INTERACTIONS BETWEEN THE OXYTOCIN AND
2-ADRENERGIC RECEPTORS IN A HUMAN MYOMETRIAL CELL LINE: FUNCTIONAL AND PHYSICAL ANALYSIS
By
PAULINA K. WRZAL
Department of Pharmacology and Therapeutics
McGill University
November 2011
A thesis submitted to the Faculty of Graduate Studies and Research of McGill University in partial fulfillment of the requirements for the degree of
DOCTOR OF PHILOSOPHY
Copyright Paulina K. Wrzal, 2011 i
ABSTRACT
The human myometrium is endowed with a vast array of receptors that transmit messages encoded in the external stimuli to the interior of the cell. These include the oxytocin receptor
(OTR) and the 2-adrenergic receptor (2AR), mediating uterine contractions and relaxation, respectively. These two receptors belong to the superfamily of G protein-coupled receptors
(GPCRs) and are important pharmacological targets because OTR antagonists and 2AR agonists are used to control pre-term uterine contractions. Although they have opposing effects on the myometrium, both receptors activate the MAP kinases ERK1/2, which have been implicated in uterine contractions and the onset of labour. However, the precise mechanisms by which the
OTR and the 2AR activate ERK1/2 in a human myometrial cells remains to be characterized.
Further, crosstalk between the 2AR and OTR signalling has been shown in the myometrium, but it is unclear what mechanisms underlie such crosstalk. In the present study, we describe a novel molecular mechanism for 2AR-mediated ERK1/2 activation in the human myometrial hTERT-
C3 cell line, which involves the activation of a pathway involving Gi-PI3kinase-PKC and Src.
We further show that this signalling cascade is dependent on the presence of the OTR. We also demonstrate physical interactions between OTR and 2AR using co-immunoprecipitation, bioluminescence resonance energy transfer (BRET) and protein-fragment complementation
(PCA) assays in HEK 293 cells. In the context of a receptor heterodimer, these interactions allow for allosteric control of one receptor partner by the other, shown here on the example of ERK1/2 activation in the hTERT-C3 cell line. This study illustrates the notion that formation of GPCR heterodimers can generate receptors with unique properties distinct from individual receptors.
Understanding how dimerization is arranged and controlled and more importantly the resulting signalling and pharmacology of such complexes will be crucial for future drug design. ii
RÉSUMÉ
Les cellules du myomètre humain expriment une vaste gamme de récepteurs transmettant un signal amorcé par des stimuli externes à l’intérieur des cellules. Parmi ces récepteurs, certains jouent un role important dans la contraction utérine, entre autre les récepteurs 2-adrénergiques
(2AR) et les récepteurs de l'oxytocine (OTR). Ces deux récepteurs qui appartiennent à la famille des récepteurs à sept domaines transmembranaires couplés aux protéines G (RCPG), représentent des cibles pharmacologiques potentielles. Ceci est essentiellement basé sur le fait que des antagonistes d’OTR et des agonistes du 2AR sont couramment utilisés pour réduire les contractions utérines en cas de risque d'accouchement prématuré. Malgré leurs effets opposés sur le myomètre, ces deux récepteurs activent les MAP kinases ERK1 et ERK2 qui jouent un rôle dans les contractions utérines. Les mécanismes exacts par lesquels ces deux récepteurs activent les MAP kinases demeurent mal définis. De plus, malgré le fait que des interactions entre les signaux mediés par les 2AR et ceux initiés par les OTR ont été démontrés dans les cellules du myomètre, les mécanismes de ces interactions demeurent méconnus. La présente étude décrit un nouveau mécanisme d’interaction et explore une nouvelle voie de signalisation par laquelle les
2AR activent la voie des MAP kinases ERK1/2 dans les lignée cellulaire hTERT-C3 du myomètre et impliquent une voie de signalisation Gi-PI3kinase-PKC et Src. De plus, nous démontrons que cette voie de signalisation nécessite la présence du OTR. Nous démontrons
également une interaction physique entre l'OTR et les 2AR dans les cellules HEK 293 par co- immunoprecipitation, essai de complémentation protéine-fragment (PCA) et grace à la technique de transfert d'énergie de résonance de bioluminescence (BRET). L’interaction entre l’OTR et le
2AR, dans un contexte de récepteur hétérodimère, permet le control allostérique de l’activation de ERK1/2 dans les lignée cellulaire hTERT-C3. Cette étude illustre l'idée que la formation iii d'hétérodimères de RCPG pourrait générer des récepteurs ayant des propriétés uniques et distinctes des récepteurs individuels. La compréhension des mécanismes de contrôle de la dimérisation, et du rôle que ceux-ci pourrait jouer dans la signalisation cellulaire, sera important pour le développement des médicaments futurs.
iv
AKNOWLEDGEMENTS
First and foremost I wish to thank my supervisors, Drs. Terry Hébert and Hans Zingg, for their patience, guidance and constant support. Not only are they outstanding scientists but they are also truly remarkable individuals and I am grateful to them for giving me the opportunity to challenge myself and develop as a student/scientist under their watchful eye. I am very thankful for the time they have dedicated to this project and the refining of this thesis.
I would also like to thank my graduate committe members Drs. Daniel Bernard, Stéphane
Laporte, Greg Miller and my academic advisor Dr. Derek Bowie for their helpful discussions and guidance throughout my studies.
I am thankful to all the past and present members of the research team for their encouragement, support, and technical advice, but most importantly for their friendship. I am particularly grateful to all my collaborators, whose names appear on my manuscripts, for their contributions.
I would like to extend a special word of gratitude to Dr. Dominic Devost for many valuable discussions as well as technical assistance and advice.
Much appreciation goes to Dr. Ahmed Bettaieb for his help in the translation of my abstract.
Finally, I extend my most sincere thanks to my parents, Grażyna and Włodzimierz Wrzal, for being the image of perseverance; to my husband, Patrik Balek, for teaching me the true meaning of dedication; to my sisters, Marta and Barbara for their love and laughter which make all things possible; and to my friends for their support.
v
DEDICATION
To Grażyna and Włodzimierz Wrzal & Halina and Stefan Chruściel vi
CONTRIBUTIONS OF THE AUTHORS This thesis is assembled in accordance with the regulations of the Faculty of Graduate Studies and Research, McGill University. It is written in a manuscript-based format, containing two original articles co-authored by myself and others. These articles have been arranged as
Chapters 2 and 3 in the same sequence as the experimental work that was conducted. The contribution of each author is as follows:
Chapter 2: Wrzal, P.K., Goupil, E., Laporte, S.A., Hébert, T.E., and Zingg, H.H. Functional interactions between OTR and 2AR: implications for ERK1/2 activation in human myometrial cells. Cellular Signalling. January 2012;24(1):333-341.
The candidate was responsible for most of the experimental work in the manuscript, including conceptualization and design of experiments, and rendering of the first draft of the manuscript. .
E. Goupil performed the confocal imaging experiments using the Zeiss LSM-510 Meta confocal microscope in the lab of S.A. Laporte. H.H. Zingg and T.E. Hébert supervised the project, participated in the design of experiments and wrote and edited different versions of the manuscript.
Chapter 3: Wrzal, P.K., Devost, D., Pétrin, D., Goupil, E., Iorio-Morin, C., Laporte, S.A.,
Zingg, H.H., and Hébert, T.E. Allosteric interactions between OTR and 2AR modulate ERK1/2 activation in human myometrial cells. Cellular Signalling. January 2012;24(1):342-350.
The candidate was responsible for most of the work in the manuscript, including rendering the first draft of the manuscript. D. Devost and C. Iorio-Morin performed the BRET experiments.
D. Pétrin performed the PCA experiments. E. Goupil performed the confocal imaging experiments using the Zeiss LSM-510 Meta confocal microscope in the lab of S.A. Laporte.
H.H. Zingg and T.E. Hébert supervised the project, participated in the design of experiments and wrote and edited different versions of the manuscript. vii
TABLE OF CONTENTS
ABSTRACT ...... i RÉSUMÉ ...... ii AKNOWLEDGEMENTS...... iv DEDICATION ...... v CONTRIBUTION OF AUTHORS...... vi TABLE OF CONTENTS ...... vii LIST OF ABBREVIATIONS ...... x LIST OF FIGURES ...... xii
CHAPTER 1 – INTRODUCTION AND LITERATURE OVERVIEW Overview ...... 1 Structural classification of GPCRs ...... 2 1.1.1. The rhodopsin receptor family of GPCRs (Class A) ...... 5
1.1.2. The secretin receptor family of GPCRs (Class B) ...... 5
1.1.3. The glutamate receptor family of GPCRs (Class C) ...... 6
Molecular mechanisms of GPCR activation ...... 8 1.1.4. Ligand binding to biogenic amine receptors ...... 8
1.1.5. Ligand binding to peptide receptors ...... 11
Heterotrimeric G proteins and their effectors ...... 17
1.1.6. The Gs family ...... 21
1.1.7. The Gi/o family ...... 22
1.1.8. The Gq family ...... 22
1.1.9. The G12 family ...... 23
1.4.5. G signalling ...... 24
1.5. Scaffolding proteins and GPCR signalling: the role of -arrestins ...... 27 1.5.1. -arrestins and GPCR dimerization ...... 27 viii
1.5.2. -arrestins and GPCR signalling ...... 30
1.6. Receptor activation models ...... 34 1.7. GPCR oligomerization ...... 38 1.7.1. GPCR heterodimerization interfaces in rhodopsin family receptors ...... 38
1.7.2. Alterations in receptor pharmacology ...... 39
1.7.3. Alterations in receptor signalling ...... 39
1.7.4. GPCR oligomerization and pathological states ...... 43
1.8. Allosteric modulation of GPCRs ...... 47
1.9. 2-adrenergic receptor signalling ...... 55
1.9.1. 2AR-mediated activation of ERK1/2 ...... 55
1.10. Oxytocin receptor signalling ...... 56 1.10.1. OTR-mediated activation of ERK1/2 ...... 57
1.11. Extracellular signal-regulated kinases as a common GPCR effector ...... 58 1.12. Human myometrium and contractility ...... 61 1.12.1. Normal labour ...... 61
1.12.2. Preterm labour and current tocoytic drugs ...... 65
1.13. Human myometrial cell model ...... 67 1.14. Project rationale and objectives ...... 70
CHAPTER 2
2.1. Preface ...... 75 2.2. Manuscript ...... 76
Allosteric interactions between the oxytocin receptor and the 2-adrenergic receptor in the modulation of ERK1/2 activation are mediated by heterodimerization.
2.3. Figures...... 99
CHAPTER 3 ix
3.1. Preface ...... 125 3.2. Manuscript ...... 126
Allosteric interactions between the oxytocin receptor and the 2-adrenergic receptor in the modulation of ERK1/2 activation are mediated by heterodimerization.
