TURUN YLIOPISTON JULKAISUJA ANNALES UNIVERSITATIS TURKUENSIS TURUN YLIOPISTON JULKAISUJA ANNALES UNIVERSITATIS TURKUENSIS

SARJA - SER. D OSA - TOM. 962 SARJA - SER. D OSA - TOM. XXX MEDICA - ODONTOLOGICA MEDICA - ODONTOLOGICA

2-ADRENOCEPTORS: STRUCTURE

α2-ADRENOCEPTORS: STRUCTURE AND LIGAND BINDING PROPERTIES AND LIGAND BINDING PROPERTIES AT THE MOLECULAR LEVEL

AT THE MOLECULAR LEVEL

by by Jonne M.M. Laurila Jonne M.M. Laurila

TURUN YLIOPISTO Turku 2011

TURUN YLIOPISTO UNIVERSITY OF TURKU Turku 2011

From the Institute of Biomedicine, From the Institute of Biomedicine, Department of Pharmacology, Drug Development and Therapeutics Department of Pharmacology, Drug Development and Therapeutics University of Turku University of Turku Turku, Finland Turku, Finland

Supervised by Supervised by Professor Mika Scheinin, MD, PhD Professor Mika Scheinin, MD, PhD Department of Pharmacology, Drug Development and Therapeutics Department of Pharmacology, Drug Development and Therapeutics University of Turku University of Turku Turku, Finland Turku, Finland

Reviewed by Reviewed by Docent Jarmo T. Laitinen, PhD Docent Jarmo T. Laitinen, PhD Institute of Biomedicine/Physiology Institute of Biomedicine/Physiology University of Eastern Finland University of Eastern Finland Kuopio, Finland Kuopio, Finland and and

Docent Ulla Petäjä-Repo, PhD Docent Ulla Petäjä-Repo, PhD Institute of Biomedicine/Department of Anatomy and Cell Biology Institute of Biomedicine/Department of Anatomy and Cell Biology University of Oulu University of Oulu Oulu, Finland Oulu, Finland

Dissertation opponent Dissertation opponent Professor Heikki Ruskoaho, MD, PhD Professor Heikki Ruskoaho, MD, PhD Institute of Biomedicine/Department of Pharmacology and Toxicology Institute of Biomedicine/Department of Pharmacology and Toxicology University of Oulu University of Oulu Oulu, Finland Oulu, Finland

ISBN 978-951-29-4606-8 (PRINT) ISBN 978-951-29-4606-8 (PRINT) ISBN 978-951-29-4607-5 (PDF) ISBN 978-951-29-4607-5 (PDF) ISSN 0355-9483 ISSN 0355-9483 Painosalama Oy – Turku, Finland 2011 Painosalama Oy – Turku, Finland 2011

In memory of my father In memory of my father Tapani Laurila (1939-2006) Tapani Laurila (1939-2006)

Jonne M.M. Laurila Jonne M.M. Laurila 2-Adrenoceptors: structure and ligand binding properties at the molecular level 2-Adrenoceptors: structure and ligand binding properties at the molecular level Institute of Biomedicine, Department of Pharmacology, Drug Development and Institute of Biomedicine, Department of Pharmacology, Drug Development and Therapeutics, University of Turku; and Drug Discovery Graduate School, Turku, Therapeutics, University of Turku; and Drug Discovery Graduate School, Turku, Finland Finland Annales Universitatis Turkuensis, Painosalama Oy, Turku, Finland 2011 Annales Universitatis Turkuensis, Painosalama Oy, Turku, Finland 2011

ABSTRACT ABSTRACT

The mouse is the most frequently used animal model in biomedical research, but the The mouse is the most frequently used animal model in biomedical research, but the use of zebrafish as a model organism to mimic human diseases is on the increase. use of zebrafish as a model organism to mimic human diseases is on the increase. Therefore it is considered important to understand their pharmacological differences Therefore it is considered important to understand their pharmacological differences from humans also at the molecular level. from humans also at the molecular level. The zebrafish 2-adrenoceptors were expressed in mammalian cells and the binding The zebrafish 2-adrenoceptors were expressed in mammalian cells and the binding affinities of 20 diverse ligands were determined and compared to the corresponding affinities of 20 diverse ligands were determined and compared to the corresponding human receptors. The pharmacological properties of the human and zebrafish 2- human receptors. The pharmacological properties of the human and zebrafish 2- adrenoceptors were found to be quite well conserved. adrenoceptors were found to be quite well conserved. Receptor models based on the crystal structures of bovine rhodopsin and the human Receptor models based on the crystal structures of bovine rhodopsin and the human 2-adrenoceptor revealed that most structural differences between the paralogous and 2-adrenoceptor revealed that most structural differences between the paralogous and orthologous 2-adrenoceptors were located within the second extracellular loop (XL2). orthologous 2-adrenoceptors were located within the second extracellular loop (XL2). Reciprocal mutations were generated in the mouse and human 2A-adrenoceptors. Reciprocal mutations were generated in the mouse and human 2A-adrenoceptors. Ligand binding experiments revealed that substitutions in XL2 reversed the binding Ligand binding experiments revealed that substitutions in XL2 reversed the binding profiles of the human and mouse 2A-adrenoceptors for yohimbine, rauwolscine and profiles of the human and mouse 2A-adrenoceptors for yohimbine, rauwolscine and RS-79948-197, evidence for a role for XL2 in the determination of species-specific RS-79948-197, evidence for a role for XL2 in the determination of species-specific ligand binding. ligand binding. Previous mutagenesis studies had not been able to explain the subtype preference of Previous mutagenesis studies had not been able to explain the subtype preference of several large 2-adrenoceptor antagonists. We prepared chimaeric 2-adrenoceptors several large 2-adrenoceptor antagonists. We prepared chimaeric 2-adrenoceptors where the first transmembrane (TM1) domain was exchanged between the three human where the first transmembrane (TM1) domain was exchanged between the three human 2-adrenoceptor subtypes. The binding affinities of spiperone, spiroxatrine and 2-adrenoceptor subtypes. The binding affinities of spiperone, spiroxatrine and chlorpromazine were observed to be significantly improved by TM1 substitutions of chlorpromazine were observed to be significantly improved by TM1 substitutions of the 2A-adrenoceptor. Docking simulations indicated that indirect effects, such as the 2A-adrenoceptor. Docking simulations indicated that indirect effects, such as allosteric modulation, are more likely to be involved in this phenomenon rather than allosteric modulation, are more likely to be involved in this phenomenon rather than specific side-chain interactions between ligands and receptors. specific side-chain interactions between ligands and receptors.

Key words: 2-adrenoceptor, GPCR, ligand binding, XL2, receptor models Key words: 2-adrenoceptor, GPCR, ligand binding, XL2, receptor models

Jonne M.M. Laurila Jonne M.M. Laurila

2-Adrenoseptorit: rakenne ja lääkeaineiden sitoutumisominaisuudet 2-Adrenoseptorit: rakenne ja lääkeaineiden sitoutumisominaisuudet molekyylitasolla molekyylitasolla Biolääketieteen laitos, Farmakologia, lääkekehitys ja lääkehoito, Turun yliopisto; Biolääketieteen laitos, Farmakologia, lääkekehitys ja lääkehoito, Turun yliopisto; Lääkekehityksen tutkijakoulu (DDGS), Turku Lääkekehityksen tutkijakoulu (DDGS), Turku Annales Universitatis Turkuensis, Painosalama Oy, Turku 2011 Annales Universitatis Turkuensis, Painosalama Oy, Turku 2011

TIIVISTELMÄ TIIVISTELMÄ

Biolääketieteen tutkimuksessa hiiri on yleisimmin käytetty koe-eläin, mutta Biolääketieteen tutkimuksessa hiiri on yleisimmin käytetty koe-eläin, mutta seeprakalan käyttö ihmisen sairauksien mallinnuksessa on lisääntymässä. Siksi on seeprakalan käyttö ihmisen sairauksien mallinnuksessa on lisääntymässä. Siksi on tärkeää tutkia ja ymmärtää näiden eläinten ja ihmisten farmakologisten ominaisuuksien tärkeää tutkia ja ymmärtää näiden eläinten ja ihmisten farmakologisten ominaisuuksien eroavaisuuksia myös molekyylitasolla. eroavaisuuksia myös molekyylitasolla. Seeprakalan 2-adrenoseptoreja tuotettiin nisäkässoluissa ja verrattiin 20 erilaisen Seeprakalan 2-adrenoseptoreja tuotettiin nisäkässoluissa ja verrattiin 20 erilaisen lääkeyhdisteen sitoutumista niihin ja ihmisen vastaaviin reseptoreihin. Ihmisen ja lääkeyhdisteen sitoutumista niihin ja ihmisen vastaaviin reseptoreihin. Ihmisen ja seeprakalan 2-adrenoseptorien farmakologiset ominaisuudet todettiin hyvin saman- seeprakalan 2-adrenoseptorien farmakologiset ominaisuudet todettiin hyvin saman- laisiksi. laisiksi. Naudan rodopsiinin ja ihmisen 2-adrenoseptorin kiderakenteisiin perustuvat 2- Naudan rodopsiinin ja ihmisen 2-adrenoseptorin kiderakenteisiin perustuvat 2- reseptorien rakennemallit osoittivat, että suurimmat rakenne-erot sekä reseptoriala- reseptorien rakennemallit osoittivat, että suurimmat rakenne-erot sekä reseptoriala- tyyppien että eri eläinlajien reseptorien välillä löytyvät reseptorien toisesta solun- tyyppien että eri eläinlajien reseptorien välillä löytyvät reseptorien toisesta solun- ulkoisesta silmukkarakenteesta (XL2). Erilaiset aminohapot vaihdettiin hiiren ja ulkoisesta silmukkarakenteesta (XL2). Erilaiset aminohapot vaihdettiin hiiren ja ihmisen 2A-adrenoseptorien välillä toistensa kaltaisiksi. Näillä muokatuilla resepto- ihmisen 2A-adrenoseptorien välillä toistensa kaltaisiksi. Näillä muokatuilla resepto- reilla tehdyt sitoutumiskokeet osoittivat kolmen testatun lääkeaineen, johimbiinin, reilla tehdyt sitoutumiskokeet osoittivat kolmen testatun lääkeaineen, johimbiinin, rauwolskiinin ja RS-79948-197:n, sitoutumishanakkuuden muuttuvan vastaavasti. rauwolskiinin ja RS-79948-197:n, sitoutumishanakkuuden muuttuvan vastaavasti. Tulos osoitti, että XL2:n rakenne osaltaan määrittää eläinlajille tyypillisen lääke- Tulos osoitti, että XL2:n rakenne osaltaan määrittää eläinlajille tyypillisen lääke- aineiden sitoutumisprofiilin sen 2A-reseptoreihin. aineiden sitoutumisprofiilin sen 2A-reseptoreihin. Aiemmat reseptorien muokkaamiseen perustuvat tutkimukset eivät olleet kyenneet Aiemmat reseptorien muokkaamiseen perustuvat tutkimukset eivät olleet kyenneet selittämään isokokoisten 2-reseptoreja salpaavien lääkeainemolekyylien valikoivaa selittämään isokokoisten 2-reseptoreja salpaavien lääkeainemolekyylien valikoivaa sitoutumista 2-reseptorialatyyppeihin. Sen vuoksi valmistettiin ns. kimeerisiä 2- sitoutumista 2-reseptorialatyyppeihin. Sen vuoksi valmistettiin ns. kimeerisiä 2- reseptoreja, joissa ensimmäistä solukalvon läpäisevää rakennejaksoa (TM1) vaihdeltiin reseptoreja, joissa ensimmäistä solukalvon läpäisevää rakennejaksoa (TM1) vaihdeltiin systemaattisesti ihmisen kolmen 2-reseptorialatyypin välillä. TM1-jakson vaihdon systemaattisesti ihmisen kolmen 2-reseptorialatyypin välillä. TM1-jakson vaihdon seurauksena kolmen reseptorinsalpaajan, spiperonin, spiroksatriinin ja klooripromatsii- seurauksena kolmen reseptorinsalpaajan, spiperonin, spiroksatriinin ja klooripromatsii- nin havaittiin sitoutuvan aiempaa selvästi hanakammin 2A-reseptoreihin. Reseptorien nin havaittiin sitoutuvan aiempaa selvästi hanakammin 2A-reseptoreihin. Reseptorien tietokonemalleilla suoritetut lääkeaineiden sovittamissimulaatiot viittasivat siihen, että tietokonemalleilla suoritetut lääkeaineiden sovittamissimulaatiot viittasivat siihen, että lisääntynyt sitoutumishanakkuus ei johdu spesifisistä vuorovaikutusten muutoksista lisääntynyt sitoutumishanakkuus ei johdu spesifisistä vuorovaikutusten muutoksista reseptorien sitomiskohtien ja lääkeaineiden välillä vaan epäsuorista vaikutuksista reseptorien sitomiskohtien ja lääkeaineiden välillä vaan epäsuorista vaikutuksista reseptorien kolmiulotteiseen muotoon. reseptorien kolmiulotteiseen muotoon.

Avainsanat: 2-adrenoseptori, GPCR, lääkeaineiden sitoutuminen, XL2, reseptori- Avainsanat: 2-adrenoseptori, GPCR, lääkeaineiden sitoutuminen, XL2, reseptori- mallit mallit

Table of Contents 7 Table of Contents 7

TABLE OF CONTENTS TABLE OF CONTENTS

ABBREVIATIONS ...... 9 ABBREVIATIONS ...... 9

LIST OF ORIGINAL COMMUNICATIONS ...... 10 LIST OF ORIGINAL COMMUNICATIONS ...... 10

1. INTRODUCTION ...... 11 1. INTRODUCTION ...... 11

2. REVIEW OF THE LITERATURE ...... 12 2. REVIEW OF THE LITERATURE ...... 12

2.1 ɑ2-Adrenoceptors, members of the G-protein coupled receptor (GPCR) 2.1 ɑ2-Adrenoceptors, members of the G-protein coupled receptor (GPCR) superfamily ...... 12 superfamily ...... 12 2.1.1 Concept of GPCRs ...... 12 2.1.1 Concept of GPCRs ...... 12 2.1.2 Subtypes of 2-adrenoceptors ...... 14 2.1.2 Subtypes of 2-adrenoceptors ...... 14

2.2 Structure-function relationships of 2-adrenoceptors ...... 16 2.2 Structure-function relationships of 2-adrenoceptors ...... 16

2.2.1 2-Adrenoceptor ligands and subtype selectivity ...... 16 2.2.1 2-Adrenoceptor ligands and subtype selectivity ...... 16 2.2.2 Experimental probing of receptor structure and ligand interactions ...... 19 2.2.2 Experimental probing of receptor structure and ligand interactions ...... 19 2.2.3 Structural determinants of 2-adrenoceptor ligand recognition and 2.2.3 Structural determinants of 2-adrenoceptor ligand recognition and binding ...... 20 binding ...... 20 2.2.4 Molecular dynamics of 2-adrenoceptor activation ...... 25 2.2.4 Molecular dynamics of 2-adrenoceptor activation ...... 25 2.2.5 Physiological functions and patterns of tissue expression of 2.2.5 Physiological functions and patterns of tissue expression of 2-adrenoceptors ...... 28 2-adrenoceptors ...... 28 2.2.6 Therapeutic applications of 2-adrenoceptors ...... 31 2.2.6 Therapeutic applications of 2-adrenoceptors ...... 31

3. AIMS OF THE STUDY ...... 33 3. AIMS OF THE STUDY ...... 33

4. MATERIALS AND METHODS ...... 34 4. MATERIALS AND METHODS ...... 34 4.1 Mutagenesis and expression vectors ...... 34 4.1 Mutagenesis and expression vectors ...... 34 4.2 Cell culture and transfections ...... 37 4.2 Cell culture and transfections ...... 37 4.3 Reverse transcription-PCR ...... 37 4.3 Reverse transcription-PCR ...... 37 4.4 Membrane preparation ...... 37 4.4 Membrane preparation ...... 37 4.5 Ligand binding assays ...... 38 4.5 Ligand binding assays ...... 38 4.5.1 Saturation experiments ...... 38 4.5.1 Saturation experiments ...... 38 4.5.2 Competition experiments ...... 39 4.5.2 Competition experiments ...... 39 4.5.3 Functional [35S]GTPS binding assay ...... 44 4.5.3 Functional [35S]GTPS binding assay ...... 44 4.6 Molecular modelling and ligand docking ...... 44 4.6 Molecular modelling and ligand docking ...... 44

5 RESULTS ...... 46 5 RESULTS ...... 46

5.1 Structural and pharmacological properties of the zebrafish 2-adrenoceptors 5.1 Structural and pharmacological properties of the zebrafish 2-adrenoceptors in comparison with the human orthologues (I) ...... 46 in comparison with the human orthologues (I) ...... 46

8 Table of Contents 8 Table of Contents

5.1.1 Characterization of receptor expression in CHO cells ...... 46 5.1.1 Characterization of receptor expression in CHO cells ...... 46 5.1.2 Competition binding assays and cluster analysis of the receptors and 5.1.2 Competition binding assays and cluster analysis of the receptors and ligands ...... 47 ligands ...... 47 5.1.3 Agonist-stimulated [35S]GTPS binding ...... 50 5.1.3 Agonist-stimulated [35S]GTPS binding ...... 50 5.1.4 Comparison of the structural models of the human and zebrafish 5.1.4 Comparison of the structural models of the human and zebrafish 2-adrenoceptors ...... 51 2-adrenoceptors ...... 51 5.2 Structural determinants involved in the interspecies difference of yohimbine 5.2 Structural determinants involved in the interspecies difference of yohimbine analogues at human and mouse 2A-adrenoceptors (II) ...... 52 analogues at human and mouse 2A-adrenoceptors (II) ...... 52

5.2.1 Comparison of the binding cavities of the human and mouse 2A- 5.2.1 Comparison of the binding cavities of the human and mouse 2A- adrenoceptors ...... 52 adrenoceptors ...... 52 3 3 5.2.2 [ H]RX821002 binding at human and mouse 2A-adrenoceptors ...... 54 5.2.2 [ H]RX821002 binding at human and mouse 2A-adrenoceptors ...... 54 5.2.3 Effects of XL2 and TM5 substitutions on ligand binding profiles ...... 54 5.2.3 Effects of XL2 and TM5 substitutions on ligand binding profiles ...... 54

5.3 Involvement of the first transmembrane domain of human 2-adrenoceptors 5.3 Involvement of the first transmembrane domain of human 2-adrenoceptors in the subtype-selectivity of bulky antagonists (III) ...... 57 in the subtype-selectivity of bulky antagonists (III) ...... 57 5.3.1 Construction of receptor chimaeras ...... 57 5.3.1 Construction of receptor chimaeras ...... 57 5.3.2 Expression of 2-adrenoceptor chimaeras in CHO cells ...... 58 5.3.2 Expression of 2-adrenoceptor chimaeras in CHO cells ...... 58 5.3.3 Characterization of antagonist binding profiles at wild-type and 5.3.3 Characterization of antagonist binding profiles at wild-type and chimaeric 2-adrenoceptors ...... 60 chimaeric 2-adrenoceptors ...... 60

6 DISCUSSION ...... 62 6 DISCUSSION ...... 62 6.1 Structural and pharmacological comparison of the orthologous 6.1 Structural and pharmacological comparison of the orthologous 2-adrenoceptors ...... 62 2-adrenoceptors ...... 62 6.2 The second extracellular loop forms part of the ligand binding site in 6.2 The second extracellular loop forms part of the ligand binding site in 2-adrenoceptors ...... 65 2-adrenoceptors ...... 65 6.3 Indirect effects may influence pharmacological profiles ...... 68 6.3 Indirect effects may influence pharmacological profiles ...... 68

7 CONCLUSIONS ...... 71 7 CONCLUSIONS ...... 71

ACKNOWLEDGEMENTS ...... 72 ACKNOWLEDGEMENTS ...... 72

REFERENCES ...... 74 REFERENCES ...... 74

ORIGINAL COMMUNICATIONS ...... 83 ORIGINAL COMMUNICATIONS ...... 83

Abbreviations 9 Abbreviations 9

ABBREVIATIONS ABBREVIATIONS

5-HT 5-Hydroxytryptamine, serotonin 5-HT 5-Hydroxytryptamine, serotonin

Bmax Maximal binding capacity Bmax Maximal binding capacity cAMP Cyclic adenosine monophosphate cAMP Cyclic adenosine monophosphate CHO Chinese hamster ovary CHO Chinese hamster ovary CNS Central nervous system CNS Central nervous system C-terminus Carboxyl terminus (of polypeptide) C-terminus Carboxyl terminus (of polypeptide) DRY Aspartic acid-arginine-tyrosine motif DRY Aspartic acid-arginine-tyrosine motif

EC50 Drug concentration that elicits 50 % of the maximal response EC50 Drug concentration that elicits 50 % of the maximal response EDTA Ethylenediaminetetraacetic acid EDTA Ethylenediaminetetraacetic acid G protein Guanine nucleotide binding protein G protein Guanine nucleotide binding protein GDP Guanosine diphosphate GDP Guanosine diphosphate GPCR G-protein coupled receptor GPCR G-protein coupled receptor GTP Guanosine triphosphate GTP Guanosine triphosphate GTPS Guanosine 5´-[gamma-thio]triphosphate GTPS Guanosine 5´-[gamma-thio]triphosphate

IC50 Drug concentration that inhibits 50 % of specific binding IC50 Drug concentration that inhibits 50 % of specific binding

Kd Equilibrium dissociation constant Kd Equilibrium dissociation constant

Ki Binding inhibition constant Ki Binding inhibition constant MAO Monoamine oxidase MAO Monoamine oxidase NSB Non-specific binding NSB Non-specific binding N-terminus Amino terminus (of polypeptide) N-terminus Amino terminus (of polypeptide) PCA Principal component analysis PCA Principal component analysis PCR Polymerase chain reaction PCR Polymerase chain reaction PTX Pertussis toxin PTX Pertussis toxin RT Room temperature RT Room temperature RT-PCR Reverse transcription PCR RT-PCR Reverse transcription PCR SCAM Substituted cysteine accessibility method SCAM Substituted cysteine accessibility method TM (TM1-TM7) Transmembrane domain (1-7) TM (TM1-TM7) Transmembrane domain (1-7) XL2 Second extracellular loop XL2 Second extracellular loop

10 List of Original Communications 10 List of Original Communications

LIST OF ORIGINAL COMMUNICATIONS LIST OF ORIGINAL COMMUNICATIONS

This thesis is based on the following original publications, which are referred to in the This thesis is based on the following original publications, which are referred to in the text by the Roman numerals I-III. In addition, some unpublished data are presented. text by the Roman numerals I-III. In addition, some unpublished data are presented.

I Ruuskanen JO, Laurila J, Xhaard H, Rantanen V-V, Vuoriluoto K, Wurster S, I Ruuskanen JO, Laurila J, Xhaard H, Rantanen V-V, Vuoriluoto K, Wurster S, Marjamäki A, Vainio M, Johnson MS, and Scheinin M (2005). Conserved Marjamäki A, Vainio M, Johnson MS, and Scheinin M (2005). Conserved structural, pharmacological and functional properties among the three human structural, pharmacological and functional properties among the three human and five zebrafish 2-adrenoceptors. Br J Pharmacol 144: 165-177. and five zebrafish 2-adrenoceptors. Br J Pharmacol 144: 165-177.

II Laurila JM, Xhaard H, Ruuskanen JO, Rantanen MJ, Karlsson HK, Johnson II Laurila JM, Xhaard H, Ruuskanen JO, Rantanen MJ, Karlsson HK, Johnson MS, and Scheinin M (2007). The second extracellular loop of 2A-adrenoceptors MS, and Scheinin M (2007). The second extracellular loop of 2A-adrenoceptors contributes to the binding of yohimbine analogues. Br J Pharmacol 151: 1293- contributes to the binding of yohimbine analogues. Br J Pharmacol 151: 1293- 1304. 1304.

III Laurila JM, Wissel G, Xhaard H, Ruuskanen JO, Johnson MS, and Scheinin M III Laurila JM, Wissel G, Xhaard H, Ruuskanen JO, Johnson MS, and Scheinin M (2011). Involvement of the first transmembrane segment of human 2A- (2011). Involvement of the first transmembrane segment of human 2A- adrenoceptors in the subtype-selective binding of chlorpromazine, spiperone and adrenoceptors in the subtype-selective binding of chlorpromazine, spiperone and spiroxatrine. Br J Pharmacol under review. spiroxatrine. Br J Pharmacol under review.

The original communications have been reproduced with kind permission of the The original communications have been reproduced with kind permission of the copyright holder. copyright holder.

Introduction 11 Introduction 11

1. INTRODUCTION 1. INTRODUCTION

Proteins embedded in cell membranes represent a major proportion (~25 %) of all Proteins embedded in cell membranes represent a major proportion (~25 %) of all proteins encoded by mammalian genomes. G-protein coupled receptors (GPCRs) are a proteins encoded by mammalian genomes. G-protein coupled receptors (GPCRs) are a large and biologically very important superfamily of cell membrane proteins. They large and biologically very important superfamily of cell membrane proteins. They mediate extracellular signals to intracellular effector pathways by activating mediate extracellular signals to intracellular effector pathways by activating membrane-associated guanine nucleotide binding regulatory proteins, or G-proteins. membrane-associated guanine nucleotide binding regulatory proteins, or G-proteins. GPCRs are important targets for pharmaceutical development as they regulate a wide GPCRs are important targets for pharmaceutical development as they regulate a wide range of physiological processes. It is estimated that almost 40 % of all currently used range of physiological processes. It is estimated that almost 40 % of all currently used therapeutic drugs mediate their effects via GPCRs. therapeutic drugs mediate their effects via GPCRs. All GPCRs share a common basic structure, consisting of seven -helical All GPCRs share a common basic structure, consisting of seven -helical transmembrane (TM) domains connected by intracellular and extracellular loops. The transmembrane (TM) domains connected by intracellular and extracellular loops. The adrenoceptors (1-, 2- and -adrenoceptors) are members of the rhodopsin-like GPCR adrenoceptors (1-, 2- and -adrenoceptors) are members of the rhodopsin-like GPCR subfamily. They mediate cellular responses to the adrenomedullary hormone subfamily. They mediate cellular responses to the adrenomedullary hormone adrenaline and the neurotransmitter noradrenaline. They posses a membrane-embedded adrenaline and the neurotransmitter noradrenaline. They posses a membrane-embedded water-accessible ligand binding cavity contained between the hydrophobic TM water-accessible ligand binding cavity contained between the hydrophobic TM domains. The amino acids that form the surface of the ligand binding cavity are domains. The amino acids that form the surface of the ligand binding cavity are considered to be potential contact sites for selective ligand recognition. Other amino considered to be potential contact sites for selective ligand recognition. Other amino acids in the TM domains lack direct contacts with ligands but have at least a structural acids in the TM domains lack direct contacts with ligands but have at least a structural role, thereby having a possibility to indirectly influence ligand binding. role, thereby having a possibility to indirectly influence ligand binding. Humans and other mammalian species have three 2-adrenoceptor subtypes (2A2B Humans and other mammalian species have three 2-adrenoceptor subtypes (2A2B and 2C) that are encoded by three distinct intronless genes. The three 2-adrenoceptors and 2C) that are encoded by three distinct intronless genes. The three 2-adrenoceptors show high structural similarity, especially within their TM domains, where 70-80 % of show high structural similarity, especially within their TM domains, where 70-80 % of the amino acids are identical between any two receptor subtypes. Their conserved the amino acids are identical between any two receptor subtypes. Their conserved structure is also reflected in their very similar ligand binding properties, especially for the structure is also reflected in their very similar ligand binding properties, especially for the endogenous catecholamines, adrenaline and noradrenaline, which bind with rather endogenous catecholamines, adrenaline and noradrenaline, which bind with rather similar affinities to the receptor subtypes within an animal species (paralogues) and to similar affinities to the receptor subtypes within an animal species (paralogues) and to homologous subtypes between species (orthologues). Nonetheless, differences have been homologous subtypes between species (orthologues). Nonetheless, differences have been found in the binding affinities of some synthetic ligands, especially for antagonist found in the binding affinities of some synthetic ligands, especially for antagonist ligands, whose binding to 2-adrenoceptors is less well understood than that of ligands, whose binding to 2-adrenoceptors is less well understood than that of catecholamines and related agonists. In spite of the advent of modern drug discovery catecholamines and related agonists. In spite of the advent of modern drug discovery technologies, current clinically available 2-adrenoceptor drugs have only marginal technologies, current clinically available 2-adrenoceptor drugs have only marginal subtype-selectivity, which limits their therapeutic usefulness because of various side- subtype-selectivity, which limits their therapeutic usefulness because of various side- effects. Subtype-selective drugs could have therapeutic applications e.g. in the treatment effects. Subtype-selective drugs could have therapeutic applications e.g. in the treatment of elevated blood pressure, in relief of pain and of opioid and alcohol withdrawal of elevated blood pressure, in relief of pain and of opioid and alcohol withdrawal symptoms, as neuropsychiatric therapeutic agents and in anaesthetic care. symptoms, as neuropsychiatric therapeutic agents and in anaesthetic care. A better understanding of the molecular basis of drug actions on 2-adrenoceptors, in A better understanding of the molecular basis of drug actions on 2-adrenoceptors, in terms of receptor structure and function, may in the future allow the design and terms of receptor structure and function, may in the future allow the design and development of new subtype-selective drug molecules. In this thesis work, computer-based development of new subtype-selective drug molecules. In this thesis work, computer-based molecular modelling and sequence comparisons between paralogous and orthologous molecular modelling and sequence comparisons between paralogous and orthologous receptor subtypes along with experimental site-directed mutagenesis and pharmacological receptor subtypes along with experimental site-directed mutagenesis and pharmacological in vitro assays were used to characterise the structural determinants of human, mouse and in vitro assays were used to characterise the structural determinants of human, mouse and zebrafish 2-adrenoceptors with respect to their specific ligand binding properties. zebrafish 2-adrenoceptors with respect to their specific ligand binding properties.

12 Review of the Literature 12 Review of the Literature

2. REVIEW OF THE LITERATURE 2. REVIEW OF THE LITERATURE

2.1 ɑ2-Adrenoceptors, members of the G-protein coupled receptor 2.1 ɑ2-Adrenoceptors, members of the G-protein coupled receptor (GPCR) superfamily (GPCR) superfamily

2.1.1 Concept of GPCRs 2.1.1 Concept of GPCRs The concept of receptor molecules was first postulated by John N. Langley at the turn The concept of receptor molecules was first postulated by John N. Langley at the turn of the 20th century. According to his theory, the effects of curare and nicotine on of the 20th century. According to his theory, the effects of curare and nicotine on skeletal muscle could be attributed to special “receptive substances” or “receptors” skeletal muscle could be attributed to special “receptive substances” or “receptors” (Langley, 1905; reviewed by Maehle, 2004). Later, this concept was broadened and (Langley, 1905; reviewed by Maehle, 2004). Later, this concept was broadened and developed to successfully explain the phenomenon that membrane-bound cell surface developed to successfully explain the phenomenon that membrane-bound cell surface receptors were the molecular entities responsible for the communication of signalling receptors were the molecular entities responsible for the communication of signalling events between the extracellular and intracellular environments. It is now known that events between the extracellular and intracellular environments. It is now known that various endogenous signalling molecules such as hormones, neurotransmitters, various endogenous signalling molecules such as hormones, neurotransmitters, neuromodulators and growth factors exert their cellular signals through these proteins neuromodulators and growth factors exert their cellular signals through these proteins (Bockaert and Pin, 1999). (Bockaert and Pin, 1999). G-protein coupled receptors (GPCRs) comprise one of the largest families of G-protein coupled receptors (GPCRs) comprise one of the largest families of proteins in humans, and all cell types express some subset of GPCRs. GPCRs are proteins in humans, and all cell types express some subset of GPCRs. GPCRs are integral membrane proteins that span the depth of the cell membrane. They transduce integral membrane proteins that span the depth of the cell membrane. They transduce extracellular signals into intracellular messages by acting on membrane-associated extracellular signals into intracellular messages by acting on membrane-associated guanine nucleotide binding regulatory proteins, or G-proteins. Genes encoding GPCRs guanine nucleotide binding regulatory proteins, or G-proteins. Genes encoding GPCRs comprise about 3 % of all genes in the human genome (Milligan and Kostenis, 2006). comprise about 3 % of all genes in the human genome (Milligan and Kostenis, 2006). GPCRs are composed of a common “canonical” motif of seven -helical GPCRs are composed of a common “canonical” motif of seven -helical transmembrane domains (TM1-TM7), connected to each other by three flexible transmembrane domains (TM1-TM7), connected to each other by three flexible intracellular and extracellular loops, an extracellular amino terminus and an intracellular and extracellular loops, an extracellular amino terminus and an intracellular carboxyl-terminal tail. The perpendicularly oriented TM domains are intracellular carboxyl-terminal tail. The perpendicularly oriented TM domains are composed of 25-35 hydrophobic amino acids, and are embedded in the phospholipid- composed of 25-35 hydrophobic amino acids, and are embedded in the phospholipid- rich plasma membrane in an anticlockwise manner (as seen from the outside), forming rich plasma membrane in an anticlockwise manner (as seen from the outside), forming a central receptor core that in many cases is involved in ligand recognition and binding a central receptor core that in many cases is involved in ligand recognition and binding (Figure 1). In order to conduct comparative structural biological comparisons in (Figure 1). In order to conduct comparative structural biological comparisons in receptor characterization, different indexing systems for the amino acids have been receptor characterization, different indexing systems for the amino acids have been established. In the Ballesteros-Weinstein convention (Ballesteros and Weinstein, established. In the Ballesteros-Weinstein convention (Ballesteros and Weinstein, 1995), the amino acid residues are numbered according to an indexing system where 1995), the amino acid residues are numbered according to an indexing system where the first number refers to the TM helix where the residue is located and the number the first number refers to the TM helix where the residue is located and the number after the decimal point refers to the residue position with respect to the most conserved after the decimal point refers to the residue position with respect to the most conserved residue in that helix, which has been arbitrarily assigned the number 50. This indexing residue in that helix, which has been arbitrarily assigned the number 50. This indexing system was further extended to encompass also the extracellular loops. system was further extended to encompass also the extracellular loops.

Review of the Literature 13 Review of the Literature 13

Figure 1. Schematic representation of TM helices organized in a counterclockwise fashion, Figure 1. Schematic representation of TM helices organized in a counterclockwise fashion, viewed from the extracellular side of the cell membrane. The putative ligand binding site is viewed from the extracellular side of the cell membrane. The putative ligand binding site is shown between the TM helices (A). A conventional 2-dimensional “snake diagram” of the shown between the TM helices (A). A conventional 2-dimensional “snake diagram” of the seven TM domains embedded in the plasma membrane. The intracellular carboxyl-terminal tail seven TM domains embedded in the plasma membrane. The intracellular carboxyl-terminal tail is attached to the plasma membrane via cysteine-linked palmitoylation (B). is attached to the plasma membrane via cysteine-linked palmitoylation (B).

The common fundamental structure of GPCRs was verified by X-ray The common fundamental structure of GPCRs was verified by X-ray crystallography of bovine rhodopsin about ten years ago (Palczewski et al., 2000). The crystallography of bovine rhodopsin about ten years ago (Palczewski et al., 2000). The lack of stability and higher-order symmetry of membrane proteins in the native state lack of stability and higher-order symmetry of membrane proteins in the native state effectively hindered the development of diffraction methods for other GPCRs for a effectively hindered the development of diffraction methods for other GPCRs for a long time. In the past few years, very significant progress in technology has made long time. In the past few years, very significant progress in technology has made crystallization and high resolution X-ray diffraction analysis possible also for some crystallization and high resolution X-ray diffraction analysis possible also for some other GPCRs, and the crystal structures of the human 2-adrenoceptor, the turkey 1- other GPCRs, and the crystal structures of the human 2-adrenoceptor, the turkey 1- adrenoceptor, and the human A2A adenosine, dopamine D3 and CXCR4 chemokine adrenoceptor, and the human A2A adenosine, dopamine D3 and CXCR4 chemokine receptors have now been characterized at the time of writing of this review (Cherezov receptors have now been characterized at the time of writing of this review (Cherezov et al., 2007, Rasmussen et al., 2007, Jaakola et al., 2008, Warne et al., 2008, Chien et et al., 2007, Rasmussen et al., 2007, Jaakola et al., 2008, Warne et al., 2008, Chien et al., 2010, Wu et al., 2010). This major advance has provided a wealth of new al., 2010, Wu et al., 2010). This major advance has provided a wealth of new information on the structural biology of GPCRs and it is highly likely that more information on the structural biology of GPCRs and it is highly likely that more receptor structures will be reported in the near future. receptor structures will be reported in the near future. GPCRs mediate effects of a wide array of endogenous and exogenous ligands, and GPCRs mediate effects of a wide array of endogenous and exogenous ligands, and regulate many important cellular functions. GPCRs are involved in almost all regulate many important cellular functions. GPCRs are involved in almost all physiological processes and have been associated with the pathology or therapy of physiological processes and have been associated with the pathology or therapy of many common diseases such as impaired vision (involving rhodopsin mutations; many common diseases such as impaired vision (involving rhodopsin mutations; Menon et al., 2001), many cardiovascular diseases (where especially 1-adrenoceptor Menon et al., 2001), many cardiovascular diseases (where especially 1-adrenoceptor antagonists are widely used; McNamara et al., 2002), and asthma (where 2- antagonists are widely used; McNamara et al., 2002), and asthma (where 2- adrenoceptor agonists are one form of therapy; Johnson, 2001). Their location in the adrenoceptor agonists are one form of therapy; Johnson, 2001). Their location in the plasma membrane makes GPCRs readily accessible to endogenous signalling plasma membrane makes GPCRs readily accessible to endogenous signalling molecules but it also makes them very attractive drug targets. It has been estimated that molecules but it also makes them very attractive drug targets. It has been estimated that about 25-40 % of the current clinically used drugs act on GPCRs, and many of them about 25-40 % of the current clinically used drugs act on GPCRs, and many of them are found among the 100 best-selling pharmaceutical products (Lagerström and are found among the 100 best-selling pharmaceutical products (Lagerström and Schiöth, 2008). On the other hand, the human genome contains approximately 100 Schiöth, 2008). On the other hand, the human genome contains approximately 100 GPCRs whose ligands and biological functions are still unknown (Suwa and Ono, GPCRs whose ligands and biological functions are still unknown (Suwa and Ono, 2009). These “orphan” receptors are very interesting for pharmacologists and the 2009). These “orphan” receptors are very interesting for pharmacologists and the pharmaceutical industry as they may be potential drug targets. pharmaceutical industry as they may be potential drug targets.

14 Review of the Literature 14 Review of the Literature

GPCRs have been highly conserved throughout evolution. These receptors are GPCRs have been highly conserved throughout evolution. These receptors are expressed in almost all living organisms, from prokaryotic bacteria to eukaryotic expressed in almost all living organisms, from prokaryotic bacteria to eukaryotic human cells. About 800 genes encode human GPCRs, of which more than 300 are non- human cells. About 800 genes encode human GPCRs, of which more than 300 are non- olfactory and about 400 are olfactory or other chemosensory receptors (Fredriksson et olfactory and about 400 are olfactory or other chemosensory receptors (Fredriksson et al., 2003, Niimura and Nei, 2006, Schiöth et al., 2007, Lagerström and Schiöth, 2008, al., 2003, Niimura and Nei, 2006, Schiöth et al., 2007, Lagerström and Schiöth, 2008, Nordström et al., 2009). In addition, many of the GPCR genes can yield several Nordström et al., 2009). In addition, many of the GPCR genes can yield several transcripts or splice variants. GPCRs are involved in a broad range of physiological transcripts or splice variants. GPCRs are involved in a broad range of physiological functions - in taste, olfaction, vision, immune responses as well as in higher functions functions - in taste, olfaction, vision, immune responses as well as in higher functions of the nervous system, e.g. memory and other cognitive functions, emotions, attention of the nervous system, e.g. memory and other cognitive functions, emotions, attention and pain. The chemical diversity of endogenous GPCR ligands is large, encompassing and pain. The chemical diversity of endogenous GPCR ligands is large, encompassing molecules like small and large peptides, monoamine neurotransmitters, nucleotides, molecules like small and large peptides, monoamine neurotransmitters, nucleotides, amino acids, lipids, even larger signalling hormones such as chemokines (Kristiansen, amino acids, lipids, even larger signalling hormones such as chemokines (Kristiansen, 2004, Lagerström and Schiöth, 2008). 2004, Lagerström and Schiöth, 2008). GPCRs are classified into subfamilies based on several classification systems GPCRs are classified into subfamilies based on several classification systems related to their sequence similarity and protein structure, ligand structure, ligand related to their sequence similarity and protein structure, ligand structure, ligand binding mode and receptor function. One of the most commonly used systems groups binding mode and receptor function. One of the most commonly used systems groups GPCRs into six classes, A-F, of which classes D, E and F contain receptors that are not GPCRs into six classes, A-F, of which classes D, E and F contain receptors that are not expressed in humans (Attwood and Findlay, 1994, Kolakowski, 1994). In this system, expressed in humans (Attwood and Findlay, 1994, Kolakowski, 1994). In this system, group A, the rhodopsin-like receptors, is the most numerous. Its subfamily 1 contains group A, the rhodopsin-like receptors, is the most numerous. Its subfamily 1 contains GPCRs for small ligands including the adrenergic, adenosine and serotonin receptors. GPCRs for small ligands including the adrenergic, adenosine and serotonin receptors. In another more recent classification system (also known as the GRAFS classification) In another more recent classification system (also known as the GRAFS classification) that is based on a phylogenetic analysis of more than 800 human GPCRs, the receptors that is based on a phylogenetic analysis of more than 800 human GPCRs, the receptors are clustered into five main families: glutamate (G, with 15 members), rhodopsin-like are clustered into five main families: glutamate (G, with 15 members), rhodopsin-like (R, n = 701), adhesion (A, n = 24), frizzled/taste2 (F, n = 24), and secretin (S, n = 15) (R, n = 701), adhesion (A, n = 24), frizzled/taste2 (F, n = 24), and secretin (S, n = 15) (Fredriksson et al., 2003). Of these, the large rhodopsin-like receptor family consists of (Fredriksson et al., 2003). Of these, the large rhodopsin-like receptor family consists of four main groups (and 13 subfamilies. This overall mapping system of the four main groups (and 13 subfamilies. This overall mapping system of the entire GPCR superfamily suggests that all GPCRs share a common evolutionary origin entire GPCR superfamily suggests that all GPCRs share a common evolutionary origin traceable to a single ancestral gene. traceable to a single ancestral gene.

2.1.2 Subtypes of 2-adrenoceptors 2.1.2 Subtypes of 2-adrenoceptors The first classification of “adrenotropic” receptors was postulated by Raymond Ahlquist The first classification of “adrenotropic” receptors was postulated by Raymond Ahlquist in 1948 and was based on the different pharmacological characteristics of adrenaline-like in 1948 and was based on the different pharmacological characteristics of adrenaline-like agonists in different tissues (Ahlquist, 1948). Ahlquist noted that there must be at least agonists in different tissues (Ahlquist, 1948). Ahlquist noted that there must be at least two types of adrenotropic receptors: those whose activation resulted in excitation and two types of adrenotropic receptors: those whose activation resulted in excitation and those whose activation evoked inhibition of the target cells. Ahlquist defined these those whose activation evoked inhibition of the target cells. Ahlquist defined these receptor types as - and -receptors. Subsequently, the “adrenotropic” receptors have receptor types as - and -receptors. Subsequently, the “adrenotropic” receptors have been divided into three different main classes and nine mammalian subtypes. been divided into three different main classes and nine mammalian subtypes. Today, receptors are mainly classified based on their primary amino acid sequences. Today, receptors are mainly classified based on their primary amino acid sequences. The first cloned adrenoceptor was the hamster 2-adrenoceptor (Dixon et al., 1986). The first cloned adrenoceptor was the hamster 2-adrenoceptor (Dixon et al., 1986). Soon thereafter, the genes of the other members of the adrenoceptor family were Soon thereafter, the genes of the other members of the adrenoceptor family were cloned. In the current scheme, adrenoceptors are classified into three main classes, 1-, cloned. In the current scheme, adrenoceptors are classified into three main classes, 1-, 2- and -adrenoceptors, based on their amino acid sequences and biological and 2- and -adrenoceptors, based on their amino acid sequences and biological and pharmacological properties, and each of these classes has been further divided into pharmacological properties, and each of these classes has been further divided into three subtypes (Bylund et al., 1992) (Figure 2). These nine adrenoceptor subtypes, three subtypes (Bylund et al., 1992) (Figure 2). These nine adrenoceptor subtypes, encoded by distinct genes, have been identified in many mammalian species. All encoded by distinct genes, have been identified in many mammalian species. All

Review of the Literature 15 Review of the Literature 15 adrenoceptors share a relatively high degree of amino acid identity, especially within adrenoceptors share a relatively high degree of amino acid identity, especially within their TM domains that form the membrane-embedded core of the proteins and contain their TM domains that form the membrane-embedded core of the proteins and contain the ligand binding pocket (66-75 % identity within TM regions between subtypes for the ligand binding pocket (66-75 % identity within TM regions between subtypes for 1-, 77-79 % for 2-, and 62-70 % for -adrenoceptors) (Xhaard et al., 2006). The 1-, 77-79 % for 2-, and 62-70 % for -adrenoceptors) (Xhaard et al., 2006). The three 2-adrenoceptor subtypes, initially designated as 2-C10 (Kobilka et al., 1987), three 2-adrenoceptor subtypes, initially designated as 2-C10 (Kobilka et al., 1987), 2-C2 (Lomasney et al., 1990) and 2-C4 (Regan et al., 1988) based on the 2-C2 (Lomasney et al., 1990) and 2-C4 (Regan et al., 1988) based on the localization of their genes on human chromosomes 10, 2 and 4, are better known as the localization of their genes on human chromosomes 10, 2 and 4, are better known as the subtypes 2A, 2B and 2C. The current molecular classification agrees with the earlier subtypes 2A, 2B and 2C. The current molecular classification agrees with the earlier receptor subtype classification that was based on their pharmacological properties. The receptor subtype classification that was based on their pharmacological properties. The three 2-adrenoceptor subtypes have many similarities but also show variations in three 2-adrenoceptor subtypes have many similarities but also show variations in protein sequence, ligand binding, regulation, as well as in patterns of tissue and cell protein sequence, ligand binding, regulation, as well as in patterns of tissue and cell expression (Eason et al., 1994, MacDonald et al., 1997, Richman and Regan, 1998). expression (Eason et al., 1994, MacDonald et al., 1997, Richman and Regan, 1998).

Adrenoceptors Adrenoceptors

1-adrenoceptors 2-adrenoceptors -adrenoceptors 1-adrenoceptors 2-adrenoceptors -adrenoceptors

1A 1B 1D 2A 2B 2C 1 2 3 1A 1B 1D 2A 2B 2C 1 2 3

Figure 2. Classification of mammalian adrenoceptors Figure 2. Classification of mammalian adrenoceptors

The rodent (rat and mouse) 2A-adrenoceptor was initially “misleadingly” identified as The rodent (rat and mouse) 2A-adrenoceptor was initially “misleadingly” identified as a fourth 2-adrenoceptor subtype, 2D, because of some differences in its pharmacological a fourth 2-adrenoceptor subtype, 2D, because of some differences in its pharmacological properties compared to the human 2A-adrenoceptor with regard to the classical 2- properties compared to the human 2A-adrenoceptor with regard to the classical 2- adrenoceptor antagonists, yohimbine and rauwolscine (Michel et al., 1989, Lanier et al., adrenoceptor antagonists, yohimbine and rauwolscine (Michel et al., 1989, Lanier et al., 1991, Simonneaux et al., 1991, Bylund et al., 1992). Subsequently, cloning of the mouse 1991, Simonneaux et al., 1991, Bylund et al., 1992). Subsequently, cloning of the mouse 2A-adrenoceptor gene demonstrated that its lower affinity for yohimbine in comparison to 2A-adrenoceptor gene demonstrated that its lower affinity for yohimbine in comparison to the human 2A-adrenoceptor was to be ascribed at least in part to a single cysteine/serine the human 2A-adrenoceptor was to be ascribed at least in part to a single cysteine/serine difference in TM5 (position 5.43) (Link et al., 1992, Cockcroft et al., 2000). difference in TM5 (position 5.43) (Link et al., 1992, Cockcroft et al., 2000). A real fourth 2-adrenoceptor subtype was cloned from zebrafish (Danio rerio), in A real fourth 2-adrenoceptor subtype was cloned from zebrafish (Danio rerio), in which it is present as two duplicates of a gene that has been lost during mammalian which it is present as two duplicates of a gene that has been lost during mammalian evolution. It appears to be present also in many other fish species and in some non- evolution. It appears to be present also in many other fish species and in some non- mammalian tetrapods (Ruuskanen et al., 2004). This receptor was named 2D mammalian tetrapods (Ruuskanen et al., 2004). This receptor was named 2D according to the nomenclature recommendations of IUPHAR (International Union of according to the nomenclature recommendations of IUPHAR (International Union of Basic and Clinical Pharmacology), in spite of the potential for confusion with regard to Basic and Clinical Pharmacology), in spite of the potential for confusion with regard to the erroneously named mouse/rat 2A-adrenoceptor. Based on the conserved syntenies the erroneously named mouse/rat 2A-adrenoceptor. Based on the conserved syntenies and identified molecular fingerprints of the gene sequences from several vertebrate and identified molecular fingerprints of the gene sequences from several vertebrate species, the set of four 2-adrenoceptor subtype genes was postulated to have arisen by species, the set of four 2-adrenoceptor subtype genes was postulated to have arisen by two rounds of duplication of the whole genome in early vertebrate evolution (Vernier two rounds of duplication of the whole genome in early vertebrate evolution (Vernier et al., 1995, Ruuskanen et al., 2004). The presence of the fourth 2-adrenoceptor et al., 1995, Ruuskanen et al., 2004). The presence of the fourth 2-adrenoceptor subtype gene in zebrafish as two duplicates (adra2da and adra2db) is in agreement subtype gene in zebrafish as two duplicates (adra2da and adra2db) is in agreement with the concept that an additional third round of large-scale duplication (~100 million with the concept that an additional third round of large-scale duplication (~100 million years ago) of at least part of the genome occurred in many teleost fish lineages years ago) of at least part of the genome occurred in many teleost fish lineages

16 Review of the Literature 16 Review of the Literature

(Postlethwait et al., 1998, Shin and Fishman, 2002, Ruuskanen et al., 2004). Based on (Postlethwait et al., 1998, Shin and Fishman, 2002, Ruuskanen et al., 2004). Based on this hypothesis, e.g. the pufferfish genome was proposed to contain as many as eight this hypothesis, e.g. the pufferfish genome was proposed to contain as many as eight 2-adrenoceptor subtype genes, making the GPCR subfamily of 2-adrenoceptor 2-adrenoceptor subtype genes, making the GPCR subfamily of 2-adrenoceptor subtypes even more complex (Ruuskanen et al., 2004, Bylund, 2005). So far, no subtypes even more complex (Ruuskanen et al., 2004, Bylund, 2005). So far, no evidence has been published about the expression patterns of these putative 2- evidence has been published about the expression patterns of these putative 2- adrenoceptor subtypes of the pufferfish. adrenoceptor subtypes of the pufferfish.

2.2 Structure-function relationships of 2-adrenoceptors 2.2 Structure-function relationships of 2-adrenoceptors

2.2.12-Adrenoceptor ligands and subtype selectivity 2.2.12-Adrenoceptor ligands and subtype selectivity The naturally occurring catecholamines, noradrenaline and adrenaline and their The naturally occurring catecholamines, noradrenaline and adrenaline and their chemical precursor dopamine, are sympathomimetic “fight-or-flight” hormones chemical precursor dopamine, are sympathomimetic “fight-or-flight” hormones (adrenaline) and neurotransmitters (dopamine and noradrenaline) that are derived from (adrenaline) and neurotransmitters (dopamine and noradrenaline) that are derived from the amino acid, tyrosine. They are rapidly degraded either by methylation by catechol- the amino acid, tyrosine. They are rapidly degraded either by methylation by catechol- O-methyltransferases (COMT) or by deamination by monoamine oxidases (MAO). O-methyltransferases (COMT) or by deamination by monoamine oxidases (MAO). Structurally, catecholamines belong to the chemical class of phenylethylamines, Structurally, catecholamines belong to the chemical class of phenylethylamines, consisting of a catechol moiety, i.e. a benzene ring with two adjacent hydroxyl groups, consisting of a catechol moiety, i.e. a benzene ring with two adjacent hydroxyl groups, and an aliphatic amine side chain (Figure 3). The biosynthesis of endogenous and an aliphatic amine side chain (Figure 3). The biosynthesis of endogenous catecholamines is driven by successive enzymatic reactions; in the case of adrenaline, a catecholamines is driven by successive enzymatic reactions; in the case of adrenaline, a total of four enzymes is involved (Figure 4). Catecholamine synthesis can be inhibited total of four enzymes is involved (Figure 4). Catecholamine synthesis can be inhibited by α-methyl-p-tyrosine (AMPT), which inhibits tyrosine hydroxylase, and has been in by α-methyl-p-tyrosine (AMPT), which inhibits tyrosine hydroxylase, and has been in clinical use in the treatment of pheochromocytomas, catecholamine-producing tumours clinical use in the treatment of pheochromocytomas, catecholamine-producing tumours of the adrenal medulla (Hengstmann et al., 1979). of the adrenal medulla (Hengstmann et al., 1979).

A B A B meta meta -hydroxyl -hydroxyl

OH OH OH OH OH para OH para NH NH NH NH 2 R' 2 R' amino catechol amino catechol group ring group ring

Figure 3. Chemical structures of the phenylethylamine scaffold (A) and (-)-noradrenaline (B). Figure 3. Chemical structures of the phenylethylamine scaffold (A) and (-)-noradrenaline (B). The meta- and para-hydroxyls, the aromatic catechol ring, the protonated amine and the chiral The meta- and para-hydroxyls, the aromatic catechol ring, the protonated amine and the chiral -hydroxyl are indicated. R’ = H in noradrenaline, CH3 in adrenaline. -hydroxyl are indicated. R’ = H in noradrenaline, CH3 in adrenaline.

