eur Neuropsychopharmacol 9:1–7. tors coexist. The extent of receptor heterooligomerization Lamey M, Thompson M, Varghese G, Chi H, Sawzdargo M, George SR, and O’Dowd BF (2002) Distinct residues in the carboxyl tail mediate agonist-induced desensi- must be significant because several changes in receptor func- tization and internalization of the human dopamine D1 receptor. J Biol Chem tion were observed, such as the generation of a novel calcium 277:9415–9421. 198 578 So et al.

Lee SP, O’Dowd BF, Ng GY, Varghese G, Akil H, Mansour A, Nguyen T, and George tial findings with [3H]raclopride versus [3H]spiperone. Mol Pharmacol 63:456– SR (2000) Inhibition of cell surface expression by mutant receptors demonstrates 462. that D2 dopamine receptors exist as oligomers in the cell. Mol Pharmacol 58:120– Terrillon S, Durroux T, Mouillac B, Breit A, Ayoub MA, Taulan M, Jockers R, 128. Barberis C, and Bouvier M (2003) Oxytocin and vasopressin V1a and V2 receptors Lee SP, So CH, Rashid AJ, Varghese G, Cheng R, Lanca AJ, O’Dowd BF, and George form constitutive homo- and heterodimers during biosynthesis. Mol Endocrinol SR (2004) Dopamine D1 and D2 receptor Co-activation generates a novel phos- 17:677–691. pholipase C-mediated calcium signal. J Biol Chem 279:35671–35678. Toews ML (2000) Radioligand-binding assays for G protein-coupled receptor inter- Lidow MS (2003) Calcium signaling dysfunction in schizophrenia: a unifying ap- nalization, in Regulation of G Protein-Coupled Receptor Function and Expression proach. Brain Res Brain Res Rev 43:70–84. (Benovic JL, ed) pp 199–230, Wiley-Liss, New York. Ma D, Zerangue N, Lin YF, Collins A, Yu M, Jan YN, and Jan LY (2001) Role of ER Vickery RG and von Zastrow M (1999) Distinct dynamin-dependent and -indepen- export signals in controlling surface potassium channel numbers. Science (Wash dent mechanisms target structurally homologous dopamine receptors to different DC) 291:316–319. endocytic membranes. J Cell Biol 144:31–43. McVey M, Ramsay D, Kellett E, Rees S, Wilson S, Pope AJ, and Milligan G (2001) Waszczak BL, Martin LP, Finlay HE, Zahr N, and Stellar JR (2002) Effects of Monitoring receptor oligomerization using time-resolved fluorescence resonance individual and concurrent stimulation of striatal D1 and D2 dopamine receptors ␦ energy transfer and bioluminescence resonance energy transfer. The human -opi- on electrophysiological and behavioral output from rat basal ganglia. J Pharmacol oid receptor displays constitutive oligomerization at the cell surface, which is not Exp Ther 300:850–861. regulated by receptor occupancy. J Biol Chem 276:14092–14099. Xu J, He J, Castleberry AM, Balasubramanian S, Lau AG, and Hall RA (2003) Ng GY, Varghese G, Chung HT, Trogadis J, Seeman P, O’Dowd BF, and George SR Heterodimerization of ␣2A- and ␤1-adrenergic receptors. J Biol Chem 278:10770– (1997) Resistance of the dopamine D2L receptor to desensitization accompanies 10777. the up-regulation of receptors on to the surface of Sf9 cells. Endocrinology 138: Zerangue N, Schwappach B, Jan YN, and Jan LY (1999) A new ER trafficking signal 4199–4206. regulates the subunit stoichiometry of plasma membrane K(ATP) channels. Neu- Piomelli D, Pilon C, Giros B, Sokoloff P, Martres MP, and Schwartz JC (1991) Dopamine activation of the arachidonic acid cascade as a basis for D1/D2 receptor ron 22:537–548. synergism. Nature (Lond) 353:164–167. Zhang LJ, Lachowicz JE, and Sibley DR (1994) The D2S and D2L dopamine receptor Pollack A (2004) Coactivation of D1 and D2 dopamine receptors: in marriage, a case isoforms are differentially regulated in Chinese hamster ovary cells. Mol Pharma- of his, hers and theirs. Sci STKE 2004:pe50. col 45:878–889. Prou D, Gu WJ, Le Crom S, Vincent JD, Salamero J, and Vernier P (2001) Intracel- lular retention of the two isoforms of the D(2) dopamine receptor promotes endo- Address correspondence to: Dr. Susan R. George, Rm. 4358, Medical Sci- plasmic reticulum disruption. J Cell Sci 114:3517–3527. ence Building, 1 King’s College Circle, University of Toronto, Toronto, Ontario, Sun W, Ginovart N, Ko F, Seeman P, and Kapur S (2003) In vivo evidence for M5S 1A8 Canada. E-mail: [email protected] dopamine-mediated internalization of D2-receptors after amphetamine: differen-

199 6.3 Dopamine Receptor Oligomerization Visualized In Living Cells

6.3.1 Summary

This report provides qualitative evidence for the assembly of dopamine receptor oligomers. Similar to the previous paper, this study showed that specific dopamine receptor subtypes with an artificially inserted nuclear localization sequence could “drag” other dopamine receptor subtypes into the nucleus by hetero-oligomerization. Furthermore, oligomerization of receptors with distinct conformations could be induced to separate by ligand binding, indicating that oligomers with conformationally dissimilar receptors were unstable and prone to dissociation.

6.3.2 Contribution by thesis author

The author generated the S198A/S199A D1 receptor mutant presented in Figures 5 and 9 and the S199A/S202A D1 receptor mutant shown in Figure 6. The thesis author also created the schematic in Figure 1A.

Dopamine Receptor Oligomerization Visualized in Living Cells* Received for publication, April 26, 2005, and in revised form, July 8, 2005 Published, JBC Papers in Press, August 22, 2005, DOI 10.1074/jbc.M504562200 Brian F. O’Dowd1, Xiaodong Ji, Mohammad Alijaniaram, Ryan D. Rajaram, Michael M. C. Kong, Asim Rashid, Tuan Nguyen, and Susan R. George2 From the Departments of Pharmacology and Medicine, University of Toronto, Toronto, Ontario M5S 1A8 and Centre for Addiction and Mental Health, Toronto, Ontario M5T 1R8, Canada

G protein-coupled receptors occur as dimers within arrays of oli- accommodating 10–20 dimers (4, 5). Rhodopsin-related GPCRs func- gomers. We visualized ensembles of dopamine receptor oligomers tion as arrays of oligomers (5, 6) and form complexes with identical or in living cells and evaluated the contributions of receptor confor- other GPCRs, generating homo- or hetero-oligomers (7–10). The struc- mation to the dynamics of oligomer association and dissociation, tural details involved in the formation of receptor dimers or oligomers using a strategy of trafficking a receptor to another cellular com- have not been elucidated, with little experimental evidence for funda- partment. We incorporated a nuclear localization sequence into the

mental questions, such as their behavior at the cell surface, whether the Downloaded from D1 dopamine receptor, which translocated from the cell surface to oligomers remain intact or separate, and if homooligomers and hetero- the nucleus. Receptor inverse agonists blocked this translocation, oligomers behave differently. That the oligomerized GPCR structures retaining the modified receptor, D1-nuclear localization signal modulate the properties and conformations of the individual constitu- (NLS), at the cell surface. D1 co-translocated with D1-NLS to the ent receptors involved has been shown for ␮- and ␦-opioid receptor -nucleus, indicating formation of homooligomers. (؉)-Butaclamol heterooligomers (8, 11, 12) and D1 and D2 dopamine receptor hetero retained both receptors at the cell surface, and removal of the drug oligomers (13). In these structures, the ligand binding properties and/or www.jbc.org allowed translocation of both receptors to the nucleus. Agonist- the coupling properties were altered, depending on whether a receptor nonbinding D1(S198A/S199A)-NLS, containing two substituted was within a homooligomer or heterooligomer. serine residues in transmembrane 5 also oligomerized with D1, and The biophysical techniques utilized to investigate GPCR oligomers, both were retained on the cell surface by (؉)-butaclamol. Drug such as bioluminescence resonance energy transfer (14) or fluorescence at University of Toronto on July 11, 2007 removal disrupted these oligomerized receptors so that D1 resonance energy transfer (9, 15), permitted the analysis of receptor- remained at the cell surface while D1(S198A/S199A)-NLS traf- receptor or receptor-protein interactions in situ, within living cells. ficked to the nucleus. Thus, receptor conformational differences However, these techniques still have only limited ability to investigate permitted oligomer disruption and showed that ligand-binding the many aspects of oligomer structure or function that still remain pocket occupancy by the inverse agonist induced a conformational unknown. There is a need to provide insight into the questions regard- change. We demonstrated robust heterooligomerization between ing receptor oligomeric complexes, such as (i) the precise sites of inter- the D2 dopamine receptor and the D1 receptor. The heterooli- actions maintaining monomers in a dimer formation; (ii) whether true gomers could not be disrupted by inverse agonists targeting either heterodimers exist or only homodimers within a heterooligomeric com- one of the receptor constituents. However, D2 did not heterooli- plex; (iii) whether the homodimer interactions differ from the het- gomerize with the structurally modified D1(S198A/S199A), indi- erodimer; (iv) the numbers of dimers in an oligomer; and (v) whether cating an impaired interface for their interaction. Thus, we describe oligomers form larger complexes and can be functionally regulated. a novel method showing that a homogeneous receptor conforma- Thus, elucidation of the mechanism underlying the formation of oligo- tion maintains the structural integrity of oligomers, whereas con- meric structures, analysis of their functional properties, or analysis of formational heterogeneity disrupts it. their behavior in cells requires new experimental paradigms. We wished to explore some of the above aspects of oligomerization to understand the dynamics governing the formation and trafficking of G protein-coupled receptors (GPCRs)3 form dimers and higher order oligomers, to ultimately understand the functional relevance of oli- oligomers, as inferred from a large body of evidence garnered from a gomers in cellular processes. We devised a strategy that engineered the variety of methodological approaches (1–3), and their static configura- trafficking of a GPCR to another cellular compartment and hypothe- tion has been visualized by atomic force microscopy in the case of rho- sized that if it took with it its oligomeric partner, this would provide dopsin to reveal dimers arranged in clustered rows, with each row definitive proof of oligomerization and provide a tool to study its dynamics in the cell. We and others determined that homodimerization in the rhodopsin-like GPCRs utilizes a transmembrane domain dimer * This work was supported by a Proof of Principle Grant from the Canadian Institutes for Health Research. The costs of publication of this article were defrayed in part by the interface (6, 16, 17, 18), and we predicted that this interaction would payment of page charges. This article must therefore be hereby marked “advertise- remain intact during the engineered receptor trafficking. Moreover, this ment” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 To whom correspondence should be addressed: Dept. of Pharmacology, Medical Sci- process may permit us to visualize homo- and heterooligomer forma- ence Bldg., University of Toronto, 1 King’s College Circle, Toronto, Ontario M5S 1A8, tion and, using conformationally altered receptor variants, to probe the Canada. Tel.: 416-978-7579; Fax: 416-978-2733; E-mail: [email protected]. contribution of the receptor structure to the stability of oligomers. To 2 Holder of a Canada Research Chair in Molecular Neuroscience. 3 The abbreviations used are: GPCR, G protein-coupled receptor; NLS, nuclear localiza- achieve our goals, a nuclear translocation pathway was exploited for tion signal; BTC, butaclamol; (ϩ)BTC and (Ϫ)BTC, (ϩ)- and (Ϫ)-butaclamol, respec- GPCRs. Translocation of proteins to the nucleus involves nuclear trans- tively; GFP, green fluorescent protein; RFP, red fluorescent protein; mRFP, monomer- ized red fluorescent protein; IC1, -2, and -3, intracellular loop 1, 2, and 3; HA, port proteins that recognize nuclear localization signal sequences hemagglutinin; FITC, fluorescein isothiocyanate. (NLSs) (19), which recruit importin carrier proteins, that mediate pro- 201 NOVEMBER 4, 2005•VOLUME 280•NUMBER 44 JOURNAL OF BIOLOGICAL CHEMISTRY 37225 Visualization of Dopamine Receptor Oligomers

tein translocation to the nucleus. Only a few GPCRs contain endoge- Fluorocytometry—50,000 cells were added to each well (96-well plate) ␮ nous NLSs (20–22). One example, the angiotensin AT1 receptor, con- and transfected with 0.5 g of cDNA. Minimum essential medium contains an endogenous NLS in helix 8, which serves to direct the receptor taining antagonists at varying concentrations were added to wells in into the nucleus in certain cells (10, 20). pentuplicate. The drugs were prepared as 1 mM concentration stock and The NLS strategy was evaluated using dopamine receptors. Dopa- diluted in growth medium to achieve a concentration between 10 nM mine D1 and D2 receptors each have been shown to form homooli- and 10 ␮M. After 48 h, cells were fixed with 4% paraformaldehyde and gomers and together to form heterooligomers (13, 23, 24). We were the incubated with the primary antibody (rat anti-HA antibody, 1:200 dilu- first to demonstrate that D2 receptors exist as homodimers in human tion; Roche Applied Science) and secondary antibody conjugated to and rat brain (25), and our demonstration of D1 and D2 receptor com- FITC (goat anti-rat antibody, 1:32 dilution; Sigma). Cell surface fluores- plexes by co-immunoprecipitation from rat brain and heterologous cence was detected using a Cytofluor 4000 (PerSeptive Biosystems). cells demonstrated that these receptors heterooligomerize (13). The Each experiment was repeated three or four times. Background fluores- functional synergism between D1 and D2 dopamine receptors was evi- cence from media, cells, plastic, etc. was evaluated in each experiment denced by the generation of a novel calcium signal by receptor co- and subtracted from the readings. activation (13). By fluorescence resonance energy transfer analysis, we DNA Constructs—All of the DNA encoding the GPCRs were from revealed that the D1 and D2 receptors exist in close proximity on the cell humans. Sequences encoding GPCRs were cloned into plasmid pEGFP, surface, presumably within a heterooligomeric complex. Furthermore, pDsRed2-N1, or pcDNA3. the D1 and D2 receptor heterooligomers displayed novel agonist-in- Receptor Constructs—The D1(S198A/S199A) and D1(S199A/S202A) duced internalization and trafficking patterns, distinct from that of D1 receptors were prepared using the Quikchange mutagenesis kit (Strat- Downloaded from and D2 receptor homooligomers (26). agene) according to the manufacturer’s instructions using the following In this report, we show that incorporating an NLS into several of the sets of primers: D1(S198A/S199A), forward (5Ј-GGACATATGCCAT- dopamine receptors mediated receptor translocation to the nucleus. GTCAGCCGCCGTAATAAGCTTTTACATCCC-3Ј) and reverse (5Ј- We used this translocation strategy as outlined (Fig. 1a) to understand GGGATGTAAAAGCTTATTACGGCGGCTGAGATGGCATATG- the ability of these GPCRs to co-traffic with their oligomerization part- TCC-3Ј); D1(S199A/S202A), forward (5Ј-GCAGGACATATGCCAT-

ners as a test of the robustness of the interaction between them and to CTCATCCGCCGTAATAGCCTTTTACATCCCTGTGG-3Ј) and re- www.jbc.org probe the structural conformation of homo- and heterooligomers after verse (5Ј-CCACAGGGATGTAAAAGGCTATTACGGCGGATGA- ligand occupancy and after introduction of structural variation by point GATGGCATATGTCCTGC-3Ј). mutagenesis. The method enabled the identification of both homo- and D1-NLS-GFP—Receptor DNA was subjected to PCR as previously heterooligomers for the dopamine receptors and demonstrated that ␮ ϫ at University of Toronto on July 11, 2007 reported (29). The reaction mixture consisted of H2O (32 l), 10 Pfu both types of interactions were robust enough to result in co-trafficking ␮ ␮ ␮ buffer (Stratagene) (5 l), dNTP (10 mM,5 l), Me2SO (5 l), oligonu- of oligomeric partners to the nucleus. Conformational homogeneity of cleotide primers (100 ng, 1 ␮l each), DNA template (100 ng), and Pfu the receptors was necessary to maintain the integrity of a homooli- enzyme (5 units). Total volume was 50 ␮l. PCR conditions were as gomer, and the interaction between similar receptors within a homo- follows: one cycle at 94 °C for 2 min and 30–35 cycles at 94 °C for 30 s, meric structure could not be disrupted. Furthermore, the introduction 55 °C for 30 s, 72 °C for 1 min per cycle and then one cycle at 72 °C for of any structural dissimilarity of the receptors within the homooli- 5 min. gomer, whether induced by drug occupancy or point mutation, resulted Primer Set for Amplification of the DNA Encoding the D1 Receptor—Primers in the ability to disrupt these oligomeric structures in living cells. How- were as follows: HD1-P1, 5Ј-GAGGACTCTGAACACCGAATTCGCCGC- ever, within the D1-D2 heterooligomer, conformational alteration by CATGGACGGGACTGGGCTGGTG-3Ј; HD1-P2, 5Ј-GTGTGGCAGGA- antagonist occupancy of one receptor was unable to affect the confor- TTCATCTGGGTACCGCGGTTGGGTGCTGACCGTT-3Ј. mation of the other, and we defined a structural alteration that pre- The restriction site EcoRI was incorporated in the primer HD1-P1, vented oligomerization of D1 and D2 receptors, indicating that the and KpnI was incorporated into HD1-P2. The PCR product, containing arrangement of the receptors within the heterooligomer may be sub- no stop codon was subcloned into vector pEGFP at EcoRI and KpnI and stantially different from that within a homooligomer. in frame with the start codon of GFP. The NLS sequence, KKFKR, was Importantly, the strategy described provides a means of testing the inserted into DNA encoding the base of TM7 (helix 8) of the D1 dopa- robustness of the interaction between receptors, which will be useful to mine receptor by PCR, replacing the sequence 336DFRKA. determine points of contact forming the oligomers. Primer Set for the Construction of DNA Encoding D1-NLS—Primers were as follows: HD1-NLSF, 5Ј-CCTAAGAGGGTTGAAAATCTTT- MATERIALS AND METHODS TAAATTTTTTAGCATTAAAGGCATAAATG-3Ј; HD1-NLSR, 5Ј- Fluorescent Proteins—cDNA sequences encoding GFP (27), pDsRed2 GCCTTTAATGCTAAAAAATTTAAAAGATTTTCAACCCTCTT- (28), and pDsRed2-nuc were obtained from CLONTECH (Palo Alto, AGGATGC-3Ј. CA). Using the DNA encoding D1-GFP as template, PCR with the primers Cell Culture—HEK cells grown on 60-mm plates in minimum essen- HD1-P1 and HD1-NLSF resulted in a product of 1000 bp (PCR1). Using tial medium were transfected with 0.5–2 ␮g of cDNA using Lipo- DNA encoding D1-GFP, PCR with primers HD1-P2 and HD1-NLSR fectamine (Invitrogen). Dopamine antagonists (ϩ)-butaclamol or SCH resulted in a product of 300 bp (PCR2). A subsequent PCR carried out 23390, when used, was added to cells at 6, 22, 30, and 42 h, and cells were with HD1-P1 and HD1-P2 primers resulted in a product of 1300 bp visualized by confocal microscopy at 48 h post-transfection. using the product from PCR1 and the product from PCR2 as templates. Microscopy—Live cells expressing GFP and pDsRed2 fusion proteins The resulting DNA encoding D1-NLS was subcloned into vector were visualized with an LSM510 Zeiss confocal laser microscope. In pEGFP at EcoRI and KpnI restriction sites. each experiment, 5–8 fields, containing 50–80 cells/field were evalu- All of the additional constructs described below were made using the ated, and the entire experiment was repeated 2–4 times (n ϭ 3–5). same PCR method and experimental conditions as described above for 202 37226 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 280•NUMBER 44•NOVEMBER 4, 2005 Visualization of Dopamine Receptor Oligomers the D1 dopamine receptor, but with the specific primers as described restriction digest with EcoRI and KpnI and subcloned into the mRFP below. vector at the same restriction sites EcoRI and KpnI and in frame with the Primer Set for the Construction of D1-IC1-NLS—Primers were as fol- mRFP. lows: D1-NLSF-IC1, 5Ј-GTGCTGCCGTTAAAAAGTTCAAACGC- Primer Set for CysLT2-NLS-GFP—Primers used were LT2-NLSF, 5Ј- CTGCGGTCCAAGG-3Ј; D1-NLSR-IC1, 5Ј-GGACCGCAGGCGTT- GCTGGGAAAAAATTTAAAAGAAGACTAAAGTCTGCAC-3Ј TGAACTTTTTAACGGCAGCACAGACC-3Ј. KKFKR was inserted and LT2-NLSR (5Ј-GTCTTCTTTTAAATTTTTTCCCAGCAAAG- into the intracellular loop, replacing 49IRFRH. TAATAGAGC-3Ј). The sequence KKFKR was inserted into the cys- Primer Set for the Construction of D1-IC2-NLS—Primers were as fol- teinyl leukotriene 2 receptor, replacing 310ENFKD. lows: D1NLSF-IC2, 5Ј-CCGGTATGAGAAAAAGTTTAAACGCAA- GGCAGCCTTC-3Ј; D1-NLSR-IC2, 5Ј-GGCTGCCTTGCGTTTAA- RESULTS Ј ACTTTTTCTCATACCGGAAAGG-3 . KKFKR was inserted into IC2, The initial objective was to generate a GPCR that would traffic to the 133 replacing RKMTP. nucleus under basal conditions and permit ligand-occupied conforma- Primer Set for the Construction of D1-IC3-NLS—Primers were as fol- tional changes to modulate this process. Ј lows: D1NLSF-IC3, 5 -GGAAAGTTCTTTTAAGAAGAAGTTCAA- Optimization of the Position of the NLS within the D1 Dopamine Ј Ј AAGAGAAAC-3 ; D1-NLSR-IC3, 5 -GTTTCTCTTTTGAACTTCT- Receptor in a Conformation-dependent Site—To determine the optimal Ј TCTTAAAAGAACTTTCC-3 . KKFKR was inserted into the IC3 site for NLS incorporation that would provide the most efficient recep- 262 segment of the D1 receptor, replacing MSFKR. tor translocation to the nucleus, the D1 dopamine receptor was modi- Primer Set for D2-NLS-GFP—Primers were as follows: HDNLSF, 5Ј-

fied. Since the proteins that carry NLS-containing proteins to the Downloaded from CACCACCTTCAACAAAAAATTCAAAAGAGCCTTCCTGAAG- nucleus, such as the importins, are cytoplasmic, the optimal placement Ј Ј ATCC-3 ; HD2-NLSR, 5 -GGATCTTCAGGAAGGCTCTTTTGAA- of the NLS was investigated by its introduction into various positions Ј TTTTTTGTTGAAGGTGGTG-3 . The sequence KKFKR was within the three intracellular loops, helix 8, and the carboxyl tail of D1; 431 inserted in the D2 receptor, replacing IEFRK. each of these locations is illustrated in Fig. 1b. Primer Set for the Construction of SPGFP-D2—The D2 cDNA was When expressed in cells, the unmodified D1 receptor tagged with isolated by the PCR method using the following set of primers: D2sp- GFP was localized on the cell surface in the majority of the cells (Ͼ90%) www.jbc.org Ј Ј BsrGI, 5 -TGTACAGCCGCCATGGATCCACTGAATCTGTCC-3 ; at 48 h post-transfection as visualized by confocal microscopy (Fig. 2a, Ј D2sp-NotI, 5 -GAGTCGCGGCCGCTTCAGCAGTGGAGGATCT- upper left panel). Cells transfected with the D1 receptor containing the Ј TCAGGAAGG-3 . This PCR product was then subcloned into the SP- NLS in helix 8 (D1-NLS) revealed a basal localization of receptor in the

GFP vector at restriction sites BsrGI and NotI. nucleus at 48 h post-transfection (Fig. 2a, upper right panel) observed in at University of Toronto on July 11, 2007 Primer Set for the Construction of SPGFP-D2-NLS—Using the over 90% of cells. The localization of the receptor in the nucleus was SPGFP-D2 as template, the NLS KKFKR was introduced into helix 8 of confirmed using a nuclear dye (Hoechst 33342) (Fig. 2a, lower panels). the D2 by the PCR method using the following set of primers: HD2- Thus, incorporation of a NLS into the D1 receptor sequence in helix 8 NLSF, 5Ј-CACCACCTTCAACAAAAAATTCAAAAGAGCCTTCC- resulted in efficient trafficking of the receptor under basal conditions to TGAAGATCC-3Ј; HD2-NLSR, 5Ј-GGATCTTCAGGAAGGCTCTT- the nucleus. A series of tomographic images (Z-stacks) obtained by TTGAATTTTTTGTTGAAGGTGGTG-3Ј. The sequence KKFRK confocal microscopy confirmed the nuclear localization of the D1-NLS. 431 was inserted in the D2 receptor, replacing IEFRK. Investigation of the effect of the NLS inserted in various other intra- Primer Set for the Construction of D5-GFP—The human D5 was iso- cellular positions of the receptor revealed that D1 with an NLS inserted lated by PCR with primers HD5-EcoRI (5Ј CTGGAATTCTGCAGAT- in the first intracellular cytoplasmic loop (D1-IC1-NLS) was expressed TCCAGCCCGAAATGCTGCCGCC-3Ј) and HD5-Kpn (5Ј-CGCCA- and detected in the nucleus in 85% of cells (Fig. 2b, left). D1 with the NLS GTGTGATGGATAATGGTACCGCATGGAATCCATTCGGGGT- inserted in the second intracellular cytoplasmic loop (D1-IC2-NLS) was G-3Ј) and subcloned into the enhanced green fluorescent protein vector expressed and detected in the nucleus in 51% of cells. In this case, over at the restriction sites EcoRI and KpnI. 40% of cells still had receptor detectable on the cell surface, indicating Primer Set for the Construction of SPGFP-D5—The human D5 was that incorporation of the NLS in this position was not as efficient in isolated by PCR with the following set of primers: D5-BsrGI, 5Ј-CCA- translocating the receptor from the cell surface. D1 with the NLS GCCCGTGTACAAATGCTGCCGCCAGGCAGC-3Ј; D5-NotI, 5Ј- inserted in the third intracellular cytoplasmic loop (D1-IC3-NLS) was GCGGCCGCTTAATGGAATCCATTCGGGG-3Ј. This PCR product expressed and detected in the nucleus of 85% of cells (Fig. 2b, middle). was then subcloned into the SP-GFP vector at restriction sites BsrGI D1 with an NLS inserted in a distal position in the carboxyl tail (D1-CT- and NotI. NLS) was expressed and was detected at the cell surface and also in the The Construction of D1-mRFP—mRFP1 in the pRSETb vector was a nucleus (Fig. 2b, right). The distribution of D1-GFP-NLS where the NLS gift from Dr. Irine Prastio (Howard Hughes Medical Institute, University of was attached distal to the GFP fused to D1, showed expression in the California, San Diego). Using this vector as template, the mRFP was isolated cytoplasm and cell surface but little expression in the nucleus (data not by PCR using the following two primers: mRFP-BAMH, 5Ј-GATAAGGA- shown). TCCGATGGCCTCCTCCGAGG-3Ј; mRFP-NOT, 5Ј-CGAATTCGCG- To determine the effect of occupancy of the binding pocket on the GCCGCTAGGCGCCGGTGGAGTGGCGG-3Ј. This PCR product was NLS within various positions of the receptor on trafficking to the then used to replace the GFP from the pEGFP-N1 vector at the restriction nucleus, we used several antagonists with inverse agonist properties to sites BamHI and NotI, thus creating the mRFP vector. Human D1 was induce a conformational change in the receptor. Although receptor excised from the D1-GFP construct by restriction digest with EcoRI and agonists would also induce conformational changes in the receptor, we KpnI and subcloned into the mRFP vector at the same restriction sites wished to preclude agonist-induced internalization mechanisms, which EcoRI and KpnI and thus in frame with the mRFP. would confound the evaluation of NLS-mediated trafficking. Cells The Construction of D1-NLS (Helix 8)-mRFP—Human D1-NLS expressing D1-NLS were treated with the dopamine receptor antago- (helix 8) was excised from the D1-NLS (helix 8)-GFP construct by nist (ϩ)-butaclamol ((ϩ)BTC) or SCH 23390 6 h post-transfection. 203 NOVEMBER 4, 2005•VOLUME 280•NUMBER 44 JOURNAL OF BIOLOGICAL CHEMISTRY 37227 Visualization of Dopamine Receptor Oligomers Downloaded from www.jbc.org at University of Toronto on July 11, 2007

FIGURE 2. a, effect of incorporation of an NLS into helix 8 of the D1 dopamine receptor (D1-NLS). The D1 dopamine receptor-GFP (D1) fusion protein was expressed in HEK cells and visualized by confocal microscopy. D1 was localized on the cell surface and was absent from the nucleus (upper left panel). D1 dopamine receptor-GFP with an NLS in helix 8 (D1-NLS) expression was located in the nucleus of the cells and not at the cell surface (upper right panel). Nuclei were identified by co-expression with DsRed2-nuc or staining with Hoechst 33342 (lower panel). b, effect of incorporation of an NLS into other positions of the D1 receptor; IC1 (D1-IC1-NLS), IC3 (D1-IC3-NLS), and the carboxyl tail (D1-CT-NLS) were expressed in HEK cells and visualized. D1-IC1-NLS and D1-IC3-NLS were localized in the nucleus, and D1-CT-NLS was localized on the cell surface and inside the cell. In some cases, nuclei were identified by co-expression with DsRed2-nuc. c, effect of exposure to antagonist. D1-NLS-expressing cells were treated with the D1 antagonist (ϩ)BTC, 1 ␮M (left), and were located at the cell surface. D1-IC1-NLS was treated with (ϩ)BTC (1 ␮M) and was localized largely on the cell surface and cytoplasm and was absent from the nucleus. D1-IC3-NLS treated with (ϩ)BTC (1 ␮M) was localized in the nucleus, indicating no effect of drug. Nuclei were identified by co-expression with DsRed2-nuc. d, demonstration of cell surface expression of D1-NLS 9 and 24 h following transfection. Cells expressing D1-NLS were treated with concanavalin A for 24 h.