3.3. Figures...... 149
CHAPTER 4 – GENERAL DISCUSSION
4.1. Summary ...... 167 4.2. Conclusion ...... 188 4.3. References ...... 189
APPENDIX
Abstracts ...... 224 Publications ...... 225 Reprints
x
LIST OF ABBREVIATIONS
2AR : 2-adrenergic receptor
AC : Adenylyl cyclase
AT : Atenolol
BRET : Bioluminescence resonance energy transfer cAMP : Cyclic adenosine monophosphate
CFP : Cyan fluorescent protein
DNA : Deoxyribonucleic acid
ECD : Extracellular domain
ECL : Extracellular loop
EGF : Epithelial growth factor
EGFR : Epidermal growth factor receptor
ERK1/2 : Extracellular signal-regulated kinases 1 and 2
GDP : Guanosine diphosphate
GFP : Green fluorescent protein
GPCR : G protein-coupled receptor
GRK : G protein-coupled receptor kinase
GTP : Guanosine triphosphate
HEK : Human embryonic kidney cells hTERT-C3 : Human telomerase reverse transcriptase-clone3 myometrial cells
IL : Intracellular loop
ISO : Isoproterenol
MAPK : Mitogen-activated protein kinase
OT : Oxytocin xi
OTA : Oxytocin receptor antagonist
OTR : Oxytocin receptor
PCA : Protein fragment complementation assay
PI3K : Phosphoinositide 3-kinase
PKA : Protein kinase A
PKC : Protein kinase C
PLC : Phospholipase C
PRO : Propranolol
PTX : Pertussis toxin
Rluc : Renilla luciferase
RNA : Ribonucleic acid
TIM : Timolol
TM : Transmembrane
YFP : Yellow fluorescent protein
xii
LIST OF FIGURES
CHAPTER 1 - INTRODUCTION AND LITERATURE REVIEW
Figure 1.1. Three major subfamilies of GPCRs ...... 3
Figure 1.2. Crystal structure of the the β2AR associated with Gs ...... 13 Figure 1.3. Model of the human oxytocin receptor ...... 15 Figure 1.4. General mechanism of heterotrimeric G protein signalling ...... 19 Figure 1.5. Modulation of GPCR function by receptor heterodimerization ...... 45 Figure 1.6. The allosteric ternary complex model (ATCM; red) and the allosteric two- state model (ATSM; cubic scheme) of GPCRs ...... 49
Figure 1.7. Allosteric ligand regulation of GPCRs ...... 53
CHAPTER 2 – FUNCTIONAL INTERACTIONS BETWEEN OTR AND β2AR MODULATE ERK1/2 ACTIVATION IN HUMAN MYOMETRIAL CELLS
Figure 2.1. Kinetics of ERK1/2 activation by OTR and 2AR in human myometrial hTERTC3 cells ...... 99 Figure 2.2. Inhibition of Gi reduces both OT- and ISO-induced ERK1/2 activation in human myometrial hTERT-C3 cells ...... 101 Figure 2.3. Inhibition of PKC reduces OT- and ISO-induced ERK1/2 activation in human myometrial hTERT-C3 cells ...... 103 Figure 2.4. Specific inhibition of PKCζ isoform reduces OT- and ISO-induced ERK1/2 activation in human myometrial hTERT-C3 cells ...... 105 Figure 2.5. Effects of Gi inhibition on ISO-induced ERK1/2 activation in presence and absence of OTR in HEK293 cells ...... 107 Figure 2.6. Effects of PKC inhibition on ISO-induced ERK1/2 activation in absence and presence of OTR in HEK293 cells ...... 109
Figure 2.7. PKC recruitment to the plasma membrane following OTR and 2AR activation in HEK293 cells ...... 111 Figure 2.8. Effect of PI3K inhibition on OT- and ISO-induced ERK1/2 activity in hTERT-C3 cells ...... 113 Figure 2.9. Effect of PI3K inhibition on ISO-induced ERK1/2activation in presence and xiii
absence of OTR in HEK293 cells ...... 115 Figure 2.10. Effect of Src inhibition on OT- and ISO-induced ERK1/2 activity in hTERT-C3 cells ...... 117 Figure 2.11. Effect of EGFR inhibition on OT- and ISO-induced ERK1/2 activity in hTERT- C3 cells ...... 119
Figure 2.12. Effect of -arrestin siRNA on OTR-mediated ERK1/2 activation in human myometrial hTERTC3 cells ...... 121 Figure 2.13. Schematic representation of a model for the signaling pathway of ERK1/2
activation via β2AR and OTR in human myometrial hTERTC3 cells ...... 123
CHAPTER 3 – ALLOSTERIC INTERACTIONS BETWEEN OTR AND β2AR MODULATE ERK1/2 ACTIVATION IN HUMAN MYOMETRIAL CELLS
Figure 3.1. β2AR agonist isoproterenol (ISO) inhibits OTR-mediated ERK1/2 signalling in human myometrial cells ...... 149
Figure 3.2. Forskolin treatment attenuates OTR-mediated ERK1/2 activation in human myometrial hTERTC3 cells ...... 151
Figure 3.3. β2AR antagonist propranolol (PRO) inhibits OTR-mediated ERK1/2 signalling ...... 153
Figure 3.4. β2AR antagonist timolol (TIM) inhibits OTR-mediated ERK1/2 signalling ...... 155
Figure 3.5. Effect of β2AR inverse agonist atenolol (AT) on OTR-mediated ERK1/2 signalling ...... 157
Figure 3.6. OTR antagonist atosiban inhibits β2AR-mediated ERK1/2 signalling ...... 159
Figure 3.7. OTR antagonist OTA inhibits β2AR-mediated ERK1/2 signalling ...... 161
Figure 3.8. β2AR and OTR co-immunoprecipitation in HEK 293 cells ...... 163
Figure 3.9. β2AR and OTR interactions shown using BRET and PCA in HEK 293 cells ....165
CHAPTER 4 – GENERAL DISCUSSION
Figure 4.1. Effects of β2AR ligands on FPR- and AT1R-mediated ERK1/2 signalling ...... 179
Figure 4.2. Effects of OTR antagonists on FPR- and AT1R-mediated ERK1/2 signalling ...181 1
CHAPTER 1 – INTRODUCTION AND LITERATURE REVIEW
1.1. OVERVIEW
G protein-coupled receptors (GPCRs) constitute the largest family of cell surface receptors represented in the human genome [1]. Due to their ubiquitous distribution, primarily at the cell surface, and their contributions to multiple physiological and pathophysiological processes, GPCRs have been exploited as drug targets, accounting for approximately 50% of currently available therapeutics [2]. All members of this superfamily of receptors share a characteristic architecture, possessing seven transmembrane-spanning domains, which have evolved to accommodate the dual roles of ligand recognition and intracellular signal transduction.
Yet, despite their common architecture, GPCRs display enormous functional versatility, responding to such diverse stimuli as light, ions, odorant molecules, amines, lipids, nucleotides, peptides, and large proteins. In addition to being activated by agonists that bind the receptor, GPCRs can also demonstrate spontaneous, ligand-independent activation and certain receptor mutations can result in constitutive signalling. Canonically, GPCRs transduce their signals via heterotrimeric G proteins giving them their name.
GPCRs participate in a vast array of protein-ligand and protein-protein interactions and the list of membrane and intracellular coupling partners is constantly expanding, highlighting the remarkable conformational flexibility inherent in GPCRs. The ability of GPCRs to present discrete ligand and partner 2 interaction sites and assume multiple conformations highlights their potential for multiple modes of drug targeting.
1.2. STRUCTURAL CLASSIFICATION OF GPCRs
GPCRs are composed of seven membrane-spanning -helices, arranged in a counter-clockwise fashion, each of which contains approximately 25-35 mostly hydrophobic amino acids. Each receptor has an extracellular N-terminus and an intracellular C-terminus (Figure 1.1). Despite the many technical difficulties involved in solubilization and purification of GPCRs for X-ray crystallography, several groups have been able to generate high-resolution three-dimensional structures of several GPCRs among them the 2AR [3-11].
GPCRs are generally classified into 6 groups (A-F) based on their primary structures [12, 13]. This system of nomenclature covers all GPCRs, in both vertebrates and invertebrates. More recently, sequencing of the human genome resulted in a re-evaluation of the established classification to apply to human
GPCRs in particular. Thus it has been proposed that there are 5 main families of human GPCRs [14]. Three of these families correspond to the A-F classification system while others are not included. For the purposes of this thesis, only human
GPCRs will be discussed in detail.
3
Figure 1.1. Three major subfamilies of GPCRs. Schematic diagrams representing prototypical members of (A) family A, (B) family B, and (C) family
C receptors. Conserved residues are shown in white circles, cysteine residues connected by disulfide bridges are shown in black circles [15].
4
5
1.2.1. The Rhodopsin Receptor Family of GPCRs (Class A)
The rhodopsin family (class A) contains the largest number of GPCRs in the superfamily. There are several highly conserved residues in members of the rhodopsin family: the transmembrane DRY (helix 3) and NPxxY (helix 7) motifs
(Fig. 1.1a). These motifs are thought to be important for regulating the conformational states of the receptor as well as G protein activation. Mutations of several key residues in these regions in different GPCRs affect G protein coupling and signalling efficacy, and in some cases ligand binding as well [16-18]. It has also been suggested that these motifs might also be important for promiscuity in G protein coupling, that is, single receptors coupled to multiple G proteins [19]. A
D136N mutation in the E/DRY motif of the OTR no longer signalled through Gi but instead exhibited increased coupling to Gq, suggesting that D136 is required both to prevent constitutive activation and for Gi coupling [20].
Receptors in this family respond to biogenic amines, peptides, photons, and nucleosides. Generally, small molecule ligands bind within the transmembrane core of a given receptor, while larger peptide ligands bind to the
N-terminal extracellular domain [14]. Rhodopsin family receptors include: adrenergic, angiotensin, chemokine, dopamine, opioid, oxytocin, prostaglandin, serotonin, somatostatin, and vasopressin receptors, among others [14].
1.2.2. The Secretin Receptor Family of GPCRs (Class B)
The secretin receptor family (class B) includes approximately 20 receptors for hormones and neuropeptides. These receptors have a large N-terminal domain, 6 containing 6 conserved cysteine residues which form 3 disulfide bridges (Fig.
1.1b). The structure of the N-terminal domain has been determined by nuclear magnetic resonance (NMR) spectroscopy and was shown to be important for ligand binding to these receptors [21-24]. The current model for ligand binding to receptors in this family suggests a two-domain mechanism where the C-terminal domain of the ligand binds the extracellular domain (ECD) and the N-terminal domain of the ligand binds the extracellular loops and transmembrane -helices
(TM domain), leading to receptor activation [25]. Although the two-domain model of ligand binding is generalizable, multiple receptor-ligand contacts are present within each of the two interacting regions and these contacts differ between receptors [25]. The core of the receptor is made up of 2 anti-parellel - sheets, 3 disulfide bonds, and variable N-terminal -helix and multiple loop regions [26]. Interactions of peptide ligands with the receptor core are likely to be important for receptor activation and second messenger generation [21, 22, 27,
28] by facilitating receptor interaction with G proteins [29]. This family of receptors includes: calcitonin, glucagon, gastric inhibitory polypeptide, secretin and parathyroid hormone receptors [30].
1.2.3. The Glutamate Receptor Family of GPCRs (Class C)
The glutamate receptor family (class C) displays a large N-terminal domain, which contains between 300-600 amino acids (Fig. 1.1c). The N-terminal domain is thought to form two lobes separated by the ligand binding site, together forming a “Venus fly trap” module, which is joined to the TM region by a cysteine-rich 7 linker [31]. This cysteine-rich linker represents a flexible spacer which allows for the movement of the binding pocket with respect to the TM region [32].
Receptors belonging to the glutamate family contain a short and highly conserved third intracellular loop, which is believed to be crucial for G protein activation, and along with the first intracellular loop and the C-terminal domain controls G protein coupling efficiency [33-36]. Examples of class C receptors include metabotropic glutamate, calcium-sensing, GABA-B and taste receptors [14].
The remaining families and/or classes of GPCRs include: human adhesion and frizzled/taste2 receptors; and invertebrate GPCR class D of fungal pheromone receptors, class E of cAMP receptors, class F of archaebacterial opsins [12-14].
The human GPCR families of adhesion and frizzled/taste2 receptors are fairly small in size, each comprised of approximately 24 GPCRs [14]. As their name suggests the human adhesion receptors are believed to be involved in cell adhesion via motifs in their N-terminal domain [37, 38]. The frizzled receptors control cell fate, proliferation, and polarity during development by mediating signals from secreted glycoproteins termed Wnt, while the taste receptors mediate gustatory sensation [14].