Adrenaline and noradrenaline are able to activate all adrenoceptor subtypes, which Adrenaline and noradrenaline are able to activate all adrenoceptor subtypes, which limits their usefulness for experimental purposes. Moreover, they share rather similar limits their usefulness for experimental purposes. Moreover, they share rather similar binding affinities and functional properties at all three 2-adrenoceptor subtypes as binding affinities and functional properties at all three 2-adrenoceptor subtypes as well as binding to other adrenoceptors. Therefore, synthetic ligands are more well as binding to other adrenoceptors. Therefore, synthetic ligands are more commonly used when studying 2-adrenoceptors and their subtypes. Several synthetic commonly used when studying 2-adrenoceptors and their subtypes. Several synthetic

Review of the Literature 17 Review of the Literature 17 agonists and antagonists have been developed, with different pharmacological agonists and antagonists have been developed, with different pharmacological properties at 2-adrenoceptors. Structurally, there is a large diversity of compounds, properties at 2-adrenoceptors. Structurally, there is a large diversity of compounds, but they can be grouped into some main classes based on their chemical scaffolds: but they can be grouped into some main classes based on their chemical scaffolds: synthetic phenylethylamines, imidazol(in)es, i.e. compounds that contain an imidazole synthetic phenylethylamines, imidazol(in)es, i.e. compounds that contain an imidazole or imidazoline moiety (e.g. atipamezole, brimonidine, clonidine, medetomidine, or imidazoline moiety (e.g. atipamezole, brimonidine, clonidine, medetomidine, oxymetazoline, idazoxan and RX821002), guanidines (guanabenz and guanfacine), oxymetazoline, idazoxan and RX821002), guanidines (guanabenz and guanfacine), yohimbine derivatives (yohimbine, MK-912 (L657.743), rauwolscine and RS-79948- yohimbine derivatives (yohimbine, MK-912 (L657.743), rauwolscine and RS-79948- 197), tricyclic antipsychotics (chlorpromazine and clozapine), and other “bulky” 197), tricyclic antipsychotics (chlorpromazine and clozapine), and other “bulky” compounds with various larger molecular structures (ARC239, prazosin, spiperone, compounds with various larger molecular structures (ARC239, prazosin, spiperone, spiroxatrine and WB-4101). spiroxatrine and WB-4101).

O O L-Tyrosine L-Tyrosine OH OH

NH2 NH2 OH OH

Tyrosine hydroxylase (TH) Tyrosine hydroxylase (TH)

O O OH L-Dihydroxyphenylalanine OH L-Dihydroxyphenylalanine OH (L-DOPA) OH (L-DOPA) NH2 NH2 OH OH

DOPA decarboxylase DOPA decarboxylase (L-Aromatic amino acid decarboxylase, AADC) (L-Aromatic amino acid decarboxylase, AADC)

OH OH Dopamine Dopamine NH2 NH2 OH OH

Dopamine -hydroxylase (DBH) Dopamine -hydroxylase (DBH)

OH OH OH OH Noradrenaline Noradrenaline NH2 NH2 OH OH Phenylethanolamine Phenylethanolamine N-methyltransferase (PNMT) N-methyltransferase (PNMT)

OH OH OH OH Adrenaline Adrenaline NH NH OH CH OH CH 3 3 Figure 4. Biosynthetic pathway of catecholamines from the amino acid tyrosine to adrenaline. Figure 4. Biosynthetic pathway of catecholamines from the amino acid tyrosine to adrenaline.

Although some synthetic ligands show preference for certain 2-adrenoceptor subtypes Although some synthetic ligands show preference for certain 2-adrenoceptor subtypes (Table 1), none of the current agents is truly 2-adrenoceptor subtype-selective. (Table 1), none of the current agents is truly 2-adrenoceptor subtype-selective. Furthermore, many of them have pharmacological effects at some other GPCRs in Furthermore, many of them have pharmacological effects at some other GPCRs in addition to 2-adrenoceptors, e.g. clozapine and chlorpromazine bind to dopamine addition to 2-adrenoceptors, e.g. clozapine and chlorpromazine bind to dopamine

18 Review of the Literature 18 Review of the Literature receptors and many other GPCRs, oxymetazoline, rauwolscine, spiroxatrine and receptors and many other GPCRs, oxymetazoline, rauwolscine, spiroxatrine and yohimbine bind to serotonin receptors, and prazosin, WB-4101, clonidine and yohimbine bind to serotonin receptors, and prazosin, WB-4101, clonidine and oxymetazoline bind to 1-adrenoceptors (http://www.iuphar-db.org/). With respect to the oxymetazoline bind to 1-adrenoceptors (http://www.iuphar-db.org/). With respect to the imidazoline-based ligands, some of their effects were suggested to be mediated by distinct imidazoline-based ligands, some of their effects were suggested to be mediated by distinct imidazoline-binding sites, instead of 2-adrenoceptors (see e.g. Bousquet et al., 1984). imidazoline-binding sites, instead of 2-adrenoceptors (see e.g. Bousquet et al., 1984). Three types of non-adrenergic imidazoline-binding sites (I1, I2 and I3) have been proposed, Three types of non-adrenergic imidazoline-binding sites (I1, I2 and I3) have been proposed, with different tissue distributions, ligand binding profiles and functions, but molecular with different tissue distributions, ligand binding profiles and functions, but molecular characterisation of these binding proteins as distinct gene products is still lacking characterisation of these binding proteins as distinct gene products is still lacking (Dardonville and Rozas, 2004, Gongadze et al., 2008). They may not be receptors at all in (Dardonville and Rozas, 2004, Gongadze et al., 2008). They may not be receptors at all in the biological sense of the term; indeed, the I2-imidazoline site was actually shown to be the biological sense of the term; indeed, the I2-imidazoline site was actually shown to be identical with the enzyme MAO-B (Sastre and Garcia-Sevilla, 1993). identical with the enzyme MAO-B (Sastre and Garcia-Sevilla, 1993).

Table 1. Binding preferences, functional properties and chemical classification of selected 2- Table 1. Binding preferences, functional properties and chemical classification of selected 2- adrenoceptor ligands. adrenoceptor ligands. Ligand Binding Agonist efficacy Chemical Ligand Binding Agonist efficacy Chemical preference classification preference classification Endogenous ligands Endogenous ligands Adrenaline 2A ≈ 2B ≈ 2C Full Phenylethylamine Adrenaline 2A ≈ 2B ≈ 2C Full Phenylethylamine Dopamine* 2A ≈ 2B ≈ 2C Full: 2A; (n.d. 2B, 2C) Phenylethylamine Dopamine* 2A ≈ 2B ≈ 2C Full: 2A; (n.d. 2B, 2C) Phenylethylamine Noradrenaline 2A ≈ 2B ≈ 2C Full Phenylethylamine Noradrenaline 2A ≈ 2B ≈ 2C Full Phenylethylamine

Synthetic agonists Synthetic agonists 2-Amino-1-phenylethanol* 2A> 2C ≈ 2B Partial Phenylethylamine 2-Amino-1-phenylethanol* 2A> 2C ≈ 2B Partial Phenylethylamine Brimonidine (UK14,304) 2A ≈ 2B > 2C Full: 2A; Partial: 2B, 2C Imidazoline Brimonidine (UK14,304) 2A ≈ 2B > 2C Full: 2A; Partial: 2B, 2C Imidazoline Biphenyline 2A > 2B ≈ 2C Partial Imidazoline Biphenyline 2A > 2B ≈ 2C Partial Imidazoline Clonidine 2A ≈ 2B ≈ 2C Partial Imidazoline Clonidine 2A ≈ 2B ≈ 2C Partial Imidazoline Dexmedetomidine 2A ≈ 2B ≈ 2C Full: 2B; Partial 2A, 2C Imidazole Dexmedetomidine 2A ≈ 2B ≈ 2C Full: 2B; Partial 2A, 2C Imidazole Guanfacine 2A >> 2B ≈ 2C Partial Guanidine Guanfacine 2A >> 2B ≈ 2C Partial Guanidine Oxymetazoline 2A >>> 2C > 2B Partial Imidazoline Oxymetazoline 2A >>> 2C > 2B Partial Imidazoline

Synthetic antagonists Synthetic antagonists ARC239 2B ≈ 2C >> 2A - Piperazine ARC239 2B ≈ 2C >> 2A - Piperazine Atipamezole 2A ≈ 2B ≈ 2C - Imidazole Atipamezole 2A ≈ 2B ≈ 2C - Imidazole BRL44408 2A >>> 2B ≈ 2C - Imidazole BRL44408 2A >>> 2B ≈ 2C - Imidazole Chlorpromazine 2B > 2C >> 2C - Phenothiazine Chlorpromazine 2B > 2C >> 2C - Phenothiazine Clozapine* 2C >> 2A ≈ 2B - Dibenzazepine Clozapine* 2C >> 2A ≈ 2B - Dibenzazepine Idazoxan 2A ≈ 2B ≈ 2C - Imidazoline Idazoxan 2A ≈ 2B ≈ 2C - Imidazoline L657.743 (MK912) 2C >> 2A ≈ 2B - Quinazoline L657.743 (MK912) 2C >> 2A ≈ 2B - Quinazoline Prazosin 2B ≈ 2C > 2A - Quinazoline Prazosin 2B ≈ 2C > 2A - Quinazoline Phentolamine 2A ≈ 2B > 2C - Imidazoline Phentolamine 2A ≈ 2B > 2C - Imidazoline Rauwolscine 2A ≈ 2B ≈ 2C - Yohimban Rauwolscine 2A ≈ 2B ≈ 2C - Yohimban RS-79948-197 2A ≈ 2B ≈ 2C - Decahydronaphthyridine RS-79948-197 2A ≈ 2B ≈ 2C - Decahydronaphthyridine Tetrahydroisoquinoline Tetrahydroisoquinoline RX821002 2A ≈ 2B ≈ 2C - Imidazoline RX821002 2A ≈ 2B ≈ 2C - Imidazoline Spiperone* 2B ≈ 2C >>> 2A  - Butyrophenone Spiperone* 2B ≈ 2C >>> 2A  - Butyrophenone Spiroxatrine 2B ≈ 2C >>> 2A  - Benzodioxane Spiroxatrine 2B ≈ 2C >>> 2A  - Benzodioxane WB4101 2B > 2A ≈ 2C - Benzodioxane WB4101 2B > 2A ≈ 2C - Benzodioxane Yohimbine 2A ≈ 2B ≈ 2C - Yohimban Yohimbine 2A ≈ 2B ≈ 2C - Yohimban Derived from Lomasney et al., 1991, Bylund et al., 1992, Devedjian et al., 1994, Uhlèn et al., Derived from Lomasney et al., 1991, Bylund et al., 1992, Devedjian et al., 1994, Uhlèn et al., 1994, Jasper et al., 1998, Peltonen et al., 1998, Peltonen et al., 2003. *Based on the binding 1994, Jasper et al., 1998, Peltonen et al., 1998, Peltonen et al., 2003. *Based on the binding affinities shown in papers I and III. affinities shown in papers I and III.

Review of the Literature 19 Review of the Literature 19

2.2.2 Experimental probing of receptor structure and ligand interactions 2.2.2 Experimental probing of receptor structure and ligand interactions The binding sites of monoamine-activated GPCRs are accessible to small cationic The binding sites of monoamine-activated GPCRs are accessible to small cationic ligands such as adrenaline and noradrenaline, and are located within the core of the ligands such as adrenaline and noradrenaline, and are located within the core of the receptor proteins formed of the seven α-helical TM domains. The general structural receptor proteins formed of the seven α-helical TM domains. The general structural architecture of GPCRs, the orientation of the TM helices and the location of the architecture of GPCRs, the orientation of the TM helices and the location of the binding site were confirmed already some years before the bovine rhodopsin 3D binding site were confirmed already some years before the bovine rhodopsin 3D structure was resolved by X-ray crystallography, by using molecular imaging structure was resolved by X-ray crystallography, by using molecular imaging techniques such as cryo-electron microscopy of bacteriorhodopsin (Henderson et al., techniques such as cryo-electron microscopy of bacteriorhodopsin (Henderson et al., 1990) and frog rhodopsin (Schertler and Hargrave, 1995) and electron paramagnetic 1990) and frog rhodopsin (Schertler and Hargrave, 1995) and electron paramagnetic resonance spectroscopy of rhodopsin (Altenbach et al., 1994). Even though the resonance spectroscopy of rhodopsin (Altenbach et al., 1994). Even though the bacterial light-sensing protein bacteriorhodopsin is not a GPCR, as it does not signal bacterial light-sensing protein bacteriorhodopsin is not a GPCR, as it does not signal through G-proteins and shares no sequence identity with GPCRs, it is considered to through G-proteins and shares no sequence identity with GPCRs, it is considered to have structural similarities with rhodopsin and other GPCRs. Based on the structures of have structural similarities with rhodopsin and other GPCRs. Based on the structures of bacteriorhodopsin and rhodopsin, the ligand binding site of GPCRs was proposed to be bacteriorhodopsin and rhodopsin, the ligand binding site of GPCRs was proposed to be contained between the TM domains, in a fold or cavity extending from the extracellular contained between the TM domains, in a fold or cavity extending from the extracellular surface of the proteins into their transmembrane core, where the surface of the binding surface of the proteins into their transmembrane core, where the surface of the binding cavity is formed by residues that constitute contact sites for the recognition and binding cavity is formed by residues that constitute contact sites for the recognition and binding of specific ligands. This has been demonstrated for retinal in the case of rhodopsin and of specific ligands. This has been demonstrated for retinal in the case of rhodopsin and for other chromophores in the case of bacteriorhodopsin (Henderson and Schertler, for other chromophores in the case of bacteriorhodopsin (Henderson and Schertler, 1990). However, in contrast to other GPCRs which provide an open access for entrance 1990). However, in contrast to other GPCRs which provide an open access for entrance of diffusible ligands, the retinal binding pocket is closed in rhodopsin to protect retinal of diffusible ligands, the retinal binding pocket is closed in rhodopsin to protect retinal from solvent molecules (Hildebrand et al., 2009). The present understanding of from solvent molecules (Hildebrand et al., 2009). The present understanding of rhodopsin-like GPCR structures is illustrated below in a comparison of the model rhodopsin-like GPCR structures is illustrated below in a comparison of the model structures of the human 2A-adrenoceptor based on the crystal structures of bovine structures of the human 2A-adrenoceptor based on the crystal structures of bovine rhodopsin and the human 2-adrenoceptor (Figure 5). rhodopsin and the human 2-adrenoceptor (Figure 5).

TM1 TM7 TM6 TM1 TM7 TM6

TM2 TM5 TM2 TM5

TM3 TM4 TM3 TM4

Figure 5. Molecular model and the binding cavity of the human 2A-adrenoceptor viewed from Figure 5. Molecular model and the binding cavity of the human 2A-adrenoceptor viewed from the extracellular side of the cell membrane, based on both the bovine rhodopsin (green) and the extracellular side of the cell membrane, based on both the bovine rhodopsin (green) and human 2-adrenoceptor structures (blue). The surface of the binding cavity is formed by amino human 2-adrenoceptor structures (blue). The surface of the binding cavity is formed by amino acids of the TM helices. Here, only some key amino acids are shown. Modified from paper III, acids of the TM helices. Here, only some key amino acids are shown. Modified from paper III, by courtesy of Henri Xhaard. by courtesy of Henri Xhaard.

20 Review of the Literature 20 Review of the Literature

Over the past 25 years, several experimental approaches have been developed and Over the past 25 years, several experimental approaches have been developed and applied to generate the current knowledge of the structural determinants that contribute applied to generate the current knowledge of the structural determinants that contribute to the specific ligand binding of various GPCRs. Site-directed mutagenesis combined to the specific ligand binding of various GPCRs. Site-directed mutagenesis combined with ligand binding experiments has been an important method for mapping the with ligand binding experiments has been an important method for mapping the molecular interactions between ligands and receptors. For example, receptor binding molecular interactions between ligands and receptors. For example, receptor binding assays performed with asymmetric ligand enantiomers or analogues with different assays performed with asymmetric ligand enantiomers or analogues with different functional groups combined with amino acid substitutions in the receptor proteins have functional groups combined with amino acid substitutions in the receptor proteins have been useful in probing the catecholamine binding site of 2-adrenoceptors at the atomic been useful in probing the catecholamine binding site of 2-adrenoceptors at the atomic level (Salminen et al., 1999, Nyrönen et al., 2001, Peltonen et al., 2003). Another level (Salminen et al., 1999, Nyrönen et al., 2001, Peltonen et al., 2003). Another mutagenesis-based method, termed the Substituted Cysteine Accessibility Method mutagenesis-based method, termed the Substituted Cysteine Accessibility Method (SCAM), has yielded information on the amino acids that contribute to the surface of (SCAM), has yielded information on the amino acids that contribute to the surface of the binding cavity. In this approach, candidate residues are mutated to cysteines, one the binding cavity. In this approach, candidate residues are mutated to cysteines, one by one, and residues facing the water-accessible binding cavity are observed to react by one, and residues facing the water-accessible binding cavity are observed to react much faster with alkylating sulphydryl-specific reagents than other residues facing the much faster with alkylating sulphydryl-specific reagents than other residues facing the interior of the receptor protein or the phospholipid bilayer of the cell membrane interior of the receptor protein or the phospholipid bilayer of the cell membrane (Liapakis et al., 2001). Several irreversible sulphydryl-reactive probes with different (Liapakis et al., 2001). Several irreversible sulphydryl-reactive probes with different pharmacological properties at 2-adrenoceptors have been employed for this purpose. pharmacological properties at 2-adrenoceptors have been employed for this purpose. These include several ligands whose covalent binding properties were initially These include several ligands whose covalent binding properties were initially identified in other contexts, e.g. N-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinole identified in other contexts, e.g. N-ethoxycarbonyl-2-ethoxy-1,2-dihydroquinole (EEDQ) (Garvey et al., 1972), benextramine (Melchiorre and Gallucci, 1983), (EEDQ) (Garvey et al., 1972), benextramine (Melchiorre and Gallucci, 1983), phenoxybenzamine (Regan et al., 1984, Frang et al., 2001), phenylmercuric chloride phenoxybenzamine (Regan et al., 1984, Frang et al., 2001), phenylmercuric chloride (Regan et al., 1986), pyrextramine (Brasili et al., 1986) and chloroethylclonidine (Regan et al., 1986), pyrextramine (Brasili et al., 1986) and chloroethylclonidine (Bultmann and Starke, 1993, Marjamäki et al., 1998). Some of these compounds were (Bultmann and Starke, 1993, Marjamäki et al., 1998). Some of these compounds were used in the initial characterization of 2-adrenoceptor structures. For example, used in the initial characterization of 2-adrenoceptor structures. For example, chloroethylclonidine and phenoxybenzamine were used in SCAM experiments to chloroethylclonidine and phenoxybenzamine were used in SCAM experiments to elucidate the critical structural determinants of TM3 and TM5 that are exposed in the elucidate the critical structural determinants of TM3 and TM5 that are exposed in the 2-adrenoceptor ligand binding pocket and contribute to receptor activation 2-adrenoceptor ligand binding pocket and contribute to receptor activation (Marjamäki et al., 1998, Marjamäki et al., 1999, Frang et al., 2001). (Marjamäki et al., 1998, Marjamäki et al., 1999, Frang et al., 2001). Although advances in gene manipulation technologies have allowed the mapping of Although advances in gene manipulation technologies have allowed the mapping of structural determinants important for ligand binding, caution is warranted when structural determinants important for ligand binding, caution is warranted when interpreting results obtained with mutated receptors, as the observed alteration of a interpreting results obtained with mutated receptors, as the observed alteration of a receptor’s pharmacological profile is not always a direct consequence of changes in the receptor’s pharmacological profile is not always a direct consequence of changes in the contact between ligand and receptor. Mutations can also disrupt the overall structure of contact between ligand and receptor. Mutations can also disrupt the overall structure of the receptors, cause conformational rearrangements, posttranslational modifications or the receptors, cause conformational rearrangements, posttranslational modifications or disturb cell surface expression, which all may have indirect effects on ligand binding. In disturb cell surface expression, which all may have indirect effects on ligand binding. In addition, differences in experimental conditions may have important effects e.g. on the addition, differences in experimental conditions may have important effects e.g. on the apparent binding affinity (Deupree et al., 1996). As a consequence, it may be very apparent binding affinity (Deupree et al., 1996). As a consequence, it may be very difficult to differentiate direct from indirect effects of experimental mutations on ligand difficult to differentiate direct from indirect effects of experimental mutations on ligand binding. binding.

2.2.3 Structural determinants of 2-adrenoceptor ligand recognition and binding 2.2.3 Structural determinants of 2-adrenoceptor ligand recognition and binding Cell membrane receptors recognize ligand molecules present in the extracellular Cell membrane receptors recognize ligand molecules present in the extracellular environment and transmit messages from the extracellular space to the intracellular environment and transmit messages from the extracellular space to the intracellular components of the cellular signal transduction pathways. Receptor-ligand interactions components of the cellular signal transduction pathways. Receptor-ligand interactions may be highly specific, because each receptor type has unique structural elements for may be highly specific, because each receptor type has unique structural elements for

Review of the Literature 21 Review of the Literature 21 ligand recognition and binding, which have developed during the course of evolution. ligand recognition and binding, which have developed during the course of evolution. For 2-adrenoceptors and other monoamine receptors, the ligand binding cavity is For 2-adrenoceptors and other monoamine receptors, the ligand binding cavity is thought to be mainly formed by residues in TM3, TM5, TM6, TM7 and the second thought to be mainly formed by residues in TM3, TM5, TM6, TM7 and the second extracellular loop (XL2). Experimental studies, for example on alkylating ligand extracellular loop (XL2). Experimental studies, for example on alkylating ligand derivatives targeted to specific residues in the proposed binding site (Marjamäki et al., derivatives targeted to specific residues in the proposed binding site (Marjamäki et al., 1999, Salminen et al., 1999, Frang et al., 2001), site-directed mutagenesis studies 1999, Salminen et al., 1999, Frang et al., 2001), site-directed mutagenesis studies (Wang et al., 1991), and competitive binding experiments where labelled ligands are (Wang et al., 1991), and competitive binding experiments where labelled ligands are displaced by other non-labelled ligands, also in terms of asymmetric (chiral) displaced by other non-labelled ligands, also in terms of asymmetric (chiral) compounds (Nyrönen et al., 2001, Peltonen et al., 2003), have revealed several amino compounds (Nyrönen et al., 2001, Peltonen et al., 2003), have revealed several amino acids that appear to play important roles in ligand binding or recognition in 2- acids that appear to play important roles in ligand binding or recognition in 2- adrenoceptors (Figure 6). A highly conserved aspartate present in all rhodopsin-like adrenoceptors (Figure 6). A highly conserved aspartate present in all rhodopsin-like GPCRs in the top portion of TM3 at position 3.32 and its negatively charged side-chain GPCRs in the top portion of TM3 at position 3.32 and its negatively charged side-chain carboxyl group have in several studies been shown to provide an important anchoring carboxyl group have in several studies been shown to provide an important anchoring point for ligands, providing direct polar ion-pair interactions for compounds that point for ligands, providing direct polar ion-pair interactions for compounds that contain a positively charged amine or imine moiety (Wang et al., 1991, Nyrönen et al., contain a positively charged amine or imine moiety (Wang et al., 1991, Nyrönen et al., 2001, Xhaard et al., 2006). Additionally, one of the two side-chain oxygens of the 2001, Xhaard et al., 2006). Additionally, one of the two side-chain oxygens of the aspartate at position 3.32 has been hypothesized to provide a hydrogen bonding site for aspartate at position 3.32 has been hypothesized to provide a hydrogen bonding site for the -hydroxyl group of the R-enantiomers of catecholamine ligands, whereas the S- the -hydroxyl group of the R-enantiomers of catecholamine ligands, whereas the S- enantiomers of catecholamines cannot undergo a similar interaction. Thus, the enantiomers of catecholamines cannot undergo a similar interaction. Thus, the affinities of S-enantiomers are much weaker than those of the R-enantiomers. affinities of S-enantiomers are much weaker than those of the R-enantiomers. Furthermore, the clearly lower affinity of dopamine at 2-adrenoceptors in comparison Furthermore, the clearly lower affinity of dopamine at 2-adrenoceptors in comparison to noradrenaline was interpreted to be a consequence of dopamine lacking the - to noradrenaline was interpreted to be a consequence of dopamine lacking the - hydroxyl moiety (Nyrönen et al., 2001). Other residues believed to interact with the hydroxyl moiety (Nyrönen et al., 2001). Other residues believed to interact with the positively charged N-methyl group of the catecholamine ligands via hydrophobic positively charged N-methyl group of the catecholamine ligands via hydrophobic contacts are two phenylalanines at positions 7.38 and 7.39 in TM7 (Nyrönen et al., contacts are two phenylalanines at positions 7.38 and 7.39 in TM7 (Nyrönen et al., 2001). Two serine residues in TM5 were shown to provide hydrogen bonding sites for 2001). Two serine residues in TM5 were shown to provide hydrogen bonding sites for the two catecholic hydroxyl groups of the catecholamines with the meta-OH probably the two catecholic hydroxyl groups of the catecholamines with the meta-OH probably interacting with the serine in position 5.42 and the para-OH with the serine in position interacting with the serine in position 5.42 and the para-OH with the serine in position 5.46 of the human 2A-adrenoceptor (Marjamäki et al., 1998, Rudling et al., 1999, 5.46 of the human 2A-adrenoceptor (Marjamäki et al., 1998, Rudling et al., 1999, Salminen et al., 1999). These interactions have also been postulated to be influenced Salminen et al., 1999). These interactions have also been postulated to be influenced by the conformational changes that occur upon receptor activation, e.g. rotation of by the conformational changes that occur upon receptor activation, e.g. rotation of TM5, and in that way to play a significant role in agonist-dependent receptor activation TM5, and in that way to play a significant role in agonist-dependent receptor activation (Nyrönen et al., 2001, Peltonen et al., 2003). In addition, they are important for the (Nyrönen et al., 2001, Peltonen et al., 2003). In addition, they are important for the orientation of ligands within the binding pocket. The side chains of a threonine at orientation of ligands within the binding pocket. The side chains of a threonine at position 3.37 in TM3 and a cysteine at position 5.43 in TM5 of the human 2A- position 3.37 in TM3 and a cysteine at position 5.43 in TM5 of the human 2A- adrenoceptor have also been shown to interact with the catecholic OH groups, but in adrenoceptor have also been shown to interact with the catecholic OH groups, but in contrast to the serines at 5.42 and 5.46, they were not observed to play a role in contrast to the serines at 5.42 and 5.46, they were not observed to play a role in receptor activation (Nyrönen et al., 2001, Peltonen et al., 2003). Other residues of the receptor activation (Nyrönen et al., 2001, Peltonen et al., 2003). Other residues of the 2A-adrenoceptor that are likely to interact through a - stacking interaction with the 2A-adrenoceptor that are likely to interact through a - stacking interaction with the aromatic ring of phenylethylamine-type ligands include a conserved valine at position aromatic ring of phenylethylamine-type ligands include a conserved valine at position 3.33 in TM3, the conserved aromatic side chain of the phenylalanine at position 5.47 in 3.33 in TM3, the conserved aromatic side chain of the phenylalanine at position 5.47 in TM5, the conserved aromatic side chains of a tryptophan at position 6.48, TM5, the conserved aromatic side chains of a tryptophan at position 6.48, phenylalanines at positions 6.51 and 6.52, and a tyrosine at position 6.55 in TM6 phenylalanines at positions 6.51 and 6.52, and a tyrosine at position 6.55 in TM6 (Nyrönen et al., 2001, Peltonen et al., 2003, Gentili et al., 2004). In docking (Nyrönen et al., 2001, Peltonen et al., 2003, Gentili et al., 2004). In docking simulations, the N-methyl group of adrenaline-like phenylethylamines was proposed to simulations, the N-methyl group of adrenaline-like phenylethylamines was proposed to become packed against aromatic side chains of phenylalanines at positions 7.38 and become packed against aromatic side chains of phenylalanines at positions 7.38 and

22 Review of the Literature 22 Review of the Literature

7.39 in TM7, but experimental support for this proposal is still lacking (Nyrönen et al., 7.39 in TM7, but experimental support for this proposal is still lacking (Nyrönen et al., 2001). Previously, also serines at positions 2.61 in TM2 and 7.46 in TM7 were claimed 2001). Previously, also serines at positions 2.61 in TM2 and 7.46 in TM7 were claimed to be involved in the stereoselectivity of the 2A-adrenoceptor for the (-)- or R- to be involved in the stereoselectivity of the 2A-adrenoceptor for the (-)- or R- enantiomers of catecholamine agonists, and in addition to play a role in the attachment enantiomers of catecholamine agonists, and in addition to play a role in the attachment of the -hydroxyl groups of the catecholamines to 2A-adrenoceptors (Hehr et al., of the -hydroxyl groups of the catecholamines to 2A-adrenoceptors (Hehr et al., 1997, Hieble et al., 1998). However, the current understanding of the molecular 1997, Hieble et al., 1998). However, the current understanding of the molecular structure of 2A-adrenoceptors points to some indirect effects being involved in these structure of 2A-adrenoceptors points to some indirect effects being involved in these findings rather than specific side-chain interactions between catecholamines and the findings rather than specific side-chain interactions between catecholamines and the receptor. receptor.

N TM6 N TM6

S5.42, C5.43 TM5 S5.42, C5.43 TM5 F6.52, Y6.55 F6.52, Y6.55 TM7 F5.47 TM7 F5.47 S5.46 S5.46 F7.38, F7.39 F7.38, F7.39 TM1 TM1 C C TM4 TM4 D3.32 D3.32 TM2 TM2 TM3 TM3

Figure 6. Major interactions between the agonist adrenaline and the human 2A-adrenoceptor. Figure 6. Major interactions between the agonist adrenaline and the human 2A-adrenoceptor. Transmembrane domains are shown as cylinders, extracellular loops and the N-terminus as Transmembrane domains are shown as cylinders, extracellular loops and the N-terminus as black lines, and intracellular loops as gray lines. black lines, and intracellular loops as gray lines.

In contrast to agonists, the binding of antagonists to 2-adrenoceptors is less well In contrast to agonists, the binding of antagonists to 2-adrenoceptors is less well understood. The chemical diversity of antagonists that bind to 2-adrenoceptors is understood. The chemical diversity of antagonists that bind to 2-adrenoceptors is much greater than that of agonists, indicating more divergent and complex modes of much greater than that of agonists, indicating more divergent and complex modes of binding. No general “antagonist binding site” has so far been reported for binding. No general “antagonist binding site” has so far been reported for adrenoceptors or any other types of GPCRs. However, in monoamine-binding adrenoceptors or any other types of GPCRs. However, in monoamine-binding GPCRs, including the adrenoceptors, antagonists are proposed to bind at least in part GPCRs, including the adrenoceptors, antagonists are proposed to bind at least in part within the same orthosteric binding site as agonists. This proposal is based on the within the same orthosteric binding site as agonists. This proposal is based on the markedly reduced binding affinity of antagonists when the receptors have been markedly reduced binding affinity of antagonists when the receptors have been mutated at the highly conserved aspartate at 3.32 in TM3 (reviewed by Shi and mutated at the highly conserved aspartate at 3.32 in TM3 (reviewed by Shi and Javitch, 2002). For 2-adrenoceptors, this was demonstrated by the failure of Javitch, 2002). For 2-adrenoceptors, this was demonstrated by the failure of yohimbine to bind to a D3.32N-mutated 2A-adrenoceptor (Wang et al., 1991), yohimbine to bind to a D3.32N-mutated 2A-adrenoceptor (Wang et al., 1991), although possible indirect mechanisms were not ruled out in that study. This although possible indirect mechanisms were not ruled out in that study. This observation was proposed to support a general concept where an ion-pair interaction observation was proposed to support a general concept where an ion-pair interaction with the aspartate at position 3.32 was a key feature also for antagonist binding. with the aspartate at position 3.32 was a key feature also for antagonist binding. Nonetheless, some conflicting reports also exist, e.g. for a D3.32N mutant of the type Nonetheless, some conflicting reports also exist, e.g. for a D3.32N mutant of the type 1A serotonin receptor (5-HT1A), and D3.32A and D3.32K mutants of the 1B- 1A serotonin receptor (5-HT1A), and D3.32A and D3.32K mutants of the 1B-

Review of the Literature 23 Review of the Literature 23 adrenoceptor, where the binding of antagonists was shown to be similar to that of the adrenoceptor, where the binding of antagonists was shown to be similar to that of the corresponding wild-type receptors, while agonist binding was impaired (Ho et al., corresponding wild-type receptors, while agonist binding was impaired (Ho et al., 1992, Porter et al., 1996). Other residues believed to be involved in antagonist 1992, Porter et al., 1996). Other residues believed to be involved in antagonist binding are a cysteine at position 3.36 that was found to interact with covalently binding are a cysteine at position 3.36 that was found to interact with covalently bound phenoxybenzamine, and which is also important for binding of imidazol(in)e bound phenoxybenzamine, and which is also important for binding of imidazol(in)e based ligands at the human 2A-adrenoceptor (Frang et al., 2001). A phenylalanine at based ligands at the human 2A-adrenoceptor (Frang et al., 2001). A phenylalanine at position 7.39 was also found to be critical for the binding of yohimbine to the 2A- position 7.39 was also found to be critical for the binding of yohimbine to the 2A- adrenoceptor (Suryanarayana et al., 1991). adrenoceptor (Suryanarayana et al., 1991). In addition to differences in the amino acids that line the binding cavity, the size of In addition to differences in the amino acids that line the binding cavity, the size of the binding pocket was also suggested to differ among the 2-adrenoceptor subtypes as the binding pocket was also suggested to differ among the 2-adrenoceptor subtypes as shown in a comparative molecular modelling study. The 2B-adrenoceptor was found shown in a comparative molecular modelling study. The 2B-adrenoceptor was found to contain the largest and the 2C-adrenoceptor the smallest cavity (Balogh et al., to contain the largest and the 2C-adrenoceptor the smallest cavity (Balogh et al., 2009). Differences in binding pocket volume may also have a role in the subtype 2009). Differences in binding pocket volume may also have a role in the subtype selectivity of some ligands. selectivity of some ligands. All adrenoceptors are proposed to have overall tertiary structures similar to All adrenoceptors are proposed to have overall tertiary structures similar to rhodopsin, but rhodopsin shares only 20-23 % sequence identity in its TM regions rhodopsin, but rhodopsin shares only 20-23 % sequence identity in its TM regions with members of the monoamine GPCR subfamily. Recently, the structures of three with members of the monoamine GPCR subfamily. Recently, the structures of three much closer homologues of the 2-adrenoceptors, namely the human 2-adrenoceptor much closer homologues of the 2-adrenoceptors, namely the human 2-adrenoceptor (Cherezov et al., 2007), the turkey 1-adrenoceptor (Warne et al., 2008) and the (Cherezov et al., 2007), the turkey 1-adrenoceptor (Warne et al., 2008) and the human dopamine D3 receptor (Chien et al., 2010), were published. Of these, the - human dopamine D3 receptor (Chien et al., 2010), were published. Of these, the - adrenoceptors share, on the average, 37-43 % identical aligned amino acids in their adrenoceptors share, on the average, 37-43 % identical aligned amino acids in their TM regions with the human 2-adrenoceptor subtypes and may thus serve as more TM regions with the human 2-adrenoceptor subtypes and may thus serve as more accurate templates for 2-adrenoceptor models than rhodopsin. A common feature of accurate templates for 2-adrenoceptor models than rhodopsin. A common feature of all rhodopsin-like GPCR crystal structures is the involvement of XL2 in ligand all rhodopsin-like GPCR crystal structures is the involvement of XL2 in ligand binding. In the crystal structure of bovine rhodopsin, XL2 folds as a -hairpin, binding. In the crystal structure of bovine rhodopsin, XL2 folds as a -hairpin, composed of two four-residue -strands, and it dives down into the pocket between composed of two four-residue -strands, and it dives down into the pocket between the TM helices, forming an aromatic “lid” over the binding cavity, and it is also in the TM helices, forming an aromatic “lid” over the binding cavity, and it is also in direct contact with covalently bound 11-cis-retinal (Palczewski et al., 2000, Teller et direct contact with covalently bound 11-cis-retinal (Palczewski et al., 2000, Teller et al., 2001, Okada et al., 2002). A single 7 Å disulphide bond connects XL2 (Cxl2.50) al., 2001, Okada et al., 2002). A single 7 Å disulphide bond connects XL2 (Cxl2.50) to TM3 (C3.25) and constrains the position of XL2 in rhodopsin. This feature is to TM3 (C3.25) and constrains the position of XL2 in rhodopsin. This feature is thought to be highly conserved across all rhodopsin-like GPCRs, and is also seen in thought to be highly conserved across all rhodopsin-like GPCRs, and is also seen in the crystal structures of the -adrenoceptors as well as in the human dopamine D3 the crystal structures of the -adrenoceptors as well as in the human dopamine D3 receptor and A2A-adenosine receptor structures (Cherezov et al., 2007, Rasmussen et receptor and A2A-adenosine receptor structures (Cherezov et al., 2007, Rasmussen et al., 2007, Jaakola et al., 2008, Warne et al., 2008, Chien et al., 2010). In spite of this al., 2007, Jaakola et al., 2008, Warne et al., 2008, Chien et al., 2010). In spite of this conserved disulphide bridge, many significant structural differences are also apparent conserved disulphide bridge, many significant structural differences are also apparent in the extracellular loops of the solved X-ray structures, as in contrast to rhodopsin, in the extracellular loops of the solved X-ray structures, as in contrast to rhodopsin, in the human 2-adrenoceptor structure (and also in the turkey 1-adrenoceptor in the human 2-adrenoceptor structure (and also in the turkey 1-adrenoceptor structure) XL2 is partly folded as an -helix, and is further stabilised by an structure) XL2 is partly folded as an -helix, and is further stabilised by an additional internal loop formed between cysteines Cxl2.43 and Cxl2.49 (Cherezov et additional internal loop formed between cysteines Cxl2.43 and Cxl2.49 (Cherezov et al., 2007). Although the conformation of XL2 in the 2-adrenoceptor is more rigid al., 2007). Although the conformation of XL2 in the 2-adrenoceptor is more rigid than in rhodopsin, it is believed to lie above the binding cavity and is in direct than in rhodopsin, it is believed to lie above the binding cavity and is in direct contact with the partial inverse agonist carazolol, and is thought to play an important contact with the partial inverse agonist carazolol, and is thought to play an important role in ligand binding by stabilizing the core of the receptor (Bokoch et al., 2010). role in ligand binding by stabilizing the core of the receptor (Bokoch et al., 2010). Also in the dopamine D2 receptor which is closely related to the 2-adrenoceptors, Also in the dopamine D2 receptor which is closely related to the 2-adrenoceptors, XL2 has been shown to be accessible to the binding cavity, as evidenced with SCAM XL2 has been shown to be accessible to the binding cavity, as evidenced with SCAM and sulphydryl-reactive probes (Shi and Javitch, 2004). In the 2A-adrenoceptor, XL2 and sulphydryl-reactive probes (Shi and Javitch, 2004). In the 2A-adrenoceptor, XL2

24 Review of the Literature 24 Review of the Literature was first implicated in antagonist binding by using a chimaeric receptor approach. was first implicated in antagonist binding by using a chimaeric receptor approach. When the region from TM3 to TM5 of the mouse 2A-adrenoceptor was replaced When the region from TM3 to TM5 of the mouse 2A-adrenoceptor was replaced with the corresponding human sequence, the binding affinity of yohimbine became with the corresponding human sequence, the binding affinity of yohimbine became closer to that seen with the human 2A-adrenoceptor (with a cysteine in position closer to that seen with the human 2A-adrenoceptor (with a cysteine in position 5.43) than to the wild-type mouse receptor (with a serine in position 5.43) or to a 5.43) than to the wild-type mouse receptor (with a serine in position 5.43) or to a S5.43C substituted mouse 2A-adrenoceptor (Link et al., 1992). In this chimaeric S5.43C substituted mouse 2A-adrenoceptor (Link et al., 1992). In this chimaeric receptor, there were no other sequence differences facing the orthosteric binding receptor, there were no other sequence differences facing the orthosteric binding cavity except for the differences within XL2 and in position 5.43, suggesting that cavity except for the differences within XL2 and in position 5.43, suggesting that XL2 is participating in antagonist binding. In the 1A- and 1B-adrenoceptors and the XL2 is participating in antagonist binding. In the 1A- and 1B-adrenoceptors and the type-1D serotonin receptor (5-HT1D) that are closely related to the 2-adrenoceptors, type-1D serotonin receptor (5-HT1D) that are closely related to the 2-adrenoceptors, site-directed mutagenesis has also revealed XL2 to affect the binding properties of site-directed mutagenesis has also revealed XL2 to affect the binding properties of antagonists (Zhao et al., 1996, Wurch et al., 1998). The predicted position of XL2 antagonists (Zhao et al., 1996, Wurch et al., 1998). The predicted position of XL2 above the binding cavity in the two model structures of the human 2A-adrenoceptor above the binding cavity in the two model structures of the human 2A-adrenoceptor is shown in Figure 7. is shown in Figure 7.

TM7 TM6 TM7 TM6

TM1 TM1 TM5 TM5

TM2 TM2 TM3 TM3

TM4 TM4

Figure 7. Molecular models (viewed in the plane of the cell membrane from the extracellular Figure 7. Molecular models (viewed in the plane of the cell membrane from the extracellular side) of the human 2A-adrenoceptor and the position of XL2, based on either the bovine side) of the human 2A-adrenoceptor and the position of XL2, based on either the bovine rhodopsin (blue) or the human -adrenoceptor (green) structure (Palczewski et al., 2000, rhodopsin (blue) or the human -adrenoceptor (green) structure (Palczewski et al., 2000, Cherezov et al., 2007). The folding of XL2 is the most marked difference in the two model Cherezov et al., 2007). The folding of XL2 is the most marked difference in the two model structures. The rhodopsin-based receptor model contains two parallel -sheets folded similarly structures. The rhodopsin-based receptor model contains two parallel -sheets folded similarly as in the bovine rhodopsin X-ray structure. The additional a-helix seen in the -adrenoceptor as in the bovine rhodopsin X-ray structure. The additional a-helix seen in the -adrenoceptor X-ray structure is not probable in the -adrenoceptor structure. Modified from paper III, X-ray structure is not probable in the -adrenoceptor structure. Modified from paper III, courtesy of Henri Xhaard. courtesy of Henri Xhaard.

The greatest extent of diversity in the amino acid compositions of the 2- The greatest extent of diversity in the amino acid compositions of the 2- adrenoceptor subtypes is present in the amino terminus, the third intracellular loop, and adrenoceptor subtypes is present in the amino terminus, the third intracellular loop, and the carboxyl terminus (Lomasney et al., 1990). The long third intracellular loop, the carboxyl terminus (Lomasney et al., 1990). The long third intracellular loop, together with other cytoplasmic domains, has been shown to be involved in G-protein together with other cytoplasmic domains, has been shown to be involved in G-protein coupling and in receptor desensitisation (Eason and Liggett, 1995, Small et al., 2000b, coupling and in receptor desensitisation (Eason and Liggett, 1995, Small et al., 2000b, Small et al., 2001, Jaakola et al., 2005). Truncation of the third intracellular loop in the Small et al., 2001, Jaakola et al., 2005). Truncation of the third intracellular loop in the human 2B-adrenoceptor altered the binding affinity of some agonist ligands, human 2B-adrenoceptor altered the binding affinity of some agonist ligands, indicating that the conformation of the third intracellular loop may also affect agonist indicating that the conformation of the third intracellular loop may also affect agonist binding to the receptor (Jaakola et al., 2005). The extracellular amino-terminal segment binding to the receptor (Jaakola et al., 2005). The extracellular amino-terminal segment

Review of the Literature 25 Review of the Literature 25 of the 2A- and 2C-adrenoceptors, in contrast to the 2B-adrenoceptor, contains two of the 2A- and 2C-adrenoceptors, in contrast to the 2B-adrenoceptor, contains two consensus sites for asparagine-linked glycosylation (Lomasney et al., 1991). The exact consensus sites for asparagine-linked glycosylation (Lomasney et al., 1991). The exact functional role of amino-terminal glycosylation is still poorly understood but it is likely functional role of amino-terminal glycosylation is still poorly understood but it is likely to improve overall receptor expression rather than affecting the intracellular trafficking to improve overall receptor expression rather than affecting the intracellular trafficking characteristics of the receptors (Keefer et al., 1994, Deslauriers et al., 1999, Angelotti characteristics of the receptors (Keefer et al., 1994, Deslauriers et al., 1999, Angelotti et al., 2010). In addition, the size of the amino terminal segment differs significantly et al., 2010). In addition, the size of the amino terminal segment differs significantly between the 2-adrenoceptor subtypes, consisting of 28, 7 and 46 amino acids in the between the 2-adrenoceptor subtypes, consisting of 28, 7 and 46 amino acids in the human 2A-, 2B- and 2C-adrenoceptors, respectively. The hydrophobic repeat human 2A-, 2B- and 2C-adrenoceptors, respectively. The hydrophobic repeat sequence ”ALAAALAAAAA” near the beginning of the amino terminal segment of sequence ”ALAAALAAAAA” near the beginning of the amino terminal segment of the 2C-adrenoceptor was found by a chimaeric 2A/2C-adrenoceptor strategy to be an the 2C-adrenoceptor was found by a chimaeric 2A/2C-adrenoceptor strategy to be an endoplasmic reticulum retention signal responsible for the intracellular retention of the endoplasmic reticulum retention signal responsible for the intracellular retention of the 2C-adrenoceptor subtype (Angelotti et al., 2010). 2C-adrenoceptor subtype (Angelotti et al., 2010). The conserved structure and the high extent of amino acid identity of the 2- The conserved structure and the high extent of amino acid identity of the 2- adrenoceptors (~75 % in the TM region; 175 out of all 450/450/461 amino acids in the adrenoceptors (~75 % in the TM region; 175 out of all 450/450/461 amino acids in the human 2A-, 2B- and 2C-adrenoceptors, respectively) would be expected to result in human 2A-, 2B- and 2C-adrenoceptors, respectively) would be expected to result in relatively similar ligand binding properties and possibly complicate the design of relatively similar ligand binding properties and possibly complicate the design of subtype-selective drugs. Therefore, any type of information on the structural subtype-selective drugs. Therefore, any type of information on the structural determinants of ligand binding and recognition at 2-adrenoceptors may be considered determinants of ligand binding and recognition at 2-adrenoceptors may be considered to be potentially important for subtype-selective drug design and development. to be potentially important for subtype-selective drug design and development.

2.2.4 Molecular dynamics of 2-adrenoceptor activation 2.2.4 Molecular dynamics of 2-adrenoceptor activation The theory of receptor activation has advanced over the past century from the first The theory of receptor activation has advanced over the past century from the first “lock-and-key” concept, formulated by Emil Fisher (reviewed by Vernier et al., “lock-and-key” concept, formulated by Emil Fisher (reviewed by Vernier et al., 1995), to much more sophisticated and dynamic models that include also allosteric 1995), to much more sophisticated and dynamic models that include also allosteric regulation of the receptors. In one generally used model based on the ternary regulation of the receptors. In one generally used model based on the ternary complex concept (Figure 8), receptor activation requires the recognition of an complex concept (Figure 8), receptor activation requires the recognition of an extracellular ligand to cause conformational changes in the receptor structure, and extracellular ligand to cause conformational changes in the receptor structure, and transforms the receptor from an inactive state (R) to an active state (R*), followed by transforms the receptor from an inactive state (R) to an active state (R*), followed by G-protein activation and the exchange of guanosine diphosphate (GDP) for G-protein activation and the exchange of guanosine diphosphate (GDP) for guanosine triphosphate (GTP) on the cognate G-protein -subunit (see e.g. Weiss et guanosine triphosphate (GTP) on the cognate G-protein -subunit (see e.g. Weiss et al., 1996, Kenakin, 2004). The heterotrimeric G-protein complex then dissociates to al., 1996, Kenakin, 2004). The heterotrimeric G-protein complex then dissociates to a GTP-bound -subunit (39-46 kDa) and a -dimer (β-subunit, 36–37 kDa and γ- a GTP-bound -subunit (39-46 kDa) and a -dimer (β-subunit, 36–37 kDa and γ- subunit, 5–10 kDa), both of which can regulate specific downstream effector systems subunit, 5–10 kDa), both of which can regulate specific downstream effector systems such as enzymes or ion channels (Pfeuffer and Helmreich, 1988, Birnbaumer, 1990). such as enzymes or ion channels (Pfeuffer and Helmreich, 1988, Birnbaumer, 1990). As a result of the intrinsic GTPase activity of the -subunit, bound GTP is rapidly As a result of the intrinsic GTPase activity of the -subunit, bound GTP is rapidly hydrolysed to GDP, followed by re-association of the-subunit with the -dimer, hydrolysed to GDP, followed by re-association of the-subunit with the -dimer, thereby completing the G-protein activity cycle. This cascade of G-protein and thereby completing the G-protein activity cycle. This cascade of G-protein and GPCR interactions represents a fundamental mechanism of cell signalling for all GPCR interactions represents a fundamental mechanism of cell signalling for all GPCRs. GPCRs.

26 Review of the Literature 26 Review of the Literature

Figure 8. Evolution of receptor occupancy and activation models (Weiss et al., 1996). A) In the Figure 8. Evolution of receptor occupancy and activation models (Weiss et al., 1996). A) In the first simple model, binding of a ligand (L) to the receptor (R) leads to the generation of a first simple model, binding of a ligand (L) to the receptor (R) leads to the generation of a response. B) In the ternary complex model, the receptor is coupled to a G-protein (G). C) In the response. B) In the ternary complex model, the receptor is coupled to a G-protein (G). C) In the extended ternary complex model, the receptor may exhibit constitutive signalling, which extended ternary complex model, the receptor may exhibit constitutive signalling, which permits spontaneous isomerisation between inactive (R) and active (R*) receptor permits spontaneous isomerisation between inactive (R) and active (R*) receptor conformations. D) The cubic ternary complex model provides an additional interaction between conformations. D) The cubic ternary complex model provides an additional interaction between the inactive receptor and the G-protein. the inactive receptor and the G-protein.