FIGURE 1. a, schematic representation of the cellular trafficking of a GPCR with an incorporated NLS co-expressed with a wild type GPCR. Oligomers form in the endoplasmic D1-NLS was expressed and treated with (ϩ)BTC (500 nM) for 48 h, reticulum (ER) and are trafficked to the cell surface and then to the nucleus. In the pres- and at this time, the receptor was located at the cell surface (100% of ence of a selective drug targeting the NLS-containing receptor, the oligomers can be ϩ retained at the cell surface. b, representation of the primary amino acid sequence of the cells). The ( )BTC was removed, and the receptor distribution was seven-transmembrane D1 dopamine receptor inserted in the plasma membrane show- examined at 3, 6, 13, 16, 19, and 24 h. Between 13 and 16 h, the receptor ing the locations of the substituted NLSs (red), IC1, IC2, IC3, helix 8, and the carboxyl tail. had left the cell surface and was distributed to the nucleus in 80% of cells. Cells expressing D1-IC1-NLS treated with either (ϩ)BTC (1 ␮M)or With (ϩ)BTC (1 ␮M) treatment for 48 h, there was a very efficient SCH 23390 (1 ␮M) also revealed retention of receptor at the cell surface retention of the D1-NLS receptor on the cell surface (85% of cells) with in 82% of cells (Fig. 2c, middle) and 77% of the cells, respectively, com- little translocation to the nucleus (Fig. 2c, left). This effect was also found pared with 76% of cells with receptor expression in the nucleus with no with the D1 receptor-selective antagonist SCH 23390. Stereoselectivity treatment. In contrast, cells expressing D1-IC3-NLS treated with antag- of this effect was demonstrated by the lack of effect of (Ϫ)BTC (1 ␮M) onist (ϩ)BTC or SCH 23390 revealed ϳ90 and 84% of cells with recep- treatment of the cells. In addition, the specificity of the D1 antagonist on tor in the nucleus, indicating no ability of the drugs to retard the trans- receptor translocation was tested by treating D1-NLS-expressing cells location of this receptor to the nucleus (Fig. 2c, right). with the D2 receptor-selective antagonist raclopride (1 ␮M), which was To investigate the cellular route by which D1-NLS reached the unable to retain the receptor at the cell surface. nucleus, cells expressing D1-NLS were examined at various times post- 204 37228 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 280•NUMBER 44•NOVEMBER 4, 2005 Visualization of Dopamine Receptor Oligomers Downloaded from

FIGURE 4. D1 receptor homooligomers remain intact when trafficking. a, D1-NLS (GFP) (green) and D1 (RFP) (red) co-translocate together to the nucleus. b, D1-NLS (green) www.jbc.org and D1 (red) were co-expressed and treated with (ϩ)BTC. (ϩ)BTC revealed robust effects to retain both receptors at the cell surface. c, D1-NLS (green) and D1 (red) visualized following removal of the (ϩ)BTC. The receptors, D1-NLS and D1, co-migrated together from the cell surface to the nucleus.

cate to the nucleus. When D1-NLS and D1 were expressed together, both at University of Toronto on July 11, 2007 receptors were located in the nucleus (93% of cells expressing both receptors) with little expression at the cell surface (Fig. 3a). Thus, the receptor not con- FIGURE 3. Homooligomerization of the D1 dopamine receptor. a, D1-NLS (GFP) (green) and D1 (RFP) (red) were co-expressed in HEK cells. D1 trafficked with D1-NLS away taining the NLS trafficked to the nucleus due to oligomer formation with from the cell surface to the nucleus. The overlap in distribution of the two receptors is D1-NLS. The figure also shows two cells expressing D1 alone (indicated by an indicated by the merged image.Anarrow denotes cells expressing predominantly D1 showing only cell surface expression. b, co-expression of D1-IC1-NLS (green) and D1 (red) arrow), demonstrating cell surface expression exclusively, confirming the visu- in HEK cells showed that D1 trafficked with D1-IC1-NLS away from the cell surface to the alization of D1 dopamine receptor homooligomerization in the cells co-ex- nucleus. c, co-expression of D1-IC3-NLS (green) and D1 (red) in HEK cells showed that D1 pressing both receptors. trafficked with D1-IC1-NLS away from the cell surface to the nucleus. d, co-expression of the receptors cysteinyl-leukotriene (CystLT2-NLS) (green) and D1 (red) in HEK cells. The When D1-IC1-NLS and D1 were expressed together, both receptors red and green colors remain distinct with no overlap, indicating that the receptors remain were located together in the nucleus (94% of cells) (Fig. 3b). Cells separate. expressing D1-IC3-NLS and D1 also revealed that both receptors were located in the nucleus (95% of cells Fig. 3c). Thus, efficient oligomer transfection and treated with concanavalin A to retard internalization of formation, as indicated by nuclear co-translocation, occurred with the receptor from the cell surface. D1-NLS was visualized and detected D1-NLS, D1-IC1-NLS, or D1-IC3-NLS, each expressed with D1. The at the cell surface between 7 and 10 h post-transfection, at a time when cysteinyl leukotriene receptor CysLT2, with low homology to the D1 no receptor was visualized in the nucleus (Fig. 2d, left), subsequently receptor was co-expressed with D1 following incorporation of a NLS. showing a progressive translocation to the nucleus (Fig. 2d, middle). The CysLT2-NLS and D1 receptors did not form oligomers as indicated D1-NLS was retained on the cell surface with no receptor located in the by the separation of receptors, one receptor visualized in the nucleus nucleus following treatment with concanavalin A for 24 h after trans- (100% of cells) and the other on the cell surface (100% of cells) (Fig. 3d). fection (Fig. 2d, right). These results indicated that the receptor traf- Investigation of Homooligomer Stability by Manipulating Receptor ficked first to the cell surface after biosynthesis and then was translo- Conformation—Cells expressing both D1-NLS and D1 showed cated to the nucleus. Cells expressing D1-IC3-NLS were treated with co-translocation to the nucleus (Fig. 4a). When cells expressing both concanavalin A for 24 h after transfection and showed undiminished D1-NLS and D1 were treated with (ϩ)BTC (1 ␮M) for 24 h, both recep- robust nuclear expression with no cell surface expression, indicating tors were located at the cell surface (100% of cells) (Fig. 4b). Following that the receptor may have traveled directly to the nucleus. 24hof(ϩ)BTC treatment, the drug was removed from the culture The incorporation of the NLS into the helix 8 of the D1 receptor did medium, cells were washed, and the distribution of the receptors in the not alter the binding pocket of the receptor, with preserved agonist- cells was reexamined after a further 24 h had elapsed. At this time, both detected high affinity and low affinity states, indicative of intact recep- D1 and D1-NLS had largely moved from the cell surface and translo- tor-G protein coupling and ligand affinities. The D1-NLS receptor had cated together to the nucleus (93% of cells) (Fig. 4c). Thus, the D1/D1- ϫ Ϫ9 ϫ Ϫ7 a Khigh value of 4.17 10 M and Klow of 1.19 10 M detected by NLS oligomer remained intact and was stable, as indicated by the two agonist SKF 81297 not different from unmodified D1 receptor. receptor types staying and trafficking together. Co-translocation of NLS-modified D1 and Wild Type D1 Receptors—The The critical residues within the binding pocket of the D1 receptor helix 8 NLS-containing D1 receptor (D1-NLS) had efficient ability to translo- include three serine residues in transmembrane domain V of the recep- 205 NOVEMBER 4, 2005•VOLUME 280•NUMBER 44 JOURNAL OF BIOLOGICAL CHEMISTRY 37229 Visualization of Dopamine Receptor Oligomers Downloaded from www.jbc.org FIGURE 6. Lack of D1 oligomer disruption by a minor receptor modification. a, D1(S199A/S202A)-NLS (A199/A202) expressed in HEK cells treated with vehicle and in the presence of (ϩ)BTC revealed robust effects to retain the receptor. b, D1-S199A/S202A)- NLS (GFP) (green) and D1 (RFP) (red) co-expressed and treated with (ϩ)BTC, which retained both receptors at the cell surface. c, cells co-expressed D1-S199A/S202A)-NLS at University of Toronto on July 11, 2007 and D1 and visualized following removal of the (ϩ)BTC. The majority of both receptors co-migrated together from the cell surface to the nucleus.

FIGURE 5. Disruption of D1 receptor oligomers by conformational alteration. a, HEK cells expressing D1(S198A/S199A) (A198/A199) showed that the receptors were located surface (100% of cells) (Fig. 5c). (ϩ)BTC was removed, and the cellular at the cell surface. b, D1(S198A/S199A)-NLS (GFP) (green) and D1 (RFP) (red), showing distribution of the receptors 24 h later was examined. At this time point, D1(S198A/S199A)-NLS trafficked with D1 to the nucleus. Cells expressing only D1 show exclusive cell surface localization (arrow). c, cells expressing D1(S198A/S199A)-NLS and D1 had remained at the cell surface, and D1(S198A/S199A)-NLS trans- D1 when treated with (ϩ)BTC revealed robust co-localization at the cell surface. d, co- located alone to the nucleus (Fig. 5d). D1(S198A/S199A)-NLS was in ϩ expression of D1(S198A/S199A)-NLS with D1, following removal of the ( )BTC. The red the nucleus in 40% of cells only, and 60% of the cells had receptor in the and green colors are distinct with no overlap, indicating that the oligomerized receptors have separated. nucleus and cell surface. These data indicated that treatment with (ϩ)BTC had probably induced a conformational change within the two tor. A D1 receptor generated with two of these serines, Ser198 and Ser199, receptors, which were now different from each other, resulting in their substituted with alanine (D1(S198A/S199A)) did not bind dopamine or separation. By this method, we could detect differences in the confor- SCH 23390, but we determined that it did bind (ϩ)BTC. The mation of the D1 and D1(S198A/S199A)-NLS receptors. Thus, D1(S198A/S199A) receptor was efficiently expressed on the cell surface D1(S198A/S199A) and D1 appear to oligomerize under basal condi- (100% of cells) (Fig. 5a). This receptor formed homooligomers as tions, but unlike the structurally homogenous D1 and D1-NLS oli- detected by Western blot analysis. Cells expressing D1(S198A/S199A)- gomers, these oligomers were more easily disrupted, probably due to NLS, when treated with (ϩ)BTC (1 ␮M) but not SCH 23390, resulted in conformational differences between the receptors. retention of receptor at the cell surface (100% of cells). To further test this hypothesis, D1(S198A/S199A) was expressed To investigate the cellular route by which this mutated receptor with D1(S198A/S199A)-NLS, and these receptors translocated together reached the nucleus, cells expressing D1(S198A/S199A)-NLS were to the nucleus (97% of cells). In the presence of (ϩ)BTC, both receptors treated with concanavalin A. The receptor was retained on the cell were located at the cell surface (100% of cells). Following the removal of surface with no receptor located in the nucleus following treatment for (ϩ)BTC, both D1(S198A/S199A) and D1(S198A/S199A)-NLS moved 24 h after transfection. This indicated that the receptor trafficked to the from the cell surface and translocated together to the nucleus (93% of cell surface prior to translocation to the nucleus (data not shown). cells). These data confirmed, therefore, that there were differences in In order to explore whether structural variations in the D1 receptor the conformations of D1 and D1(S198A/S199A) receptors, not just could affect oligomerization, D1 was expressed with D1(S198A/ within the binding pocket, as predicted by the mutations, but in the S199A)-NLS, and these receptors were visualized to translocate overall structure. Under basal conditions, the difference was such that it together to the nucleus, indicating efficient oligomer formation (90% of permitted efficient oligomer formation, but it was more exaggerated by cells) (Fig. 5b). Cells in the microscopic field expressing only D1 showed antagonist binding; hence, the oligomer was disrupted. Thus, as shown cell surface expression exclusively (indicated by an arrow). Cells by the pairs D1 and D1-NLS, and D1(S198A/S199A) and D1(S198A/ expressing both D1 and D1(S198A/S199A)-NLS were treated with S199A)-NLS, stable oligomers were formed only between conforma- (ϩ)BTC (1 ␮M), and both receptors were seen to be located on the cell tionally identical receptors. 206 37230 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 280•NUMBER 44•NOVEMBER 4, 2005 Visualization of Dopamine Receptor Oligomers Downloaded from www.jbc.org

FIGURE 7. Expression of HA-D1-NLS and treatment with antagonist with detection of cell surface receptor density by fluorometric analysis. a, HA-tagged dopamine D1 receptor with the helix 8 incorporated NLS (HA-D1-NLS) expressed in HEK cells. Cells FIGURE 8. Heterooligomers of D1 and D2 dopamine receptors. a, amino-D2-NLS (GFP) at University of Toronto on July 11, 2007 expressing HA-D1-NLS receptor were treated with varying doses of (ϩ)BTC antagonist or (green) and D1 (RFP) (red) co-translocation to the nucleus. Amino-D2-NLS is a fusion vehicle, and cell surface receptors were quantified by the fluorescent signal of an FITC- protein containing a GFP in the amino terminus of the receptor. b, amino-D2-NLS and D1 conjugated second antibody directed at the HA antibody. Treatment with the antago- co-expressed in cells and treated with the D1-selective antagonist SCH 23390 showed nist (ϩ)BTC revealed robust dose-dependent effects to retain HA-D1-NLS at the cell unaltered trafficking to the nucleus in the presence or after removal of drug. c, amino- surface, indicating that the antagonist reduced translocation from the cell surface. b, D2-NLS and D1 co-expressed in cells and treated with D2-selective antagonist raclo- HA-tagged dopamine D1 receptor (HA-D1) expressed in HEK cells. Cells expressing pride, which retained the heterooligomer at the cell surface. Removal of the raclopride HA-D1 were co-expressed with varying amounts of FLAG-D1-NLS. The HA-D1 cell surface enabled trafficking of the heterooligomer to the nucleus. d, amino-D2-NLS and D1 co- receptors were quantified by the fluorescent signal of a FITC-conjugated second anti- expressed in cells and treated with nonselective dopamine antagonist (ϩ)BTC. (ϩ)BTC body directed at the HA antibody. Increasing quantities of FLAG-D1-NLS resulted in a retained the heterooligomer at the cell surface, and removal of the (ϩ)BTC enabled decrease in cell surface expression of HA-D1, indicating that FLAG-D1-NLS formed oli- trafficking of the heterooligomer to the nucleus. gomers with HA-D1.

The effect of structural variation was analyzed further by evaluating obtained (Fig. 7a), showing retention of the receptor on the cell surface, the ability of D1(S199A/S202A), containing Ala substitutions at Ser199 indicating that the antagonist reduced D1-NLS trafficking from the cell and Ser202, also in transmembrane domain V, to oligomerize with D1. surface. Stereoselectivity of this effect was demonstrated by the lack of This receptor was incapable of binding dopamine but bound antago- effect of (Ϫ)BTC (1 ␮M). HA-tagged D1 receptors were co-expressed nists such as SCH 23390 normally (30). The receptor D1(S199A/ S202A)-NLS was as efficiently translocated to the nucleus as D1-NLS with increasing amounts of a FLAG-tagged D1-NLS. There was a con- and was retained on the cell surface by (ϩ)BTC (100% of cells Fig. 6a). centration-dependent effect of FLAG-D1-NLS to remove the HA-D1 When D1 was co-expressed with D1(S199A/S202A)-NLS, both recep- from the cell surface (Fig. 7b). tors co-trafficked to the nucleus (100% of cells). Treatment with Heterooligomerization of D1 and D2 Dopamine Receptors—The (ϩ)BTC (1 ␮M) resulted in retention of both receptors at the cell surface D2-NLS, with the NLS in helix 8, did not translocate as efficiently as (100% of cells) (Fig. 6b). Following the removal of (ϩ)BTC for 24 h, some D1-NLS from the cell surface to the nucleus, presumed to be due to the D1 remained at the cell surface, but some translocated with D1(S199A/ absence of an extended carboxyl tail on the D2 receptor, placing the S202A)-NLS to the nucleus (100% of cells) (Fig. 6c), suggesting an attachment of the GFP in too close proximity to the helix 8 NLS and incomplete separation of the two receptors. These data indicated that impeding importin access. A D2 receptor with GFP located in the amino considerable conformational similarity existed between these receptors terminus of the receptor (amino-D2-NLS) was therefore prepared. capable of forming stable oligomers. When the GFP was placed in the receptor amino-terminal sequence, Evaluating the Ability of D1-NLS Receptor to Remove D1 from the Cell there was very efficient D2 receptor translocation from the cell surface Surface Using Fluorometric Analysis—Analysis of the density of cell to the nucleus (95% of cells). Cells transfected with amino-D2-NLS and surface receptors was used to monitor the ability of D1-NLS to translo- D1 together revealed both receptors were located in the nucleus (92% of cate D1 off the cell surface. Using fluorescent labeling of epitope-tagged cells) (Fig. 8a). Thus, the D1 receptor not containing the NLS trafficked D1 receptors, cell surface receptor detection was achieved, and the sig- to the nucleus, due to oligomerization with amino-D2-NLS. This exper- nal was detected by a plate reader fluorometer. Cells expressing HA- iment demonstrated that the D1 and D2 dopamine receptors formed tagged D1-NLS receptors were treated with a range of (ϩ)BTC concen- heterooligomers. However, in some of the cells, both receptors trations (10 nM to 1 ␮M). A dose-dependent effect of (ϩ)BTC was appeared to remain at the cell surface, when the ratio of the D2-NLS to 207 NOVEMBER 4, 2005•VOLUME 280•NUMBER 44 JOURNAL OF BIOLOGICAL CHEMISTRY 37231 Visualization of Dopamine Receptor Oligomers

FIGURE 9. The effect of a structural conformational difference on heterooligomer formation. a, cells co-expressing amino-D2-NLS (GFP) (green) and D1(S198A/S199A) (RFP) (red) showed that amino-D2-NLS had trafficked to the nucleus, whereas D1(S198A/S199A) remained on the cell ϩ surface. b, when treated with ( )BTC, both recep- Downloaded from tors were found on the cell surface, but after drug wash off, the receptors separated, with amino-D2- NLS translocating to the nucleus and D1(S198A/ S199A) remaining on the cell surface. c, amino-D2- NLS co-expressed with D1(S199A/S202A) showed that both trafficked together to the nucleus. D, when treated with (ϩ)BTC, both receptors were

co-localized on the cell surface. e, after drug wash www.jbc.org off, both receptors trafficked together to the nucleus, indicating an intact heterooligomer. at University of Toronto on July 11, 2007

D1 expression was not equivalent and markedly lesser amounts of the together to the nucleus when the drug was removed (35% of cells in NLS-containing receptor were present. nucleus only, 65% of cells in nucleus and cell surface). Thus, the confor- In order to investigate the effect of conformational change in one or mational change occurring in one constituent of the heterooligomer the other or both protomers within the heterooligomer, we examined could not be imparted to the other constituent, as indicated by the the effect of selectively or jointly manipulating the receptor conforma- inability of the SCH 23390 occupying the D1 receptor to affect the tions. Cells expressing amino-D2-NLS and D1 were treated with the conformation of the D2-NLS. selective D1 receptor antagonist SCH 23390 (500 nM) for 24 h, and cells We also located the NLS in intracellular loop one of the D2 receptor were observed 24 h after drug wash off. In the presence of SCH 23390, (D2-IC1-NLS), distant from the carboxyl tail, and this receptor also both amino-D2-NLS and D1 receptors trafficked together robustly to translocated efficiently to the nucleus (100% of cells). The heterooligo- the nucleus (90% of cells) (Fig. 8b). The receptor distributions remained meric association between D1 and D2 receptors was confirmed with unaltered after drug removal. Treatment with the D2-selective antago- this construct as well, since the D1 receptor translocated with the nist raclopride (500 nM) in a similar experimental design revealed, in the D2-IC1-NLS to the nucleus (100% of cells). presence of drug, that both receptors were localized together on the cell To probe the effect of a structural conformational difference in the surface (100% of cells) (Fig. 8c). After raclopride removal, both receptors stability of the heterooligomer, we co-expressed amino-D2-NLS and moved off the cell surface and translocated to the nucleus (56% of cells in D1(S198A/S199A). These cells showed that amino-D2-NLS had traf- the nucleus only and 44% of cells nucleus and cell surface). The effect of ficked to the nucleus (100% of cells), whereas D1(S198A/S199A) co-manipulation of both receptor conformations simultaneously was remained on the cell surface (100% of cells) (Fig. 9a). When treated with tested by treating cells expressing amino-D2-NLS and D1 receptors (ϩ)BTC or raclopride (500 nM), both receptors were found on the cell with (ϩ)BTC (500 nM). When (ϩ)BTC was present, both receptors surface, but after drug wash off, the receptors were again seen to sepa- were localized on the cell surface (100% of cells) (Fig. 8d) and moved off rate, with amino-D2-NLS translocating to the nucleus (100% of cells) 208 37232 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 280•NUMBER 44•NOVEMBER 4, 2005 Visualization of Dopamine Receptor Oligomers and D1(S198A/S199A) remaining on the cell surface (100% of cells) (Fig. 9b). We also expressed amino-D2-NLS with D1(S199A/S202A), and both trafficked together to the nucleus under basal conditions (100% of cells) (Fig. 9c). When treated with (ϩ)BTC (500 nM), both receptors were co-localized on the cell surface (100% of cells) (Fig. 9d). After drug wash off, both receptors trafficked together to the nucleus (97% of cells) (Fig. 9e), indicating that this heterooligomer remained intact. The incorporation of the NLS into the D2 receptor did not alter the binding pocket of the receptor, with preserved agonist-detected high affinity and low affinity states, indicative of intact receptor-G protein ϫ Ϫ9 coupling. The D2 receptor had a KHigh value of 1.51 10 M and KLow ϫ Ϫ6 of 6.67 10 M for quinpirole. Similarly, D2-NLS had a KHigh value of ϫ Ϫ9 ϫ Ϫ6 3.22 10 M and KLow of 4.16 10 M for quinpirole, not different from the D2 receptor. Formation of D5 and D1 Dopamine Receptor Heterooligomers—The D5 dopamine receptor structure shares extensive homology with the D1 receptor; however, there are pronounced sequence differences in the carboxyl tail regions of these receptors. When expressed in cells, the D5 Downloaded from was located primarily at the cell surface and also in the cytoplasm. The D5 was modified with the NLS incorporated in helix 8 (D5-NLS). How- ever, D5-NLS did not translocate as efficiently as D1-NLS to the nucleus. Additionally, the D5 receptor with an NLS incorporated into intracellular loop one (D5-IC1-NLS) also did not translocate as efficiently to the nucleus, compared with D1-IC1-NLS. These experiments FIGURE 10. Oligomerization of D1 and D5 dopamine receptors. a, D1-NLS (GFP) www.jbc.org indicated significant differences in the structural conformation of the (green) and D5 (RFP) (red) co-translocation to the nucleus. b, HA-tagged dopamine D5 receptor (HA-D5) expressed in HEK cells. Cells expressing HA-D5 were co-expressed with D5 carboxyl tail compared with D1. In support of this, D5-IC1-NLS with FLAG-D1-NLS. The HA-D5 cell surface receptors were quantified by the fluorescent signal a carboxyl tail truncated at Asp399 was very efficiently translocated to of an FITC-conjugated second antibody directed at the HA antibody. Co-expression with the nucleus (95% of cells). From these results, we suspected that the GFP FLAG-D1-NLS resulted in a decrease in cell surface expression of HA-D5, indicating that at University of Toronto on July 11, 2007 FLAG-D1-NLS formed oligomers with HA-D5. moiety attached to the carboxyl tail of the D5 dopamine receptor in the above experiments may be positioned by the carboxyl tail conformation native for the study of properties of GPCR oligomers not possible to sterically hinder access by the importin protein to the NLS. A con- otherwise. struct with the GFP moiety placed in the amino terminus of the receptor We visualized that the D1 receptor formed homooligomers with (amino-D5) when expressed with D1-NLS showed that these receptors D1-NLS and readily trafficked with it to the nuclear compartment. translocated together to the nucleus (85% of cells), confirming that the When the co-expressed D1 and D1-NLS receptors were treated with D1 and D5 receptors formed heterooligomers (Fig. 10a). (ϩ)butaclamol or SCH 23390, both forms remained on the cell surface, Removal of D5 Receptor from the Cell Surface by D1-NLS—Using and then, when the drug was removed, both trafficked together off the fluorescence cell surface labeling, HA-D5 cell surface detection was cell surface to the nucleus. These data indicated that these receptors measured. Cells co-expressing HA-D5 receptors with increasing interacted to form an oligomer, robustly enough to survive co-traffick- amounts of FLAG-D1-NLS revealed a robust effect of FLAG-D1-NLS to ing to another cellular compartment. Since any conformational change remove the HA-D1 from the cell surface (Fig. 10b). in the receptors induced by the drug would occur in all the receptors within the D1/D1-NLS oligomer and would therefore be identical, the DISCUSSION oligomer was seen to remain intact through antagonist occupancy and after drug wash off. To investigate the characteristics of the oligomers We developed a novel strategy that enabled elucidation of some formed between D1 and D1-NLS, two amino acid mutations were intro- structural features and architecture of dopamine receptor homooli- duced into transmembrane domain V of the receptor at Ser198 and gomers and heterooligomers in living cells. The method, unlike static Ser199. The D1(S198A/S199A) receptor did not bind to agonist or the in-cell fluorescent methodologies currently available, was able to probe antagonist/inverse agonist SCH 23390 but was shown to bind the antag- structural variations within dopamine receptor oligomers and test the onist/inverse agonist (ϩ)-butaclamol. When expressed individually robustness of the receptor-receptor interactions, including demonstra- with its NLS-containing counterpart, these receptors were co-trafficked tion of disruption of the oligomers from the cell surface in living cells. and oligomerized. The oligomers between D1(S198A/S199A)-NLS and These are aspects that cannot be evaluated by any current methodology. D1(S198A/S199A) were not separated and remained intact, whereas the The occupancy of the ligand-binding pocket by an antagonist/inverse oligomers formed between D1 and D1(S198A/S199A) receptors, could agonist was able to shift the conformation of the D1 receptor sufficiently be disrupted and separated. This suggested that these two receptors, the to disrupt the oligomer and result in separation of the components when D1 and D1(S198A/S199A), with the small structural difference in trans- there was a receptor with a structural variation expressed with the wild type membrane V, were structurally similar in the basal state to oligomerize receptor. These data provide, for the first time, evidence of the ability to but sufficiently distinct conformationally when the wild type receptor regulate GPCR oligomers at the cell surface, showing that the relative was bound by (ϩ)-butaclamol for the oligomer to separate. An overall activation state or the conformation of oligomers contributes to interpretation of these data is that stable oligomers only form between whether oligomers remain together or separate from one another. receptors that have identical conformations and that introduction of a These capabilities inherent within this method provide a unique alter- conformational difference between them may be a mechanism by which 209 NOVEMBER 4, 2005•VOLUME 280•NUMBER 44 JOURNAL OF BIOLOGICAL CHEMISTRY 37233 Visualization of Dopamine Receptor Oligomers oligomer size is delimited and regulated. It is intriguing to speculate that fluorometric analysis, the HA-tagged D5-NLS was efficiently translo- this may provide a physiological mechanism for oligomer disruption, cated from the cell surface. This indicated that the positioning of the whereby the ligand-occupied receptor, which would be in an activated GFP impeded importin access to NLS binding. In confirmation of this, conformation, could separate from neighboring receptors that are not D5-IC1-NLS with a truncated carboxyl tail or D5 with the GFP posi- activated and therefore not in an identical conformation. However, tioned at the amino terminus translocated efficiently to the nucleus. since the building block of GPCR oligomers appears to be a dimer, it Our strategy to examine the properties of the dopamine receptor remains to be determined whether this process could separate the oligomers capitalized on the physiological mechanism whereby receptors into monomers or rather break up an oligomer into dimeric sequences specifying an NLS, when incorporated into cytoplasmic pro- components. Indeed, it has been suggested that the weakest interaction teins, enabled the proteins to be relocated to the nucleus by the importin may be between dimer rows within oligomeric arrays (5). translocation pathway. This is a well characterized process, the mecha- D1(S199A/S202A) bound both antagonist/inverse agonist ligands nism of which is well understood (10, 19, 32). Therefore, the incorpora- similarly to the D1 receptor, implying that this receptor was less altered tion of an NLS into a GPCR and its translocation from the cell surface conformationally, compared with wild-type D1, than D1(S198A/ toward and into the nucleus formed the basis for the investigation of S199A). This was confirmed by the robustness of its oligomerization to dopamine receptor oligomerization. D1 and the inability to separate these proteins by ligand occupancy. In summary, we have developed an engineered translocation method The specific sites of interaction maintaining the monomeric receptor that enables the study of GPCR oligomer complexes directly in living components in both dimer and oligomer formation have yet to be com- cells. This method provides insight into the dynamics of these completely elucidated. We and others have reported that transmembrane 4 plexes in situ following changes in configuration of the component is involved in receptor dimer formation (5, 16, 17), and other transmem- receptors, introduced by selective mutagenesis or that occur with antag- Downloaded from branes have been reported to participate in oligomer formation (18, 31). onist occupancy of the binding pocket. This strategy provides an Thus, alterations in the receptor hydrophobic core, particularly residues entirely novel approach to examine homo- and heterooligomerization containing the ligand-binding site, predictably may affect the receptor in Family A GPCRs. We have shown that an oligomer composed of conformation, such as in D1(S198A/S199A). homogenous structural units behaves as a single complex, its integrity We also showed the ability of dopamine D1 and D2 receptors to