It has been stated that the different GPCR families show limited structural similarities to each other or between species [39-41]. However, human GPCRs do share a common feature aside from seven trans-membrane domain architecture. All the human GPCR families possess two conserved cysteines, one between TM1 and 2 (in ECL2) and the second between TM 3 and 4 (in ECL3).
These cysteines are thought to form a disulfide bridge important for receptor structural integrity. The distance between these two residues varies with the 8 length of the receptor and the formation of the disulfide bridges does not appear to require a specific structural surrounding [14]. Although, these bridges have not been demonstrated in all the different GPCR families, they are functionally crucial for some receptors in the rhodopsin family [14]. An additional highly conserved cysteine is present in the C-terminal domain, generally palmitoylated, and appears to be important for receptor function in the human 2AR [42]. A mutation of the 2AR Cys341 to a glycine resulted in a nonpalmitoylated form of the receptor, which exhibited a reduction in the ability of the receptor to stimulate adenylyl cyclase (AC) in response to isoproterenol [42].
1.3. MOLECULAR MECHANISMS OF GPCR ACTIVATION
This thesis will focus on two GPCRs: the 2AR and the oxytocin receptor
(OTR). As mentioned earlier, these receptors belong to the rhodopsin receptor family (class A). This section will briefly describe receptor conformational changes and the process of activation that occurs upon ligand binding as pertaining to the receptors in question. Although both 2AR and OTR are classified as rhodopsin-like (class A) GPCRs, they certainly can be further subdivided into 2 different groups in terms of their ligand binding properties, biogenic amine and peptide receptors, respectively.
1.3.1. Ligand Binding to Biogenic Amine Receptors
The binding cleft of rhodopsin-like (class A) GPCRs such as 2AR is located deep within the hydrophobic core of the receptor molecule and is framed by 9 residues in transmembrane segments (TM) 3,5,6,7 [43]. Site-directed mutagenesis experiments indicate that certain residues in ECL2 are also important for binding of both agonists and antagonists to GPCRs in this subfamily such as the adenosine receptors [44]. A direct contact between ligands and residues in
ECL2 is possible, especially in light of the putative disulfide bridge formed between a cysteine in ECL2 and a cysteine near the N-terminal domain of TM3, which would place ECL2 in physical proximity to the ligand binding site [44].
The structure of the 2AR has been extensively studied over the last couple of decades. Several domains have been identified to be involved in ligand binding and G protein coupling through mutagenesis and chimeric receptor studies [45, 46]. More recently, the structure of 2AR was determined using X- ray crystallography (Figure 1.2) [4, 10, 11, 47, 48]. Taken together, these studies show that the ligand binding site of the 2AR is lined with hydrophobic residues, which bind the ligand via van der Waals contacts, and some polar residues, which form stronger interactions with the ligand. Among notable binding site residues are: Asp113, forming ion pairs with the amine group of ligands; Ser203, Ser204 and Ser207, which can form hydrogen bonds with the catechol ring of ligands; and Gln312, which can accept hydrogen bonds from -hydroxyl-amine motifs of ligands [49-51]. Ligand specificity of the 2AR has been attributed to residues
Asn312 and Asn293, where Asn312 has been shown to be important for binding affinity to antagonists and Asn293 was shown to be involved in recognition of agonist and receptor activation [52, 53]. A residue highly conserved among 10 rhodopsin-like family of GPCRs is Asp79, however, it has been shown to be more important for receptor activation than ligand binding per se [54].
There are several structural changes that occur upon ligand activation of
GPCRs within this subfamily. The recently obtained crystal structures of 2AR reveal small local structural changes in the binding pocket that are related to changes in an inter-helical hydrogen bond network involving residues in TM3, 5,
6 and 7 [10, 48]. The interaction of the agonist with Ser203 and Ser207 stabilizes a receptor conformation that shows a 2.1 Å inward movement of TM5 at Ser207 and a 1.4 Å movement of Pro211 relative to the inactive structure [10]. This leads to several conformational changes in residues Ile 121, Phe 282 and Trp 286 as well as rearrangements in TM5 and TM6 [48, 55]. The crystal structure of 2AR-
Gs protein complex reveals the interface of the receptor and G protein to be formed by ICL2, TM5 and TM6 of the 2AR, and by the 5-helix, N-1 junction, top of 3-strand and 4-helix of the GsRas domain [11]. Although there are no specific residues in GPCRs that confer G protein coupling specificity there are several residues in Gs that may play a role in receptor-Gs interactions. These residues include Phe139, Val217, Phe376, Cys379, Arg380, and Ile383 [11]. Phe139 of Gs subunit of the G protein usually corresponds to either a Phe or a Leu residues on most receptors which couple to Gs, however, these residues are more variable on receptors that couple to other G proteins [11].
Further, no direct interactions were detected between 2AR and the G subunit in the recent T4L-2AR-Gs-Nb35 complex crystal structure. Nevertheless, some 11 models of GPCR dimers propose that one protomer interacts with the G subunit and the other with the G subunit [56, 57].
1.3.2. Ligand Binding to Peptide Receptors
The OTR belongs to a subgroup of rhodopsin-like (class A) GPCR family which mediates the physiological roles of endogenous peptide hormones. In contrast to rhodopsin-like (class A) receptors that bind small molecule ligands, mutational mapping of ligand binding sites in peptide receptors has shown the importance of the N-terminal domain and the extracellular domains in binding the larger peptide ligands [15, 58-65]. Some peptide ligands may also have additional points of interactions in the TM domains, thus it is possible that these ligands may enter the TM binding crevice to different degrees depending on the ligand and receptor [15].
The OTR binds the agonist OT, mainly through interactions in the N- terminal domain (E1) and extracellular loops E2 and E3 of the receptor (Figure
1.3) [66, 67]. In particular, Arg34 in the N-terminal domain (E1) has been identified as essential for high-affinity OT binding using site-directed mutagenesis
[67]. Molecular modelling studies suggest the OT-binding site to be formed by the upper parts of the transmembrane helices 3-7 and regions E2 and E3 [68].
Interestingly, OT antagonists were not reported to bind the N-terminus of the
OTR [66, 67, 69-72]. Rather, it has been proposed that OTR antagonists bind to transmembrane helices 1 and 2, and the upper parts of helix 7, at sites distinct from the agonist-binding domain [70]. The LVK sequence in the upper part of 12
TM3 was identified as a binding site for non-selective antagonists [71].
Antagonists with high affinity for the OTR such as barusiban have been reported to interact between transmembrane domains 1and 2 of OTR [72]. The differences in binding profiles of agonists versus antagonists may suggest the presence of multiple conformational states the receptor can assume to elicit different signals.
However, the precise OTR residues that interact with the ligands still need to be determined. On the other hand, the RVSSVKL segment of the C-terminal domain of the third intracellular domain of OTR appears to be important for Gq coupling
[73-75]. The serines and valine residues in the RVSSVKL segment appear to be particularly important in the coupling of OTR to G proteins as mutations of these residues to alanine resulted in a compromised ability of the receptor to increase phosphatidylinositol (PI) turnover without affecting receptor plasma membrane expression and ligand affinity [74, 75]. 13
Figure 1.2. Crystal structure of the the β2AR associated with Gs. Side view showing the β2AR (green) bound to an agonist (yellow spheres) and engaged in interactions with Gs (orange). Gs along with G (cyan) and G (purple) constitute the heterotrimeric G protein Gs. The crystallization of the receptor is facilitated by the presence of the Gs-binding nanobody Nb35 (red) and the T4 lysozyme (magenta) which is fused to the N-terminal domain of the β2AR [11]. 14
15
Figure 1.3. Model of the human oxytocin receptor. (A) Side view of the receptor structure. The rectangle marks the section shown in panel B from the top.
(B) Extracellular view of the marked section. The transmembrane helices are arranged in an anti-clockwise fashion around a central pocket site.
Transmembrane helices 4 and 5 are believed to be involved in receptor homodimerization. Agonists and antagonists bind at different sites around the central pocket site. The black dot represents Arg34, which is involved in oxytocin binding to the receptor [76].
16
17
1.4. HETEROTRIMERIC G PROTEINS AND THEIR EFFECTORS
One of the most fascinating subjects, which preoccupies GPCR researchers, is to understand the mechanisms by which the receptor, ligand-bound or not, signals via activation of heterotrimeric guanine nucleotide-binding proteins
(G proteins) to the cell interior. The extracellular portion of the receptor binds ligand, while the intracellular surface is crucial for G protein activation.
The standard model for transmission of signals inside the cell by way of G protein activation is cyclic and reversible. Heterotrimeric G proteins are composed of , , and subunits. The inactive G protein heterotrimer is bound to guanosine diphosphate (GDP) via the G subunit. Upon ligand binding to the
GPCR, the active receptor acts as a guanine nucleotide exchange factor (GEF), and via induction of conformational changes in the G protein, guanosine triphosphate (GTP) is exchanged for GDP on the G subunit simply due to the higher concentrations of GTP in the cell. This then promotes G protein activation and the GTP-G subunit complex dissociates from the G dimer. Both the G and the G go on to regulate effectors and/or generate second messengers. The
G subunit then hydrolyzes GTP to GDP via an intrinsic GTPase domain. This promotes re-association of the heterotrimeric G protein terminating signalling
(Figure 1.4) [77].
In humans, at least 15 distinct G protein subunits, 5 different and 12 different subunits have been identified, with and subunits considered to form single functional units [78]. Although some interactions are more favoured than 18 others, the variety of combinations formed confers increased diversity and potential specificity of signalling.
G proteins interact specifically with various effectors such as ion channels and enzymes, the latter in turn regulate the production of second messengers.
GPCRs are described in accordance to the G subunit they are coupled to.
Furthermore, a receptor may couple to different G proteins increasing the diversity of signalling responses [79]. This also increases the level of complexity in GPCR signalling, whereby differential signalling can be obtained by the same
GPCR in the same cell. The coupling of a receptor to different G proteins represents just one level of such complexity. Parameters such as the time and location of signals may also play a role in this diversity, as could the interactions of the receptor with other proteins inside the cell.
There are 4 main families of heterotrimeric G proteins, based on the sequence homology of the subunits: Gs, Gi, Gq, and G12 [77]. The main members of each family couple to specific effectors thus conferring a specific signal. OTR has been shown to couple to both Gq and Gi [80-83]. Studies of the β2AR have shown that it mainly couples to Gs, however, additional studies have demonstrated that the β2AR can undergo PKA-dependent phosphorylation leading to activation of Gi [84-86]. This type of G protein “switching” has also been reported for other AR subtypes as well as the prostacyclin receptor [87-90]. 19
Figure 1.4. General mechanism of heterotrimeric G protein signalling. The conversion of a G protein heterotrimer from the inactive, guanosine diphosphate
(GDP)-bound, state to the active, guanosine triphosphate (GTP)-bound, state occurs via an interaction with a guanine nucleotide exchange factor (GEF), the most common of which are GPCRs. Upon ligand binding to the GPCR, the receptor induces conformational changes in the G protein resulting in the exchange of GTP for GDP on the G subunit. Subsequent conformational changes lead to the separation of the GTP-bound G subunit from the G dimer.