In humans, 17 genes encode G-protein -subunits, 6 encode -subunits and 12 In humans, 17 genes encode G-protein -subunits, 6 encode -subunits and 12 encode -subunits (Jones and Assmann, 2004, Sprang et al., 2007), and some of these encode -subunits (Jones and Assmann, 2004, Sprang et al., 2007), and some of these genes are further expressed as splice variants. Based on their sequence, functional genes are further expressed as splice variants. Based on their sequence, functional properties and toxin sensitivities, the -subunits of G-proteins are classified into four properties and toxin sensitivities, the -subunits of G-proteins are classified into four main families, Gs, Gi/o, Gq and G12 (Birnbaumer, 1990, Sprang et al., 2007). The main families, Gs, Gi/o, Gq and G12 (Birnbaumer, 1990, Sprang et al., 2007). The specificity of a given receptor for the -isoforms generally determines the specific specificity of a given receptor for the -isoforms generally determines the specific intracellular messenger pathways of each receptor but also the -subunits regulate intracellular messenger pathways of each receptor but also the -subunits regulate certain intracellular effects. All three 2-adrenoceptor subtypes are preferentially certain intracellular effects. All three 2-adrenoceptor subtypes are preferentially coupled to pertussis toxin-sensitive Gi/o-type G-proteins (Cerione et al., 1986, coupled to pertussis toxin-sensitive Gi/o-type G-proteins (Cerione et al., 1986, Cotecchia et al., 1990, Eason et al., 1992). The activation of Gi/o proteins leads Cotecchia et al., 1990, Eason et al., 1992). The activation of Gi/o proteins leads primarily to reduced formation of cyclic adenosine monophosphate (cAMP) by primarily to reduced formation of cyclic adenosine monophosphate (cAMP) by adenylyl cyclase enzymes and increased K+ efflux via ligand-regulated K+-channels, adenylyl cyclase enzymes and increased K+ efflux via ligand-regulated K+-channels,

Review of the Literature 27 Review of the Literature 27 but also to stimulation of Ca2+ release from intracellular stores (Limbird, 1988, but also to stimulation of Ca2+ release from intracellular stores (Limbird, 1988, Cotecchia et al., 1990, Dorn et al., 1997, Arima et al., 1998, Kukkonen et al., 1998) Cotecchia et al., 1990, Dorn et al., 1997, Arima et al., 1998, Kukkonen et al., 1998) and to inhibition of N-, L- and P/Q-type voltage-activated Ca2+ channels (Soini et al., and to inhibition of N-, L- and P/Q-type voltage-activated Ca2+ channels (Soini et al., 1998, Delmas et al., 1999, Kolaj and Renaud, 2001, Timmons et al., 2004). These 1998, Delmas et al., 1999, Kolaj and Renaud, 2001, Timmons et al., 2004). These signalling pathways have been demonstrated for all 2-adrenoceptor subtypes in many signalling pathways have been demonstrated for all 2-adrenoceptor subtypes in many different types of tissue- and cell-based experiments. Some additional coupling different types of tissue- and cell-based experiments. Some additional coupling mechanisms have also been reported, e.g. coupling to activation of mitogen-activated mechanisms have also been reported, e.g. coupling to activation of mitogen-activated protein (MAP) kinases (Richman and Regan, 1998), activation of phospholipase A2 protein (MAP) kinases (Richman and Regan, 1998), activation of phospholipase A2 (Jones et al., 1991), phospholipase C (Seuwen et al., 1990, Koch et al., 1994, Dorn et (Jones et al., 1991), phospholipase C (Seuwen et al., 1990, Koch et al., 1994, Dorn et al., 1997) and phospholipase D (MacNulty et al., 1992). al., 1997) and phospholipase D (MacNulty et al., 1992). In contrast to the original ternary complex model, it is now appreciated that In contrast to the original ternary complex model, it is now appreciated that receptors may exhibit spontaneous signalling activity also in the absence of activating receptors may exhibit spontaneous signalling activity also in the absence of activating agonist ligands, referred to as the phenomenon of constitutive activity or basal activity agonist ligands, referred to as the phenomenon of constitutive activity or basal activity (Lefkowitz et al., 1993). Mutations of the conserved DRY motif at the cytoplamic end (Lefkowitz et al., 1993). Mutations of the conserved DRY motif at the cytoplamic end of TM3 have clearly increased the level of constitutive activity of certain GPCRs of TM3 have clearly increased the level of constitutive activity of certain GPCRs including some adrenoceptors (Wurch et al., 1999, Ge et al., 2003). On the basis of including some adrenoceptors (Wurch et al., 1999, Ge et al., 2003). On the basis of constitutive activity, it is currently considered that receptors must exist in an constitutive activity, it is currently considered that receptors must exist in an equilibrium between a resting state and an active state, and that agonist ligands equilibrium between a resting state and an active state, and that agonist ligands stabilise the receptor in its active conformation whereas inverse agonists suppress basal stabilise the receptor in its active conformation whereas inverse agonists suppress basal activity by stabilising the receptor in its inactive, resting conformation. Neutral activity by stabilising the receptor in its inactive, resting conformation. Neutral antagonists that prevent other ligands, agonists and inverse agonists, from binding are antagonists that prevent other ligands, agonists and inverse agonists, from binding are capable of binding to both receptor conformations with (nearly) equal affinity and thus capable of binding to both receptor conformations with (nearly) equal affinity and thus block their functional effects (Samama et al., 1994). block their functional effects (Samama et al., 1994). This established view of receptor-G-protein-effector coupling presents a linear This established view of receptor-G-protein-effector coupling presents a linear signalling cascade that has recently been challenged by suggestions of more complex signalling cascade that has recently been challenged by suggestions of more complex signalling networks. Activated receptors are known to recruit not only G-proteins but signalling networks. Activated receptors are known to recruit not only G-proteins but also arrestins, and arrestin-mediated signalling may contribute to the net effects of also arrestins, and arrestin-mediated signalling may contribute to the net effects of receptor activation. Arrestins not only modulate G-protein signalling (Lohse et al., receptor activation. Arrestins not only modulate G-protein signalling (Lohse et al., 1990), a classical effect demonstrated already 25 years ago in the light-sensing cells of 1990), a classical effect demonstrated already 25 years ago in the light-sensing cells of the retina (Wilden et al., 1986), but they can also mediate G-protein-independent the retina (Wilden et al., 1986), but they can also mediate G-protein-independent signalling in cells via pathways largely related to pro-survival and anti-apoptosis signalling in cells via pathways largely related to pro-survival and anti-apoptosis signals (Lefkowitz and Shenoy, 2005). Indeed, “biased ligands” may selectively signals (Lefkowitz and Shenoy, 2005). Indeed, “biased ligands” may selectively activate G-protein- or arrestin-mediated signalling, resulting in a situation where the activate G-protein- or arrestin-mediated signalling, resulting in a situation where the same receptor can mediate different cellular effects, depending on the activating same receptor can mediate different cellular effects, depending on the activating agonist. This may provide novel opportunities for drug development (Violin and agonist. This may provide novel opportunities for drug development (Violin and Lefkowitz, 2007). Lefkowitz, 2007). In spite of the availability of crystal structures of seven different GPCRs, little is In spite of the availability of crystal structures of seven different GPCRs, little is known at the time of writing this review about the structural basis of GPCR function: known at the time of writing this review about the structural basis of GPCR function: how does binding of an agonist lead to the conformational changes that are propagated how does binding of an agonist lead to the conformational changes that are propagated from the extracellular portion of the receptor molecule to the cytoplasmic surface from the extracellular portion of the receptor molecule to the cytoplasmic surface involved in G-protein binding? Each of the crystal structures of the GPCRs is just a involved in G-protein binding? Each of the crystal structures of the GPCRs is just a “snapshot” of one receptor conformation, and they probably represent inactive “snapshot” of one receptor conformation, and they probably represent inactive conformations of the receptors, and therefore do not necessarily improve our conformations of the receptors, and therefore do not necessarily improve our understanding of the conformation changes occurring upon receptor activation. understanding of the conformation changes occurring upon receptor activation. Receptor activation has, however, been extensively studied with other methods. Receptor activation has, however, been extensively studied with other methods. Different biochemical and biophysical approaches have also been applied for some Different biochemical and biophysical approaches have also been applied for some

28 Review of the Literature 28 Review of the Literature adrenoceptors, including site-directed spin labelling techniques, site-directed adrenoceptors, including site-directed spin labelling techniques, site-directed mutagenesis and various fluorescent technologies (Javitch et al., 1997, Jensen and mutagenesis and various fluorescent technologies (Javitch et al., 1997, Jensen and Gether, 2000, Jaakola et al., 2005, Zurn et al., 2009). One general assumption is that Gether, 2000, Jaakola et al., 2005, Zurn et al., 2009). One general assumption is that the overall conserved receptor structures support a common molecular activation the overall conserved receptor structures support a common molecular activation mechanism for all GPCRs (Oliveira et al., 2003, Schwartz et al., 2006). Nonetheless, mechanism for all GPCRs (Oliveira et al., 2003, Schwartz et al., 2006). Nonetheless, the exact mechanisms are not yet fully understood and there are at least three proposed the exact mechanisms are not yet fully understood and there are at least three proposed models for this: (1) in the “pivot” model, activation is driven by the rotation of TM3 models for this: (1) in the “pivot” model, activation is driven by the rotation of TM3 and TM6, wherein the proline (P6.50) -induced kink in TM6 acts as a pivot, leading to and TM6, wherein the proline (P6.50) -induced kink in TM6 acts as a pivot, leading to subsequent rearrangements and conformational interchanges at the intracellular portion subsequent rearrangements and conformational interchanges at the intracellular portion of the receptor (Chen et al., 2002); (2) in the “see-saw” model, the mechanism is of the receptor (Chen et al., 2002); (2) in the “see-saw” model, the mechanism is similar but is based on vertical see-saw movements around the proline-induced bend in similar but is based on vertical see-saw movements around the proline-induced bend in TM6 (Schwartz et al., 2006); (3) in the “toggle switch” model, receptor activation TM6 (Schwartz et al., 2006); (3) in the “toggle switch” model, receptor activation alters the configuration of the proline-induced kink in TM6 and causes the subsequent alters the configuration of the proline-induced kink in TM6 and causes the subsequent rotameric movement of the cytoplasmic end of TM6, e.g. allowing the straightening of rotameric movement of the cytoplasmic end of TM6, e.g. allowing the straightening of the proline kink (Shi and Javitch, 2002). One common feature of all of these models is the proline kink (Shi and Javitch, 2002). One common feature of all of these models is that the intracellular segments of the TMs, especially TM6, move apart during receptor that the intracellular segments of the TMs, especially TM6, move apart during receptor activation and thereby expose receptor epitopes, such as the DRY motif at the end of activation and thereby expose receptor epitopes, such as the DRY motif at the end of TM3, allowing recognition of/interactions with intracellular signalling molecules, such TM3, allowing recognition of/interactions with intracellular signalling molecules, such as G-proteins. as G-proteins. If the mechanisms of receptor activation are not well known, the molecular If the mechanisms of receptor activation are not well known, the molecular mechanisms of basal (constitutive) activity and inhibition of basal activity by inverse mechanisms of basal (constitutive) activity and inhibition of basal activity by inverse agonists are even more poorly understood and difficult to study, because the agonists are even more poorly understood and difficult to study, because the constitutively active state is short-lived and represents a minor fraction of possible constitutively active state is short-lived and represents a minor fraction of possible receptor conformations. Nonetheless, basal activity is thought to be physiologically receptor conformations. Nonetheless, basal activity is thought to be physiologically relevant, and the ability of a drug to inhibit constitutive receptor activity may influence relevant, and the ability of a drug to inhibit constitutive receptor activity may influence its therapeutic properties. its therapeutic properties.

2.2.5 Physiological functions and patterns of tissue expression of 2-adrenoceptors 2.2.5 Physiological functions and patterns of tissue expression of 2-adrenoceptors Three basic types of techniques have been used to characterise the tissue distributions Three basic types of techniques have been used to characterise the tissue distributions of the mammalian 2-adrenoceptor subtypes, especially in humans and in rodents, i.e. of the mammalian 2-adrenoceptor subtypes, especially in humans and in rodents, i.e. techniques to detect ligand binding (receptor autoradiography and radioligand binding techniques to detect ligand binding (receptor autoradiography and radioligand binding assays with tissue samples) (Boyajian et al., 1987), techniques to monitor receptor assays with tissue samples) (Boyajian et al., 1987), techniques to monitor receptor gene expression on the messenger RNA (mRNA) level (i.e. in situ hybridisation and gene expression on the messenger RNA (mRNA) level (i.e. in situ hybridisation and other methods to detect and quantitate mRNA species) (Nicholas et al., 1993), and other methods to detect and quantitate mRNA species) (Nicholas et al., 1993), and antibody-based methods (immunohistochemistry, Western blotting and related antibody-based methods (immunohistochemistry, Western blotting and related methods) (Kurose et al., 1993). The three 2-adrenoceptor subtypes were observed to methods) (Kurose et al., 1993). The three 2-adrenoceptor subtypes were observed to have unique patterns of tissue distribution both in the central nervous system (CNS) have unique patterns of tissue distribution both in the central nervous system (CNS) and in peripheral tissues (MacDonald et al., 1997). The 2A-adrenoceptor is expressed and in peripheral tissues (MacDonald et al., 1997). The 2A-adrenoceptor is expressed widely throughout the CNS, abundantly in the locus coeruleus and in other widely throughout the CNS, abundantly in the locus coeruleus and in other noradrenergic cell body regions, but also in peripheral tissues. The 2B-adrenoceptor noradrenergic cell body regions, but also in peripheral tissues. The 2B-adrenoceptor appears to be expressed primarily in peripheral tissues and only low levels of appears to be expressed primarily in peripheral tissues and only low levels of expression are found in the CNS, especially in the thalamus. The 2C-adrenoceptor expression are found in the CNS, especially in the thalamus. The 2C-adrenoceptor appears to be expressed mainly in the CNS, but with a different expression pattern appears to be expressed mainly in the CNS, but with a different expression pattern from the 2A-adrenoceptor, being most abundant in the striatum (Nicholas et al., 1993, from the 2A-adrenoceptor, being most abundant in the striatum (Nicholas et al., 1993,

Review of the Literature 29 Review of the Literature 29

Scheinin et al., 1994). Patterns of the tissue expression of the three 2-adrenoceptor Scheinin et al., 1994). Patterns of the tissue expression of the three 2-adrenoceptor subtypes are summarized in Table 2. subtypes are summarized in Table 2.

Table 2. Tissue distributions of 2-adrenoceptor subtypes in the central nervous system (CNS) Table 2. Tissue distributions of 2-adrenoceptor subtypes in the central nervous system (CNS) and peripheral tissues assessed by receptor autoradiography, in situ hybridisation, radioligand and peripheral tissues assessed by receptor autoradiography, in situ hybridisation, radioligand binding assays, immunohistochemistry and reverse transcriptase-PCR (RT-PCR). binding assays, immunohistochemistry and reverse transcriptase-PCR (RT-PCR). Subtype CNS Peripheral tissues Subtype CNS Peripheral tissues

2A Amygdala Kidney 2A Amygdala Kidney Locus coeruleus Vasculature Locus coeruleus Vasculature Lateral septum Urethra Lateral septum Urethra Brain stem Heart Brain stem Heart Cerebral cortex Blood platelets Cerebral cortex Blood platelets Thalamus Spleen Thalamus Spleen Hypothalamus Salivary glands Hypothalamus Salivary glands Hippocampus Pancreas Hippocampus Pancreas Spinal cord Fat cells Spinal cord Fat cells Olfactory nucleus Olfactory nucleus Retina Retina

2B Thalamus Kidney 2B Thalamus Kidney Placenta Placenta Liver Liver Vasculature Vasculature

2C Striatum Kidney 2C Striatum Kidney Olfactory tubercle Adrenal gland Olfactory tubercle Adrenal gland Locus coeruleus Vasculature Locus coeruleus Vasculature Hippocampus Pancreas Hippocampus Pancreas Cerebral cortex Cerebral cortex Amygdala Amygdala Substantia nigra Substantia nigra Derived from: Perälä et al., 1992, McCune et al., 1993, Nicholas et al., 1993, Ruffolo and Derived from: Perälä et al., 1992, McCune et al., 1993, Nicholas et al., 1993, Ruffolo and Hieble, 1994, Scheinin et al., 1994, MacDonald and Scheinin, 1995, Rosin et al., 1996, Tavares Hieble, 1994, Scheinin et al., 1994, MacDonald and Scheinin, 1995, Rosin et al., 1996, Tavares et al., 1996, Trendelenburg et al., 1997, Winzer-Serhan et al., 1997, Lee et al., 1998, Saunders et al., 1996, Trendelenburg et al., 1997, Winzer-Serhan et al., 1997, Lee et al., 1998, Saunders and Limbird, 1999, Dossin et al., 2000, Holmberg et al., 2003, Peterhoff et al., 2003, Chotani et and Limbird, 1999, Dossin et al., 2000, Holmberg et al., 2003, Peterhoff et al., 2003, Chotani et al., 2004, Happe et al., 2004, Schambra et al., 2005. al., 2004, Happe et al., 2004, Schambra et al., 2005.

The first and most classical physiological function described for 2-adrenoceptors is The first and most classical physiological function described for 2-adrenoceptors is to mediate presynaptic feedback inhibition of neurotransmitter release from to mediate presynaptic feedback inhibition of neurotransmitter release from noradrenergic nerve endings (Langer, 1976). Today, based on molecular cloning, ligand noradrenergic nerve endings (Langer, 1976). Today, based on molecular cloning, ligand binding and anatomical and functional studies of the different 2-adrenoceptor subtypes, binding and anatomical and functional studies of the different 2-adrenoceptor subtypes, it is evident that 2-adrenoceptors are also involved in many postsynaptic and it is evident that 2-adrenoceptors are also involved in many postsynaptic and extrasynaptic actions. However, the precise physiological functions of each of the three extrasynaptic actions. However, the precise physiological functions of each of the three 2-adrenoceptor subtypes have been difficult to elucidate, largely because of the lack of 2-adrenoceptor subtypes have been difficult to elucidate, largely because of the lack of subtype-selective pharmacological probes. Anatomical data on receptor expression have subtype-selective pharmacological probes. Anatomical data on receptor expression have not always been possible to definitively link to specific physiological functions of a not always been possible to definitively link to specific physiological functions of a particular receptor subtype. More recently, mice with altered 2-adrenoceptor gene particular receptor subtype. More recently, mice with altered 2-adrenoceptor gene expression have become important tools to elucidate the distinct physiological functions expression have become important tools to elucidate the distinct physiological functions

30 Review of the Literature 30 Review of the Literature of the 2-adrenoceptor subtypes. Several genetically engineered mouse lines have been of the 2-adrenoceptor subtypes. Several genetically engineered mouse lines have been generated, e.g. 2A-, 2B-, 2C- and double 2AC-deficient (knockout) mouse lines with generated, e.g. 2A-, 2B-, 2C- and double 2AC-deficient (knockout) mouse lines with disrupted gene expression of one or several of the 2-adrenoceptor subtypes (Link et al., disrupted gene expression of one or several of the 2-adrenoceptor subtypes (Link et al., 1996, Altman et al., 1999, Hein et al., 1999), a mouse line with increased 2C- 1996, Altman et al., 1999, Hein et al., 1999), a mouse line with increased 2C- adrenoceptor expression (overexpressing mouse line) (Link et al., 1995, Sallinen et al., adrenoceptor expression (overexpressing mouse line) (Link et al., 1995, Sallinen et al., 1997) and a mouse line containing an inactivating point mutation of the 2A- 1997) and a mouse line containing an inactivating point mutation of the 2A- adrenoceptor (D2.50N substitution in the D79N mouse line) (MacMillan et al., 1996, adrenoceptor (D2.50N substitution in the D79N mouse line) (MacMillan et al., 1996, Lakhlani et al., 1997). Studies on these gene-manipulated mice have emphasized the Lakhlani et al., 1997). Studies on these gene-manipulated mice have emphasized the crucial role for 2A-adrenoceptors in sedation, analgesia, hypotension, sympathetic crucial role for 2A-adrenoceptors in sedation, analgesia, hypotension, sympathetic inhibition, pupil control, alleviation of opioid and alcohol withdrawal symptoms, inhibition, pupil control, alleviation of opioid and alcohol withdrawal symptoms, regulation of body temperature and blood glucose homeostasis and modulation of regulation of body temperature and blood glucose homeostasis and modulation of seizure susceptibility, all previously identified as important pharmacological effects seizure susceptibility, all previously identified as important pharmacological effects mediated by 2-adrenoceptors (Ruffolo et al., 1993). Nonetheless, few of these effects mediated by 2-adrenoceptors (Ruffolo et al., 1993). Nonetheless, few of these effects can be totally and unequivocally attributed to a single receptor subtype. The 2A- can be totally and unequivocally attributed to a single receptor subtype. The 2A- adrenoceptor has generally been observed to have a much more prominent role in CNS adrenoceptor has generally been observed to have a much more prominent role in CNS regulation than the 2C-adrenoceptor (Trendelenburg et al., 2001). Peripherally regulation than the 2C-adrenoceptor (Trendelenburg et al., 2001). Peripherally expressed 2B-adrenoceptors appear to be involved in the regulation of blood pressure expressed 2B-adrenoceptors appear to be involved in the regulation of blood pressure by controlling contraction and relaxation of vascular smooth muscle cells (Link et al., by controlling contraction and relaxation of vascular smooth muscle cells (Link et al., 1996), but the evidence is still somewhat uncertain. Examples of some 2-adrenoceptor- 1996), but the evidence is still somewhat uncertain. Examples of some 2-adrenoceptor- mediated physiological functions identified by experiments performed in transgenic mediated physiological functions identified by experiments performed in transgenic mice with altered gene expression are summarised in Table 3. mice with altered gene expression are summarised in Table 3.

Table 3. Overview of physiological functions and pharmacological responses mediated by 2- Table 3. Overview of physiological functions and pharmacological responses mediated by 2- adrenoceptor subtypes, as proposed from experiments with genetically engineered mice. adrenoceptor subtypes, as proposed from experiments with genetically engineered mice. Receptor Physiological functions and responses References Receptor Physiological functions and responses References

2A Analgesia Hunter et al., 1997 2A Analgesia Hunter et al., 1997 Bradycardia and hypotension MacMillan et al., 1996 Bradycardia and hypotension MacMillan et al., 1996 Hypothermia Lähdesmäki et al., 2003 Hypothermia Lähdesmäki et al., 2003 Inhibition of epileptic seizures Janumpalli et al., 1998, Szot et al., 2004 Inhibition of epileptic seizures Janumpalli et al., 1998, Szot et al., 2004 Presynaptic inhibition of Altman et al., 1999, Hein et al., 1999, Presynaptic inhibition of Altman et al., 1999, Hein et al., 1999, neurotransmitter release Kable et al., 2000 neurotransmitter release Kable et al., 2000 Anxiety-like behaviour Schramm et al., 2001 Anxiety-like behaviour Schramm et al., 2001 Sedation and anaesthesia Lakhlani et al., 1997 Sedation and anaesthesia Lakhlani et al., 1997 Regulation of blood glucose and insulin Fagerholm et al., 2004 Regulation of blood glucose and insulin Fagerholm et al., 2004 homeostasis homeostasis  Decrease of intraocular pressure Wikberg-Matsson and Simonsen, 2001  Decrease of intraocular pressure Wikberg-Matsson and Simonsen, 2001  Inhibition of gastrointestinal motility Scheibner et al., 2002  Inhibition of gastrointestinal motility Scheibner et al., 2002

2B Placental angiogenesis Philipp et al., 2002 2B Placental angiogenesis Philipp et al., 2002 Salt-induced hypertension Makaritsis et al., 1999 Salt-induced hypertension Makaritsis et al., 1999 Vascular smooth muscle contraction Hein et al., 1999 Vascular smooth muscle contraction Hein et al., 1999

2C Presynaptic inhibition of catecholamine Hein et al., 1999, Brede et al., 2003 2C Presynaptic inhibition of catecholamine Hein et al., 1999, Brede et al., 2003 release release Modulation of motor behaviour Sallinen et al., 1999 Modulation of motor behaviour Sallinen et al., 1999 Regulation of dopamine and serotonin Sallinen et al., 1998 Regulation of dopamine and serotonin Sallinen et al., 1998 balance in the brain balance in the brain Vascular smooth muscle contraction Hein et al., 1999 Vascular smooth muscle contraction Hein et al., 1999

Review of the Literature 31 Review of the Literature 31

On the cellular level, many GPCRs have been found to form homodimers and On the cellular level, many GPCRs have been found to form homodimers and heterodimers, possibly implying increased biological and pharmacological complexity heterodimers, possibly implying increased biological and pharmacological complexity in receptor functions. The formation of dimers and higher-order oligomers may occur in receptor functions. The formation of dimers and higher-order oligomers may occur during the purification of receptor proteins with apparent molecular weights larger than during the purification of receptor proteins with apparent molecular weights larger than expected. In living cells, 2A-adrenoceptors and muscarinic M4 receptors as well as expected. In living cells, 2A-adrenoceptors and muscarinic M4 receptors as well as 2A-adrenoceptors and µ- and -type opioid receptors have been proposed to form 2A-adrenoceptors and µ- and -type opioid receptors have been proposed to form functional dimers (Jordan et al., 2003, Rios et al., 2004, Zhang and Limbird, 2004, functional dimers (Jordan et al., 2003, Rios et al., 2004, Zhang and Limbird, 2004, Nobles et al., 2005). These findings point to more complicated characteristics of 2- Nobles et al., 2005). These findings point to more complicated characteristics of 2- adrenoceptors than initially believed; perhaps very few functions of 2-adrenoceptors adrenoceptors than initially believed; perhaps very few functions of 2-adrenoceptors can ultimately be ascribed to a single 2-adrenoceptor subtype acting alone. On the can ultimately be ascribed to a single 2-adrenoceptor subtype acting alone. On the other hand, functional complexity based on receptor oligomerisation and other protein other hand, functional complexity based on receptor oligomerisation and other protein interactions may yield increased possibilities for drug discovery (Ferrè et al., 2010). interactions may yield increased possibilities for drug discovery (Ferrè et al., 2010). The localisation of GPCRs in the plasma membrane is dynamic rather than static, The localisation of GPCRs in the plasma membrane is dynamic rather than static, with several highly regulated processes determining the density and spatial distribution with several highly regulated processes determining the density and spatial distribution of the receptors in different domains of the plasma membrane and in intracellular of the receptors in different domains of the plasma membrane and in intracellular compartments. GPCRs are synthesised in the endoplasmic reticulum and then further compartments. GPCRs are synthesised in the endoplasmic reticulum and then further processed in the Golgi apparatus before their transport to the plasma membrane. processed in the Golgi apparatus before their transport to the plasma membrane. Agonist binding to the receptor not only leads to the generation of second messengers Agonist binding to the receptor not only leads to the generation of second messengers and intracellular responses, but may also trigger a series of adaptive changes that and intracellular responses, but may also trigger a series of adaptive changes that decrease the subsequent responsiveness of the receptor. This phenomenon is referred to decrease the subsequent responsiveness of the receptor. This phenomenon is referred to as desensitisation, which is often accompanied by internalisation of the receptors. as desensitisation, which is often accompanied by internalisation of the receptors. Desensitised receptors may become resensitised and recycled back into the plasma Desensitised receptors may become resensitised and recycled back into the plasma membrane ready to function in ligand recognition and signal transduction. membrane ready to function in ligand recognition and signal transduction. Alternatively, internalised receptors may be transported to other intracellular Alternatively, internalised receptors may be transported to other intracellular compartments for further processing, leading to receptor “down-regulation”, and compartments for further processing, leading to receptor “down-regulation”, and subsequently to decreased receptor density. subsequently to decreased receptor density. Agonist-induced trafficking has also been reported for 2-adrenoceptors. The Agonist-induced trafficking has also been reported for 2-adrenoceptors. The human 2A- and 2B-adrenoceptors were observed to internalise into distinct human 2A- and 2B-adrenoceptors were observed to internalise into distinct intracellular vesicles in transfected Madin-Darby canine kidney (MDCK) cells, rat intracellular vesicles in transfected Madin-Darby canine kidney (MDCK) cells, rat pheochromocytoma (PC12) cells and in human embryonic kidney (HEK-293) cells pheochromocytoma (PC12) cells and in human embryonic kidney (HEK-293) cells after agonist exposure (Keefer and Limbird, 1993, Daunt et al., 1997, Olli-Lähdesmäki after agonist exposure (Keefer and Limbird, 1993, Daunt et al., 1997, Olli-Lähdesmäki et al., 1999, Taraviras et al., 2002, Olli-Lähdesmäki et al., 2003). Dynamic regulation et al., 1999, Taraviras et al., 2002, Olli-Lähdesmäki et al., 2003). Dynamic regulation of receptor density and signaling after exposure to an agonist has significant of receptor density and signaling after exposure to an agonist has significant consequences for the pharmacological responsiveness to many drugs, and is therefore consequences for the pharmacological responsiveness to many drugs, and is therefore an important cellular phenomenon. an important cellular phenomenon.

2.2.6 Therapeutic applications of 2-adrenoceptors 2.2.6 Therapeutic applications of 2-adrenoceptors

Current clinically available 2-adrenoceptor drugs show only marginal subtype Current clinically available 2-adrenoceptor drugs show only marginal subtype selectivity, and partly as a consequence of this, several unwanted side-effects limit their selectivity, and partly as a consequence of this, several unwanted side-effects limit their therapeutic usefulness. This lack of agents that are at the same time subtype-selective and therapeutic usefulness. This lack of agents that are at the same time subtype-selective and specific for 2-adrenoceptors has also precluded the detailed elucidation of the specific for 2-adrenoceptors has also precluded the detailed elucidation of the physiological roles of the 2-adrenoceptor subtypes and the clinical contexts where drugs physiological roles of the 2-adrenoceptor subtypes and the clinical contexts where drugs acting on 2-adrenoceptors could be applicable. Mice with genetically modified receptor acting on 2-adrenoceptors could be applicable. Mice with genetically modified receptor gene expression are now available and are used extensively to study the physiological gene expression are now available and are used extensively to study the physiological and pharmacological functions of the 2-adrenoceptor subtypes. Gene-targeted mice are and pharmacological functions of the 2-adrenoceptor subtypes. Gene-targeted mice are

32 Review of the Literature 32 Review of the Literature expected to predict the pharmacological and therapeutic properties of subtype-selective expected to predict the pharmacological and therapeutic properties of subtype-selective drugs, but they are not expected to totally display the same phenotypic effects as could be drugs, but they are not expected to totally display the same phenotypic effects as could be seen by administering pharmacological agents to human subjects (MacDonald et al., seen by administering pharmacological agents to human subjects (MacDonald et al., 1997). In addition to species differences between mice and humans and possible 1997). In addition to species differences between mice and humans and possible adaptations taking place in transgenic mice because of the life-long absence of a receptor adaptations taking place in transgenic mice because of the life-long absence of a receptor type, genetic variation in the human population could also play a role. For example, type, genetic variation in the human population could also play a role. For example, several common polymorphisms have been identified within the coding regions of the several common polymorphisms have been identified within the coding regions of the human 2-adrenoceptor genes (Heinonen et al., 1999, Small et al., 2000a, Small et al., human 2-adrenoceptor genes (Heinonen et al., 1999, Small et al., 2000a, Small et al., 2000b). Some of these sequence variants, e.g. substitutions or deletions of amino acids in 2000b). Some of these sequence variants, e.g. substitutions or deletions of amino acids in the third intracellular loop, have been demonstrated to affect receptor function at least in the third intracellular loop, have been demonstrated to affect receptor function at least in cell models in vitro (Small et al., 2000b, Small and Liggett, 2001). Clinical genetic cell models in vitro (Small et al., 2000b, Small and Liggett, 2001). Clinical genetic studies have associated 2-adrenoceptor sequence variants with certain clinical studies have associated 2-adrenoceptor sequence variants with certain clinical phenotypes (Snapir et al., 2001, Snapir et al., 2003), and genetic variation may at least in phenotypes (Snapir et al., 2001, Snapir et al., 2003), and genetic variation may at least in part explain the observed substantial inter-individual variability in responses to 2- part explain the observed substantial inter-individual variability in responses to 2- adrenoceptor agonists and antagonists (Kohli et al., 2010). From a different viewpoint, adrenoceptor agonists and antagonists (Kohli et al., 2010). From a different viewpoint, genetic receptor variants may also be considered as potential drug targets in the future, in genetic receptor variants may also be considered as potential drug targets in the future, in an era of individualized medicine. an era of individualized medicine. Several subtype non-selective 2-adrenoceptor agonists are in clinical use with Several subtype non-selective 2-adrenoceptor agonists are in clinical use with well-established therapeutic indications, e.g. clonidine, -methyldopa, guanfacine, well-established therapeutic indications, e.g. clonidine, -methyldopa, guanfacine, guanabenz, moxonidine and rilmenidine in the treatment of hypertension (Ruffolo and guanabenz, moxonidine and rilmenidine in the treatment of hypertension (Ruffolo and Hieble, 1994). Para-aminoclonidine (apraclonidine) and brimonidine (bromoxidine) Hieble, 1994). Para-aminoclonidine (apraclonidine) and brimonidine (bromoxidine) are dispensed as eyedrops to treat glaucoma (Serle, 1994). A structural analogue of are dispensed as eyedrops to treat glaucoma (Serle, 1994). A structural analogue of clonidine, tizanidine, is an effective muscle relaxant and useful in the treatment of clonidine, tizanidine, is an effective muscle relaxant and useful in the treatment of spasticity in several neurological diseases such as multiple sclerosis and spinal cord spasticity in several neurological diseases such as multiple sclerosis and spinal cord injury (Bes et al., 1988, Eyssette et al., 1988, Mathias et al., 1989). Oxymetazoline, an injury (Bes et al., 1988, Eyssette et al., 1988, Mathias et al., 1989). Oxymetazoline, an imidazoline derivative, is used as a nasal spray formulation as a decongestant to imidazoline derivative, is used as a nasal spray formulation as a decongestant to provide rapid relief of nasal obstruction (Fox et al., 1979). provide rapid relief of nasal obstruction (Fox et al., 1979). In veterinary surgery, 2-adrenoceptor agonists such as xylazine, detomidine, In veterinary surgery, 2-adrenoceptor agonists such as xylazine, detomidine, medetomidine and dexmedetomidine are widely used to achieve sedation, analgesia, medetomidine and dexmedetomidine are widely used to achieve sedation, analgesia, anxiolysis, hypnosis and immobilization of animals, while the 2-adrenoceptor antagonist anxiolysis, hypnosis and immobilization of animals, while the 2-adrenoceptor antagonist atipamezole is available for reversal of sedation and immobilization (Savola et al., 1986, atipamezole is available for reversal of sedation and immobilization (Savola et al., 1986, Virtanen, 1989, Schwartz and Clark, 1998, Paddleford and Harvey, 1999). Virtanen, 1989, Schwartz and Clark, 1998, Paddleford and Harvey, 1999). Dexmedetomidine is also used clinically in human medicine in the intensive care setting in Dexmedetomidine is also used clinically in human medicine in the intensive care setting in the United States and in about 40 other non-EU countries. It has sympatholytic and the United States and in about 40 other non-EU countries. It has sympatholytic and antinociceptive effects that provide hemodynamic stability during surgery and other painful antinociceptive effects that provide hemodynamic stability during surgery and other painful and stressful procedures, and has some beneficial therapeutic effects in comparison with and stressful procedures, and has some beneficial therapeutic effects in comparison with other clinically used anaesthetics (Aantaa and Jalonen, 2006, Arcangeli et al., 2009). other clinically used anaesthetics (Aantaa and Jalonen, 2006, Arcangeli et al., 2009). Even though many functional effects have been described for 2-adrenoceptors, Even though many functional effects have been described for 2-adrenoceptors, there are currently almost no human therapeutic applications for 2-adrenoceptor there are currently almost no human therapeutic applications for 2-adrenoceptor antagonists. Yohimbine and phentolamine were used by some in the treatment of male antagonists. Yohimbine and phentolamine were used by some in the treatment of male sexual impotence before modern phosphodiesterase type 5 (PDE5) inhibitors became sexual impotence before modern phosphodiesterase type 5 (PDE5) inhibitors became available (Andersson and Stief, 2001). Mianserin and mirtazapine are commonly used available (Andersson and Stief, 2001). Mianserin and mirtazapine are commonly used as antidepressants, but these drugs possess many other pharmacological effects in as antidepressants, but these drugs possess many other pharmacological effects in addition to antagonism of 2-adrenoceptors (Stimmel et al., 1997). Some other addition to antagonism of 2-adrenoceptors (Stimmel et al., 1997). Some other conditions have been hypothesized to be targeted by 2-adrenoceptor antagonists, e.g. conditions have been hypothesized to be targeted by 2-adrenoceptor antagonists, e.g. non-insulin-dependent diabetes and obesity (Ruffolo and Hieble, 1994). non-insulin-dependent diabetes and obesity (Ruffolo and Hieble, 1994).

Aims of the Study 33 Aims of the Study 33

3. AIMS OF THE STUDY 3. AIMS OF THE STUDY

The three mammalian 2-adrenoceptor subtypes bind the endogenous catecholamines The three mammalian 2-adrenoceptor subtypes bind the endogenous catecholamines adrenaline and noradrenaline with similar affinities, while their binding affinities for adrenaline and noradrenaline with similar affinities, while their binding affinities for many synthetic ligands differ. The present clinically available 2-adrenoceptor drugs, many synthetic ligands differ. The present clinically available 2-adrenoceptor drugs, agonists and antagonists, are not subtype-selective, which limits their therapeutic agonists and antagonists, are not subtype-selective, which limits their therapeutic usefulness. Understanding of the molecular basis of drug actions on 2-adrenoceptors, usefulness. Understanding of the molecular basis of drug actions on 2-adrenoceptors, in terms of receptor structure and function, is expected to provide valuable information in terms of receptor structure and function, is expected to provide valuable information for subtype-selective drug development. Site-directed mutagenesis, ligand binding for subtype-selective drug development. Site-directed mutagenesis, ligand binding experiments, computer-based molecular modelling and sequence comparisons between experiments, computer-based molecular modelling and sequence comparisons between paralogous and orthologous receptors are strategies used to reveal the key domains and paralogous and orthologous receptors are strategies used to reveal the key domains and amino acids that are important for specific ligand recognition and binding. amino acids that are important for specific ligand recognition and binding. Improved understanding of differences between the 2-adrenoceptor subtypes on Improved understanding of differences between the 2-adrenoceptor subtypes on the molecular level is expected to facilitate the design and development of new truly the molecular level is expected to facilitate the design and development of new truly subtype-selective 2-adrenoceptor drugs and to improve the accuracy of molecular subtype-selective 2-adrenoceptor drugs and to improve the accuracy of molecular models of the 2-adrenoceptor proteins. From this point of view, the aims of the models of the 2-adrenoceptor proteins. From this point of view, the aims of the present studies were as follows: present studies were as follows:

1. Characterisation of the pharmacological properties of the recently cloned 1. Characterisation of the pharmacological properties of the recently cloned zebrafish 2-adrenoceptor subtypes and their comparison with the corresponding zebrafish 2-adrenoceptor subtypes and their comparison with the corresponding human orthologues. This information was also expected to suggest amino acid human orthologues. This information was also expected to suggest amino acid domains and positions for subsequent mutagenesis experiments (study I). domains and positions for subsequent mutagenesis experiments (study I).

2. Elucidation of the structural determinants that contribute to the different 2. Elucidation of the structural determinants that contribute to the different pharmacological profiles of the human and mouse (rodent) 2A-adrenoceptors pharmacological profiles of the human and mouse (rodent) 2A-adrenoceptors (study II). (study II).

3. Elucidation of the possible roles of the extracellular amino-terminal region and 3. Elucidation of the possible roles of the extracellular amino-terminal region and the first transmembrane domain of the receptors in the subtype-selective binding the first transmembrane domain of the receptors in the subtype-selective binding of antagonist ligands to human 2-adrenoceptors (study III). of antagonist ligands to human 2-adrenoceptors (study III).

34 Materials and Methods 34 Materials and Methods

4. MATERIALS AND METHODS 4. MATERIALS AND METHODS

4.1 Mutagenesis and expression vectors 4.1 Mutagenesis and expression vectors Mutagenesis is a process where DNA, the molecular basis of genetic information, is Mutagenesis is a process where DNA, the molecular basis of genetic information, is permanently changed, either in nature or experimentally by the use of chemicals or permanently changed, either in nature or experimentally by the use of chemicals or radiation. For geneticists, the study of mutagenesis is important since mutations may radiation. For geneticists, the study of mutagenesis is important since mutations may reveal the genetic mechanisms underlying heredity and gene expression. In the life reveal the genetic mechanisms underlying heredity and gene expression. In the life sciences and in molecular biology, mutagenesis is utilized as a tool to study protein sciences and in molecular biology, mutagenesis is utilized as a tool to study protein function and structure. In site-directed mutagenesis, a pre-planned mutation is created function and structure. In site-directed mutagenesis, a pre-planned mutation is created at a defined site in a DNA molecule. This requires that the wild-type gene sequence is at a defined site in a DNA molecule. This requires that the wild-type gene sequence is known. known. Mutagenesis based on polymerase chain reaction (PCR) technology is a relatively Mutagenesis based on polymerase chain reaction (PCR) technology is a relatively simple modern method for generating site-directed mutations. PCR can be used to simple modern method for generating site-directed mutations. PCR can be used to produce mutations such as nucleotide substitutions, insertions and deletions in double- produce mutations such as nucleotide substitutions, insertions and deletions in double- stranded DNA without the need of subcloning into a bacteriophage plasmid. The stranded DNA without the need of subcloning into a bacteriophage plasmid. The procedure involves a PCR reaction using template DNA and synthetic oligonucleotide procedure involves a PCR reaction using template DNA and synthetic oligonucleotide primers containing the desired mutation, i.e. the “mismatch”, which are complementary primers containing the desired mutation, i.e. the “mismatch”, which are complementary to the opposite strands of the template DNA. to the opposite strands of the template DNA. The wild-type cDNAs encoding human, mouse and zebrafish 2-adrenoceptors used The wild-type cDNAs encoding human, mouse and zebrafish 2-adrenoceptors used in the present studies were inserted either into the vector pREP4, pcDNA3 or in the present studies were inserted either into the vector pREP4, pcDNA3 or pcDNA3.1(+) (Invitrogen Life Technologies Inc., Rockville, MD, USA). In study I, pcDNA3.1(+) (Invitrogen Life Technologies Inc., Rockville, MD, USA). In study I, fragments containing the coding sequences of the zebrafish 2A-, 2B-, 2C-, 2Da- and fragments containing the coding sequences of the zebrafish 2A-, 2B-, 2C-, 2Da- and 2Db-adrenoceptors were generated either from genomic DNA or phage or cosmid 2Db-adrenoceptors were generated either from genomic DNA or phage or cosmid clones with primers flanking the coding regions and containing artificial restriction clones with primers flanking the coding regions and containing artificial restriction enzyme recognition sites for ligation into the pREP4 vector (see Ruuskanen et al., enzyme recognition sites for ligation into the pREP4 vector (see Ruuskanen et al., 2004). In study II, the cDNA encoding the human 2A-adrenoceptor in the vector 2004). In study II, the cDNA encoding the human 2A-adrenoceptor in the vector pcDNA3(+) was obtained from the UMR cDNA Resource Center (University of pcDNA3(+) was obtained from the UMR cDNA Resource Center (University of Missouri-Rolla, Rolla, MO, USA) and the cDNA encoding the mouse 2A- Missouri-Rolla, Rolla, MO, USA) and the cDNA encoding the mouse 2A- adrenoceptor (clone M2-10H; Link et al., 1992) in the vector pcDNA3 was originally adrenoceptor (clone M2-10H; Link et al., 1992) in the vector pcDNA3 was originally provided by Dr. Brian Kobilka (Stanford University, Stanford, CA, USA). In study III, provided by Dr. Brian Kobilka (Stanford University, Stanford, CA, USA). In study III, the cDNAs encoding human 2A-, 2B- and 2C-adrenoceptors were cloned into the the cDNAs encoding human 2A-, 2B- and 2C-adrenoceptors were cloned into the pREP4 (2A and 2B) and pcDNA3 (2C) vectors, and were originally provided by Dr. pREP4 (2A and 2B) and pcDNA3 (2C) vectors, and were originally provided by Dr. Brian Kobilka. Brian Kobilka. Site-directed mutagenesis (in studies II and III) was performed utilising the Site-directed mutagenesis (in studies II and III) was performed utilising the GeneEditorTM in vitro Site-Directed Mutagenesis System (Promega, Madison, WI, GeneEditorTM in vitro Site-Directed Mutagenesis System (Promega, Madison, WI, USA). The system is designed for use with plasmid vectors conferring ampicillin USA). The system is designed for use with plasmid vectors conferring ampicillin resistance to the E. coli bacterial hosts, encoded by the TEM-1 -lactamase (ampicillin resistance to the E. coli bacterial hosts, encoded by the TEM-1 -lactamase (ampicillin resistance) gene. Synthetic mutagenic oligonucleotides complementary to the target resistance) gene. Synthetic mutagenic oligonucleotides complementary to the target templates but containing the desired mutations were annealed to denatured single- templates but containing the desired mutations were annealed to denatured single- stranded DNA together with a selection oligonucleotide complementary to the - stranded DNA together with a selection oligonucleotide complementary to the - lactamase gene, except for a 7-nucleotide sequence that encodes resistance to the lactamase gene, except for a 7-nucleotide sequence that encodes resistance to the GeneEditorTM Antibiotic Selection Mix. The hybridised oligonucleotide was extended GeneEditorTM Antibiotic Selection Mix. The hybridised oligonucleotide was extended

Materials and Methods 35 Materials and Methods 35 with DNA polymerase to form double-stranded DNA. Subsequently, the constructed with DNA polymerase to form double-stranded DNA. Subsequently, the constructed cDNA was transformed into E. coli host cells (BMH71-18mutS and JM109 competent cDNA was transformed into E. coli host cells (BMH71-18mutS and JM109 competent cells) and grown with the GeneEditorTM Antibiotic Selection Mix for selection of cells) and grown with the GeneEditorTM Antibiotic Selection Mix for selection of plasmids derived from the mutant strand. plasmids derived from the mutant strand. Chimaeric receptor constructs based on the 2A- and 2B-adrenoceptors (in study Chimaeric receptor constructs based on the 2A- and 2B-adrenoceptors (in study III) were constructed using PCR-based mutagenesis and two pairs of primers designed III) were constructed using PCR-based mutagenesis and two pairs of primers designed for each subtype. Exchange of the TM1 domain and the preceding N-terminal segment for each subtype. Exchange of the TM1 domain and the preceding N-terminal segment between receptor subtypes was performed utilizing conserved threonine-serine sites at between receptor subtypes was performed utilizing conserved threonine-serine sites at the distal end of TM1. A reverse primer to the distal part of TM1 and a forward primer the distal end of TM1. A reverse primer to the distal part of TM1 and a forward primer to the first intracellular loop was designed to contain an artificial SpeI-site; recognition to the first intracellular loop was designed to contain an artificial SpeI-site; recognition sequence ACTAGT, coding for threonine-serine. In addition, the N- and C-terminal sequence ACTAGT, coding for threonine-serine. In addition, the N- and C-terminal primers contained restriction sites for subcloning into the pREP4 expression vector. A primers contained restriction sites for subcloning into the pREP4 expression vector. A list of the primers used in studies II and III is shown in Table 4. list of the primers used in studies II and III is shown in Table 4.

36

Table 4. A list of primers used for mutagenesis in studies II and III. Nucleotides that encode an artificial restriction enzyme recognition site or mutation are highlighted in colour: KpnI (blue), SpeI (red), HindIII (green) and mutation (bold black). GeneEditorTM in vitro Site-Directed Mutagenesis System primers Primer name Primer sequence Study Human ADRA2A (RCExl2.49-xl2.51SCK) 3'-TCG GCC GGC TCG GCT CAA CGC CTT AGT TGC TGG TCT TCA CCA T-5' II Mouse ADRA2A (SCKxl2.49-xl2.51RCE): 5'-GGT CGT TGA TCT CGC ATC TTG GCT CGG CC-3' II ADRA2C-SpeI 5’-GCG CAG CGC CCG ACT AGT CAG CAC GGC GAT-3’ III

ADRA2C (ADRA2A Nter-TM1)-NheI 5´-TGC CAC CGC GCC CGC GCT AGC CTG GCC GCG CGG CGG-3´ III Materials andMethods ADRA2C (ADRA2A Nter-TM1) (restore) 5´-CAC CTG CAG GCT ATA CTG GCC GCG CG-3´ III ADRA2A (ADCA2C Nter-TM1)-NheI 5´-CAG CGT CAC CTG CAG GCT AGC AGG GGT GGC CCG GGC-3´ III ADRA2A (ADCA2C Nter-TM1) (restore) 5´-CCG CGC CCG CGC TAT AAG GGG TGG CCC GGG C-3´ III

PCR-based mutagenesis primers III Primer name Primer sequence III ADRA2A-TM1-KpnI (forward) 5'-TCG GTA CCA TGG GCT CCC TGC AGC CGG ACG CGG GCA ACG CGA GCT GGA A-3' III ADRA2A-TM1-SpeI (reverse) 5'-GAC TAG TGA ACA CGG CGA TGA TGA CGA GCA CGT T-3' III ADRA2A-cterm-SpeI (forward) 5'-AAC TAG TCG CGC GCT CAA GGC GCC CCA AAA CCT CTT-3' III ADRA2A-cterm-HindIII (reverse) 5'-TCG AGA AGC TTT CAC ACG ATC CGC TTC CTG T-3' III ADRA2B-TM1-KpnI (forward) 5'-TCG GTA CCA TGG ACC ACC AGG ACC CCT ACT CCG TGC AGG CCG CAG CGG CCA TA-3' III ADRA2B-TM1-SpeI (reverse) 5'-TAC TAG TCA ACA CAG CCA GGA TGA CCA GCG CGT T-3' III ADRA2B-cterm-SpeI (forward) 5'-AAC TAG TCG CTC GCT GCG CGC CCC TCG GAA-3' III ADRA2B-cterm-HindIII (reverse) 5'-TTA GGA AGC TTT CAC CAG GCC GTC TGG GTC CAC GGG CGG CAC AGG AT-3' III 5'-TCG GTA CCA TGG CGT CCC CGG CGC TGG CGG CGG CGC TGG CGG TGG CGG CAG CGG ADRA2C-TM1-KpnI (forward) CGG GCC CCA AT-3' III ADRA2C-TM1-SpeI (reverse) 5'-TAC TAG TCA GCA CGG CGA TCA CCA CCA GCA CGT T-3' III

36

Table 4. A list of primers used for mutagenesis in studies II and III. Nucleotides that encode an artificial restriction enzyme recognition site or mutation are highlighted in colour: KpnI (blue), SpeI (red), HindIII (green) and mutation (bold black). GeneEditorTM in vitro Site-Directed Mutagenesis System primers Primer name Primer sequence Study Human ADRA2A (RCExl2.49-xl2.51SCK) 3'-TCG GCC GGC TCG GCT CAA CGC CTT AGT TGC TGG TCT TCA CCA T-5' II Mouse ADRA2A (SCKxl2.49-xl2.51RCE): 5'-GGT CGT TGA TCT CGC ATC TTG GCT CGG CC-3' II ADRA2C-SpeI 5’-GCG CAG CGC CCG ACT AGT CAG CAC GGC GAT-3’ III

ADRA2C (ADRA2A Nter-TM1)-NheI 5´-TGC CAC CGC GCC CGC GCT AGC CTG GCC GCG CGG CGG-3´ III Materials andMethods ADRA2C (ADRA2A Nter-TM1) (restore) 5´-CAC CTG CAG GCT ATA CTG GCC GCG CG-3´ III ADRA2A (ADCA2C Nter-TM1)-NheI 5´-CAG CGT CAC CTG CAG GCT AGC AGG GGT GGC CCG GGC-3´ III ADRA2A (ADCA2C Nter-TM1) (restore) 5´-CCG CGC CCG CGC TAT AAG GGG TGG CCC GGG C-3´ III

PCR-based mutagenesis primers III Primer name Primer sequence III ADRA2A-TM1-KpnI (forward) 5'-TCG GTA CCA TGG GCT CCC TGC AGC CGG ACG CGG GCA ACG CGA GCT GGA A-3' III ADRA2A-TM1-SpeI (reverse) 5'-GAC TAG TGA ACA CGG CGA TGA TGA CGA GCA CGT T-3' III ADRA2A-cterm-SpeI (forward) 5'-AAC TAG TCG CGC GCT CAA GGC GCC CCA AAA CCT CTT-3' III ADRA2A-cterm-HindIII (reverse) 5'-TCG AGA AGC TTT CAC ACG ATC CGC TTC CTG T-3' III ADRA2B-TM1-KpnI (forward) 5'-TCG GTA CCA TGG ACC ACC AGG ACC CCT ACT CCG TGC AGG CCG CAG CGG CCA TA-3' III ADRA2B-TM1-SpeI (reverse) 5'-TAC TAG TCA ACA CAG CCA GGA TGA CCA GCG CGT T-3' III ADRA2B-cterm-SpeI (forward) 5'-AAC TAG TCG CTC GCT GCG CGC CCC TCG GAA-3' III ADRA2B-cterm-HindIII (reverse) 5'-TTA GGA AGC TTT CAC CAG GCC GTC TGG GTC CAC GGG CGG CAC AGG AT-3' III 5'-TCG GTA CCA TGG CGT CCC CGG CGC TGG CGG CGG CGC TGG CGG TGG CGG CAG CGG ADRA2C-TM1-KpnI (forward) CGG GCC CCA AT-3' III ADRA2C-TM1-SpeI (reverse) 5'-TAC TAG TCA GCA CGG CGA TCA CCA CCA GCA CGT T-3' III

Materials and Methods 37 Materials and Methods 37 4.2 Cell culture and transfections 4.2 Cell culture and transfections To express the constructed receptor proteins for subsequent ligand binding assays, To express the constructed receptor proteins for subsequent ligand binding assays, the vector-based constructs were transfected into mammalian cells. Adherent Chinese the vector-based constructs were transfected into mammalian cells. Adherent Chinese hamster ovary (CHO-K1) cells (American Type Culture Collection, Rockville, MD, hamster ovary (CHO-K1) cells (American Type Culture Collection, Rockville, MD, USA) were cultured in -MEM medium (-Minimum Essential Medium, GIBCOTM, USA) were cultured in -MEM medium (-Minimum Essential Medium, GIBCOTM, Invitrogen) supplemented with 26 mM NaHCO3, 50 IU/ml penicillin, 50 g/ml Invitrogen) supplemented with 26 mM NaHCO3, 50 IU/ml penicillin, 50 g/ml streptomycin and 5 % heat-inactivated fetal bovine serum (FBS). Cells were grown streptomycin and 5 % heat-inactivated fetal bovine serum (FBS). Cells were grown o o at 37 C in a humidified atmosphere containing 5 % CO2. The pcDNA3 and at 37 C in a humidified atmosphere containing 5 % CO2. The pcDNA3 and pcDNA3.1(+) -based expression constructs were linearised prior to transfection to pcDNA3.1(+) -based expression constructs were linearised prior to transfection to increase genomic incorporation and to obtain stable expression. The construct increase genomic incorporation and to obtain stable expression. The construct cDNAs were transfected into the cells using either the Lipofectin or the cDNAs were transfected into the cells using either the Lipofectin or the Lipofectamine 2000 reagent kit (Invitrogen Life Technologies Inc., Rockville, MD, Lipofectamine 2000 reagent kit (Invitrogen Life Technologies Inc., Rockville, MD, USA). Selection of transfected cells was performed using either the neomycin USA). Selection of transfected cells was performed using either the neomycin analogue G418 (Geneticin®) (Sigma-Aldrich, St. Louis, MO, USA) or Hygromycin analogue G418 (Geneticin®) (Sigma-Aldrich, St. Louis, MO, USA) or Hygromycin B (Roche Molecular Biochemicals, Mannheim, Germany), depending on the B (Roche Molecular Biochemicals, Mannheim, Germany), depending on the expression vector used. The production of stable CHO cell lines with expression expression vector used. The production of stable CHO cell lines with expression vectors based on pMAMneo for expression of wild-type 2A-, 2B- and 2C- vectors based on pMAMneo for expression of wild-type 2A-, 2B- and 2C- adrenoceptors has been described previously (Marjamäki et al., 1992, Marjamäki et adrenoceptors has been described previously (Marjamäki et al., 1992, Marjamäki et al., 1993, Pohjanoksa et al., 1997). Also the production of Shionogi 115 mouse al., 1993, Pohjanoksa et al., 1997). Also the production of Shionogi 115 mouse mammary tumour cell lines (S115) stably expressing wild-type 2A-, 2B- and 2C- mammary tumour cell lines (S115) stably expressing wild-type 2A-, 2B- and 2C- adrenoceptors (used in study I) has been described previously (Marjamäki et al., adrenoceptors (used in study I) has been described previously (Marjamäki et al., 1992). 1992).

4.3 Reverse transcription-PCR 4.3 Reverse transcription-PCR Total RNA was isolated from cultured CHO cells using the RNeasy Mini Kit Total RNA was isolated from cultured CHO cells using the RNeasy Mini Kit (Qiagen, Valencia, CA, USA), and cDNA was synthesised with random hexamer (Qiagen, Valencia, CA, USA), and cDNA was synthesised with random hexamer primers using the DyNAmoTM cDNA Synthesis Kit (Finnzymes, Espoo, Finland). primers using the DyNAmoTM cDNA Synthesis Kit (Finnzymes, Espoo, Finland). Reverse transcription PCR (RT-PCR) was performed with gene-specific primers Reverse transcription PCR (RT-PCR) was performed with gene-specific primers designed for the chimaeric human 2C-adrenoceptor-based construct containing TM1 designed for the chimaeric human 2C-adrenoceptor-based construct containing TM1 and the N-terminus from the 2B-adrenoceptor. Primers were designed before and and the N-terminus from the 2B-adrenoceptor. Primers were designed before and after the TM1 cut-off using the program PrimerSelect (DNASTAR Inc., Madison, after the TM1 cut-off using the program PrimerSelect (DNASTAR Inc., Madison, WI, USA). Control reactions for RT-PCR were run with RNA from untransfected WI, USA). Control reactions for RT-PCR were run with RNA from untransfected CHO cells. CHO cells.

4.4 Membrane preparation 4.4 Membrane preparation Confluent CHO cells (>90 %) were detached from culture flasks with 0.25 % Confluent CHO cells (>90 %) were detached from culture flasks with 0.25 % trypsin/0.02 % ethylenediamine tetraacetic acid (EDTA), centrifuged at 1500 rpm for 5 trypsin/0.02 % ethylenediamine tetraacetic acid (EDTA), centrifuged at 1500 rpm for 5 min at 4 oC, washed twice with chilled phosphate-buffered saline (PBS), and stored min at 4 oC, washed twice with chilled phosphate-buffered saline (PBS), and stored frozen at -20 oC until used. The harvested cell pellets were suspended in ice-cold frozen at -20 oC until used. The harvested cell pellets were suspended in ice-cold homogenization buffer (10 mM Tris, 0.1 mM EDTA, 0.32 mM sucrose, pH 7.5), homogenization buffer (10 mM Tris, 0.1 mM EDTA, 0.32 mM sucrose, pH 7.5), followed by homogenization with an Ultra-Turrax homogenizer (model T25, Janke & followed by homogenization with an Ultra-Turrax homogenizer (model T25, Janke & Kunkel, Staufen, Germany) (3 x 10 s at 800 rpm.). Cell homogenates were centrifuged Kunkel, Staufen, Germany) (3 x 10 s at 800 rpm.). Cell homogenates were centrifuged at 108 g for 15 min at 4 oC, and supernatants were collected. The pooled supernatants at 108 g for 15 min at 4 oC, and supernatants were collected. The pooled supernatants

38 Materials and Methods 38 Materials and Methods were further centrifuged at 50 227 g for 30 min at 4 oC. Membrane pellets were washed were further centrifuged at 50 227 g for 30 min at 4 oC. Membrane pellets were washed with sucrose-free Tris-EDTA buffer, aliquoted in homogenization buffer, and stored at with sucrose-free Tris-EDTA buffer, aliquoted in homogenization buffer, and stored at -70 oC. Protein concentrations were determined with the method of Bradford (1976) -70 oC. Protein concentrations were determined with the method of Bradford (1976) using bovine serum albumin as reference. using bovine serum albumin as reference.