due to receptor-receptor interactions maintained even when trafficking www.jbc.org oligomerize and traffic together. In this case, the use of selective antag- to a distant cellular compartment. However, an oligomer composed of onists to attempt to introduce a conformational difference showed that heterogeneous units, even if generated from the same GPCR, was prone manipulation of one receptor could not influence the conformation of to easier disruption by altering the conformation of the components. the other. An explanation for this may be that the D1 and D2 receptors This may provide a means of physiological regulation of the size and at University of Toronto on July 11, 2007 do not form heterodimers, although they form heterooligomers. It is kinetics of an oligomer. In addition, the clear segregation of the differ- possible that heterooligomers are formed between arrays of entially tagged component receptors upon oligomer disruption sug- homodimers but heterodimers are not formed. The separation induced gested that they were heterooligomeric and probably not heterodimeric by the antagonist with heterooligomers may suggest that they were in nature. Furthermore, we were also able to demonstrate that small never heterodimers. The inability of the D1(S198A/S199A) to oligomer- conformational changes could prevent heterooligomerization, suggest- ize with the D2 receptor indicated that the conformational alteration ing that alterations in the structure may affect the interaction interfaces. induced in the D1 receptor impaired an interface necessary for D1 and We have shown, therefore, that the multisubunit GPCR oligomeric D2 heterooligomerization. However, since the D1(S198A/S199A) was complexes can be regulated at the cell surface, and these data provide able to oligomerize with the wild type D1 receptor, it indicated a lesser insight into the dynamic nature of the structure and the stability of these conformational difference between these two receptors. Whether this protein complexes within cells. indicates that different interfaces are responsible for the assembly of homooligomers versus heterooligomers remains to be shown. We have REFERENCES shown that the basal intrinsic activity of D1(S198A/S199A) on adenylyl 1. George, S. R., O’Dowd, B. F., and Lee, S. P. (2002) Nat. Rev. Drug Discovery 1, 808–820 cyclase activation was comparable with that of the wild type D1 recep- 2. Milligan, G. (2004) Mol. Pharmacol. 66, 1–7 tor.4 This, together with the demonstrated ability of this receptor to 3. Bouvier, M. (2001) Nat. Rev. Neurosci. 2, 274–286 bind and respond to (ϩ)-butaclamol, indicated that the conformational 4. Fotiadis, D., Liang, Y., Filipek, S., Saperstein, D. A., Engel, A., and Palczewski, K. (2003) Nature 421, 127–128 change resulting from the mutation of the two serine residues in trans- 5. Liang, Y., Fotiadis, D., Filipek, S., Saperstein, D. A., Palczewski, K., and Engel, A. (2003) membrane domain V has not led to a major disruption of receptor J. Biol. Chem. 278, 21655–21662 structure but has been relatively small. 6. Lee, S. P., O’Dowd, B. F., Ng, G. Y., Varghese, G., Akil, H., Mansour, A., Nguyen, T., The data also reveal differences in the nuclear translocation of the and George, S. R. (2000) Mol. Pharmacol. 58, 120–128 7. Jordan, B. A., and Devi, L. A. (1999) Nature 399, 697–700 highly homologous D1 and D5 receptors when modified with the inser- 8. George, S. R., Fan, T., Xie, Z., Tse, R., Tam, V., Varghese, G., O’Dowd, B. F. (2000) tion of a NLS, which indicated the novel conclusion of significant con- J. Biol. Chem. 275, 26128–26135 formational differences between the intracellular domains of these 9. McVey, M., Ramsay, D., Kellett, E., Rees, S., Wilson, S., Pope, A.J., and Milligan, G. receptors, despite the extensive similarities in the primary structures (2001) J. Biol. Chem. 276, 14092–14099 10. Lee, D. K., Lanca, A. J., Cheng, R., Nguyen, T., Ji, X. D., Gobeil, F., Jr., Chemtob, S., between D1 and D5 receptors (75% overall). The D1 dopamine receptor, George, S. R., and O’Dowd, B. F. (2004) J. Biol. Chem. 279, 7901–7908 fused with GFP with the NLS incorporated in IC1, IC3, or helix 8 each 11. Gomes, I., Jordan, B.A., Gupta, A., Trapaidze, N., Nagy, V., and Devi, L. A. (2000) resulted in a robust translocation to the nucleus, whereas each of these J. Neurosci. 20, Rapid Communication 110, 1–5 equivalent modifications made with the D5 (also fused with GFP) 12. Levac, B. A., O’Dowd, B. F., and George, S. R. (2002) Curr. Opin. Pharmacol. 2, 76–81 showed less robust nuclear localization. However, as determined by 13. Lee, S. P., So, C. H., Rashid, A. J., Varghese, G., Cheng, R., Lanca A. J., O’Dowd, B. F., and George, S. R. (2004) J. Biol. Chem. 279, 35671–35678 14. Gales, C., Rebois, R. V., Hogue, M., Trieu, P., Breit, A., Hebert, T. E., and Bouvier, M. (2005) Nat. Methods 2, 177–184 4 M. C. Kong, G. Varghese, T. Fan, S. P. Lee, B. F. O’Dowd, and S. R. George, submitted for 15. Rocheville, M., Lange, D. C., Kumar, U., Patel, S. C., Patel, R. C., and Patel, Y. C. (2000) publication. Science 288, 154–157 210 37234 JOURNAL OF BIOLOGICAL CHEMISTRY VOLUME 280•NUMBER 44•NOVEMBER 4, 2005 Visualization of Dopamine Receptor Oligomers

16. Lee, S. P., O’Dowd, B. F., Rajaram, R. D., Nguyen, T., and George, S. R. (2003) Bio- 25. Zawarynski, P., Tallerico, T., Seeman, P., Lee, S. P. O’Dowd, B. F., and George, S. R. chemistry 42, 11023–11031 (1998) FEBS Lett. 441, 383–386 17. Guo, W., Shi, L., and Javitch, J. A. (2003) J. Biol. Chem. 278, 4385–4388 26. So, C. H., Varghese, G., Curley, K. J., Kong, M. M., Alijaniaram, M., Ji, X., Nguyen, T., 18. Carrillo, J. J., Lopez-Gimenez, J. F., and Milligan, G. (2004) Mol. Pharmacol. 66, O’Dowd, B. F., and George, S. R. (2005) Mol. Pharmacol. 68, 568–578 1123–1137 27. Prasher, D. C., Eckenrode, V. K., Ward, W. W., Prendergast, F. G., and Cormier, M. J. 19. Jans, D. A., Xiao, C-Y., and Lam, M. H. C. (2000) BioEssays 22, 532–544 (1992) Gene (Amst.) 111, 229–233 20. Lu, D., Yang, H., Shaw, G., and Raizada, M. K. (1998) Endocrinology 139, 365–375 28. Matz, M. V., Fradkov, A. F., Labas, Y. A., Savitsky, A. P., Zaraisky, A. G., Markelov, 21. Chen, R., Mukhin, Y. V., Garnovskaya, M. N., Thielen, T. E., Iijima, Y., Huang, C., M. L., and Lukyanov, S. A. (1999) Nat. Biotechnol. 17, 969–973 Raymond, J. R., Ullian, M. E., and Paul, R. V. (2000) Am. J. Physiol. 279, F440–F448 29. Marchese, A., George, S. R., and O’Dowd, B. F. (1998) in Identification and Expression 22. Watson, P. H., Fraher, L. J., Natale, B. V., Kisiel, M., Hendy, G. N., and Hodsman, A. B. of G Protein Coupled Receptors (Lynch, K., ed) pp. 1–26, Wiley-Liss, New York (2000) Bone 26, 221–225 30. Tomic, M., Seeman, P., George, S. R., and O’Dowd, B. F. (1993) Biochem. Biophys. Res. 23. Ng, G. Y. K., Trogadis, J., Stevens, J., Bouvier, M., O’Dowd, B. F., and George, S. R. Commun. 191, 1020–1027 (1995) Proc. Natl. Acad. Sci. U. S. A. 92, 10157–10161 31. Overton, M. C., and Blumer, K. J. (2002) J. Biol. Chem. 277, 41463–41472 24. Ng, G. Y., O’Dowd, B. F., Lee, S. P., Chung, H. T., Brann, M. R., Seeman, P., George, 32. Jong, Y. J., Kumar, V., Kingston, A. E., Romano, C., and O’Malley, K. L. (2005) J. Biol. S. R. (1996) Biochem. Biophys. Res. Commun. 227, 200–204 Chem. 280, 30469–30480 Downloaded from www.jbc.org at University of Toronto on July 11, 2007

211 NOVEMBER 4, 2005•VOLUME 280•NUMBER 44 JOURNAL OF BIOLOGICAL CHEMISTRY 37235 6.4 D1-D2 Dopamine Receptor Hetero-oligomers With Unique Pharmacology Are

Coupled To Rapid Activation Of Gq/11 In The Striatum

6.4.1 Summary

This study demonstrates the existence of a novel D1-D2 dopamine receptor hetero-

olgiomeric complex that can trigger calcium release through coupling to Gq/11. This novel

signalling complex could be activated specifically by SKF83959, resulting in phospholipase C

activation by full and partial agonist activity at D1 and D2 receptors within the oligomer,

respectively. This signaling complex could be detected more readily in older mice and activation

through Gq/11 could also trigger increases in CaMKIIα suggesting a potential role for this in synaptic plasticity.

6.4.2 Contribution by thesis author

The author generated the data presented in the Supplementary Data, Figure 8, and assisted in the preparation of the manuscript.

212 D1–D2 dopamine receptor heterooligomers with unique pharmacology are coupled to rapid activation of Gq/11 in the striatum Asim J. Rashid*†, Christopher H. So*, Michael M. C. Kong*, Teresa Furtak*, Muﬁda El-Ghundi*, Regina Cheng†, Brian F. O’Dowd*†, and Susan R. George*†‡§

Departments of *Pharmacology and ‡Medicine, University of Toronto, Toronto, Ontario, Canada M5S 1A8; and †Centre for Addiction and Mental Health, Toronto, Ontario, Canada M5T 1R8

Edited by Robert J. Lefkowitz, Duke University Medical Center, Durham, NC, and approved November 9, 2006 (received for review May 17, 2006) We demonstrate a heteromeric D1–D2 dopamine receptor signal- Results ing complex in brain that is coupled to Gq/11 and requires agonist We examined calcium signaling through D1 and D2 dopamine binding to both receptors for G protein activation and intracellular receptors that were stably coexpressed in human embryonic kidney calcium release. The D1 agonist SKF83959 was identified as a cells (D1–D2HEK cells). A robust dose-dependent transient rise in specific agonist for the heteromer that activated Gq/11 by func- calcium caused by release from intracellular stores was observed tioning as a full agonist for the D1 receptor and a high-affinity after coapplication of the D1 receptor agonist SKF81297 and the partial agonist for a pertussis toxin-resistant D2 receptor within D2 receptor agonist quinpirole (Fig. 1 a and b) or with application the complex. We provide evidence that the D1–D2 signaling com- of dopamine [supporting information (SI) Fig. 6]. The rise in plex can be more readily detected in mice that are 8 months in age calcium was abolished by the selective D1 receptor antagonist compared with animals that are 3 months old, suggesting that SCH23390 and by the selective D2 receptor antagonist raclopride calcium signaling through the D1–D2 dopamine receptor complex (Fig. 1c). In D1–D2HEK cells treated with SKF81297, there was a is relevant for function in the postadolescent brain. Activation of smaller rise in calcium (Fig. 1 a and b) not seen in cells expressing Gq/11 through the heteromer increases levels of calcium/calmod- D1 alone (data not shown), which could also be blocked by ␣ ulin-dependent protein kinase II in the nucleus accumbens, unlike SCH23390 or raclopride (Fig. 1c). The ability of raclopride to blunt activation of Gs/olf-coupled D1 receptors, indicating a mechanism the signal indicated a role for the D2 receptor in the signal by which D1–D2 dopamine receptor complexes may contribute to generated by SKF81297 and suggested that this agonist could synaptic plasticity. directly activate the D2 receptor. Because treatment of D1–D2HEK cells or D2 cells with quinpirole alone did not stimulate calcium ͉ ͉ ͉ HEK heterooligomerization SKF83959 calcium signaling release (data not shown), calcium release appeared to depend on ␣ calcium/calmodulin-dependent protein kinase II coordinated activation of both D1 and D2 receptors. We demonstrated that the effects of D1–D2 receptor activation iverse roles for each of the five dopamine receptors (D1–D5) occurred by Gq/11 activation of PLC, producing inositol trisphos- Dhave been shown to be initiated primarily through stimulation phate (IP3), which can act on intracellular IP3 receptors to release or inhibition of adenylyl cyclase (AC) via Gs/olf or Gi/o signaling calcium (Fig. 1d), and was independent of AC modulation (SI Fig. proteins, respectively (1). There have been reports, however, of a 7). Treatment of D1–D2HEK cells with the PLC inhibitor U71322 or D1-like receptor in brain that is coupled to Gq/11, stimulating thapsigargin, a depletor of intracellular calcium stores, eliminated phospholipase C (PLC) and intracellular calcium release (2–5). the calcium signal. The signal was also eliminated by 2-aminoe- Activation of this Gq/11-coupled D1-like receptor by specific re- thoxydiphenyl borate, an antagonist of intracellular IP3 receptors. ceptor agonists does not correlate with the ability of these same A definitive role for Gq/11 as the initiator of this cascade was agonists to activate AC (4), suggesting that the Gq/11-coupled established by using the Gq/11 inhibitor YM254890 (8), which D1-like receptor is a molecular entity distinct from the Gs/olf- abolished rises in calcium in response to SKF81297 and quinpirole coupled D1 receptor. (Fig. 1d). Molecular identification of the Gq/11-coupled D1-like receptor D1 receptor agonists have varying abilities to activate AC or has proven elusive because D1 receptor coupling to PLC has not phosphoinositide (PI) hydrolysis in brain (2, 4). Although been demonstrated in a variety of cell types in which the D1 SKF81297 is a potent activator of both AC and PI turnover, receptor was expressed. We had postulated that Gq/11 activation by SKF83822 has been shown to activate only AC, and SKF83959 D1 receptor agonists in brain could occur by concurrent activation selectively triggers PI hydrolysis. To see whether there was a similar of the D1 receptor and the D2 receptor (6). We have shown that heterologously coexpressed D1 and D2 dopamine receptors formed heterooligomers (7) and that coactivation of these receptors re- Author contributions: B.F.O. and S.R.G. contributed equally to this paper; A.J.R., C.H.S., sulted in a PLC-dependent rise in intracellular calcium (6). We also M.M.C.K., and S.R.G. designed research; A.J.R., C.H.S., M.M.C.K., T.F., and R.C. performed demonstrated that D1 and D2 receptors could be coimmunopre- research; M.E.-G. contributed new reagents/analytic tools; A.J.R., C.H.S., M.M.C.K., T.F., B.F.O., and S.R.G. analyzed data; and A.J.R., B.F.O., and S.R.G. wrote the paper. cipitated from striatal membranes (6). These results suggested the possibility of a unique signaling complex in brain composed of The authors declare no conflict of interest. PLC-coupled D1–D2 receptor heterooligomers. This article is a PNAS direct submission. In this work we report the presence of such a D1–D2 dopamine Abbreviations: AC, adenylyl cyclase; CaMKII␣, calcium/calmodulin-dependent protein kinase II␣; GTP␥S, guanosine 5Ј-␥-thiotriphosphate; IP3, inositol trisphosphate; PLC, phos- receptor signaling complex in striatum that is coupled to rapid pholipase C; PTX, pertussis toxin. G /11 signaling on activation of both receptors and which can be q §To whom correspondence should be addressed at: Department of Pharmacology, Univer- defined by a unique pharmacology. The complex was more readily sity of Toronto, MSB Room 4358, Toronto, ON, Canada M5S 1A8. E-mail: s.george@ detected in older mice and could modulate levels of calcium/cal- utoronto.ca. modulin-dependent protein kinase II␣ (CaMKII␣) in the nucleus This article contains supporting information online at www.pnas.org/cgi/content/full/ accumbens, indicating a potential role for the D1–D2 heteromer in 0604049104/DC1. synaptic plasticity in the postadolescent brain. © 2006 by The National Academy of Sciences of the USA

213 654–659 ͉ PNAS ͉ January 9, 2007 ͉ vol. 104 ͉ no. 2 www.pnas.org͞cgi͞doi͞10.1073͞pnas.0604049104 Fig. 2. Specificity of D1 receptor agonists for the D1–D2 calcium signal. (a) SKF83959 (1 ␮M) or SKF81297 (1 ␮M) stimulated calcium in D1–D2HEK cells (n ϭ 6), which was increased by 2.18 Ϯ 0.071-fold and 2.66 Ϯ 0.038-fold, respectively, by the coaddition of 1 ␮M quinpirole (n ϭ 8). SKF83822 (1 ␮M) application did not stimulate any increases in intracellular calcium (n ϭ 4), and coapplication of quinpirole had no effect (n ϭ 4). AFU, arbitrary fluorescence units. (b) Quantification of 35S-labeled guanosine 5Ј-␥-thiotriphosphate ([35S]GTP␥S) incorporation into immunoprecipitated G proteins demonstrates activation of Gq/11 (67.2 Ϯ 8.4%) and Gi3 (20.0 Ϯ 3.3%) but not Gs (Ϫ6.9 Ϯ 12.1%) in response to SKF83959 and quinpirole (10 ␮M each) (n ϭ 4). SKF83959 alone (10 ␮M) increased incorporation into Gq/11 (28.6 Ϯ 9.3%), did not significantly affect Gi (8.2 Ϯ 4.6%), and incorporation into Gs was slightly decreased (Ϫ17.0 Ϯ 11.0%) (n ϭ 4). Treatment of cells with pertussis toxin (PTX) abolished Gi3 activation by SKF83959 and quinpirole treatment but only slightly affected Gq/11 activation (38.1 Ϯ 11.6%) (n ϭ 3). SKF83822 application resulted in robust activation of Gs (140.0 Ϯ 37.0%), modest activation of Gq/11 Fig. 1. Calcium signaling through stably expressed D1 and D2 dopamine (15.2 Ϯ 4.3%), and no significant activation of Gi3 (12.9 Ϯ 6.3%) (n ϭ 4). The 35 receptors is caused by Gq/11-mediated activation of PLC. (a) Changes in dashed line represents basal levels of [ S]GTP␥S incorporation in the absence fluorescence corresponding to changes in intracellular calcium levels on treat- of agonist. *, P Ͻ 0.05; **, P Ͻ 0.005; Student’s t test (for b, compared with 35 ment of D1–D2HEK cells with SKF81297 (1 ␮M) or SKF81297 and quinpirole (1 [ S]GTP␥S incorporation in the absence of agonist).

␮M each). The time of agonist addition is indicated with open arrow. AFU, NEUROSCIENCE arbitrary fluorescence units. (b) Dose–response curves of peak calcium levels in response to agonist (EC50(SKF81297ϩquinpirole), 50.8 Ϯ 8.8 nM, n ϭ 8; EC50(SKF81297), 147.6 Ϯ 46.9 nM, n ϭ 8). (c) Treatment of cells with 10 ␮M SCH23390 (SCH) or did not stimulate any rises in intracellular calcium, with or without raclopride (Rac) abolished agonist (1 ␮M)-mediated rises in calcium (n ϭ 5). (d) coapplication of quinpirole. Rises in calcium in response to SKF81297 and quinpirole (1 ␮M each) were The differential ability of SKF83822 and SKF83959 to modulate eliminated by the IP3 receptor blocker 2-aminoethoxydiphenyl borate (2-APB; intracellular levels of cAMP and calcium indicates differences in 100 ␮M) as well as by depletion of intracellular calcium stores with thapsigar- their ability to activate G -coupled D1 receptors and G /11-coupled ␮ ␮ ϭ s q gin (TG; 1 M) or inhibition of PLC with U71322 (50 M) (n 6 for all). The D1–D2 receptor complexes. To test this theory, we treated mem- inactive isomer of U71322, U73343, did not abolish the calcium signal, although the effect of SKF81297 and quinpirole was reduced by 18.9 Ϯ 7.9% branes from D1–D2HEK cells with agonists in the presence of 35 ␥ 35 (n ϭ 4). The Gq/11 inhibitor YM-254890 (YM; 100 nM) blocked increases in [ S]GTP S and quantified the incorporation of S into immuno- calcium in D1–D2HEK cells in response to SKF81297 and quinpirole (n ϭ 5). precipitated G␣ proteins as a measure of their activation. In Background levels of fluorescence were qualitatively determined from indi- response to SKF83959 and quinpirole, [35S]GTP␥S incorporation vidual fluorescence profiles but were generally considered to be below 3,500 into G /11 and G was increased over basal levels, whereas there was ϩ Ͻ Ͻ q i AFU. , P 0.05; **, P 0.0001; Student’s t test compared with corresponding no change in incorporation into G (Fig. 2b). Preincubation of control. s membranes with PTX eliminated incorporation into Gi and did not significantly affect incorporation into Gq/11 when compared with agonist selectivity profile for these drugs on the D1–D2 calcium that in the absence of PTX, indicating that the effects of quinpirole signal, we first confirmed that only SKF81297 and SKF83822 could on calcium release were mediated through potentiation of Gq/11 activate AC through D1 receptors in a dose-dependent manner, activation and were independent of Gi/o activation. Treatment with 35 whereas SKF83959 could not (SI Fig. 8). The SKF compounds were SKF83959 alone increased [ S]GTP␥S incorporation into Gq/11 then compared for their ability to trigger intracellular calcium but to a lesser extent than with quinpirole coapplication. Neither Gs release in D1–D2HEK cells (Fig. 2a). In response to SKF83959 or nor Gi was activated by SKF83959 alone. SKF81297, a rise in calcium was observed that was significantly In contrast to SKF83959, SKF83822 minimally activated Gq/11, increased by coapplication of quinpirole. By comparison, SKF83822 and it did not affect Gi but robustly activated Gs (Fig. 2b).

214 Rashid et al. PNAS ͉ January 9, 2007 ͉ vol. 104 ͉ no. 2 ͉ 655 that depends on the presence and possibly activation of the D1 receptor. To test this hypothesis, the ability of the SKF agonists to displace [3H]raclopride binding to the D2 receptor competitively was examined in D1–D2HEK cells and D2HEK cells (Fig. 3 and Table 1). Competition binding profiles of SKF83959 on [3H]raclopride binding in D1–D2HEK cells revealed a high-affinity binding site for SKF83959 on the D2 receptor that was not observed in D2HEK cells (Fig. 3a). Pretreatment of D1–D2HEK cells with PTX modestly reduced but did not eliminate the proportion of high-affinity binding sites for SKF83959 (Fig. 3b), demonstrating that the majority of SKF83959 binding was to PTX-resistant and not Gi/o-coupled D2 receptors. This PTX-resistant site overlapped with or was the same as the binding site for quinpirole because incubation of membranes with quinpirole competitively displaced SKF83959 binding from the high-affinity site (Fig. 3c)(nH Ϸ1) (Table 1). Incubation of membranes with SCH23390 did not affect high-affinity binding of SKF83959 to the D2 receptor (Fig. 3d), indicating that the site was distinct from the D1 receptor and that it was present in the D2 receptor basal state in the absence of ligand occupancy or activation of the D1 receptor. Competition by SKF81297 of [3H]raclopride binding similarly revealed a high-affinity PTX-resistant binding site on the D2 receptor in D1–D2HEK cells (data not shown). However, PTX- sensitive binding of SKF81297 to the D2 receptor in D2HEK cells was also observed. For SKF83822, a proportion of [3H]raclopride binding could be displaced by agonist in both D2HEK cells and D1–D2HEK cells that could be abolished by PTX (Fig. 3e), reflecting high-affinity binding to Gi/o-coupled D2 receptors in both cell lines. The competition binding results reveal a distinct pharmacology Fig. 3. SKF83959 binds with high affinity to PTX-resistant D2 receptors only of the D2 receptor in D1–D2HEK cells such that the D1 receptor in the presence of D1 receptors. Competition of [3H]raclopride binding by agonists SKF81297 and SKF83959 but not SKF83822 can act as SKF83959 or SKF83822 is shown. Data from three to eight independent ligands for a PTX-resistant D2 receptor when it is coexpressed with experiments conducted in duplicate were normalized and fit to one-site or the D1 receptor. Taken together with the calcium-signaling data, two-site analysis. (a) Comparison of binding on membranes from D2HEK cells binding of the D1 receptor agonists to the D2 receptor site indicates and D1–D2HEK cells reveals a high-affinity binding site for SKF83959 only in Ϯ Ϯ partial agonism of D2 receptors within the Gq/11-coupled D1–D2 D1–D2 cells (KH, 2.4 0.8 nM; %KH,19 1.5). (b) High-affinity binding of receptor complex, therefore allowing a single agonist to activate SKF83959 to D2 receptors in D1–D2HEK cells was only slightly affected by pretreatment with PTX (K , 1.9 Ϯ 1.3 nM; %K ,11Ϯ 3.3). (c) Incubation of both members of the heteromer. In accordance with this concept, H H the addition of quinpirole along with SKF81297 or SKF83959 would D1–D2HEK membranes with quinpirole (10 nM) eliminated high-affinity binding of SKF83959. (d) Incubation of D1–D2HEK membranes with SCH23390 (10 result in full agonism at the PTX-resistant high-affinity state of the nM) did not reduce affect high-affinity binding of SKF83959. (e) Competition D2 receptor within the complex and a greater calcium signal. 3 binding of [ H]raclopride by SKF83822 indicated high-affinity binding of the To determine whether Gq/11-coupled D1–D2 receptor signaling 35 agonist to D2 receptors in D1–D2HEK cells that was eliminated by pretreatment complexes exist in the brain, [ S]GTP␥S incorporation into Gq/11 of cells with PTX. from murine striatal membranes was quantified after membranes had been treated with SKF83959 alone or with equivalent concentrations of quinpirole. Initial experiments used striata from 12- Coapplication of quinpirole with SKF83822 did not affect activation week-old male mice, but Gq/11 activation in response to agonists of Gq/11orGs (data not shown). was not reliably observed. A consistent agonist-dependent increase 35 The generation of a raclopride-sensitive calcium signal with in [ S]GTP␥S incorporation into Gq/11 could be elicited, however, SKF81297 or SKF83959 treatment of D1–D2HEK cells suggests that when older animals (Ն8 months old) were used (Fig. 4a). Treat- these two drugs can act as agonists for D2 receptors in a manner ment with SKF83959 and quinpirole gave significant increases in

Table 1. Competition binding studies with ͓3H͔raclopride

Row Cell line Agonist Treatment nH KH,nM KL,nM Ki,nM RH,% n

† a D1–D2HEK SKF83959 — Ϫ0.68 Ϯ 0.02 2.38 Ϯ 0.80 319 Ϯ 37 19.1 Ϯ 1.5 8 Ϫ Ϯ ‡ Ϯ bD2HEK SKF83959 — 0.90 0.10* NA NA 346 31 NA 3 Ϫ Ϯ Ϯ Ϯ Ϯ c D1–D2HEK SKF83959 PTX 0.79 0.06 1.90 1.3* 351 27 11 3.3* 4 Ϫ Ϯ Ϯ d D1–D2HEK SKF83959 Quinpirole 0.94 0.08* NA NA 246 12 NA 3 Ϫ Ϯ Ϯ Ϯ Ϯ e D1–D2HEK SKF83959 SCH23390 0.65 0.05 3.49 0.7* 712 38 16.2 1.3* 3 f D1–D2HEK SKF83822 — Ϫ0.59 Ϯ 0.09 0.39 Ϯ 0.03 3,927 Ϯ 980 11 Ϯ 3.2 3 Ϫ Ϯ Ϯ g D1–D2HEK SKF83822 PTX 0.93 0.08** NA NA 3,974 609 NA 3 ͓3H͔Raclopride binding to membranes from D2 or D1–D2 cells in the presence of increasing concentrations of SKF83959 or SKF83822 is shown. Data from three to eight independent experiments were analyzed and pooled (nH, Hill coefficient; KH, high-affinity dissociation constant; KL, low-affinity dissociation constant; RH, percentage of receptors in high-affinity state). In rows b, d, and f, where Hill coefficients were 1.0, binding data were analyzed to fit to a single site, and the Ͻ Ͻ Ki was calculated. *, P 0.05 for nH, KH, and RH values compared with those values in row a; **, P 0.05 compared with nH in row f. †—, not done. ‡NA, not applicable.