Both elements then go on to regulate effectors and/or generate second messengers. The intrinsic GTPase activity of the G subunit hydrolyzes GTP to
GDP, facilitating the re-association of the G-GDP and the G dimer, which terminates signalling [77]. 20
21
1.4.1 The Gs Family
The Gs family is known as the adenylyl cyclase (AC) stimulatory G proteins and includes Gs and Golf. Members of this family of G proteins signal via the effector adenylyl cyclase (AC), a family of enzymes spanning the plasma membrane, which catalyze the production of cyclic adenosine monophosphate
(cAMP) from adenosine triphosphate (ATP) [91]. The subunit of Gs also acts as a substrate for mono-ADP ribosylation catalyzed by cholera toxin, an ADP- ribosyl transferease which covalently attaches ADP-ribose to Gs. This results in elevated and sustained levels of cAMP, as the cholera toxin-modified Gs cannot hydrolyze GTP back to GDP [77]. Activation of Gs leads to the activation of cAMP-dependent protein kinase A (PKA), a serine/threonine kinase, which consists of 2 regulatory and 2 catalytic subunits. PKA is activated upon the binding of cAMP molecules to each regulatory subunit thus releasing the active catalytic subunits [92]. PKA targets include metabolic enzymes, ion channels and transcriptional regulators.
Golf has been shown to be involved in olfaction, being activated upon odorant binding of an olfactory GPCR. Much like Gs, activation of Golf activates adenylyl cyclase, increasing levels of cAMP which can then bind cyclic nucleotide-gated channels, causing them to open. An influx of cations through the channel leads to the generation of action potentials, which travel along the axon of the primary neuron to signal to the brain [93].
22
1.4.2. The Gi/o Family
In contrast to the Gs family of G proteins, the members of the Gi/o family, by definition, inhibit AC. This family of G proteins includes Gi1-3, Go, Gg, Gt1-
2, and Gz and all members except Gz are sensitive to and inactivated by pertussis toxin [77]. Pertussis toxin is an ADP-ribosyl-transferase, much like cholera toxin, but instead of activating Gi proteins it inactivates them [94].
Gi1-3 are products of different genes [95]. Go is the most abundant G protein in the mammalian brain and has been reported to be involved in neurite outgrowth [96]. It is the product of a gene that can undergo alternative splicing to produce at least 2 different proteins, Go1 and Go2 [77]. Gg, Gt1, and Gt2 have been shown to be involved in sensory responses, with Gg mediating signals from taste receptors and Gt1, and Gt2 involved in mediating signals in the retina
[77]. Finally, Gz was found to couple to dopamine D2-like receptors and is expressed in platelets, adrenal chromaffin cells, neurons and neurosecretory cells
[97].
1.4.3. The Gq Family
The Gq family includes Gq, G11, G14 and G15/16. Members of this family are insensitive to either pertussis or cholera toxins. Gq and G11 are ubiquitously expressed while other family members have more restricted tissue expression profiles [77]. The Gq family members activate phospholipase C
(PLC), which hydrolyzes phosphotidylinositol 4, 5-bisphosphate (PIP2) into diacylglycerol (DAG) and inositol 1, 4, 5-trisphosphate (IP3). IP3 is a second 23 messenger which mobilizes calcium (Ca2+) by binding to its receptor on the endoplasmic reticulum, causing the opening of Ca2+ channels and increasing levels of cytosolic Ca2+. Consequently, this activates Ca2+-sensitive enzymes such as protein kinase C (PKC) isoforms. PKC can also be activated by DAG [98].
PLC is a soluble protein that is localized mainly in the cytosol and translocates to the plasma membrane. There are four members of the PLC family, PLC1-4 and they exhibit different affinities for the Gq and subunits.
PLC1 and 4 exhibit higher affinity for the Gq subunit, while PLC2 and 3 are more sensitive to G subunits [98, 99]. This may represent a mechanism for achieving specificity in GPCR signalling.
1.4.4. The G12 Family
The G12 family has 2 members: G12 and G13, which have been reported to be involved in regulation of cell morphology and proliferation and are expressed ubiquitously in most tissues [100]. Constitutively active (GTPase deficient) forms of these G proteins have been associated with stimulation of DNA synthesis, cell proliferation and malignant transformation; formation of stress fibers and focal adhesions resulting in changes in cell shape; activation of phospholipase C-
(PLC-); as well as activation of Rho, a small monomeric G protein involved in regulation of actin dynamics [101-105]. Although a number of effectors have been defined for G12 and G13, there is no one effector that is unique to the activation of these G proteins, and an overlap with other G protein families exists in the regulation of effectors. For example, Rho can be activated by Gq- as well 24
as G12- and G13-coupled receptors, however, agonists such as thrombin activate
Rho through G12 or G13 in mouse fibroblasts at concentrations much lower than those required for Gq-mediated Rho activation [106]. Further, the EC50 values for agonists such as thrombin and thromboxane A2 to induce changes in platelet shape and aggregation was found to be increased upon deletion of G13 [107].
1.4.5. G Signalling
G subunits have been shown to participate in signalling downstream of activated GPCRs by modulating many effectors by direct interaction, which in many cases are also regulated by G subunits [108]. However, originally, the G subunit was mainly thought to terminate G signalling by binding the activated
G subunit and anchoring it to the plasma membrane [109]. As illustrated in
Figure 1.4, upon GPCR activation by agonist the G protein converts from an inactive GDP-bound state to an active GTP-bound state. Conformational changes that occur during this conversion result in dissociation of the G and the G subunits of the G protein; at which point each element goes on to regulate effector proteins leading to signal transduction. The G subunit has GTPase activity and hydrolyses the terminal phosphate of the bound GTP, thus, terminating its function. The G subunit, having high affinity for the GDP- bound G subunit reassociates with it and terminates effector regulation [77]. The first evidence that the G complex played a direct role in GPCR signalling was reported in 1987 where it was found that G purified from bovine brain could activate muscarinic acetylcholine receptor-gated potassium channels (an inwardly rectifying 25 potassium channel) in chick embryonic atrial cells [110]. Since the discovery of the first G effector, Kir3 potassium channels, many others have been identified including PLC, some AC isoforms, voltage-gated calcium channels, phosphoinositide 3-kinase (PI3K), mitogen activated protein kinases (MAPK), protein kinase D (PKD), adipocyte enhancer-binding protein (AEBP1), RGS7 binding protein (R7BP), nucleaosite diphosphate kinase B (NDPK-B), receptor for activated C kinase (RACK1), activators for G protein signalling (AGS), glucocorticoid receptor (GR), histone deacetylase 5 (HDAC5), t-complex testis expressed-1(Tctex-1), and the SNAP (soluble NSF attachment protein) receptor
(SNARE) [110-127].
In mammals, there are 5 G and 12 G subunits and since they form a complex under physiological conditions the activity of the G complex is derived from the identity of both - and -subtypes [128]. The number of possible , , and subunit combinations is considerable, however, there is little information about which actually do form in vivo. Studies of subunit expression in different physiological systems have provided some information on the probable subunit combinations. Briefly, the G1-subtype interacts with most G - subtypes (preference for G12); the G2-subtype was shown not to interact with
G1 and G11; the G3-subtype can interact with G2, G5, G8cone, G12; and the
G4-subtype was shown to interact with all G-subtypes (preference for G5 and
G12) [129-136]. The G5-subtype was shown to interact with G2, G4, G7, and unlike the other G-subtypes it was shown to bind RGS proteins [132, 137, 138].
Some studies suggest that chaperone proteins such as cytosolic chaperonin 26 complex (CCT) [139] and dopamine receptor-interacting protein 78 (DRiP78) may play a role in the specificity of G complex formation [140]. Other studies propose that the possible combinations of these subunits in vivo may be restricted by expression and regulated at the levels of transcription, translation and post- translation processing [141-145]. The presence of the receptor may be an additional factor influencing the interactions between specific G and G subunits [146, 147]. For example, the G3 and G4 subtypes were shown to poorly interact with Gs and Gq, respectively [147]. However, G complexes containing the G3 and G4 subtypes were found to couple to Gs and Gq, respectively, in the presence of a receptor [134, 148]. Thus, the specificity of
G complex formation with the different , and subtypes suggests a role for these proteins in modulation of G protein signalling.
The role of G and the G subunits of G proteins in regulating effector systems has often been studied under the premise that the G protein complex separates or dissociates upon GPCR activation. However, there are an increasing number of reports that have challenged this concept. Several studies have shown that G protein dissociation is not necessary for activation of effectors and in some cases the binding between the G and G subunits may be required for signalling [149-152]. In such cases, instead of dissociation, a conformational rearrangement of the complex is thought to occur which exposes previously buried interfaces between the G and G subunits that can then interact with their respective effectors [153, 154]. This model termed the “clamshell” model predicts that the G subunits are not interchangeable, as they might be in the 27 dissociation model, since they remain closely associated with the activated receptor [155]. Further, the cellular mobility of the activated G subunit by its susceptibility to undergo dissociation from the G subunit upon activation is thought to be influenced by the associated - and -subtypes in the G protein complex and may represent a mechanism for signalling specificity distinct from the receptor itself [156-159].
1.5. SCAFFOLDING PROTEINS AND GPCR SIGNALLING: THE ROLE
OF β-ARRESTINS
In addition to interacting with G proteins, GPCRs can interact directly to modulate the activity of a growing list of other signalling partners [160]. Among these are the β-arrestins, which have received a considerable amount of attention in the recent years. Initially, β-arrestins were only thought to play a role in desensitization of GPCRs, terminating G protein coupling and mediating endocytosis of the receptor, however, they are now recognized as scaffolding proteins that can engage a diversity of signalling molecules [161].
1.5.1. β-Arrestins and GPCR Desensitization
There are 4 known arrestins. Arrestin-1, known as visual arrestin and arrestin-4, known as cone arrestin, are restricted to the visual system [162]. The other 2 arrestins, arrestin-2 (β-arrestin-1) and -3 (β-arrestin-2), are ubiquitously expressed outside the retina and interact with the majority of GPCRs [163]. 28
GPCRs undergo phosphorylation-dependent desensitization that limits the efficiency of the receptor-G protein coupling [164]. Phosphorylation of specific residues of the intracellular C-terminal domains of GPCRs by protein kinases such as protein kinase A (PKA) or C (PKC) results in heterologous desensitization that is independent of ligand occupancy of the receptor [162].
Phosphorylation of agonist-occupied GPCRs by G protein-coupled receptor kinases (GRKs) on serine or threonine residues within the C-terminal domain or third intracellular loop leads to homologous desensitization via direct interactions with β-arrestins [163, 165]. β-arrestins bind GRK-phosphorylated receptors and prevent further interactions with the G protein. The polar core of arrestin, located between the two globular domains of the protein, interacts directly with the GRK- phosphorylated residues on the receptor C-terminal domain. This exposes the inner surfaces of the globular domains and allows for their interaction with the receptor, which leads to conformational changes within arrestin [166, 167].
Arrestin binding stabilizes a high agonist affinity state of the receptor, similar to that which exists between agonist-receptor-G protein in the ternary complex
[168]. The 2 non-visual arrestins, β-arrestin-1 and -2, decrease G protein signalling even further by prompting the receptor to undergo clathrin-dependent endocytosis [169]. The C-terminal domain of β-arrestins, exposed after interactions with the receptor, can now bind to the clathrin heavy chain and the 2 adaptin subunit of the adaptor protein (AP)-2 complex [170-172]. Consequently, the binding between clathrin and AP-2 complex leads to the clustering of arrestin- bound receptors in clathrin-coated pits, which are pinched off the plasma 29 membrane by the motor protein, dynamin [173]. This process removes receptors from the cell surface and leads to a decreased response to subsequent stimuli thus resulting in GPCR desensitization. Nevertheless, some GPCRs have a higher affinity for β-arrestins than others, which results in differences in a more stable receptor-arrestin interaction after internalization [174]. The majority of GPCRs can be divided into 2 classes based on their interactions with β-arrestins. GPCRs that exhibit higher affinity for β-arrestin-2 than -1 form transient receptor-arrestin complexes that dissociate soon after internalization. Such GPCRs, like the β2AR, resensitize and recycle rapidly to the plasma membrane, most likely directly from the early endosomal compartment after being dephosphorylated [175-177].