4.5 Ligand binding assays 4.5 Ligand binding assays The radioligand binding assay is a relatively simple but powerful tool to study The radioligand binding assay is a relatively simple but powerful tool to study receptor-ligand interactions of GPCRs. There are three basic types of experiments receptor-ligand interactions of GPCRs. There are three basic types of experiments based on this technique: (1) saturation experiments from which the affinity of the based on this technique: (1) saturation experiments from which the affinity of the radioligand for the receptor and the binding site density can be determined; (2) radioligand for the receptor and the binding site density can be determined; (2) competition experiments where the affinity of a competing, unlabeled compound for competition experiments where the affinity of a competing, unlabeled compound for the receptor can be determined; and (3) kinetic experiments from which the association the receptor can be determined; and (3) kinetic experiments from which the association and dissociation rate constants for radioligand binding can be determined (reviewed and dissociation rate constants for radioligand binding can be determined (reviewed extensively in Bylund and Toews, 1993). extensively in Bylund and Toews, 1993).

4.5.1 Saturation experiments 4.5.1 Saturation experiments In a saturation experiment, a receptor preparation (R) is incubated together with a In a saturation experiment, a receptor preparation (R) is incubated together with a radioligand (L) for a period of time until equilibrium is reached with regard to radioligand (L) for a period of time until equilibrium is reached with regard to formation of a receptor-radioligand complex (RL). This can be represented by: formation of a receptor-radioligand complex (RL). This can be represented by:

L + R RL L + R RL

The amount of RL is measured as a function of the free radioligand concentration The amount of RL is measured as a function of the free radioligand concentration [L]. The parameters obtained from a saturation experiment are the affinity, expressed [L]. The parameters obtained from a saturation experiment are the affinity, expressed as the dissociation constant, Kd (receptor affinity), and the number of binding sites as the dissociation constant, Kd (receptor affinity), and the number of binding sites present in the receptor preparation, Bmax (maximal binding capacity). present in the receptor preparation, Bmax (maximal binding capacity). In the present studies, transfected cell lines were screened for binding of three In the present studies, transfected cell lines were screened for binding of three 3 3 2-adrenoceptor specific antagonist radioligands [methyl- H]rauwolscine (NEN Life 2-adrenoceptor specific antagonist radioligands [methyl- H]rauwolscine (NEN Life Science Products, Inc., Boston, MA, USA), [ethyl-3H]RS79948-197 and Science Products, Inc., Boston, MA, USA), [ethyl-3H]RS79948-197 and [3H]RX821002 (Amersham Pharmacia Biotech, Buckinghamshire, UK) (Table 5). [3H]RX821002 (Amersham Pharmacia Biotech, Buckinghamshire, UK) (Table 5). Saturation binding assays were performed in 50 mM potassium phosphate buffer Saturation binding assays were performed in 50 mM potassium phosphate buffer (pH 7.4) as reported previously (Halme et al., 1995, Marjamäki et al., 1998). Either (pH 7.4) as reported previously (Halme et al., 1995, Marjamäki et al., 1998). Either whole-cell homogenates (20-120 g of protein per sample) or membrane whole-cell homogenates (20-120 g of protein per sample) or membrane preparations (4-20 g of protein per sample) were incubated in 250 l of buffer with preparations (4-20 g of protein per sample) were incubated in 250 l of buffer with serial dilutions of radioligand at 25 oC. Non-specific binding (NSB) was defined in serial dilutions of radioligand at 25 oC. Non-specific binding (NSB) was defined in parallel assay tubes in the presence of 10 M phentolamine, an 2-adrenoceptor parallel assay tubes in the presence of 10 M phentolamine, an 2-adrenoceptor antagonist. After 30 min incubation, reactions were terminated and bound antagonist. After 30 min incubation, reactions were terminated and bound radioactivity was separated by rapid filtration through pre-soaked glass fibre filters radioactivity was separated by rapid filtration through pre-soaked glass fibre filters (Whatman GF/B) using a Brandel M-48R cell harvester (Medical Research and (Whatman GF/B) using a Brandel M-48R cell harvester (Medical Research and Development Laboratories, Gaithersburg, MD, USA) in a cold room (6 oC). Filters Development Laboratories, Gaithersburg, MD, USA) in a cold room (6 oC). Filters were washed twice with 5 ml of ice-cold buffer (50 mM Tris-HCl, 10 mM EDTA, were washed twice with 5 ml of ice-cold buffer (50 mM Tris-HCl, 10 mM EDTA, pH 7.4) and placed into the bottom of scintillation vials with OptiPhase ‘HiSafe’ III pH 7.4) and placed into the bottom of scintillation vials with OptiPhase ‘HiSafe’ III (Wallac, Turku, Finland). The bound radioactivity on the filters was quantitated by (Wallac, Turku, Finland). The bound radioactivity on the filters was quantitated by

Materials and Methods 39 Materials and Methods 39 liquid scintillation counting (Wallac 1410) after 8 h incubation at room temperature liquid scintillation counting (Wallac 1410) after 8 h incubation at room temperature (RT). Specific binding was defined as the difference between total binding and (RT). Specific binding was defined as the difference between total binding and NSB. For each of the investigated cell lines, saturation experiments were performed NSB. For each of the investigated cell lines, saturation experiments were performed in duplicate or triplicate and repeated at least three times. The Kd and Bmax values in duplicate or triplicate and repeated at least three times. The Kd and Bmax values were calculated from the results of saturation binding experiments using the were calculated from the results of saturation binding experiments using the GraphPad Prism software package (GrapPad Prism Software, San Diego, CA, USA). GraphPad Prism software package (GrapPad Prism Software, San Diego, CA, USA).

Table 5. Details of the employed radiolabelled compounds. * indicates the position of the Table 5. Details of the employed radiolabelled compounds. * indicates the position of the radioactive isotope. radioactive isotope. Radioligand Chemical structure Specific Binding Radioligand Chemical structure Specific Binding radioactivity preference radioactivity preference

[Methyl- 77.5-71 2A ≈ 2B < 2C [Methyl- 77.5-71 2A ≈ 2B < 2C 3H]Rauwolscine Ci/mmol 3H]Rauwolscine Ci/mmol

[Ethyl- 88-81 Ci/mmol 2A ≈ 2B ≈ 2C [Ethyl- 88-81 Ci/mmol 2A ≈ 2B ≈ 2C 3H]RS79948-197 3H]RS79948-197

3 3 [ H]RX821002 60-41 Ci/mmol 2A ≈ 2C > 2B [ H]RX821002 60-41 Ci/mmol 2A ≈ 2C > 2B

4.5.2 Competition experiments 4.5.2 Competition experiments In a competition experiment, the amount of receptors and the radioligand concentration In a competition experiment, the amount of receptors and the radioligand concentration are kept constant, while the concentration of the unlabeled competing drug is varied. are kept constant, while the concentration of the unlabeled competing drug is varied. When the concentration of the unlabeled compound is zero, all receptors in the sample When the concentration of the unlabeled compound is zero, all receptors in the sample are free to be occupied by the radioligand. When the concentration of the unlabeled are free to be occupied by the radioligand. When the concentration of the unlabeled compound is increased, the unlabeled compound will compete with the radioligand for compound is increased, the unlabeled compound will compete with the radioligand for the receptor binding sites, decreasing the amount of receptor-radioligand complex the receptor binding sites, decreasing the amount of receptor-radioligand complex being present. The parameters obtained from competition experiments are IC50 values being present. The parameters obtained from competition experiments are IC50 values

40 Materials and Methods 40 Materials and Methods

(the concentration of the competing drug that inhibits 50 % of the specific binding in a (the concentration of the competing drug that inhibits 50 % of the specific binding in a given experiment) and inhibition constants, Ki (the affinity of the competitor for the given experiment) and inhibition constants, Ki (the affinity of the competitor for the receptor). An IC50 is converted into a Ki-value using the Cheng-Prusoff equation, receptor). An IC50 is converted into a Ki-value using the Cheng-Prusoff equation, where [L] represents the free radioligand concentration and Kd represents the where [L] represents the free radioligand concentration and Kd represents the dissociation constant of the radioligand (Cheng and Prusoff, 1973): dissociation constant of the radioligand (Cheng and Prusoff, 1973):

Ligands tested in the competition binding assays were chosen mainly on the basis of Ligands tested in the competition binding assays were chosen mainly on the basis of their chemical structure and known preference for different 2-adrenoceptor subtypes their chemical structure and known preference for different 2-adrenoceptor subtypes (Table 6). The competition binding assays were implemented using either a Beckman (Table 6). The competition binding assays were implemented using either a Beckman Biomek 2000 Laboratory Automation Workstation (Beckman Instruments Inc., Palo Biomek 2000 Laboratory Automation Workstation (Beckman Instruments Inc., Palo Alto, CA, USA) with 96-well plate format, or a MultiScreen Vacuum Manifold system Alto, CA, USA) with 96-well plate format, or a MultiScreen Vacuum Manifold system (Millipore Corporation, Bedford, MA, USA) with Millipore MultiScreen MSFBN 96- (Millipore Corporation, Bedford, MA, USA) with Millipore MultiScreen MSFBN 96- well glass fibre filtration plates. The experiments were performed in 50 mM potassium well glass fibre filtration plates. The experiments were performed in 50 mM potassium phosphate buffer (pH 7.4) using radioligands at concentrations close to their affinity phosphate buffer (pH 7.4) using radioligands at concentrations close to their affinity constants (Kd) for each particular receptor, and 6-8 serial dilutions of the competing constants (Kd) for each particular receptor, and 6-8 serial dilutions of the competing ligands, which were incubated with 2-10 g of cell membrane protein per sample. ligands, which were incubated with 2-10 g of cell membrane protein per sample. After 30 min incubation at RT, reactions were terminated by rapid vacuum filtration. After 30 min incubation at RT, reactions were terminated by rapid vacuum filtration. The filter plates were washed three times with ice-cold potassium phosphate buffer, The filter plates were washed three times with ice-cold potassium phosphate buffer, dried and impregnated with Meltilex B/HS scintillation wax (1205-422, Wallac, Turku, dried and impregnated with Meltilex B/HS scintillation wax (1205-422, Wallac, Turku, Finland) or 50 l SuperMix scintillation cocktail (Wallac), depending on the system Finland) or 50 l SuperMix scintillation cocktail (Wallac), depending on the system used. The incorporated radioactivity was determined in a Wallac 1205 or Wallac 1450 used. The incorporated radioactivity was determined in a Wallac 1205 or Wallac 1450 Betaplate liquid scintillation counter. The apparent affinities (Ki) of the competing Betaplate liquid scintillation counter. The apparent affinities (Ki) of the competing ligands at each receptor were determined via non-linear regression analysis (GraphPad ligands at each receptor were determined via non-linear regression analysis (GraphPad Prism), assuming homogeneous one-site binding. The statistical significance of Prism), assuming homogeneous one-site binding. The statistical significance of differences between the apparent affinities of two receptors was evaluated with differences between the apparent affinities of two receptors was evaluated with unpaired t-tests. unpaired t-tests.

Materials and Methods 41 Materials and Methods 41

Table 6. Ligands used in competition binding assays Table 6. Ligands used in competition binding assays Ligand Chemical structure Function Study Ligand Chemical structure Function Study (-)-Adrenaline OH Agonist I, II (-)-Adrenaline OH Agonist I, II H H OH N OH N CH3 CH3

OH OH 2-Amino-1-phenylethanol OH Agonist I 2-Amino-1-phenylethanol OH Agonist I

NH2 NH2

ARC239 3CH CH3 Antagonist I, II, III ARC239 3CH CH3 Antagonist I, II, III O O

N CH N CH N O 3 N O 3 O N O N

Atipamezole CH3 Antagonist I, II, III Atipamezole CH3 Antagonist I, II, III H H N N

N N Brimonidine (UK-14.304) N Agonist I Brimonidine (UK-14.304) N Agonist I N N

N N N N N N H H H H Br Br

Chlorpromazine CH3 Antagonist I, II, III Chlorpromazine CH3 Antagonist I, II, III N N 3CH 3CH N Cl N Cl

S S Clonidine Cl Agonist I Clonidine Cl Agonist I H H H H N N N N

N N Cl Cl Clozapine CH3 Antagonist I Clozapine CH3 Antagonist I N N

N N N N Cl Cl

N N H H Table continued on next page. Table continued on next page.

42 Materials and Methods 42 Materials and Methods

Ligand Chemical structure Function Study Ligand Chemical structure Function Study

Dexmedetomidine CH3 CH3 Agonist I Dexmedetomidine CH3 CH3 Agonist I

N CH3 N CH3 H H N N H H

Dopamine OH NH2 Agonist I, II Dopamine OH NH2 Agonist I, II

OH OH

Doxazosin CH3 NH2 Antagonist II Doxazosin CH3 NH2 Antagonist II O O N N O O O N N O N N H H CH N CH N 3 O 3 O O O Idazoxan N Antagonist I, III Idazoxan N Antagonist I, III O O N N H H O O MK-912 (L657.743) Antagonist I, II MK-912 (L657.743) Antagonist I, II N N O O CH3 CH3 N N

O O N N

CH3 CH3 (-)-Noradrenaline OH Agonist I, II (-)-Noradrenaline OH Agonist I, II

OH NH2 OH NH2

OH OH

Oxymetazoline OH CH3 Agonist I Oxymetazoline OH CH3 Agonist I 3CH N 3CH N

3CH 3CH CH N CH N 3 H 3 H CH3 CH3

Prazosin CH3 NH2 Antagonist I, II, III Prazosin CH3 NH2 Antagonist I, II, III O O N N

O N N O N N

CH3 N CH3 N O O O O Table continued on next page. Table continued on next page.

Materials and Methods 43 Materials and Methods 43

Ligand Chemical structure Function Study Ligand Chemical structure Function Study Rauwolscine Antagonist I, II, III Rauwolscine Antagonist I, II, III N N H H N H N H H H H H

O O CH CH 3 OH 3 OH O O RS79948-197 O Antagonist II RS79948-197 O Antagonist II 3CH 3CH N N H H H H

H H N N 3CH S 3CH S O O O O Spiperone Antagonist I, III Spiperone Antagonist I, III

N N N NH N NH

F O F O O O Spiroxatrine Antagonist I, III Spiroxatrine Antagonist I, III

N N N NH N NH O O O O O O

WB-4101 CH3 Antagonist I, II WB-4101 CH3 Antagonist I, II O O O O H H N N O O O O O O CH3 CH3 Yohimbine Antagonist I, II Yohimbine Antagonist I, II N N H H N H N H H H H H

O O CH CH 3 OH 3 OH O O

44 Materials and Methods 44 Materials and Methods

4.5.3 Functional [35S]GTPS binding assay 4.5.3 Functional [35S]GTPS binding assay The [35S]GTPS binding assay is a simple method to monitor the first step of the The [35S]GTPS binding assay is a simple method to monitor the first step of the intracellular signalling cascade after receptor activation by exposure to an agonist intracellular signalling cascade after receptor activation by exposure to an agonist ligand. Radiolabelled [35S]GTPS is a poorly hydrolysable GTP analogue, which binds ligand. Radiolabelled [35S]GTPS is a poorly hydrolysable GTP analogue, which binds almost irreversibly to G-protein -subunits upon receptor activation. almost irreversibly to G-protein -subunits upon receptor activation. Functional [35S]GTPS binding assays were conducted with membrane suspensions Functional [35S]GTPS binding assays were conducted with membrane suspensions expressing zebrafish and human 2-adrenoceptor subtypes (study I). The experiments expressing zebrafish and human 2-adrenoceptor subtypes (study I). The experiments were carried out in a reaction mixture based on 50 mM Tris-HCl buffer and containing were carried out in a reaction mixture based on 50 mM Tris-HCl buffer and containing 1 mM EDTA, 5 mM MgCl2, 1 mM dithiothreitol (DTT), 150 mM NaCl, 1 M GDP 1 mM EDTA, 5 mM MgCl2, 1 mM dithiothreitol (DTT), 150 mM NaCl, 1 M GDP and 30 M ascorbic acid (pH 7.4) as final concentrations. An aliquot of 5-10 g of the and 30 M ascorbic acid (pH 7.4) as final concentrations. An aliquot of 5-10 g of the thawed membrane protein suspension per sample was added to the reaction buffer in a thawed membrane protein suspension per sample was added to the reaction buffer in a total volume of 0.3 ml. The samples were preincubated for 30 min at RT with serial total volume of 0.3 ml. The samples were preincubated for 30 min at RT with serial dilutions of the agonists before 0.1 nM [35S]GTPS (NEN Life Science Products, Inc., dilutions of the agonists before 0.1 nM [35S]GTPS (NEN Life Science Products, Inc., Boston, MA, USA) was added for the 60 min incubation period. The incubations were Boston, MA, USA) was added for the 60 min incubation period. The incubations were terminated by rapid filtration through Whatman GF/B filters, the filters were rinsed terminated by rapid filtration through Whatman GF/B filters, the filters were rinsed (with cold 20 mM Tris-HCl, 1 mM EDTA, 5 mM MgCl2, pH 7.4) and the bound (with cold 20 mM Tris-HCl, 1 mM EDTA, 5 mM MgCl2, pH 7.4) and the bound radioactivity was counted in a liquid scintillation counter. Agonist-induced stimulation radioactivity was counted in a liquid scintillation counter. Agonist-induced stimulation of [35S]GTPS binding was determined for the natural ligands (-)-adrenaline and (-)- of [35S]GTPS binding was determined for the natural ligands (-)-adrenaline and (-)- noradrenaline and the synthetic agonists brimonidine, oxymetazoline and noradrenaline and the synthetic agonists brimonidine, oxymetazoline and dexmedetomidine, displaying different efficacies and potencies at the three human 2- dexmedetomidine, displaying different efficacies and potencies at the three human 2- adrenoceptor subtypes. adrenoceptor subtypes.

4.6 Molecular modelling and ligand docking 4.6 Molecular modelling and ligand docking Molecular modelling is a computational technique to mimic the atomic-level Molecular modelling is a computational technique to mimic the atomic-level description of the investigated molecules. In structural biology and chemistry, description of the investigated molecules. In structural biology and chemistry, molecular modelling is used to visualize the three-dimensional shapes of the proteins molecular modelling is used to visualize the three-dimensional shapes of the proteins of interest and interacting ligands. If the structure of the target protein is not known, of interest and interacting ligands. If the structure of the target protein is not known, homology modelling may be applied by using sequence alignment comparisons and a homology modelling may be applied by using sequence alignment comparisons and a homologous protein with a known structure as template. For validation of homology homologous protein with a known structure as template. For validation of homology models, it is important that experimental findings correlate with the structural features models, it is important that experimental findings correlate with the structural features of the modelled protein. of the modelled protein. Structural models of various orthologous and paralogous 2-adrenoceptors (studies Structural models of various orthologous and paralogous 2-adrenoceptors (studies I and II) were constructed using the X-ray structure of bovine rhodopsin as template I and II) were constructed using the X-ray structure of bovine rhodopsin as template (Palczewski et al., 2000). The structure of an inactive state conformation of bovine (Palczewski et al., 2000). The structure of an inactive state conformation of bovine rhodopsin, bound with 11-cis-retinal, solved at 2.6 Å resolution, was obtained from the rhodopsin, bound with 11-cis-retinal, solved at 2.6 Å resolution, was obtained from the Protein Data Bank (PDB) (PDBcodes 1F88, 1HZX, 1LPH) (Palczewski et al., 2000, Protein Data Bank (PDB) (PDBcodes 1F88, 1HZX, 1LPH) (Palczewski et al., 2000, Teller et al., 2001, Okada et al., 2002, Li et al., 2004). A modified human 2A- Teller et al., 2001, Okada et al., 2002, Li et al., 2004). A modified human 2A- adrenoceptor structural model was reconstructed in study III based on the new X-ray adrenoceptor structural model was reconstructed in study III based on the new X-ray structure of the human 2-adrenoceptor (PDB code 2RH1) (Cherezov et al., 2007, structure of the human 2-adrenoceptor (PDB code 2RH1) (Cherezov et al., 2007, Rasmussen et al., 2007). The human 2-adrenoceptor structure was also solved in an Rasmussen et al., 2007). The human 2-adrenoceptor structure was also solved in an inactive state in complex with the inverse agonist, carazolol. The human 2- inactive state in complex with the inverse agonist, carazolol. The human 2- adrenoceptor structure was believed to provide a more accurate molecular model of the adrenoceptor structure was believed to provide a more accurate molecular model of the 2-adrenoceptors than the bovine rhodopsin structure, as it belongs to the same 2-adrenoceptors than the bovine rhodopsin structure, as it belongs to the same subfamily of amine GPCRs and shares with the 2-adrenoceptors both a higher level of subfamily of amine GPCRs and shares with the 2-adrenoceptors both a higher level of

Materials and Methods 45 Materials and Methods 45 sequence identity – 37-43 % in the TM regions – and higher structural similarities e.g. sequence identity – 37-43 % in the TM regions – and higher structural similarities e.g. with regard to the amino acids involved in the recognition of catecholamines. with regard to the amino acids involved in the recognition of catecholamines. In order to construct homology models of the 2-adrenoceptor subtypes, multiple In order to construct homology models of the 2-adrenoceptor subtypes, multiple pairwise sequence alignments of the 2-adrenoceptor sequences with the template pairwise sequence alignments of the 2-adrenoceptor sequences with the template sequence were first generated with the program MALIGN (Johnson and Overington, sequence were first generated with the program MALIGN (Johnson and Overington, 1993) in the Bodil modelling environment (http://users.abo.fi/bodil/about.php) 1993) in the Bodil modelling environment (http://users.abo.fi/bodil/about.php) (Lehtonen et al., 2004). Regions that were too dissimilar to be aligned, i.e. the N- (Lehtonen et al., 2004). Regions that were too dissimilar to be aligned, i.e. the N- terminal segment and the third intracellular loop, were deleted from the alignments and terminal segment and the third intracellular loop, were deleted from the alignments and were not included in the models. The modelling procedures of 2-adrenoceptors were were not included in the models. The modelling procedures of 2-adrenoceptors were performed using the program MODELLER (versions 6.0, 8.0 and 8.2) (Sali and performed using the program MODELLER (versions 6.0, 8.0 and 8.2) (Sali and Blundell, 1993) based on the template structures. Here, the deleted sequence segments Blundell, 1993) based on the template structures. Here, the deleted sequence segments were corrected manually in order to ensure that the modelled conformations of the 2- were corrected manually in order to ensure that the modelled conformations of the 2- adrenoceptors were relaxed and free of any undesirable atomic contacts. Both in the adrenoceptors were relaxed and free of any undesirable atomic contacts. Both in the bovine rhodopsin and the human 2-adrenoceptor structures, two cysteines, C3.25 in bovine rhodopsin and the human 2-adrenoceptor structures, two cysteines, C3.25 in TM3 and Cxl2.50 in XL2, form a disulphide bridge, which constrains XL2 to fold TM3 and Cxl2.50 in XL2, form a disulphide bridge, which constrains XL2 to fold above the binding cavity. In the rhodopsin structure, XL2 directly interacts with bound above the binding cavity. In the rhodopsin structure, XL2 directly interacts with bound 11-cis-retinal, and in the 2-adrenoceptor structure, it interacts with carazolol 11-cis-retinal, and in the 2-adrenoceptor structure, it interacts with carazolol (Palczewski et al., 2000, Cherezov et al., 2007). In all 2-adrenoceptors, cysteines are (Palczewski et al., 2000, Cherezov et al., 2007). In all 2-adrenoceptors, cysteines are present at positions 3.25 and xl2.50, which suggests a similar position of XL2 as well present at positions 3.25 and xl2.50, which suggests a similar position of XL2 as well as direct interactions of XL2 with bound ligands in the model structures of the 2- as direct interactions of XL2 with bound ligands in the model structures of the 2- adrenoceptors. adrenoceptors. In the modelling procedures, 10-20 models for each of the receptor subtypes were In the modelling procedures, 10-20 models for each of the receptor subtypes were constructed that varied in the side-chain conformations of the amino acids present in constructed that varied in the side-chain conformations of the amino acids present in the putative ligand binding site. For those amino acids that are conserved among the the putative ligand binding site. For those amino acids that are conserved among the 2-adrenoceptors and the employed template, the conformations of the side-chains 2-adrenoceptors and the employed template, the conformations of the side-chains were used mainly as found in the template. For non-conserved residues, variable side- were used mainly as found in the template. For non-conserved residues, variable side- chain conformations were produced by MODELLER (for details, see Xhaard et al., chain conformations were produced by MODELLER (for details, see Xhaard et al., 2005). 2005). Ligands docked into the structural models of the 2-adrenoceptors varied in their Ligands docked into the structural models of the 2-adrenoceptors varied in their size, structure, function and in the level of preference for individual 2-adrenoceptor size, structure, function and in the level of preference for individual 2-adrenoceptor subtypes. Docking simulations were performed using either manual (I and II) or subtypes. Docking simulations were performed using either manual (I and II) or automated docking (III), where the program GOLD (version 4.0) (Jones et al., 1997) automated docking (III), where the program GOLD (version 4.0) (Jones et al., 1997) was allowed to perform automated docking. The manual docking simulations for was allowed to perform automated docking. The manual docking simulations for antagonists were performed in two different ways; first, based on the classical antagonists were performed in two different ways; first, based on the classical hypothesis of ion pair formation between the side-chain oxygen of D3.32 and the hypothesis of ion pair formation between the side-chain oxygen of D3.32 and the protonated nitrogen atom of the ligands, and secondly, using an alternative binding protonated nitrogen atom of the ligands, and secondly, using an alternative binding mode hypothesis based on cation- interactions with aromatic residues, whereby mode hypothesis based on cation- interactions with aromatic residues, whereby ligands interact with D3.32 via carboxylate-aromatic interactions (Xhaard et al., 2005). ligands interact with D3.32 via carboxylate-aromatic interactions (Xhaard et al., 2005).

46 Results 46 Results

5 RESULTS 5 RESULTS

5.1 Structural and pharmacological properties of the zebrafish 2- 5.1 Structural and pharmacological properties of the zebrafish 2- adrenoceptors in comparison with the human orthologues (I) adrenoceptors in comparison with the human orthologues (I)

5.1.1 Characterization of receptor expression in CHO cells 5.1.1 Characterization of receptor expression in CHO cells In saturation binding assays, the binding affinities of all three radioligands, [methyl- In saturation binding assays, the binding affinities of all three radioligands, [methyl- 3H]rauwolscine, [ethyl-3H]RS79948-197 and [3H]RX821002, were observed to be in 3H]rauwolscine, [ethyl-3H]RS79948-197 and [3H]RX821002, were observed to be in the same low nanomolar range towards the five zebrafish 2-adrenoceptors and their the same low nanomolar range towards the five zebrafish 2-adrenoceptors and their human orthologs (Table 7). Receptor densities of the studied cell lines are summarized human orthologs (Table 7). Receptor densities of the studied cell lines are summarized in Table 8. In the recombinant CHO cell lines, the overall binding characteristics of in Table 8. In the recombinant CHO cell lines, the overall binding characteristics of [ethyl-3H]RS79948-197 were observed to be comparable with values obtained in [ethyl-3H]RS79948-197 were observed to be comparable with values obtained in membrane preparations from zebrafish brain homogenate, where material collected membrane preparations from zebrafish brain homogenate, where material collected from 21 pooled zebrafish brains yielded a Kd of 0.1 nM and a Bmax of 475 fmol/mg from 21 pooled zebrafish brains yielded a Kd of 0.1 nM and a Bmax of 475 fmol/mg protein. protein.

Table 7. Binding affinities of three 2-adrenoceptor preferring radioligands in CHO cell Table 7. Binding affinities of three 2-adrenoceptor preferring radioligands in CHO cell membranes expressing human and zebrafish 2-adrenoceptors. membranes expressing human and zebrafish 2-adrenoceptors. Receptor (species, subtype) Receptor (species, subtype) Kd (nM) Kd (nM)

Radioligand h 2A z 2A h 2B z 2B h 2C z 2C z 2Da z 2Db Radioligand h 2A z 2A h 2B z 2B h 2C z 2C z 2Da z 2Db [methyl-3H]rauwolscine 0.9 0.2 1.3 6.3 0.3 1.5 0.3 0.05 [methyl-3H]rauwolscine 0.9 0.2 1.3 6.3 0.3 1.5 0.3 0.05 [3H]RX821002 1.1 5.0 4.6 3.3 0.9 1.7 6.0 6.8 [3H]RX821002 1.1 5.0 4.6 3.3 0.9 1.7 6.0 6.8 [ethyl-3H]RS79948-197 0.2 0.6 0.3 0.3 0.2 0.1 0.3 0.3 [ethyl-3H]RS79948-197 0.2 0.6 0.3 0.3 0.2 0.1 0.3 0.3 Results are expressed as mean values from at least three independent saturation binding Results are expressed as mean values from at least three independent saturation binding experiments performed in duplicate or triplicate. h = human, z = zebrafish experiments performed in duplicate or triplicate. h = human, z = zebrafish

Table 8. Receptor densities in recombinant CHO cell lines obtained from saturation binding Table 8. Receptor densities in recombinant CHO cell lines obtained from saturation binding experiments using [ethyl-3H]RS79948-197 as radioligand. experiments using [ethyl-3H]RS79948-197 as radioligand. Receptor (species, subtype) Receptor (species, subtype) Bmax (fmol/mg protein) Bmax (fmol/mg protein) h 2A z 2A h 2B z 2B h 2C z 2C z 2Da z 2Db h 2A z 2A h 2B z 2B h 2C z 2C z 2Da z 2Db 5 500 ± 300 3 100 ± 100 1 300 ± 100 460 ± 10 1 700 ± 100 550 ± 10 2 300 ± 200 2 300 ± 300 5 500 ± 300 3 100 ± 100 1 300 ± 100 460 ± 10 1 700 ± 100 550 ± 10 2 300 ± 200 2 300 ± 300 Results are expressed as means ± s.e.m. of at least three independent experiments. h = human, z Results are expressed as means ± s.e.m. of at least three independent experiments. h = human, z = zebrafish = zebrafish

Results 47 Results 47

5.1.2 Competition binding assays and cluster analysis of the receptors and ligands 5.1.2 Competition binding assays and cluster analysis of the receptors and ligands Binding affinities of unlabelled compounds were determined in competition binding Binding affinities of unlabelled compounds were determined in competition binding experiments performed mainly with the 2-adrenoceptor antagonist radioligand experiments performed mainly with the 2-adrenoceptor antagonist radioligand [3H]RX821002, but also [methyl-3H]rauwolscine and [ethyl-3H]RS79948-197 were [3H]RX821002, but also [methyl-3H]rauwolscine and [ethyl-3H]RS79948-197 were used in some instances. In the conversion of IC50 into Ki values with the Cheng-Prusoff used in some instances. In the conversion of IC50 into Ki values with the Cheng-Prusoff equation (Cheng and Prusoff, 1973), different Kd values for different radioligands and equation (Cheng and Prusoff, 1973), different Kd values for different radioligands and receptors were applied. The affinities of 20 structurally different ligands at three receptors were applied. The affinities of 20 structurally different ligands at three human and five zebrafish 2-adrenoceptors are summarized in Table 9 as apparent Ki human and five zebrafish 2-adrenoceptors are summarized in Table 9 as apparent Ki values. These radioligand binding assays were performed with receptors expressed in values. These radioligand binding assays were performed with receptors expressed in CHO-K1 cells, with the exception of clonidine, dexmedetomidine, oxymetazoline, CHO-K1 cells, with the exception of clonidine, dexmedetomidine, oxymetazoline, WB-4101, dopamine and (-)-noradrenaline binding assays with the human 2A- WB-4101, dopamine and (-)-noradrenaline binding assays with the human 2A- adrenoceptor, which were conducted with receptors expressed in the S115 host cell adrenoceptor, which were conducted with receptors expressed in the S115 host cell line. This difference in assay performance should not have any effect on the outcome, line. This difference in assay performance should not have any effect on the outcome, as 11 ligands were tested with 2A-adrenoceptors in both CHO-K1 and S115 cell lines as 11 ligands were tested with 2A-adrenoceptors in both CHO-K1 and S115 cell lines with an excellent correlation between the host cell types (Figure 9). with an excellent correlation between the host cell types (Figure 9). Cell line correlation Cell line correlation

4 4

3 3

2 2 (CHO-K1) (CHO-K1) i 1 i 1

log K log 0 K log 0 r=0.96 r=0.96 -1 -1 -1 0 1 2 3 4 -1 0 1 2 3 4 log Ki (S115) log Ki (S115) Figure 9. Correlation of ligand binding affinities of a set of 11 ligands obtained from CHO-K1 Figure 9. Correlation of ligand binding affinities of a set of 11 ligands obtained from CHO-K1 and S115 host cell lines stably expressing human 2A-adrenoceptors. and S115 host cell lines stably expressing human 2A-adrenoceptors.

The zebrafish and human 2-adrenoceptors share generally similar ligand binding The zebrafish and human 2-adrenoceptors share generally similar ligand binding properties as seen in Table 9, i.e. the ligand binding affinities show greater similarities properties as seen in Table 9, i.e. the ligand binding affinities show greater similarities between receptor orthologues in comparison to the receptor paralogues. For receptor between receptor orthologues in comparison to the receptor paralogues. For receptor orthologues, the largest differences, up to 40-fold, are seen for chlorpromazine and orthologues, the largest differences, up to 40-fold, are seen for chlorpromazine and spiroxatrine towards the human and zebrafish 2B-adrenoceptors. Whereas for receptor spiroxatrine towards the human and zebrafish 2B-adrenoceptors. Whereas for receptor paralogues, the largest differences, up to 100-fold, are seen for ARC239, paralogues, the largest differences, up to 100-fold, are seen for ARC239, oxymetazoline, spiperone, spiroxatrine and chlorpromazine, wherein the human and oxymetazoline, spiperone, spiroxatrine and chlorpromazine, wherein the human and zebrafish 2A-adrenoceptors bind these compounds with clearly lower affinity than the zebrafish 2A-adrenoceptors bind these compounds with clearly lower affinity than the B and C subtypes. For the zebrafish 2Da- and 2Db-adrenoceptors, the pharmacological B and C subtypes. For the zebrafish 2Da- and 2Db-adrenoceptors, the pharmacological binding profiles were observed to be almost identical (r=0.98; Figure 10) as expected binding profiles were observed to be almost identical (r=0.98; Figure 10) as expected based on their close phylogenetic distance and high sequence similarity, with 75 % of based on their close phylogenetic distance and high sequence similarity, with 75 % of the amino acids being identical within their entire sequences (Ruuskanen et al., 2004). the amino acids being identical within their entire sequences (Ruuskanen et al., 2004).

48 Results 48 Results

For comparison of the binding affinity data of Table 9, two independent clustering For comparison of the binding affinity data of Table 9, two independent clustering methods were applied: the binary tree approach (see Figure 2A in I) and principal methods were applied: the binary tree approach (see Figure 2A in I) and principal component analysis (PCA) (Figure 2B in I). From these visualizations, the close component analysis (PCA) (Figure 2B in I). From these visualizations, the close clustering of the zebrafish 2Da- and 2Db-adrenoceptors can be clearly seen, but for the clustering of the zebrafish 2Da- and 2Db-adrenoceptors can be clearly seen, but for the human and zebrafish 2B- and 2C-adrenoceptors, the pairing is not resolved based on human and zebrafish 2B- and 2C-adrenoceptors, the pairing is not resolved based on the binding affinity data. For the 20 investigated ligands, the binary tree approach and the binding affinity data. For the 20 investigated ligands, the binary tree approach and PCA clustering of the binding affinity data (Table 9) resulted in the segregation of the PCA clustering of the binding affinity data (Table 9) resulted in the segregation of the ligands based on their chemical structures and functional characteristics (see Figure 4 ligands based on their chemical structures and functional characteristics (see Figure 4 in I). In the binary tree-based clustering, all agonist ligands (clonidine, oxymetazoline, in I). In the binary tree-based clustering, all agonist ligands (clonidine, oxymetazoline, brimonidine, 2-amino-1-phenylethanol, dopamine, (-)-adrenaline and (-)- brimonidine, 2-amino-1-phenylethanol, dopamine, (-)-adrenaline and (-)- noradrenaline) except one (dexmedetomidine) clustered together and away from the noradrenaline) except one (dexmedetomidine) clustered together and away from the inverse agonist/antagonist ligands. inverse agonist/antagonist ligands.

3 a Table 9. Competition binding affinities of 20 compounds at three human and five zebrafish 2-adrenoceptors using [methyl- H]rauwolscine( ), [ethyl-3H]RS79948-197(b) and [3H]RX821002 as radioligands.

Ligand human 2A zebrafish 2A human 2B zebrafish2B human 2C zebrafish 2C zebrafish 2Da zebrafish 2Db 1 Atipamezole 1.6 (1.3-2.1) 13 (7.4-24) ** 1.5 (0.8-2.8)a 5.0 (3.2-7.9)a ** 4.3 (2.5-7.7)a 2.1 (1.9-2.3)a 5.1 (3.0-8.8) 6.90 (4.1-12) 250 2 Clonidine 10 (5-23)a 89 (62-130) *** 44 (21-94)a ** 110 (51-240)a 55 (46-67)a 120 (65-210) 150 (98-240) (190-320)a 3 Dexmedetomidine 1.3 (0.5-3.4)a 2.2 (1.5-3.0) 4.7 (2.2-10)a 7.6 (5.0-12)a 6.5 (2.8-16)a 12 (9.9-15)a 4.1 (3.3-5.0) 3.7 (2.6-5.3) 4 Idazoxan 17 (13-21) 85 (48-150) ** 24 (18-34)a 40 (23-68) 17 (6.7-26)a 17 (13-22) 52 (31-85) 94 (42-210) 1100 (530- 1200 5 Oxymetazoline 2.1 (1.5-3.0)a 5.1 (3.2-8.0) * 130 (70-250)a 1300 (860-2000)a ** 1100 (540-2400) 440 (300-660) 2300)a (1000-1400)a 1200 6 Brimonidine 32 (28-37) 40 (23-69) 320 (200-510)a ** 190 (140-270)a 700 (510-970)a *** 260 (200-320) 280 (210-370) (940-1600)a

7 L657.743 0.8 (0.5-1.2) 6.9 (3.7-13) * 0.7 (0.5-1.2)a 1.2 (1.0-1.6)a 0.09 (0.06-0.14)a 1.0 (0.9-1.1)a *** 1.6 (1.1-2.2) 1.3 (0.7-2.1) 8 Rauwolscine 1.9 (1.3-3.0) 1.0 (0.6-1.6) 1.1 (0.7-1.8)a 1.4 (0.6-3.2) 0.2 (0.1-0.5)a 0.5 (0.3-0.7) 2.3 (1.2-4.4) 2.3 (1.7-3.1) 9 Yohimbine 5.9 (4.9-7.1)b 5.2 (3.4-7.8) 7.5 (6.4-8.9)b 9.3 (7.0-12)a 4.6 (3.9-5.5)b 3.4 (3.0-3.7)a * 6.4 (4.5-9.2) 4.0 (3.1-5.1)

a a a a 10 Chlorpromazine 990 (500-1900) 110 (36-330) ** 43 (20-100) 1.1 (0.8-1.6) * 330 (220-1900) 83 (72-95) * 18 (13-26) 19 (10-36) Results 11 Clozapine 32 (15-66) 3.3 (2.6-4.3) *** 12 (5.0-28)a 9.3 (5.8-14) 2.1 (1.1-3.9)a 3.2 (1.8-5.7) 12 (9.5-16) 24 (15-38)

12 ARC239 2100 (860-5100) 1800 (870-3900) 9.6 (3.9-26)a 36 (31-42)a * 66 (36-630)a 280 (210-370)a 55 (46.8-65) 44 (19-110) 300 13 Prazosin 1030 (540-2050) 330 (200-540) 66 (31-150)a *** 31 (18-56)a 100 (78-130)a * 68 (50-92) 64 (48-86) (260-360)a 14 Spiperone 540 (440-660) 45 (28-71) *** 12 (3.7-38)a 51 (38-67)a ** 11 (5.4-23)a 63 (44-91)a 15 (11-22) 18 (9.6-33) 15 Spiroxatrine 320 (240-430)b 150 (92-240) * 2.4 (1.2-5.0)a 93 (62-140)a *** 3.1 (1.7-5.6)a 35 (22-56) ** 11 (3.7-33) 11 (4.7-86) 16 WB-4101 5.4 (2.0-15)a 11 (3.7-30) 60 (30-120)a 51 (45-59)a 1.9 (1.0-3.9)a 19 (16-24)a *** 31 (22-43) 16 (11-24)

2-amino-1- 5400 4200 9400 8100 3700 4000 17 1300 (940-1700)b ** 5100 (3700-7000)a phenylethanol (3100-9400) (2500-7100)b (7000-12600)a (6200-10600)b (2700-5100) (2200-7200) 2000 6300 4400 18 Dopamine 790 (520-1200) *** 1200 (710-2300)a 3900( 2100-7200)a 1300 (820-1900) 1700 (790-3700) (1300-3000)a (2800-14400)a (3800-5200)a 710 (330- 910 19 (-)-Adrenaline 150 (83-250) 140 (94-220) 130 (65-270)a 1080 (700-1700)a 500 (240-1100) 470 (300-740) 1600)a (720-1200)a 680 (360- 647 20 (-)-Noradrenaline 110 (32-400)a 260 (140-490) 250 (120-550)a 580 (330-1000)a 380 (270-540) 510 (310-840) 1300)a (500-830)a

Results are expressed as apparent Ki values (nM) and their 95 % confidence intervals (CÍ) from three to five separate experiments. Ligands are grouped into five clusters based on their chemical structures. Statistical significance of differences between the receptor orthologues A, B or C, *p<0.05, **p<0.01, ***p<0.001. 49

3 a Table 9. Competition binding affinities of 20 compounds at three human and five zebrafish 2-adrenoceptors using [methyl- H]rauwolscine( ), [ethyl-3H]RS79948-197(b) and [3H]RX821002 as radioligands.

Ligand human 2A zebrafish 2A human 2B zebrafish2B human 2C zebrafish 2C zebrafish 2Da zebrafish 2Db 1 Atipamezole 1.6 (1.3-2.1) 13 (7.4-24) ** 1.5 (0.8-2.8)a 5.0 (3.2-7.9)a ** 4.3 (2.5-7.7)a 2.1 (1.9-2.3)a 5.1 (3.0-8.8) 6.90 (4.1-12) 250 2 Clonidine 10 (5-23)a 89 (62-130) *** 44 (21-94)a ** 110 (51-240)a 55 (46-67)a 120 (65-210) 150 (98-240) (190-320)a 3 Dexmedetomidine 1.3 (0.5-3.4)a 2.2 (1.5-3.0) 4.7 (2.2-10)a 7.6 (5.0-12)a 6.5 (2.8-16)a 12 (9.9-15)a 4.1 (3.3-5.0) 3.7 (2.6-5.3) 4 Idazoxan 17 (13-21) 85 (48-150) ** 24 (18-34)a 40 (23-68) 17 (6.7-26)a 17 (13-22) 52 (31-85) 94 (42-210) 1100 (530- 1200 5 Oxymetazoline 2.1 (1.5-3.0)a 5.1 (3.2-8.0) * 130 (70-250)a 1300 (860-2000)a ** 1100 (540-2400) 440 (300-660) 2300)a (1000-1400)a 1200 6 Brimonidine 32 (28-37) 40 (23-69) 320 (200-510)a ** 190 (140-270)a 700 (510-970)a *** 260 (200-320) 280 (210-370) (940-1600)a

7 L657.743 0.8 (0.5-1.2) 6.9 (3.7-13) * 0.7 (0.5-1.2)a 1.2 (1.0-1.6)a 0.09 (0.06-0.14)a 1.0 (0.9-1.1)a *** 1.6 (1.1-2.2) 1.3 (0.7-2.1) 8 Rauwolscine 1.9 (1.3-3.0) 1.0 (0.6-1.6) 1.1 (0.7-1.8)a 1.4 (0.6-3.2) 0.2 (0.1-0.5)a 0.5 (0.3-0.7) 2.3 (1.2-4.4) 2.3 (1.7-3.1) 9 Yohimbine 5.9 (4.9-7.1)b 5.2 (3.4-7.8) 7.5 (6.4-8.9)b 9.3 (7.0-12)a 4.6 (3.9-5.5)b 3.4 (3.0-3.7)a * 6.4 (4.5-9.2) 4.0 (3.1-5.1)

a a a a 10 Chlorpromazine 990 (500-1900) 110 (36-330) ** 43 (20-100) 1.1 (0.8-1.6) * 330 (220-1900) 83 (72-95) * 18 (13-26) 19 (10-36) Results 11 Clozapine 32 (15-66) 3.3 (2.6-4.3) *** 12 (5.0-28)a 9.3 (5.8-14) 2.1 (1.1-3.9)a 3.2 (1.8-5.7) 12 (9.5-16) 24 (15-38)

12 ARC239 2100 (860-5100) 1800 (870-3900) 9.6 (3.9-26)a 36 (31-42)a * 66 (36-630)a 280 (210-370)a 55 (46.8-65) 44 (19-110) 300 13 Prazosin 1030 (540-2050) 330 (200-540) 66 (31-150)a *** 31 (18-56)a 100 (78-130)a * 68 (50-92) 64 (48-86) (260-360)a 14 Spiperone 540 (440-660) 45 (28-71) *** 12 (3.7-38)a 51 (38-67)a ** 11 (5.4-23)a 63 (44-91)a 15 (11-22) 18 (9.6-33) 15 Spiroxatrine 320 (240-430)b 150 (92-240) * 2.4 (1.2-5.0)a 93 (62-140)a *** 3.1 (1.7-5.6)a 35 (22-56) ** 11 (3.7-33) 11 (4.7-86) 16 WB-4101 5.4 (2.0-15)a 11 (3.7-30) 60 (30-120)a 51 (45-59)a 1.9 (1.0-3.9)a 19 (16-24)a *** 31 (22-43) 16 (11-24)

2-amino-1- 5400 4200 9400 8100 3700 4000 17 1300 (940-1700)b ** 5100 (3700-7000)a phenylethanol (3100-9400) (2500-7100)b (7000-12600)a (6200-10600)b (2700-5100) (2200-7200) 2000 6300 4400 18 Dopamine 790 (520-1200) *** 1200 (710-2300)a 3900( 2100-7200)a 1300 (820-1900) 1700 (790-3700) (1300-3000)a (2800-14400)a (3800-5200)a 710 (330- 910 19 (-)-Adrenaline 150 (83-250) 140 (94-220) 130 (65-270)a 1080 (700-1700)a 500 (240-1100) 470 (300-740) 1600)a (720-1200)a 680 (360- 647 20 (-)-Noradrenaline 110 (32-400)a 260 (140-490) 250 (120-550)a 580 (330-1000)a 380 (270-540) 510 (310-840) 1300)a (500-830)a

Results are expressed as apparent Ki values (nM) and their 95 % confidence intervals (CÍ) from three to five separate experiments. Ligands are grouped into five clusters based on their chemical structures. Statistical significance of differences between the receptor orthologues A, B or C, *p<0.05, **p<0.01, ***p<0.001. 49

50 Results 50 Results A B A B

17 17 4 17 4 4 17 4 18 18 12 6 12 6 3 3 19 5 3 3 19 5 18 18 20 13 20 20 13 20 13 2 13 2 2 4 19 15 2 4 19 15 2 10 2 15 2 10 2 15 2B 2B 2A 14 2A 14 6 14 4 16 6 14 4 16 D D D 12 D 12 1 16 1 16 1 7 1 11 1 7 1 11 1 1 5 9 3 9 5 9 3 9 11 11 3 3 10 10

zebrafish 0 8 zebrafish 0 8 8 zebrafish 0 7 8 zebrafish 0 7 r=0.85 r=0.85 r=0.87 r=0.87 -1 s=0.71 -1 -1 s=0.71 -1 s=0.89 s=0.89 p<0.0001 p<0.0001 p<0.0001 p<0.0001 -2 -2 -2 -2 -2 -1 0 1 2 3 4 -2 -1 0 1 2 3 4 -2 -1 0 1 2 3 4 -2 -1 0 1 2 3 4 human D2A human D 2B human D2A human D 2B C D C D

4 4 4 4 18 17 17 18 17 17 18 18 5 5 6 6 3 19 3 20 3 19 3 20 12 20 5 12 20 5 6 19 6 19 2 2 13 13 2 14 2 4 2 14 2 4 2C 2C

15 10 2Db 13 15 10 2Db 13

D 2 D 2 16 D 11 10 12 16 D 11 10 12 4 4 3 15 16 3 15 16 1 1 1 14 1 1 1 14 9 9 8 9 8 9 11 3 11 3 1 1 0 7 0 7 0 7 0 7 zebrafish zebrafish 8 zebrafish 8 zebrafish r=0.90 r=0.90 -1 s=0.88 -1 r=0.98 -1 s=0.88 -1 r=0.98 p<0.0001 s=0.99 p<0.0001 s=0.99 p<0.0001 p<0.0001 -2 -2 -2 -2 -2 -1 0 1 2 3 4 -2 -1 0 1 2 3 4 -2 -1 0 1 2 3 4 -2 -1 0 1 2 3 4 human D 2C zebrafish D 2Da human D 2C zebrafish D 2Da

Figure 10. Linear regression analysis of the logarithmic Ki values (nM) for 20 competing Figure 10. Linear regression analysis of the logarithmic Ki values (nM) for 20 competing ligands between the three orthologous human and zebrafish 2-adrenoceptor subtypes (A-C) ligands between the three orthologous human and zebrafish 2-adrenoceptor subtypes (A-C) and the paralogous zebrafish 2Da- and 2Db-adrenoceptors (D). Regression lines, slopes (s), and the paralogous zebrafish 2Da- and 2Db-adrenoceptors (D). Regression lines, slopes (s), Spearman’s correlation coefficients (r) and P-values for difference of the linear regression line Spearman’s correlation coefficients (r) and P-values for difference of the linear regression line from a line with zero slope are shown. Ligands are numbered in the order of Table 9. Modified from a line with zero slope are shown. Ligands are numbered in the order of Table 9. Modified from publication I. from publication I.

5.1.3 Agonist-stimulated [35S]GTPS binding 5.1.3 Agonist-stimulated [35S]GTPS binding The efficacies and potencies of (-)-adrenaline, (-)-noradrenaline, brimonidine, The efficacies and potencies of (-)-adrenaline, (-)-noradrenaline, brimonidine, oxymetazoline and dexmedetomidine on the three human and the five zebrafish 2- oxymetazoline and dexmedetomidine on the three human and the five zebrafish 2- adrenoceptors were determined using the functional [35S]GTPS binding assay (Table 2 adrenoceptors were determined using the functional [35S]GTPS binding assay (Table 2 in paper I). For the human and zebrafish 2A-adrenoceptors, the natural agonists (-)- in paper I). For the human and zebrafish 2A-adrenoceptors, the natural agonists (-)- adrenaline and (-)-noradrenaline exhibited high efficacy but rather low potency, while adrenaline and (-)-noradrenaline exhibited high efficacy but rather low potency, while brimonidine, oxymetazoline and dexmedetomidine displayed only partial agonism but brimonidine, oxymetazoline and dexmedetomidine displayed only partial agonism but

Results 51 Results 51 higher potency. For the human 2B-adrenoceptor, the maximal signal obtained with the higher potency. For the human 2B-adrenoceptor, the maximal signal obtained with the natural ligands was smaller than for the human 2A- and 2C-adrenoceptors. On the natural ligands was smaller than for the human 2A- and 2C-adrenoceptors. On the other hand, they revealed even less efficacy at the zebrafish 2B- and 2C- other hand, they revealed even less efficacy at the zebrafish 2B- and 2C- adrenoceptors. Moreover, no clear stimulation was seen for brimonidine or adrenoceptors. Moreover, no clear stimulation was seen for brimonidine or oxymetazoline at the zebrafish 2B- and 2C-adrenoceptors. In addition, oxymetazoline at the zebrafish 2B- and 2C-adrenoceptors. In addition, dexmedetomidine did not appreciably stimulate [35S]GTPS binding at the zebrafish dexmedetomidine did not appreciably stimulate [35S]GTPS binding at the zebrafish 2B-adrenoceptor. When 2-adrenoceptor expressing CHO cells were treated with 2B-adrenoceptor. When 2-adrenoceptor expressing CHO cells were treated with pertussis toxin (PTX) before being harvested, the responses to (-)-adrenaline were pertussis toxin (PTX) before being harvested, the responses to (-)-adrenaline were completely abolished for all adrenoceptor subtypes, indicating coupling to Gi proteins, completely abolished for all adrenoceptor subtypes, indicating coupling to Gi proteins, which are inactivated by PTX (Freissmuth et al., 1999). which are inactivated by PTX (Freissmuth et al., 1999).

5.1.4 Comparison of the structural models of the human and zebrafish 2- 5.1.4 Comparison of the structural models of the human and zebrafish 2- adrenoceptors adrenoceptors

Similarity in the pharmacological profiles of the human and zebrafish 2-adrenoceptors Similarity in the pharmacological profiles of the human and zebrafish 2-adrenoceptors can be thought to reflect similar requirements to bind endogenous catecholamines. In can be thought to reflect similar requirements to bind endogenous catecholamines. In the structural models of the receptor proteins, the amino acids that were predicted to the structural models of the receptor proteins, the amino acids that were predicted to constitute the membrane-embedded ligand-accessible surface of the receptors are constitute the membrane-embedded ligand-accessible surface of the receptors are located in TM2-TM7 and in XL2, and most of them were found to be highly conserved located in TM2-TM7 and in XL2, and most of them were found to be highly conserved (Figure 11). Only three residues (at positions 2.57, 5.39 and 5.43) of 27 ligand- (Figure 11). Only three residues (at positions 2.57, 5.39 and 5.43) of 27 ligand- accessible amino acids were found to differ within the TM domains in the structural accessible amino acids were found to differ within the TM domains in the structural models. Of these, the role of serine/cysteine variation at position 5.43 in the 2- models. Of these, the role of serine/cysteine variation at position 5.43 in the 2- adrenoceptors has been studied extensively with regard to catecholamine binding adrenoceptors has been studied extensively with regard to catecholamine binding (Rudling et al., 1999, Nyrönen et al., 2001, Peltonen et al., 2003). Other amino acids (Rudling et al., 1999, Nyrönen et al., 2001, Peltonen et al., 2003). Other amino acids known to be important for the binding of small catecholic ligands in TM3-TM6, D3.32, known to be important for the binding of small catecholic ligands in TM3-TM6, D3.32, V3.33, C3.36, S5.42, S5.46, F5.47, W6.48, F6.51, F6.52 and Y6.55, are conserved V3.33, C3.36, S5.42, S5.46, F5.47, W6.48, F6.51, F6.52 and Y6.55, are conserved among all of the human and zebrafish 2-adrenoceptor subtypes (Marjamäki et al., among all of the human and zebrafish 2-adrenoceptor subtypes (Marjamäki et al., 1999, Nyrönen et al., 2001, Peltonen et al., 2003). Most of the amino acid variation 1999, Nyrönen et al., 2001, Peltonen et al., 2003). Most of the amino acid variation within the proposed binding cavity of the human and zebrafish 2-adrenoceptors was within the proposed binding cavity of the human and zebrafish 2-adrenoceptors was found to be concentrated in XL2. In particular, three amino acids, at positions xl2.49, found to be concentrated in XL2. In particular, three amino acids, at positions xl2.49, xl2.51 and xl2.52, were predicted to be exposed to the binding cavity, and were found xl2.51 and xl2.52, were predicted to be exposed to the binding cavity, and were found to differ both between paralogous and orthologous receptor subtypes. to differ both between paralogous and orthologous receptor subtypes.