215 656 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0604049104 Rashid et al. absence of specific [3H]raclopride binding in striatal membranes from D2Ϫ/Ϫ mice (Fig. 4e). For both SKF83959 and SKF83959 plus quinpirole treatments, activation of Gq/11 was prevented by blockade of either D1 or D2 receptors with SCH23390 or raclopride, respectively (Fig. 4c). The D2 antagonist sulpiride also blocked activation of Gq/11. These results indicate that in the striatum, activation of both D1 and D2 receptors are necessary for rapid Gq/11 activation and that activation of the D1 receptor alone is not sufficient. The agonist specificity of the striatal D1–D2 receptor signaling complex was similar to that established in the D1–D2 stable cell line in that SKF83959 could activate Gq/11 signaling through the complex and did not have any effect on Gs/olf activation, whereas SKF83822 robustly activated Gs/olf (Fig. 4d). Notably, activation of Gq/11 by SKF83822 in striatum could not be observed, suggesting the small degree of Gq/11 activation by this compound in D1– D2HEK cells is not reflected in vivo. Competition binding experiments of [3H]raclopride and SKF83959 or SKF83822 on murine striatal membranes revealed high-affinity binding of both D1 receptor agonists to D2 receptors, and saturation binding isotherms from striata of 3- and 8-month-old male mice gave similar Bmax (Ϸ220 fmol/mg) and Kd values (Ϸ1 nM) for [3H]raclopride. The ability of a single dose of SKF83822 or SKF83959 to displace [3H]raclopride (1 nM) was then quantified (Fig. 4f), as described in ref. 9. SKF83822 (10 nM) displaced [3H]raclopride to similar degrees in 3- and 8-month-old mice as well as in 8-month-old D1Ϫ/Ϫ animals. In contrast, there was a 27% increase in displacement of [3H]raclopride binding by 10 nM SKF83959 in 8-month-old mice compared with 3-month-old mice, and displacement was almost completely eliminated in D1Ϫ/Ϫ Fig. 4. Coactivation of striatal D1 and D2 dopamine receptors activates mice. Therefore, high-affinity binding of SKF83959 to the D2 35 ␥ Gq/11. Agonist-dependent [ S]GTP S incorporation into G proteins is shown. receptor in brain depends on the presence of the D1 receptor, (a) A dose-dependent increase in activation of Gq/11 was observed after unlike binding of SKF83822, and is greater in older animals. membranes from wild-type (WT) mouse striatum (D1ϩ/ϩ) were treated for 1 min with agonists. SKF83959 stimulated relative increases in [35S]GTP␥S incor- To identify functional consequences of calcium signaling by the D1–D2 receptor complex, the effect of D1–D2 heterooligomer poration into Gq/11 of 13.3 Ϯ 2.7% for 10 ␮M, 30.8 Ϯ 8.5% for 50 ␮M, and 45.9 Ϯ 1.4% for 100 ␮M agonist (n ϭ 6 for all). Cotreatment with SKF83959 activation on CaMKII␣ was examined. CaMKII␣ plays a funda- (SKF) and quinpirole [(Quin) 10, 50, and 100 ␮M each] resulted in increases of mental role in synaptic plasticity, and both its translation and 26.3 Ϯ 4.9%, 44.7 Ϯ 5.1%, and 116.2 Ϯ 34.0% over baseline, respectively (n ϭ activity can be regulated by increases in intracellular calcium (10, ϭ 6). Quinpirole alone did not stimulate activation of Gq/11 (n 6). (b)No 11), typically subsequent to NMDA receptor activation. Immuno- activation of Gq/11 was observed in membranes from D1 or D2 mutant mice histochemical labeling for CaMKII␣ was performed after i.p. (D1Ϫ/Ϫ;D2Ϫ/Ϫ)(n ϭ 5, n ϭ 2). Basal level of [35S]GTP␥S incorporation is dopamine receptor agonist administration to animals (Fig. 5 a–h) indicated with the dashed line. (c) Comparison of Gq/11 activation by dopa- NEUROSCIENCE mine, SKF83959, and quinpirole or SKF83959 alone (100 ␮M agonist for each). and quantified (Fig. 5 i and j) as described. For both total and Activation was prevented by pretreatment of membranes with SCH23390 activated CaMKII␣, a large increase in both the intensity and (SCH), raclopride (RAC), or sulpiride (SLP) (n ϭ 6 for each). (d) SKF83959 number of immunolabeled neurons in the nucleus accumbens of activated Gq/11 but not Gs/olf, in contrast to SKF83822, which activated Gs/olf adult male rats was observed within 10 min of SKF83959 and 3 but not Gq/11 (n ϭ 4 for each). (e) Binding of 1 nM [ H]raclopride in striatal Ϯ Ϫ Ϫ quinpirole coadministration (Fig. 5a). There was no change in the membranes from WT mice (97.1 3.3 fmol/mg) was reduced by 23% in D1 / ␣ mice (74.9 Ϯ 3.1 fmol/mg) and was completely absent in D2Ϫ/Ϫ mice. (f) number of CaMKII -positive neurons in response to SKF83959 or SKF83822 displaced [3H]raclopride (1 nM) binding to similar degrees in quinpirole individually, although there was a moderate increase in 3-month and 8-month-old mice (49.5 Ϯ 1.9% and 46.5 Ϯ 2.8%) (n ϭ 7), and the intensity of labeling per cell for either drug (Fig. 5 b and c). The displacement in 8-month-old D1Ϫ/Ϫ mice (38.5 Ϯ 1.4%) was not statistically agonist-mediated increases in CaMKII␣ could be blocked by different from that in 8-month-old wild type (n ϭ 3). In contrast, there was a pretreating animals with either SCH23390 or raclopride (Fig. 5 e 3 27% increase in displacement of [ H]raclopride binding by SKF83959 in and f) indicating the necessity for both D1 and D2 receptors. 8-month-old mice (32.9 Ϯ 1.4%) in contrast to 3-month-old mice (26.0 Ϯ 1.3%) (n ϭ 6), and displacement was almost completely eliminated in D1Ϫ/Ϫ mice Furthermore, the agonist-mediated increase was also detected in Ϫ Ϫ Ϫ Ϫ (4.87 Ϯ 3.7%) (n ϭ 3). *, P Ͻ 0.05; **, P Ͻ 0.005; Student’s t test compared with wild-type mice but was absent in both D1 / and D2 / mice (SI normalized baseline values. Fig. 9). Significantly, there was no increase in CaMKII␣ in animals treated with SKF83822 or SKF83822 and quinpirole (Fig. 5g), indicating that the effect of dopamine agonist on CaMKII␣ was 35 [ S]GTP␥S incorporation into Gq/11 over baseline that were specific to activation of Gq/11-coupled receptor complexes. Pre- greater than with SKF83959 alone. Quinpirole alone did not treatment of animals with the NMDA receptor antagonist MK-801 stimulate activation of Gq/11 at any of the doses tested. Also, it was did not affect the SKF83959- and quinpirole-mediated increase in confirmed that it was the striatal D1 and D2 receptor subtypes that the number of CaMKII␣-positive neurons, although the increase in formed this signaling complex because we could not elicit Gq/11 the intensity of immunolabel per cell was slightly lower (Fig. 5h). activation with agonist treatments of membranes from mice that Overall, these results point to a role for D1–D2 receptor complexes lacked functional D1 (D1Ϫ/Ϫ) or D2 receptors (D2Ϫ/Ϫ) (Fig. 4b). in direct modulation of CaMKII␣ levels through activation of Gq/11 The involvement of the D2 subtype was further confirmed by the and release of intracellular calcium.

216 Rashid et al. PNAS ͉ January 9, 2007 ͉ vol. 104 ͉ no. 2 ͉ 657 Discussion In this work we have identified a heteromeric signaling complex in brain composed of D1 and D2 dopamine receptor subtypes which rapidly activates Gq/11 on agonist binding to both receptors within the complex. The receptor complex possesses a unique pharmacology such that a specific subset of D1 receptor agonists, SKF81297 and SKF83959, can activate the heteromer by acting concurrently on both the D1 receptor and a distinct conformation of the D2 receptor that depends on the presence of the D1 receptor. Because SKF83959 does not activate AC-coupled D1 or D2 recep- torsorGq/11 through D1 receptor homomeric units, we propose that this D1-like receptor agonist is in fact a specific agonist for Gq/11-coupled D1–D2 receptor heterooligomers. We also present evidence indicating that the D1–D2 receptor complex is more prevalent in murine striatum at 8 months of age and that it can increase levels of total and activated CaMKII␣ in the nucleus accumbens, providing a distinct mechanism of dopaminergic modulation of neuronal function in later adulthood. Coimmunoprecipitation of D1 and D2 dopamine receptors from rodent striata had provided direct evidence that these receptors could oligomerize in vivo (6). We now show that both the D1 and D2 receptors within the striatal Gq/11-coupled signaling unit possess a distinct pharmacology and rank order of the agonists that can activate the complex, consistent with the creation of unique ligand- binding pockets and G protein coupling resulting from receptor heterooligomerization. Specifically, although each of the D1 agonists tested has equivalent ability to bind with high affinity to the D1 receptor, only SKF81297 or SKF83959, and not SKF83822, could activate Gq/11 through D1 receptors in conjunction with D2 receptor activation by quinpirole. In the absence of quinpirole, SKF81297 or SKF83959 could activate the complex by acting as full agonists for the D1 receptor and partial agonists for the D2 receptor. This unique D2 receptor pharmacology was induced by the presence of D1 receptors and was independent of D1 receptor activation. Strikingly, the ability of the agonists SKF81297 and SKF83959 but not SKF83822 to activate D1–D2 heterooligomers correlates with their specificity in stimulating PI turnover in brain (2, 4). The apparent increase in Gq/11-coupled D1–D2 receptor complexes in striata from older animals was unexpected because studies have shown that after 60 days of age, the density of binding sites for D1 and D2 receptors in the nucleus accumbens and striatum of male rats either decreases or does not change significantly (12, 13). Because our data indicated no difference in the total density of D2 receptors in the striata of 3-month-old and 8-month-old mice, there appears to be a shift in the proportion of D2 receptors associated with D1 receptors with increasing age. Because most studies of dopamine receptor function in the brain use rodents that are 3–4 months of age, these results may explain the limited reports of putative functions of Gq/11-coupled dopamine receptors in brain. Furthermore, potential changes in the relative proportion of dopamine-activated Gs/olf, Gi/o, and Gq/11 signaling pathways in Fig. 5. Activation of Gq/11-coupled D1–D2 dopamine receptor complexes different brain regions could have important implications for our increases CaMKII␣ levels in the nucleus accumbens. (a) Injection of animals understanding of the age-related regulation of dopamine function ␣ with SKF83959 and quinpirole increased the number of CaMKII -immunola- in brain. beled cells and the intensity of labeling per cell compared with saline-injected controls (d). (b) SKF83959 caused a small increase in immunolabel intensity but Calcium signaling has profound effects on almost all aspects of no change in the number of CaMKII␣-positive neurons. (c) Quinpirole gave neuronal function, notably regulation of intercellular communica- results similar to those in b.(e and f) Pretreatment with SCH23390 or raclo- tion and neuronal plasticity (14, 15). The possibility that Gq/11- pride prevented the SKF83959- and quinpirole-mediated increase in CaMKII␣. coupled dopamine receptors can modulate synaptic plasticity by (g) Injections of SKF83822 and quinpirole gave no net changes in CaMKII␣.(h) activating CaMKII␣ has been suggested previously (1, 16). Our The NMDA receptor antagonist MK-801 slightly reduced the immunolabel results showing that G /11-coupled D1–D2 receptor complexes can ␣ q intensity of the SKF83959/quinpirole-mediated increase in CaMKII but did increase CaMKII␣ in the nucleus accumbens provide the molecular not affect the increase in the number of CaMKII␣-positive neurons. (i and j) basis for a direct link among dopamine action, calcium signaling, Quantiﬁcation of data from four independent experiments. Within each ␣ experiment, treatments were performed in duplicate, and two or three slices and CaMKII activation. Furthermore, these data may provide an were analyzed from each animal. *, P Ͻ 0.005; Student’s t test compared with explanation for reports showing that in the nucleus accumbens both saline controls. ac, anterior commissure. coactivation of D1 and D2 receptors as well as activation of

217 658 ͉ www.pnas.org͞cgi͞doi͞10.1073͞pnas.0604049104 Rashid et al. CaMKII␣ are necessary for the induction of behavioral sensitiza- after the addition of the radioisotope, and the reaction was allowed tion to psychostimulants such as cocaine (17, 18). to proceed for 1 or 3 min (for Gq/11 and Gs activation). Membranes In summary, in brain there is a Gq/11-coupled signaling unit were collected and solubilized, and 5 ␮gofanti-G␣ antibody was composed of D1 and D2 dopamine receptors that can be identified added (Gq/11, Gs,Gs/olf, Gi3; Santa Cruz Biotechnology, Santa by its unique pharmacology and that requires concurrent activation Cruz, CA). Specificity of the antibodies has been described (21). of both receptors for signaling. The formation of a distinct dopa- Immunoprecipitation occurred overnight at 4°C. Protein G–agar- minergic signaling unit by two receptors that signal through sepa- ose was added, and the reaction was left for an additional 90 min rate pathways when homooligomeric is significant in that it provides at 4°C. The agarose was washed five times with solubilization buffer, a greater repertoire of signaling pathways by which dopamine can and incorporation of [35S]GTP␥S was measured by liquid scintil- modulate neuronal function than would be possible by each of the lation spectrometry. five different dopamine receptor subtypes acting solely as separate units. Characterization of changes in this signaling unit with age and Radioligand-Binding Assays. Binding experiments were performed the functional consequences of signaling through the complex will on 20-␮g membrane extracts with 1–2 nM [3H]raclopride or 1 nM increase our understanding of how D1–D2 heteromers contribute [3H]SCH23390 in the presence of agonist as described (8). Data to neuronal function as well as the role that this pathway may play points were analyzed by nonlinear least-squares regression (Prism in the etiology or pathophysiology of disorders in which altered dopamine signaling is implicated, such as schizophrenia. Notably, a 3.0 software; GraphPad, San Diego, CA). diminished link between D1 and D2 dopamine receptors has been noted in the brains of schizophrenic patients (19), and it has been Immunohistochemistry. Adult male Sprague–Dawley rats were in- proposed that disruption in calcium homeostasis is the central jected i.p. with saline or agonists SKF83959 (3 mg/kg) or SKF83822 factor underlying the molecular pathology of schizophrenia (20). (3 mg/kg), either alone or with quinpirole (2 mg/kg). Stock solutions Our data provide a mechanism by which to converge these lines of of drugs were diluted so that the DMSO concentration did not evidence and significant impetus to determine whether the D1–D2 exceed 5%, and the total volume of injection was 500 ␮l. For receptor signaling complex is altered in neuropsychiatric disease. antagonist experiments, SCH23390 (1 mg/kg), raclopride (2 mg/kg), MK-801 (1 mg/kg), or saline was injected 10 min before agonist Materials and Methods administration. Animals were anesthetized 10 min after injection Measurement of Intracellular Calcium Levels. Calcium mobilization and perfused intracardially with 4% paraformaldehyde, and 16-␮m assays on stable HEK 293 cell lines expressing human D1 and/or D2 cryostat sections were prepared as described (7) for immunostain- receptors (Ϸ1 pmol/mg protein Ϯ 0.2 pmol) were performed as ing by using the Elite ABC kit (Vector Laboratories, Burlingame, described previously (6) with modifications. Cells were seeded at CA) as indicated by the manufacturer. Primary antibodies for total 1.2 ϫ 105 cells per well in 96-well plates and maintained in advanced and activated (i.e., Thr286-phosphorylated) CaMKII␣ (rabbit anti- minimum essential medium (Invitrogen, Carlsbad, CA). All exper- CaMKII␣ and rabbit anti-Thr286–CaMKII␣; Santa Cruz Biotech- iments were performed in the presence of EGTA. For responses to nology) were used at 1:200. Images were obtained by using an SKF83822, a portion of the signal was identified as nondopamin- Axioplan2 microscope (Carl Zeiss, Thornwood, NY) and quanti- ergic because it could not be eliminated by incubation of cells with fied by using ImageJ software (National Institutes of Health, SCH23390 or raclopride. Dopamine receptor-specific responses Bethesda, MD). were obtained by subtracting the antagonist-resistant signal. For antagonist and inhibition studies, cells were incubated at 37°C in Statistical Analysis. All values are reported as mean Ϯ SEM. saline plus the appropriate compound for 20 min before the assay. Comparisons of means were performed by using Student’s t test For PTX treatments, cells were incubated before dye loading for (two-tailed, unpaired). 18–24 h in PTX (0.25 ␮g/ml) diluted in growth medium. We thank Yamanouchi Pharmaceuticals for generously providing GTP␥S Assay. Agonist-mediated [35S]GTP␥S incorporation into

YM254890, Dr. Derek van der Kooy and Dr. Jose Nobrega for the D2 NEUROSCIENCE specific G proteins was assessed as described (21). GDP (final ␮ ␮ mutant mice, and Jennifer Ng for technical assistance. This work was concentration, 1 M) was added to 100- g membranes from HEK supported by National Institute on Drug Abuse Grant DA007223 (to cells or striata of wild-type, D1 mutant mice (22), or D2 mutant S.R.G. and B.F.O.) and Canadian Institutes of Health Research Grant mice (23), and the assay mixture was incubated on ice for 10 min. MOP12180 (to S.R.G. and B.F.O.), an Ontario Mental Health Foun- The mixture was moved to 30°C and then incubated for 5 min before dation Postdoctoral Fellowship (to A.J.R.), and an Ontario Graduate adding [35S]GTP␥S (1,250 Ci/mmol) (PerkinElmer, Wellesley, Scholarship (to C.H.S.). S.R.G. holds a Canada Research Chair in MA) to a final concentration of 2.5 nM. Agonist was added 30 sec Molecular Neuroscience.

1. Neve KA, Seamans JK, Trantham-Davidson H (2004) J Rec Sig Trans Res 13. Teicher MH, Andersen SL, Hostetter JC, Jr (1995) Dev Brain Res 89:167–172. 24:165–205. 14. Berridge MJ (1998) Neuron 21:13–26. 2. Jin LQ, Goswami S, Cai G, Zhen X, Friedman E (2003) J Neurochem 85:378–386. 15. Verkhratsky A (2005) Physiol Rev 85:201–279. 3. Tang TS, Bezprozvanny I (2004) J Biol Chem 279:42082–42094. 16. Zhen X, Goswami S, Abdali SA, Gil M, Bakshi K, Friedman E (2004) Mol 4. Undie AS, Weinstock J, Sarau HM, Friedman E (1994) J Neurochem 62:2045– Pharmacol 66:1500–1507. 2048. 17. Capper-Loup C, Canales JJ, Kadaba N, Graybiel AM (2002) J Neurosci 5. Wang HY, Undie AS, Friedman E (1995) Mol Pharmacol 48:988–994. 22:6218–6227. 6. Lee SP, So CH, Rashid AJ, Varghese G, Cheng R, Lanca AJ, O’Dowd BF, 18. Pierce RC, Quick EA, Reeder DC, Morgan ZR, Kalivas PW (1998) J Phar- George SR (2004) J Biol Chem 279:35671–35678. macol Exp Ther 286:1171–1176. 7. So CH, Varghese G, Curley KJ, Kong MM, Alijaniaram M, Ji X, Nguyen T, 19. Seeman P, Niznik HB, Guan HC, Booth G, Ulpian C (1989) Proc Natl Acad O’Dowd BF, George SR (2005) Mol Pharmacol 68:568–578. Sci USA 86:10156–10160. 8. Takasaki J, Saito T, Taniguchi M, Kawasaki T, Moritani Y, Hayashi K, Kobori M (2004) J Biol Chem 279:47438–47445. 20. Lidow MS (2003) Brain Res Rev 43:70–84. 9. Torvinen M, Marcellino D, Canals M, Agnati LF, Lluis C, Franco R, Fuxe K 21. Dowling MR, Nahorski SA, Challis RAJ (2004) in Receptor Signal Transduction (2005) Mol Pharmacol 67:400–407. Protocols, eds Willars GB, Challis RAJ (Humana, Totowa, NJ), pp 197–206. 10. Lisman J, Schulman H, Cline H (2002) Nat Rev Neurosci 3:175–190. 22. Drago J, Gerfen CR, Lachowicz JE, Steiner H, Hollon TR, Love PE, Ooi GT, 11. Scheetz AJ, Nairn AC, Constantine-Paton M (2000) Nat Neurosci 3:211–216. Grinberg A, Lee EJ, Huang SP, et al. (1994) Proc Natl Acad Sci USA 12. Suzuki M, Hatano K, Sakiyama Y, Kawasumi Y, Kato T, Ito K (2001) Synapse 91:2564–2568. 41:285–293. 23. Kippin TE, Kapur S, van der Kooy D (2005) J Neurosci 25:5815–5823.

218 Rashid et al. PNAS ͉ January 9, 2007 ͉ vol. 104 ͉ no. 2 ͉ 659 7 GENERAL CONCLUSIONS AND FUTURE DIRECTIONS

7.1 General Conclusions

Prior to the beginning of these studies, there was considerable debate in regards to our

understanding of various facets of GPCR regulation, in particular, a) how GPCR oligomers were

assembled in the cell and b) how such GPCR complexes were organized at the plasma

membrane. A concerted effort using a diverse repertoire of biochemical, pharmacological, and

biophysical tools have helped to clarify the nature of these biological processes. Indeed, energy

transfer techniques as well as dominant negative studies from specific intracellularly sequestered

GPCR mutants have provided evidence suggesting that most, if not all, GPCR oligomers are

assembled during biosynthesis prior to receptor trafficking. This is in contrast to the agonist-

dependent model of GPCR oligomerization, in which evidence supporting this is mostly derived

from energy transfer increases that are most likely explained by proximity shifts in response to

agonist-induced conformational re-arrangments. Following maturation, these GPCR multimers

are localized to the plasma membrane where they likely form stable pre-assembled complexes

with the corresponding G protein and effector. Although this notion is contrary to the traditional

collision-coupling model of signalling complex formation, it provides a more plausible framework with which to describe how GPCR signalling pathways are so highly organized given

the diversity of G protein, receptor, and effector subtypes within a single cell (Rebois and

Hebert, 2003). Furthermore, this model appears consistent with prior evidence that GPCR

signalling molecules are not “free-floating” but are likely constrained within the lipid bilayer

with limited lateral mobility and diffusion potential (Neubig, 1994). Indeed, this stability and

requirement for consistent signalling fidelity is best described by their confinement within

219 caveolae and lipid raft microdomains. Nevertheless, the structural determinants and regulatory

mechanisms that define how these different signalling molecules function within these

microenvironments remain unclear.

Further studies are now required to determine the functional significance of GPCR homo- oligomerization as well as the factors that govern how these oligomers are partitioned into lipid- enriched microdomains. Is there a relationship between these two processes or do they represent independent events within the complex life cycle of a G protein-coupled receptor ?

The findings presented in this thesis support and shed further light on these two areas of

GPCR regulation using the D1 dopamine receptor as a model example. The overall findings for each chapter and suggestions for future studies are presented.

7.1.1 Conformational Checkpoints Regulate D1 Dopamine Receptor Homo-

oligomerization

The findings in Chapter Three provide evidence for a conformational checkpoint in the cell that selectively permits certain oligomeric configurations to be trafficked to the cell surface.

It is difficult to define the precise nature of these configurations but we have presented some plausible hypotheses addressing this. The quality control mechanisms governing assembly of

GPCR dimers and oligomers requires further understanding. A novel finding of the present study is that the relative conformation of protomers within an oligomer is recognized by a quality control mechanism that monitors the structural integrity of oligomers prior to cell surface trafficking. Two conformationally distinct receptors, which were found to traffic normally as homo-oligomers when expressed individually, were retained intracellularly when interacting with each other. However, because cell surface restoration of the oligomer could be achieved by

220 pharmacological intervention using only cell permeable agonists, we propose that a specific cellular checkpoint monitors for certain conformational prerequisites of a homo-oligomer prior to its trafficking through the distal secretory pathway. One of these prerequisites may be the requirement for a homogenous conformational arrangement of an oligomeric complex where all the constituent protomers must either be in an activated or unactivated state.

Future work will be required to further validate this hypothesis as the data currently presented is unable to unequivocally support this conclusion. One potential route of study to adequately show that a homogenously activated or unactivated oligomeric configuration is required for plasma membrane trafficking is by assessing the effect of a spectrum of constitutively activated D1 dopamine receptor mutants on wild-type receptor expression. Indeed, under this model, a set of receptor mutants with a graded range of constitutive activity (and hence, activated conformations) would be predicted to have a graded dominant negative effect on the cell surface expression of co-expressed wild-type receptors. Such a model would predict that the most constitutively active receptor mutant would exert the greatest inhibitory effect on wild- type receptor expression. There have been a number of artificially generated constitutively active

D1 dopamine receptors reported with different levels of constitutive activity. Most of these receptor mutants were designed to incorporate or swap either specific residues or fragments of the D5 receptor into the corresponding location in the D1 receptor, thus rendering D5-like constitutively active properties (Chaar et al., 2001; Iwasiow et al., 1999; Tumova et al., 2003;

Tumova et al., 2004). Alternatively, a graded range of conformational states may be attained by mutating a specific amino acid residue critical for activation/de-activation to structurally distinct amino acids. This strategy was elegantly demonstrated with the α1b-adrenergic receptor in which sequential mutation of residue A293 (in IC3) to every amino acid possibility rendered uniquely

221 different basal levels of inositol phosphate production for every amino acid substitution

(Kjelsberg et al., 1992). This residue was predicted to be involved in Gq-coupling suggesting that

each amino acid substitution likely had a unique propensity to shift the receptor away from the

inactive, G protein-coupled state of the wild-type receptor. Although the D1 dopamine receptor

does not have an analogous residue in IC3, a similar mutagenesis strategy in residues required for Gs-coupling may yield similar results.