Meanwhile, GPCRs that have equivalent affinities for both β-arrestins form more stable receptor-arrestin complexes that remain intact as the receptor undergoes endosomal sorting [178, 179]. These receptors, such as the V2 vasopressin receptor (V2R), are sequestered into endosomes and tend to recycle slowly through the perinuclear recycling endosomal compartment or are transported to lysosomes for degradation [180, 181]. The OTR was reported to belong to this second class of GPCRs, which stably associate with β-arrestins via serine clusters in the C-terminal domain of these receptors [182]. Although the OTR was found to recycle to the plasma membrane and not to be degraded, its association with β- arrestin-2 was found to play a role in determining the rate of OTR recycling [151].
Nevertheless, prolonged agonist exposure of GPCRs may alter the trafficking profile of the receptor. During prolonged exposure to agonist, the β2AR was found to utilize the slow recycling pathway through the perinuclear recycling endosomal compartment and with time some β2ARs were redirected to lysosomes 30 for degradation [183-185]. In human myometrial cells, prolonged exposure of the
OTR to OT was found to result in slow desensitization of the OTR and was accompanied by down-regulation of the OTR mRNA [186, 187]. Thus, the fate of the receptor after desensitization does not simply depend on interactions with
β-arrestins and involves other factors such as Ras-like small GTPases termed Rab
GTPases, which are implicated in regulation of intracellular trafficking [188].
Rab GTPases are involved in nearly all aspects of trafficking from vesicle budding to its motility and fusion. Different Rab GTPases are localized to distinct organelles and their microdomains to mediate trafficking to specific destinations such as the plasma membrane, lysosome or the trans-Golgi network [128, 155,
188]. The human OTR, when internalized has been shown to localize in vesicles containing Rab4 and Rab5 and recycle via short cycle to the cell surface after agonist application, however, this can be modulated by the association of OTR with β-arrestin-2 [151].
1.5.2. β-Arrestins and GPCR Signalling
The first experimental evidence showing the role of β-arrestins as mediators of GPCR signalling was reported for the β2AR-induced activation of extracellular signal regulated kinases-1 and -2 (ERK1/2) [189]. β-arrestins were shown to serve as scaffolds for the β2AR activation of ERK1/2 by forming complexes with Src, linking the receptor to downstream signalling pathways [190-
192]. Since then, much evidence has accumulated to show an important role for 31
-arrestins in GPCR signal transduction, linking activated GPCRs to numerous effector pathways with a number of proteins that bind to β-arrestins and are recruited to agonist-occupied GPCRs. These include: components of the ERK1/2 and c-Jun N-terminal kinase 3 (JNK3) MAP kinase cascades [193-195], the E3 ubiquitin ligase, Mdm2 [196], cAMP phosphodiesterases (PDE), PDE4D3/5
[197], diacylglycerol kinase [198], the inhibitor of nuclear factor (NF)B, IB
[199], the Ral-GDP dissociation stimulator (GDS), Ral-GDS [200], and the
Ser/Thr protein phosphatase (PP)2A [201]. Many of these effectors are not known to be regulated by heterotrimeric G protein subunits, suggesting that perhaps GPCR-β-arrestin signalling functions in parallel to GPCR-G protein signalling [162]. The role of β-arrestins in signalling is further supported by a recent proteomic screen which identified approximately 337 different proteins that were shown to interact with either β-arrestin-1 or -2 [202].
The binding of β-arrestins to the receptor uncouples it from the G protein, thus at least in theory G protein-dependent and β-arrestin-dependent signals should be mutually exclusive. β-arrestin binding is thought to “switch” off G protein- dependent signalling and thus creates two temporally distinct signalling modes. A comparison of the time course of G protein-dependent and β-arrestin-dependent
ERK1/2 activation for several GPCRs, including 2AR, shows that the onset of the β-arrestin-dependent ERK1/2 activation corresponds to the end of the G protein-dependent signalling and persists as receptors internalize [203-206]. - arrestin-dependent GPCR signalling is also spatially distinct as receptors that form stable complexes with -arrestins, such as the angiotensin AT1A and 32 vasopressin V2 receptors, activate a pool of ERK1/2 that accumulates at early endosomes alongside the receptor and does not translocate to the nucleus to elicit transcriptional responses, in contrast to ERK1/2 activated by G protein-dependent pathways [192, 193, 195, 207]. However, GPCRs that form transient complexes with -arrestins, such as the 2AR or the LPA receptor, which dissociate upon internalization of the receptor, do not constrain ERK1/2 activity to the endosome generating a pool of ERK1/2 that can activate gene transcription [204, 206]. In fact, exchange of the C-terminal domain of the V2R for that of the 2AR, converting receptor-β-arrestin interactions from stable to transient, resulted in vasopressin-stimulated ERK1/2 able to translocate to the nucleus and promote cell proliferation [208]. Thus stability of the receptor- β-arrestin complex, at least partially determines the action of activated ERK1/2.
Nevertheless, some -arrestin-mediated signals were found to oppose G protein signalling. For instance, the D2 dopamine receptor (D2DR), a Gi-coupled receptor that normally activates PI3K and Akt, elicits some of its effects by dephosphorylating and inactivating Akt through activation of the phosphatase
PP2A. A recent study showed -arrestin-2 is involved in scaffolding both Akt and PP2A and that D2R-mediated Akt dephosphorylation requires -arrestin-2
[201]. On the other hand, some GPCRs utilize -arrestin in a way that is synergistic with G protein activation. One such example is the 2AR, which responds to catecholamines by activating both Gs-PKA and -arrestin-mediated signalling pathways, triggering DNA damage and suppression of p53 levels 33 respectively, thus synergistically leading to the accumulation of DNA damage in response to chronic stress [209].
Furthermore, a number of β-arrestin-mediated signals were shown not to require G protein activity at all. Findings obtained using mutants functionally uncoupled from G proteins suggest that β-arrestin-dependent ERK1/2 activation via the 2AR, AT1R, and the PTH1R occurs independently of G protein signalling, however, it remains to be clarified whether such signalling mechanisms are physiologically relevant [205, 206, 210-212]. Isoform-selective silencing of GRKs showed that phosphorylation of the AT1 angiotensin II receptor and V2R by GRK2 and GRK3 was responsible for β-arrestin-dependent desensitization of these receptors, while phosphorylation by GRK5 and GRK6 was exclusively responsible for β-arrestin-dependent ERK1/2 activation [213,
214]. The exact mechanism of β-arrestin recruitment in the absence of G protein activation is not yet well understood. Experiments where a G protein-uncoupled neurokinin NK-1 receptor-β-arrestin-1 chimera was overexpressed in a heterologous cell system suggest that β-arrestins may simply act as scaffolds, tethered to an agonist-bound GPCR, summoning together the appropriate signalling components resulting in the activation of a transduction pathway, such as the ERK1/2 cascade [215].
Although β-arrestin-mediated signalling cascades may appear to be G protein-independent, under physiological conditions they follow G protein activation and thus these two signalling modes may be inherently linked in vivo
[162]. Studies using β-arrestin knockout mice cannot separate the loss of β- 34 arrestin-dependent signalling from the loss of GPCR desensitization and sequestration because of the crucial role that β-arrestins play in receptor desensitization [216-218]. Further, they cannot determine whether prior physical association with G protein is required for -arrestin-dependent signalling [219,
220]. Finally, studies using biased agonists selective for the β-arrestin-dependent signalling pathways such as the AR ligands ICI 118551, propranolol and carvedilol act as partial inverse agonists with respect to Gs-dependent activation of AC, but function as partial agonists for the activation of ERK1/2 and were shown to recruit β-arrestins [210, 221, 222]. These biased ligands block G protein signalling in response to endogenous ligands. This may occur through the stabilization of different conformations of the receptor which in turn may facilitate the recruitment of β-arrestins and subsequent internalization of the signalling complex resulting in qualitatively different types of signalling.
1.6. RECEPTOR ACTIVATION MODELS
The effect of a ligand on a signalling behaviour of a receptor, that is the degree of receptor activation, is described by efficacy. Ligands can be grouped into (1) agonists, which activate the receptor; (2) partial agonists, which produce submaximal activity; and (3) antagonists, which block the binding of other ligands to the receptor but do not alter its basal activity. Some GPCRs exhibit agonist- independent activity that can be suppressed by ligands termed inverse agonists
[223]. GPCRs can couple to more than one G protein, signal through G protein- 35 independent pathways, undergo complex regulatory processes, and be allosterically modulated by other proteins. They can mediate a wide array of signalling events and exhibit regulatory behaviour that can be modulated in a ligand-specific manner. Given this complexity, the functional versatility of most
GPCRs cannot be explained by a simple on-off switch model of receptor activation, as is possible with rhodopsin, for example. There are increasing reports showing that GPCRs are dynamic proteins and that different ligands can stabilize specific receptor conformations [224, 225]. Several studies have shown that single mutations in GPCRs can lead to constitutive active mutants (CAMs) that produce a signal in the absence of agonist. The first of such mutations was an amino acid substitution at Ala293 of the α1BAR [226]. This region of GPCRs, near the intracellular region of TM VI, was shown to be important for constitutive activity of the receptor as substitutions using each of the 19 amino acids resulted in variable levels of agonist-independent activity [227]. Since then a large number of CAMs have been identified indicating that GPCR activation can be triggered by changes in structure to several different regions of the receptor, and that some of these regions cannot be easily linked to regions of agonist binding or
G protein interactions [228]. There are several mutations in the 2AR that lead to increased constitutive activity leading to increased agonist-independent AC activity [229, 230]. Further constitutively active mutants of 2AR were found not only to exhibit constitutive activity but also constitutive desensitization and downregulation [231]. The OTR CAM R137A and the constitutively inactive mutant D85A were generated to study the role of conserved Arg137 (DRY motif) 36 and Asp85 in OTR activation [68, 232]. The data suggests that the receptor exists in numerous distinct active states, however, it is not known how many [233]. It has been proposed that different ligands selectively stabilize certain receptor conformations [234]. Some of these conformation states may activate G proteins; others may activate β-arrestins while others still may produce no cellular signalling. In their “resting” state, the majority of receptors in a population are thought to be inactive. Upon stimulation with a given ligand, the ligand can stabilize certain states thus resulting, for instance, in predominantly G protein- dependent responses [233].
For example, switches in conformation upon activation by an agonist can be different depending on the different ligand efficacies [235]. Differences in EC loop conformations were detected upon agonist vs. inverse agonist binding in the
2AR [236]. Further, kinetic FRET experiments showed that the efficacy of an agonist was dependent on the rate of conformational change in the GPCR [237].
Hence the concept of biased signalling emerged whereby stabilization of a distinct activation state of a receptor by a particular ligand could lead to the activation of different cellular signalling profiles [233]. Experiments on the selectivity or efficacy of different 2AR ligands in the activation of distinct signalling pathways have shown that a variety of traditional 2AR antagonists such as propranolol are able to modulate both cAMP and ERK1/2 activity with variable efficacy [238].
More recent studies assessed the ability of ligands to selectively recruit β-arrestin and activate ERK1/2. For example, carvedilol demonstrated a unique signalling profile of negative efficacy for Gs-dependent AC activation, while weakly 37 promoting β-arrestin recruitment, internalization and activation of ERK1/2 signalling by the 2AR [221]. Carvedilol and alprenolol were both shown to promote 1AR-mediated EGFR transactivation that occurs through recruitment and activation of β-arrestin [239]. The OTR antagonist atosiban has been shown to selectively activate Gi-mediated ERK1/2 while blocking the Gq-mediated IP3 production [240]. Most recently, an investigation that used a quantitative mass spectroscopy method to assess the dynamic conformational changes in the 2AR showed that even ligands that exhibit similar signalling profiles can stabilize distinct conformations in the 2AR [241].