52 Results 52 Results

TM6 TM6 TM6 TM6 A TM7 B TM7 A TM7 B TM7 Y Y Y Y 6.55 5.35NDQK W5.39 6.55 5.35NEWQA5.39 6.55 5.35NDQK W5.39 6.55 5.35NEWQA5.39 5.35 5.39 5.35 5.39 5.35 5.39 5.35 5.39 F F F NEWKK TM5 F F F NDER W TM5 F F F NEWKK TM5 F F F NDER W TM5 7.39 6.51 6.52 7.39 6.51 6.52 7.39 6.51 6.52 7.39 6.51 6.52 Y G W Y G W Y G W Y G W 7.43 7.42 6.48 V I 7.43 7.42 6.48 I 7.43 7.42 6.48 V I 7.43 7.42 6.48 I 5.39 5.39 5.39 5.39 N F Y N F Y N F Y N F Y 7.45 6.44 C 5.38 7.45 6.44 S T 5.38 7.45 6.44 C 5.38 7.45 6.44 S T 5.38 5.43 5.43 5.43 5.43 xl2.48PEAPQPGGGGGxl2.38 xl2.48PRGRPQPGxl2.41 xl2.48PEAPQPGGGGGxl2.38 xl2.48PRGRPQPGxl2.41 xl2.48 xl2.45 S xl2.48 xl2.40 S xl2.48 xl2.45 S xl2.48 xl2.40 S Xl2 DGES F 5.42 Xl2 PLAGSEGGG F 5.42 Xl2 DGES F 5.42 Xl2 PLAGSEGGG F 5.42 5.47 5.47 5.47 5.47 R I C E D I S Q C K Q L S R I C E D I S Q C K Q L S xl2.49 xl2.50 xl2.51 xl2.52 5.46 xl2.49 xl2.50 xl2.51 xl2.52 5.46 xl2.49 xl2.50 xl2.51 xl2.52 5.46 xl2.49 xl2.50 xl2.51 xl2.52 5.46

4.67KEIKSI4.62 4.67 QDGKYI4.62 4.67KEIKSI4.62 4.67 QDGKYI4.62 4.66KTIKM 4.62 4.68KNKSL NM 4.62 4.66KTIKM 4.62 4.68KNKSL NM 4.62 C C C C 3.25 3.25 3.25 3.25 Y L Y L Y L Y L TM1 3.28 3.29 P TM1 3.28 3.29 P TM1 3.28 3.29 P TM1 3.28 3.29 P V 4.60 I 4.60 V 4.60 I 4.60 2.57 2.57 2.57 2.57 D V V D V V D V V D V V V 3.32 3.33 4.56 V 3.32 3.33 4.56 V 3.32 3.33 4.56 V 3.32 3.33 4.56 2.53 2.53 2.53 2.53 C T I C T I C T I C T I 3.36 3.37 4.52 3.36 3.37 4.52 3.36 3.37 4.52 3.36 3.37 4.52

TM3 I TM3 I TM3 I TM3 I TM2 3.40 TM4 TM2 3.40 TM4 TM2 3.40 TM4 TM2 3.40 TM4

TM6 TM6 TM6 TM6 C TM7 D TM7 C TM7 D TM7 Y Y Y Y 6.55 5.35NDET W5.39 6.55 5.35NDET W5.39 6.55 5.35NDET W5.39 6.55 5.35NDET W5.39 5.35 5.39 5.35 5.39 5.35 5.39 5.35 5.39 F F F NDHT W TM5 F F F NEWNT TM5 F F F NDHT W TM5 F F F NEWNT TM5 7.39 6.51 6.52 7.39 6.51 6.52 7.39 6.51 6.52 7.39 6.51 6.52 Y G W Y G W Y G W Y G W 7.43 7.42 6.48 I 7.43 7.42 6.48 I 7.43 7.42 6.48 I 7.43 7.42 6.48 I 5.39 5.39 5.39 5.39 N F Y N F Y N F Y N F Y 7.45 6.44 C S 5.38 7.45 6.44 C 5.38 7.45 6.44 C S 5.38 7.45 6.44 C 5.38 5.43 5.43 xl2.48 xl2.43 5.43 xl2.48 xl2.46 xl2.48 xl2.43 5.43 xl2.48 xl2.46 PYAAGD S KED S PYAAGD S KED S Xl2 xl2.48PRREDVxl2.43 F 5.42 Xl2 xl2.48LED xl2.46 F 5.42 Xl2 xl2.48PRREDVxl2.43 F 5.42 Xl2 xl2.48LED xl2.46 F 5.42 5.47 5.47 5.47 5.47 Q M C G Q L S E C L I L S Q M C G Q L S E C L I L S xl2.49 xl2.50 xl2.51 xl2.52 5.46 xl2.49 xl2.50 xl2.51 xl2.52 5.46 xl2.49 xl2.50 xl2.51 xl2.52 5.46 xl2.49 xl2.50 xl2.51 xl2.52 5.46

4.68PQRSV YL 4.62 4.66HTMKI4.62 4.68PQRSV YL 4.62 4.66HTMKI4.62 4.68TNRSI DM 4.62 4.66HTMKL 4.62 4.68TNRSI DM 4.62 4.66HTMKL 4.62 C C C C 3.25 3.25 3.25 3.25 Y L Y L Y L Y L TM1 3.28 3.29 P TM1 3.28 3.29 P TM1 3.28 3.29 P TM1 3.28 3.29 P V I 4.60 V 4.60 V I 4.60 V 4.60 2.57 2.57 2.57 2.57 D V V D V V D V V D V V V 3.32 3.33 4.56 V 3.32 3.33 4.56 V 3.32 3.33 4.56 V 3.32 3.33 4.56 2.53 2.53 2.53 2.53 C T I C T I C T I C T I 3.36 3.37 4.52 3.36 3.37 4.52 3.36 3.37 4.52 3.36 3.37 4.52

TM3 I TM3 I TM3 I TM3 I TM2 3.40 TM4 TM2 3.40 TM4 TM2 3.40 TM4 TM2 3.40 TM4 Figure 11. Schematic presentation of the putative ligand binding cavities of the human and Figure 11. Schematic presentation of the putative ligand binding cavities of the human and zebrafish 2-adrenoceptors based on the three-dimensional model structures. Only residues zebrafish 2-adrenoceptors based on the three-dimensional model structures. Only residues predicted to be exposed in the binding cavity are shown. Human versus zebrafish (A) 2A, predicted to be exposed in the binding cavity are shown. Human versus zebrafish (A) 2A, (B)2B, (C)2C and (D) the zebrafish2Da and 2Db duplicates. Positional conservation and (B)2B, (C)2C and (D) the zebrafish2Da and 2Db duplicates. Positional conservation and variation are indicated as follows: conserved positions in the compared receptors are circled and variation are indicated as follows: conserved positions in the compared receptors are circled and the amino acids are indicated in the single letter amino-acid code; a dot on the circle indicates the amino acids are indicated in the single letter amino-acid code; a dot on the circle indicates that the position is not conserved across one or more of the paralogous receptors; variable that the position is not conserved across one or more of the paralogous receptors; variable positions in the compared receptors are boxed and the variation is indicated with the appropriate positions in the compared receptors are boxed and the variation is indicated with the appropriate single letters (in panels A-C the amino acid in human is followed by that in zebrafish; in panel single letters (in panels A-C the amino acid in human is followed by that in zebrafish; in panel D the amino acid in 2Da is followed by that in2Db). Residues whose location with respect to D the amino acid in 2Da is followed by that in2Db). Residues whose location with respect to the proposed binding site is uncertain, belonging to XL2, are shown in dashed boxes. Shades of the proposed binding site is uncertain, belonging to XL2, are shown in dashed boxes. Shades of gray are used to indicate the location of amino acids with respect to the extracellular surface: gray are used to indicate the location of amino acids with respect to the extracellular surface: light towards the surface and darker away from the surface. Taken from publication I. light towards the surface and darker away from the surface. Taken from publication I.

5.2 Structural determinants involved in the interspecies difference of 5.2 Structural determinants involved in the interspecies difference of yohimbine analogues at human and mouse 2A-adrenoceptors (II) yohimbine analogues at human and mouse 2A-adrenoceptors (II)

5.2.1 Comparison of the binding cavities of the human and mouse 2A-adrenoceptors 5.2.1 Comparison of the binding cavities of the human and mouse 2A-adrenoceptors In order to predict the amino acids that differ within the binding cavities, molecular In order to predict the amino acids that differ within the binding cavities, molecular models of the human and mouse 2A-adrenoceptors were constructed based on the models of the human and mouse 2A-adrenoceptors were constructed based on the bovine rhodopsin crystal structure (Palczewski et al., 2000) (Figure 12). Including the bovine rhodopsin crystal structure (Palczewski et al., 2000) (Figure 12). Including the

Results 53 Results 53 expanded binding site proposal of Surgand et al. (2006), 34 amino acids in the TM expanded binding site proposal of Surgand et al. (2006), 34 amino acids in the TM domains were found to face the ligand binding cavity, and only one (5.43) of them domains were found to face the ligand binding cavity, and only one (5.43) of them differed between the human and mouse 2A-adrenoceptors. The XL2 domain, thought differed between the human and mouse 2A-adrenoceptors. The XL2 domain, thought to form an aromatic “lid” over the binding pocket, contains 25 amino acids but only to form an aromatic “lid” over the binding pocket, contains 25 amino acids but only four residues of these were thought to face the binding cavity, of which two (xl2.49 four residues of these were thought to face the binding cavity, of which two (xl2.49 and xl2.51) were different between the human and mouse 2A-adrenoceptors. Based on and xl2.51) were different between the human and mouse 2A-adrenoceptors. Based on the model structure comparisons, the three amino acids that differ between the binding the model structure comparisons, the three amino acids that differ between the binding cavities of the mouse and human 2A-adrenoceptors were replaced with the cavities of the mouse and human 2A-adrenoceptors were replaced with the corresponding reciprocal amino acids by the use of site-directed mutagenesis, and the corresponding reciprocal amino acids by the use of site-directed mutagenesis, and the effects of the mutations on ligand binding were tested. A total of five receptor mutants effects of the mutations on ligand binding were tested. A total of five receptor mutants were constructed: a human 2A-adrenoceptor with two mutations in XL2 (Rxl2.49S were constructed: a human 2A-adrenoceptor with two mutations in XL2 (Rxl2.49S and Exl2.51K) and the reciprocal mouse 2A-adrenoceptor double mutant (Sxl2.49R and Exl2.51K) and the reciprocal mouse 2A-adrenoceptor double mutant (Sxl2.49R and K2.51E); a human TM5 mutant (C5.43S) and the reciprocal mouse TM5 mutant and K2.51E); a human TM5 mutant (C5.43S) and the reciprocal mouse TM5 mutant (S5.43C); and a triple-mutated (S5.43, Sxl2.9R and K2.51E) mouse 2A-adrenoceptor. (S5.43C); and a triple-mutated (S5.43, Sxl2.9R and K2.51E) mouse 2A-adrenoceptor.

Figure 12. Schematic representations of the 2A-adrenoceptors from (A) human, (B) mouse and Figure 12. Schematic representations of the 2A-adrenoceptors from (A) human, (B) mouse and rat, and (C) corresponding amino acid codes according to the Ballesteros and Weinstein (1995) rat, and (C) corresponding amino acid codes according to the Ballesteros and Weinstein (1995) numbering scheme. Amino acids facing the ligand binding cavity in the TM and XL2 domains numbering scheme. Amino acids facing the ligand binding cavity in the TM and XL2 domains are indicated with one-letter codes. Residues differing between the human and mouse receptors are indicated with one-letter codes. Residues differing between the human and mouse receptors are indicated with grey. Four residues in XL2 suggested to face the ligand binding cavity are are indicated with grey. Four residues in XL2 suggested to face the ligand binding cavity are indicated with (*). Amino acids from the expanded binding site close to TM1 as proposed by indicated with (*). Amino acids from the expanded binding site close to TM1 as proposed by Surgand et al. (2006) are boxed. Two cysteines at 3.25 and xl2.50 are connected by a disulphide Surgand et al. (2006) are boxed. Two cysteines at 3.25 and xl2.50 are connected by a disulphide bridge. Modified from II. bridge. Modified from II.

54 Results 54 Results

3 3 5.2.2 [ H]RX821002 binding at human and mouse 2A-adrenoceptors 5.2.2 [ H]RX821002 binding at human and mouse 2A-adrenoceptors The antagonist radioligand [3H]RX821002 was found to have an about two-fold The antagonist radioligand [3H]RX821002 was found to have an about two-fold difference in its binding affinity towards the human and mouse 2A-adrenoceptors difference in its binding affinity towards the human and mouse 2A-adrenoceptors 3 3 (Table 10). The human 2A-adrenoceptor bound [ H]RX821002 with lower binding (Table 10). The human 2A-adrenoceptor bound [ H]RX821002 with lower binding affinity than the mouse 2A-adrenoceptor. This difference was reciprocally changed affinity than the mouse 2A-adrenoceptor. This difference was reciprocally changed when the cysteine-serine difference at position 5.43 in TM5 was reversed. Otherwise, when the cysteine-serine difference at position 5.43 in TM5 was reversed. Otherwise, 3 3 the binding affinity of [ H]RX821002 at the mutant 2A-adrenoceptors was observed to the binding affinity of [ H]RX821002 at the mutant 2A-adrenoceptors was observed to be in the same nanomolar range as found in the wild-type receptor orthologues. be in the same nanomolar range as found in the wild-type receptor orthologues.

Table 10. Binding affinities of [3H]RX821002 and receptor densities in CHO cell membranes Table 10. Binding affinities of [3H]RX821002 and receptor densities in CHO cell membranes containing wild-type and mutated 2A-adrenoceptors. containing wild-type and mutated 2A-adrenoceptors.

Receptor Kd (nM) Bmax (pmol/mg protein) Receptor Kd (nM) Bmax (pmol/mg protein) h2A C/R/E (wild-type) 1.10 ± 0.17 29.5 ± 2.3 h2A C/R/E (wild-type) 1.10 ± 0.17 29.5 ± 2.3 h2A C/S/K (xl2) 1.20 ± 0.12 14.5 ± 1.6 h2A C/S/K (xl2) 1.20 ± 0.12 14.5 ± 1.6 h2A S/R/E (TM5) 0.50 ± 0.02 43.3 ± 1.5 h2A S/R/E (TM5) 0.50 ± 0.02 43.3 ± 1.5 m2A S/S/K (wild-type) 0.54 ± 0.02 8.49 ± 0.03 m2A S/S/K (wild-type) 0.54 ± 0.02 8.49 ± 0.03 m2A S/R/E (xl2) 0.93 ± 0.10 28.0 ± 5.3 m2A S/R/E (xl2) 0.93 ± 0.10 28.0 ± 5.3 m2A C/S/K (TM5) 1.88 ± 0.13 5.33 ± 0.10 m2A C/S/K (TM5) 1.88 ± 0.13 5.33 ± 0.10 m2A C/R/E (TM5 & xl2) 1.59 ± 0.28 2.57 ± 0.25 m2A C/R/E (TM5 & xl2) 1.59 ± 0.28 2.57 ± 0.25 Receptor variants are named based on the amino acids at positions 5.43/xl2.49/xl2.51. Results Receptor variants are named based on the amino acids at positions 5.43/xl2.49/xl2.51. Results are expressed as means ± s.e.m of three independent experiments. h = human, m = mouse are expressed as means ± s.e.m of three independent experiments. h = human, m = mouse

5.2.3 Effects of XL2 and TM5 substitutions on ligand binding profiles 5.2.3 Effects of XL2 and TM5 substitutions on ligand binding profiles

The binding affinities (apparent Ki) of 10 antagonists and 3 natural agonists at wild- The binding affinities (apparent Ki) of 10 antagonists and 3 natural agonists at wild- type human and mouse 2A-adrenoceptors, as well as at five mutant receptors, were type human and mouse 2A-adrenoceptors, as well as at five mutant receptors, were characterized using competition binding assays (Table 11). The natural 2- characterized using competition binding assays (Table 11). The natural 2- adrenoceptor ligands, adrenaline, noradrenaline and dopamine, bound to the wild-type adrenoceptor ligands, adrenaline, noradrenaline and dopamine, bound to the wild-type human and mouse 2A-adrenoceptors with similar binding affinities in the following human and mouse 2A-adrenoceptors with similar binding affinities in the following potency order: adrenaline > noradrenaline > dopamine. As quite often observed, the potency order: adrenaline > noradrenaline > dopamine. As quite often observed, the affinity values of agonists exhibited greater experimental variability – larger affinity values of agonists exhibited greater experimental variability – larger confidence intervals – than those of antagonists. This is partly explained by the effects confidence intervals – than those of antagonists. This is partly explained by the effects of variation in the receptor expression levels, as a high level of expression leads to a of variation in the receptor expression levels, as a high level of expression leads to a predominance of the low-affinity conformation of the receptor, thus shifting the agonist predominance of the low-affinity conformation of the receptor, thus shifting the agonist binding curve to the right. This effect was also seen here where the expression levels binding curve to the right. This effect was also seen here where the expression levels were quite variable between the studied cell lines (see Table 11). were quite variable between the studied cell lines (see Table 11). With respect to the antagonist ligands, yohimbine was observed to prefer the human With respect to the antagonist ligands, yohimbine was observed to prefer the human wild-type 2A-adrenoceptor by about 15-fold over the mouse orthologue, and for its wild-type 2A-adrenoceptor by about 15-fold over the mouse orthologue, and for its chiral analogue rauwolscine, this ratio was even greater, about 20-fold. These chiral analogue rauwolscine, this ratio was even greater, about 20-fold. These observations agree with previous findings (Link et al., 1992, Uhlèn et al., 1998), but observations agree with previous findings (Link et al., 1992, Uhlèn et al., 1998), but here, the binding affinities of another four antagonists were also observed to here, the binding affinities of another four antagonists were also observed to

Results 55 Results 55 discriminate between the human and mouse 2A-adrenoceptors, but with more discriminate between the human and mouse 2A-adrenoceptors, but with more moderate effect sizes: WB4101 (seven-fold) and RS-79948-197 (four-fold) preferred moderate effect sizes: WB4101 (seven-fold) and RS-79948-197 (four-fold) preferred the human 2A-adrenoceptor, while both ARC239 and atipamezole preferred the mouse the human 2A-adrenoceptor, while both ARC239 and atipamezole preferred the mouse orthologue by somewhat less than four-fold. orthologue by somewhat less than four-fold. Reciprocal mutations in XL2 and TM5 were observed to lead to reciprocal changes Reciprocal mutations in XL2 and TM5 were observed to lead to reciprocal changes in the binding affinity profiles of the four antagonist ligands, rauwolscine, yohimbine, in the binding affinity profiles of the four antagonist ligands, rauwolscine, yohimbine, RS-79948-197 and WB4101, that preferred the human 2A-adrenoceptor over the RS-79948-197 and WB4101, that preferred the human 2A-adrenoceptor over the mouse receptor orthologue (Figure 13). However, for WB4101, only the mutation in mouse receptor orthologue (Figure 13). However, for WB4101, only the mutation in the mouse 2A-adrenoceptor at position S5.43C was observed to improve the binding the mouse 2A-adrenoceptor at position S5.43C was observed to improve the binding affinity, but this conclusion must be viewed with caution because of large overlapping affinity, but this conclusion must be viewed with caution because of large overlapping confidence intervals of the affinity estimates. With regard to the other three ligands, the confidence intervals of the affinity estimates. With regard to the other three ligands, the double substitution Rxl2.49S and Exl2.51K in XL2 of the human 2A-adrenoceptor double substitution Rxl2.49S and Exl2.51K in XL2 of the human 2A-adrenoceptor increased significantly the Ki for rauwolscine (four-fold), yohimbine (six-fold) and RS- increased significantly the Ki for rauwolscine (four-fold), yohimbine (six-fold) and RS- 79948-197 (seven-fold). The reciprocal double mutation Sxl2.49R and Kxl2.51E in the 79948-197 (seven-fold). The reciprocal double mutation Sxl2.49R and Kxl2.51E in the mouse receptor led to opposite “gain-of-function” effects on the binding of these mouse receptor led to opposite “gain-of-function” effects on the binding of these ligands as seen significant decreases in the Ki for rauwolscine (less than two-fold), ligands as seen significant decreases in the Ki for rauwolscine (less than two-fold), yohimbine (four-fold) and RS-79948-197 (four-fold). For rauwolscine and yohimbine, yohimbine (four-fold) and RS-79948-197 (four-fold). For rauwolscine and yohimbine, but not for RS-79948-197, corresponding reciprocal effects were seen in receptors but not for RS-79948-197, corresponding reciprocal effects were seen in receptors mutated at position 5.43 in TM5, as the C5.43S substitution in the human receptor led mutated at position 5.43 in TM5, as the C5.43S substitution in the human receptor led to an increase in the Ki for rauwolscine (eight-fold) and yohimbine (six-fold), whereas to an increase in the Ki for rauwolscine (eight-fold) and yohimbine (six-fold), whereas the reciprocal S5.43C substitution in the mouse receptor led to decreases in the Ki for the reciprocal S5.43C substitution in the mouse receptor led to decreases in the Ki for rauwolscine (three-fold) and yohimbine (six-fold). When all three positions (S5.43C, rauwolscine (three-fold) and yohimbine (six-fold). When all three positions (S5.43C, Sxl2.49R and Kxl2.51E) were simultaneously mutated in the mouse 2A-adrenoceptor, Sxl2.49R and Kxl2.51E) were simultaneously mutated in the mouse 2A-adrenoceptor, the affinities of yohimbine, rauwolscine and RS-79948-197 were observed to be the affinities of yohimbine, rauwolscine and RS-79948-197 were observed to be comparable with those of the human wild-type 2A-adrenoceptor: Ki was decreased for comparable with those of the human wild-type 2A-adrenoceptor: Ki was decreased for yohimbine (16-fold), rauwolscine (six-fold) and RS-79948-197 (six-fold), when yohimbine (16-fold), rauwolscine (six-fold) and RS-79948-197 (six-fold), when compared to the wild-type mouse 2A-adrenoceptor. compared to the wild-type mouse 2A-adrenoceptor.

54

Table 11. Competition binding affinities of different ligands obtained with [3H]RX821002 at human and mouse wild-type (WT) and point mutated 2A-adrenoceptors expressed in CHO cells.

Human 2A Mouse 2A TM5 & xl2 Ligand WT (C/R/E) xl2 (C/S/K) TM5 (S/R/E) WT (S/S/K) xl2 (S/R/E) TM5 (C/S/K) (C/R/E) Rauwolscine 1.1 (0.48-2.5)●● 4.7 (3.4-6.5)** 9.0 (6.8-12)*** 22 (15-32) 15 (12-20) 6.9 (4.6-10)* 3.9 (2.6-5.7)** RS-79948-197 0.16 (0.054-0.53) 1.1 (0.72-1.7)*** 0.19 (0.15-0.24) 0.64 (0.33-1.2) 0.15 (0.087-0.28)* 0.63 (0.51-0.78) 0.11 (0.046-0.28)* WB4101 4.5 (2.0-7.4)●● 5.9 (3.2-11) 2.7 (0.46-17) 32 (9.0-100) 44 (7.8-170) 11 (1.7-85) 8.1 (2.3-32) Yohimbine 2.1 (1.6-4.6)●● 13 (6.3-23)*** 12 (8.8-16)*** 31 (21-46) 7.0 (3.4-15)** 4.9 (2.2-11)** 2.0 (1.2-4.1)***

ARC239 1800 3000 (1300-9200) 2400 (1000-5600) 500 (200-1300) 550 (240-1700) 650 (210-2100) 280 (120-710) (1400-2400)●●● ●●●

Atipamezole 2.3 (1.3-3.9) 3.4 (1.5-8.4) 1.4 (0.47-5.3) 0.63 (0.52-0.76) 1.9 (0.84-5.6)** 3.6 (1.2-11)** 4.3 (1.4-13)*** Results Chlorpromazine 200 (67-630) 280 (95-840) 210 (82-570) 100 (38-280) 100 (43-260) 200 (75-540) 150 (66-370) Doxazosin 710 (330-1600) 1600 (670-5100) 1300 (270-6500) 1600 (340-8100) 1600 (470-8900) 2300 (630-8800) 720 (330-2700) L-657.743 2.5 (1.2-5.6) 1.5 (0.64-3.4) 1.2 (0.87-1.6) 2.7 (1.7-4.2) 2.6 (1.4-4.7) 2.5 (1.1-5.8) 2.8 (1.8-4.3) Prazosin 880 (250-3400) 1600 (770-3700) 650 (200-2100) 510 (290-1400) 780 (180-2400) 390 (130-1300) 190 (68-510)

Adrenaline 880 (260-2100) 1900 (510-7200) 1900 (1400-2700) 910 (200-4300) 5300 (530-61 000) 570 (240-2000) 500 (140-2200) Noradrenaline 2400 (940-6800) 2200 (1000-5200) 1000 (340-4100) 1900 (400-12 000) 6200 (900-48 000) 1400 (510-3900) 590 (180-2000) Dopamine 4200 5900 3200 (1100-9500) 8000 (1500-72 000) 21 000 3400 (760-17 000) 5000 (1000-18 000) (1500-13 000) (3300-11 000) (4900-110 000) Chlorpromazine/ yohimbine Ki ratio 95 22 18 3 14 41 75

Results are expressed as apparent Ki (nM) and 95 % confidence intervals of three to five independent experiments. Ligands are grouped as 1) antagonists with more than 4-fold different affinity between the human and mouse wild-type receptors, 2) antagonists with less than 4-fold difference in affinity between the human and mouse wild-type receptors, 3) endogenous agonists. The pharmacological comparison of receptors is illustrated by ● ●● ●●● the ratios of the Ki values for chlorpromazine and yohimbine.* P<0.05; ** P<0.01; *** P<0.001 mutant vs. wild-type; P<0.05; P<0.01; P<0.001 human vs. mouse.

54

Table 11. Competition binding affinities of different ligands obtained with [3H]RX821002 at human and mouse wild-type (WT) and point mutated 2A-adrenoceptors expressed in CHO cells.

Human 2A Mouse 2A TM5 & xl2 Ligand WT (C/R/E) xl2 (C/S/K) TM5 (S/R/E) WT (S/S/K) xl2 (S/R/E) TM5 (C/S/K) (C/R/E) Rauwolscine 1.1 (0.48-2.5)●● 4.7 (3.4-6.5)** 9.0 (6.8-12)*** 22 (15-32) 15 (12-20) 6.9 (4.6-10)* 3.9 (2.6-5.7)** RS-79948-197 0.16 (0.054-0.53) 1.1 (0.72-1.7)*** 0.19 (0.15-0.24) 0.64 (0.33-1.2) 0.15 (0.087-0.28)* 0.63 (0.51-0.78) 0.11 (0.046-0.28)* WB4101 4.5 (2.0-7.4)●● 5.9 (3.2-11) 2.7 (0.46-17) 32 (9.0-100) 44 (7.8-170) 11 (1.7-85) 8.1 (2.3-32) Yohimbine 2.1 (1.6-4.6)●● 13 (6.3-23)*** 12 (8.8-16)*** 31 (21-46) 7.0 (3.4-15)** 4.9 (2.2-11)** 2.0 (1.2-4.1)***

ARC239 1800 3000 (1300-9200) 2400 (1000-5600) 500 (200-1300) 550 (240-1700) 650 (210-2100) 280 (120-710) (1400-2400)●●● ●●●

Atipamezole 2.3 (1.3-3.9) 3.4 (1.5-8.4) 1.4 (0.47-5.3) 0.63 (0.52-0.76) 1.9 (0.84-5.6)** 3.6 (1.2-11)** 4.3 (1.4-13)*** Results Chlorpromazine 200 (67-630) 280 (95-840) 210 (82-570) 100 (38-280) 100 (43-260) 200 (75-540) 150 (66-370) Doxazosin 710 (330-1600) 1600 (670-5100) 1300 (270-6500) 1600 (340-8100) 1600 (470-8900) 2300 (630-8800) 720 (330-2700) L-657.743 2.5 (1.2-5.6) 1.5 (0.64-3.4) 1.2 (0.87-1.6) 2.7 (1.7-4.2) 2.6 (1.4-4.7) 2.5 (1.1-5.8) 2.8 (1.8-4.3) Prazosin 880 (250-3400) 1600 (770-3700) 650 (200-2100) 510 (290-1400) 780 (180-2400) 390 (130-1300) 190 (68-510)

Adrenaline 880 (260-2100) 1900 (510-7200) 1900 (1400-2700) 910 (200-4300) 5300 (530-61 000) 570 (240-2000) 500 (140-2200) Noradrenaline 2400 (940-6800) 2200 (1000-5200) 1000 (340-4100) 1900 (400-12 000) 6200 (900-48 000) 1400 (510-3900) 590 (180-2000) Dopamine 4200 5900 3200 (1100-9500) 8000 (1500-72 000) 21 000 3400 (760-17 000) 5000 (1000-18 000) (1500-13 000) (3300-11 000) (4900-110 000) Chlorpromazine/ yohimbine Ki ratio 95 22 18 3 14 41 75

Results are expressed as apparent Ki (nM) and 95 % confidence intervals of three to five independent experiments. Ligands are grouped as 1) antagonists with more than 4-fold different affinity between the human and mouse wild-type receptors, 2) antagonists with less than 4-fold difference in affinity between the human and mouse wild-type receptors, 3) endogenous agonists. The pharmacological comparison of receptors is illustrated by ● ●● ●●● the ratios of the Ki values for chlorpromazine and yohimbine.* P<0.05; ** P<0.01; *** P<0.001 mutant vs. wild-type; P<0.05; P<0.01; P<0.001 human vs. mouse.

Results 57 Results 57

YOHIMBINE RAUWOLSCINE YOHIMBINE RAUWOLSCINE

S/S/K (wild-type) S/S/K (wild-type) S/S/K (wild-type) S/S/K (wild-type) ** ** S/R/E (xl2) S/R/E (xl2) S/R/E (xl2) S/R/E (xl2) ** ** C/S/K (TM5) C/S/K (TM5) * C/S/K (TM5) C/S/K (TM5) * *** ** *** ** C/R/E (TM5 & xl2) C/R/E (TM5 & xl2) C/R/E (TM5 & xl2) C/R/E (TM5 & xl2) Mouse 2A Mouse 2A Mouse 2A Mouse 2A Human 2A Human 2A Human 2A Human 2A C/R/E (wild-type) C/R/E (wild-type) C/R/E (wild-type) C/R/E (wild-type) ** ** C/S/K (xl2) *** C/S/K (xl2) C/S/K (xl2) *** C/S/K (xl2) *** *** *** *** S/R/E (TM5) S/R/E (TM5) S/R/E (TM5) S/R/E (TM5) 0.01 0.1 1 10 100 1000 0.01 0.1 1 10 100 1000 0.01 0.1 1 10 100 1000 0.01 0.1 1 10 100 1000 logKi (nM) logKi (nM) logKi (nM) logKi (nM)

RS-79948-197 WB4101 RS-79948-197 WB4101

S/S/K (wild-type) S/S/K (wild-type) S/S/K (wild-type) S/S/K (wild-type) * * S/R/E (xl2) S/R/E (xl2) S/R/E (xl2) S/R/E (xl2) C/S/K (TM5) C/S/K (TM5) C/S/K (TM5) C/S/K (TM5) C/R/E (TM5 & xl2) * C/R/E (TM5 & xl2) C/R/E (TM5 & xl2) * C/R/E (TM5 & xl2) Mouse 2A Mouse 2A Mouse 2A Mouse 2A Human  Human  Human  Human  C/R/E (wild-type) 2A C/R/E (wild-type) 2A C/R/E (wild-type) 2A C/R/E (wild-type) 2A *** *** C/S/K (xl2) C/S/K (xl2) C/S/K (xl2) C/S/K (xl2) S/R/E (TM5) S/R/E (TM5) S/R/E (TM5) S/R/E (TM5) 0.01 0.1 1 10 100 1000 0.01 0.1 1 10 100 1000 0.01 0.1 1 10 100 1000 0.01 0.1 1 10 100 1000 logKi (nM) logKi (nM) logKi (nM) logKi (nM)

Figure 13. Comparison of the binding affinities of yohimbine, rauwolscine, RS-79948-197 and Figure 13. Comparison of the binding affinities of yohimbine, rauwolscine, RS-79948-197 and WB4101 towards the human and mouse 2A-adrenoceptors and their mutants. Error bars WB4101 towards the human and mouse 2A-adrenoceptors and their mutants. Error bars represent 95 % confidence intervals of 3-5 independent experiments. Statistical significance represent 95 % confidence intervals of 3-5 independent experiments. Statistical significance compared to the wild-type receptor (unpaired t-tests): *P<0.05, **P<0.01, ***P<0.001. compared to the wild-type receptor (unpaired t-tests): *P<0.05, **P<0.01, ***P<0.001. Modified from II. Modified from II.

5.3 Involvement of the first transmembrane domain of human 2- 5.3 Involvement of the first transmembrane domain of human 2- adrenoceptors in the subtype-selectivity of bulky antagonists (III) adrenoceptors in the subtype-selectivity of bulky antagonists (III)

5.3.1 Construction of receptor chimaeras 5.3.1 Construction of receptor chimaeras At the start of the project, all receptor mutants (Figure 14) were planned to be prepared At the start of the project, all receptor mutants (Figure 14) were planned to be prepared based on PCR and primers containing the desired mutations and an artificial restriction based on PCR and primers containing the desired mutations and an artificial restriction enzyme recognition site for subcloning into the pREP4 vector. However, only enzyme recognition site for subcloning into the pREP4 vector. However, only chimaeras based on the 2A- and 2B-adrenoceptors were constructed using PCR-based chimaeras based on the 2A- and 2B-adrenoceptors were constructed using PCR-based mutagenesis and two pairs of primers, designed for each subtype. The exchange of mutagenesis and two pairs of primers, designed for each subtype. The exchange of TM1 and the preceding amino terminal sequence was performed utilizing a conserved TM1 and the preceding amino terminal sequence was performed utilizing a conserved threonine-serine site at the intracellular end of TM1. Thus, reverse primers until the threonine-serine site at the intracellular end of TM1. Thus, reverse primers until the end of TM1 and forward primers from the start of the first intracellular loop were end of TM1 and forward primers from the start of the first intracellular loop were designed to contain an artificial SpeI-site; recognition sequence ACTAGT, coding for designed to contain an artificial SpeI-site; recognition sequence ACTAGT, coding for threonine-serine. In addition, the N- and C-terminal primers were designed to contain threonine-serine. In addition, the N- and C-terminal primers were designed to contain restriction sites for subcloning into the pREP4 expression vector. Finally, the PCR- restriction sites for subcloning into the pREP4 expression vector. Finally, the PCR- amplified fragments were digested with the appropriate restriction enzymes and ligated amplified fragments were digested with the appropriate restriction enzymes and ligated into pREP4 for expression in mammalian cells. into pREP4 for expression in mammalian cells. After several failures in PCR-amplification of the C-terminal fragment of the 2C- After several failures in PCR-amplification of the C-terminal fragment of the 2C- adrenoceptor, chimaeras based on this receptor subtype were constructed using the adrenoceptor, chimaeras based on this receptor subtype were constructed using the

58 Results 58 Results

Gene EditorTM in vitro Site-Directed Mutagenesis System. In this approach, the Gene EditorTM in vitro Site-Directed Mutagenesis System. In this approach, the original nucleotide sequence encoding conserved threonine-serine amino acids at the original nucleotide sequence encoding conserved threonine-serine amino acids at the end of TM1 were mutated to contain the sequence ACTAGT, a site for SpeI end of TM1 were mutated to contain the sequence ACTAGT, a site for SpeI recognition. Subsequently, the 2C-adrenoceptor-based chimaeras were constructed recognition. Subsequently, the 2C-adrenoceptor-based chimaeras were constructed using the artificial SpeI-site and PCR-amplified TM1 (and N-terminal) fragments of using the artificial SpeI-site and PCR-amplified TM1 (and N-terminal) fragments of the 2A- and 2B-adrenoceptors, and finally, the digested fragments were ligated and the 2A- and 2B-adrenoceptors, and finally, the digested fragments were ligated and subcloned into the pcDNA3.1(+) expression vector. subcloned into the pcDNA3.1(+) expression vector. Additional 2A- and 2C-adrenoceptor chimaeras with only TM1-region Additional 2A- and 2C-adrenoceptor chimaeras with only TM1-region substitutions were constructed using the GeneEditorTM in vitro Site-Directed substitutions were constructed using the GeneEditorTM in vitro Site-Directed Mutagenesis System and the previous chimaeras as templates; the nucleotide sequence Mutagenesis System and the previous chimaeras as templates; the nucleotide sequence coding for the conserved tyrosine-serine pair at the end of the N-terminus was mutated coding for the conserved tyrosine-serine pair at the end of the N-terminus was mutated to contain an NheI-site (recognition sequence GCTAGC, coding for alanine-serine). to contain an NheI-site (recognition sequence GCTAGC, coding for alanine-serine). Subsequently, using this artificial NheI-site, the N-terminal fragments were digested, Subsequently, using this artificial NheI-site, the N-terminal fragments were digested, isolated from agarose gel, and ligated to the appropriate 2A- and 2C-adrenoceptor- isolated from agarose gel, and ligated to the appropriate 2A- and 2C-adrenoceptor- based chimaeras. Thereafter, the artificial NheI-sites were restored to encode the based chimaeras. Thereafter, the artificial NheI-sites were restored to encode the original tyrosine-serine pair. original tyrosine-serine pair.

AAA BBA CCA ACA AAA BBA CCA ACA

BBB AAB CCB BBB AAB CCB

2A 2A

2B 2B

2C 2C

CCC AAC BBC CAC CCC AAC BBC CAC

Figure 14. Schematic representation of the constructed 2-adrenoceptor chimaeras. Receptors Figure 14. Schematic representation of the constructed 2-adrenoceptor chimaeras. Receptors are named based on the origin of the sequence of the N-terminal segment, the TM1 domain and are named based on the origin of the sequence of the N-terminal segment, the TM1 domain and the body of the receptor, corresponding to the first, second and third letter, respectively. the body of the receptor, corresponding to the first, second and third letter, respectively. Modified from III. Modified from III.

5.3.2 Expression of 2-adrenoceptor chimaeras in CHO cells 5.3.2 Expression of 2-adrenoceptor chimaeras in CHO cells

The cDNAs of eight chimaeric 2-adrenoceptor constructs were transfected into CHO The cDNAs of eight chimaeric 2-adrenoceptor constructs were transfected into CHO cells using lipid-based transfection mixtures. Cell cultures were screened for their cells using lipid-based transfection mixtures. Cell cultures were screened for their capacity to bind the radioligand [3H]RX821002 and the cell populations with the capacity to bind the radioligand [3H]RX821002 and the cell populations with the

Results 59 Results 59 highest levels of receptor expression were chosen for further binding experiments. highest levels of receptor expression were chosen for further binding experiments. Seven out of eight receptor constructs were successfully expressed in CHO cells, as Seven out of eight receptor constructs were successfully expressed in CHO cells, as evidenced by saturation binding experiments with [3H]RX821002 (Table 12). For the evidenced by saturation binding experiments with [3H]RX821002 (Table 12). For the wild-type receptors, [3H]RX821002 had 5- to 10-fold lower affinity towards the human wild-type receptors, [3H]RX821002 had 5- to 10-fold lower affinity towards the human 2B-adrenoceptor in comparison to the 2A- and 2C-adrenoceptors. This observation 2B-adrenoceptor in comparison to the 2A- and 2C-adrenoceptors. This observation agrees with previous findings (O'Rourke et al., 1994a, Deupree et al., 1996, Uhlèn et agrees with previous findings (O'Rourke et al., 1994a, Deupree et al., 1996, Uhlèn et al., 1998). When the N-terminal and TM1 regions of the 2B-adrenoceptor were al., 1998). When the N-terminal and TM1 regions of the 2B-adrenoceptor were replaced with the corresponding regions of the 2A- and 2C-adrenoceptors, the binding replaced with the corresponding regions of the 2A- and 2C-adrenoceptors, the binding 3 3 affinity of [ H]RX821002 was increased 3-fold towards that of the wild-type 2A- and affinity of [ H]RX821002 was increased 3-fold towards that of the wild-type 2A- and 2C-adrenoceptors. Reciprocal effects were seen in the 2A-based chimaera containing 2C-adrenoceptors. Reciprocal effects were seen in the 2A-based chimaera containing the N-terminal and TM1 regions from the 2B-adrenoceptor, i.e. an almost 3-fold the N-terminal and TM1 regions from the 2B-adrenoceptor, i.e. an almost 3-fold decrease in affinity relative to the wild-type 2A-adrenoceptor. decrease in affinity relative to the wild-type 2A-adrenoceptor.

Table 12. Binding affinities of [3H]RX821002 and receptor densities in recombinant CHO cell Table 12. Binding affinities of [3H]RX821002 and receptor densities in recombinant CHO cell lines expressing wild-type and mutated human 2B-adrenoceptor subtypes. lines expressing wild-type and mutated human 2B-adrenoceptor subtypes.

Receptor Kd (nM) Bmax (fmol/mg protein) Receptor Kd (nM) Bmax (fmol/mg protein) 0.45 ± 0.04 6700 ± 100 0.45 ± 0.04 6700 ± 100 2A (AAA) 2A (AAA)

2A BBA 1.2  0.7 400  300 2A BBA 1.2  0.7 400  300

2A CCA 0.39  0.06 5600  500 2A CCA 0.39  0.06 5600  500

2A ACA 0.27 ± 0.02* 19 000 ± 2 800 2A ACA 0.27 ± 0.02* 19 000 ± 2 800

4.6 ± 1.0 1500 ± 300 4.6 ± 1.0 1500 ± 300 2B (BBB) 2B (BBB)

2B AAB 1.4  0.2* 1600  100 2B AAB 1.4  0.2* 1600  100

2B CCB 1.6  0.2* 6100  500 2B CCB 1.6  0.2* 6100  500

2C (CCC) 0.87 ± 0.06 1800 ± 40 2C (CCC) 0.87 ± 0.06 1800 ± 40

2C AAC 0.71 ± 0.04 470 ± 10 2C AAC 0.71 ± 0.04 470 ± 10

2C BBC - - 2C BBC - -

2C CAC 0.55 ± 0.15 710 ± 80 2C CAC 0.55 ± 0.15 710 ± 80 Receptors are named based on the exchanged domains (see Figure 14). Results are expressed as Receptors are named based on the exchanged domains (see Figure 14). Results are expressed as means ± s.e.m. of three independent experiments. Statistical significance of differences between means ± s.e.m. of three independent experiments. Statistical significance of differences between the chimaeric receptors and the respective wild-type receptors were tested with Student’s t-test: the chimaeric receptors and the respective wild-type receptors were tested with Student’s t-test: * P<0.05 * P<0.05

Cells transfected with the 2C-adrenoceptor-based construct containing TM1 and Cells transfected with the 2C-adrenoceptor-based construct containing TM1 and 3 3 the N-terminus from the 2B-adrenoceptor failed to show [ H]RX821002 binding in the N-terminus from the 2B-adrenoceptor failed to show [ H]RX821002 binding in the saturation binding experiments, in spite of several transfection experiments the saturation binding experiments, in spite of several transfection experiments with several batches of plasmid DNA. To confirm the transfection of the construct, with several batches of plasmid DNA. To confirm the transfection of the construct, total RNA was isolated from transfected, antibiotic-resistant CHO cells and total RNA was isolated from transfected, antibiotic-resistant CHO cells and converted to cDNA. PCR amplification was performed with gene-specific converted to cDNA. PCR amplification was performed with gene-specific primers designed to both sites used for nucleotide sequence cut-off (upper primers designed to both sites used for nucleotide sequence cut-off (upper 5´-TCATGGGCGTGTTCGTGCTCTGCT-3´; lower 5´-ATCGCCGGAAATCCTGGTTGAAG-3´). 5´-TCATGGGCGTGTTCGTGCTCTGCT-3´; lower 5´-ATCGCCGGAAATCCTGGTTGAAG-3´). Control reactions were run using RNA from non-transfected cells (Figure 15). Control reactions were run using RNA from non-transfected cells (Figure 15).

60 Results 60 Results

Figure 15. RT-PCR reactions using 2C BBC gene-specific primers run on a 1 % agarose gel. Figure 15. RT-PCR reactions using 2C BBC gene-specific primers run on a 1 % agarose gel. Lanes 1 and 6: DNA size marker. Lane 2: PCR control (water used instead of template DNA) to Lanes 1 and 6: DNA size marker. Lane 2: PCR control (water used instead of template DNA) to prove that no DNA contamination exists. Lane 3: a 198-bp control fragment amplified from prove that no DNA contamination exists. Lane 3: a 198-bp control fragment amplified from plasmid DNA (pcDNA3.1-2C BBC). Lane 4: RT-PCR using RNA isolated from non- plasmid DNA (pcDNA3.1-2C BBC). Lane 4: RT-PCR using RNA isolated from non- transfected CHO-K1 cells. Lane 5: a 198-bp amplification fragment from RT-PCR using RNA transfected CHO-K1 cells. Lane 5: a 198-bp amplification fragment from RT-PCR using RNA isolated from 2C BBC transfected CHO cells. isolated from 2C BBC transfected CHO cells.

5.3.3 Characterization of antagonist binding profiles at wild-type and chimaeric 2- 5.3.3 Characterization of antagonist binding profiles at wild-type and chimaeric 2- adrenoceptors adrenoceptors

The binding affinities (apparent Ki) of nine antagonist ligands at the seven receptor The binding affinities (apparent Ki) of nine antagonist ligands at the seven receptor chimaeras and the three human wild-type 2-adrenoceptors were characterized using chimaeras and the three human wild-type 2-adrenoceptors were characterized using competition binding assays and [3H]RX821002 as radioligand (Table 13). For wild- competition binding assays and [3H]RX821002 as radioligand (Table 13). For wild- type receptors, six ligands (ARC239, prazosin, spiperone, spiroxatrine, chlorpromazine type receptors, six ligands (ARC239, prazosin, spiperone, spiroxatrine, chlorpromazine and clozapine) showed significantly lower affinity towards the 2A-adrenoceptor in and clozapine) showed significantly lower affinity towards the 2A-adrenoceptor in comparison to the 2B- and 2C-adrenoceptor subtypes, as highlighted by the 10-100 - comparison to the 2B- and 2C-adrenoceptor subtypes, as highlighted by the 10-100 - fold differences in the binding affinities. Only two of the ligands (chlorpromazine and fold differences in the binding affinities. Only two of the ligands (chlorpromazine and spiroxatrine) were observed to significantly discriminate between the 2B- and 2C- spiroxatrine) were observed to significantly discriminate between the 2B- and 2C- wild-type receptors (see Table 2 in III). For three ligands (spiperone, spiroxatrine and wild-type receptors (see Table 2 in III). For three ligands (spiperone, spiroxatrine and chlorpromazine), the binding affinities were found to be significantly improved at the chlorpromazine), the binding affinities were found to be significantly improved at the 2A-adrenoceptor when TM1 and the N-terminus were replaced with the corresponding 2A-adrenoceptor when TM1 and the N-terminus were replaced with the corresponding regions from the 2C-adrenoceptor subtype. Also insertion of TM1 and the N-terminus regions from the 2C-adrenoceptor subtype. Also insertion of TM1 and the N-terminus of the 2B-adrenoceptor into the 2A-subtype significantly improved the binding of the 2B-adrenoceptor into the 2A-subtype significantly improved the binding affinities of two of these ligands (spiroxatrine and chlorpromazine) in comparison to affinities of two of these ligands (spiroxatrine and chlorpromazine) in comparison to the wild-type 2A-adrenoceptor. Similar effects were also seen for ARC239 and the wild-type 2A-adrenoceptor. Similar effects were also seen for ARC239 and prazosin, but the differences failed to reach statistical significance. Reciprocal effects prazosin, but the differences failed to reach statistical significance. Reciprocal effects were not seen in chimaeras based on the human 2B- and 2C-adrenoceptor subtypes. were not seen in chimaeras based on the human 2B- and 2C-adrenoceptor subtypes. When only TM1 was imported into the 2A-adrenoceptor from the 2C-adrenoceptor, When only TM1 was imported into the 2A-adrenoceptor from the 2C-adrenoceptor, the binding affinities did not differ from those observed for the chimaera containing the binding affinities did not differ from those observed for the chimaera containing both TM1 and the N-terminus from the 2C-adrenoceptor, indicating involvement of both TM1 and the N-terminus from the 2C-adrenoceptor, indicating involvement of TM1 in the observed binding affinity changes. TM1 in the observed binding affinity changes.

Results 61 Results 61

Table 13. Competition binding affinities of nine antagonist ligands obtained with Table 13. Competition binding affinities of nine antagonist ligands obtained with 3 3 [ H]RX821002 at wild-type (WT) and chimaeric 2-adrenoceptors expressed in CHO cells. [ H]RX821002 at wild-type (WT) and chimaeric 2-adrenoceptors expressed in CHO cells.

Human 2A Human 2A Ligand WT (AAA) BBA CCA  ACA Ligand WT (AAA) BBA CCA  ACA ARC239 1600 (1000-2600) 450 (160-1400) 820 (600-1110) 760 (480-1200) ARC239 1600 (1000-2600) 450 (160-1400) 820 (600-1110) 760 (480-1200) Prazosin 1500 (800-3100) 1400 (490-2400) 820 (600-1100) 910 (600-1500) Prazosin 1500 (800-3100) 1400 (490-2400) 820 (600-1100) 910 (600-1500) Spiperone 1200 (870-1600) 870 (450-1800) 560 (440-730)* 500 (320-770)** Spiperone 1200 (870-1600) 870 (450-1800) 560 (440-730)* 500 (320-770)** Spiroxatrine 550 (240-1300) 69 (34-140)** 71 (39-130)** 91 (48-180)** Spiroxatrine 550 (240-1300) 69 (34-140)** 71 (39-130)** 91 (48-180)** Chlorpromazine 600 (430-870) 160 (83-310)*** 120 (56-260)*** 160 (100-230)*** Chlorpromazine 600 (430-870) 160 (83-310)*** 120 (56-260)*** 160 (100-230)*** Atipamezole 2.1 (1.5-3.1) 2.0 (0.99-4.3) 1.2 (0.95-1.5) 1.1 (0.77-1.5) Atipamezole 2.1 (1.5-3.1) 2.0 (0.99-4.3) 1.2 (0.95-1.5) 1.1 (0.77-1.5) Idazoxan 22 (15-31) 15 (2.7-45) 17 (13-24) 13 (8.3-22) Idazoxan 22 (15-31) 15 (2.7-45) 17 (13-24) 13 (8.3-22) Clozapine 89 (54-150) 67 (32-150) 34 (17-67) 51 (35-77) Clozapine 89 (54-150) 67 (32-150) 34 (17-67) 51 (35-77) Rauwolscine 1.8 (0.78-4.5) 4.4 (2.0-9.8) 0.85 (0.40-2.1) 1.6 (0.82-3.1) Rauwolscine 1.8 (0.78-4.5) 4.4 (2.0-9.8) 0.85 (0.40-2.1) 1.6 (0.82-3.1)

Human 2B Human 2B Ligand WT (BBB) AAB CCB  Ligand WT (BBB) AAB CCB  ARC239 150 (40-570)♦♦♦ 240 (120-480) 98 (48-220) ARC239 150 (40-570)♦♦♦ 240 (120-480) 98 (48-220) Prazosin 47 (17-140)♦♦ 200 (120-360) 170 (89-310) Prazosin 47 (17-140)♦♦ 200 (120-360) 170 (89-310) Spiperone 12 (3.7-38)♦♦♦ 26 (11-65) 14 (6.0-38) Spiperone 12 (3.7-38)♦♦♦ 26 (11-65) 14 (6.0-38) Spiroxatrine 2.4 (1.2-5.0)♦♦♦ 6.8 (1.8-28) 1.6 (0.64-5.9) Spiroxatrine 2.4 (1.2-5.0)♦♦♦ 6.8 (1.8-28) 1.6 (0.64-5.9) Chlorpromazine 43 (20-100)♦♦♦ 53 (20-200) 46 (18-120) Chlorpromazine 43 (20-100)♦♦♦ 53 (20-200) 46 (18-120) Atipamezole 2.7 (0.56-14) 3.0 (1.7-5.6) 1.7 (0.80-4.0) Atipamezole 2.7 (0.56-14) 3.0 (1.7-5.6) 1.7 (0.80-4.0) Idazoxan 24 (18-34) 37 (29-48) 30 (19-51) Idazoxan 24 (18-34) 37 (29-48) 30 (19-51) Clozapine 12 (5.0-28)♦♦♦ 18 (8.7-45) 6.7 (1.9-23) Clozapine 12 (5.0-28)♦♦♦ 18 (8.7-45) 6.7 (1.9-23) Rauwolscine 1.1 (0.7-1.8) 1.2 (0.66-2.4) 2.7 (1.1-6.9) Rauwolscine 1.1 (0.7-1.8) 1.2 (0.66-2.4) 2.7 (1.1-6.9)

Human 2C Human 2C Ligand WT (CCC) AAC BBC  CAC Ligand WT (CCC) AAC BBC  CAC ARC239 130 (66-260)♦♦♦ 180 (86-380) n.d. 77 (34-180) ARC239 130 (66-260)♦♦♦ 180 (86-380) n.d. 77 (34-180) Prazosin 45 (21-81)♦♦ 110 (50-230) n.d. 120 (55-260) Prazosin 45 (21-81)♦♦ 110 (50-230) n.d. 120 (55-260) Spiperone 29 (9.2-81)♦♦♦ 26 (10-75) n.d. 39 (18-89) Spiperone 29 (9.2-81)♦♦♦ 26 (10-75) n.d. 39 (18-89) Spiroxatrine 13 (5.9-28)♦♦♦ 12 (5.8-27) n.d. 18 (8.4-31) Spiroxatrine 13 (5.9-28)♦♦♦ 12 (5.8-27) n.d. 18 (8.4-31) Chlorpromazine 260 (150-350)♦♦ 260 (110-720) n.d. 230 (130-400) Chlorpromazine 260 (150-350)♦♦ 260 (110-720) n.d. 230 (130-400) Atipamezole 4.1 (2.0-9.0) 5.3 (4.8-8.3) n.d. 10 (5.0-23) Atipamezole 4.1 (2.0-9.0) 5.3 (4.8-8.3) n.d. 10 (5.0-23) Idazoxan 71 (20-251) 48 (21-110) n.d. 18 (12-34) Idazoxan 71 (20-251) 48 (21-110) n.d. 18 (12-34) Clozapine 6.5 (2.5-18)♦♦♦ 5.1 (2.6-8.2) n.d. 1.8 (0.2-17) Clozapine 6.5 (2.5-18)♦♦♦ 5.1 (2.6-8.2) n.d. 1.8 (0.2-17) Rauwolscine 0.47 (0.21-1.2) 0.27 (0.12-0.63) n.d. 0.39 (0.20-0.83) Rauwolscine 0.47 (0.21-1.2) 0.27 (0.12-0.63) n.d. 0.39 (0.20-0.83)

Results are expressed as an apparent Ki (nM) and 95 % confidence intervals of three to six Results are expressed as an apparent Ki (nM) and 95 % confidence intervals of three to six independent experiments; n.d. (not determined). Receptors are named based on the exchanged independent experiments; n.d. (not determined). Receptors are named based on the exchanged domains (see Figure 14). Statistical significances are shown with symbols (one) P<0.05; (two) domains (see Figure 14). Statistical significances are shown with symbols (one) P<0.05; (two) P<0.01; (three) P<0.001. (*) chimaera versus wild-type 2A; (♦) receptor wild-type versus wild- P<0.01; (three) P<0.001. (*) chimaera versus wild-type 2A; (♦) receptor wild-type versus wild- type 2A. type 2A.