The role of molecular chaperones in the assembly and trafficking of oligomeric proteins

has been well documented for a number of integral membrane proteins including various

receptors and channels. Given the recent finding that calnexin can interact with the D1 dopamine

receptor and is required for its cell surface trafficking (Free et al., 2007), it would be of future

interest to determine whether this interaction is required for proper maturation of D1DR oligomers. Indeed, if glycosylation is a prerequisite for D1DR trafficking, the metabolic labeling

of newly synthesized WT/D103A oligomers could be studied to determine whether glycosylation

is compromised in this complex as a result of aberrant conformational folding. This may prevent

the interaction (by co-precipitation) of molecular chaperones that include but are not limited to

calnexin.

Alternatively, equally possible is the notion that this intracellularly sequestered complex

yields a conformation that irregularly exposes ER retention motifs (RXR, for multispanning

integral membrane proteins (Ma and Jan, 2002)) that would normally be masked in the wild-type

receptor oligomer. The binding of cell permeable agonists to ligand or allosteric binding sites in

the oligomeric complex might induce a conformational change that subsequently masks this

motif allowing receptor maturation. Hence, it would be of interest to examine the effect of

mutating the RXR motif in IC1 in each of the D103A and wild-type receptors. If involved in

222 oligomeric sequestration, the ablation of this sequence would facilitate proper trafficking of the

WT/D103A oligomer to the plasma membrane. In a similar strategy, candidate ER export motifs

such as the FXXXFXXXF sequence at the TM7-carboxyl tail junction of D1DR could be

explored. Although masking of this motif is temporally regulated by binding to DRiP78 (Bermak

et al., 2001), an abnormal WT/D103A conformation could conceivably mask this motif thereby

permanently sequestering this oligomeric complex. Thus, it would be of interest to determine

whether DRiP78 can maintain access to this critical region of either WT or D103A receptors

within an oligomeric complex.

7.1.2 D1 Dopamine Receptors are Localized in and Functionally Regulated by Caveolae

The findings in Chapter Four demonstrate the importance of lipid and cholesterol- enriched microdomains in regulating turnover and signalling of the D1 receptor. Although this

receptor traditionally undergoes agonist-dependent clathrin mediated internalization, we

demonstrate that even with this option available, D1 receptors can be sequestered through a

caveolar pathway. This route of endocytosis is kinetically slower than the arrestin-mediated

pathway and appears to occur independent of classical GPCR phosphorylation events mediated

by PKA or GRK2. Receptor internalization appears to require an intact putative caveolin binding

motif in transmembrane domain 7. The disruption of this motif at specific aromatic amino acid

residues severely compromises the ability of the D1 receptor to internalize without affecting cell

surface expression or its pharmacological profile. The inability of caveolin-1 to interact with

these mutant receptors strongly implicates a role for this scaffolding protein in sequestering D1

receptors. In addition to the role of caveolae in D1DR endocytosis, receptor localization in

caveolae appears to have an inhibitory effect on Gs activation and cAMP production. This is

223 contrary to the traditional idea that lipid rafts act to enhance signalling, as is the case for many

Gq-coupled receptors.

Although we have provided evidence for caveolar-mediated D1DR internalization, the

role for caveolae in receptor recycling is less clear. Our single cell time course studies using confocal microscopy suggest that D1 receptors are not degraded but recycled to the plasma membrane within 60 min. Indeed, studies from other cell lines have shown that recycling

endosomes can be enriched in sphingolipids and cholesterol as well as specific lipid raft markers

such as caveolin-1 and flotillin (Gagescu et al., 2000). It would be of future interest to dissect

the molecular mechanisms behind this as we are not currently aware of any reports

demonstrating the role of caveolae in receptor recycling.

Based on the observation that D1DR localizes to both caveolin and non-caveolin

enriched sucrose gradient fractions, it would also be of future interest to identify the molecular

determinants that dictate the basal localization of D1DRs at the plasma membrane. Why is there

a caveolin-associated and caveolin-independent population of receptors ? Do these two

populations of receptors signal differently or are they regulated differently ? One recent finding that sheds light on this issue and has broad implications for most of the work presented in this thesis is that the active metabolite, Act-Met, of the antithrombotic drug, clopidogrel, has selective binding preference for P2Y12 receptor oligomers specifically localized in caveolin enriched domains (Savi et al., 2006). Upon agonist stimulation, these receptor oligomers were found to dissociate into monomers and dimers which subsequently partitioned into non-caveolin enriched domains and rendered non-functional. Hence, the raft associated oligomeric species represented the functional form of the receptor which interacted with a free thiol group on Act-

Met, providing a mechanism that is believed to account for the antithrombotic properties of

224 clopidogrel. Although the relationship between caveolae localization and D1 receptor

oligomerization was not specifically studied in this thesis, it would be of future interest to determine whether “monomerization” by receptor mutagenesis alters the functional regulation of

these receptors by caveolae. For instance, targeting of the analogous phenylalanine residue in

TM4 of the D2 dopamine receptor, believed to form part of the oligomeric interface (Guo et al.,

2003; Lee et al., 2003b), may have an equivalent effect in disrupting D1 dopamine receptor

oligomers. These mutants could be further studied to determine whether caveolin-1 interaction or

caveolae association is preserved.

Finally, the physiological relevance of caveolae in D1 dopamine receptor function would

be of particular interest in the brain, especially given the fact that most neurons and

neuroblastoma cells don’t appreciably express caveolin or have morphologically identifiable

caveolae (Allen et al., 2007). Therefore, the co-immunoprecipitation of D1DR with caveolin-1

from rat brain strongly suggests a role for this interaction in glial cells, which have been shown

to express D1 dopamine receptors (Miyazaki et al., 2004; Zanassi et al., 1999) and functionally

respond to atypical antipsychotics (Reuss and Unsicker, 2001). Although the precise function of

D1 receptors in these cells is not known, the use of striatal astrocytes, for instance, might be a

useful tool in forging a novel link between the activity of neuroleptic drugs at D1DRs and

caveolar localization.

7.1.3 Palmitoylation of D1 Dopamine Receptors Regulates Caveolar Internalization

The findings in Chapter Five suggest a role for palmitoylation as a regulatory

modification that assists in controlling the kinetics of and facilitating the caveolar internalization

of D1 dopamine receptors. Although palmitate incorporation does not appear to be required for

225 basal localization of the D1 receptor in caveolin-enriched microdomains, its absence significantly accelerates the rate of caveolar internalization. This is not completely explained by a switch in the route of endocytosis from caveolae to clathrin coated pits since internalization is still sensitive to caveolae disruption and resistant to clathrin inhibition, both albeit, to a lesser degree than wild-type D1 receptors. Although the mechanism describing this effect is unknown, palmitoylation likely has a role in maintaining a “controlled” rate of agonist-induced internalization, possibly by preventing either the hyperphosphorylation or more rapid phosphorylation of the receptor.

The ablation of palmitoylatable cysteines in D1DR would likely lead to the structural instability of the carboxyl tail which would no longer be anchored to the lipid bilayer. This might reduce some steric hindrance making this particularly lengthy peptide sequence more amenable to modifications by enzymes involved in receptor internalization such as GRK or PKA. Thus, it would be of future interest to determine whether the accelerated internalization kinetics of

C347A/C351A is due to increased phosphorylation by these kinases. One approach to facilitate this study would be to determine which serines and threonines in the carboxyl tail are better substrates for phosphorylation in a palmitoylation-less background. The incorporation of a proteolytic cleavage site just upstream of residue C347 would allow specific isolation of the carboxyl tail fragment which could be immunodetected for changes in serine or threonine phosphorylation independent of other regions of the receptor such as the cytosolic loops. One cleavage site that has been successfully used to study GPCR structure (Zeng et al., 1999) is the

IEDR sequence, a motif specifically cleaved by the protease, Factor Xa. Alanine scanning mutagenesis of carboxyl tail serines and threonines could subsequently be used to pinpoint specific residues that are subject to enhanced phosphorylation.

226 7.2 Final Thoughts

Much progress has been made in elucidating the neurophysiology and molecular mechanisms, including desensitization and signalling, of D1 dopamine receptors. The novel finding that these receptors form higher order protein complexes with themselves, other GPCRs, and other scaffolding proteins adds another layer of complexity to understanding how these receptors function at the molecular level. An integrative approach to studying these different facets of receptor function is needed to fully appreciate the role of these receptors in neurodegenerative, neuropsychiatric, and renal disorders.

227 8 REFERENCES

AbdAlla S, Zaki E, Lother H and Quitterer U (1999) Involvement of the amino terminus of the B(2) receptor in agonist-induced receptor dimerization. J Biol Chem 274(37):26079- 26084.

Allen JA, Halverson-Tamboli RA and Rasenick MM (2007) Lipid raft microdomains and neurotransmitter signalling. Nat Rev Neurosci 8(2):128-140.

Allen JA, Yu JZ, Donati RJ and Rasenick MM (2005) Beta-adrenergic receptor stimulation promotes G alpha s internalization through lipid rafts: a study in living cells. Mol Pharmacol 67(5):1493-1504.

Amenta F, Gallo P, Rossodivita A and Ricci A (1993) Radioligand binding and autoradiographic analysis of dopamine receptors in the human heart. Naunyn Schmiedebergs Arch Pharmacol 347(2):147-154.

Anderson RG (1998) The caveolae membrane system. Annu Rev Biochem 67:199-225.

Angers S, Salahpour A, Joly E, Hilairet S, Chelsky D, Dennis M and Bouvier M (2000) Detection of beta 2-adrenergic receptor dimerization in living cells using bioluminescence resonance energy transfer (BRET). Proc Natl Acad Sci U S A 97(7):3684-3689.

Aoki T, Nomura R and Fujimoto T (1999) Tyrosine phosphorylation of caveolin-1 in the endothelium. Exp Cell Res 253(2):629-636.

Aridor M and Balch WE (1996) Membrane fusion: timing is everything. Nature 383(6597):220- 221.

Armstrong D and Strange PG (2001) Dopamine D2 receptor dimer formation: evidence from ligand binding. J Biol Chem 276(25):22621-22629.

Arnal JF, Dinh-Xuan AT, Pueyo M, Darblade B and Rami J (1999) Endothelium-derived nitric oxide and vascular physiology and pathology. Cell Mol Life Sci 55(8-9):1078-1087.

Arni S, Keilbaugh SA, Ostermeyer AG and Brown DA (1998) Association of GAP-43 with detergent-resistant membranes requires two palmitoylated cysteine residues. J Biol Chem 273(43):28478-28485.

Audigier Y, Friedlander M and Blobel G (1987) Multiple topogenic sequences in bovine opsin. Proc Natl Acad Sci U S A 84(16):5783-5787.

228 Ayoub MA, Couturier C, Lucas-Meunier E, Angers S, Fossier P, Bouvier M and Jockers R (2002) Monitoring of ligand-independent dimerization and ligand-induced conformational changes of melatonin receptors in living cells by bioluminescence resonance energy transfer. J Biol Chem 277(24):21522-21528.

Bai M, Trivedi S, Kifor O, Quinn SJ and Brown EM (1999) Intermolecular interactions between dimeric calcium-sensing receptor monomers are important for its normal function. Proc Natl Acad Sci U S A 96(6):2834-2839.

Baker DA, Fuchs RA, Specio SE, Khroyan TV and Neisewander JL (1998) Effects of intraaccumbens administration of SCH-23390 on cocaine-induced locomotion and conditioned place preference. Synapse 30(2):181-193.

Bakker RA, Dees G, Carrillo JJ, Booth RG, Lopez-Gimenez JF, Milligan G, Strange PG and Leurs R (2004) Domain swapping in the human histamine H1 receptor. J Pharmacol Exp Ther 311(1):131-138.

Balasubramanian S, Teissere JA, Raju DV and Hall RA (2004) Hetero-oligomerization between GABAA and GABAB receptors regulates GABAB receptor trafficking. J Biol Chem 279(18):18840-18850.

Baldwin JM (1994) Structure and function of receptors coupled to G proteins. Curr Opin Cell Biol 6(2):180-190.

Baneres JL and Parello J (2003) Structure-based analysis of GPCR function: evidence for a novel pentameric assembly between the dimeric leukotriene B4 receptor BLT1 and the G- protein. J Mol Biol 329(4):815-829.

Barbier P, Colelli A, Maggio R, Bravi D and Corsini GU (1997) Pergolide binds tightly to dopamine D2 short receptors and induces receptor sequestration. J Neural Transm 104(8- 9):867-874.

Bari M, Battista N, Fezza F, Finazzi-Agro A and Maccarrone M (2005) Lipid rafts control signaling of type-1 cannabinoid receptors in neuronal cells. Implications for anandamide- induced apoptosis. J Biol Chem 280(13):12212-12220.

Bass J, Chiu G, Argon Y and Steiner DF (1998) Folding of insulin receptor monomers is facilitated by the molecular chaperones calnexin and calreticulin and impaired by rapid dimerization. J Cell Biol 141(3):637-646.

Befort K, Zilliox C, Filliol D, Yue S and Kieffer BL (1999) Constitutive activation of the delta opioid receptor by mutations in transmembrane domains III and VII. J Biol Chem 274(26):18574-18581.

229 Bendayan M and Rasio EA (1996) Transport of insulin and albumin by the microvascular endothelium of the rete mirabile. J Cell Sci 109 (Pt 7):1857-1864.

Beninger RJ, Hoffman DC and Mazurski EJ (1989) Receptor subtype-specific dopaminergic agents and conditioned behavior. Neurosci Biobehav Rev 13(2-3):113-122.

Benkirane M, Jin DY, Chun RF, Koup RA and Jeang KT (1997) Mechanism of transdominant inhibition of CCR5-mediated HIV-1 infection by ccr5delta32. J Biol Chem 272(49):30603-30606.

Bergson C, Mrzljak L, Smiley JF, Pappy M, Levenson R and Goldman-Rakic PS (1995) Regional, cellular, and subcellular variations in the distribution of D1 and D5 dopamine receptors in primate brain. J Neurosci 15(12):7821-7836.

Bermak JC, Li M, Bullock C and Zhou QY (2001) Regulation of transport of the dopamine D1 receptor by a new membrane-associated ER protein. Nat Cell Biol 3(5):492-498.

Bernier V, Bichet DG and Bouvier M (2004) Pharmacological chaperone action on G-protein- coupled receptors. Curr Opin Pharmacol 4(5):528-533.

Bhatnagar A, Sheffler DJ, Kroeze WK, Compton-Toth B and Roth BL (2004) Caveolin-1 interacts with 5-HT2A serotonin receptors and profoundly modulates the signaling of selected Galphaq-coupled protein receptors. J Biol Chem 279(33):34614-34623.

Bian D, Mahanivong C, Yu J, Frisch SM, Pan ZK, Ye RD and Huang S (2006) The G12/13- RhoA signaling pathway contributes to efficient lysophosphatidic acid-stimulated cell migration. Oncogene 25(15):2234-2244.

Blendy JA (2006) The role of CREB in depression and antidepressant treatment. Biol Psychiatry 59(12):1144-1150.

Blobel G (1980) Intracellular protein topogenesis. Proc Natl Acad Sci U S A 77(3):1496-1500.

Bouschet T, Martin S and Henley JM (2005) Receptor-activity-modifying proteins are required for forward trafficking of the calcium-sensing receptor to the plasma membrane. J Cell Sci 118(Pt 20):4709-4720.

Bouvier M (2001) Oligomerization of G-protein-coupled transmitter receptors. Nat Rev Neurosci 2(4):274-286.

Boyd GW, Low P, Dunlop JI, Robertson LA, Vardy A, Lambert JJ, Peters JA and Connolly CN (2002) Assembly and cell surface expression of homomeric and heteromeric 5-HT3

230 receptors: the role of oligomerization and chaperone proteins. Mol Cell Neurosci 21(1):38-50.

Bradbury FA, Kawate N, Foster CM and Menon KM (1997) Post-translational processing in the Golgi plays a critical role in the trafficking of the luteinizing hormone/human chorionic gonadotropin receptor to the cell surface. J Biol Chem 272(9):5921-5926.

Braun JE and Madison DV (2000) A novel SNAP25-caveolin complex correlates with the onset of persistent synaptic potentiation. J Neurosci 20(16):5997-6006.

Bressan RA and Crippa JA (2005) The role of dopamine in reward and pleasure behaviour-- review of data from preclinical research. Acta Psychiatr Scand Suppl(427):14-21.

Brown DA and London E (1998) Functions of lipid rafts in biological membranes. Annu Rev Cell Dev Biol 14:111-136.

Bruns RR and Palade GE (1968) Studies on blood capillaries. II. Transport of ferritin molecules across the wall of muscle capillaries. J Cell Biol 37(2):277-299.

Bu J, Bruckner SR, Sengoku T, Geddes JW and Estus S (2003) Glutamate regulates caveolin expression in rat hippocampal neurons. J Neurosci Res 72(2):185-190.

Bulenger S, Marullo S and Bouvier M (2005) Emerging role of homo- and heterodimerization in G-protein-coupled receptor biosynthesis and maturation. Trends Pharmacol Sci 26(3):131-137.

Bunzow JR, Van Tol HH, Grandy DK, Albert P, Salon J, Christie M, Machida CA, Neve KA and Civelli O (1988) Cloning and expression of a rat D2 dopamine receptor cDNA. Nature 336(6201):783-787.

Calver AR, Robbins MJ, Cosio C, Rice SQ, Babbs AJ, Hirst WD, Boyfield I, Wood MD, Russell RB, Price GW, Couve A, Moss SJ and Pangalos MN (2001) The C-terminal domains of the GABA(b) receptor subunits mediate intracellular trafficking but are not required for receptor signaling. J Neurosci 21(4):1203-1210.

Cameron PL, Ruffin JW, Bollag R, Rasmussen H and Cameron RS (1997) Identification of caveolin and caveolin-related proteins in the brain. J Neurosci 17(24):9520-9535.

Canals M, Marcellino D, Fanelli F, Ciruela F, De Benedetti P, Goldberg SR, Fuxe K, Agnati LF, Woods AS, Ferre S, Lluis C, Bouvier M and Franco R (2003) Adenosine A2A-dopamine D2 receptor-receptor heteromerization. Qualitative and quantitative assessment by fluorescence and bioluminescence energy transfer. J Biol Chem.

231 Cao TT, Deacon HW, Reczek D, Bretscher A and von Zastrow M (1999) A kinase-regulated PDZ-domain interaction controls endocytic sorting of the beta2-adrenergic receptor. Nature 401(6750):286-290.

Cao TT, Mays RW and von Zastrow M (1998) Regulated endocytosis of G-protein-coupled receptors by a biochemically and functionally distinct subpopulation of clathrin-coated pits. J Biol Chem 273(38):24592-24602.

Carman CV, Lisanti MP and Benovic JL (1999) Regulation of G protein-coupled receptor kinases by caveolin. J Biol Chem 274(13):8858-8864.

Carrillo JJ, Lopez-Gimenez JF and Milligan G (2004) Multiple interactions between transmembrane helices generate the oligomeric alpha1b-adrenoceptor. Mol Pharmacol 66(5):1123-1137.

Carrillo JJ, Pediani J and Milligan G (2003) Dimers of class A G protein-coupled receptors function via agonist-mediated trans-activation. J Biol Chem.

Casey JR and Reithmeier RA (1993) Detergent interaction with band 3, a model polytopic membrane protein. Biochemistry 32(4):1172-1179.

Chaar ZY, Jackson A and Tiberi M (2001) The cytoplasmic tail of the D1A receptor subtype: identification of specific domains controlling dopamine cellular responsiveness. J Neurochem 79(5):1047-1058.

Chabre M, Cone R and Saibil H (2003) Biophysics: is rhodopsin dimeric in native retinal rods? Nature 426(6962):30-31; discussion 31.

Chaipatikul V, Erickson-Herbrandson LJ, Loh HH and Law PY (2003) Rescuing the traffic- deficient mutants of rat mu-opioid receptors with hydrophobic ligands. Mol Pharmacol 64(1):32-41.

Chang W, Gelman MS and Prives JM (1997) Calnexin-dependent enhancement of nicotinic acetylcholine receptor assembly and surface expression. J Biol Chem 272(46):28925- 28932.

Chang WJ, Ying YS, Rothberg KG, Hooper NM, Turner AJ, Gambliel HA, De Gunzburg J, Mumby SM, Gilman AG and Anderson RG (1994) Purification and characterization of smooth muscle cell caveolae. J Cell Biol 126(1):127-138.

Chapple JP and Cheetham ME (2003) The chaperone environment at the cytoplasmic face of the endoplasmic reticulum can modulate rhodopsin processing and inclusion formation. J Biol Chem 278(21):19087-19094.

232 Charpentier S, Jarvie KR, Severynse DM, Caron MG and Tiberi M (1996) Silencing of the constitutive activity of the dopamine D1B receptor. Reciprocal mutations between D1 receptor subtypes delineate residues underlying activation properties. J Biol Chem 271(45):28071-28076.

Cheng ZJ and Miller LJ (2001) Agonist-dependent dissociation of oligomeric complexes of G protein-coupled cholecystokinin receptors demonstrated in living cells using bioluminescence resonance energy transfer. J Biol Chem 276(51):48040-48047.

Chini B and Parenti M (2004) G-protein coupled receptors in lipid rafts and caveolae: how, when and why do they go there? J Mol Endocrinol 32(2):325-338.

Ciruela F, Casado V, Mallol J, Canela EI, Lluis C and Franco R (1995) Immunological identification of A1 adenosine receptors in brain cortex. J Neurosci Res 42(6):818-828.

Ciruela F, Escriche M, Burgueno J, Angulo E, Casado V, Soloviev MM, Canela EI, Mallol J, Chan WY, Lluis C, McIlhinney RA and Franco R (2001) Metabotropic glutamate 1alpha and adenosine A1 receptors assemble into functionally interacting complexes. J Biol Chem 276(21):18345-18351.

Coge F, Guenin SP, Renouard-Try A, Rique H, Ouvry C, Fabry N, Beauverger P, Nicolas JP, Galizzi JP, Boutin JA and Canet E (1999) Truncated isoforms inhibit [3H]prazosin binding and cellular trafficking of native human alpha1A-adrenoceptors. Biochem J 343 Pt 1:231-239.

Cornea A, Janovick JA, Maya-Nunez G and Conn PM (2001) Gonadotropin-releasing hormone receptor microaggregation. Rate monitored by fluorescence resonance energy transfer. J Biol Chem 276(3):2153-2158.

Cosson P, Lankford SP, Bonifacino JS and Klausner RD (1991) Membrane protein association by potential intramembrane charge pairs. Nature 351(6325):414-416.

Couet J, Li S, Okamoto T, Ikezu T and Lisanti MP (1997) Identification of peptide and protein ligands for the caveolin-scaffolding domain. Implications for the interaction of caveolin with caveolae-associated proteins. J Biol Chem 272(10):6525-6533.

Couve A, Filippov AK, Connolly CN, Bettler B, Brown DA and Moss SJ (1998) Intracellular retention of recombinant GABAB receptors. J Biol Chem 273(41):26361-26367.

Crawford CA, Zavala AR, Karper PE and McDougall SA (2000) Long-term effects of postnatal amphetamine treatment on striatal protein kinase A activity, dopamine D(1)-like and D(2)-like binding sites, and dopamine content. Neurotoxicol Teratol 22(6):799-804.

233 Damjanovich S, Vereb G, Schaper A, Jenei A, Matko J, Starink JP, Fox GQ, Arndt-Jovin DJ and Jovin TM (1995) Structural hierarchy in the clustering of HLA class I molecules in the plasma membrane of human lymphoblastoid cells. Proc Natl Acad Sci U S A 92(4):1122- 1126.

Dangoria NS, Breau WC, Anderson HA, Cishek DM and Norkin LC (1996) Extracellular simian virus 40 induces an ERK/MAP kinase-independent signalling pathway that activates primary response genes and promotes virus entry. J Gen Virol 77 (Pt 9):2173-2182.

de Weerd WF and Leeb-Lundberg LM (1997) Bradykinin sequesters B2 bradykinin receptors and the receptor-coupled Galpha subunits Galphaq and Galphai in caveolae in DDT1 MF-2 smooth muscle cells. J Biol Chem 272(28):17858-17866.

Devi LA (2001) Heterodimerization of G-protein-coupled receptors: pharmacology, signaling and trafficking. Trends Pharmacol Sci 22(10):532-537.

Di Guglielmo GM, Le Roy C, Goodfellow AF and Wrana JL (2003) Distinct endocytic pathways regulate TGF-beta receptor signalling and turnover. Nat Cell Biol 5(5):410-421.

Dietzen DJ, Hastings WR and Lublin DM (1995) Caveolin is palmitoylated on multiple cysteine residues. Palmitoylation is not necessary for localization of caveolin to caveolae. J Biol Chem 270(12):6838-6842.

Digby GJ, Lober RM, Sethi PR and Lambert NA (2006) Some G protein heterotrimers physically dissociate in living cells. Proc Natl Acad Sci U S A 103(47):17789-17794.

Dong C, Filipeanu CM, Duvernay MT and Wu G (2007) Regulation of G protein-coupled receptor export trafficking. Biochim Biophys Acta 1768(4):853-870.

Drab M, Verkade P, Elger M, Kasper M, Lohn M, Lauterbach B, Menne J, Lindschau C, Mende F, Luft FC, Schedl A, Haller H and Kurzchalia TV (2001) Loss of caveolae, vascular dysfunction, and pulmonary defects in caveolin-1 gene-disrupted mice. Science 293(5539):2449-2452.

Dracheva S, Xu M, Kelley KA, Haroutunian V, Holstein GR, Haun S, Silverstein JH and Sealfon SC (1999) Paradoxical locomotor behavior of dopamine D1 receptor transgenic mice. Exp Neurol 157(1):169-179.

Drago J, Gerfen CR, Westphal H and Steiner H (1996) D1 dopamine receptor-deficient mouse: cocaine-induced regulation of immediate-early gene and substance P expression in the striatum. Neuroscience 74(3):813-823.

Drews J (2000) Drug discovery: a historical perspective. Science 287(5460):1960-1964.

234 Du D, Raaka BM, Grimberg H, Lupu-Meiri M, Oron Y and Gershengorn MC (2005) Carboxyl tail cysteine mutants of the thyrotropin-releasing hormone receptor type 1 exhibit constitutive signaling: role of palmitoylation. Mol Pharmacol 68(1):204-209.

Dupre DJ and Hebert TE (2006) Biosynthesis and trafficking of seven transmembrane receptor signalling complexes. Cell Signal 18(10):1549-1559.

Duthey B, Caudron S, Perroy J, Bettler B, Fagni L, Pin JP and Prezeau L (2002) A single subunit (GB2) is required for G-protein activation by the heterodimeric GABA(B) receptor. J Biol Chem 277(5):3236-3241.

Dykstra ML, Cherukuri A and Pierce SK (2001) Floating the raft hypothesis for immune receptors: access to rafts controls receptor signaling and trafficking. Traffic 2(3):160-166.

Edidin M, Zuniga MC and Sheetz MP (1994) Truncation mutants define and locate cytoplasmic barriers to lateral mobility of membrane glycoproteins. Proc Natl Acad Sci U S A 91(8):3378-3382.

Eidne KA, Kroeger KM and Hanyaloglu AC (2002) Applications of novel resonance energy transfer techniques to study dynamic hormone receptor interactions in living cells. Trends Endocrinol Metab 13(10):415-421.

Eilers M, Hornak V, Smith SO and Konopka JB (2005) Comparison of class A and D G protein- coupled receptors: common features in structure and activation. Biochemistry 44(25):8959-8975.

El-Ghundi M, Fletcher PJ, Drago J, Sibley DR, O'Dowd BF and George SR (1999) Spatial learning deficit in dopamine D(1) receptor knockout mice. Eur J Pharmacol 383(2):95- 106.

El-Ghundi M, George SR, Drago J, Fletcher PJ, Fan T, Nguyen T, Liu C, Sibley DR, Westphal H and O'Dowd BF (1998) Disruption of dopamine D1 receptor gene expression attenuates alcohol-seeking behavior. Eur J Pharmacol 353(2-3):149-158.