Thus, GPCRs are more than just on and off switches for single signalling pathways, they represent signalling hubs that can regulate alternative subsets of signalling modes, depending on the receptor conformation stabilized by a specific ligand. The 2AR engages both the stimulatory and inhibitory G proteins Gs and
Gi to control adenylyl cyclase and MAPK activity [84]. Further, the 2AR can recruit β-arrestin and lead to G protein-independent MAPK activation [242].
GPCRs can act as a “scaffolds” or hubs directing intracellular signalling.
Conceptually a scaffold protein is like a platform composed of multiple interaction domains, each able to interact with and coordinate other proteins
[243]. GPCRs like 2AR are just one example of such scaffold proteins or signalling hubs. Naturally, GPCR oligomers, heterooligomers in particular, can also function as signalling hubs, further expanding the repertoire of pathway responses. 38
1.7. GPCR OLIGOMERIZATION
Initially, GPCRs were thought to function as monomeric structural units coupling to a single G protein heterotrimer. In the last 15 years or so, there have been an increasing number of reports describing GPCR homo- and hetero-dimers as well as higher order oligomers (reviewed in [244-246]). Nevertheless, dimerization is not necessarily required for functional coupling of the GPCR to G proteins [247]. Dimerization does, however, seem to be important for intracellular trafficking and more recently it has been shown to regulate the function and properties of several GPCRs [244]. Further, it has also been implicated in pathological events [248-251].
1.7.1. GPCR Heterodimerization Interfaces for Rhodopsin Family
Receptors
Most of the research on class A GPCR dimerization suggests that receptor oligomers are formed through non-covalent interactions of TM domains.
Formation of oligomeric complexes has been primarily associated with residues in
TM1 and TM4 on individual receptors [252, 253]. Furthermore, TM5 as well as the cytoplasmic helical domain (H8) have also been reported to be important for oligomer formation in class A GPCRs [251, 254]. A putative dimerization motif,
276GxxxGxxxL284, initially identified in TM6 of the 2AR was shown to be important for homodimerization of the receptor [156]. Mutations in this motif resulted in ER retention of the receptor, suggesting a role for homodimerization in
ER export and cell surface targeting of the receptor [255]. 39
1.7.2. Alterations in Receptor Pharmacology
Heterodimerization of GPCRs was found to modulate the binding properties of several receptor pairs [256-258]. This is thought to be a consequence of conformational changes in the binding pockets of associated receptors resulting from receptor oligomerization. For instance, a decrease in affinity and potency of agonists was observed for - and - opioid receptors
(DOR-MOR) and somatostatin sst2A-sst3 receptors [259, 260]. Selective antagonists of the DOR were shown to modulate the pharmacology of the MOR by increasing the number of binding sites and enhancing the receptor signalling
[256, 261]. Another example of heterodimerization affecting the affinity of partner receptor was observed for 2AR ligands in the case of 1AR-2AR pair
[262]. More recently, the formation of a heterooligomeric complex of chemokine receptors CCR2, CCR5, and CXCR4, was observed to be responsible for negative binding cooperativity between the binding pockets of these receptors [263].
1.7.3. Alterations in Receptor Signalling
Several reports have shown that heterodimerization results in changes in G protein coupling, signalling and receptor internalization. For example, a change in G protein coupling was observed for the 2AAR/MOR heterodimer, where morphine binding inhibits norepinephrine-mediated activation of Gi and the
ERK signalling pathway [261]. It has been suggested that heterodimerization induces a change in receptor conformation which results in the generation of new properties of the receptor [264]. Further, a shift in the type of G protein coupled 40 to the receptor can also change as a result of heterodimerization. In the case of the heterodimer between DOR and sensory neuron-specific receptor-4, a switch from Gi/o to Gq-mediated signalling has been observed [265].
Heterodimerization between D1 and D2 dopamine receptors leads to a switch from Gs (D1R) or Gi (D2R) to Gq-mediated signalling, meanwhile heterodimerization between DOR and MOR results in a switch from Gi to Gz signalling [266].
2AR association with 1AR and 3AR has been shown to reduce the rate of agonist-induced internalization as well as the ability of 2AR to activate
ERK1/2 [267, 268]. Internalization of one receptor following agonist treatment of the putative partner receptor has been reported for heterodimers of 1AR and
2AR with 2AAR and heterodimers of 2AR and opioid receptors [269, 270].
Inhibition of partner receptor signalling has also been reported by antagonists of
2AR and angiotensin AT1 receptor in the context of a heterodimer [271].
Heterodimerization of 1DAR with 2AR promotes cell surface expression and functional activity of 1DAR [272].
More recently, heterodimer formation has been shown to modulate the recruitment of signalling molecules to the heterodimeric receptor thus leading to changes in transcription factor activation and gene expression.
Heterodimerization of MOR and DOR leads to a switch from a predominantly G protein-mediated to a -arrestin mediated pathway, in turn resulting in differential activation of transcription factors [273]. It is not completely clear what mechanisms underlie these types of changes in signalling pathways. In the case 41
of 2AAR-MOR, it has been proposed that one protomer may function as a scaffold to recruit signalling proteins which promote distinct signalling of the other protomer [261]. This mechanism could represent a means to diversify
GPCR function. Heterodimerization of GPCRs provides a mechanism for generating and regulating GPCR diversity and specificity (Figure 1.5). Thus, targeting of a receptor within a heterodimer could allow the activation of a discrete signalling pathway, limiting side effects in other processes the receptor mediates when not in a heterodimer. Development of bitopic modulators such as bivalent orthosteric ligands or bivalent orthosteric-allosteric ligands for heterodimers could represent a possibility of achieving functional selectivity; however, this remains to be determined [274, 275].
The stability of such complexes as heterodimers or oligomers in the plasma membrane has recently been questioned. GPCR oligomerization has been shown to be important in receptor maturation and delivery to the plasma membrane [253, 255]. However, recent reports also propose the existence of a dynamic equilibrium between monomeric and dimeric/oligomeric states of
GPCRs [276, 277]. The lifetime of these transient dimers was reported to be significantly longer than those of simple bimolecular collisions in the membrane.
These studies proposed that such transient dimers can still be important for specificity in cellular signalling and regulation of receptor trafficking, however, on a much shorter time scale [276]. Surprisingly, the monomer-dimer dynamic was unaffected by the addition of ligand [276]. The ultimate meaning of such discoveries remains to be determined. 42
There are many stages in the life cycle of a GPCR from synthesis, through quality control and maturation to presentation at the plasma membrane and endocytosis into endosomes. The GPCR can be present in one of several active states and in complex with a G protein as well as other scaffolding proteins. It is possible that the same receptor may exist in different oligomerization states, whether they be monomers, dimers or higher-order oligomers, each with a unique set of associated proteins [278].
43
1.7.4. GPCR Oligomerization and Pathological States
Some of the changes in receptor pharmacology and/or signalling attributed to GPCR oligomerization were found to be involved in different pathologies. For example, alterations in AT1R properties which contribute to the pathological effects of angiotensin II have been attributed to the heterodimerization of AT1R with other GPCRs, such as the B2 bradykinin receptor and the 1DAR in the case of specific forms of hypertension, the apelin receptor in atherosclerosis, and most recently, the type 1 cannabinoid receptor in liver fibrosis [248, 279-281].
Heterodimerization between metabotropic glutamate receptors (mGluR) and serotonin 5-HT2A receptors (5-HT2AR) was reported to trigger unique cellular responses when targeted by drugs to treat psychosis, implicating the complex in the altered cortical processes of schizophrenia [251]. Yet another study has demonstrated a physical interaction between melanocortin 4 receptor (MC4R) and
GPR7 as well as melanocortin 3 receptor (MC3R) and growth hormone secretagogue receptor. Because both MC3R and MC4R play an important role in hypothalamic weight regulation the identification of new signalling complexes involving these receptors may provide important information for the development of therapeutic strategies to treat obesity [250]. Formation of heteromeric complexes appears to also serve an evolutionary function for some pathogens.
The Epstein-Barr virus, a human herpes virus that infects B lymphocytes and is associated with tumour development, was shown to express BILF1, a GPCR that can form heterodimers with chemokine receptors like CXCR4 and impair Gi- mediated signalling and function, thus changing the responsiveness of B 44 lymphocytes to chemokines [282]. The same study also showed that BILF1 was able to heterodimerize with the human histamine H (4) receptor (H(4)R) and inhibit its Gi-mediated signalling in a similar way.
45
Figure 1.5. Modulation of GPCR function by receptor heterodimerization.
Heterodimerization can lead to: (A) formation of altered ligand binding site, (B) alteration in coupling to the G protein, (C) coupling to distinct G proteins, and/or
(D) alterations in the balance of receptor coupling to G proteins or other pathways such as β-arrestin [283]. 46
47
1.8. ALLOSTERIC MODULATION OF GPCRs
GPCRs move within the plasma membrane and interact with other proteins to transmit external stimuli into the cell interior and modulate cellular activities.
The sites of interaction of the GPCR agonist and that of accessory interacting cellular proteins are different and topographically distinct. Thus, GPCRs are naturally allosteric proteins in that they possess more than one type of binding site, with the G protein itself being the best-known allosteric modulator of agonist binding to GPCRs [284]. Allosteric protein-protein interactions are highly likely to occur for GPCRs because these receptors interact with a variety of cellular proteins, in addition to or independently of the G proteins [285].
The term “allosteric” with respect to GPCRs, encompasses a variety of related mechanisms by which protein function can be regulated in either a positive or negative direction. There are two main concepts associated with allosterism:
(1) GPCRs possess more than one binding site (orthosteric binding site for the endogenous agonist and additional allosteric binding site(s) distinct from the orthosteric site); and (2) GPCRs are able to undergo conformational changes which result in binding pockets with different affinities for ligands [286]. Thus an allosteric modulator/ligand is one which binds an allosteric binding site on a protein and modulates the binding and/or signalling properties of the orthosteric site [287].
The concept of allosterism was first formalized with the Monod-Wyman-
Changeux (MWC) model, which was built on observations related to the phenomenon of end-product inhibition in enzymes whereby many inhibitors were 48 found to be very diverse in structure from the enzyme’s structure [288]. The
MWC model proposed that a protein exists in a spontaneous equilibrium between active and inactive states in the absence of ligand, and the binding of either an orthosteric or allosteric modulator to their respective binding sites stabilized one state over the other. This model stated that allosteric proteins are oligomeric with some level of ligand-independent (basal) activity and that the allosteric transition between conformational states occurs in an all-or-none fashion [289]. The first model to describe allosteric mechanisms of GPCRs was the ternary complex model developed to explain the reciprocal effects that GPCR orthosteric ligands and G proteins have on each other [290]. This soon developed into the allosteric ternary complex model (ATCM) which accounts for modulation by more than one type of ligand (Figure 1.6) [291]. The ATCM describes the simplest allosteric effect where the orthosteric and allosteric ligands bind reversibly and saturably to their respective binding sites on a free receptor, and their interaction is driven by ligand concentrations, their equilibrium dissociation constants and the cooperativity factor. The cooperativity factor depicts the magnitude of change in the dissociation constant of each ligand as modulated by the presence of the other ligand [289]. However, this model does not account for the presence of distinct conformational states of GPCRs or the idea that allosteric modulators can change the signalling efficacy of an orthosteric ligand without necessarily affecting its binding affinity [289]. Thus the ATCM was extended to accommodate the dynamic behaviour of GPCRs and the allosteric two-state model (ATSM) was introduced (Figure 1.6) [292]. The ATSM incorporates the isomerisation of a receptor between active (R*) and inactive (R) states and introduces additional 49
Figure 1.6. The allosteric ternary complex model (ATCM; red) and the allosteric two-state model (ATSM; cubic scheme) of GPCRs. The R* and R denote the active and inactive states of a GPCR, respectively. KA and KB denote the equilibrium dissociation constants of orthosteric, A, and allosteric , B, ligands, respectively. L denotes the isomerization constant governing the transition between active and inactive receptor states. The parameter denotes binding cooperativity factor for interaction at the ground state of the receptor, the parameters and denote the intrinsic efficacies of orthosteric and allosteric ligands, respectively (their ability to stabilize the active receptor state), and the parameter denotes the activation cooperativity factor for the ternary complex
[[289]].