62 Discussion 62 Discussion

6 DISCUSSION 6 DISCUSSION

6.1 Structural and pharmacological comparison of the orthologous 6.1 Structural and pharmacological comparison of the orthologous 2-adrenoceptors 2-adrenoceptors

The ligand binding characteristics of the mammalian 2-adrenoceptor orthologues were The ligand binding characteristics of the mammalian 2-adrenoceptor orthologues were known to be very similar (see e.g. Uhlèn et al., 1998). This can be considered as another known to be very similar (see e.g. Uhlèn et al., 1998). This can be considered as another reflection of the highly conserved structure of the 2-adrenoceptors among various reflection of the highly conserved structure of the 2-adrenoceptors among various mammalian species. The zebrafish is a teleost fish and, though only distantly related to mammalian species. The zebrafish is a teleost fish and, though only distantly related to mammals, has recently gained increasing interest as a model organism to study mammals, has recently gained increasing interest as a model organism to study development and genetics, since many mutations in its genes cause phenotypes development and genetics, since many mutations in its genes cause phenotypes reminiscent of human diseases (Patton and Zon, 2001, Golling et al., 2002, Shin and reminiscent of human diseases (Patton and Zon, 2001, Golling et al., 2002, Shin and Fishman, 2002). Three genes coding for orthologues of the mammalian 2-adrenoceptors Fishman, 2002). Three genes coding for orthologues of the mammalian 2-adrenoceptors and two genes coding for an additional duplicated 2-adrenoceptor subtype have been and two genes coding for an additional duplicated 2-adrenoceptor subtype have been cloned from zebrafish (Ruuskanen et al., 2004). Although ~350 million years of cloned from zebrafish (Ruuskanen et al., 2004). Although ~350 million years of evolution separate mammalian and zebrafish 2-adrenoceptor genes from each other evolution separate mammalian and zebrafish 2-adrenoceptor genes from each other (Shin and Fishman, 2002, Ruuskanen et al., 2004), the ligand binding characteristics of (Shin and Fishman, 2002, Ruuskanen et al., 2004), the ligand binding characteristics of the human and zebrafish 2-adrenoceptors were found to be surprisingly similar. This the human and zebrafish 2-adrenoceptors were found to be surprisingly similar. This pharmacological conservation is perhaps constrained by evolution as all mammalian as pharmacological conservation is perhaps constrained by evolution as all mammalian as well as zebrafish 2-adrenoceptors recognise, bind and are activated in vivo by the same well as zebrafish 2-adrenoceptors recognise, bind and are activated in vivo by the same natural ligands, adrenaline and noradrenaline. However, there is no obvious reason why natural ligands, adrenaline and noradrenaline. However, there is no obvious reason why the binding affinities of synthetic ligands should also be similarly conserved, unless all the binding affinities of synthetic ligands should also be similarly conserved, unless all ligands bind to the receptors in a similar mode. One explanation may be that the ligands ligands bind to the receptors in a similar mode. One explanation may be that the ligands examined were chosen mainly based on their affinity differences between the human 2- examined were chosen mainly based on their affinity differences between the human 2- adrenoceptor subtypes. Differences between receptor orthologues could have emerged adrenoceptor subtypes. Differences between receptor orthologues could have emerged had a broader range of ligands been used. had a broader range of ligands been used. The similarity of the pharmacological profiles of the orthologous 2-adrenoceptors The similarity of the pharmacological profiles of the orthologous 2-adrenoceptors was not totally unexpected, if one compares the model structures of the three human was not totally unexpected, if one compares the model structures of the three human and five zebrafish 2-adrenoceptors with each other. The model comparison revealed and five zebrafish 2-adrenoceptors with each other. The model comparison revealed that only six of thirty-two amino acid residues vary within the proposed binding cavity that only six of thirty-two amino acid residues vary within the proposed binding cavity (Figure 6 in I). The duplicated zebrafish 2D-adrenoceptors are most similar with each (Figure 6 in I). The duplicated zebrafish 2D-adrenoceptors are most similar with each other, having only a single conservative exchange (at position xl2.52) within the other, having only a single conservative exchange (at position xl2.52) within the modelled binding cavity. In the receptor orthologues, three differences (at positions modelled binding cavity. In the receptor orthologues, three differences (at positions xl2.49, xl2.51 and 5.39) are present between the 2A-adrenoceptors of human and xl2.49, xl2.51 and 5.39) are present between the 2A-adrenoceptors of human and zebrafish. For 2B-adrenoceptors, only two residues (at positions xl2.51 and 5.43) are zebrafish. For 2B-adrenoceptors, only two residues (at positions xl2.51 and 5.43) are different between the human and zebrafish orthologues. The 2C-adrenoceptor different between the human and zebrafish orthologues. The 2C-adrenoceptor orthologues were found to be most divergent, having four variable residues (at orthologues were found to be most divergent, having four variable residues (at positions 2.57, xl2.49, xl2.51 and 5.43) in the proposed binding site. The overall positions 2.57, xl2.49, xl2.51 and 5.43) in the proposed binding site. The overall binding cavity of 2-adrenoceptors seems to be evolutionarily strictly conserved as it is binding cavity of 2-adrenoceptors seems to be evolutionarily strictly conserved as it is quite hydrophobic, with one-third of the side chains being aromatic and containing two quite hydrophobic, with one-third of the side chains being aromatic and containing two polar regions in TM3 and TM5, important for the binding of small catecholic ligands. polar regions in TM3 and TM5, important for the binding of small catecholic ligands. In terms of pharmacology, the human and zebrafish 2-adrenoceptors were In terms of pharmacology, the human and zebrafish 2-adrenoceptors were clustered based on the binding affinities of 20 chemically divergent ligands using the clustered based on the binding affinities of 20 chemically divergent ligands using the

Discussion 63 Discussion 63 binary tree approach and PCA (Figure 2 in I). Here, the duplicated zebrafish 2D- binary tree approach and PCA (Figure 2 in I). Here, the duplicated zebrafish 2D- adrenoceptors clustered together, as did also the human and zebrafish 2A-adrenoceptor adrenoceptors clustered together, as did also the human and zebrafish 2A-adrenoceptor orthologues. With respect to the 2B-adrenoceptors, the pharmacology of the human orthologues. With respect to the 2B-adrenoceptors, the pharmacology of the human receptor resembled more the characteristics of the duplicated zebrafish 2D- receptor resembled more the characteristics of the duplicated zebrafish 2D- adrenoceptor than those of its zebrafish 2B-orthologue, although more amino acids (4- adrenoceptor than those of its zebrafish 2B-orthologue, although more amino acids (4- 5) are variable between the human 2B-adrenoceptor and the zebrafish 2D- 5) are variable between the human 2B-adrenoceptor and the zebrafish 2D- adrenoceptors than between the 2B-orthologues, where only two amino acids differ. adrenoceptors than between the 2B-orthologues, where only two amino acids differ. The 2C-adrenoceptor orthologues appeared to be no more similar with each other than The 2C-adrenoceptor orthologues appeared to be no more similar with each other than with the other human and zebrafish 2-adrenoceptors. Obviously, as also evidenced with the other human and zebrafish 2-adrenoceptors. Obviously, as also evidenced here, clustering of receptors according to their ligand binding or other pharmacological here, clustering of receptors according to their ligand binding or other pharmacological properties is neither a sufficient nor a conclusive way to classify receptor subtypes, as properties is neither a sufficient nor a conclusive way to classify receptor subtypes, as compared with molecular phylogeny analyses. This is to some extent obvious as a compared with molecular phylogeny analyses. This is to some extent obvious as a small number of amino acid replacements may largely or totally explain the observed small number of amino acid replacements may largely or totally explain the observed pharmacological differences. Therefore, caution must be exercised when conducting pharmacological differences. Therefore, caution must be exercised when conducting receptor subtype classifications according to pharmacological observations. receptor subtype classifications according to pharmacological observations. In spite of the generally conserved pharmacology of the human and zebrafish 2- In spite of the generally conserved pharmacology of the human and zebrafish 2- adrenoceptors, some differences have been observed in the binding affinities of certain adrenoceptors, some differences have been observed in the binding affinities of certain compounds at orthologous 2-adrenoceptors. The best-known pharmacological compounds at orthologous 2-adrenoceptors. The best-known pharmacological difference characterized for orthologous 2-adrenoceptors is the binding affinity difference characterized for orthologous 2-adrenoceptors is the binding affinity difference of yohimbine and rauwolscine towards 2A-adrenoceptors from various difference of yohimbine and rauwolscine towards 2A-adrenoceptors from various species (Table 14). Mouse, rat, guinea-pig and bovine 2A-adrenoceptors bind species (Table 14). Mouse, rat, guinea-pig and bovine 2A-adrenoceptors bind rauwolscine and yohimbine with clearly lower affinity in comparison to the rauwolscine and yohimbine with clearly lower affinity in comparison to the orthologous receptors of human, pig, rabbit, chicken and zebrafish (Bylund et al., orthologous receptors of human, pig, rabbit, chicken and zebrafish (Bylund et al., 1988, Link et al., 1992, O'Rourke et al., 1994b, Svensson et al., 1996, Uhlèn et al., 1988, Link et al., 1992, O'Rourke et al., 1994b, Svensson et al., 1996, Uhlèn et al., 1998, Naselsky et al., 2001). Indeed, based on these pharmacological differences, the 1998, Naselsky et al., 2001). Indeed, based on these pharmacological differences, the rodent 2A-adrenoceptor was at one time misnamed as a fourth subtype, the 2D- rodent 2A-adrenoceptor was at one time misnamed as a fourth subtype, the 2D- adrenoceptor (Simonneaux et al., 1991). As rodents are frequently used as animal adrenoceptor (Simonneaux et al., 1991). As rodents are frequently used as animal models in pharmaceutical development, it is extremely important to understand, also at models in pharmaceutical development, it is extremely important to understand, also at the molecular level, the origin of their pharmacological differences from humans. the molecular level, the origin of their pharmacological differences from humans.

Table 14. Pharmacological comparison of 2A-adrenoceptors from various animal species. Ki Table 14. Pharmacological comparison of 2A-adrenoceptors from various animal species. Ki values (nM) derived from radioligand binding experiments. values (nM) derived from radioligand binding experiments. Human(a) Zebrafish(b) Pig Rabbit Chicken Mouse(a) Rat Bovine Human(a) Zebrafish(b) Pig Rabbit Chicken Mouse(a) Rat Bovine ARC239 1800 1800 365 100 000 133 500 749 352 ARC239 1800 1800 365 100 000 133 500 749 352 Rauwolscine 1.1 1 0.37 11.3 0.25 22 19 9.5 Rauwolscine 1.1 1 0.37 11.3 0.25 22 19 9.5 Ratio 1636 1800 986 8850 605 23 39 37 Ratio 1636 1800 986 8850 605 23 39 37 Adapted from Bylund, 2005. (a) Derived from Study II; (b) Derived from Study I Adapted from Bylund, 2005. (a) Derived from Study II; (b) Derived from Study I

Structural models of the 2A-adrenoceptors from eight animal species revealed that Structural models of the 2A-adrenoceptors from eight animal species revealed that most of the amino acid variability within the predicted binding cavity is located along most of the amino acid variability within the predicted binding cavity is located along XL2, while in the TM domains, only four amino acids (1.39, 3.33, 5.39 and 5.43) XL2, while in the TM domains, only four amino acids (1.39, 3.33, 5.39 and 5.43) differed among the 2A-orthologues (Figure 16). Of these, the serine-cysteine difference differed among the 2A-orthologues (Figure 16). Of these, the serine-cysteine difference at position 5.43 was previously shown to account for some of the variation in yohimbine at position 5.43 was previously shown to account for some of the variation in yohimbine and rauwolscine binding affinities between the human and mouse 2A-adrenoceptors and rauwolscine binding affinities between the human and mouse 2A-adrenoceptors (Link et al., 1992, Cockcroft et al., 2000). Across the entire species comparison, the (Link et al., 1992, Cockcroft et al., 2000). Across the entire species comparison, the presence of a cysteine or a serine at position 5.43 correlated with the receptor’s affinity presence of a cysteine or a serine at position 5.43 correlated with the receptor’s affinity

64 Discussion 64 Discussion for yohimbine, with serine being linked to lower affinity and cysteine to higher affinity. for yohimbine, with serine being linked to lower affinity and cysteine to higher affinity. However, the chicken receptor represents an exception to this rule, as it has a serine at However, the chicken receptor represents an exception to this rule, as it has a serine at position 5.43 but a pharmacological profile similar to the human receptor, suggesting that position 5.43 but a pharmacological profile similar to the human receptor, suggesting that also other regions of the binding site may have effects on yohimbine binding. Indeed, the also other regions of the binding site may have effects on yohimbine binding. Indeed, the chicken receptor differs from the mouse receptor within XL2 at several positions, as one chicken receptor differs from the mouse receptor within XL2 at several positions, as one amino acid downstream of xl2.49 is deleted in the chicken receptor, and glycine is found amino acid downstream of xl2.49 is deleted in the chicken receptor, and glycine is found at xl2.49 whereas serine is present in the mouse receptor. This suggests that XL2 may at xl2.49 whereas serine is present in the mouse receptor. This suggests that XL2 may also contribute to the binding of yohimbine at the chicken 2A-adrenoceptor. also contribute to the binding of yohimbine at the chicken 2A-adrenoceptor.

Figure 16. Schematic Figure 16. Schematic comparison of the 2A- comparison of the 2A- adrenoceptors from several adrenoceptors from several species. Residues different species. Residues different from the human receptor are from the human receptor are indicated with gray. Amino indicated with gray. Amino acid deletions are shown as acid deletions are shown as triangles. Receptors with triangles. Receptors with “mouse-like” low affinity for “mouse-like” low affinity for yohimbine are boxed. The yohimbine are boxed. The residues in XL2 believed to residues in XL2 believed to face the binding cavity are face the binding cavity are indicated with stars. indicated with stars.

Discussion 65 Discussion 65

Using site-directed mutagenesis targeted to residues at positions xl2.49 and xl2.51 Using site-directed mutagenesis targeted to residues at positions xl2.49 and xl2.51 of the human and mouse 2A-adrenoceptors, we found that these positions influence the of the human and mouse 2A-adrenoceptors, we found that these positions influence the binding preference of yohimbine and rauwolscine of the human versus mouse 2A- binding preference of yohimbine and rauwolscine of the human versus mouse 2A- adrenoceptors. Reciprocal mutations for arginine/serine at xl2.49 and glutamate/lysine adrenoceptors. Reciprocal mutations for arginine/serine at xl2.49 and glutamate/lysine at xl2.51, as well as cysteine/serine at position 5.43 reversed the binding profiles of the at xl2.51, as well as cysteine/serine at position 5.43 reversed the binding profiles of the human and mouse 2A-adrenoceptors for yohimbine, rauwolscine and RS-79948-197, human and mouse 2A-adrenoceptors for yohimbine, rauwolscine and RS-79948-197, pointing to a role of XL2 in the determination of species-specific ligand binding pointing to a role of XL2 in the determination of species-specific ligand binding profiles for yohimbine-like compounds. Nonetheless, the amino acid at position 5.43 is profiles for yohimbine-like compounds. Nonetheless, the amino acid at position 5.43 is still believed to be a major determinant of the binding profile, as the human, rabbit, still believed to be a major determinant of the binding profile, as the human, rabbit, bovine and guinea-pig receptors are identical at xl2.49 and xl2.51 but have different bovine and guinea-pig receptors are identical at xl2.49 and xl2.51 but have different pharmacological profiles. Of these, the bovine and guinea-pig receptors have a serine pharmacological profiles. Of these, the bovine and guinea-pig receptors have a serine at position 5.43 and show mouse-like pharmacology, while the human and rabbit at position 5.43 and show mouse-like pharmacology, while the human and rabbit receptors have a cysteine at 5.43 and show human-like pharmacology. Yohimbine receptors have a cysteine at 5.43 and show human-like pharmacology. Yohimbine binding affinity thus appears to segregate only according to the residue present at binding affinity thus appears to segregate only according to the residue present at position 5.43. However, the similarity of the residues at xl2.49 and xl2.51 cannot be position 5.43. However, the similarity of the residues at xl2.49 and xl2.51 cannot be interpreted as evidence that these amino acids are not involved in mediating the interpreted as evidence that these amino acids are not involved in mediating the interactions also in these receptors. interactions also in these receptors. In addition to the differences in XL2 and at position 5.43, some other amino acids In addition to the differences in XL2 and at position 5.43, some other amino acids facing the ligand binding cavity in the TM regions were also found to differ between facing the ligand binding cavity in the TM regions were also found to differ between the 2A-orthologues. For example, the chicken and zebrafish receptors have an the 2A-orthologues. For example, the chicken and zebrafish receptors have an isoleucine (I) at position 5.39 instead of a valine (V) that is present in the other species, isoleucine (I) at position 5.39 instead of a valine (V) that is present in the other species, the rabbit receptor has a methionine (M) at position 3.33 instead of a valine in the the rabbit receptor has a methionine (M) at position 3.33 instead of a valine in the others and the guinea-pig and zebrafish receptors have a valine at position 1.39 instead others and the guinea-pig and zebrafish receptors have a valine at position 1.39 instead of an alanine (A). These differences may to some extent also contribute to the binding of an alanine (A). These differences may to some extent also contribute to the binding profiles of the receptors towards different ligands, but so far this has not been profiles of the receptors towards different ligands, but so far this has not been experimentally investigated. experimentally investigated.

6.2 The second extracellular loop forms part of the ligand binding site in 6.2 The second extracellular loop forms part of the ligand binding site in 2-adrenoceptors 2-adrenoceptors Residues in the TM domains are generally held responsible for ligand binding in Residues in the TM domains are generally held responsible for ligand binding in adrenoceptors, as they contain the main contact sites involved in the binding of the adrenoceptors, as they contain the main contact sites involved in the binding of the endogenous catecholamines, adrenaline and noradrenaline. However, with the recent endogenous catecholamines, adrenaline and noradrenaline. However, with the recent appearance of crystal structures of -adrenoceptors it has been established that the appearance of crystal structures of -adrenoceptors it has been established that the orthosteric ligand binding site of adrenoceptors extends from the conserved TM core to orthosteric ligand binding site of adrenoceptors extends from the conserved TM core to the extracellular surface where the extracellular regions form the top of the binding site the extracellular surface where the extracellular regions form the top of the binding site (Cherezov et al., 2007, Rasmussen et al., 2007, Warne et al., 2008). In -adrenoceptors (Cherezov et al., 2007, Rasmussen et al., 2007, Warne et al., 2008). In -adrenoceptors as well as in other crystal structures of GPCRs, the conserved cysteine at xl2.50 is as well as in other crystal structures of GPCRs, the conserved cysteine at xl2.50 is attached with a disulphide bond to another cysteine, C3.25 in TM3, constraining the attached with a disulphide bond to another cysteine, C3.25 in TM3, constraining the position of XL2 above the ligand binding site. As these cysteines are highly conserved position of XL2 above the ligand binding site. As these cysteines are highly conserved among rhodopsin-like GPCRs, it is very likely that XL2 is folded similarly also in 2- among rhodopsin-like GPCRs, it is very likely that XL2 is folded similarly also in 2- adrenoceptors. This would make the positions in the vicinity of the conserved Cxl2.50 adrenoceptors. This would make the positions in the vicinity of the conserved Cxl2.50 exposed to the binding cavity and allow their involvement in ligand binding. In exposed to the binding cavity and allow their involvement in ligand binding. In addition, the sequence alignment comparisons of orthologous and paralogous 2- addition, the sequence alignment comparisons of orthologous and paralogous 2- adrenoceptors (see e.g. Supplementary material in Ruuskanen et al., 2004) support the adrenoceptors (see e.g. Supplementary material in Ruuskanen et al., 2004) support the

66 Discussion 66 Discussion concept that not only the TM domains but also XL2 is conserved during evolution, as concept that not only the TM domains but also XL2 is conserved during evolution, as in many species a polar asparagine (D) is present at position xl2.53 and other polar in many species a polar asparagine (D) is present at position xl2.53 and other polar residues such as arginine (R), glutamate (E), lysine (K) and glutamine (Q) are often residues such as arginine (R), glutamate (E), lysine (K) and glutamine (Q) are often found at xl2.49 and xl2.51, pointing to a functional role for this domain. Previously, in found at xl2.49 and xl2.51, pointing to a functional role for this domain. Previously, in other monoamine receptors closely related to 2-adrenoceptors, residues in XL2 have other monoamine receptors closely related to 2-adrenoceptors, residues in XL2 have been implicated in ligand binding e.g. in the 1-adrenoceptor subtypes A and B, where been implicated in ligand binding e.g. in the 1-adrenoceptor subtypes A and B, where substitutions at xl2.51-xl2.53 affected the subtype-selectivity of phentolamine and substitutions at xl2.51-xl2.53 affected the subtype-selectivity of phentolamine and WB4101 (Zhao et al., 1996). In the dopamine D2 receptor, point mutations of xl2.49 WB4101 (Zhao et al., 1996). In the dopamine D2 receptor, point mutations of xl2.49 and xl2.51-xl2.52 reduced the binding affinity of N-methylspiperone (Shi and Javitch, and xl2.51-xl2.52 reduced the binding affinity of N-methylspiperone (Shi and Javitch, 2004). In the type-1D serotonin receptor, replacements in XL2 affected the binding of 2004). In the type-1D serotonin receptor, replacements in XL2 affected the binding of ketanserin (Wurch et al., 1998, Wurch and Pauwels, 2000). As a follow-up to these ketanserin (Wurch et al., 1998, Wurch and Pauwels, 2000). As a follow-up to these studies, results from our xl2.49- and xl2.51-substituted human and mouse 2A- studies, results from our xl2.49- and xl2.51-substituted human and mouse 2A- adrenoceptors provided the first evidence that XL2 is involved in ligand binding also in adrenoceptors provided the first evidence that XL2 is involved in ligand binding also in 2-adrenoceptors. Molecular models, based on either bovine rhodopsin or the human 2-adrenoceptors. Molecular models, based on either bovine rhodopsin or the human 2-adrenoceptor, suggested that the ligand binding cavity of 2-adrenoceptors is 2-adrenoceptor, suggested that the ligand binding cavity of 2-adrenoceptors is elongated and oriented horizontally with respect to the cell membrane, with XL2 elongated and oriented horizontally with respect to the cell membrane, with XL2 folded above the binding cavity and forming part of the binding site, although the folded above the binding cavity and forming part of the binding site, although the entire three-dimensional structure is likely to be more similar to the -adrenoceptor entire three-dimensional structure is likely to be more similar to the -adrenoceptor structures than to rhodopsin. In the -adrenoceptor structures (Cherezov et al., 2007, structures than to rhodopsin. In the -adrenoceptor structures (Cherezov et al., 2007, Warne et al., 2008), XL2 forms a hairpin that contains an -helical segment and a Warne et al., 2008), XL2 forms a hairpin that contains an -helical segment and a second internal disulphide bridge that is not present in the 2-adrenoceptor subtypes or second internal disulphide bridge that is not present in the 2-adrenoceptor subtypes or in other crystal structures of GPCRs. Again in contrast to rhodopsin, in the - in other crystal structures of GPCRs. Again in contrast to rhodopsin, in the - adrenoceptor structures XL2 is located clearly deeper towards the centre of the binding adrenoceptor structures XL2 is located clearly deeper towards the centre of the binding cavity and is shifted away from TM5 towards TM1. As a consequence of this shift, in cavity and is shifted away from TM5 towards TM1. As a consequence of this shift, in 2-adrenoceptor-based receptor models xl2.49 is less exposed to the binding cavity 2-adrenoceptor-based receptor models xl2.49 is less exposed to the binding cavity than in rhodopsin-based models, suggesting that there is a relatively bigger impact for than in rhodopsin-based models, suggesting that there is a relatively bigger impact for xl2.51 on binding affinity changes observed for the human and mouse 2A- xl2.51 on binding affinity changes observed for the human and mouse 2A- adrenoceptors. On the other hand, this shift also allows position xl2.54 to contribute to adrenoceptors. On the other hand, this shift also allows position xl2.54 to contribute to the binding cavity surface. However, an aspartic acid (D) at this position is rather the binding cavity surface. However, an aspartic acid (D) at this position is rather conserved among 2A-adrenoceptor orthologues as the zebrafish receptor makes the conserved among 2A-adrenoceptor orthologues as the zebrafish receptor makes the only exception to this by having a lysine (K) at xl2.54. Therefore this position cannot only exception to this by having a lysine (K) at xl2.54. Therefore this position cannot account for the interspecies differences in the binding of yohimbine. account for the interspecies differences in the binding of yohimbine. Most available experimental evidence suggests that antagonists bind at least in part Most available experimental evidence suggests that antagonists bind at least in part within the same orthosteric binding pocket where agonists bind in 2-adrenoceptors, within the same orthosteric binding pocket where agonists bind in 2-adrenoceptors, and that ligands containing a positively charged amine moiety form a direct ion-pair and that ligands containing a positively charged amine moiety form a direct ion-pair interaction with the conserved aspartic acid (D3.32) in TM3 (Frang et al., 2001, Shi interaction with the conserved aspartic acid (D3.32) in TM3 (Frang et al., 2001, Shi and Javitch, 2002). However, when yohimbine, rauwolscine and RS-79948-197 were and Javitch, 2002). However, when yohimbine, rauwolscine and RS-79948-197 were docked into the orthosteric pocket of the human 2A-adrenoceptor model, direct docked into the orthosteric pocket of the human 2A-adrenoceptor model, direct interactions with the amino acid residues at 5.43, xl2.49 and xl2.51 as well as F7.39 interactions with the amino acid residues at 5.43, xl2.49 and xl2.51 as well as F7.39 (Suryanarayana et al., 1991) were possible only when the compounds were placed (Suryanarayana et al., 1991) were possible only when the compounds were placed close to TM5, where direct ionic interactions with D3.32 became unlikely (Figure 17). close to TM5, where direct ionic interactions with D3.32 became unlikely (Figure 17). This different binding mode of antagonists was also supported by other docking results This different binding mode of antagonists was also supported by other docking results when twelve antagonist ligands were automatically docked to rhodopsin-based 2A- when twelve antagonist ligands were automatically docked to rhodopsin-based 2A- adrenoceptor models (see Xhaard et al., 2005). Here, ion-pair formation with D3.32 adrenoceptor models (see Xhaard et al., 2005). Here, ion-pair formation with D3.32 would have required deviation of the models from the template structure, e.g. a rotation would have required deviation of the models from the template structure, e.g. a rotation of TM3 to relocate D3.32 more centrally within the binding cavity, and creation of new of TM3 to relocate D3.32 more centrally within the binding cavity, and creation of new

Discussion 67 Discussion 67 space near TM2/TM7 so that atoms of the antagonist ligands would not overlap with space near TM2/TM7 so that atoms of the antagonist ligands would not overlap with TM5. Thus, it was postulated that many antagonist ligands, though not the TM5. Thus, it was postulated that many antagonist ligands, though not the quinazolines, are unlikely to form ion-pairs with D3.32 (Xhaard et al., 2005). An quinazolines, are unlikely to form ion-pairs with D3.32 (Xhaard et al., 2005). An alternative binding mode for these antagonists was proposed, whereby the positively alternative binding mode for these antagonists was proposed, whereby the positively charged nitrogen of the antagonist ligands is stabilized by cation-π interactions with charged nitrogen of the antagonist ligands is stabilized by cation-π interactions with aromatic residues (e.g., F6.51), and the antagonists simultaneously interact with D3.32 aromatic residues (e.g., F6.51), and the antagonists simultaneously interact with D3.32 via carboxylate-aromatic interactions. In this proposal, position 5.43 may provide a via carboxylate-aromatic interactions. In this proposal, position 5.43 may provide a better anchoring point for antagonist binding than D3.32. In addition, this hypothesis is better anchoring point for antagonist binding than D3.32. In addition, this hypothesis is also supported by the impairment of ligand binding seen for receptors mutated at D3.32 also supported by the impairment of ligand binding seen for receptors mutated at D3.32 (Ho et al., 1992, Porter et al., 1996). Although the assumption of different binding (Ho et al., 1992, Porter et al., 1996). Although the assumption of different binding modes for antagonists would enable better contact with XL2, which is in good modes for antagonists would enable better contact with XL2, which is in good agreement with the experimental results obtained for yohimbine, rauwolscine and RS- agreement with the experimental results obtained for yohimbine, rauwolscine and RS- 79948-197, this cannot be generalized to all antagonist ligands, e.g. as seen for MK- 79948-197, this cannot be generalized to all antagonist ligands, e.g. as seen for MK- 912. When this compound was docked on top of yohimbine, equally close to TM5, the 912. When this compound was docked on top of yohimbine, equally close to TM5, the ligand interfered with the receptor structure, especially with W6.48 in TM6, evidence ligand interfered with the receptor structure, especially with W6.48 in TM6, evidence of different binding modes for MK-912 and yohimbine (Figure 17d). This is consistent of different binding modes for MK-912 and yohimbine (Figure 17d). This is consistent with the present experimental results, where the binding of MK-912 was affected with the present experimental results, where the binding of MK-912 was affected neither by mutations at 5.43 nor at xl2.49/xl2.51. neither by mutations at 5.43 nor at xl2.49/xl2.51.

A B A B

C D C D

Figure 17. Close-up view of (A) yohimbine, (B) rauwolscine, (C) RS-79948-97 and (D) MK- Figure 17. Close-up view of (A) yohimbine, (B) rauwolscine, (C) RS-79948-97 and (D) MK- 912 docked into the molecular model of the human 2A-adrenoceptor. TM6 and TM7 helices 912 docked into the molecular model of the human 2A-adrenoceptor. TM6 and TM7 helices are shown in dark. Three residues found to affect the binding of yohimbine, rauwolscine and are shown in dark. Three residues found to affect the binding of yohimbine, rauwolscine and RS-79948-197 are shown; cysteine 5.43 (left of the ligand), glutamate xl2.51 (left above the RS-79948-197 are shown; cysteine 5.43 (left of the ligand), glutamate xl2.51 (left above the ligand) and arginine xl2.49 (right above the ligand). Modified from paper II. ligand) and arginine xl2.49 (right above the ligand). Modified from paper II.

XL2 is thus involved in ligand binding to 2-adrenoceptors and many other types of XL2 is thus involved in ligand binding to 2-adrenoceptors and many other types of GPCRs. For drug discovery and development, XL2 is an ideal target, also for 2- GPCRs. For drug discovery and development, XL2 is an ideal target, also for 2- adrenoceptors, as it is highly variable in contrast to the TM domains. Due to the adrenoceptors, as it is highly variable in contrast to the TM domains. Due to the constrained position of XL2, especially the amino acids in the vicinity of the conserved constrained position of XL2, especially the amino acids in the vicinity of the conserved cysteine at xl2.50 are likely to face the orthosteric binding site. In the human 2- cysteine at xl2.50 are likely to face the orthosteric binding site. In the human 2-

68 Discussion 68 Discussion adrenoceptor subtypes, the amino acids at these positions are rather variable: adrenoceptor subtypes, the amino acids at these positions are rather variable: [R/Q/Q]xl2.49 ([human 2A/human 2B/human 2C]), [E/K/G]xl2.51 and xl2.52 [R/Q/Q]xl2.49 ([human 2A/human 2B/human 2C]), [E/K/G]xl2.51 and xl2.52 [I/L/L]. One would expect these differences to have at least some influence on the [I/L/L]. One would expect these differences to have at least some influence on the binding properties of certain ligands, especially antagonists. However, the binding properties of certain ligands, especially antagonists. However, the experimental results obtained with single- and double-mutated human 2A- and 2C- experimental results obtained with single- and double-mutated human 2A- and 2C- adrenoceptors at positions xl2.49 and xl2.51 showed only marginal effects on the adrenoceptors at positions xl2.49 and xl2.51 showed only marginal effects on the binding affinity of 19 ligands (unpublished results). The other positions along XL2 binding affinity of 19 ligands (unpublished results). The other positions along XL2 suggested to be exposed in the binding site, e.g. xl2.52 [I/L/L] and xl2.54 [D/Q/D], suggested to be exposed in the binding site, e.g. xl2.52 [I/L/L] and xl2.54 [D/Q/D], have so far not been evaluated experimentally. have so far not been evaluated experimentally. In addition to ligand binding, XL2 was recently proposed to have a functional role In addition to ligand binding, XL2 was recently proposed to have a functional role in receptor activation. In the -adrenoceptor structure, the carboxyl group of the polar in receptor activation. In the -adrenoceptor structure, the carboxyl group of the polar Dxl2.51 forms a salt bridge with a positively charged lysine, K7.32, and ligands may Dxl2.51 forms a salt bridge with a positively charged lysine, K7.32, and ligands may have allosteric effects by influencing this interaction, as an agonist, an antagonist and have allosteric effects by influencing this interaction, as an agonist, an antagonist and an inverse agonist were found to stabilise distinct conformations of the extracellular an inverse agonist were found to stabilise distinct conformations of the extracellular loops (Bokoch et al., 2010). For the 1B-adrenoceptor, mutations at position 7.32 were loops (Bokoch et al., 2010). For the 1B-adrenoceptor, mutations at position 7.32 were also shown to affect the binding affinity of noradrenaline and adrenaline, evidence for also shown to affect the binding affinity of noradrenaline and adrenaline, evidence for the importance of this ion-pair interaction between XL2 and TM7 for the resting the importance of this ion-pair interaction between XL2 and TM7 for the resting receptor conformation (Porter et al., 1996). If this salt bridge exists also in 2- receptor conformation (Porter et al., 1996). If this salt bridge exists also in 2- adrenoceptors, it may have a functional role in regulation of receptor conformations adrenoceptors, it may have a functional role in regulation of receptor conformations and in addition be able to influence ligand binding. The corresponding amino acids are and in addition be able to influence ligand binding. The corresponding amino acids are [R/H/G]7.32 and [E/K/G]xl2.51 in the human 2-adrenoceptors, which would make an [R/H/G]7.32 and [E/K/G]xl2.51 in the human 2-adrenoceptors, which would make an analogous salt bridge possible in the 2A-subtype, but not in the 2B- and 2C-subtypes. analogous salt bridge possible in the 2A-subtype, but not in the 2B- and 2C-subtypes.

6.3 Indirect effects may influence pharmacological profiles 6.3 Indirect effects may influence pharmacological profiles It is tempting to assume that an experimentally observed effect on ligand binding to a It is tempting to assume that an experimentally observed effect on ligand binding to a mutated receptor is a direct consequence of changed atomic contacts between the mutated receptor is a direct consequence of changed atomic contacts between the ligand and the amino acids exposed in the binding cavity of the receptor. However, a ligand and the amino acids exposed in the binding cavity of the receptor. However, a mutation may cause many indirect effects on the receptor’s structure, e.g. affect protein mutation may cause many indirect effects on the receptor’s structure, e.g. affect protein folding, alter its post-translational modifications, or disrupt its cell surface expression folding, alter its post-translational modifications, or disrupt its cell surface expression (Shi and Javitch, 2002). Even the presence of a mutant receptor in the plasma (Shi and Javitch, 2002). Even the presence of a mutant receptor in the plasma membrane is not evidence of normal protein folding, as structural rearrangements do membrane is not evidence of normal protein folding, as structural rearrangements do not always lead to intracellular retention. Mutations may also change side-chain not always lead to intracellular retention. Mutations may also change side-chain volumes, charge distributions, hydrophobicity and other physico-chemical properties of volumes, charge distributions, hydrophobicity and other physico-chemical properties of the receptor which may indirectly affect its normal folding. Therefore, it can be the receptor which may indirectly affect its normal folding. Therefore, it can be extremely difficult to differentiate between the direct and indirect effects of mutations extremely difficult to differentiate between the direct and indirect effects of mutations on receptor function. on receptor function. The process of ligand binding is more dynamic than was considered in the era of The process of ligand binding is more dynamic than was considered in the era of classical receptor pharmacology, and the binding affinity is determined by more factors classical receptor pharmacology, and the binding affinity is determined by more factors than the direct interactions of the ligand and the receptor. Even the dynamic flexibility than the direct interactions of the ligand and the receptor. Even the dynamic flexibility of different segments of the receptor protein may have effects on ligand recognition. In of different segments of the receptor protein may have effects on ligand recognition. In addition, there are many examples of positive and negative allosteric modulators that addition, there are many examples of positive and negative allosteric modulators that can regulate binding affinity (Milligan and Smith, 2007). The presence of charged can regulate binding affinity (Milligan and Smith, 2007). The presence of charged groups even outside of the binding pocket may affect the protonation state of groups even outside of the binding pocket may affect the protonation state of compounds and in that way indirectly influence their binding to the receptor. Some compounds and in that way indirectly influence their binding to the receptor. Some

Discussion 69 Discussion 69 residues facing away from the binding cavity can undergo ionic interactions with other residues facing away from the binding cavity can undergo ionic interactions with other residues of the receptor, and ligands may have allosteric effects through influences on residues of the receptor, and ligands may have allosteric effects through influences on these interactions, as was recently reported for the 2-adrenoceptor where Dxl2.51 these interactions, as was recently reported for the 2-adrenoceptor where Dxl2.51 forms a salt bridge with K7.32 (Bokoch et al., 2010). It is actually possible that the forms a salt bridge with K7.32 (Bokoch et al., 2010). It is actually possible that the effects of the present XL2 mutations on the binding of yohimbine analogues at human effects of the present XL2 mutations on the binding of yohimbine analogues at human and mouse 2A-adrenoceptors may be mediated through this ionic interaction instead of and mouse 2A-adrenoceptors may be mediated through this ionic interaction instead of direct interactions between ligand and receptor. Position 7.32 in TM7 of the 2- direct interactions between ligand and receptor. Position 7.32 in TM7 of the 2- adrenoceptors has so far not been subjected to mutagenesis experiments, but this would adrenoceptors has so far not been subjected to mutagenesis experiments, but this would be highly interesting from this point of view. Another example of allosteric effects on be highly interesting from this point of view. Another example of allosteric effects on ligand binding is that mutations at position 2.50 of the 2A-adrenoceptor lead to ligand binding is that mutations at position 2.50 of the 2A-adrenoceptor lead to decreased agonist binding affinity, although this position is relatively far from the decreased agonist binding affinity, although this position is relatively far from the orthosteric agonist binding site (Ceresa and Limbird, 1994). In monoamine-binding orthosteric agonist binding site (Ceresa and Limbird, 1994). In monoamine-binding GPCRs, the aspartate at 2.50 has been shown to form a sodium-binding site that plays a GPCRs, the aspartate at 2.50 has been shown to form a sodium-binding site that plays a role in receptor activation and thus allows allosteric modulation of the receptor-ligand role in receptor activation and thus allows allosteric modulation of the receptor-ligand interaction (Neve et al., 2001). interaction (Neve et al., 2001). The competition binding results obtained with the chimaeric human 2- The competition binding results obtained with the chimaeric human 2- adrenoceptors pointed to the involvement of TM1 in defining the specific adrenoceptors pointed to the involvement of TM1 in defining the specific pharmacological profile of the human 2A-adrenoceptor for antagonists with a large pharmacological profile of the human 2A-adrenoceptor for antagonists with a large molecular structure. However, molecular models indicated that TM1 has very limited molecular structure. However, molecular models indicated that TM1 has very limited exposure in the binding pocket, as it is distant from the orthosteric binding cavity. exposure in the binding pocket, as it is distant from the orthosteric binding cavity. Actually, in the 2-adrenoceptor-based 2-adrenoceptor models, the distance of TM1 Actually, in the 2-adrenoceptor-based 2-adrenoceptor models, the distance of TM1 from TM5 and the orthosteric binding site is even longer than in rhodopsin-based from TM5 and the orthosteric binding site is even longer than in rhodopsin-based models (26.5 Å vs. 23.5 Å, respectively) (see Figure 3 in III). Furthermore, access to models (26.5 Å vs. 23.5 Å, respectively) (see Figure 3 in III). Furthermore, access to TM1 is blocked by side-chain atoms from TM2 and TM7 (see Figure 4 in III). Thus, TM1 is blocked by side-chain atoms from TM2 and TM7 (see Figure 4 in III). Thus, indirect effects are more likely to be involved rather than direct receptor-ligand indirect effects are more likely to be involved rather than direct receptor-ligand interactions. It may be too speculative to try to define the exact mechanisms of such interactions. It may be too speculative to try to define the exact mechanisms of such indirect effects, especially as no reciprocal effects were observed in the 2B- and 2C- indirect effects, especially as no reciprocal effects were observed in the 2B- and 2C- adrenoceptor-based chimaeras. In addition to this study, there are also some other adrenoceptor-based chimaeras. In addition to this study, there are also some other previous investigations on 2-adrenoceptors where indirect effects could be involved previous investigations on 2-adrenoceptors where indirect effects could be involved rather than specific side-chain interactions. For example, substitution of five amino rather than specific side-chain interactions. For example, substitution of five amino acids in the third cytoplasmic loop of the rat 2A-adrenoceptor was shown to affect the acids in the third cytoplasmic loop of the rat 2A-adrenoceptor was shown to affect the binding affinity of several ligands including e.g. oxymetazoline and prazosin binding affinity of several ligands including e.g. oxymetazoline and prazosin (Venkataraman et al., 1997). Mutations of the serines at positions 2.61 in TM2 and (Venkataraman et al., 1997). Mutations of the serines at positions 2.61 in TM2 and 7.46 in TM7 reduced the stereoselective binding of the (-)-enantiomers of 7.46 in TM7 reduced the stereoselective binding of the (-)-enantiomers of catecholamine agonists to the 2A-adrenoceptor (Hehr et al., 1997, Hieble et al., 1998). catecholamine agonists to the 2A-adrenoceptor (Hehr et al., 1997, Hieble et al., 1998). It is possible that some of these findings may be explained by effects related to the It is possible that some of these findings may be explained by effects related to the charge distribution or overall shape of the binding pocket. charge distribution or overall shape of the binding pocket. Although the recent crystal structures of some GPCRs have very significantly Although the recent crystal structures of some GPCRs have very significantly improved our understanding of structural determinants important for ligand binding, improved our understanding of structural determinants important for ligand binding, they have not so far produced real breakthroughs in GPCR drug discovery. This could they have not so far produced real breakthroughs in GPCR drug discovery. This could be because these structures are static “snapshots” of receptor structure, and indeed, all be because these structures are static “snapshots” of receptor structure, and indeed, all of them represent inactive receptor conformations as they were determined while the of them represent inactive receptor conformations as they were determined while the receptor was complexed with antagonist ligands (the human 2-adrenoceptor with receptor was complexed with antagonist ligands (the human 2-adrenoceptor with timolol (Hanson et al., 2008); the turkey 1-adrenoceptor with cyanopindolol (Warne timolol (Hanson et al., 2008); the turkey 1-adrenoceptor with cyanopindolol (Warne et al., 2008); the human A2A adenosine receptor with ZM241385 (Jaakola et al., 2008); et al., 2008); the human A2A adenosine receptor with ZM241385 (Jaakola et al., 2008); the human dopamine D3 receptor with eticlopride (Chien et al., 2010); the CXCR4 the human dopamine D3 receptor with eticlopride (Chien et al., 2010); the CXCR4

70 Discussion 70 Discussion chemokine receptor with IT1t (Wu et al., 2010)) or inverse agonists (the human 2- chemokine receptor with IT1t (Wu et al., 2010)) or inverse agonists (the human 2- adrenoceptor with carazolol (Cherezov et al., 2007, Rasmussen et al., 2007, adrenoceptor with carazolol (Cherezov et al., 2007, Rasmussen et al., 2007, Rosenbaum et al., 2007) and bovine rhodopsin covalently bound to 11-cis retinal Rosenbaum et al., 2007) and bovine rhodopsin covalently bound to 11-cis retinal (Palczewski et al., 2000, Okada et al., 2004)). There is no doubt that these structures (Palczewski et al., 2000, Okada et al., 2004)). There is no doubt that these structures are very useful for in silico drug design, but the need for additional structures, are very useful for in silico drug design, but the need for additional structures, especially structures in an active state, is apparent. In addition, receptor structures in especially structures in an active state, is apparent. In addition, receptor structures in complex with other receptors (heterodimers) as well as structures which are linked with complex with other receptors (heterodimers) as well as structures which are linked with signaling molecules like G-proteins are needed for structure-based drug design. A good signaling molecules like G-proteins are needed for structure-based drug design. A good example of this is a recent molecular docking study where the human 2-adrenoceptor example of this is a recent molecular docking study where the human 2-adrenoceptor crystal structure was used as the target and six new 2-adrenoceptor-selective ligands crystal structure was used as the target and six new 2-adrenoceptor-selective ligands were identified, all of them exhibiting inverse agonist activity (Kolb et al., 2009). were identified, all of them exhibiting inverse agonist activity (Kolb et al., 2009). Therefore, when designing new compounds that should at the same time be selective Therefore, when designing new compounds that should at the same time be selective for a certain receptor subtype and in addition have the desired functional activity, any for a certain receptor subtype and in addition have the desired functional activity, any information on the structural determinants of ligand binding, also including allosteric information on the structural determinants of ligand binding, also including allosteric effects and other indirect effects, should be considered as potentially valuable. effects and other indirect effects, should be considered as potentially valuable.

Conclusions 71 Conclusions 71

7 CONCLUSIONS 7 CONCLUSIONS

The present ligand binding studies in combination with site-directed mutagenesis, The present ligand binding studies in combination with site-directed mutagenesis, molecular modelling and docking simulations of 2-adrenoceptors indicated structural molecular modelling and docking simulations of 2-adrenoceptors indicated structural determinants that contribute to specific ligand binding and recognition at 2- determinants that contribute to specific ligand binding and recognition at 2- adrenoceptor subtypes from different animal species. These findings will hopefully adrenoceptor subtypes from different animal species. These findings will hopefully promote the design and development of new 2-adrenoceptor selective drug. The main promote the design and development of new 2-adrenoceptor selective drug. The main results and conclusions were: results and conclusions were:

1. The human and zebrafish 2-adrenoceptors have surprisingly similar ligand 1. The human and zebrafish 2-adrenoceptors have surprisingly similar ligand binding profiles in spite of their long evolutionary distance. Comparisons of the binding profiles in spite of their long evolutionary distance. Comparisons of the structural models of the human and zebrafish 2-adrenoceptors indicate that structural models of the human and zebrafish 2-adrenoceptors indicate that amino acid differences especially in XL2 may contribute to the observed amino acid differences especially in XL2 may contribute to the observed differences in ligand binding affinity. differences in ligand binding affinity.

2. Site-directed mutagenesis of the amino acids at positions 5.43, xl2.49 and xl2.51 2. Site-directed mutagenesis of the amino acids at positions 5.43, xl2.49 and xl2.51 reversed the binding profiles of the human and mouse 2A-adrenoceptors for reversed the binding profiles of the human and mouse 2A-adrenoceptors for yohimbine, rauwolscine and RS-79948-197, pointing to a role of XL2 in the yohimbine, rauwolscine and RS-79948-197, pointing to a role of XL2 in the determination of species-specific ligand binding profiles. determination of species-specific ligand binding profiles.

3. Chimaeric 2-adrenoceptors where TM1 was exchanged between the human 2- 3. Chimaeric 2-adrenoceptors where TM1 was exchanged between the human 2- adrenoceptor subtypes revealed that TM1 is involved in defining the specific adrenoceptor subtypes revealed that TM1 is involved in defining the specific pharmacological profile of the human 2A-adrenoceptor. The binding affinities pharmacological profile of the human 2A-adrenoceptor. The binding affinities of three antagonists (spiperone, spiroxatrine and chlorpromazine) were of three antagonists (spiperone, spiroxatrine and chlorpromazine) were significantly improved by TM1 substitutions in the 2A-adrenoceptor. TM1 may significantly improved by TM1 substitutions in the 2A-adrenoceptor. TM1 may thus influence the pharmacological profiles of the human 2-adrenoceptors, but thus influence the pharmacological profiles of the human 2-adrenoceptors, but the observed effects are believed to be indirect rather than mediated by direct the observed effects are believed to be indirect rather than mediated by direct orthosteric receptor contacts. orthosteric receptor contacts.

72 Acknowledgements 72 Acknowledgements

ACKNOWLEDGEMENTS ACKNOWLEDGEMENTS

This thesis work was carried out in the Department of Pharmacology, Drug This thesis work was carried out in the Department of Pharmacology, Drug Development and Therapeutics, Institute of Biomedicine, University of Turku. I want Development and Therapeutics, Institute of Biomedicine, University of Turku. I want to express my gratitude to the Heads of the Department, Professors Mika Scheinin, to express my gratitude to the Heads of the Department, Professors Mika Scheinin, Risto Huupponen, Markku Koulu and Erkka Syvälahti for the opportunity to work in Risto Huupponen, Markku Koulu and Erkka Syvälahti for the opportunity to work in such an excellent research facility. such an excellent research facility. I owe my deepest gratitude to Professor Mika Scheinin who has been the principal I owe my deepest gratitude to Professor Mika Scheinin who has been the principal supervisor of my thesis work. I will never stop admiring his wisdom, calm and endless supervisor of my thesis work. I will never stop admiring his wisdom, calm and endless interest towards all kinds of knowledge, whether related to scientific or non-scientific interest towards all kinds of knowledge, whether related to scientific or non-scientific issues. Mika has managed to keep me motivated throughout the project by encouraging issues. Mika has managed to keep me motivated throughout the project by encouraging me, especially during the hard times. His patience, optimistic attitude and relatively me, especially during the hard times. His patience, optimistic attitude and relatively blind trust have been the elements that have carried this thesis to completion. I am blind trust have been the elements that have carried this thesis to completion. I am grateful to Mika for many reasons, including the opportunity to visit Professor Marc grateful to Mika for many reasons, including the opportunity to visit Professor Marc Caron’s famous research laboratory in Duke University in the summer of 2002 and to Caron’s famous research laboratory in Duke University in the summer of 2002 and to collaborate with several esteemed colleagues as well as for funding support, and his collaborate with several esteemed colleagues as well as for funding support, and his guidance in science as well as in different teaching tasks. I really feel privileged to guidance in science as well as in different teaching tasks. I really feel privileged to have worked with him and to have learned to know him. have worked with him and to have learned to know him. The Drug Discovery Graduate School (DDGS) and its coordinator Eeva Valve are The Drug Discovery Graduate School (DDGS) and its coordinator Eeva Valve are acknowledged for providing financial support and arranging several travel acknowledged for providing financial support and arranging several travel opportunities as well as the chance to interact with many other young scientists. opportunities as well as the chance to interact with many other young scientists. I was very fortunate to have Professor Mark Johnson and his research group close I was very fortunate to have Professor Mark Johnson and his research group close to me. I thank Mark for his valuable direct contributions to the work and for his role as to me. I thank Mark for his valuable direct contributions to the work and for his role as a member of my supervisory committee. It is a fact that without the Structural a member of my supervisory committee. It is a fact that without the Structural Bioinformatics Laboratory and the facilities and knowledge that Mark and his group Bioinformatics Laboratory and the facilities and knowledge that Mark and his group have provided, this work would never have been completed. have provided, this work would never have been completed. The reviewers of this thesis, Docent Jarmo T. Laitinen and Docent Ulla Petäjä- The reviewers of this thesis, Docent Jarmo T. Laitinen and Docent Ulla Petäjä- Repo, are acknowledged for their valuable comments and constructive criticism. I wish Repo, are acknowledged for their valuable comments and constructive criticism. I wish to thank them for significantly improving the contents of this book. Dr. Ewen to thank them for significantly improving the contents of this book. Dr. Ewen MacDonald is acknowledged for revision of the language of this thesis and for his MacDonald is acknowledged for revision of the language of this thesis and for his encouraging comments. encouraging comments. My special thanks belong to Henri Xhaard and Jori Ruuskanen who have been my closest My special thanks belong to Henri Xhaard and Jori Ruuskanen who have been my closest fellows from the beginning of my thesis studies. Henri, your contribution to all papers fellows from the beginning of my thesis studies. Henri, your contribution to all papers included in this thesis as well as the comments related to all kinds of issues in structural included in this thesis as well as the comments related to all kinds of issues in structural biology have been extremely valuable and didactic. It has been amazing to see how you biology have been extremely valuable and didactic. It has been amazing to see how you zoom into the middle of a receptor and see what’s here and there and what is actually zoom into the middle of a receptor and see what’s here and there and what is actually relevant. Jori, your contribution and guidance in molecular biology and help in the planning relevant. Jori, your contribution and guidance in molecular biology and help in the planning of the binding experiments have been very important, especially when I first joined the of the binding experiments have been very important, especially when I first joined the laboratory. I also want to thank Jori for his good companionship while we were attending laboratory. I also want to thank Jori for his good companionship while we were attending several international meetings as well as for being a member of my thesis committee. several international meetings as well as for being a member of my thesis committee. All my other co-authors, Gloria Wissel, Matias Rantanen, Henry Karlsson, Ville- All my other co-authors, Gloria Wissel, Matias Rantanen, Henry Karlsson, Ville- Veikko Rantanen, Minna Vainio, Anne Marjamäki, Karoliina Vuoriluoto, Siegfried Veikko Rantanen, Minna Vainio, Anne Marjamäki, Karoliina Vuoriluoto, Siegfried Wurster and Tiina Salminen, are acknowledged for their valuable contributions to the Wurster and Tiina Salminen, are acknowledged for their valuable contributions to the

Acknowledgements 73 Acknowledgements 73 articles included in this thesis. Gloria, your input in the last sub-project was essential in articles included in this thesis. Gloria, your input in the last sub-project was essential in getting the paper finally submitted for publication. getting the paper finally submitted for publication. All fellows and members of the 2-group, past and present, are acknowledged. This All fellows and members of the 2-group, past and present, are acknowledged. This huge team with diverse backgrounds has offered a really multi-disciplinary template to huge team with diverse backgrounds has offered a really multi-disciplinary template to create a worldwide collaboration network. It has been great to be able to work with you: create a worldwide collaboration network. It has been great to be able to work with you: Erik Aro, Susann Björk, Mia Engström, Veronica Fagerholm, Heini Frang, Huifang Ge, Erik Aro, Susann Björk, Mia Engström, Veronica Fagerholm, Heini Frang, Huifang Ge, Paula Heinonen, Maya Holmberg, Anna Huhtinen, Maija Ivaska, Jaana Kallio, Eeva Paula Heinonen, Maya Holmberg, Anna Huhtinen, Maija Ivaska, Jaana Kallio, Eeva Kronqvist, Hanna Laine, Ilkka Lähdesmäki, Janne Lähdesmäki, Marja Nykänen, Tuire Kronqvist, Hanna Laine, Ilkka Lähdesmäki, Janne Lähdesmäki, Marja Nykänen, Tuire Olli-Lähdesmäki, Katariina Pohjanoksa, Jussi Posti, Hanna-Mari Raussi, Saku Olli-Lähdesmäki, Katariina Pohjanoksa, Jussi Posti, Hanna-Mari Raussi, Saku Ruohonen, Suvi Ruohonen, Jukka Sallinen, Amir Snapir, Mona Suikkari, Tomi Streng, Ruohonen, Suvi Ruohonen, Jukka Sallinen, Amir Snapir, Mona Suikkari, Tomi Streng, Raisa Vuorinen, Ümit Özdogan, and all others I may not remember. Finally, from the 2- Raisa Vuorinen, Ümit Özdogan, and all others I may not remember. Finally, from the 2- group, I want to express my special thanks to Ms. Ulla Uoti, “mum” of the 2-group, for group, I want to express my special thanks to Ms. Ulla Uoti, “mum” of the 2-group, for all her care, enormous help and skilful technical assistance in the lab. all her care, enormous help and skilful technical assistance in the lab. A big thanks to all the people in Farmis over the years. It is too difficult to A big thanks to all the people in Farmis over the years. It is too difficult to remember you all as a list here, but you know who I mean. I want to express my remember you all as a list here, but you know who I mean. I want to express my special thanks to Ms. Anja Similä, Ms. Hanna Tuominen and Ms. Ulla Hurme for their special thanks to Ms. Anja Similä, Ms. Hanna Tuominen and Ms. Ulla Hurme for their patience and valuable secretarial assistance in office matters related to teaching tasks or patience and valuable secretarial assistance in office matters related to teaching tasks or administrative concerns. It has also been a pleasure to share my office with Sanna administrative concerns. It has also been a pleasure to share my office with Sanna Tikka. Our unofficial talks about life and events outside of the lab and work have been Tikka. Our unofficial talks about life and events outside of the lab and work have been a reason to come to work on many a Monday morning. a reason to come to work on many a Monday morning. I want to thank all members of the Health Biosciences class of ’98 at the University I want to thank all members of the Health Biosciences class of ’98 at the University of Turku: Anna, Anna-Stiina, Anne, Hanna, Jenni B., Jenni V., Jonna, Jukka, of Turku: Anna, Anna-Stiina, Anne, Hanna, Jenni B., Jenni V., Jonna, Jukka, Karoliina, Liisa, Mari, Matias, Niina, Riikka, Saku, Tiina, and Tuire. The spirit of our Karoliina, Liisa, Mari, Matias, Niina, Riikka, Saku, Tiina, and Tuire. The spirit of our class has been indescribable. I am glad that this collegial attitude is still so strongly class has been indescribable. I am glad that this collegial attitude is still so strongly alive after the M.Sc. studies. I wish all of you success in the future. alive after the M.Sc. studies. I wish all of you success in the future. I would also like to thank my best friends outside of the academic world, Toni, Risto I would also like to thank my best friends outside of the academic world, Toni, Risto and Pasi for your indispensable support and patience in listening when I have started to and Pasi for your indispensable support and patience in listening when I have started to talk about things about which you could have no idea, and Antti for accommodating me talk about things about which you could have no idea, and Antti for accommodating me and giving me a home several times during my research or other visits to Helsinki. and giving me a home several times during my research or other visits to Helsinki. Dear mother, Arja Laurila, I want to thank you for the love and constant support Dear mother, Arja Laurila, I want to thank you for the love and constant support that you have given to me during my life. You have always been present as a tower of that you have given to me during my life. You have always been present as a tower of strength when I have had difficulties. Your help in baby-sitting as well as in many strength when I have had difficulties. Your help in baby-sitting as well as in many practical things have made my life much easier in many ways. My siblings, Tatu, Anna practical things have made my life much easier in many ways. My siblings, Tatu, Anna and Lotta, I am happy that we have such close relationships, where all happiness and and Lotta, I am happy that we have such close relationships, where all happiness and sorrows are shared. I want to thank you for your support and help in my life. sorrows are shared. I want to thank you for your support and help in my life. Last, but certainly not least, I am deeply grateful to the most important person in my Last, but certainly not least, I am deeply grateful to the most important person in my life, my lovely daughter Venla. When I look at you, I really understand what is most life, my lovely daughter Venla. When I look at you, I really understand what is most important in life. important in life. This thesis was financially supported by DDGS, the Finnish Cultural Foundation, This thesis was financially supported by DDGS, the Finnish Cultural Foundation, the Finnish Society of Science and Letters, the Turku University Foundation, the the Finnish Society of Science and Letters, the Turku University Foundation, the Orion-Farmos Research Foundation, the Ida Montin Foundation and the Finnish- Orion-Farmos Research Foundation, the Ida Montin Foundation and the Finnish- Norwegian Medical Foundation. Norwegian Medical Foundation.