El-Ghundi M, O'Dowd BF, Erclik M and George SR (2003) Attenuation of sucrose reinforcement in dopamine D1 receptor deficient mice. Eur J Neurosci 17(4):851-862.

El-Ghundi M, O'Dowd BF and George SR (2001) Prolonged fear responses in mice lacking dopamine D1 receptor. Brain Res 892(1):86-93.

El-Ghundi M, O'Dowd BF and George SR (2007) Insights into the role of dopamine receptor systems in learning and memory. Rev Neurosci 18(1):37-66.

235 el-Husseini Ael D and Bredt DS (2002) Protein palmitoylation: a regulator of neuronal development and function. Nat Rev Neurosci 3(10):791-802.

Elliot EE, Sibley DR and Katz JL (2003) Locomotor and discriminative-stimulus effects of cocaine in dopamine D5 receptor knockout mice. Psychopharmacology (Berl) 169(2):161-168.

Elmhurst JL, Xie Z, O'Dowd BF and George SR (2000) The splice variant D3nf reduces ligand binding to the D3 dopamine receptor: evidence for heterooligomerization. Brain Res Mol Brain Res 80(1):63-74.

Ernst OP, Meyer CK, Marin EP, Henklein P, Fu WY, Sakmar TP and Hofmann KP (2000) Mutation of the fourth cytoplasmic loop of rhodopsin affects binding of transducin and peptides derived from the carboxyl-terminal sequences of transducin alpha and gamma subunits. J Biol Chem 275(3):1937-1943.

Escriba PV, Wedegaertner PB, Goni FM and Vogler O (2006) Lipid-protein interactions in GPCR-associated signaling. Biochim Biophys Acta.

Farrens DL, Altenbach C, Yang K, Hubbell WL and Khorana HG (1996) Requirement of rigid- body motion of transmembrane helices for light activation of rhodopsin. Science 274(5288):768-770.

Ferguson SS (2001) Evolving concepts in G protein-coupled receptor endocytosis: the role in receptor desensitization and signaling. Pharmacol Rev 53(1):1-24.

Filizola M, Olmea O and Weinstein H (2002) Prediction of heterodimerization interfaces of G- protein coupled receptors with a new subtractive correlated mutation method. Protein Eng 15(11):881-885.

Filizola M and Weinstein H (2002) Structural models for dimerization of G-protein coupled receptors: the opioid receptor homodimers. Biopolymers 66(5):317-325.

Filizola M and Weinstein H (2005) The study of G-protein coupled receptor oligomerization with computational modeling and bioinformatics. Febs J 272(12):2926-2938.

Fotiadis D, Liang Y, Filipek S, Saperstein DA, Engel A and Palczewski K (2003) Atomic-force microscopy: Rhodopsin dimers in native disc membranes. Nature 421(6919):127-128.

Fotiadis D, Liang Y, Filipek S, Saperstein DA, Engel A and Palczewski K (2004) The G protein- coupled receptor rhodopsin in the native membrane. FEBS Lett 564(3):281-288.

236 Fredriksson R, Lagerstrom MC, Lundin LG and Schioth HB (2003) The G-protein-coupled receptors in the human genome form five main families. Phylogenetic analysis, paralogon groups, and fingerprints. Mol Pharmacol 63(6):1256-1272.

Free RB, Hazelwood LA, Cabrera DM, Spalding HN, Namkung Y, Rankin ML and Sibley DR (2007) D1 and D2 dopamine receptor expression is regulated by direct interaction with the chaperone protein calnexin. J Biol Chem.

Friedlander M and Blobel G (1985) Bovine opsin has more than one signal sequence. Nature 318(6044):338-343.

Friedman E, Jin LQ, Cai GP, Hollon TR, Drago J, Sibley DR and Wang HY (1997) D1-like dopaminergic activation of phosphoinositide hydrolysis is independent of D1A dopamine receptors: evidence from D1A knockout mice. Mol Pharmacol 51(1):6-11.

Fu H, Subramanian RR and Masters SC (2000) 14-3-3 proteins: structure, function, and regulation. Annu Rev Pharmacol Toxicol 40:617-647.

Fujimoto T, Kogo H, Nomura R and Une T (2000) Isoforms of caveolin-1 and caveolar structure. J Cell Sci 113 Pt 19:3509-3517.

Fukushima Y, Saitoh T, Anai M, Ogihara T, Inukai K, Funaki M, Sakoda H, Onishi Y, Ono H, Fujishiro M, Ishikawa T, Takata K, Nagai R, Omata M and Asano T (2001) Palmitoylation of the canine histamine H2 receptor occurs at Cys(305) and is important for cell surface targeting. Biochim Biophys Acta 1539(3):181-191.

Furthmayr H and Marchesi VT (1976) Subunit structure of human erythrocyte glycophorin A. Biochemistry 15(5):1137-1144.

Gaborik Z, Szaszak M, Szidonya L, Balla B, Paku S, Catt KJ, Clark AJ and Hunyady L (2001) Beta-arrestin- and dynamin-dependent endocytosis of the AT1 angiotensin receptor. Mol Pharmacol 59(2):239-247.

Gagescu R, Demaurex N, Parton RG, Hunziker W, Huber LA and Gruenberg J (2000) The recycling endosome of Madin-Darby canine kidney cells is a mildly acidic compartment rich in raft components. Mol Biol Cell 11(8):2775-2791.

Galbiati F, Volonte D, Gil O, Zanazzi G, Salzer JL, Sargiacomo M, Scherer PE, Engelman JA, Schlegel A, Parenti M, Okamoto T and Lisanti MP (1998) Expression of caveolin-1 and - 2 in differentiating PC12 cells and dorsal root ganglion neurons: caveolin-2 is up- regulated in response to cell injury. Proc Natl Acad Sci U S A 95(17):10257-10262.

237 Galbiati F, Volonte D, Meani D, Milligan G, Lublin DM, Lisanti MP and Parenti M (1999) The dually acylated NH2-terminal domain of gi1alpha is sufficient to target a green fluorescent protein reporter to caveolin-enriched plasma membrane domains. Palmitoylation of caveolin-1 is required for the recognition of dually acylated g-protein alpha subunits in vivo. J Biol Chem 274(9):5843-5850.

Gales C, Van Durm JJ, Schaak S, Pontier S, Percherancier Y, Audet M, Paris H and Bouvier M (2006) Probing the activation-promoted structural rearrangements in preassembled receptor-G protein complexes. Nat Struct Mol Biol 13(9):778-786.

Galvez T, Duthey B, Kniazeff J, Blahos J, Rovelli G, Bettler B, Prezeau L and Pin JP (2001) Allosteric interactions between GB1 and GB2 subunits are required for optimal GABA(B) receptor function. Embo J 20(9):2152-2159.

Gama L, Wilt SG and Breitwieser GE (2001) Heterodimerization of calcium sensing receptors with metabotropic glutamate receptors in neurons. J Biol Chem 276(42):39053-39059.

Garcia-Cardena G, Martasek P, Masters BS, Skidd PM, Couet J, Li S, Lisanti MP and Sessa WC (1997) Dissecting the interaction between nitric oxide synthase (NOS) and caveolin. Functional significance of the nos caveolin binding domain in vivo. J Biol Chem 272(41):25437-25440.

Gaudreault SB, Blain JF, Gratton JP and Poirier J (2005) A role for caveolin-1 in post-injury reactive neuronal plasticity. J Neurochem 92(4):831-839.

Gazi L, Lopez-Gimenez JF, Rudiger MP and Strange PG (2003) Constitutive oligomerization of human D2 dopamine receptors expressed in Spodoptera frugiperda 9 (Sf9) and in HEK293 cells. Analysis using co-immunoprecipitation and time-resolved fluorescence resonance energy transfer. Eur J Biochem 270(19):3928-3938.

George SR, Fan T, Xie Z, Tse R, Tam V, Varghese G and O'Dowd BF (2000) Oligomerization of mu- and delta-opioid receptors. Generation of novel functional properties. J Biol Chem 275(34):26128-26135.

George SR, Lee SP, Varghese G, Zeman PR, Seeman P, Ng GY and O'Dowd BF (1998) A transmembrane domain-derived peptide inhibits D1 dopamine receptor function without affecting receptor oligomerization. J Biol Chem 273(46):30244-30248.

George SR, O'Dowd BF and Lee SP (2002) G-protein-coupled receptor oligomerization and its potential for drug discovery. Nat Rev Drug Discov 1(10):808-820.

Gether U and Kobilka BK (1998) G protein-coupled receptors. II. Mechanism of agonist activation. J Biol Chem 273(29):17979-17982.

238 Gether U, Lin S, Ghanouni P, Ballesteros JA, Weinstein H and Kobilka BK (1997) Agonists induce conformational changes in transmembrane domains III and VI of the beta2 adrenoceptor. Embo J 16(22):6737-6747.

Ghanouni P, Gryczynski Z, Steenhuis JJ, Lee TW, Farrens DL, Lakowicz JR and Kobilka BK (2001) Functionally different agonists induce distinct conformations in the G protein coupling domain of the beta 2 adrenergic receptor. J Biol Chem 276(27):24433-24436.

Ghitescu L, Fixman A, Simionescu M and Simionescu N (1986) Specific binding sites for albumin restricted to plasmalemmal vesicles of continuous capillary endothelium: receptor-mediated transcytosis. J Cell Biol 102(4):1304-1311.

Giguere V, Gallant MA, de Brum-Fernandes AJ and Parent JL (2004) Role of extracellular cysteine residues in dimerization/oligomerization of the human prostacyclin receptor. Eur J Pharmacol 494(1):11-22.

Gines S, Hillion J, Torvinen M, Le Crom S, Casado V, Canela EI, Rondin S, Lew JY, Watson S, Zoli M, Agnati LF, Verniera P, Lluis C, Ferre S, Fuxe K and Franco R (2000) Dopamine D1 and adenosine A1 receptors form functionally interacting heteromeric complexes. Proc Natl Acad Sci U S A 97(15):8606-8611.

Glatt CE and Snyder SH (1993) Cloning and expression of an adenylyl cyclase localized to the corpus striatum. Nature 361(6412):536-538.

Glusman G, Yanai I, Rubin I and Lancet D (2001) The complete human olfactory subgenome. Genome Res 11(5):685-702.

Gomes I, Jordan BA, Gupta A, Trapaidze N, Nagy V and Devi LA (2000) Heterodimerization of mu and delta opioid receptors: A role in opiate synergy. J Neurosci 20(22):RC110.

Gouldson PR, Higgs C, Smith RE, Dean MK, Gkoutos GV and Reynolds CA (2000) Dimerization and domain swapping in G-protein-coupled receptors: a computational study. Neuropsychopharmacology 23(4 Suppl):S60-77.

Gouldson PR, Snell CR, Bywater RP, Higgs C and Reynolds CA (1998) Domain swapping in G- protein coupled receptor dimers. Protein Eng 11(12):1181-1193.

Grosse R, Schoneberg T, Schultz G and Gudermann T (1997) Inhibition of gonadotropin- releasing hormone receptor signaling by expression of a splice variant of the human receptor. Mol Endocrinol 11(9):1305-1318.

Guo W, Shi L and Javitch JA (2003) The fourth transmembrane segment forms the interface of the dopamine D2 receptor homodimer. J Biol Chem 278(7):4385-4388.

239 Guzzi F, Zanchetta D, Cassoni P, Guzzi V, Francolini M, Parenti M and Chini B (2002) Localization of the human oxytocin receptor in caveolin-1 enriched domains turns the receptor-mediated inhibition of cell growth into a proliferative response. Oncogene 21(11):1658-1667.

Hague C, Uberti MA, Chen Z, Bush CF, Jones SV, Ressler KJ, Hall RA and Minneman KP (2004a) Olfactory receptor surface expression is driven by association with the beta2- adrenergic receptor. Proc Natl Acad Sci U S A 101(37):13672-13676.

Hague C, Uberti MA, Chen Z, Hall RA and Minneman KP (2004b) Cell surface expression of alpha1D-adrenergic receptors is controlled by heterodimerization with alpha1B- adrenergic receptors. J Biol Chem 279(15):15541-15549.

Hamm HE (2001) How activated receptors couple to G proteins. Proc Natl Acad Sci U S A 98(9):4819-4821.

Hancock JF, Paterson H and Marshall CJ (1990) A polybasic domain or palmitoylation is required in addition to the CAAX motif to localize p21ras to the plasma membrane. Cell 63(1):133-139.

Harikumar KG, Puri V, Singh RD, Hanada K, Pagano RE and Miller LJ (2005) Differential effects of modification of membrane cholesterol and sphingolipids on the conformation, function, and trafficking of the G protein-coupled cholecystokinin receptor. J Biol Chem 280(3):2176-2185.

Harmar AJ (2001) Family-B G-protein-coupled receptors. Genome Biol 2(12):REVIEWS3013.

Hawtin SR, Tobin AB, Patel S and Wheatley M (2001) Palmitoylation of the vasopressin V1a receptor reveals different conformational requirements for signaling, agonist-induced receptor phosphorylation, and sequestration. J Biol Chem 276(41):38139-38146.

Hayashi MK and Haga T (1997) Palmitoylation of muscarinic acetylcholine receptor m2 subtypes: reduction in their ability to activate G proteins by mutation of a putative palmitoylation site, cysteine 457, in the carboxyl-terminal tail. Arch Biochem Biophys 340(2):376-382.

Head BP, Patel HH, Roth DM, Murray F, Swaney JS, Niesman IR, Farquhar MG and Insel PA (2006) Microtubules and actin microfilaments regulate lipid raft/caveolae localization of adenylyl cyclase signaling components. J Biol Chem 281(36):26391-26399.

Heldin CH (1995) Dimerization of cell surface receptors in signal transduction. Cell 80(2):213- 223.

240 Hemar A, Olivo JC, Williamson E, Saffrich R and Dotti CG (1997) Dendroaxonal transcytosis of transferrin in cultured hippocampal and sympathetic neurons. J Neurosci 17(23):9026- 9034.

Henley JR, Krueger EW, Oswald BJ and McNiven MA (1998) Dynamin-mediated internalization of caveolae. J Cell Biol 141(1):85-99.

Herve D, Le Moine C, Corvol JC, Belluscio L, Ledent C, Fienberg AA, Jaber M, Studler JM and Girault JA (2001) Galpha(olf) levels are regulated by receptor usage and control dopamine and adenosine action in the striatum. J Neurosci 21(12):4390-4399.

Heuser JE and Anderson RG (1989) Hypertonic media inhibit receptor-mediated endocytosis by blocking clathrin-coated pit formation. J Cell Biol 108(2):389-400.

Hiol A, Davey PC, Osterhout JL, Waheed AA, Fischer ER, Chen CK, Milligan G, Druey KM and Jones TL (2003) Palmitoylation regulates regulators of G-protein signaling (RGS) 16 function. I. Mutation of amino-terminal cysteine residues on RGS16 prevents its targeting to lipid rafts and palmitoylation of an internal cysteine residue. J Biol Chem 278(21):19301-19308.

Houndolo T, Boulay PL and Claing A (2005) G protein-coupled receptor endocytosis in ADP- ribosylation factor 6-depleted cells. J Biol Chem 280(7):5598-5604.

Hukovic N, Panetta R, Kumar U, Rocheville M and Patel YC (1998) The cytoplasmic tail of the human somatostatin receptor type 5 is crucial for interaction with adenylyl cyclase and in mediating desensitization and internalization. J Biol Chem 273(33):21416-21422.

Hurtley SM, Bole DG, Hoover-Litty H, Helenius A and Copeland CS (1989) Interactions of misfolded influenza virus hemagglutinin with binding protein (BiP). J Cell Biol 108(6):2117-2126.

Igarashi J and Michel T (2000) Agonist-modulated targeting of the EDG-1 receptor to plasmalemmal caveolae. eNOS activation by sphingosine 1-phosphate and the role of caveolin-1 in sphingolipid signal transduction. J Biol Chem 275(41):32363-32370.

Ikezu T, Ueda H, Trapp BD, Nishiyama K, Sha JF, Volonte D, Galbiati F, Byrd AL, Bassell G, Serizawa H, Lane WS, Lisanti MP and Okamoto T (1998) Affinity-purification and characterization of caveolins from the brain: differential expression of caveolin-1, -2, and -3 in brain endothelial and astroglial cell types. Brain Res 804(2):177-192.

Iversen L (2006) Neurotransmitter transporters and their impact on the development of psychopharmacology. Br J Pharmacol 147 Suppl 1:S82-88.

241 Iwasiow RM, Nantel MF and Tiberi M (1999) Delineation of the structural basis for the activation properties of the dopamine D1 receptor subtypes. J Biol Chem 274(45):31882- 31890.

Janovick JA, Goulet M, Bush E, Greer J, Wettlaufer DG and Conn PM (2003) Structure-activity relations of successful pharmacologic chaperones for rescue of naturally occurring and manufactured mutants of the gonadotropin-releasing hormone receptor. J Pharmacol Exp Ther 305(2):608-614.

Jenei A, Varga S, Bene L, Matyus L, Bodnar A, Bacso Z, Pieri C, Gaspar R, Jr., Farkas T and Damjanovich S (1997) HLA class I and II antigens are partially co-clustered in the plasma membrane of human lymphoblastoid cells. Proc Natl Acad Sci U S A 94(14):7269-7274.

Jensen AA, Hansen JL, Sheikh SP and Brauner-Osborne H (2002) Probing intermolecular protein-protein interactions in the calcium-sensing receptor homodimer using bioluminescence resonance energy transfer (BRET). Eur J Biochem 269(20):5076-5087.

Jiang D and Sibley DR (1999) Regulation of D(1) dopamine receptors with mutations of protein kinase phosphorylation sites: attenuation of the rate of agonist-induced desensitization. Mol Pharmacol 56(4):675-683.

Jin H, Xie Z, George SR and O'Dowd BF (1999) Palmitoylation occurs at cysteine 347 and cysteine 351 of the dopamine D(1) receptor. Eur J Pharmacol 386(2-3):305-312.

Jin H, Zastawny R, George SR and O'Dowd BF (1997) Elimination of palmitoylation sites in the human dopamine D1 receptor does not affect receptor-G protein interaction. Eur J Pharmacol 324(1):109-116.

Jin LQ, Goswami S, Cai G, Zhen X and Friedman E (2003) SKF83959 selectively regulates phosphatidylinositol-linked D1 dopamine receptors in rat brain. J Neurochem 85(2):378- 386.

Jones KA, Borowsky B, Tamm JA, Craig DA, Durkin MM, Dai M, Yao WJ, Johnson M, Gunwaldsen C, Huang LY, Tang C, Shen Q, Salon JA, Morse K, Laz T, Smith KE, Nagarathnam D, Noble SA, Branchek TA and Gerald C (1998) GABA(B) receptors function as a heteromeric assembly of the subunits GABA(B)R1 and GABA(B)R2. Nature 396(6712):674-679.

Jordan BA and Devi LA (1999) G-protein-coupled receptor heterodimerization modulates receptor function. Nature 399(6737):697-700.

242 Kamen BA, Wang MT, Streckfuss AJ, Peryea X and Anderson RG (1988) Delivery of folates to the cytoplasm of MA104 cells is mediated by a surface membrane receptor that recycles. J Biol Chem 263(27):13602-13609.

Kapoor M, Srinivas H, Kandiah E, Gemma E, Ellgaard L, Oscarson S, Helenius A and Surolia A (2003) Interactions of substrate with calreticulin, an endoplasmic reticulum chaperone. J Biol Chem 278(8):6194-6200.

Karasinska JM, George SR, Cheng R and O'Dowd BF (2005) Deletion of dopamine D1 and D3 receptors differentially affects spontaneous behaviour and cocaine-induced locomotor activity, reward and CREB phosphorylation. Eur J Neurosci 22(7):1741-1750.

Karnik SS, Ridge KD, Bhattacharya S and Khorana HG (1993) Palmitoylation of bovine opsin and its cysteine mutants in COS cells. Proc Natl Acad Sci U S A 90(1):40-44.

Karpa KD, Lidow MS, Pickering MT, Levenson R and Bergson C (1999) N-linked glycosylation is required for plasma membrane localization of D5, but not D1, dopamine receptors in transfected mammalian cells. Mol Pharmacol 56(5):1071-1078.

Karpa KD, Lin R, Kabbani N and Levenson R (2000) The dopamine D3 receptor interacts with itself and the truncated D3 splice variant d3nf: D3-D3nf interaction causes mislocalization of D3 receptors. Mol Pharmacol 58(4):677-683.

Kaupmann K, Malitschek B, Schuler V, Heid J, Froestl W, Beck P, Mosbacher J, Bischoff S, Kulik A, Shigemoto R, Karschin A and Bettler B (1998) GABA(B)-receptor subtypes assemble into functional heteromeric complexes. Nature 396(6712):683-687.

Kebabian JW, Petzold GL and Greengard P (1972) Dopamine-sensitive adenylate cyclase in caudate nucleus of rat brain, and its similarity to the "dopamine receptor". Proc Natl Acad Sci U S A 69(8):2145-2149.

Kennedy ME and Limbird LE (1993) Mutations of the alpha 2A-adrenergic receptor that eliminate detectable palmitoylation do not perturb receptor-G-protein coupling. J Biol Chem 268(11):8003-8011.

Kennedy ME and Limbird LE (1994) Palmitoylation of the alpha 2A-adrenergic receptor. Analysis of the sequence requirements for and the dynamic properties of alpha 2A- adrenergic receptor palmitoylation. J Biol Chem 269(50):31915-31922.

Kifor O, Diaz R, Butters R, Kifor I and Brown EM (1998) The calcium-sensing receptor is localized in caveolin-rich plasma membrane domains of bovine parathyroid cells. J Biol Chem 273(34):21708-21713.

243 Kim OJ, Ariano MA, Lazzarini RA, Levine MS and Sibley DR (2002) Neurofilament-M interacts with the D1 dopamine receptor to regulate cell surface expression and desensitization. J Neurosci 22(14):5920-5930.

Kjelsberg MA, Cotecchia S, Ostrowski J, Caron MG and Lefkowitz RJ (1992) Constitutive activation of the alpha 1B-adrenergic receptor by all amino acid substitutions at a single site. Evidence for a region which constrains receptor activation. J Biol Chem 267(3):1430-1433.

Klco JM, Lassere TB and Baranski TJ (2003) C5a receptor oligomerization. I. Disulfide trapping reveals oligomers and potential contact surfaces in a G protein-coupled receptor. J Biol Chem 278(37):35345-35353.

Kniazeff J, Bessis AS, Maurel D, Ansanay H, Prezeau L and Pin JP (2004a) Closed state of both binding domains of homodimeric mGlu receptors is required for full activity. Nat Struct Mol Biol 11(8):706-713.

Kniazeff J, Saintot PP, Goudet C, Liu J, Charnet A, Guillon G and Pin JP (2004b) Locking the dimeric GABA(B) G-protein-coupled receptor in its active state. J Neurosci 24(2):370- 377.

Koleske AJ, Baltimore D and Lisanti MP (1995) Reduction of caveolin and caveolae in oncogenically transformed cells. Proc Natl Acad Sci U S A 92(5):1381-1385.

Kraft K, Olbrich H, Majoul I, Mack M, Proudfoot A and Oppermann M (2001) Characterization of sequence determinants within the carboxyl-terminal domain of chemokine receptor CCR5 that regulate signaling and receptor internalization. J Biol Chem 276(37):34408- 34418.

Kroeger KM, Hanyaloglu AC, Seeber RM, Miles LE and Eidne KA (2001) Constitutive and agonist-dependent homo-oligomerization of the thyrotropin-releasing hormone receptor. Detection in living cells using bioluminescence resonance energy transfer. J Biol Chem 276(16):12736-12743.

Kunishima N, Shimada Y, Tsuji Y, Sato T, Yamamoto M, Kumasaka T, Nakanishi S, Jingami H and Morikawa K (2000) Structural basis of glutamate recognition by a dimeric metabotropic glutamate receptor. Nature 407(6807):971-977.

Lamb ME, Zhang C, Shea T, Kyle DJ and Leeb-Lundberg LM (2002) Human B1 and B2 bradykinin receptors and their agonists target caveolae-related lipid rafts to different degrees in HEK293 cells. Biochemistry 41(48):14340-14347.

244 Lamey M, Thompson M, Varghese G, Chi H, Sawzdargo M, George SR and O'Dowd BF (2002) Distinct residues in the carboxyl tail mediate agonist-induced desensitization and internalization of the human dopamine D1 receptor. J Biol Chem 277(11):9415-9421.

Le Gouill C, Parent JL, Caron CA, Gaudreau R, Volkov L, Rola-Pleszczynski M and Stankova J (1999) Selective modulation of wild type receptor functions by mutants of G-protein- coupled receptors. J Biol Chem 274(18):12548-12554.

Le Page SL, Bi Y and Williams JA (2003) CCK-A receptor activates RhoA through G alpha 12/13 in NIH3T3 cells. Am J Physiol Cell Physiol 285(5):C1197-1206.

Leclerc PC, Auger-Messier M, Lanctot PM, Escher E, Leduc R and Guillemette G (2002) A polyaromatic caveolin-binding-like motif in the cytoplasmic tail of the type 1 receptor for angiotensin II plays an important role in receptor trafficking and signaling. Endocrinology 143(12):4702-4710.

Lee H, Park DS, Razani B, Russell RG, Pestell RG and Lisanti MP (2002a) Caveolin-1 mutations (P132L and null) and the pathogenesis of breast cancer: caveolin-1 (P132L) behaves in a dominant-negative manner and caveolin-1 (-/-) null mice show mammary epithelial cell hyperplasia. Am J Pathol 161(4):1357-1369.

Lee KW, Hong JH, Choi IY, Che Y, Lee JK, Yang SD, Song CW, Kang HS, Lee JH, Noh JS, Shin HS and Han PL (2002b) Impaired D2 dopamine receptor function in mice lacking type 5 adenylyl cyclase. J Neurosci 22(18):7931-7940.

Lee SP, O'Dowd BF and George SR (2003a) Homo- and hetero-oligomerization of G protein- coupled receptors. Life Sci 74(2-3):173-180.

Lee SP, O'Dowd BF, Ng GY, Varghese G, Akil H, Mansour A, Nguyen T and George SR (2000a) Inhibition of cell surface expression by mutant receptors demonstrates that D2 dopamine receptors exist as oligomers in the cell. Mol Pharmacol 58(1):120-128.

Lee SP, O'Dowd BF, Rajaram RD, Nguyen T and George SR (2003b) D2 dopamine receptor homodimerization is mediated by multiple sites of interaction, including an intermolecular interaction involving transmembrane domain 4. Biochemistry 42(37):11023-11031.

Lee SP, So CH, Rashid AJ, Varghese G, Cheng R, Lanca AJ, O'Dowd BF and George SR (2004) Dopamine D1 and D2 receptor Co-activation generates a novel phospholipase C- mediated calcium signal. J Biol Chem 279(34):35671-35678.

Lee SP, Xie Z, Varghese G, Nguyen T, O'Dowd BF and George SR (2000b) Oligomerization of dopamine and serotonin receptors. Neuropsychopharmacology 23(4 Suppl):S32-40.

245 Lefkowitz RJ (1998) G protein-coupled receptors. III. New roles for receptor kinases and beta- arrestins in receptor signaling and desensitization. J Biol Chem 273(30):18677-18680.

Lefkowitz RJ, Cotecchia S, Samama P and Costa T (1993) Constitutive activity of receptors coupled to guanine nucleotide regulatory proteins. Trends Pharmacol Sci 14(8):303-307.

Lei Y, Hagen GM, Smith SM, Barisas BG and Roess DA (2005) Chimeric GnRH-LH receptors and LH receptors lacking C-terminus palmitoylation sites do not localize to plasma membrane rafts. Biochem Biophys Res Commun 337(2):430-434.

Lemmon MA and Engelman DM (1994) Specificity and promiscuity in membrane helix interactions. FEBS Lett 346(1):17-20.

Lemmon MA, Flanagan JM, Hunt JF, Adair BD, Bormann BJ, Dempsey CE and Engelman DM (1992) Glycophorin A dimerization is driven by specific interactions between transmembrane alpha-helices. J Biol Chem 267(11):7683-7689.

Levac BA, O'Dowd BF and George SR (2002) Oligomerization of opioid receptors: generation of novel signaling units. Curr Opin Pharmacol 2(1):76-81.