50
51 constants to account for the selective stabilization of these states by orthosteric and allosteric ligands [289].
These concepts of allosteric interactions between conformationally linked sites can be applied to heterodimers, in which one protomer may regulate the binding and signalling of the other protomer via allosteric interactions (Figure
1.6) [287]. In the context of GPCR heterodimers, allosterism can occur in a number of ways. Binding at the orthosteric or an allosteric site of protomer A in a receptor heterodimer can result in allosteric modulation of ligand binding affinity at an allosteric or the orthosteric site of protomer B. The same type of interactions can modulate efficacy in the heterodimer. Further, some ligands can act to allosterically influence only a subset of functions of the partner protomer
[287]. Thus, conformational changes in protomer A in response to ligands binding to protomer B are indicative of allosterism across a dimer [287]. As described in section 1.5.2 and 1.5.3, such changes can be observed as changes in radioligand binding kinetics and/or signalling properties such as efficacy or complete signal switching. For example, dopamine receptor affinity is modulated in the dopamine D1-D3 heterodimer as the affinity of the D1 agonist ligand
SKF81297 is increased by the occupancy of the D3 receptor with a selective agonist [293]. The ability of serotonin 5-HT2A receptor agonist DOI to enhance
Gi coupling of the receptor was reduced by co-addition of the mGlu2/3 agonist
LY379268 in cells expressing both receptors, thus demonstrating that allosterism between receptor partners can affect properties such as G protein coupling [251].
The same study also reported that an mGlu2/3 antagonist was able to increase locomotor and vertical activities mediated by the 5-HT2A receptor in mice, this 52 effect was abolished in 5-HT2A receptor knock-out animals. This highlights the relevance of heterodimers as drug targets [251]. 53
Figure 1.7. Allosteric ligand regulation of GPCRs. (A) Traditional view of
GPCRs as monomers where an allosteric modulator binds the same receptor as the orthosteric (endogenous) agonist but at a different binding site. (B) This concept can be expanded to GPCR heterodimers, where the binding of orthosteric ligand to one heterodimer protomer allosterically modulates receptor properties of the other protomer and vice versa [294].
54
55
1.9. 2-ADRENERGIC RECEPTOR SIGNALLING
The 2-adrenergic receptor (2AR) represents a subtype of -adrenergic receptors, 1AR and 3AR being the other two subtypes. ARs are a part of a larger family of adrenoreceptors which facilitate sympathetic nervous system responses of “flight or fight”. 2ARs are widely expressed throughout the cardiovascular, respiratory, metabolic and musculoskeletal systems [295]).
Stimulation of 2ARs causes smooth muscle relaxation in many organs, such as gastrointestinal tract and lungs [295]. These receptors interact with a wide variety of signalling pathways, some of which depend directly on G protein-dependent signalling meanwhile others involve agonist-dependent recruitment of G protein- coupled receptor kinases (GRKs) and -arrestins [162].
1.9.1. 2AR-mediated activation of ERK1/2
Stimulation of 2AR also results in activation of extracellular signal regulated kinases 1/2 (ERK1/2). Most studies investigating β2AR-mediated ERK1/2 activation have been performed in HEK 293 or COS-7 cells with transfected receptors. Some studies suggest that β2AR-mediated ERK1/2 activation involves
Gs subunits, rather than the Gs-cAMP-PKA pathway, which then act on a
Ras-dependent pathway leading to activation of ERK1/2 [296]. While others report that β2AR activated ERK1/2 via Gs-PKA-dependent pathway [297].
Moreover, as discussed above, additional studies both in cultured cell lines and and in vitro reconstituted system have demonstrated that, in response to agonist, 56
the β2AR can undergo PKA-dependent phosphorylation leading to a loss of Gs signalling and a switch to activation of Gi [84-86, 154, 157, 298]. β2AR- mediated ERK1/2 activation has also been reported to involve both G protein and
β-arrestin components [206]. Other findings suggest that Src activation plays a role in β2AR stimulation of ERK1/2 [299, 300]. The ability of β2AR to couple to many different signalling pathways reinforces the idea that GPCRs can act as platforms coordinating many different yet specific signals inside the cell.
1.10. OXYTOCIN RECEPTOR SIGNALLING
The oxytocin (OT) receptor (OTR) along with the three vasopressin receptor subtypes V1a, V1b and V2, forms a subfamily of structurally-related GPCRs. The
OTR mediates a very wide spectrum of physiological actions: The OTR is expressed in different specific brain regions where they mediate behavioral functions, ranging from maternal behavior to specific sexual and social behaviours [301, 302]. In the periphery, the OTR mediates effects on uterine contractions, mammary gland milk ejection and differentiation, pituitary prolactin secretion, sodium excretion, T-cell function, cardiovascular control, cardiomyocyte and osteoblast differentiation, and endothelial cell function [303-
306]. The OTR is also widely expressed in different cancers [307].
The great diversity of the expression sites and proposed function of the OTR is paralleled by a diversity of its signalling pathways, many of which have still remained unexplored. A Gq/11-mediated pathway, leading to stimulation of phospholipase C (PLC) inducing increased intracellular calcium and inositol 57 trisphosphate production, has been clearly defined [308]; however, the precise mechanisms by which OT exerts its multiple biological actions have not been fully established.
1.10.1. OTR-mediated activation of ERK1/2
The OTR is also able to activate the MAP kinases ERK1/2 [308, 309]. In Chinese hamster ovary (CHO) cells overexpressing OTR, activation of ERK1/2 occurred mainly via the Gq-PLC-PKC pathway [82]. However, OTR was shown to activate Gi in CHO cells as well as in pregnant rat myometrium [82, 310].
Further, in cultured myometrial cells, OTR-mediated ERK1/2 activation was shown to be Gi-dependent [311]. Finally, OTR-mediated ERK1/2 phosphorylation has been reported to involve both PKC and Gi-Gβγ pathways as well as transactivation of the epithelial growth factor receptor (EGFR) in PMH1 myometrial cells [83]. Most recently, it has been shown that the scaffolding protein -arrestin2 is involved in OT-mediated ERK1/2 activation in HEK 293 cells overexpressing OTR as well as in immortalized human myometrial ULTR cells [312, 313]. Although there are likely cell-specific differences, these reports suggest that OTR-mediated ERK1/2 activation which involves -arrestins occurs downstream of Gq- and Gi-mediated signalling and contributes to approximately
50% of the ERK1/2 phosphorylation via OTR in the rapid phase of ERK1/2 signalling [312, 313].
58
1.11. EXTRACELLULAR SIGNAL ACTIVATED KINASES 1 AND 2
(ERK1/2) AS A COMMON GPCR EFFECTOR
The ERK1/2 belong to a subfamily of mitogen activated protein kinases
(MAPK) and share approximately 83% sequence homology. ERK1/2 is expressed to various extents in all tissues, with particularly high levels in the brain, skeletal muscle, thymus, and heart [314]. ERK1/2 can be activated by growth factors such as platelet-derived growth factor (PDGF), epidermal growth factor (EGF), and nerve growth factor (NGF), and in response to insulin. They are also activated by GPCR ligands, cytokines, osmotic stress, and microtubule disorganization [315]. In order to be activated, ERK1/2 are each phosphorylated on 2 residues (threonine and tyrosine) within a conserved Thr-Glu-Tyr (TEY) motif in their activation loop [165, 169].
ERK1/2 are activated as a result of a cascade of phosphorylation by specific protein kinases. The ERK1/2 module consists of the MAPK kinase kinases
(MAPKKKs) A-Raf, B-Raf and Raf-1 (or C-Raf), the MAPK kinases (MAPKKs)
MEK1 and 2, and the MAPKs ERK1 and 2. Although Raf isoforms are the principal MAPKKKs in the ERK1/2 module, the protein kinases MEKK1, Mos, and Tpl2 were also identified to function as MAPKKKs in a cell type-restricted and stimulus-specific way [315, 316]. The ERK1/2 module is mainly activated by cell surface receptors, like receptor tyrosine kinases (RTKs), which become phosphorylated upon activation.
GPCRs have also been shown to regulate ERK1/2 either via G protein- dependent or -arrestin-dependent signalling pathways. Gs-mediated ERK1/2 59 activation was shown to involve cAMP-mediated activation of Rap-1 by its specific GEF, EPAC (exchange protein directly activated by cAMP) [317-319].
Activated Rap-1 activates B-Raf and the downstream ERK1/2 module [320].
However, for some receptors like the 2AR, activated PKA also appeared to be important for ERK1/2 activation [321]. 2AR –mediated ERK1/2 activation required the stimulation of Ras and Rap-1 Further studies have indicated that another GEF, the C3G (Crk SH3 domain-binding guanine nucleotide releasing factor) can activate Rap-1 in some cell types (HEK293, AtT20, COS, v-Ki-ras- transformed NIH 3T3 cell lines and insect cells), however, this appears to require
PKA-mediated phosphorylation of Src [322-325]. Gs-cAMP-mediated or PKA- mediated activation of Ras was also shown to be involved in ERK1/2 activation
[326-330]. The ERK1/2 pathway can also be inhibited by PKA-mediated phosphorylation of Raf-1 or its sequestration by Rap-1[331-334]. Gi-dependent activation of ERK1/2 was shown to involve suppression of the inhibitory effects on Raf-1 of PKA and Rap-1 from Gs-mediated signalling [105, 335]. G subunits released from the Gi-heterotrimer complex were also implicated in
ERK1/2 activation. This mechanism involves stimulation of PLC leading to IP3- mediated rise in cytosolic Ca2+ and the subsequent Ca2+-mediated activation of
Pyk2 kinase. The activated Pyk2 kinase initiates a cascade involving Src, Shc adaptor protein and Ras [336-339]. In some cellular contexts, the G subunits activate ERK1/2 via a pathway involving PI3K and the dynamin-Grb2 complex formation which results in activation of Ras and then ERK1/2 [340, 341]. As seen in the case of -arrestin-mediated ERK1/2 activation, it is possible that G- 60 mediated ERK1/2 activation is dependent on dynamin-mediated endocytosis and does not result in transcriptional activity [342, 343]. Gq can stimulate ERK1/2 by three distinct mechanisms: (1) a PLC-DAG-PKC-mediated activation of Raf-1;
2+ 2+ (2) a PLC-IP3-Ca -Pyk2-Src-dependent activation of Ras; and (3) Ca -DAG- stimulated Rap-1 activation of B-Raf [344-347]. In some cases, such as for the
2+ 1AR, ERK1/2 activation can involve both PKC- and Ca - mediated pathways
[339], while in other cases, like the OTR, the involvement of G subunits and epidermal growth factor receptor (EGFR) transactivation was shown to result in
ERK1/2 activation in the human myometrium PHM1 cell line [348]. The authors suggested that the mechanism of OT-mediated ERK1/2 activation in these cells involves a PLC-independent pathway mediated by G subunits released from activated Gq, which leads to activation of EGFR in a Ca2+-dependent manner which can trigger Ras-dependent activation of ERK1/2 [348]. Finally, in contrast to the other G proteins described here, G12 and G13 were reported to mainly attenuate the activation of ERK1/2 through a mechanism that is not yet fully understood [349].