Turku, April 2011 Turku, April 2011

Jonne Laurila Jonne Laurila

74 References 74 References

REFERENCES REFERENCES

Aantaa R, Jalonen J (2006). Perioperative use of 2- Bokoch MP, Zou Y, Rasmussen SG, Liu CW, Nygaard Aantaa R, Jalonen J (2006). Perioperative use of 2- Bokoch MP, Zou Y, Rasmussen SG, Liu CW, Nygaard adrenoceptor agonists and the cardiac patient. Eur J R, Rosenbaum DM, et al. (2010). Ligand-specific adrenoceptor agonists and the cardiac patient. Eur J R, Rosenbaum DM, et al. (2010). Ligand-specific Anaesthesiol. 23: 361-72. regulation of the extracellular surface of a G-protein- Anaesthesiol. 23: 361-72. regulation of the extracellular surface of a G-protein- Ahlquist RP (1948). A study of the adrenotropic coupled receptor. Nature. 463: 108-12. Ahlquist RP (1948). A study of the adrenotropic coupled receptor. Nature. 463: 108-12. receptors. Am J Physiol. 153: 586-600. Bousquet P, Feldman J and Schwartz J (1984). Central receptors. Am J Physiol. 153: 586-600. Bousquet P, Feldman J and Schwartz J (1984). Central Altenbach C, Greenhalgh DA, Khorana HG and cardiovascular effects of alpha adrenergic drugs: Altenbach C, Greenhalgh DA, Khorana HG and cardiovascular effects of alpha adrenergic drugs: Hubbell WL (1994). A collision gradient method to Differences between catecholamines and Hubbell WL (1994). A collision gradient method to Differences between catecholamines and determine the immersion depth of nitroxides in lipid imidazolines. J Pharmacol Exp Ther. 230: 232-6. determine the immersion depth of nitroxides in lipid imidazolines. J Pharmacol Exp Ther. 230: 232-6. bilayers: Application to spin-labeled mutants of Boyajian CL, Loughlin SE and Leslie FM (1987). bilayers: Application to spin-labeled mutants of Boyajian CL, Loughlin SE and Leslie FM (1987). bacteriorhodopsin. Proc Natl Acad Sci U S A. 91: Anatomical evidence for 2- adrenoceptor bacteriorhodopsin. Proc Natl Acad Sci U S A. 91: Anatomical evidence for 2- adrenoceptor 1667-71. heterogeneity: Differential autoradiographic 1667-71. heterogeneity: Differential autoradiographic 3 3 3 3 Altman JD, Trendelenburg AU, MacMillan L, distributions of [ H]rauwolscine and [ H]idazoxan in Altman JD, Trendelenburg AU, MacMillan L, distributions of [ H]rauwolscine and [ H]idazoxan in Bernstein D, Limbird L, Starke K, et al. (1999). rat brain. J Pharmacol Exp Ther. 241: 1079-91. Bernstein D, Limbird L, Starke K, et al. (1999). rat brain. J Pharmacol Exp Ther. 241: 1079-91. Abnormal regulation of the sympathetic nervous Bradford MM (1976). A rapid and sensitive method for Abnormal regulation of the sympathetic nervous Bradford MM (1976). A rapid and sensitive method for system in 2A-adrenergic receptor knockout mice. the quantitation of microgram quantities of protein system in 2A-adrenergic receptor knockout mice. the quantitation of microgram quantities of protein Mol Pharmacol. 56: 154-61. utilizing the principle of protein-dye binding. Anal Mol Pharmacol. 56: 154-61. utilizing the principle of protein-dye binding. Anal Andersson KE, Stief C (2001). Oral - adrenoceptor Biochem. 7: 248-54. Andersson KE, Stief C (2001). Oral - adrenoceptor Biochem. 7: 248-54. blockade as a treatment of erectile dysfunction. Brasili L, Cassinelli A, Angeli P and Melchiorre C blockade as a treatment of erectile dysfunction. Brasili L, Cassinelli A, Angeli P and Melchiorre C World J Urol. 19: 9-13. (1986). Thiol group may be involved in the World J Urol. 19: 9-13. (1986). Thiol group may be involved in the Angelotti T, Daunt D, Shcherbakova OG, Kobilka B irreversible blockade of presynaptic 2- Angelotti T, Daunt D, Shcherbakova OG, Kobilka B irreversible blockade of presynaptic 2- and Hurt CM (2010). Regulation of G-protein adrenoreceptors by pyrextramine and benextramine and Hurt CM (2010). Regulation of G-protein adrenoreceptors by pyrextramine and benextramine coupled receptor traffic by an evolutionary in the isolated guinea pig ileum. Life Sci. 38: 1633- coupled receptor traffic by an evolutionary in the isolated guinea pig ileum. Life Sci. 38: 1633- conserved hydrophobic signal. Traffic. 11: 560-78. 40. conserved hydrophobic signal. Traffic. 11: 560-78. 40. Arcangeli A, D'Alo C and Gaspari R (2009). Brede M, Nagy G, Philipp M, Sorensen JB, Lohse MJ Arcangeli A, D'Alo C and Gaspari R (2009). Brede M, Nagy G, Philipp M, Sorensen JB, Lohse MJ Dexmedetomidine use in general anaesthesia. Curr and Hein L (2003). Differential control of adrenal Dexmedetomidine use in general anaesthesia. Curr and Hein L (2003). Differential control of adrenal Drug Targets. 10: 687-95. and sympathetic catecholamine release by 2- Drug Targets. 10: 687-95. and sympathetic catecholamine release by 2- adrenoceptor subtypes. Mol Endocrinol. 17: 1640-6. adrenoceptor subtypes. Mol Endocrinol. 17: 1640-6. Arima J, Kubo C, Ishibashi H and Akaike N (1998). Arima J, Kubo C, Ishibashi H and Akaike N (1998). Bultmann R, Starke K (1993). Chloroethylclonidine: Bultmann R, Starke K (1993). Chloroethylclonidine: 2-Adrenoceptor-mediated potassium currents in 2-Adrenoceptor-mediated potassium currents in acutely dissociated rat locus coeruleus neurones. J An irreversible agonist at prejunctional2- acutely dissociated rat locus coeruleus neurones. J An irreversible agonist at prejunctional2- Physiol. 508: 57-66. adrenoceptors in rat vas deferens. Br J Pharmacol. Physiol. 508: 57-66. adrenoceptors in rat vas deferens. Br J Pharmacol. 108: 336-41. 108: 336-41. Attwood TK, Findlay JB (1994). Fingerprinting G- Attwood TK, Findlay JB (1994). Fingerprinting G- protein-coupled receptors. Protein Eng. 7: 195-203. Bylund DB (2005). Alpha-2 adrenoceptor subtypes: protein-coupled receptors. Protein Eng. 7: 195-203. Bylund DB (2005). Alpha-2 adrenoceptor subtypes: Are more better? Br J Pharmacol. 144: 159-60. Are more better? Br J Pharmacol. 144: 159-60. Ballesteros JA, Weinstein H (1995). Integrated Ballesteros JA, Weinstein H (1995). Integrated methods for the construction of three-dimensional Bylund DB, Blaxall HS, Iversen LJ, Caron MG, methods for the construction of three-dimensional Bylund DB, Blaxall HS, Iversen LJ, Caron MG, models and computational probing of structure- Lefkowitz RJ and Lomasney JW (1992). models and computational probing of structure- Lefkowitz RJ and Lomasney JW (1992). function relations in G protein-coupled receptors: In: Pharmacological characteristics of 2-adrenergic function relations in G protein-coupled receptors: In: Pharmacological characteristics of 2-adrenergic Sealfon SC (ed). Receptor Molecular Biology. receptors: Comparison of pharmacologically defined Sealfon SC (ed). Receptor Molecular Biology. receptors: Comparison of pharmacologically defined Academic Press Inc.: San Diego, pp 366-427. subtypes with subtypes identified by molecular Academic Press Inc.: San Diego, pp 366-427. subtypes with subtypes identified by molecular cloning. Mol Pharmacol. 42: 1-5. cloning. Mol Pharmacol. 42: 1-5. Balogh B, Szilagyi A, Gyires K, Bylund DB and Balogh B, Szilagyi A, Gyires K, Bylund DB and Matyus P (2009). Molecular modelling of subtypes Bylund DB, Rudeen PK, Petterborg LJ and Ray- Matyus P (2009). Molecular modelling of subtypes Bylund DB, Rudeen PK, Petterborg LJ and Ray- Prenger C (1988). Identification of 2-adrenergic Prenger C (1988). Identification of 2-adrenergic (2A, 2B and 2C) of 2-adrenoceptors: A (2A, 2B and 2C) of 2-adrenoceptors: A comparative study. Neurochem Int. 55: 355-61. receptors in chicken pineal gland using comparative study. Neurochem Int. 55: 355-61. receptors in chicken pineal gland using [3H]rauwolscine. J Neurochem. 51: 81-6. [3H]rauwolscine. J Neurochem. 51: 81-6. Bes A, Eyssette M, Pierrot-Deseilligny E, Rohmer F Bes A, Eyssette M, Pierrot-Deseilligny E, Rohmer F and Warter JM (1988). A multi-centre, double-blind Bylund DB, Toews ML (1993). Radioligand binding and Warter JM (1988). A multi-centre, double-blind Bylund DB, Toews ML (1993). Radioligand binding trial of tizanidine, a new antispastic agent, in methods: Practical guide and tips. Am J Physiol. trial of tizanidine, a new antispastic agent, in methods: Practical guide and tips. Am J Physiol. spasticity associated with hemiplegia. Curr Med Res 265: 421-9. spasticity associated with hemiplegia. Curr Med Res 265: 421-9. Opin. 10: 709-18. Ceresa BP, Limbird LE (1994). Mutation of an Opin. 10: 709-18. Ceresa BP, Limbird LE (1994). Mutation of an Birnbaumer L (1990). G proteins in signal aspartate residue highly conserved among G-protein- Birnbaumer L (1990). G proteins in signal aspartate residue highly conserved among G-protein- transduction. Annu Rev Pharmacol Toxicol. 30: 675- coupled receptors results in nonreciprocal disruption transduction. Annu Rev Pharmacol Toxicol. 30: 675- coupled receptors results in nonreciprocal disruption 705. of 2-adrenergic receptor-G-protein interactions. A 705. of 2-adrenergic receptor-G-protein interactions. A negative charge at amino acid residue 79 forecasts negative charge at amino acid residue 79 forecasts Bockaert J, Pin JP (1999). Molecular tinkering of G 2A-adrenergic receptor sensitivity to allosteric Bockaert J, Pin JP (1999). Molecular tinkering of G 2A-adrenergic receptor sensitivity to allosteric protein-coupled receptors: An evolutionary success. modulation by monovalent cations and fully protein-coupled receptors: An evolutionary success. modulation by monovalent cations and fully EMBO J. 18: 1723-9. EMBO J. 18: 1723-9.

References 75 References 75

3 3 3 3 effective receptor/G-protein coupling. J Biol Chem. [ H]rauwolscine and [ H]RX821002 to the 2- effective receptor/G-protein coupling. J Biol Chem. [ H]rauwolscine and [ H]RX821002 to the 2- 269: 29557-64. adrenergic receptor subtypes. J Pharmacol Exp 269: 29557-64. adrenergic receptor subtypes. J Pharmacol Exp Cerione RA, Regan JW, Nakata H, Codina J, Benovic Ther. 278: 1215-27. Cerione RA, Regan JW, Nakata H, Codina J, Benovic Ther. 278: 1215-27. JL, Gierschik P, et al. (1986). Functional Devedjian JC, Esclapez F, Denis-Pouxviel C and Paris JL, Gierschik P, et al. (1986). Functional Devedjian JC, Esclapez F, Denis-Pouxviel C and Paris reconstitution of the 2-adrenergic receptor with H (1994). Further characterization of human 2- reconstitution of the 2-adrenergic receptor with H (1994). Further characterization of human 2- guanine nucleotide regulatory proteins in adrenoceptor subtypes: [3H]RX821002 binding and guanine nucleotide regulatory proteins in adrenoceptor subtypes: [3H]RX821002 binding and phospholipid vesicles. J Biol Chem. 261: 3901-9. definition of additional selective drugs. Eur J phospholipid vesicles. J Biol Chem. 261: 3901-9. definition of additional selective drugs. Eur J Chen S, Lin F, Xu M and Graham RM (2002). Pharmacol. 252: 43-9. Chen S, Lin F, Xu M and Graham RM (2002). Pharmacol. 252: 43-9. Phe(303) in TMVI of the 1B-adrenergic receptor is Dixon RA, Kobilka BK, Strader DJ, Benovic JL, Phe(303) in TMVI of the 1B-adrenergic receptor is Dixon RA, Kobilka BK, Strader DJ, Benovic JL, a key residue coupling TM helical movements to G- Dohlman HG, Frielle T, et al. (1986). Cloning of the a key residue coupling TM helical movements to G- Dohlman HG, Frielle T, et al. (1986). Cloning of the protein activation. Biochemistry. 41: 588-96. gene and cDNA for mammalian -adrenergic protein activation. Biochemistry. 41: 588-96. gene and cDNA for mammalian -adrenergic Cheng Y, Prusoff WH (1973). Relationship between receptor and homology with rhodopsin. Nature. 321: Cheng Y, Prusoff WH (1973). Relationship between receptor and homology with rhodopsin. Nature. 321: the inhibition constant (K1) and the concentration of 75-9. the inhibition constant (K1) and the concentration of 75-9. inhibitor which causes 50 per cent inhibition (I50) of Dorn GW,2nd, Oswald KJ, McCluskey TS, Kuhel DG inhibitor which causes 50 per cent inhibition (I50) of Dorn GW,2nd, Oswald KJ, McCluskey TS, Kuhel DG an enzymatic reaction. Biochem Pharmacol. 22: and Liggett SB (1997). 2A-Adrenergic receptor an enzymatic reaction. Biochem Pharmacol. 22: and Liggett SB (1997). 2A-Adrenergic receptor 3099-108. stimulated calcium release is transduced by Gi- 3099-108. stimulated calcium release is transduced by Gi- Cherezov V, Rosenbaum DM, Hanson MA, associated G-mediated activation of phospholipase Cherezov V, Rosenbaum DM, Hanson MA, associated G-mediated activation of phospholipase Rasmussen SG, Thian FS, Kobilka TS, et al. (2007). C. Biochemistry. 36: 6415-23. Rasmussen SG, Thian FS, Kobilka TS, et al. (2007). C. Biochemistry. 36: 6415-23. High-resolution crystal structure of an engineered Dossin O, Mouledous L, Baudry X, Tafani JA, High-resolution crystal structure of an engineered Dossin O, Mouledous L, Baudry X, Tafani JA, human 2-adrenergic G protein-coupled receptor. Mazarguil H and Zajac JM (2000). Characterization human 2-adrenergic G protein-coupled receptor. Mazarguil H and Zajac JM (2000). Characterization Science. 318: 1258-65. of a new radioiodinated probe for the 2C- Science. 318: 1258-65. of a new radioiodinated probe for the 2C- Chien EY, Liu W, Zhao Q, Katritch V, Han GW, adrenoceptor in the mouse brain. Neurochem Int. 36: Chien EY, Liu W, Zhao Q, Katritch V, Han GW, adrenoceptor in the mouse brain. Neurochem Int. 36: Hanson MA, et al. (2010). Structure of the human 7-18. Hanson MA, et al. (2010). Structure of the human 7-18. dopamine D3 receptor in complex with a D2/D3 Eason MG, Jacinto MT, Theiss CT and Liggett SB dopamine D3 receptor in complex with a D2/D3 Eason MG, Jacinto MT, Theiss CT and Liggett SB selective antagonist. Science. 330: 1091-5. (1994). The palmitoylated cysteine of the selective antagonist. Science. 330: 1091-5. (1994). The palmitoylated cysteine of the Chotani MA, Mitra S, Su BY, Flavahan S, Eid AH, cytoplasmic tail of 2A-adrenergic receptors confers Chotani MA, Mitra S, Su BY, Flavahan S, Eid AH, cytoplasmic tail of 2A-adrenergic receptors confers subtype-specific agonist-promoted downregulation. subtype-specific agonist-promoted downregulation. Clark KR, et al. (2004). Regulation of 2- Clark KR, et al. (2004). Regulation of 2- adrenoceptors in human vascular smooth muscle Proc Natl Acad Sci U S A. 91: 11178-82. adrenoceptors in human vascular smooth muscle Proc Natl Acad Sci U S A. 91: 11178-82. cells. Am J Physiol Heart Circ Physiol. 286: H59- Eason MG, Kurose H, Holt BD, Raymond JR and cells. Am J Physiol Heart Circ Physiol. 286: H59- Eason MG, Kurose H, Holt BD, Raymond JR and 67. Liggett SB (1992). Simultaneous coupling of 2- 67. Liggett SB (1992). Simultaneous coupling of 2- Cockcroft V, Frang H, Pihlavisto M, Marjamäki A and adrenergic receptors to two G-proteins with Cockcroft V, Frang H, Pihlavisto M, Marjamäki A and adrenergic receptors to two G-proteins with Scheinin M (2000). Ligand recognition of serine- opposing effects. subtype-selective coupling of Scheinin M (2000). Ligand recognition of serine- opposing effects. subtype-selective coupling of cysteine amino acid exchanges in transmembrane 2C10, 2C4, and 2C2 adrenergic receptors to Gi cysteine amino acid exchanges in transmembrane 2C10, 2C4, and 2C2 adrenergic receptors to Gi and Gs. J Biol Chem. 267: 15795-801. and Gs. J Biol Chem. 267: 15795-801. domain 5 of 2-adrenergic receptors by UK 14,304. domain 5 of 2-adrenergic receptors by UK 14,304. J Neurochem. 74: 1705-10. Eason MG, Liggett SB (1995). Identification of a Gs J Neurochem. 74: 1705-10. Eason MG, Liggett SB (1995). Identification of a Gs Cotecchia S, Kobilka BK, Daniel KW, Nolan RD, coupling domain in the amino terminus of the third Cotecchia S, Kobilka BK, Daniel KW, Nolan RD, coupling domain in the amino terminus of the third Lapetina EY, Caron MG, et al. (1990). Multiple intracellular loop of the 2A-adrenergic receptor. Lapetina EY, Caron MG, et al. (1990). Multiple intracellular loop of the 2A-adrenergic receptor. second messenger pathways of -adrenergic evidence for distinct structural determinants that second messenger pathways of -adrenergic evidence for distinct structural determinants that receptor subtypes expressed in eukaryotic cells. J confer Gs versus Gi coupling. J Biol Chem. 270: receptor subtypes expressed in eukaryotic cells. J confer Gs versus Gi coupling. J Biol Chem. 270: Biol Chem. 265: 63-9. 24753-60. Biol Chem. 265: 63-9. 24753-60. Dardonville C, Rozas I (2004). Imidazoline binding Eyssette M, Rohmer F, Serratrice G, Warter JM and Dardonville C, Rozas I (2004). Imidazoline binding Eyssette M, Rohmer F, Serratrice G, Warter JM and sites and their ligands: An overview of the different Boisson D (1988). Multi-centre, double-blind trial of sites and their ligands: An overview of the different Boisson D (1988). Multi-centre, double-blind trial of chemical structures. Med Res Rev. 24: 639-61. a novel antispastic agent, tizanidine, in spasticity chemical structures. Med Res Rev. 24: 639-61. a novel antispastic agent, tizanidine, in spasticity associated with multiple sclerosis. Curr Med Res associated with multiple sclerosis. Curr Med Res Daunt DA, Hurt C, Hein L, Kallio J, Feng F and Opin. 10: 699-708. Daunt DA, Hurt C, Hein L, Kallio J, Feng F and Opin. 10: 699-708. Kobilka BK (1997). Subtype-specific intracellular Kobilka BK (1997). Subtype-specific intracellular Fagerholm V, Grönroos T, Marjamäki P, Viljanen T, Fagerholm V, Grönroos T, Marjamäki P, Viljanen T, trafficking of 2-adrenergic receptors. Mol trafficking of 2-adrenergic receptors. Mol Pharmacol. 51: 711-20. Scheinin M and Haaparanta M (2004). Altered Pharmacol. 51: 711-20. Scheinin M and Haaparanta M (2004). Altered glucose homeostasis in 2A-adrenoceptor knockout glucose homeostasis in 2A-adrenoceptor knockout Delmas P, Abogadie FC, Milligan G, Buckley NJ and mice. Eur J Pharmacol. 505: 243-52. Delmas P, Abogadie FC, Milligan G, Buckley NJ and mice. Eur J Pharmacol. 505: 243-52. Brown DA (1999). Betagamma dimers derived from Brown DA (1999). Betagamma dimers derived from Ferrè S, Navarro G, Casadò V, Cortès A, Mallol J, Ferrè S, Navarro G, Casadò V, Cortès A, Mallol J, Go and Gi proteins contribute different components Go and Gi proteins contribute different components of adrenergic inhibition of Ca2+ channels in rat Canela EI et al. (2010). G protein-coupled receptor of adrenergic inhibition of Ca2+ channels in rat Canela EI et al. (2010). G protein-coupled receptor sympathetic neurones. J Physiol. 518: 23-36. heteromers as new targets for drug development. sympathetic neurones. J Physiol. 518: 23-36. heteromers as new targets for drug development. Prog Mol Biol Transl Sci. 91: 41-52. Prog Mol Biol Transl Sci. 91: 41-52. Deslauriers B, Ponce C, Lombard C, Larguier R, Deslauriers B, Ponce C, Lombard C, Larguier R, Bonnafous JC and Marie J (1999). N-glycosylation Fox SL, Samson CR and Danzig MR (1979). Bonnafous JC and Marie J (1999). N-glycosylation Fox SL, Samson CR and Danzig MR (1979). Oxymetazoline in the treatment of allergic and non- Oxymetazoline in the treatment of allergic and non- requirements for the AT1a angiotensin II receptor requirements for the AT1a angiotensin II receptor delivery to the plasma membrane. Biochem J. 339: infectious conjunctivitis. J Int Med Res. 7: 528-30. delivery to the plasma membrane. Biochem J. 339: infectious conjunctivitis. J Int Med Res. 7: 528-30. 397-405. Frang H, Cockcroft V, Karskela T, Scheinin M and 397-405. Frang H, Cockcroft V, Karskela T, Scheinin M and Deupree JD, Hinton KA, Cerutis DR and Bylund DB Marjamäki A (2001). Phenoxybenzamine binding Deupree JD, Hinton KA, Cerutis DR and Bylund DB Marjamäki A (2001). Phenoxybenzamine binding (1996). Buffers differentially alter the binding of reveals the helical orientation of the third (1996). Buffers differentially alter the binding of reveals the helical orientation of the third

76 References 76 References transmembrane domain of adrenergic receptors. J receptor that is associated with reduced basal transmembrane domain of adrenergic receptors. J receptor that is associated with reduced basal Biol Chem. 276: 31279-84. metabolic rate in obese subjects. J Clin Endocrinol Biol Chem. 276: 31279-84. metabolic rate in obese subjects. J Clin Endocrinol Fredriksson R, Lagerstrom MC, Lundin LG and Metab. 84: 2429-33. Fredriksson R, Lagerstrom MC, Lundin LG and Metab. 84: 2429-33. Schiöth HB (2003). The G-protein-coupled receptors Henderson R, Baldwin JM, Ceska TA, Zemlin F, Schiöth HB (2003). The G-protein-coupled receptors Henderson R, Baldwin JM, Ceska TA, Zemlin F, in the human genome form five main families. Beckmann E and Downing KH (1990). Model for in the human genome form five main families. Beckmann E and Downing KH (1990). Model for Phylogenetic analysis, paralogon groups, and the structure of bacteriorhodopsin based on high- Phylogenetic analysis, paralogon groups, and the structure of bacteriorhodopsin based on high- fingerprints. Mol Pharmacol. 63: 1256-72. resolution electron cryo-microscopy. J Mol Biol. fingerprints. Mol Pharmacol. 63: 1256-72. resolution electron cryo-microscopy. J Mol Biol. Freissmuth M, Waldhoer M, Bofill-Cardona E and 213: 899-929. Freissmuth M, Waldhoer M, Bofill-Cardona E and 213: 899-929. Nanoff C (1999). G protein antagonists. Trends Henderson R and Schertler GF (1990). The structure of Nanoff C (1999). G protein antagonists. Trends Henderson R and Schertler GF (1990). The structure of Pharmacol Sci. 20: 237-45. bacteriorhodopsin and its relevance to the visual Pharmacol Sci. 20: 237-45. bacteriorhodopsin and its relevance to the visual Garvey HL, Ling GM and Melville KI (1972). opsins and other seven-helix G-protein coupled Garvey HL, Ling GM and Melville KI (1972). opsins and other seven-helix G-protein coupled Cardiovascular actions of N-ethoxycarbonyl-2- receptors. Philos Trans R Soc Lond B Biol Sci, 326: Cardiovascular actions of N-ethoxycarbonyl-2- receptors. Philos Trans R Soc Lond B Biol Sci, 326: ethoxy-1,2-dihydroquinoline (EEDQ) a new alpha- 379-89. ethoxy-1,2-dihydroquinoline (EEDQ) a new alpha- 379-89. adrenergic receptor antagonist. Eur J Pharmacol. Hengstmann JH, Gugler R and Dengler HJ (1979). adrenergic receptor antagonist. Eur J Pharmacol. Hengstmann JH, Gugler R and Dengler HJ (1979). 18: 148-58. Malignant pheochromocytoma. effect of oral alpha- 18: 148-58. Malignant pheochromocytoma. effect of oral alpha- Ge H, Scheinin M and Kallio J (2003). Constitutive methyl-p-tyrosine upon catecholamine metabolism. Ge H, Scheinin M and Kallio J (2003). Constitutive methyl-p-tyrosine upon catecholamine metabolism. Klin Wochenschr. 57: 351-5. Klin Wochenschr. 57: 351-5. precoupling to Gi and increased agonist potency in precoupling to Gi and increased agonist potency in the2B-adrenoceptor. Biochem Biophys Res Hieble JP, Hehr A, Li YO and Ruffolo RR Jr (1998). the2B-adrenoceptor. Biochem Biophys Res Hieble JP, Hehr A, Li YO and Ruffolo RR Jr (1998). Commun. 306: 959-65. Molecular basis for the stereoselective interactions Commun. 306: 959-65. Molecular basis for the stereoselective interactions Gentili F, Ghelfi F, Giannella M, Piergentili A, Pigini of catecholamines with alpha-adrenoceptors. Proc Gentili F, Ghelfi F, Giannella M, Piergentili A, Pigini of catecholamines with alpha-adrenoceptors. Proc West Pharmacol Soc. 41: 225-8. West Pharmacol Soc. 41: 225-8. M, Quaglia W, et al. (2004). 2-Adrenoreceptors M, Quaglia W, et al. (2004). 2-Adrenoreceptors profile modulation. 2. Biphenyline analogues as Hildebrand PW, Scheerer P, Park JH, Choe HW, profile modulation. 2. Biphenyline analogues as Hildebrand PW, Scheerer P, Park JH, Choe HW, tools for selective activation of the 2C-subtype. J Piechnick R, Ernst OP, et al. (2009). A ligand tools for selective activation of the 2C-subtype. J Piechnick R, Ernst OP, et al. (2009). A ligand Med Chem. 47: 6160-73. channel through the G protein coupled receptor Med Chem. 47: 6160-73. channel through the G protein coupled receptor Golling G, Amsterdam A, Sun Z, Antonelli M, opsin. PLoS One. 4: e4382. Golling G, Amsterdam A, Sun Z, Antonelli M, opsin. PLoS One. 4: e4382. Maldonado E, Chen W, et al. (2002). Insertional Ho BY, Karschin A, Branchek T, Davidson N and Maldonado E, Chen W, et al. (2002). Insertional Ho BY, Karschin A, Branchek T, Davidson N and mutagenesis in zebrafish rapidly identifies genes Lester HA (1992). The role of conserved aspartate mutagenesis in zebrafish rapidly identifies genes Lester HA (1992). The role of conserved aspartate essential for early vertebrate development. Nat and serine residues in ligand binding and in function essential for early vertebrate development. Nat and serine residues in ligand binding and in function Genet. 31: 135-40. of the 5-HT1A receptor: A site-directed mutation Genet. 31: 135-40. of the 5-HT1A receptor: A site-directed mutation Gongadze NV, Antelava NA and Kezeli TD (2008). study. FEBS Lett. 312: 259-62. Gongadze NV, Antelava NA and Kezeli TD (2008). study. FEBS Lett. 312: 259-62. Imidazoline receptors. Georgian Med News. (160- Holmberg M, Fagerholm V and Scheinin M (2003). Imidazoline receptors. Georgian Med News. (160- Holmberg M, Fagerholm V and Scheinin M (2003). 161): 44-7. Regional distribution of 2C-adrenoceptors in brain 161): 44-7. Regional distribution of 2C-adrenoceptors in brain Halme M, Sjöholm B, Savola JM and Scheinin M and spinal cord of control mice and transgenic mice Halme M, Sjöholm B, Savola JM and Scheinin M and spinal cord of control mice and transgenic mice overexpressing the 2C-subtype: An overexpressing the 2C-subtype: An (1995). Recombinant human 2-adrenoceptor 3 (1995). Recombinant human 2-adrenoceptor 3 3 autoradiographic study with [ H]RX821002 and 3 autoradiographic study with [ H]RX821002 and subtypes: Comparison of [ H]rauwolscine, 3 subtypes: Comparison of [ H]rauwolscine, 3 [3H]atipamezole and [3H]RX821002 as radioligands. [ H]rauwolscine. Neuroscience. 117: 875-98. [3H]atipamezole and [3H]RX821002 as radioligands. [ H]rauwolscine. Neuroscience. 117: 875-98. Biochim Biophys Acta. 1266: 207-14. Hunter JC, Fontana DJ, Hedley LR, Jasper JR, Lewis Biochim Biophys Acta. 1266: 207-14. Hunter JC, Fontana DJ, Hedley LR, Jasper JR, Lewis Hanson MA, Cherezov V, Griffith MT, Roth CB, R, Link RE, et al. (1997). Assessment of the role of Hanson MA, Cherezov V, Griffith MT, Roth CB, R, Link RE, et al. (1997). Assessment of the role of Jaakola VP, Chien EY, et al. (2008). A specific 2-adrenoceptor subtypes in the antinociceptive, Jaakola VP, Chien EY, et al. (2008). A specific 2-adrenoceptor subtypes in the antinociceptive, cholesterol binding site is established by the 2.8 Å sedative and hypothermic action of cholesterol binding site is established by the 2.8 Å sedative and hypothermic action of dexmedetomidine in transgenic mice. Br J dexmedetomidine in transgenic mice. Br J structure of the human 2-adrenergic receptor. structure of the human 2-adrenergic receptor. Structure. 16: 897-905. Pharmacol. 122: 1339-44. Structure. 16: 897-905. Pharmacol. 122: 1339-44. Happe HK, Coulter CL, Gerety ME, Sanders JD, Jaakola VP, Griffith MT, Hanson MA, Cherezov V, Happe HK, Coulter CL, Gerety ME, Sanders JD, Jaakola VP, Griffith MT, Hanson MA, Cherezov V, Chien EY, Lane JR, et al. (2008). The 2.6 angstrom Chien EY, Lane JR, et al. (2008). The 2.6 angstrom O'Rourke M, Bylund DB, et al. (2004). 2- O'Rourke M, Bylund DB, et al. (2004). 2- Adrenergic receptor development in rat CNS: An crystal structure of a human A2A adenosine receptor Adrenergic receptor development in rat CNS: An crystal structure of a human A2A adenosine receptor autoradiographic study. Neuroscience. 123: 167-78. bound to an antagonist. Science. 322: 1211-7. autoradiographic study. Neuroscience. 123: 167-78. bound to an antagonist. Science. 322: 1211-7. Hehr A, Hieble JP, Li YO, Bergsma DJ, Swift AM, Jaakola VP, Vainio M, Sen S, Rehn M, Heimo H, Hehr A, Hieble JP, Li YO, Bergsma DJ, Swift AM, Jaakola VP, Vainio M, Sen S, Rehn M, Heimo H, Ganguly S, et al. (1997). Ser165 of transmembrane Scheinin M, et al. (2005). Intracellularly truncated Ganguly S, et al. (1997). Ser165 of transmembrane Scheinin M, et al. (2005). Intracellularly truncated helix IV is not involved in the interaction of human 2B-adrenoceptors: Stable and functional helix IV is not involved in the interaction of human 2B-adrenoceptors: Stable and functional GPCRs for structural studies. J Recept Signal GPCRs for structural studies. J Recept Signal catecholamines with the 2A-adrenoceptor. catecholamines with the 2A-adrenoceptor. Pharmacology. 55: 18-24. Transduct Res. 25: 99-124. Pharmacology. 55: 18-24. Transduct Res. 25: 99-124. Hein L, Limbird LE, Eglen RM and Kobilka BK Janumpalli S, Butler LS, MacMillan LB, Limbird LE Hein L, Limbird LE, Eglen RM and Kobilka BK Janumpalli S, Butler LS, MacMillan LB, Limbird LE (1999). Gene substitution/knockout to delineate the and McNamara JO (1998). A point mutation (D79N) (1999). Gene substitution/knockout to delineate the and McNamara JO (1998). A point mutation (D79N) of the 2A-adrenergic receptor abolishes the of the 2A-adrenergic receptor abolishes the role of 2-adrenoceptor subtypes in mediating role of 2-adrenoceptor subtypes in mediating central effects of catecholamines and imidazolines. antiepileptogenic action of endogenous central effects of catecholamines and imidazolines. antiepileptogenic action of endogenous Ann N Y Acad Sci. 881: 265-71. norepinephrine. J Neurosci. 18: 2004-8. Ann N Y Acad Sci. 881: 265-71. norepinephrine. J Neurosci. 18: 2004-8. Heinonen P, Koulu M, Pesonen U, Karvonen MK, Jasper JR, Lesnick JD, Chang LK, Yamanishi SS, Heinonen P, Koulu M, Pesonen U, Karvonen MK, Jasper JR, Lesnick JD, Chang LK, Yamanishi SS, Rissanen A, Laakso M, et al. (1999). Identification Chang TK, Hsu SA, et al. (1998). Ligand efficacy Rissanen A, Laakso M, et al. (1999). Identification Chang TK, Hsu SA, et al. (1998). Ligand efficacy and potency at recombinant 2-adrenergic receptors: and potency at recombinant 2-adrenergic receptors: of a three-amino acid deletion in the 2B-adrenergic of a three-amino acid deletion in the 2B-adrenergic

References 77 References 77

Agonist-mediated [35S]GTPS binding. Biochem Kohli U, Muszkat M, Sofowora GG, Harris PA, Agonist-mediated [35S]GTPS binding. Biochem Kohli U, Muszkat M, Sofowora GG, Harris PA, Pharmacol. 55: 1035-43. Friedman EA, Dupont WD, et al. (2010). Effects of Pharmacol. 55: 1035-43. Friedman EA, Dupont WD, et al. (2010). Effects of Javitch JA, Fu D, Liapakis G and Chen J (1997). variation in the human 2A- and 2C-adrenoceptor Javitch JA, Fu D, Liapakis G and Chen J (1997). variation in the human 2A- and 2C-adrenoceptor Constitutive activation of the beta2 adrenergic genes on cognitive tasks and pain perception. Eur J Constitutive activation of the beta2 adrenergic genes on cognitive tasks and pain perception. Eur J receptor alters the orientation of its sixth membrane- Pain. 14: 154-9. receptor alters the orientation of its sixth membrane- Pain. 14: 154-9. spanning segment. J Biol Chem. 272: 18546-9. Kolaj M, Renaud LP (2001). Norepinephrine acts via spanning segment. J Biol Chem. 272: 18546-9. Kolaj M, Renaud LP (2001). Norepinephrine acts via Jensen AD, Gether U (2000). Assessing adrenergic 2-adrenergic receptors to suppress N-type calcium Jensen AD, Gether U (2000). Assessing adrenergic 2-adrenergic receptors to suppress N-type calcium receptor conformation using chemically reactive channels in dissociated rat median preoptic nucleus receptor conformation using chemically reactive channels in dissociated rat median preoptic nucleus fluorescent probes. Methods Mol Biol. 126: 345-61. neurons. Neuropharmacology. 41: 472-9. fluorescent probes. Methods Mol Biol. 126: 345-61. neurons. Neuropharmacology. 41: 472-9. Kolakowski LF,Jr (1994). GCRDb: A G-protein- Kolakowski LF,Jr (1994). GCRDb: A G-protein- Johnson M (2001). 2-Adrenoceptors: Mechanisms of Johnson M (2001). 2-Adrenoceptors: Mechanisms of action of  -agonists. Paediatr Respir Rev. 2: 57-62. coupled receptor database. Receptors Channels. 2: action of  -agonists. Paediatr Respir Rev. 2: 57-62. coupled receptor database. Receptors Channels. 2: 2 1-7. 2 1-7. Johnson MS, Overington JP (1993). A structural basis Johnson MS, Overington JP (1993). A structural basis for sequence comparisons. an evaluation of scoring Kolb P, Rosenbaum DM, Irwin JJ, Fung JJ, Kobilka for sequence comparisons. an evaluation of scoring Kolb P, Rosenbaum DM, Irwin JJ, Fung JJ, Kobilka methodologies. J Mol Biol. 233: 716-38. BK and Shoichet BK (2009). Structure-based methodologies. J Mol Biol. 233: 716-38. BK and Shoichet BK (2009). Structure-based discovery of 2-adrenergic receptor ligands. Proc discovery of 2-adrenergic receptor ligands. Proc Jones AM, Assmann SM (2004). Plants: The latest Natl Acad Sci U S A. 106: 6843-8. Jones AM, Assmann SM (2004). Plants: The latest Natl Acad Sci U S A. 106: 6843-8. model system for G-protein research. EMBO Rep. 5: model system for G-protein research. EMBO Rep. 5: 572-8. Kristiansen K (2004). Molecular mechanisms of ligand 572-8. Kristiansen K (2004). Molecular mechanisms of ligand binding, signaling, and regulation within the binding, signaling, and regulation within the Jones G, Willett P, Glen RC, Leach AR and Taylor R superfamily of G-protein-coupled receptors: Jones G, Willett P, Glen RC, Leach AR and Taylor R superfamily of G-protein-coupled receptors: (1997). Development and validation of a genetic Molecular modeling and mutagenesis approaches to (1997). Development and validation of a genetic Molecular modeling and mutagenesis approaches to algorithm for flexible docking. J Mol Biol. 267: 727- receptor structure and function. Pharmacol Ther. algorithm for flexible docking. J Mol Biol. 267: 727- receptor structure and function. Pharmacol Ther. 48. 103: 21-80. 48. 103: 21-80.

Jones SB, Halenda SP and Bylund DB (1991). 2- Kukkonen JP, Renvaktar A, Shariatmadari R and Jones SB, Halenda SP and Bylund DB (1991). 2- Kukkonen JP, Renvaktar A, Shariatmadari R and Adrenergic receptor stimulation of phospholipase A2 Akerman KE (1998). Ligand- and subtype-selective Adrenergic receptor stimulation of phospholipase A2 Akerman KE (1998). Ligand- and subtype-selective ++ ++ and of adenylate cyclase in transfected chinese coupling of human 2-adrenoceptors to Ca and of adenylate cyclase in transfected chinese coupling of human 2-adrenoceptors to Ca hamster ovary cells is mediated by different elevation in chinese hamster ovary cells. J hamster ovary cells is mediated by different elevation in chinese hamster ovary cells. J mechanisms. Mol Pharmacol. 39: 239-45. Pharmacol Exp Ther. 287: 667-71. mechanisms. Mol Pharmacol. 39: 239-45. Pharmacol Exp Ther. 287: 667-71. Jordan BA, Gomes I, Rios C, Filipovska J and Devi Kurose H, Arriza JL and Lefkowitz RJ (1993). Jordan BA, Gomes I, Rios C, Filipovska J and Devi Kurose H, Arriza JL and Lefkowitz RJ (1993). LA (2003). Functional interactions between  opioid Characterization of 2-adrenergic receptor subtype- LA (2003). Functional interactions between  opioid Characterization of 2-adrenergic receptor subtype- and 2A-adrenergic receptors. Mol Pharmacol. 64: specific antibodies. Mol Pharmacol. 43: 444-50. and 2A-adrenergic receptors. Mol Pharmacol. 64: specific antibodies. Mol Pharmacol. 43: 444-50. 1317-24. 1317-24. Lagerström MC, Schiöth HB (2008). Structural Lagerström MC, Schiöth HB (2008). Structural Kable JW, Murrin LC and Bylund DB (2000). In vivo diversity of G protein-coupled receptors and Kable JW, Murrin LC and Bylund DB (2000). In vivo diversity of G protein-coupled receptors and gene modification elucidates subtype-specific significance for drug discovery. Nat Rev Drug gene modification elucidates subtype-specific significance for drug discovery. Nat Rev Drug functions of 2-adrenergic receptors. J Pharmacol Discov. 7: 339-57. functions of 2-adrenergic receptors. J Pharmacol Discov. 7: 339-57. Exp Ther. 293: 1-7. Exp Ther. 293: 1-7. Lähdesmäki J, Sallinen J, MacDonald E, Sirviö J and Lähdesmäki J, Sallinen J, MacDonald E, Sirviö J and Keefer JR, Kennedy ME and Limbird LE (1994). Scheinin M (2003). 2-Adrenergic drug effects on Keefer JR, Kennedy ME and Limbird LE (1994). Scheinin M (2003). 2-Adrenergic drug effects on Unique structural features important for stabilization brain monoamines, locomotion, and body Unique structural features important for stabilization brain monoamines, locomotion, and body versus polarization of the 2A-adrenergic receptor on temperature are largely abolished in mice lacking the versus polarization of the 2A-adrenergic receptor on temperature are largely abolished in mice lacking the the basolateral membrane of Madin-Darby canine 2A-adrenoceptor subtype. Neuropharmacology. 44: the basolateral membrane of Madin-Darby canine 2A-adrenoceptor subtype. Neuropharmacology. 44: kidney cells. J Biol Chem. 269: 16425-32. 882-92. kidney cells. J Biol Chem. 269: 16425-32. 882-92.

Keefer JR, Limbird LE (1993). The 2A-adrenergic Lakhlani PP, MacMillan LB, Guo TZ, McCool BA, Keefer JR, Limbird LE (1993). The 2A-adrenergic Lakhlani PP, MacMillan LB, Guo TZ, McCool BA, receptor is targeted directly to the basolateral Lovinger DM, Maze M, et al. (1997). Substitution of receptor is targeted directly to the basolateral Lovinger DM, Maze M, et al. (1997). Substitution of membrane domain of Madin-Darby canine kidney a mutant 2A-adrenergic receptor via "hit and run" membrane domain of Madin-Darby canine kidney a mutant 2A-adrenergic receptor via "hit and run" cells independent of coupling to pertussis toxin- gene targeting reveals the role of this subtype in cells independent of coupling to pertussis toxin- gene targeting reveals the role of this subtype in sensitive GTP-binding proteins. J Biol Chem. 268: sedative, analgesic, and anesthetic-sparing responses sensitive GTP-binding proteins. J Biol Chem. 268: sedative, analgesic, and anesthetic-sparing responses 11340-7. in vivo. Proc Natl Acad Sci USA. 94: 9950-5. 11340-7. in vivo. Proc Natl Acad Sci USA. 94: 9950-5. Kenakin T (2004). Principles: Receptor theory in Langer SZ (1976). The role of alpha- and beta- Kenakin T (2004). Principles: Receptor theory in Langer SZ (1976). The role of alpha- and beta- pharmacology. Trends Pharmacol Sci. 25: 186-92. presynaptic receptors in the regulation of pharmacology. Trends Pharmacol Sci. 25: 186-92. presynaptic receptors in the regulation of Kobilka BK, Matsui H, Kobilka TS, Yang-Feng TL, noradrenaline release elicited by nerve stimulation. Kobilka BK, Matsui H, Kobilka TS, Yang-Feng TL, noradrenaline release elicited by nerve stimulation. Francke U, Caron MG, et al. (1987). Cloning, Clin Sci Mol Med Suppl. 3: 423-6. Francke U, Caron MG, et al. (1987). Cloning, Clin Sci Mol Med Suppl. 3: 423-6. sequencing, and expression of the gene coding for Lanier SM, Downing S, Duzic E and Homcy CJ sequencing, and expression of the gene coding for Lanier SM, Downing S, Duzic E and Homcy CJ the human platelet 2-adrenergic receptor. Science. (1991). Isolation of rat genomic clones encoding the human platelet 2-adrenergic receptor. Science. (1991). Isolation of rat genomic clones encoding 238: 650-6. subtypes of the 2-adrenergic receptor. Identification 238: 650-6. subtypes of the 2-adrenergic receptor. Identification Koch WJ, Hawes BE, Inglese J, Luttrell LM and of a unique receptor subtype. J Biol Chem. 266: Koch WJ, Hawes BE, Inglese J, Luttrell LM and of a unique receptor subtype. J Biol Chem. 266: Lefkowitz RJ (1994). Cellular expression of the 10470-8. Lefkowitz RJ (1994). Cellular expression of the 10470-8. carboxyl terminus of a G protein-coupled receptor Lee A, Wissekerke AE, Rosin DL and Lynch KR carboxyl terminus of a G protein-coupled receptor Lee A, Wissekerke AE, Rosin DL and Lynch KR kinase attenuates G-mediated signaling. J Biol (1998). Localization of 2C-adrenergic receptor kinase attenuates G-mediated signaling. J Biol (1998). Localization of 2C-adrenergic receptor Chem. 269: 6193-7. immunoreactivity in catecholaminergic neurons in Chem. 269: 6193-7. immunoreactivity in catecholaminergic neurons in the rat central nervous system. Neuroscience. 84: the rat central nervous system. Neuroscience. 84: 1085-96. 1085-96.