Li S, Okamoto T, Chun M, Sargiacomo M, Casanova JE, Hansen SH, Nishimoto I and Lisanti MP (1995) Evidence for a regulated interaction between heterotrimeric G proteins and caveolin. J Biol Chem 270(26):15693-15701.

Liang Y, Fotiadis D, Filipek S, Saperstein DA, Palczewski K and Engel A (2003) Organization of the G protein-coupled receptors rhodopsin and opsin in native membranes. J Biol Chem 278(24):21655-21662.

Liao RM, Chang YH and Wang SH (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(2):85-92.

Limbird LE, Meyts PD and Lefkowitz RJ (1975) Beta-adrenergic receptors: evidence for negative cooperativity. Biochem Biophys Res Commun 64(4):1160-1168.

Linder ME and Deschenes RJ (2007) Palmitoylation: policing protein stability and traffic. Nat Rev Mol Cell Biol 8(1):74-84.

Lisanti MP, Scherer PE, Vidugiriene J, Tang Z, Hermanowski-Vosatka A, Tu YH, Cook RF and Sargiacomo M (1994) Characterization of caveolin-rich membrane domains isolated from an endothelial-rich source: implications for human disease. J Cell Biol 126(1):111-126.

246 Liu K, Bergson C, Levenson R and Schmauss C (1994) On the origin of mRNA encoding the truncated dopamine D3-type receptor D3nf and detection of D3nf-like immunoreactivity in human brain. J Biol Chem 269(46):29220-29226.

Liu P, Ying Y, Ko YG and Anderson RG (1996) Localization of platelet-derived growth factor- stimulated phosphorylation cascade to caveolae. J Biol Chem 271(17):10299-10303.

Lopez-Gimenez JF, Canals M, Pediani JD and Milligan G (2007) The {alpha}1b-adrenoceptor exists as a higher-order oligomer: effective oligomerization is required for receptor maturation, surface delivery and function. Mol Pharmacol.

Lu K and Guidotti G (1996) Identification of the cysteine residues involved in the class I disulfide bonds of the human insulin receptor: properties of insulin receptor monomers. Mol Biol Cell 7(5):679-691.

Ma D and Jan LY (2002) ER transport signals and trafficking of potassium channels and receptors. Curr Opin Neurobiol 12(3):287-292.

Maggio R, Barbier P, Colelli A, Salvadori F, Demontis G and Corsini GU (1999) G protein- linked receptors: pharmacological evidence for the formation of heterodimers. J Pharmacol Exp Ther 291(1):251-257.

Maggio R, Vogel Z and Wess J (1993) Coexpression studies with mutant muscarinic/adrenergic receptors provide evidence for intermolecular "cross-talk" between G-protein-linked receptors. Proc Natl Acad Sci U S A 90(7):3103-3107.

Mahan LC, Burch RM, Monsma FJ, Jr. and Sibley DR (1990) Expression of striatal D1 dopamine receptors coupled to inositol phosphate production and Ca2+ mobilization in Xenopus oocytes. Proc Natl Acad Sci U S A 87(6):2196-2200.

Malaba L, Smeland S, Senoo H, Norum KR, Berg T, Blomhoff R and Kindberg GM (1995) Retinol-binding protein and asialo-orosomucoid are taken up by different pathways in liver cells. J Biol Chem 270(26):15686-15692.

Manders EMM, Verbeek FJ and Aten JA (1993) Measurement of co-localisation of objects in dual colour confocal images. J Microsc 169:375-382.

Manolios N, Bonifacino JS and Klausner RD (1990) Transmembrane helical interactions and the assembly of the T cell receptor complex. Science 249(4966):274-277.

Marchese A and Benovic JL (2001) Agonist-promoted ubiquitination of the G protein-coupled receptor CXCR4 mediates lysosomal sorting. J Biol Chem 276(49):45509-45512.

247 Margeta-Mitrovic M, Jan YN and Jan LY (2000) A trafficking checkpoint controls GABA(B) receptor heterodimerization. Neuron 27(1):97-106.

Margeta-Mitrovic M, Jan YN and Jan LY (2001) Function of GB1 and GB2 subunits in G protein coupling of GABA(B) receptors. Proc Natl Acad Sci U S A 98(25):14649-14654.

Marin EP, Krishna AG, Zvyaga TA, Isele J, Siebert F and Sakmar TP (2000) The amino terminus of the fourth cytoplasmic loop of rhodopsin modulates rhodopsin-transducin interaction. J Biol Chem 275(3):1930-1936.

Marti T (1998) Refolding of bacteriorhodopsin from expressed polypeptide fragments. J Biol Chem 273(15):9312-9322.

Martin NP, Lefkowitz RJ and Shenoy SK (2003) Regulation of V2 vasopressin receptor degradation by agonist-promoted ubiquitination. J Biol Chem 278(46):45954-45959.

Mason JN, Kozell LB and Neve KA (2002) Regulation of dopamine D(1) receptor trafficking by protein kinase A-dependent phosphorylation. Mol Pharmacol 61(4):806-816.

Mastick CC, Brady MJ and Saltiel AR (1995) Insulin stimulates the tyrosine phosphorylation of caveolin. J Cell Biol 129(6):1523-1531.

Mattera R, Pitts BJ, Entman ML and Birnbaumer L (1985) Guanine nucleotide regulation of a mammalian myocardial muscarinic receptor system. Evidence for homo- and heterotropic cooperativity in ligand binding analyzed by computer-assisted curve fitting. J Biol Chem 260(12):7410-7421.

McLatchie LM, Fraser NJ, Main MJ, Wise A, Brown J, Thompson N, Solari R, Lee MG and Foord SM (1998) RAMPs regulate the transport and ligand specificity of the calcitonin- receptor-like receptor. Nature 393(6683):333-339.

McVey M, Ramsay D, Kellett E, Rees S, Wilson S, Pope AJ and Milligan G (2001) Monitoring receptor oligomerization using time-resolved fluorescence resonance energy transfer and bioluminescence resonance energy transfer. The human delta -opioid receptor displays constitutive oligomerization at the cell surface, which is not regulated by receptor occupancy. J Biol Chem 276(17):14092-14099.

Menard L, Ferguson SS, Zhang J, Lin FT, Lefkowitz RJ, Caron MG and Barak LS (1997) Synergistic regulation of beta2-adrenergic receptor sequestration: intracellular complement of beta-adrenergic receptor kinase and beta-arrestin determine kinetics of internalization. Mol Pharmacol 51(5):800-808.

248 Mercier JF, Salahpour A, Angers S, Breit A and Bouvier M (2002) Quantitative assessment of beta 1- and beta 2-adrenergic receptor homo- and heterodimerization by bioluminescence resonance energy transfer. J Biol Chem 277(47):44925-44931.

Mesnier D and Baneres JL (2004) Cooperative conformational changes in a G-protein-coupled receptor dimer, the leukotriene B(4) receptor BLT1. J Biol Chem 279(48):49664-49670.

Michel CC and Curry FE (1999) Microvascular permeability. Physiol Rev 79(3):703-761.

Miggin SM, Lawler OA and Kinsella BT (2002) Investigation of a functional requirement for isoprenylation by the human prostacyclin receptor. Eur J Biochem 269(6):1714-1725.

Mineo C, James GL, Smart EJ and Anderson RG (1996) Localization of epidermal growth factor-stimulated Ras/Raf-1 interaction to caveolae membrane. J Biol Chem 271(20):11930-11935.

Mineo C, Ying YS, Chapline C, Jaken S and Anderson RG (1998) Targeting of protein kinase Calpha to caveolae. J Cell Biol 141(3):601-610.

Miner LL, Drago J, Chamberlain PM, Donovan D and Uhl GR (1995) Retained cocaine conditioned place preference in D1 receptor deficient mice. Neuroreport 6(17):2314- 2316.

Missale C, Castelletti L, Memo M, Carruba MO and Spano PF (1988) Identification and characterization of postsynaptic D1- and D2-dopamine receptors in the cardiovascular system. J Cardiovasc Pharmacol 11(6):643-650.

Missale C, Nash SR, Robinson SW, Jaber M and Caron MG (1998) Dopamine receptors: from structure to function. Physiol Rev 78(1):189-225.

Miura Y, Hanada K and Jones TL (2001) G(s) signaling is intact after disruption of lipid rafts. Biochemistry 40(50):15418-15423.

Miyazaki I, Asanuma M, Diaz-Corrales FJ, Miyoshi K and Ogawa N (2004) Direct evidence for expression of dopamine receptors in astrocytes from basal ganglia. Brain Res 1029(1):120-123.

Moffett S, Adam L, Bonin H, Loisel TP, Bouvier M and Mouillac B (1996) Palmitoylated cysteine 341 modulates phosphorylation of the beta2-adrenergic receptor by the cAMP- dependent protein kinase. J Biol Chem 271(35):21490-21497.

Moffett S, Brown DA and Linder ME (2000) Lipid-dependent targeting of G proteins into rafts. J Biol Chem 275(3):2191-2198.

249 Moffett S, Mouillac B, Bonin H and Bouvier M (1993) Altered phosphorylation and desensitization patterns of a human beta 2-adrenergic receptor lacking the palmitoylated Cys341. Embo J 12(1):349-356.

Monnot C, Bihoreau C, Conchon S, Curnow KM, Corvol P and Clauser E (1996) Polar residues in the transmembrane domains of the type 1 angiotensin II receptor are required for binding and coupling. Reconstitution of the binding site by co-expression of two deficient mutants. J Biol Chem 271(3):1507-1513.

Morello JP, Salahpour A, Laperriere A, Bernier V, Arthus MF, Lonergan M, Petaja-Repo U, Angers S, Morin D, Bichet DG and Bouvier M (2000) Pharmacological chaperones rescue cell-surface expression and function of misfolded V2 vasopressin receptor mutants. J Clin Invest 105(7):887-895.

Morimoto R, Uehara S, Yatsushiro S, Juge N, Hua Z, Senoh S, Echigo N, Hayashi M, Mizoguchi T, Ninomiya T, Udagawa N, Omote H, Yamamoto A, Edwards RH and Moriyama Y (2006) Secretion of L-glutamate from osteoclasts through transcytosis. Embo J 25(18):4175-4186.

Morris JB, Huynh H, Vasilevski O and Woodcock EA (2006) Alpha1-adrenergic receptor signaling is localized to caveolae in neonatal rat cardiomyocytes. J Mol Cell Cardiol 41(1):17-25.

Mouillac B, Caron M, Bonin H, Dennis M and Bouvier M (1992) Agonist-modulated palmitoylation of beta 2-adrenergic receptor in Sf9 cells. J Biol Chem 267(30):21733- 21737.

Munshi UM, Clouser CL, Peegel H and Menon KM (2005) Evidence that palmitoylation of carboxyl terminus cysteine residues of the human luteinizing hormone receptor regulates postendocytic processing. Mol Endocrinol 19(3):749-758.

Nairn AC, Svenningsson P, Nishi A, Fisone G, Girault JA and Greengard P (2004) The role of DARPP-32 in the actions of drugs of abuse. Neuropharmacology 47 Suppl 1:14-23.

Nash SR, Godinot N and Caron MG (1993) Cloning and characterization of the opossum kidney cell D1 dopamine receptor: expression of identical D1A and D1B dopamine receptor mRNAs in opossum kidney and brain. Mol Pharmacol 44(5):918-925.

Navratil AM, Bliss SP, Berghorn KA, Haughian JM, Farmerie TA, Graham JK, Clay CM and Roberson MS (2003) Constitutive localization of the gonadotropin-releasing hormone (GnRH) receptor to low density membrane microdomains is necessary for GnRH signaling to ERK. J Biol Chem 278(34):31593-31602.

250 Nelson G, Hoon MA, Chandrashekar J, Zhang Y, Ryba NJ and Zuker CS (2001) Mammalian sweet taste receptors. Cell 106(3):381-390.

Neubig RR (1994) Membrane organization in G-protein mechanisms. Faseb J 8(12):939-946.

Ng GY, Mouillac B, George SR, Caron M, Dennis M, Bouvier M and O'Dowd BF (1994) Desensitization, phosphorylation and palmitoylation of the human dopamine D1 receptor. Eur J Pharmacol 267(1):7-19.

Ng GY, Trogadis J, Stevens J, Bouvier M, O'Dowd BF and George SR (1995) Agonist-induced desensitization of dopamine D1 receptor-stimulated adenylyl cyclase activity is temporally and biochemically separated from D1 receptor internalization. Proc Natl Acad Sci U S A 92(22):10157-10161.

Nichols BJ, Kenworthy AK, Polishchuk RS, Lodge R, Roberts TH, Hirschberg K, Phair RD and Lippincott-Schwartz J (2001) Rapid cycling of lipid raft markers between the cell surface and Golgi complex. J Cell Biol 153(3):529-541.

Nichols BJ and Lippincott-Schwartz J (2001) Endocytosis without clathrin coats. Trends Cell Biol 11(10):406-412.

Nimchinsky EA, Hof PR, Janssen WG, Morrison JH and Schmauss C (1997) Expression of dopamine D3 receptor dimers and tetramers in brain and in transfected cells. J Biol Chem 272(46):29229-29237.

Niv H, Gutman O, Kloog Y and Henis YI (2002) Activated K-Ras and H-Ras display different interactions with saturable nonraft sites at the surface of live cells. J Cell Biol 157(5):865-872.

Nohe A, Keating E, Underhill TM, Knaus P and Petersen NO (2005) Dynamics and interaction of caveolin-1 isoforms with BMP-receptors. J Cell Sci 118(Pt 3):643-650.

O'Dowd BF, Hnatowich M, Caron MG, Lefkowitz RJ and Bouvier M (1989) Palmitoylation of the human beta 2-adrenergic receptor. Mutation of Cys341 in the carboxyl tail leads to an uncoupled nonpalmitoylated form of the receptor. J Biol Chem 264(13):7564-7569.

O'Dowd BF, Ji X, Alijaniaram M, Rajaram RD, Kong MM, Rashid A, Nguyen T and George SR (2005) Dopamine receptor oligomerization visualized in living cells. J Biol Chem 280(44):37225-37235.

Oakley RH, Laporte SA, Holt JA, Barak LS and Caron MG (1999) Association of beta-arrestin with G protein-coupled receptors during clathrin-mediated endocytosis dictates the profile of receptor resensitization. J Biol Chem 274(45):32248-32257.

251 Oakley RH, Laporte SA, Holt JA, Barak LS and Caron MG (2001) Molecular determinants underlying the formation of stable intracellular G protein-coupled receptor-beta-arrestin complexes after receptor endocytosis*. J Biol Chem 276(22):19452-19460.

Oakley RH, Laporte SA, Holt JA, Caron MG and Barak LS (2000) Differential affinities of visual arrestin, beta arrestin1, and beta arrestin2 for G protein-coupled receptors delineate two major classes of receptors. J Biol Chem 275(22):17201-17210.

Oh P, McIntosh DP and Schnitzer JE (1998) Dynamin at the neck of caveolae mediates their budding to form transport vesicles by GTP-driven fission from the plasma membrane of endothelium. J Cell Biol 141(1):101-114.

Oh P and Schnitzer JE (2001) Segregation of heterotrimeric G proteins in cell surface microdomains. G(q) binds caveolin to concentrate in caveolae, whereas G(i) and G(s) target lipid rafts by default. Mol Biol Cell 12(3):685-698.

Okada T, Fujiyoshi Y, Silow M, Navarro J, Landau EM and Shichida Y (2002) Functional role of internal water molecules in rhodopsin revealed by X-ray crystallography. Proc Natl Acad Sci U S A 99(9):5982-5987.

Okada T, Sugihara M, Bondar AN, Elstner M, Entel P and Buss V (2004) The retinal conformation and its environment in rhodopsin in light of a new 2.2 A crystal structure. J Mol Biol 342(2):571-583.

Okamoto Y, Ninomiya H, Miwa S and Masaki T (2000) Cholesterol oxidation switches the internalization pathway of endothelin receptor type A from caveolae to clathrin-coated pits in Chinese hamster ovary cells. J Biol Chem 275(9):6439-6446.

Orlandi PA and Fishman PH (1998) Filipin-dependent inhibition of cholera toxin: evidence for toxin internalization and activation through caveolae-like domains. J Cell Biol 141(4):905-915.

Ostrom RS, Gregorian C, Drenan RM, Xiang Y, Regan JW and Insel PA (2001) Receptor number and caveolar co-localization determine receptor coupling efficiency to adenylyl cyclase. J Biol Chem 276(45):42063-42069.

Overton MC and Blumer KJ (2000) G-protein-coupled receptors function as oligomers in vivo. Curr Biol 10(6):341-344.

Overton MC and Blumer KJ (2002) The extracellular N-terminal domain and transmembrane domains 1 and 2 mediate oligomerization of a yeast G protein-coupled receptor. J Biol Chem 277(44):41463-41472.

252 Overton MC, Chinault SL and Blumer KJ (2003) Oligomerization, biogenesis, and signaling is promoted by a glycophorin A-like dimerization motif in transmembrane domain 1 of a yeast G protein-coupled receptor. J Biol Chem 278(49):49369-49377.

Palczewski K, Kumasaka T, Hori T, Behnke CA, Motoshima H, Fox BA, Le Trong I, Teller DC, Okada T, Stenkamp RE, Yamamoto M and Miyano M (2000) Crystal structure of rhodopsin: A G protein-coupled receptor. Science 289(5480):739-745.

Parameswaran N and Spielman WS (2006) RAMPs: The past, present and future. Trends Biochem Sci 31(11):631-638.

Parton RG, Joggerst B and Simons K (1994) Regulated internalization of caveolae. J Cell Biol 127(5):1199-1215.

Parton RG and Simons K (2007) The multiple faces of caveolae. Nat Rev Mol Cell Biol 8(3):185-194.

Pelkmans L, Kartenbeck J and Helenius A (2001) Caveolar endocytosis of simian virus 40 reveals a new two-step vesicular-transport pathway to the ER. Nat Cell Biol 3(5):473- 483.

Percherancier Y, Planchenault T, Valenzuela-Fernandez A, Virelizier JL, Arenzana-Seisdedos F and Bachelerie F (2001) Palmitoylation-dependent control of degradation, life span, and membrane expression of the CCR5 receptor. J Biol Chem 276(34):31936-31944.

Perez DM, Hwa J, Gaivin R, Mathur M, Brown F and Graham RM (1996) Constitutive activation of a single effector pathway: evidence for multiple activation states of a G protein-coupled receptor. Mol Pharmacol 49(1):112-122.

Petaja-Repo UE, Hogue M, Bhalla S, Laperriere A, Morello JP and Bouvier M (2002) Ligands act as pharmacological chaperones and increase the efficiency of delta opioid receptor maturation. Embo J 21(7):1628-1637.

Petaja-Repo UE, Hogue M, Leskela TT, Markkanen PM, Tuusa JT and Bouvier M (2006) Distinct subcellular localization for constitutive and agonist-modulated palmitoylation of the human delta opioid receptor. J Biol Chem 281(23):15780-15789.

Pierce KL, Premont RT and Lefkowitz RJ (2002) Seven-transmembrane receptors. Nat Rev Mol Cell Biol 3(9):639-650.

Pike LJ and Casey L (1996) Localization and turnover of phosphatidylinositol 4,5-bisphosphate in caveolin-enriched membrane domains. J Biol Chem 271(43):26453-26456.

253 Pike LJ and Miller JM (1998) Cholesterol depletion delocalizes phosphatidylinositol bisphosphate and inhibits hormone-stimulated phosphatidylinositol turnover. J Biol Chem 273(35):22298-22304.

Pin JP, Kniazeff J, Binet V, Liu J, Maurel D, Galvez T, Duthey B, Havlickova M, Blahos J, Prezeau L and Rondard P (2004) Activation mechanism of the heterodimeric GABA(B) receptor. Biochem Pharmacol 68(8):1565-1572.

Pollock NJ, Manelli AM, Hutchins CW, Steffey ME, MacKenzie RG and Frail DE (1992) Serine mutations in transmembrane V of the dopamine D1 receptor affect ligand interactions and receptor activation. J Biol Chem 267(25):17780-17786.

Ponimaskin E, Dumuis A, Gaven F, Barthet G, Heine M, Glebov K, Richter DW and Oppermann M (2005) Palmitoylation of the 5-hydroxytryptamine4a receptor regulates receptor phosphorylation, desensitization, and beta-arrestin-mediated endocytosis. Mol Pharmacol 67(5):1434-1443.

Ponimaskin EG, Heine M, Joubert L, Sebben M, Bickmeyer U, Richter DW and Dumuis A (2002) The 5-hydroxytryptamine(4a) receptor is palmitoylated at two different sites, and acylation is critically involved in regulation of receptor constitutive activity. J Biol Chem 277(4):2534-2546.

Ponimaskin EG, Schmidt MF, Heine M, Bickmeyer U and Richter DW (2001) 5- Hydroxytryptamine 4(a) receptor expressed in Sf9 cells is palmitoylated in an agonist- dependent manner. Biochem J 353(Pt 3):627-634.

Popot JL, Gerchman SE and Engelman DM (1987) Refolding of bacteriorhodopsin in lipid bilayers. A thermodynamically controlled two-stage process. J Mol Biol 198(4):655-676.

Porter JE, Hwa J and Perez DM (1996) Activation of the alpha1b-adrenergic receptor is initiated by disruption of an interhelical salt bridge constraint. J Biol Chem 271(45):28318-28323.

Predescu D, Horvat R, Predescu S and Palade GE (1994) Transcytosis in the continuous endothelium of the myocardial microvasculature is inhibited by N-ethylmaleimide. Proc Natl Acad Sci U S A 91(8):3014-3018.

Predescu D, Predescu S, McQuistan T and Palade GE (1998) Transcytosis of alpha1-acidic glycoprotein in the continuous microvascular endothelium. Proc Natl Acad Sci U S A 95(11):6175-6180.

Predescu D, Simionescu M, Simionescu N and Palade GE (1988) Binding and transcytosis of glycoalbumin by the microvascular endothelium of the murine myocardium: evidence that glycoalbumin behaves as a bifunctional ligand. J Cell Biol 107(5):1729-1738.

254 Predescu SA, Predescu DN and Palade GE (1997) Plasmalemmal vesicles function as transcytotic carriers for small proteins in the continuous endothelium. Am J Physiol 272(2 Pt 2):H937-949.

Prior IA, Harding A, Yan J, Sluimer J, Parton RG and Hancock JF (2001) GTP-dependent segregation of H-ras from lipid rafts is required for biological activity. Nat Cell Biol 3(4):368-375.

Prior IA, Muncke C, Parton RG and Hancock JF (2003) Direct visualization of Ras proteins in spatially distinct cell surface microdomains. J Cell Biol 160(2):165-170.

Pruitt DL, Bolanos CA and McDougall SA (1995) Effects of dopamine D1 and D2 receptor antagonists on cocaine-induced place preference conditioning in preweanling rats. Eur J Pharmacol 283(1-3):125-131.

Puri V, Watanabe R, Singh RD, Dominguez M, Brown JC, Wheatley CL, Marks DL and Pagano RE (2001) Clathrin-dependent and -independent internalization of plasma membrane sphingolipids initiates two Golgi targeting pathways. J Cell Biol 154(3):535-547.

Qanbar R and Bouvier M (2003) Role of palmitoylation/depalmitoylation reactions in G-protein- coupled receptor function. Pharmacol Ther 97(1):1-33.

Rajendran L and Simons K (2005) Lipid rafts and membrane dynamics. J Cell Sci 118(Pt 6):1099-1102.

Ramsay D, Kellett E, McVey M, Rees S and Milligan G (2002) Homo- and hetero-oligomeric interactions between G-protein-coupled receptors in living cells monitored by two variants of bioluminescence resonance energy transfer (BRET): hetero-oligomers between receptor subtypes form more efficiently than between less closely related sequences. Biochem J 365(Pt 2):429-440.

Rankin ML, Marinec PS, Cabrera DM, Wang Z, Jose PA and Sibley DR (2006) The D1 dopamine receptor is constitutively phosphorylated by G protein-coupled receptor kinase 4. Mol Pharmacol 69(3):759-769.

Rapacciuolo A, Suvarna S, Barki-Harrington L, Luttrell LM, Cong M, Lefkowitz RJ and Rockman HA (2003) Protein kinase A and G protein-coupled receptor kinase phosphorylation mediates beta-1 adrenergic receptor endocytosis through different pathways. J Biol Chem 278(37):35403-35411.

Rashid AJ, So CH, Kong MM, Furtak T, El-Ghundi M, Cheng R, O'Dowd BF and George SR (2007) D1-D2 dopamine receptor heterooligomers with unique pharmacology are

255 coupled to rapid activation of Gq/11 in the striatum. Proc Natl Acad Sci U S A 104(2):654-659.

Razani B, Engelman JA, Wang XB, Schubert W, Zhang XL, Marks CB, Macaluso F, Russell RG, Li M, Pestell RG, Di Vizio D, Hou H, Jr., Kneitz B, Lagaud G, Christ GJ, Edelmann W and Lisanti MP (2001) Caveolin-1 null mice are viable but show evidence of hyperproliferative and vascular abnormalities. J Biol Chem 276(41):38121-38138.

Razani B, Rubin CS and Lisanti MP (1999) Regulation of cAMP-mediated signal transduction via interaction of caveolins with the catalytic subunit of protein kinase A. J Biol Chem 274(37):26353-26360.

Rebois RV and Hebert TE (2003) Protein complexes involved in heptahelical receptor-mediated signal transduction. Receptors Channels 9(3):169-194.

Reuss B and Unsicker K (2001) Atypical neuroleptic drugs downregulate dopamine sensitivity in rat cortical and striatal astrocytes. Mol Cell Neurosci 18(2):197-209.

Ridge KD, Lee SS and Abdulaev NG (1996) Examining rhodopsin folding and assembly through expression of polypeptide fragments. J Biol Chem 271(13):7860-7867.

Ridge KD, Lee SS and Yao LL (1995) In vivo assembly of rhodopsin from expressed polypeptide fragments. Proc Natl Acad Sci U S A 92(8):3204-3208.

Ringerike T, Blystad FD, Levy FO, Madshus IH and Stang E (2002) Cholesterol is important in control of EGF receptor kinase activity but EGF receptors are not concentrated in caveolae. J Cell Sci 115(Pt 6):1331-1340.

Robbins MJ, Calver AR, Filippov AK, Hirst WD, Russell RB, Wood MD, Nasir S, Couve A, Brown DA, Moss SJ and Pangalos MN (2001) GABA(B2) is essential for g-protein coupling of the GABA(B) receptor heterodimer. J Neurosci 21(20):8043-8052.

Robinson MS (1994) The role of clathrin, adaptors and dynamin in endocytosis. Curr Opin Cell Biol 6(4):538-544.

Robinson PR, Cohen GB, Zhukovsky EA and Oprian DD (1992) Constitutively active mutants of rhodopsin. Neuron 9(4):719-725.

Rocheville M, Lange DC, Kumar U, Patel SC, Patel RC and Patel YC (2000a) Receptors for dopamine and somatostatin: formation of hetero-oligomers with enhanced functional activity. Science 288(5463):154-157.

256 Rocheville M, Lange DC, Kumar U, Sasi R, Patel RC and Patel YC (2000b) Subtypes of the somatostatin receptor assemble as functional homo- and heterodimers. J Biol Chem 275(11):7862-7869.

Roettger BF, Rentsch RU, Pinon D, Holicky E, Hadac E, Larkin JM and Miller LJ (1995) Dual pathways of internalization of the cholecystokinin receptor. J Cell Biol 128(6):1029- 1041.

Romano C, Yang WL and O'Malley KL (1996) Metabotropic glutamate receptor 5 is a disulfide- linked dimer. J Biol Chem 271(45):28612-28616.

Roseberry AG and Hosey MM (2001) Internalization of the M2 muscarinic acetylcholine receptor proceeds through an atypical pathway in HEK293 cells that is independent of clathrin and caveolae. J Cell Sci 114(Pt 4):739-746.

Rovati GE, Capra V and Neubig RR (2006) The Highly Conserved DRY Motif of Class A GPCRs: Beyond the Ground State. Mol Pharmacol.

Roy S, Luetterforst R, Harding A, Apolloni A, Etheridge M, Stang E, Rolls B, Hancock JF and Parton RG (1999) Dominant-negative caveolin inhibits H-Ras function by disrupting cholesterol-rich plasma membrane domains. Nat Cell Biol 1(2):98-105.

Ruprecht JJ, Mielke T, Vogel R, Villa C and Schertler GF (2004) Electron crystallography reveals the structure of metarhodopsin I. Embo J 23(18):3609-3620.