78 References 78 References

Lefkowitz RJ, Cotecchia S, Samama P and Costa T the 2A-adrenergic receptor subtype. Science. 273: Lefkowitz RJ, Cotecchia S, Samama P and Costa T the 2A-adrenergic receptor subtype. Science. 273: (1993). Constitutive activity of receptors coupled to 801-3. (1993). Constitutive activity of receptors coupled to 801-3. guanine nucleotide regulatory proteins. Trends MacNulty EE, McClue SJ, Carr IC, Jess T, Wakelam guanine nucleotide regulatory proteins. Trends MacNulty EE, McClue SJ, Carr IC, Jess T, Wakelam Pharmacol Sci. 14: 303-7. Pharmacol Sci. 14: 303-7. MJ and Milligan G (1992). 2-C10 adrenergic MJ and Milligan G (1992). 2-C10 adrenergic Lefkowitz RJ and Shenoy SK (2005). Transduction of receptors expressed in rat 1 fibroblasts can regulate Lefkowitz RJ and Shenoy SK (2005). Transduction of receptors expressed in rat 1 fibroblasts can regulate receptor signals by -arrestins. Science. 308: 512-7. both adenylylcyclase and phospholipase D-mediated receptor signals by -arrestins. Science. 308: 512-7. both adenylylcyclase and phospholipase D-mediated Lehtonen JV, Still DJ, Rantanen VV, Ekholm J, hydrolysis of phosphatidylcholine by interacting Lehtonen JV, Still DJ, Rantanen VV, Ekholm J, hydrolysis of phosphatidylcholine by interacting Björklund D, Iftikhar Z, et al. (2004). BODIL: A with pertussis toxin-sensitive guanine nucleotide- Björklund D, Iftikhar Z, et al. (2004). BODIL: A with pertussis toxin-sensitive guanine nucleotide- molecular modeling environment for structure- binding proteins. J Biol Chem. 267: 2149-56. molecular modeling environment for structure- binding proteins. J Biol Chem. 267: 2149-56. function analysis and drug design. J Comput Aided Maehle AH (2004). "Receptive substances": John function analysis and drug design. J Comput Aided Maehle AH (2004). "Receptive substances": John Mol Des. 18: 401-19. Newport Langley (1852-1925) and his path to a Mol Des. 18: 401-19. Newport Langley (1852-1925) and his path to a Li J, Edwards PC, Burghammer M, Villa C and receptor theory of drug action. Med Hist. 48: 153-74. Li J, Edwards PC, Burghammer M, Villa C and receptor theory of drug action. Med Hist. 48: 153-74. Schertler GF (2004). Structure of bovine rhodopsin Makaritsis KP, Handy DE, Johns C, Kobilka B, Gavras Schertler GF (2004). Structure of bovine rhodopsin Makaritsis KP, Handy DE, Johns C, Kobilka B, Gavras in a trigonal crystal form. J Mol Biol. 343: 1409-38. I and Gavras H (1999). Role of the 2B-adrenergic in a trigonal crystal form. J Mol Biol. 343: 1409-38. I and Gavras H (1999). Role of the 2B-adrenergic Liapakis G, Simpson MM and Javitch JA (2001). The receptor in the development of salt-induced Liapakis G, Simpson MM and Javitch JA (2001). The receptor in the development of salt-induced substituted-cysteine accessibility method (SCAM) to hypertension. Hypertension. 33: 14-7. substituted-cysteine accessibility method (SCAM) to hypertension. Hypertension. 33: 14-7. elucidate membrane protein structure. Curr Protoc Marjamäki A, Ala-Uotila S, Luomala K, Peräla M, elucidate membrane protein structure. Curr Protoc Marjamäki A, Ala-Uotila S, Luomala K, Peräla M, Neurosci. Chapter 4: Unit 4.15. Jansson C, Jalkanen M, et al. (1992). Stable Neurosci. Chapter 4: Unit 4.15. Jansson C, Jalkanen M, et al. (1992). Stable Limbird LE (1988). Receptors linked to inhibition of expression of recombinant human 2-adrenoceptor Limbird LE (1988). Receptors linked to inhibition of expression of recombinant human 2-adrenoceptor subtypes in two mammalian cell lines: subtypes in two mammalian cell lines: adenylate cyclase: Additional signaling mechanisms. 3 adenylate cyclase: Additional signaling mechanisms. 3 FASEB J. 2: 2686-95. Characterization with [ H]rauwolscine binding, FASEB J. 2: 2686-95. Characterization with [ H]rauwolscine binding, inhibition of adenylate cyclase and RNase protection inhibition of adenylate cyclase and RNase protection Link R, Daunt D, Barsh G, Chruscinski A and Kobilka assay. Biochim Biophys Acta. 1134: 169-77. Link R, Daunt D, Barsh G, Chruscinski A and Kobilka assay. Biochim Biophys Acta. 1134: 169-77. B (1992). Cloning of two mouse genes encoding 2- B (1992). Cloning of two mouse genes encoding 2- adrenergic receptor subtypes and identification of a Marjamäki A, Frang H, Pihlavisto M, Hoffrèn AM, adrenergic receptor subtypes and identification of a Marjamäki A, Frang H, Pihlavisto M, Hoffrèn AM, Salminen T, Johnson MS, et al. (1999). Salminen T, Johnson MS, et al. (1999). single amino acid in the mouse 2-C10 homolog single amino acid in the mouse 2-C10 homolog responsible for an interspecies variation in Chloroethylclonidine and 2-aminoethyl responsible for an interspecies variation in Chloroethylclonidine and 2-aminoethyl antagonist binding. Mol Pharmacol. 42: 16-27. methanethiosulfonate recognize two different antagonist binding. Mol Pharmacol. 42: 16-27. methanethiosulfonate recognize two different conformations of the human 2A-adrenergic conformations of the human 2A-adrenergic Link RE, Desai K, Hein L, Stevens ME, Chruscinski receptor. J Biol Chem. 274: 21867-72. Link RE, Desai K, Hein L, Stevens ME, Chruscinski receptor. J Biol Chem. 274: 21867-72. A, Bernstein D, et al. (1996). Cardiovascular A, Bernstein D, et al. (1996). Cardiovascular Marjamäki A, Luomala K, Ala-Uotila S and Scheinin Marjamäki A, Luomala K, Ala-Uotila S and Scheinin regulation in mice lacking 2-adrenergic receptor regulation in mice lacking 2-adrenergic receptor subtypes b and c. Science. 273: 803-5. M (1993). Use of recombinant human 2- subtypes b and c. Science. 273: 803-5. M (1993). Use of recombinant human 2- adrenoceptors to characterize subtype selectively of adrenoceptors to characterize subtype selectively of Link RE, Stevens MS, Kulatunga M, Scheinin M, antagonist binding. Eur J Pharmacol. 246: 219-26. Link RE, Stevens MS, Kulatunga M, Scheinin M, antagonist binding. Eur J Pharmacol. 246: 219-26. Barsh GS and Kobilka BK (1995). Targeted Barsh GS and Kobilka BK (1995). Targeted Marjamäki A, Pihlavisto M, Cockcroft V, Heinonen P, Marjamäki A, Pihlavisto M, Cockcroft V, Heinonen P, inactivation of the gene encoding the mouse 2C- inactivation of the gene encoding the mouse 2C- adrenoceptor homolog. Mol Pharmacol. 48: 48-55. Savola JM and Scheinin M (1998). adrenoceptor homolog. Mol Pharmacol. 48: 48-55. Savola JM and Scheinin M (1998). Chloroethylclonidine binds irreversibly to exposed Chloroethylclonidine binds irreversibly to exposed Lohse MJ, Benovic JL, Codina J, Caron MG and cysteines in the fifth membrane-spanning domain of Lohse MJ, Benovic JL, Codina J, Caron MG and cysteines in the fifth membrane-spanning domain of Lefkowitz RJ (1990). -Arrestin: a protein that the human 2A-adrenergic receptor. Mol Pharmacol. Lefkowitz RJ (1990). -Arrestin: a protein that the human 2A-adrenergic receptor. Mol Pharmacol. regulates -adrenergic receptor function. Science. 53: 370-6. regulates -adrenergic receptor function. Science. 53: 370-6. 248: 1547-50. 248: 1547-50. Mathias CJ, Luckitt J, Desai P, Baker H, el Masri W Mathias CJ, Luckitt J, Desai P, Baker H, el Masri W Lomasney JW, Cotecchia S, Lefkowitz RJ and Caron and Frankel HL (1989). Pharmacodynamics and Lomasney JW, Cotecchia S, Lefkowitz RJ and Caron and Frankel HL (1989). Pharmacodynamics and MG (1991). Molecular biology of alpha-adrenergic pharmacokinetics of the oral antispastic agent MG (1991). Molecular biology of alpha-adrenergic pharmacokinetics of the oral antispastic agent receptors: Implications for receptor classification tizanidine in patients with spinal cord injury. J receptors: Implications for receptor classification tizanidine in patients with spinal cord injury. J and for structure-function relationships. Biochim Rehabil Res Dev. 26: 9-16. and for structure-function relationships. Biochim Rehabil Res Dev. 26: 9-16. Biophys Acta. 1095: 127-39. Biophys Acta. 1095: 127-39. McCune SK, Voigt MM and Hill JM (1993). McCune SK, Voigt MM and Hill JM (1993). Lomasney JW, Lorenz W, Allen LF, King K, Regan Expression of multiple alpha adrenergic receptor Lomasney JW, Lorenz W, Allen LF, King K, Regan Expression of multiple alpha adrenergic receptor JW, Yang-Feng TL, et al. (1990). Expansion of the subtype messenger RNAs in the adult rat brain. JW, Yang-Feng TL, et al. (1990). Expansion of the subtype messenger RNAs in the adult rat brain. 2-adrenergic receptor family: Cloning and Neuroscience. 57: 143-51. 2-adrenergic receptor family: Cloning and Neuroscience. 57: 143-51. characterization of a human 2-adrenergic receptor characterization of a human 2-adrenergic receptor subtype, the gene for which is located on McNamara DM, MacGowan GA and London B subtype, the gene for which is located on McNamara DM, MacGowan GA and London B chromosome 2. Proc Natl Acad Sci U S A. 87: 5094- (2002). Clinical importance of -adrenoceptor chromosome 2. Proc Natl Acad Sci U S A. 87: 5094- (2002). Clinical importance of -adrenoceptor 8. polymorphisms in cardiovascular disease. Am J 8. polymorphisms in cardiovascular disease. Am J Pharmacogenomics. 2: 73-8. Pharmacogenomics. 2: 73-8. MacDonald E, Kobilka BK and Scheinin M (1997). MacDonald E, Kobilka BK and Scheinin M (1997). Melchiorre C, Gallucci P (1983). The irreversible Melchiorre C, Gallucci P (1983). The irreversible Gene targeting--homing in on 2-adrenoceptor- Gene targeting--homing in on 2-adrenoceptor- subtype function. Trends Pharmacol Sci. 18: 211-9. alpha-blocker benextramine interacts with two subtype function. Trends Pharmacol Sci. 18: 211-9. alpha-blocker benextramine interacts with two different thiol groups. Farmaco Sci. 38: 950-6. different thiol groups. Farmaco Sci. 38: 950-6. MacDonald E, Scheinin M (1995). Distribution and MacDonald E, Scheinin M (1995). Distribution and Menon ST, Han M and Sakmar TP (2001). Rhodopsin: Menon ST, Han M and Sakmar TP (2001). Rhodopsin: pharmacology of 2-adrenoceptors in the central pharmacology of 2-adrenoceptors in the central nervous system. J Physiol Pharmacol. 46: 241-58. Structural basis of molecular physiology. Physiol nervous system. J Physiol Pharmacol. 46: 241-58. Structural basis of molecular physiology. Physiol Rev. 81: 1659-88. Rev. 81: 1659-88. MacMillan LB, Hein L, Smith MS, Piascik MT and MacMillan LB, Hein L, Smith MS, Piascik MT and Limbird LE (1996). Central hypotensive effects of Limbird LE (1996). Central hypotensive effects of

References 79 References 79

Michel AD, Loury DN and Whiting RL (1989). Olli-Lähdesmäki T, Scheinin M, Pohjanoksa K and Michel AD, Loury DN and Whiting RL (1989). Olli-Lähdesmäki T, Scheinin M, Pohjanoksa K and Differences between the 2-adrenoceptor in rat Kallio J (2003). Agonist-dependent trafficking of 2- Differences between the 2-adrenoceptor in rat Kallio J (2003). Agonist-dependent trafficking of 2- submaxillary gland and the2A-and 2B- adrenoceptor subtypes: Dependence on receptor submaxillary gland and the2A-and 2B- adrenoceptor subtypes: Dependence on receptor adrenoceptor subtypes. Br J Pharmacol. 98: 890-7. subtype and employed agonist. Eur J Cell Biol. 82: adrenoceptor subtypes. Br J Pharmacol. 98: 890-7. subtype and employed agonist. Eur J Cell Biol. 82: Milligan G, Kostenis E (2006). Heterotrimeric G- 231-9. Milligan G, Kostenis E (2006). Heterotrimeric G- 231-9. proteins: A short history. Br J Pharmacol. 147: 46- O'Rourke MF, Blaxall HS, Iversen LJ and Bylund DB proteins: A short history. Br J Pharmacol. 147: 46- O'Rourke MF, Blaxall HS, Iversen LJ and Bylund DB 55. (1994a). Characterization of [3H]RX821002 binding 55. (1994a). Characterization of [3H]RX821002 binding Milligan G, Smith NJ (2007). Allosteric modulation of to 2-adrenergic receptor subtypes. J Pharmacol Exp Milligan G, Smith NJ (2007). Allosteric modulation of to 2-adrenergic receptor subtypes. J Pharmacol Exp heterodimeric G-protein-coupled receptors. Trends Ther. 268: 1362-7. heterodimeric G-protein-coupled receptors. Trends Ther. 268: 1362-7. Pharmacol Sci. 28: 615-20. O'Rourke MF, Iversen LJ, Lomasney JW and Bylund Pharmacol Sci. 28: 615-20. O'Rourke MF, Iversen LJ, Lomasney JW and Bylund Murakami M, Kouyama T (2008). Crystal structure of DB (1994b). Species orthologs of the 2A-adrenergic Murakami M, Kouyama T (2008). Crystal structure of DB (1994b). Species orthologs of the 2A-adrenergic squid rhodopsin. Nature. 453: 363-7. receptor: The pharmacological properties of the squid rhodopsin. Nature. 453: 363-7. receptor: The pharmacological properties of the bovine and rat receptors differ from the human and bovine and rat receptors differ from the human and Naselsky DP, Ashton D, Ruffolo RR,Jr and Hieble JP porcine receptors. J Pharmacol Exp Ther. 271: 735- Naselsky DP, Ashton D, Ruffolo RR,Jr and Hieble JP porcine receptors. J Pharmacol Exp Ther. 271: 735- (2001). Rabbit 2-adrenoceptors: Both platelets and 40. (2001). Rabbit 2-adrenoceptors: Both platelets and 40. adipocytes have 2A-pharmacology. J Pharmacol adipocytes have 2A-pharmacology. J Pharmacol Exp Ther. 298: 219-25. Paddleford RR, Harvey RC (1999). 2-Agonists and Exp Ther. 298: 219-25. Paddleford RR, Harvey RC (1999). 2-Agonists and antagonists. Vet Clin North Am Small Anim Pract. antagonists. Vet Clin North Am Small Anim Pract. Neve KA, Cumbay MG, Thompson KR, Yang R, Buck 29: 737-45. Neve KA, Cumbay MG, Thompson KR, Yang R, Buck 29: 737-45. DC, Watts VJ, et al. (2001). Modeling and DC, Watts VJ, et al. (2001). Modeling and mutational analysis of a putative sodium-binding Palczewski K, Kumasaka T, Hori T, Behnke CA, mutational analysis of a putative sodium-binding Palczewski K, Kumasaka T, Hori T, Behnke CA, Motoshima H, Fox BA, et al. (2000). Crystal Motoshima H, Fox BA, et al. (2000). Crystal pocket on the dopamine D2 receptor. Mol pocket on the dopamine D2 receptor. Mol Pharmacol. 60: 373-81. structure of rhodopsin: A G protein-coupled Pharmacol. 60: 373-81. structure of rhodopsin: A G protein-coupled receptor. Science. 289: 739-45. receptor. Science. 289: 739-45. Nicholas AP, Pieribone V and Hökfelt T (1993). Nicholas AP, Pieribone V and Hökfelt T (1993). Patton EE, Zon LI (2001). The art and design of Patton EE, Zon LI (2001). The art and design of Distributions of mRNAs for 2-adrenergic receptor Distributions of mRNAs for 2-adrenergic receptor subtypes in rat brain: An in situ hybridization study. genetic screens: Zebrafish. Nat Rev Genet. 2: 956- subtypes in rat brain: An in situ hybridization study. genetic screens: Zebrafish. Nat Rev Genet. 2: 956- J Comp Neurol. 328: 575-94. 66. J Comp Neurol. 328: 575-94. 66. Niimura Y, Nei M (2006). Evolutionary dynamics of Peltonen JM, Nyrönen T, Wurster S, Pihlavisto M, Niimura Y, Nei M (2006). Evolutionary dynamics of Peltonen JM, Nyrönen T, Wurster S, Pihlavisto M, olfactory and other chemosensory receptor genes in Hoffrèn AM, Marjamäki A, et al. (2003). Molecular olfactory and other chemosensory receptor genes in Hoffrèn AM, Marjamäki A, et al. (2003). Molecular vertebrates. J Hum Genet. 51: 505-17. mechanisms of ligand-receptor interactions in vertebrates. J Hum Genet. 51: 505-17. mechanisms of ligand-receptor interactions in transmembrane domain V of the 2A-adrenoceptor. transmembrane domain V of the 2A-adrenoceptor. Nobles M, Benians A and Tinker A (2005). Br J Pharmacol. 140: 347-58. Nobles M, Benians A and Tinker A (2005). Br J Pharmacol. 140: 347-58. Heterotrimeric G proteins precouple with G protein- Heterotrimeric G proteins precouple with G protein- Peltonen JM, Pihlavisto M and ScheininM (1998). Peltonen JM, Pihlavisto M and ScheininM (1998). coupled receptors in living cells. Proc Natl Acad Sci 35 coupled receptors in living cells. Proc Natl Acad Sci 35 U S A. 102: 18706-11. Subtype-specific stimulation of [ S]GTPS binding U S A. 102: 18706-11. Subtype-specific stimulation of [ S]GTPS binding by recombinant 2-adrenoceptors. Eur J Pharmacol by recombinant 2-adrenoceptors. Eur J Pharmacol Nordström KJ, Lagerström MC, Waller LM, 355: 275-9. Nordström KJ, Lagerström MC, Waller LM, 355: 275-9. Fredriksson R and Schiöth HB (2009). The secretin Fredriksson R and Schiöth HB (2009). The secretin GPCRs descended from the family of adhesion Perälä M, Hirvonen H, Kalimo H, Ala-Uotila S, Regan GPCRs descended from the family of adhesion Perälä M, Hirvonen H, Kalimo H, Ala-Uotila S, Regan GPCRs. Mol Biol Evol. 26: 71-84. JW, Akerman KE, et al. (1992). Differential GPCRs. Mol Biol Evol. 26: 71-84. JW, Akerman KE, et al. (1992). Differential expression of two 2-adrenergic receptor subtype expression of two 2-adrenergic receptor subtype Nyrönen T, Pihlavisto M, Peltonen JM, Hoffrèn AM, mRNAs in human tissues. Brain Res Mol Brain Res. Nyrönen T, Pihlavisto M, Peltonen JM, Hoffrèn AM, mRNAs in human tissues. Brain Res Mol Brain Res. Varis M, Salminen T, et al. (2001). Molecular 16: 57-63. Varis M, Salminen T, et al. (2001). Molecular 16: 57-63. mechanism for agonist-promoted 2A-adrenoceptor mechanism for agonist-promoted 2A-adrenoceptor activation by norepinephrine and epinephrine. Mol Peterhoff M, Sieg A, Brede M, Chao CM, Hein L and activation by norepinephrine and epinephrine. Mol Peterhoff M, Sieg A, Brede M, Chao CM, Hein L and Pharmacol. 59: 1343-54. Ullrich S (2003). Inhibition of insulin secretion via Pharmacol. 59: 1343-54. Ullrich S (2003). Inhibition of insulin secretion via distinct signaling pathways in 2-adrenoceptor distinct signaling pathways in 2-adrenoceptor Okada T, Fujiyoshi Y, Silow M, Navarro J, Landau knockout mice. Eur J Endocrinol. 149: 343-50. Okada T, Fujiyoshi Y, Silow M, Navarro J, Landau knockout mice. Eur J Endocrinol. 149: 343-50. EM and Shichida Y (2002). Functional role of EM and Shichida Y (2002). Functional role of internal water molecules in rhodopsin revealed by Pfeuffer T, Helmreich EJ (1988). Structural and internal water molecules in rhodopsin revealed by Pfeuffer T, Helmreich EJ (1988). Structural and X-ray crystallography. Proc Natl Acad Sci U S A. functional relationships of guanosine triphosphate X-ray crystallography. Proc Natl Acad Sci U S A. functional relationships of guanosine triphosphate 99: 5982-7. binding proteins. Curr Top Cell Regul. 29: 129-216. 99: 5982-7. binding proteins. Curr Top Cell Regul. 29: 129-216. Okada T, Sugihara M, Bondar AN, Elstner M, Entel P Philipp M, Brede ME, Hadamek K, Gessler M, Lohse Okada T, Sugihara M, Bondar AN, Elstner M, Entel P Philipp M, Brede ME, Hadamek K, Gessler M, Lohse and Buss V (2004). The retinal conformation and its MJ and Hein L (2002). Placental 2-adrenoceptors and Buss V (2004). The retinal conformation and its MJ and Hein L (2002). Placental 2-adrenoceptors environment in rhodopsin in light of a new 2.2 Å control vascular development at the interface environment in rhodopsin in light of a new 2.2 Å control vascular development at the interface crystal structure. J Mol Biol. 342: 571-83. between mother and embryo. Nat Genet. 31: 311-5. crystal structure. J Mol Biol. 342: 571-83. between mother and embryo. Nat Genet. 31: 311-5. Oliveira L, Paiva PB, Paiva AC and Vriend G (2003). Pohjanoksa K, Jansson CC, Luomala K, Marjamäki A, Oliveira L, Paiva PB, Paiva AC and Vriend G (2003). Pohjanoksa K, Jansson CC, Luomala K, Marjamäki A, Sequence analysis reveals how G protein-coupled Savola JM and Scheinin M (1997). 2-Adrenoceptor Sequence analysis reveals how G protein-coupled Savola JM and Scheinin M (1997). 2-Adrenoceptor receptors transduce the signal to the G protein. regulation of adenylyl cyclase in CHO cells: receptors transduce the signal to the G protein. regulation of adenylyl cyclase in CHO cells: Proteins. 52: 553-60. Dependence on receptor density, receptor subtype Proteins. 52: 553-60. Dependence on receptor density, receptor subtype and current activity of adenylyl cyclase. Eur J and current activity of adenylyl cyclase. Eur J Olli-Lähdesmäki T, Kallio J and Scheinin M (1999). Pharmacol. 335: 53-63. Olli-Lähdesmäki T, Kallio J and Scheinin M (1999). Pharmacol. 335: 53-63. Receptor subtype-induced targeting and subtype- Receptor subtype-induced targeting and subtype- Porter JE, Hwa J and Perez DM (1996). Activation of Porter JE, Hwa J and Perez DM (1996). Activation of specific internalization of human 2-adrenoceptors specific internalization of human 2-adrenoceptors in PC12 cells. J Neurosci. 19: 9281-8. the 1b-adrenergic receptor is initiated by disruption in PC12 cells. J Neurosci. 19: 9281-8. the 1b-adrenergic receptor is initiated by disruption

80 References 80 References of an interhelical salt bridge constraint. J Biol Chem. Sallinen J, Haapalinna A, MacDonald E, Viitamaa T, of an interhelical salt bridge constraint. J Biol Chem. Sallinen J, Haapalinna A, MacDonald E, Viitamaa T, 271: 28318-23. Lähdesmäki J, Rybnikova E, et al. (1999). Genetic 271: 28318-23. Lähdesmäki J, Rybnikova E, et al. (1999). Genetic Postlethwait JH, Yan YL, Gates MA, Horne S, Amores alteration of the 2-adrenoceptor subtype c in mice Postlethwait JH, Yan YL, Gates MA, Horne S, Amores alteration of the 2-adrenoceptor subtype c in mice A, Brownlie A, et al. (1998). Vertebrate genome affects the development of behavioral despair and A, Brownlie A, et al. (1998). Vertebrate genome affects the development of behavioral despair and evolution and the zebrafish gene map. Nat Genet. stress-induced increases in plasma corticosterone evolution and the zebrafish gene map. Nat Genet. stress-induced increases in plasma corticosterone 18: 345-9. levels. Mol Psychiatry. 4: 443-52. 18: 345-9. levels. Mol Psychiatry. 4: 443-52. Rasmussen SG, Choi HJ, Rosenbaum DM, Kobilka Sallinen J, Haapalinna A, Viitamaa T, Kobilka BK and Rasmussen SG, Choi HJ, Rosenbaum DM, Kobilka Sallinen J, Haapalinna A, Viitamaa T, Kobilka BK and TS, Thian FS, Edwards PC, et al. (2007). Crystal Scheinin M (1998). Adrenergic 2C-receptors TS, Thian FS, Edwards PC, et al. (2007). Crystal Scheinin M (1998). Adrenergic 2C-receptors modulate the acoustic startle reflex, prepulse modulate the acoustic startle reflex, prepulse structure of the human 2-adrenergic G-protein- structure of the human 2-adrenergic G-protein- coupled receptor. Nature. 450: 383-7. inhibition, and aggression in mice. J Neurosci. 18: coupled receptor. Nature. 450: 383-7. inhibition, and aggression in mice. J Neurosci. 18: 3035-42. 3035-42. Regan JW, DeMarinis RM, Caron MG and Lefkowitz Regan JW, DeMarinis RM, Caron MG and Lefkowitz RJ (1984). Identification of the subunit-binding site Sallinen J, Link RE, Haapalinna A, Viitamaa T, RJ (1984). Identification of the subunit-binding site Sallinen J, Link RE, Haapalinna A, Viitamaa T, Kulatunga M, Sjöholm B, et al. (1997). Genetic Kulatunga M, Sjöholm B, et al. (1997). Genetic of 2-adrenergic receptors using of 2-adrenergic receptors using [3H]phenoxybenzamine. J Biol Chem. 259: 7864-9. alteration of 2C-adrenoceptor expression in mice: [3H]phenoxybenzamine. J Biol Chem. 259: 7864-9. alteration of 2C-adrenoceptor expression in mice: Influence on locomotor, hypothermic, and Influence on locomotor, hypothermic, and Regan JW, Kobilka TS, Yang-Feng TL, Caron MG, neurochemical effects of dexmedetomidine, a Regan JW, Kobilka TS, Yang-Feng TL, Caron MG, neurochemical effects of dexmedetomidine, a Lefkowitz RJ and Kobilka BK (1988). Cloning and subtype-nonselective 2-adrenoceptor agonist. Mol Lefkowitz RJ and Kobilka BK (1988). Cloning and subtype-nonselective 2-adrenoceptor agonist. Mol expression of a human kidney cDNA for an 2- Pharmacol. 51: 36-46. expression of a human kidney cDNA for an 2- Pharmacol. 51: 36-46. adrenergic receptor subtype. Proc Natl Acad Sci U S adrenergic receptor subtype. Proc Natl Acad Sci U S A. 85: 6301-5. Salminen T, Varis M, Nyrönen T, Pihlavisto M, A. 85: 6301-5. Salminen T, Varis M, Nyrönen T, Pihlavisto M, Hoffrèn AM, Lönnberg T, et al. (1999). Three- Hoffrèn AM, Lönnberg T, et al. (1999). Three- Regan JW, Nakata H, DeMarinis RM, Caron MG and dimensional models of 2A-adrenergic receptor Regan JW, Nakata H, DeMarinis RM, Caron MG and dimensional models of 2A-adrenergic receptor Lefkowitz RJ (1986). Purification and complexes provide a structural explanation for Lefkowitz RJ (1986). Purification and complexes provide a structural explanation for characterization of the human platelet 2-adrenergic ligand binding. J Biol Chem. 274: 23405-13. characterization of the human platelet 2-adrenergic ligand binding. J Biol Chem. 274: 23405-13. receptor. J Biol Chem. 261: 3894-900. receptor. J Biol Chem. 261: 3894-900. Samama P, Pei G, Costa T, Cotecchia S and Lefkowitz Samama P, Pei G, Costa T, Cotecchia S and Lefkowitz Richman JG, Regan JW (1998). 2-Adrenergic RJ (1994). Negative antagonists promote an inactive Richman JG, Regan JW (1998). 2-Adrenergic RJ (1994). Negative antagonists promote an inactive receptors increase cell migration and decrease F- conformation of the 2-adrenergic receptor. Mol receptors increase cell migration and decrease F- conformation of the 2-adrenergic receptor. Mol actin labeling in rat aortic smooth muscle cells. Am J Pharmacol. 45: 390-4. actin labeling in rat aortic smooth muscle cells. Am J Pharmacol. 45: 390-4. Physiol. 274: 654-62. Physiol. 274: 654-62. Sastre M, Garcia-Sevilla JA (1993). Opposite age- Sastre M, Garcia-Sevilla JA (1993). Opposite age- Rios C, Gomes I and Devi LA (2004). Interactions dependent changes of 2A-adrenoceptors and Rios C, Gomes I and Devi LA (2004). Interactions dependent changes of 2A-adrenoceptors and 3 3 between  opioid receptors and -adrenoceptors. nonadrenoceptor [ H]idazoxan binding sites (I2- between  opioid receptors and -adrenoceptors. nonadrenoceptor [ H]idazoxan binding sites (I2- Clin Exp Pharmacol Physiol. 31: 833-6. imidazoline sites) in the human brain: Strong Clin Exp Pharmacol Physiol. 31: 833-6. imidazoline sites) in the human brain: Strong Rosenbaum DM, Cherezov V, Hanson MA, correlation of I2 with monoamine oxidase-B sites. J Rosenbaum DM, Cherezov V, Hanson MA, correlation of I2 with monoamine oxidase-B sites. J Rasmussen SG, Thian FS, Kobilka TS, et al. (2007). Neurochem. 61: 881-9. Rasmussen SG, Thian FS, Kobilka TS, et al. (2007). Neurochem. 61: 881-9. GPCR engineering yields high-resolution structural Saunders C, Limbird LE (1999). Localization and GPCR engineering yields high-resolution structural Saunders C, Limbird LE (1999). Localization and insights into 2-adrenergic receptor function. trafficking of 2-adrenergic receptor subtypes in insights into 2-adrenergic receptor function. trafficking of 2-adrenergic receptor subtypes in Science. 318: 1266-73. cells and tissues. Pharmacol Ther. 84: 193-205. Science. 318: 1266-73. cells and tissues. Pharmacol Ther. 84: 193-205. Rosin DL, Talley EM, Lee A, Stornetta RL, Gaylinn Savola JM, Ruskoaho H, Puurunen J, Salonen JS and Rosin DL, Talley EM, Lee A, Stornetta RL, Gaylinn Savola JM, Ruskoaho H, Puurunen J, Salonen JS and BD, Guyenet PG, et al. (1996). Distribution of 2C- Karki NT (1986). Evidence for medetomidine as a BD, Guyenet PG, et al. (1996). Distribution of 2C- Karki NT (1986). Evidence for medetomidine as a adrenergic receptor-like immunoreactivity in the rat selective and potent agonist at 2-adrenoreceptors. J adrenergic receptor-like immunoreactivity in the rat selective and potent agonist at 2-adrenoreceptors. J central nervous system. J Comp Neurol. 372: 135- Auton Pharmacol. 6: 275-84. central nervous system. J Comp Neurol. 372: 135- Auton Pharmacol. 6: 275-84. 65. 65. Schambra UB, Mackensen GB, Stafford-Smith M, Schambra UB, Mackensen GB, Stafford-Smith M, Rudling JE, Kennedy K and Evans PD (1999). The Haines DE and Schwinn DA (2005). Neuron Rudling JE, Kennedy K and Evans PD (1999). The Haines DE and Schwinn DA (2005). Neuron effect of site-directed mutagenesis of two specific -adrenergic receptor expression in human effect of site-directed mutagenesis of two specific -adrenergic receptor expression in human transmembrane serine residues on agonist-specific cerebellum: Implications for emerging cerebellar transmembrane serine residues on agonist-specific cerebellum: Implications for emerging cerebellar coupling of a cloned human 2A-adrenoceptor to roles in neurologic disease. Neuroscience. 135: 507- coupling of a cloned human 2A-adrenoceptor to roles in neurologic disease. Neuroscience. 135: 507- adenylyl cyclase. Br J Pharmacol. 127: 877-86. 23. adenylyl cyclase. Br J Pharmacol. 127: 877-86. 23. Ruffolo RR,Jr, Hieble JP (1994). -Adrenoceptors. Scheibner J, Trendelenburg AU, Hein L, Starke K and Ruffolo RR,Jr, Hieble JP (1994). -Adrenoceptors. Scheibner J, Trendelenburg AU, Hein L, Starke K and Pharmacol Ther. 61: 1-64. Blandizzi C (2002). 2-Adrenoceptors in the enteric Pharmacol Ther. 61: 1-64. Blandizzi C (2002). 2-Adrenoceptors in the enteric Ruffolo RR,Jr, Nichols AJ, Stadel JM and Hieble JP nervous system: A study in 2A-adrenoceptor- Ruffolo RR,Jr, Nichols AJ, Stadel JM and Hieble JP nervous system: A study in 2A-adrenoceptor- (1993). Pharmacologic and therapeutic applications deficient mice. Br J Pharmacol. 135: 697-704. (1993). Pharmacologic and therapeutic applications deficient mice. Br J Pharmacol. 135: 697-704. of 2-adrenoceptor subtypes. Annu Rev Pharmacol Scheinin M, Lomasney JW, Hayden-Hixson DM, of 2-adrenoceptor subtypes. Annu Rev Pharmacol Scheinin M, Lomasney JW, Hayden-Hixson DM, Toxicol. 33: 243-79. Schambra UB, Caron MG, Lefkowitz RJ, et al. Toxicol. 33: 243-79. Schambra UB, Caron MG, Lefkowitz RJ, et al. Ruuskanen JO, Xhaard H, Marjamäki A, Salaneck E, (1994). Distribution of 2-adrenergic receptor Ruuskanen JO, Xhaard H, Marjamäki A, Salaneck E, (1994). Distribution of 2-adrenergic receptor Salminen T, Yan YL, et al. (2004). Identification of subtype gene expression in rat brain. Brain Res Mol Salminen T, Yan YL, et al. (2004). Identification of subtype gene expression in rat brain. Brain Res Mol Brain Res. 21: 133-49. Brain Res. 21: 133-49. duplicated fourth 2-adrenergic receptor subtype by duplicated fourth 2-adrenergic receptor subtype by cloning and mapping of five receptor genes in Schertler GF, Hargrave PA (1995). Projection structure cloning and mapping of five receptor genes in Schertler GF, Hargrave PA (1995). Projection structure zebrafish. Mol Biol Evol. 21: 14-28. of frog rhodopsin in two crystal forms. Proc Natl zebrafish. Mol Biol Evol. 21: 14-28. of frog rhodopsin in two crystal forms. Proc Natl Sali A, Blundell TL (1993). Comparative protein Acad Sci U S A. 92: 11578-82. Sali A, Blundell TL (1993). Comparative protein Acad Sci U S A. 92: 11578-82. modelling by satisfaction of spatial restraints. J Mol Schiöth HB, Nordström KJ and Fredriksson R (2007). modelling by satisfaction of spatial restraints. J Mol Schiöth HB, Nordström KJ and Fredriksson R (2007). Biol. 234: 779-815. Mining the gene repertoire and ESTs for G protein- Biol. 234: 779-815. Mining the gene repertoire and ESTs for G protein-

References 81 References 81

coupled receptors with evolutionary perspective. insertion/deletion polymorphism in the a2B- coupled receptors with evolutionary perspective. insertion/deletion polymorphism in the a2B- Acta Physiol (Oxf). 190: 21-31. adrenergic receptor gene is a novel genetic risk Acta Physiol (Oxf). 190: 21-31. adrenergic receptor gene is a novel genetic risk Schramm NL, McDonald MP and Limbird LE (2001). factor for acute coronary events. J Am Coll Cardiol. Schramm NL, McDonald MP and Limbird LE (2001). factor for acute coronary events. J Am Coll Cardiol. 37: 1516-22. 37: 1516-22. The 2A-adrenergic receptor plays a protective role The 2A-adrenergic receptor plays a protective role in mouse behavioral models of depression and Snapir A, Mikkelsson J, Perola M, Penttilä A, Scheinin in mouse behavioral models of depression and Snapir A, Mikkelsson J, Perola M, Penttilä A, Scheinin anxiety. J Neurosci. 21: 4875-82. M and Karhunen PJ (2003). Variation in the 2B- anxiety. J Neurosci. 21: 4875-82. M and Karhunen PJ (2003). Variation in the 2B- Schwartz DD, Clark TP (1998). Selectivity of adrenoceptor gene as a risk factor for prehospital Schwartz DD, Clark TP (1998). Selectivity of adrenoceptor gene as a risk factor for prehospital fatal myocardial infarction and sudden cardiac death. fatal myocardial infarction and sudden cardiac death. atipamezole, yohimbine and tolazoline for 2- atipamezole, yohimbine and tolazoline for 2- adrenergic receptor subtypes: Implications for J Am Coll Cardiol. 41: 190-4. adrenergic receptor subtypes: Implications for J Am Coll Cardiol. 41: 190-4. clinical reversal of 2- adrenergic receptor mediated Soini SL, Duzic E, Lanier SM and Akerman KE clinical reversal of 2- adrenergic receptor mediated Soini SL, Duzic E, Lanier SM and Akerman KE sedation in sheep. J Vet Pharmacol Ther. 21: 342-7. (1998). Dual modulation of calcium channel current sedation in sheep. J Vet Pharmacol Ther. 21: 342-7. (1998). Dual modulation of calcium channel current Schwartz TW, Frimurer TM, Holst B, Rosenkilde MM via recombinant 2-adrenoceptors in Schwartz TW, Frimurer TM, Holst B, Rosenkilde MM via recombinant 2-adrenoceptors in and Elling CE (2006). Molecular mechanism of pheochromocytoma (PC-12) cells. Pflugers Arch. and Elling CE (2006). Molecular mechanism of pheochromocytoma (PC-12) cells. Pflugers Arch. 7TM receptor activation--a global toggle switch 435: 280-5. 7TM receptor activation--a global toggle switch 435: 280-5. model. Annu Rev Pharmacol Toxicol. 46: 481-519. Sprang SR, Chen Z and Du X (2007). Structural basis model. Annu Rev Pharmacol Toxicol. 46: 481-519. Sprang SR, Chen Z and Du X (2007). Structural basis Serle JB (1994). Pharmacological advances in the of effector regulation and signal termination in Serle JB (1994). Pharmacological advances in the of effector regulation and signal termination in treatment of glaucoma. Drugs Aging. 5: 156-70. heterotrimeric galpha proteins. Adv Protein Chem. treatment of glaucoma. Drugs Aging. 5: 156-70. heterotrimeric galpha proteins. Adv Protein Chem. 74: 1-65. 74: 1-65. Seuwen K, Magnaldo I, Kobilka BK, Caron MG, Seuwen K, Magnaldo I, Kobilka BK, Caron MG, Stimmel GL, Dopheide JA and Stahl SM (1997). Stimmel GL, Dopheide JA and Stahl SM (1997). Regan JW, Lefkowitz RJ, et al. (1990). 2- Regan JW, Lefkowitz RJ, et al. (1990). 2- Adrenergic agonists stimulate DNA synthesis in Mirtazapine: An antidepressant with noradrenergic Adrenergic agonists stimulate DNA synthesis in Mirtazapine: An antidepressant with noradrenergic chinese hamster lung fibroblasts transfected with a and specific serotonergic effects. Pharmacotherapy. chinese hamster lung fibroblasts transfected with a and specific serotonergic effects. Pharmacotherapy. 17: 10-21. 17: 10-21. human 2-adrenergic receptor gene. Cell Regul. 1: human 2-adrenergic receptor gene. Cell Regul. 1: 445-51. Surgand JS, Rodrigo J, Kellenberger E, Rognan D 445-51. Surgand JS, Rodrigo J, Kellenberger E, Rognan D Shi L, Javitch JA (2004). The second extracellular loop (2006). A chemogenomic analysis of the Shi L, Javitch JA (2004). The second extracellular loop (2006). A chemogenomic analysis of the transmembrane binding cavity of human G-protein- transmembrane binding cavity of human G-protein- of the dopamine D2 receptor lines the binding-site of the dopamine D2 receptor lines the binding-site crevice. Proc Natl Acad Sci U S A. 101: 440-5. coupled receptors. Proteins 62: 509-38. crevice. Proc Natl Acad Sci U S A. 101: 440-5. coupled receptors. Proteins 62: 509-38. Shi L, Javitch JA (2002). The binding site of aminergic Suryanarayana S, Daunt DA, Von Zastrow M and Shi L, Javitch JA (2002). The binding site of aminergic Suryanarayana S, Daunt DA, Von Zastrow M and G protein-coupled receptors: The transmembrane Kobilka BK (1991). A point mutation in the seventh G protein-coupled receptors: The transmembrane Kobilka BK (1991). A point mutation in the seventh segments and second extracellular loop. Annu Rev hydrophobic domain of the 2-adrenergic receptor segments and second extracellular loop. Annu Rev hydrophobic domain of the 2-adrenergic receptor Pharmacol Toxicol. 42: 437-67. increases its affinity for a family of - receptor Pharmacol Toxicol. 42: 437-67. increases its affinity for a family of - receptor antagonists. J Biol Chem. 266: 15488-92. antagonists. J Biol Chem. 266: 15488-92. Shin JT, Fishman MC (2002). From zebrafish to Shin JT, Fishman MC (2002). From zebrafish to human: Modular medical models. Annu Rev Suwa M, Ono Y (2009). Computational overview of human: Modular medical models. Annu Rev Suwa M, Ono Y (2009). Computational overview of Genomics Hum Genet. 3: 311-40. GPCR gene universe to support reverse chemical Genomics Hum Genet. 3: 311-40. GPCR gene universe to support reverse chemical genomics study. Methods Mol Biol. 577: 41-54. genomics study. Methods Mol Biol. 577: 41-54. Simonneaux V, Ebadi M and Bylund DB (1991). Simonneaux V, Ebadi M and Bylund DB (1991). Svensson SP, Bailey TJ, Porter AC, Richman JG and Svensson SP, Bailey TJ, Porter AC, Richman JG and Identification and characterization of 2D-adrenergic Identification and characterization of 2D-adrenergic receptors in bovine pineal gland. Mol Pharmacol. Regan JW (1996). Heterologous expression of the receptors in bovine pineal gland. Mol Pharmacol. Regan JW (1996). Heterologous expression of the 40: 235-41. cloned guinea pig 2A-, 2B-, and 2C-adrenoceptor 40: 235-41. cloned guinea pig 2A-, 2B-, and 2C-adrenoceptor subtypes. radioligand binding and functional subtypes. radioligand binding and functional Small KM, Brown KM, Forbes SL and Liggett SB coupling to a CAMP-responsive reporter gene. Small KM, Brown KM, Forbes SL and Liggett SB coupling to a CAMP-responsive reporter gene. (2001). Polymorphic deletion of three intracellular Biochem Pharmacol. 51: 291-300. (2001). Polymorphic deletion of three intracellular Biochem Pharmacol. 51: 291-300. acidic residues of the 2B-adrenergic receptor acidic residues of the 2B-adrenergic receptor decreases G protein-coupled receptor kinase- Szot P, Lester M, Laughlin ML, Palmiter RD, Liles LC decreases G protein-coupled receptor kinase- Szot P, Lester M, Laughlin ML, Palmiter RD, Liles LC mediated phosphorylation and desensitization. J Biol and Weinshenker D (2004). The anticonvulsant and mediated phosphorylation and desensitization. J Biol and Weinshenker D (2004). The anticonvulsant and Chem. 276: 4917-22. proconvulsant effects of 2-adrenoreceptor agonists Chem. 276: 4917-22. proconvulsant effects of 2-adrenoreceptor agonists are mediated by distinct populations of 2A- are mediated by distinct populations of 2A- Small KM, Forbes SL, Brown KM and Liggett SB adrenoreceptors. Neuroscience. 126: 795-803. Small KM, Forbes SL, Brown KM and Liggett SB adrenoreceptors. Neuroscience. 126: 795-803. (2000a). An Asn to Lys polymorphism in the third (2000a). An Asn to Lys polymorphism in the third Taraviras S, Olli-Lähdesmäki T, Lymperopoulos A, Taraviras S, Olli-Lähdesmäki T, Lymperopoulos A, intracellular loop of the human 2A-adrenergic intracellular loop of the human 2A-adrenergic receptor imparts enhanced agonist-promoted Gi Charitonidou D, Mavroidis M, Kallio J, et al. receptor imparts enhanced agonist-promoted Gi Charitonidou D, Mavroidis M, Kallio J, et al. coupling. J Biol Chem. 275: 38518-23. (2002). Subtype-specific neuronal differentiation of coupling. J Biol Chem. 275: 38518-23. (2002). Subtype-specific neuronal differentiation of PC12 cells transfected with 2-adrenergic receptors. PC12 cells transfected with 2-adrenergic receptors. Small KM, Forbes SL, Rahman FF, Bridges KM and Eur J Cell Biol. 81: 363-74. Small KM, Forbes SL, Rahman FF, Bridges KM and Eur J Cell Biol. 81: 363-74. Liggett SB (2000b). A four amino acid deletion Liggett SB (2000b). A four amino acid deletion polymorphism in the third intracellular loop of the Tavares A, Handy DE, Bogdanova NN, Rosene DL polymorphism in the third intracellular loop of the Tavares A, Handy DE, Bogdanova NN, Rosene DL and Gavras H (1996). Localization of 2A- and 2B- and Gavras H (1996). Localization of 2A- and 2B- human 2C-adrenergic receptor confers impaired human 2C-adrenergic receptor confers impaired coupling to multiple effectors. J Biol Chem. 275: adrenergic receptor subtypes in brain. Hypertension. coupling to multiple effectors. J Biol Chem. 275: adrenergic receptor subtypes in brain. Hypertension. 23059-64. 27: 449-55. 23059-64. 27: 449-55. Small KM, Liggett SB (2001). Identification and Teller DC, Okada T, Behnke CA, Palczewski K and Small KM, Liggett SB (2001). Identification and Teller DC, Okada T, Behnke CA, Palczewski K and Stenkamp RE (2001). Advances in determination of Stenkamp RE (2001). Advances in determination of functional characterization of 2-adrenoceptor functional characterization of 2-adrenoceptor polymorphisms. Trends Pharmacol Sci. 22: 471-7. a high-resolution three-dimensional structure of polymorphisms. Trends Pharmacol Sci. 22: 471-7. a high-resolution three-dimensional structure of rhodopsin, a model of G-protein-coupled receptors rhodopsin, a model of G-protein-coupled receptors Snapir A, Heinonen P, Tuomainen TP, Alhopuro P, (GPCRs). Biochemistry. 40: 7761-72. Snapir A, Heinonen P, Tuomainen TP, Alhopuro P, (GPCRs). Biochemistry. 40: 7761-72. Karvonen MK, Lakka TA, et al. (2001). An Karvonen MK, Lakka TA, et al. (2001). An

82 References 82 References

Timmons SD, Geisert E, Stewart AE, Lorenzon NM brimonidine in porcine ciliary arteries. Invest Timmons SD, Geisert E, Stewart AE, Lorenzon NM brimonidine in porcine ciliary arteries. Invest and Foehring RC (2004). 2-Adrenergic receptor- Ophthalmol Vis Sci. 42: 2049-55. and Foehring RC (2004). 2-Adrenergic receptor- Ophthalmol Vis Sci. 42: 2049-55. mediated modulation of calcium current in Wilden U, Hall SW and Kühn H (1986). mediated modulation of calcium current in Wilden U, Hall SW and Kühn H (1986). neocortical pyramidal neurons. Brain Res. 1014: Phosphodiesterase activation by photoexcited neocortical pyramidal neurons. Brain Res. 1014: Phosphodiesterase activation by photoexcited 184-96. rhodopsin is quenched when rhodopsin is 184-96. rhodopsin is quenched when rhodopsin is Trendelenburg AU, Klebroff W, Hein L and Starke K phosphorylated and binds the intrinsic 48-kDa Trendelenburg AU, Klebroff W, Hein L and Starke K phosphorylated and binds the intrinsic 48-kDa (2001). A study of presynaptic 2-autoreceptors in protein of rod outer segments. Proc Natl Acad Sci U (2001). A study of presynaptic 2-autoreceptors in protein of rod outer segments. Proc Natl Acad Sci U 2A/D-, 2B- and 2C-adrenoceptor-deficient mice. S A. 83: 1174-8. 2A/D-, 2B- and 2C-adrenoceptor-deficient mice. S A. 83: 1174-8. Naunyn Schmiedebergs Arch Pharmacol. 364: 117- Winzer-Serhan UH, Raymon HK, Broide RS, Chen Y Naunyn Schmiedebergs Arch Pharmacol. 364: 117- Winzer-Serhan UH, Raymon HK, Broide RS, Chen Y 30. 30. and Leslie FM (1997). Expression of 2- and Leslie FM (1997). Expression of 2- Trendelenburg AU, Sutej I, Wahl CA, Molderings GJ, adrenoceptors during rat brain development--II. Trendelenburg AU, Sutej I, Wahl CA, Molderings GJ, adrenoceptors during rat brain development--II. Rump LC and Starke K (1997). A re-investigation of alpha 2C messenger RNA expression and Rump LC and Starke K (1997). A re-investigation of alpha 2C messenger RNA expression and questionable subclassifications of presynaptic 2- [3H]rauwolscine binding. Neuroscience. 76: 261-72. questionable subclassifications of presynaptic 2- [3H]rauwolscine binding. Neuroscience. 76: 261-72. autoreceptors: Rat vena cava, rat atria, human kidney Wu B, Chien EY, Mol CD, Fenalti G, Liu W, Katritch autoreceptors: Rat vena cava, rat atria, human kidney Wu B, Chien EY, Mol CD, Fenalti G, Liu W, Katritch and guinea-pig urethra. Naunyn Schmiedebergs Arch V, et al. (2010). Structures of the CXCR4 and guinea-pig urethra. Naunyn Schmiedebergs Arch V, et al. (2010). Structures of the CXCR4 Pharmacol. 356: 721-37. chemokine GPCR with small-molecule and cyclic Pharmacol. 356: 721-37. chemokine GPCR with small-molecule and cyclic Uhlèn S, Dambrova M, Näsman J, Schiöth HB, Gu Y, peptide antagonists. Science. 330: 1066-71. Uhlèn S, Dambrova M, Näsman J, Schiöth HB, Gu Y, peptide antagonists. Science. 330: 1066-71. Wikberg-Matsson A, et al. (1998). 3H]RS79948-197 Wurch T, Colpaert FC and Pauwels PJ (1999). G- Wikberg-Matsson A, et al. (1998). 3H]RS79948-197 Wurch T, Colpaert FC and Pauwels PJ (1999). G- binding to human, rat, guinea pig and pig 2A-, 2B- protein activation by putative antagonists at mutant binding to human, rat, guinea pig and pig 2A-, 2B- protein activation by putative antagonists at mutant and 2C-adrenoceptors. comparison with MK912, and 2C-adrenoceptors. comparison with MK912, Thr373Lys 2A- adrenergic receptors. Br J Thr373Lys 2A- adrenergic receptors. Br J RX821002, rauwolscine and yohimbine. Eur J Pharmacol. 126: 939-48. RX821002, rauwolscine and yohimbine. Eur J Pharmacol. 126: 939-48. Pharmacol. 343: 93-101. Pharmacol. 343: 93-101. Wurch T, Colpaert FC and Pauwels PJ (1998). Wurch T, Colpaert FC and Pauwels PJ (1998). Uhlèn S, Porter AC and Neubig RR (1994). The novel Uhlèn S, Porter AC and Neubig RR (1994). The novel 3 Chimeric receptor analysis of the ketanserin binding 3 Chimeric receptor analysis of the ketanserin binding 2-adrenergic radioligand [ H]-MK912 is 2C- 2-adrenergic radioligand [ H]-MK912 is 2C- site in the human 5-hydroxytryptamine1D receptor: site in the human 5-hydroxytryptamine1D receptor: selective among human 2A-, 2B-and 2C- Importance of the second extracellular loop and fifth selective among human 2A-, 2B-and 2C- Importance of the second extracellular loop and fifth adrenoceptors. J Pharmacol Exp Ther. 271: 1558- transmembrane domain in antagonist binding. Mol adrenoceptors. J Pharmacol Exp Ther. 271: 1558- transmembrane domain in antagonist binding. Mol 65. Pharmacol. 54: 1088-96. 65. Pharmacol. 54: 1088-96. Venkataraman V, Duda T and Sharma RK (1997). A Wurch T, Pauwels PJ (2000). Coupling of canine Venkataraman V, Duda T and Sharma RK (1997). A Wurch T, Pauwels PJ (2000). Coupling of canine role for amino acid residues in the third cytoplasmic role for amino acid residues in the third cytoplasmic serotonin 5-HT1B and 5-HT1D receptor subtypes to serotonin 5-HT1B and 5-HT1D receptor subtypes to loop in defining the ligand binding characteristics of the formation of inositol phosphates by dual loop in defining the ligand binding characteristics of the formation of inositol phosphates by dual the 2D-adrenergic receptor. Mol Cell Biochem. 177: the 2D-adrenergic receptor. Mol Cell Biochem. 177: interactions with endogenous Gi/o and recombinant interactions with endogenous Gi/o and recombinant 125-9. 125-9. G15 proteins. J Neurochem. 75: 1180-9. G15 proteins. J Neurochem. 75: 1180-9. Vernier P, Cardinaud B, Valdenaire O, Philippe H and Xhaard H, Nyrönen T, Rantanen VV, Ruuskanen JO, Vernier P, Cardinaud B, Valdenaire O, Philippe H and Xhaard H, Nyrönen T, Rantanen VV, Ruuskanen JO, Vincent JD (1995). An evolutionary view of drug- Laurila J, Salminen T et al. (2005). Model structures Vincent JD (1995). An evolutionary view of drug- Laurila J, Salminen T et al. (2005). Model structures receptor interaction: The bioamine receptor family. receptor interaction: The bioamine receptor family. of 2- adrenoceptors in complex with automatically of 2- adrenoceptors in complex with automatically Trends Pharmacol Sci. 16: 375-81. docked antagonist ligands raise the possibility of Trends Pharmacol Sci. 16: 375-81. docked antagonist ligands raise the possibility of Violin JD and Lefkowitz RJ (2007). -Arrestin-biased interactions dissimilar from agonist ligands. J Struct Violin JD and Lefkowitz RJ (2007). -Arrestin-biased interactions dissimilar from agonist ligands. J Struct ligands at seven-transmembrane receptors. Trends Biol. 150: 126-43. ligands at seven-transmembrane receptors. Trends Biol. 150: 126-43. Pharmacol Sci 28: 416-22. Xhaard H, Rantanen VV, Nyrönen T and Johnson MS Pharmacol Sci 28: 416-22. Xhaard H, Rantanen VV, Nyrönen T and Johnson MS Virtanen R (1989). Pharmacological profiles of (2006). Molecular evolution of adrenoceptors and Virtanen R (1989). Pharmacological profiles of (2006). Molecular evolution of adrenoceptors and medetomidine and its antagonist, atipamezole. Acta dopamine receptors: Implications for the binding of medetomidine and its antagonist, atipamezole. Acta dopamine receptors: Implications for the binding of Vet Scand Suppl. 85: 29-37. catecholamines. J Med Chem. 49: 1706-19. Vet Scand Suppl. 85: 29-37. catecholamines. J Med Chem. 49: 1706-19. Wang CD, Buck MA and Fraser CM (1991). Site- Zhang YQ, Limbird LE (2004). Hetero-oligomers of Wang CD, Buck MA and Fraser CM (1991). Site- Zhang YQ, Limbird LE (2004). Hetero-oligomers of directed mutagenesis of 2A-adrenergic receptors: 2A-adrenergic and -opioid receptors do not lead to directed mutagenesis of 2A-adrenergic receptors: 2A-adrenergic and -opioid receptors do not lead to Identification of amino acids involved in ligand transactivation of G-proteins or altered endocytosis Identification of amino acids involved in ligand transactivation of G-proteins or altered endocytosis binding and receptor activation by agonists. Mol profiles. Biochem Soc Trans. 32: 856-60. binding and receptor activation by agonists. Mol profiles. Biochem Soc Trans. 32: 856-60. Pharmacol. 40: 168-79. Zhao MM, Hwa J and Perez DM (1996). Identification Pharmacol. 40: 168-79. Zhao MM, Hwa J and Perez DM (1996). Identification Warne T, Serrano-Vega MJ, Baker JG, of critical extracellular loop residues involved in 1- Warne T, Serrano-Vega MJ, Baker JG, of critical extracellular loop residues involved in 1- Moukhametzianov R, Edwards PC, Henderson R, et adrenergic receptor subtype-selective antagonist Moukhametzianov R, Edwards PC, Henderson R, et adrenergic receptor subtype-selective antagonist al. (2008). Structure of a 1-adrenergic G-protein- binding. Mol Pharmacol. 50: 1118-26. al. (2008). Structure of a 1-adrenergic G-protein- binding. Mol Pharmacol. 50: 1118-26. coupled receptor. Nature. 454: 486-91. Zurn A, Zabel U, Vilardaga JP, Schindelin H, Lohse coupled receptor. Nature. 454: 486-91. Zurn A, Zabel U, Vilardaga JP, Schindelin H, Lohse Weiss JM, Morgan PH, Lutz MW and Kenakin TP MJ and Hoffmann C (2009). Fluorescence resonance Weiss JM, Morgan PH, Lutz MW and Kenakin TP MJ and Hoffmann C (2009). Fluorescence resonance (1996). The cubic ternary complex receptor- energy transfer analysis of 2A-adrenergic receptor (1996). The cubic ternary complex receptor- energy transfer analysis of 2A-adrenergic receptor occupancy model. III. resurrecting efficacy. J Theor activation reveals distinct agonist-specific occupancy model. III. resurrecting efficacy. J Theor activation reveals distinct agonist-specific Biol. 181: 381-97. conformational changes. Mol Pharmacol. 75: 534- Biol. 181: 381-97. conformational changes. Mol Pharmacol. 75: 534- 41. 41. Wikberg-Matsson A, Simonsen U (2001). Potent 2A- Wikberg-Matsson A, Simonsen U (2001). Potent 2A- adrenoceptor-mediated vasoconstriction by adrenoceptor-mediated vasoconstriction by