Rybin VO, Xu X, Lisanti MP and Steinberg SF (2000) Differential targeting of beta -adrenergic receptor subtypes and adenylyl cyclase to cardiomyocyte caveolae. A mechanism to functionally regulate the cAMP signaling pathway. J Biol Chem 275(52):41447-41457.

Ryman-Rasmussen JP, Nichols DE and Mailman RB (2005) Differential activation of adenylate cyclase and receptor internalization by novel dopamine D1 receptor agonists. Mol Pharmacol 68(4):1039-1048.

Sabatini DD, Kreibich G, Morimoto T and Adesnik M (1982) Mechanisms for the incorporation of proteins in membranes and organelles. J Cell Biol 92(1):1-22.

Sabourin T, Bastien L, Bachvarov DR and Marceau F (2002) Agonist-induced translocation of the kinin B(1) receptor to caveolae-related rafts. Mol Pharmacol 61(3):546-553.

Sadeghi HM, Innamorati G, Dagarag M and Birnbaumer M (1997) Palmitoylation of the V2 vasopressin receptor. Mol Pharmacol 52(1):21-29.

257 Salahpour A, Angers S, Mercier JF, Lagace M, Marullo S and Bouvier M (2004) Homodimerization of the beta2-adrenergic receptor as a prerequisite for cell surface targeting. J Biol Chem 279(32):33390-33397.

Salim K, Fenton T, Bacha J, Urien-Rodriguez H, Bonnert T, Skynner HA, Watts E, Kerby J, Heald A, Beer M, McAllister G and Guest PC (2002) Oligomerization of G-protein- coupled receptors shown by selective co-immunoprecipitation. J Biol Chem 277(18):15482-15485.

Salom D, Le Trong I, Pohl E, Ballesteros JA, Stenkamp RE, Palczewski K and Lodowski DT (2006a) Improvements in G protein-coupled receptor purification yield light stable rhodopsin crystals. J Struct Biol 156(3):497-504.

Salom D, Lodowski DT, Stenkamp RE, Le Trong I, Golczak M, Jastrzebska B, Harris T, Ballesteros JA and Palczewski K (2006b) Crystal structure of a photoactivated deprotonated intermediate of rhodopsin. Proc Natl Acad Sci U S A 103(44):16123-16128.

Satoh T, Cohen HT and Katz AI (1993) Different mechanisms of renal Na-K-ATPase regulation by protein kinases in proximal and distal nephron. Am J Physiol 265(3 Pt 2):F399-405.

Savi P, Zachayus JL, Delesque-Touchard N, Labouret C, Herve C, Uzabiaga MF, Pereillo JM, Culouscou JM, Bono F, Ferrara P and Herbert JM (2006) The active metabolite of Clopidogrel disrupts P2Y12 receptor oligomers and partitions them out of lipid rafts. Proc Natl Acad Sci U S A 103(29):11069-11074.

Scarselli M, Novi F, Schallmach E, Lin R, Baragli A, Colzi A, Griffon N, Corsini GU, Sokoloff P, Levenson R, Vogel Z and Maggio R (2001) D2/D3 dopamine receptor heterodimers exhibit unique functional properties. J Biol Chem 276(32):30308-30314.

Schneuwly S, Shortridge RD, Larrivee DC, Ono T, Ozaki M and Pak WL (1989) Drosophila ninaA gene encodes an eye-specific cyclophilin (cyclosporine A binding protein). Proc Natl Acad Sci U S A 86(14):5390-5394.

Schnitzer JE, Oh P and McIntosh DP (1996) Role of GTP hydrolysis in fission of caveolae directly from plasma membranes. Science 274(5285):239-242.

Schnitzer JE, Oh P, Pinney E and Allard J (1994) Filipin-sensitive caveolae-mediated transport in endothelium: reduced transcytosis, scavenger endocytosis, and capillary permeability of select macromolecules. J Cell Biol 127(5):1217-1232.

Schoneberg T, Liu J and Wess J (1995) Plasma membrane localization and functional rescue of truncated forms of a G protein-coupled receptor. J Biol Chem 270(30):18000-18006.

258 Schubert W, Frank PG, Razani B, Park DS, Chow CW and Lisanti MP (2001) Caveolae- deficient endothelial cells show defects in the uptake and transport of albumin in vivo. J Biol Chem 276(52):48619-48622.

Schulein R, Liebenhoff U, Muller H, Birnbaumer M and Rosenthal W (1996) Properties of the human arginine vasopressin V2 receptor after site-directed mutagenesis of its putative palmitoylation site. Biochem J 313 (Pt 2):611-616.

Schulz A, Grosse R, Schultz G, Gudermann T and Schoneberg T (2000) Structural implication for receptor oligomerization from functional reconstitution studies of mutant V2 vasopressin receptors. J Biol Chem 275(4):2381-2389.

Schwencke C, Yamamoto M, Okumura S, Toya Y, Kim SJ and Ishikawa Y (1999) Compartmentation of cyclic adenosine 3',5'-monophosphate signaling in caveolae. Mol Endocrinol 13(7):1061-1070.

Sealfon SC and Olanow CW (2000) Dopamine receptors: from structure to behavior. Trends Neurosci 23(10 Suppl):S34-40.

Seeman P and Van Tol HH (1994) Dopamine receptor pharmacology. Trends Pharmacol Sci 15(7):264-270.

Shaul PW, Smart EJ, Robinson LJ, German Z, Yuhanna IS, Ying Y, Anderson RG and Michel T (1996) Acylation targets emdothelial nitric-oxide synthase to plasmalemmal caveolae. J Biol Chem 271(11):6518-6522.

Shenoy-Scaria AM, Dietzen DJ, Kwong J, Link DC and Lublin DM (1994) Cysteine3 of Src family protein tyrosine kinase determines palmitoylation and localization in caveolae. J Cell Biol 126(2):353-363.

Shenoy SK, McDonald PH, Kohout TA and Lefkowitz RJ (2001) Regulation of receptor fate by ubiquitination of activated beta 2-adrenergic receptor and beta-arrestin. Science 294(5545):1307-1313.

Shieh BH, Stamnes MA, Seavello S, Harris GL and Zuker CS (1989) The ninaA gene required for visual transduction in Drosophila encodes a homologue of cyclosporin A-binding protein. Nature 338(6210):67-70.

Shin N, Coates E, Murgolo NJ, Morse KL, Bayne M, Strader CD and Monsma FJ, Jr. (2002) Molecular modeling and site-specific mutagenesis of the histamine-binding site of the histamine H4 receptor. Mol Pharmacol 62(1):38-47.

259 Silva WI, Maldonado HM, Lisanti MP, Devellis J, Chompre G, Mayol N, Ortiz M, Velazquez G, Maldonado A and Montalvo J (1999) Identification of caveolae and caveolin in C6 glioma cells. Int J Dev Neurosci 17(7):705-714.

Silva WI, Maldonado HM, Velazquez G, Rubio-Davila M, Miranda JD, Aquino E, Mayol N, Cruz-Torres A, Jardon J and Salgado-Villanueva IK (2005) Caveolin isoform expression during differentiation of C6 glioma cells. Int J Dev Neurosci 23(7):599-612.

Singer SJ, Maher PA and Yaffe MP (1987) On the transfer of integral proteins into membranes. Proc Natl Acad Sci U S A 84(7):1960-1964.

Singer SJ and Nicolson GL (1972) The fluid mosaic model of the structure of cell membranes. Science 175(23):720-731.

Smart EJ, Graf GA, McNiven MA, Sessa WC, Engelman JA, Scherer PE, Okamoto T and Lisanti MP (1999) Caveolins, liquid-ordered domains, and signal transduction. Mol Cell Biol 19(11):7289-7304.

Smart EJ, Ying Y, Donzell WC and Anderson RG (1996) A role for caveolin in transport of cholesterol from endoplasmic reticulum to plasma membrane. J Biol Chem 271(46):29427-29435.

Smart EJ, Ying YS, Conrad PA and Anderson RG (1994) Caveolin moves from caveolae to the Golgi apparatus in response to cholesterol oxidation. J Cell Biol 127(5):1185-1197.

Smiley JF, Levey AI, Ciliax BJ and Goldman-Rakic PS (1994) D1 dopamine receptor immunoreactivity in human and monkey cerebral cortex: predominant and extrasynaptic localization in dendritic spines. Proc Natl Acad Sci U S A 91(12):5720-5724.

Song KS, Li S, Okamoto T, Quilliam LA, Sargiacomo M and Lisanti MP (1996) Co-purification and direct interaction of Ras with caveolin, an integral membrane protein of caveolae microdomains. Detergent-free purification of caveolae microdomains. J Biol Chem 271(16):9690-9697.

Sotgia F, Lee JK, Das K, Bedford M, Petrucci TC, Macioce P, Sargiacomo M, Bricarelli FD, Minetti C, Sudol M and Lisanti MP (2000) Caveolin-3 directly interacts with the C- terminal tail of beta -dystroglycan. Identification of a central WW-like domain within caveolin family members. J Biol Chem 275(48):38048-38058.

Stenkamp RE, Teller DC and Palczewski K (2005) Rhodopsin: a structural primer for G-protein coupled receptors. Arch Pharm (Weinheim) 338(5-6):209-216.

260 Strader CD, Candelore MR, Hill WS, Sigal IS and Dixon RA (1989a) Identification of two serine residues involved in agonist activation of the beta-adrenergic receptor. J Biol Chem 264(23):13572-13578.

Strader CD, Fong TM, Tota MR, Underwood D and Dixon RA (1994) Structure and function of G protein-coupled receptors. Annu Rev Biochem 63:101-132.

Strader CD, Sigal IS, Candelore MR, Rands E, Hill WS and Dixon RA (1988) Conserved aspartic acid residues 79 and 113 of the beta-adrenergic receptor have different roles in receptor function. J Biol Chem 263(21):10267-10271.

Strader CD, Sigal IS and Dixon RA (1989b) Genetic approaches to the determination of structure-function relationships of G protein-coupled receptors. Trends Pharmacol Sci Suppl:26-30.

Sunahara RK, Guan HC, O'Dowd BF, Seeman P, Laurier LG, Ng G, George SR, Torchia J, Van Tol HH and Niznik HB (1991) Cloning of the gene for a human dopamine D5 receptor with higher affinity for dopamine than D1. Nature 350(6319):614-619.

Syme CA, Zhang L and Bisello A (2006) Caveolin-1 regulates cellular trafficking and function of the glucagon-like Peptide 1 receptor. Mol Endocrinol 20(12):3400-3411.

Teller DC, Okada T, Behnke CA, Palczewski K and Stenkamp RE (2001) Advances in determination of a high-resolution three-dimensional structure of rhodopsin, a model of G-protein-coupled receptors (GPCRs). Biochemistry 40(26):7761-7772.

Terrell J, Shih S, Dunn R and Hicke L (1998) A function for monoubiquitination in the internalization of a G protein-coupled receptor. Mol Cell 1(2):193-202.

Terrillon S, Durroux T, Mouillac B, Breit A, Ayoub MA, Taulan M, Jockers R, Barberis C and Bouvier M (2003) Oxytocin and Vasopressin V1a and V2 Receptors Form Constitutive Homo- and Heterodimers during Biosynthesis. Mol Endocrinol 17(4):677-691.

Thomsen P, Roepstorff K, Stahlhut M and van Deurs B (2002) Caveolae are highly immobile plasma membrane microdomains, which are not involved in constitutive endocytic trafficking. Mol Biol Cell 13(1):238-250.

Tiberi M, Jarvie KR, Silvia C, Falardeau P, Gingrich JA, Godinot N, Bertrand L, Yang-Feng TL, Fremeau RT, Jr. and 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(17):7491-7495.

261 Tiruppathi C, Song W, Bergenfeldt M, Sass P and Malik AB (1997) Gp60 activation mediates albumin transcytosis in endothelial cells by tyrosine kinase-dependent pathway. J Biol Chem 272(41):25968-25975.

Toki S, Donati RJ and Rasenick MM (1999) Treatment of C6 glioma cells and rats with antidepressant drugs increases the detergent extraction of G(s alpha) from plasma membrane. J Neurochem 73(3):1114-1120.

Toya Y, Schwencke C, Couet J, Lisanti MP and Ishikawa Y (1998) Inhibition of adenylyl cyclase by caveolin peptides. Endocrinology 139(4):2025-2031.

Trejo J and Coughlin SR (1999) The cytoplasmic tails of protease-activated receptor-1 and substance P receptor specify sorting to lysosomes versus recycling. J Biol Chem 274(4):2216-2224.

Trivedi M, Narkar VA, Hussain T and Lokhandwala MF (2004) Dopamine recruits D1A receptors to Na-K-ATPase-rich caveolar plasma membranes in rat renal proximal tubules. Am J Physiol Renal Physiol 287(5):F921-931.

Tsuji Y, Shimada Y, Takeshita T, Kajimura N, Nomura S, Sekiyama N, Otomo J, Usukura J, Nakanishi S and Jingami H (2000) Cryptic dimer interface and domain organization of the extracellular region of metabotropic glutamate receptor subtype 1. J Biol Chem 275(36):28144-28151.

Tumova K, Iwasiow RM and Tiberi M (2003) Insight into the mechanism of dopamine D1-like receptor activation. Evidence for a molecular interplay between the third extracellular loop and the cytoplasmic tail. J Biol Chem 278(10):8146-8153.

Tumova K, Zhang D and Tiberi M (2004) Role of the fourth intracellular loop of D1-like dopaminergic receptors in conferring subtype-specific signaling properties. FEBS Lett 576(3):461-467.

Tuomisto J and Mannisto P (1985) Neurotransmitter regulation of anterior pituitary hormones. Pharmacol Rev 37(3):249-332.

Uberti MA, Hague C, Oller H, Minneman KP and Hall RA (2005) Heterodimerization with beta2-adrenergic receptors promotes surface expression and functional activity of alpha1D-adrenergic receptors. J Pharmacol Exp Ther 313(1):16-23.

Uberti MA, Hall RA and Minneman KP (2003) Subtype-specific dimerization of alpha 1- adrenoceptors: effects on receptor expression and pharmacological properties. Mol Pharmacol 64(6):1379-1390.

262 Uittenbogaard A, Ying Y and Smart EJ (1998) Characterization of a cytosolic heat-shock protein-caveolin chaperone complex. Involvement in cholesterol trafficking. J Biol Chem 273(11):6525-6532.

Ullrich A and Schlessinger J (1990) Signal transduction by receptors with tyrosine kinase activity. Cell 61(2):203-212.

Undie AS, Weinstock J, Sarau HM and Friedman E (1994) Evidence for a distinct D1-like dopamine receptor that couples to activation of phosphoinositide metabolism in brain. J Neurochem 62(5):2045-2048.

Unger VM, Hargrave PA, Baldwin JM and Schertler GF (1997) Arrangement of rhodopsin transmembrane alpha-helices. Nature 389(6647):203-206.

van Duyl BY, Rijkers DT, de Kruijff B and Killian JA (2002) Influence of hydrophobic mismatch and palmitoylation on the association of transmembrane alpha-helical peptides with detergent-resistant membranes. FEBS Lett 523(1-3):79-84.

Venema VJ, Ju H, Zou R and Venema RC (1997) Interaction of neuronal nitric-oxide synthase with caveolin-3 in skeletal muscle. Identification of a novel caveolin scaffolding/inhibitory domain. J Biol Chem 272(45):28187-28190.

Venter JC, Adams MD, Myers EW, Li PW, Mural RJ, Sutton GG, Smith HO, Yandell M, Evans CA, Holt RA, Gocayne JD, Amanatides P, Ballew RM, Huson DH, Wortman JR, Zhang Q, Kodira CD, Zheng XH, Chen L, Skupski M, Subramanian G, Thomas PD, Zhang J, Gabor Miklos GL, Nelson C, Broder S, Clark AG, Nadeau J, McKusick VA, Zinder N, Levine AJ, Roberts RJ, Simon M, Slayman C, Hunkapiller M, Bolanos R, Delcher A, Dew I, Fasulo D, Flanigan M, Florea L, Halpern A, Hannenhalli S, Kravitz S, Levy S, Mobarry C, Reinert K, Remington K, Abu-Threideh J, Beasley E, Biddick K, Bonazzi V, Brandon R, Cargill M, Chandramouliswaran I, Charlab R, Chaturvedi K, Deng Z, Di Francesco V, Dunn P, Eilbeck K, Evangelista C, Gabrielian AE, Gan W, Ge W, Gong F, Gu Z, Guan P, Heiman TJ, Higgins ME, Ji RR, Ke Z, Ketchum KA, Lai Z, Lei Y, Li Z, Li J, Liang Y, Lin X, Lu F, Merkulov GV, Milshina N, Moore HM, Naik AK, Narayan VA, Neelam B, Nusskern D, Rusch DB, Salzberg S, Shao W, Shue B, Sun J, Wang Z, Wang A, Wang X, Wang J, Wei M, Wides R, Xiao C, Yan C, Yao A, Ye J, Zhan M, Zhang W, Zhang H, Zhao Q, Zheng L, Zhong F, Zhong W, Zhu S, Zhao S, Gilbert D, Baumhueter S, Spier G, Carter C, Cravchik A, Woodage T, Ali F, An H, Awe A, Baldwin D, Baden H, Barnstead M, Barrow I, Beeson K, Busam D, Carver A, Center A, Cheng ML, Curry L, Danaher S, Davenport L, Desilets R, Dietz S, Dodson K, Doup L, Ferriera S, Garg N, Gluecksmann A, Hart B, Haynes J, Haynes C, Heiner C, Hladun S, Hostin D, Houck J, Howland T, Ibegwam C, Johnson J, Kalush F, Kline L, Koduru S, Love A, Mann F, May D, McCawley S, McIntosh T, McMullen I, Moy M, Moy L, Murphy B, Nelson K, Pfannkoch C, Pratts E, Puri V, Qureshi H, Reardon M, Rodriguez R, Rogers YH, Romblad D, Ruhfel B, Scott R, Sitter C, Smallwood M, Stewart E, Strong R, Suh E, Thomas R, Tint NN, Tse S, Vech C, Wang G, Wetter J, Williams S, Williams

263 M, Windsor S, Winn-Deen E, Wolfe K, Zaveri J, Zaveri K, Abril JF, Guigo R, Campbell MJ, Sjolander KV, Karlak B, Kejariwal A, Mi H, Lazareva B, Hatton T, Narechania A, Diemer K, Muruganujan A, Guo N, Sato S, Bafna V, Istrail S, Lippert R, Schwartz R, Walenz B, Yooseph S, Allen D, Basu A, Baxendale J, Blick L, Caminha M, Carnes-Stine J, Caulk P, Chiang YH, Coyne M, Dahlke C, Mays A, Dombroski M, Donnelly M, Ely D, Esparham S, Fosler C, Gire H, Glanowski S, Glasser K, Glodek A, Gorokhov M, Graham K, Gropman B, Harris M, Heil J, Henderson S, Hoover J, Jennings D, Jordan C, Jordan J, Kasha J, Kagan L, Kraft C, Levitsky A, Lewis M, Liu X, Lopez J, Ma D, Majoros W, McDaniel J, Murphy S, Newman M, Nguyen T, Nguyen N, Nodell M, Pan S, Peck J, Peterson M, Rowe W, Sanders R, Scott J, Simpson M, Smith T, Sprague A, Stockwell T, Turner R, Venter E, Wang M, Wen M, Wu D, Wu M, Xia A, Zandieh A and Zhu X (2001) The sequence of the human genome. Science 291(5507):1304-1351.

Vereb G, Matko J, Vamosi G, Ibrahim SM, Magyar E, Varga S, Szollosi J, Jenei A, Gaspar R, Jr., Waldmann TA and Damjanovich S (2000) Cholesterol-dependent clustering of IL- 2Ralpha and its colocalization with HLA and CD48 on T lymphoma cells suggest their functional association with lipid rafts. Proc Natl Acad Sci U S A 97(11):6013-6018.

Vickery RG and von Zastrow M (1999) Distinct dynamin-dependent and -independent mechanisms target structurally homologous dopamine receptors to different endocytic membranes. J Cell Biol 144(1):31-43.

Volonte D, Galbiati F, Pestell RG and Lisanti MP (2001) Cellular stress induces the tyrosine phosphorylation of caveolin-1 (Tyr(14)) via activation of p38 mitogen-activated protein kinase and c-Src kinase. Evidence for caveolae, the actin cytoskeleton, and focal adhesions as mechanical sensors of osmotic stress. J Biol Chem 276(11):8094-8103.

Vrecl M, Anderson L, Hanyaloglu A, McGregor AM, Groarke AD, Milligan G, Taylor PL and Eidne KA (1998) Agonist-induced endocytosis and recycling of the gonadotropin- releasing hormone receptor: effect of beta-arrestin on internalization kinetics. Mol Endocrinol 12(12):1818-1829.

Waldo GL, Northup JK, Perkins JP and Harden TK (1983) Characterization of an altered membrane form of the beta-adrenergic receptor produced during agonist-induced desensitization. J Biol Chem 258(22):13900-13908.

Wang H, Haas M, Liang M, Cai T, Tian J, Li S and Xie Z (2004) Ouabain assembles signaling cascades through the caveolar Na+/K+-ATPase. J Biol Chem 279(17):17250-17259.

Wang HY, Undie AS and 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(6):988-994.

264 Wang Y, Zhou Y, Szabo K, Haft CR and Trejo J (2002) Down-regulation of protease-activated receptor-1 is regulated by sorting nexin 1. Mol Biol Cell 13(6):1965-1976.

Weiner DM, Levey AI, Sunahara RK, Niznik HB, O'Dowd BF, Seeman P and Brann MR (1991) D1 and D2 dopamine receptor mRNA in rat brain. Proc Natl Acad Sci U S A 88(5):1859- 1863.

Wessels HP and Spiess M (1988) Insertion of a multispanning membrane protein occurs sequentially and requires only one signal sequence. Cell 55(1):61-70.

Whistler JL, Enquist J, Marley A, Fong J, Gladher F, Tsuruda P, Murray SR and Von Zastrow M (2002) Modulation of postendocytic sorting of G protein-coupled receptors. Science 297(5581):615-620.

White JH, Wise A, Main MJ, Green A, Fraser NJ, Disney GH, Barnes AA, Emson P, Foord SM and Marshall FH (1998) Heterodimerization is required for the formation of a functional GABA(B) receptor. Nature 396(6712):679-682.

Williams TM and Lisanti MP (2004) The Caveolin genes: from cell biology to medicine. Ann Med 36(8):584-595.

Williams TM and Lisanti MP (2005) Caveolin-1 in oncogenic transformation, cancer, and metastasis. Am J Physiol Cell Physiol 288(3):C494-506.

Wilson MH, Highfield HA and Limbird LE (2001) The role of a conserved inter-transmembrane domain interface in regulating alpha(2a)-adrenergic receptor conformational stability and cell-surface turnover. Mol Pharmacol 59(4):929-938.

Wise A, Gearing K and Rees S (2002) Target validation of G-protein coupled receptors. Drug Discov Today 7(4):235-246.

Wreggett KA and Wells JW (1995) Cooperativity manifest in the binding properties of purified cardiac muscarinic receptors. J Biol Chem 270(38):22488-22499.

Wyse BD, Prior IA, Qian H, Morrow IC, Nixon S, Muncke C, Kurzchalia TV, Thomas WG, Parton RG and Hancock JF (2003) Caveolin interacts with the angiotensin II type 1 receptor during exocytic transport but not at the plasma membrane. J Biol Chem 278(26):23738-23746.

Xiang Y, Rybin VO, Steinberg SF and Kobilka B (2002) Caveolar localization dictates physiologic signaling of beta 2-adrenoceptors in neonatal cardiac myocytes. J Biol Chem 277(37):34280-34286.

265 Xie Z, Lee SP, O'Dowd BF and George SR (1999) Serotonin 5-HT1B and 5-HT1D receptors form homodimers when expressed alone and heterodimers when co-expressed. FEBS Lett 456(1):63-67.

Xu M, Hu XT, Cooper DC, Moratalla R, Graybiel AM, White FJ and Tonegawa S (1994) Elimination of cocaine-induced hyperactivity and dopamine-mediated neurophysiological effects in dopamine D1 receptor mutant mice. Cell 79(6):945-955.

Xu W, Yoon SI, Huang P, Wang Y, Chen C, Chong PL and Liu-Chen LY (2006) Localization of the kappa opioid receptor in lipid rafts. J Pharmacol Exp Ther 317(3):1295-1306.

Yamabhai M and Anderson RG (2002) Second cysteine-rich region of epidermal growth factor receptor contains targeting information for caveolae/rafts. J Biol Chem 277(28):24843- 24846.

Yamaguchi I, Jose PA, Mouradian MM, Canessa LM, Monsma FJ, Jr., Sibley DR, Takeyasu K and Felder RA (1993) Expression of dopamine D1A receptor gene in proximal tubule of rat kidneys. Am J Physiol 264(2 Pt 2):F280-285.

Yamaguchi I, Yao L, Sanada H, Ozono R, Mouradian MM, Jose PA, Carey RM and Felder RA (1997) Dopamine D1A receptors and renin release in rat juxtaglomerular cells. Hypertension 29(4):962-968.

Yamaguchi T, Murata Y, Fujiyoshi Y and Doi T (2003) Regulated interaction of endothelin B receptor with caveolin-1. Eur J Biochem 270(8):1816-1827.

Yamamoto M, Toya Y, Schwencke C, Lisanti MP, Myers MG, Jr. and Ishikawa Y (1998) Caveolin is an activator of insulin receptor signaling. J Biol Chem 273(41):26962-26968.

Youdim MB, Edmondson D and Tipton KF (2006) The therapeutic potential of monoamine oxidase inhibitors. Nat Rev Neurosci 7(4):295-309.

Yu P, Yang Z, Jones JE, Wang Z, Owens SA, Mueller SC, Felder RA and Jose PA (2004) D1 dopamine receptor signaling involves caveolin-2 in HEK-293 cells. Kidney Int 66(6):2167-2180.

Yuan H, Michelsen K and Schwappach B (2003) 14-3-3 dimers probe the assembly status of multimeric membrane proteins. Curr Biol 13(8):638-646.

Zacharias DA, Violin JD, Newton AC and Tsien RY (2002) Partitioning of lipid-modified monomeric GFPs into membrane microdomains of live cells. Science 296(5569):913- 916.

266 Zanassi P, Paolillo M, Montecucco A, Avvedimento EV and Schinelli S (1999) Pharmacological and molecular evidence for dopamine D(1) receptor expression by striatal astrocytes in culture. J Neurosci Res 58(4):544-552.

Zawarynski P, Tallerico T, Seeman P, Lee SP, O'Dowd BF and George SR (1998) Dopamine D2 receptor dimers in human and rat brain. FEBS Lett 441(3):383-386.

Zeng FY, Hopp A, Soldner A and Wess J (1999) Use of a disulfide cross-linking strategy to study muscarinic receptor structure and mechanisms of activation. J Biol Chem 274(23):16629-16640.

Zeng FY and Wess J (1999) Identification and molecular characterization of m3 muscarinic receptor dimers. J Biol Chem 274(27):19487-19497.

Zhang JX, Braakman I, Matlack KE and Helenius A (1997) Quality control in the secretory pathway: the role of calreticulin, calnexin and BiP in the retention of glycoproteins with C-terminal truncations. Mol Biol Cell 8(10):1943-1954.

Zhang W, Trible RP and Samelson LE (1998) LAT palmitoylation: its essential role in membrane microdomain targeting and tyrosine phosphorylation during T cell activation. Immunity 9(2):239-246.

Zhao GQ, Zhang Y, Hoon MA, Chandrashekar J, Erlenbach I, Ryba NJ and Zuker CS (2003) The receptors for mammalian sweet and umami taste. Cell 115(3):255-266.

Zhen X, Uryu K, Wang HY and Friedman E (1998) D1 dopamine receptor agonists mediate activation of p38 mitogen-activated protein kinase and c-Jun amino-terminal kinase by a protein kinase A-dependent mechanism in SK-N-MC human neuroblastoma cells. Mol Pharmacol 54(3):453-458.

Zhu H, Wang H and Ascoli M (1995) The lutropin/choriogonadotropin receptor is palmitoylated at intracellular cysteine residues. Mol Endocrinol 9(2):141-150.

Zhu X and Wess J (1998) Truncated V2 vasopressin receptors as negative regulators of wild-type V2 receptor function. Biochemistry 37(45):15773-15784.

267