In addition to G protein-dependent ERK1/2 activation, GPCRs can also activate the ERK1/2 module via -arrestins. -arrestin-1 was shown to bind both
GRK-phosphorylated GPCRs like the 2AR, and Src, targeting them to clathrin- coated pits where the complex acts as a signalling hub for the activation of
ERK1/2 [190]. Further, -arrestin-2 was shown to sequester components of the
ERK1/2 module such as Raf-1, MEK1/2 and ERK1/2 by its interaction with Raf-1
[350]. The -arrestin scaffold containing the GPCR and components of the 61
ERK1/2 module are internalized by clathrin-coated pits into endosomes, where
ERK1/2 activation takes place. The kinetics of such -arrestin-mediated ERK1/2 activation appear to be slower than those mediated by G proteins and the pool of
ERK1/2 activated in this way is not involved in nuclear signalling but instead appears to be involved in cytosolic responses that require slower but persistent
ERK1/2 activation [193, 207, 342].
Once activated ERK1/2 phosphorylates numerous substrates within the cytosol and the nucleus, where it regulates gene transcription [351]. The ERK1/2 signalling pathway is one of the most ubiquitous signal transduction cascades.
However, the mechanisms of ERK1/2 activation by GPCRs often depend on cell type and receptor involved, i.e. molecular context is an all-important determinant of the nature and duration of a particular ERK1/2 signal. Because of the complex nature of GPCR signalling it is important to take note of the molecular context of the signalling pathways studied in order to interpret results properly.
1.12. HUMAN MYOMETRIUM AND CONTRACTILITY
1.12.1. Normal labour
The myometrium makes a remarkable transition from a relatively relaxed state throughout pregnancy to the onset of contractions at labour. This transition occurs as a result of stimulatory inputs and in turn the loss of uterine quiescence.
The significantly increased expression of contraction-associated proteins (CAPs) is a crucial event in labour that allows for the initiation and progression of 62 powerful rhythmic contractions that force the fetus through the softening cervix at term [228]. There are three types of CAPs: those that (1) enhance interactions between the actin and myosin proteins that cause muscle contraction, (2) increase excitability of individual myometrial cells, and (3) promote intercellular connectivity that permits the development of synchronous contractions [167].
Proteins that promote myocyte contractility increase interactions between actin and myosin, which are necessary for contraction to take place. Filamentous actin, attached to the cytoskeleton at focal points in the cell membrane linked to the underlying matrix, overlaps myosin filaments in such a way that when activated myosin can bind and pull the actin leading to the development of tension and thus contraction of the myocyte. Contraction begins when a stimulus, electrical or chemical, opens voltage- or ligand-regulated Ca2+-channels on the cell membrane or the sarcoplamic reticulum allowing Ca2+ to enter the cytoplasm.
Upon entry into the cytoplasm, Ca2+ binds calmodulin (CaM) and the Ca2+-CaM complex then activates myosin light chain kinase (MLCK). In turn, MLCK phosphorylates and thus activates myosin, which when activated, has ATPase activity and is able to bind actin leading to myocyte contraction. Relaxation of the smooth muscle fibre takes place upon removal of Ca2+ from the cytoplasm by
Na+/Ca2+ exchangers as well as plasma membrane (PMCA) and endoplasmic reticulum (SERCA) Ca2+-ATPases, and the dephosphorylation of myosin by a myosin light chain phosphatase. Myosin regulation can also be supplemented by regulation of actin. Actin can bind caldesmon, a regulatory protein, which inhibits actin-myosin biding. However, when caldesmon is phosphorylated by 63 protein kinases its interaction with actin is prevented thus allowing actin and myosin binding and promoting contractions [168].
As previously mentioned the opening of Ca2+ channels and the subsequent release of Ca2+ into the cytosol is necessary for myocyte contractions. There are some signalling molecules which impact Ca2+ homeostasis directly, while others stimulate signalling pathways that impact Ca2+ homeostasis indirectly. In a pathway used by a number of myometrial contractants such as oxytocin and the prostaglandins, activated receptors coupled to Gq activate PLC isoforms, resulting in the hydrolysis of plasma membrane phosphatidylinositide bisphosphate (PIP2) to produce inositol 1,4,5-triphosphate (IP3) and diacylglycerol
2+ (DAG) [352]. IP3 binds to IP3 receptors in the SR leading to a release of Ca into the cytoplasm in myometrial cells [286, 353]. Hormones that lead to myometrial contractions can also indirectly stimulate a change in membrane potential, triggering Ca2+ entry through voltage-gated Ca2+ channels as well as signal- regulated Ca2+ channels [288, 290, 292, 354-356]. In contrast, myometrial relaxants such as -adrenergic ligands and relaxin activate Gs pathways which oppose these pathways by activating AC, generating cAMP and activating PKA.
PKA-mediated phosphorylation affects a number of components regulating intracellular Ca2+, which include Ca2+ transporting ATPases; RGS4, enhancing its ability to inhibit Gq signalling; GRK2, contributing to inhibition of ligand- dependent PLC1 activation; and Ca2+-activated K+ channels [137, 357, 358].
Another relaxant signalling pathway present in the myometrium is the nitric oxide
(NO) guanylyl cyclase/cGMP pathway. Nitric oxide is a vasodilator produced in 64 myometrial cells and is important for the maintenance of smooth muscle tone
[359]. NO interacts with soluble guanylyl cyclase, present in nearby effector cells, which results in the synthesis of cyclic guanosine monophosphate (cGMP) in these target cells [359]. The increased cGMP content in smooth muscle cells inactivates MLCK resulting in smooth muscle relaxation [121, 360]. There is a shift in the balance of these opposing systems at the onset of labour which promotes myocyte contraction [227, 361].
The levels of Ca2+ in the cytoplasm can also be regulated indirectly by mechanisms such as activation of K+ channels, which hyperpolarize the cell and decrease Ca2+ entry through voltage-gated Ca2+ channels [241]. There are changes in the distribution and function of K+ channels that occur at the time of labour which result in modulation of the myocyte excitability, lowering the intensity of the stimulus required to depolarize myocytes, increase Ca2+ entry and generate contractions [229, 230, 241]. The coupling of 2- and 3-adrenergic receptors to K+ channels also declines at labour, removing their inhibitory action on myocyte excitability [160, 362].
Finally, an important aspect of myometrial activity at labour is the development of synchronized contractions. As parturition progresses there is an increase in synchronization of the electrical activity of the uterus [166, 189, 190].
At the cellular level, this synchrony is achieved by electrical conduction through connecting myofibrils, which transmit electrical activity to nearby muscle fibers.
The activated myocytes produce prostaglandins, which act in a paracrine fashion to depolarize neighbouring myocytes. This leads to a wave of activity as more 65 and more myocytes are recruited into the contraction [167]. At the molecular level, myocytes are connected by channels or gap junctions that are created by multimers of connexion 43 (Cx43) [167]. At the onset of labour, clusters of Cx43 gap junctions increase in size and number in the myometrium [129]. The Cx43 mRNA expression also increases in the days leading up to parturition reaching maximal levels at the time of labour [131, 136, 363].
1.12.2. Preterm labour and current tocolytic drugs
Preterm labour is defined as birth before the completion of 37 weeks of gestation. Preterm birth can be categorized either indicated or spontaneous.
Indicated preterm birth occurs when delivery of the infant is necessary due to medical or obstetrical complications which endanger the health of the mother or fetus [359]. Spontaneous preterm birth occurs as a result of spontaneous labour or preterm rapture of fetal membranes before the onset of labour [359]. Preterm labour is difficult to predict or diagnose because the mechanisms regulating the transition of the uterus from implantation through pregnancy to parturition are not well understood [137]. The management of preterm labour is thus symptomatic.
Once the threat of preterm labour is established, attenuation of uterine contractions with tocolytic drugs becomes important for the safe transportation of the mother to a medical center with appropriate obstetric and neonatal facilities.
Unfortunately, currently available tocolytic agents can only prolong uterine contractions for approximately 24 to 48 hours, which is not sufficient to improve perinatal outcomes, but they can prolong labour long enough for administration of glucocorticoids to improve fetal lung maturation [137, 364]. 66
Current tocolytic treatment strategies include: (1) Ca2+ channel blockers,
(2) NO donors, (3) magnesium sulfate, (4) cyclooxygenase inhibitors, (5) 2AR agonists, and (6) OTR antagonists. The Ca2+ channel blockers, such as nifedipine, have been used extensively to suppress preterm contractions [359]. Although nifedipine was shown to be an effective treatment, it may cause hypotension, tachycardia and neuromuscular blockage in pregnant women [111, 365-367]. NO donors, such as nitroglycerine, have been shown to be effective in prolonging labour for 48 hours, however, hypotension, dizziness and flushing have been reported as maternal side effects [113, 116, 368]. Magnesium sulphate has been used as a tocolytic agent because high levels of Mg2+ are thought to compete with
Ca2+ entry into the cell and inhibit actin-myosin interactions [137]. However, magnesium sulphate has been shown ineffective and is associated with several serious side effects in the mother (respiratory paralysis, suppression of heart rate and contractility, neuromuscular blockage) as well as an increased risk of infant mortality [119, 123, 127, 365]. Cyclooxygenase (COX or prostaglandin synthase) inhibitors decrease prostaglandin production either by general inhibition of COX-
1 and -2 or by specific inhibition of COX-2 [137]. The general Cox inhibitor, indomethacin, is the most commonly used tocolytic agent of this class and has been shown to reduce preterm births in three trials [124]. On the other hand, although specific COX-2 inhibitor nimesulide has been shown to reduce myometrial contractility in animal studies, there is not enough evidence from human studies to suggest COX-2 inhibitors for this indication [118, 122, 125,
126]. 67
The most widely used tocolytic drugs have been 2AR agonists, ritodrine and terbutaline. Although treatment with these tocolytics delays preterm labour
(48 hours), prolonged stimulation of 2AR leads to receptor desensitization and downregulation resulting in no real benefit [369, 370]. Worse, the use of 2AR agonists is associated with potentially serious cardiovascular side effects such as tachycardia and hypotension which has resulted in the reduction in their clinical use [367]. Finally, the OTR antagonists, such as atosiban and barusiban, have been used in the management of preterm labour. Clinical trials have shown that atosiban is as effective as 2AR agonists in reducing preterm labour and has a much lower frequency of side effects [117].
The currently available tocolytic drugs are not ideal. This is most likely because these drugs do not change the fundamental process leading to myometrial activation [359]. Hence the crucial need for more effective treatments of preterm labour can only be addressed by developing a better understanding of the complex signalling systems at work in the human myometrium.
1.13. HUMAN MYOMETRIAL CELL MODEL
We have used the myometrial human telomerase reverse transcriptase- clone 3 (hTERT-C3) cell line as a native cell model in the studies presented in this thesis. hTERT-C3 cells are a cell line of course and not primary myometrial cells per se. hTERT-C3 cells represent a selected subclone that was obtained by serial clonal dilution of the myometrial human telomerase reverse transcriptase
(hTERT)-HM cells and has been described previously [371]. The parent hTERT- 68
HM cells were obtained from Dr. W. E. Rainey represent a cell population of human myometrial cells that was immortalized by transfection with an expression vector containing hTERT [372]. This procedure maintains telomere length and endows the cells with an unlimited life span in culture [372]. The primary human myometrial cells used for this procedure were obtained from the anterior wall of the uterine fundus from premenopausal women undergoing hysterectomy with no indications of uterine disease [372]. hTERT-C3 cells exhibit a phenotype very similar to the primary cells from which they were derived in terms of morphology, OTR density, coupling of the activated OTR to the MAPK pathway, and the OT-induced contraction response curves [371]. hTERT-C3 cells thus represent a good model for the study of myometrial cell biology during the onset of labour since: (1) these cells express similarly high levels of OTRs as myometrial cells at the onset of labour [140], and (2) these cells show a concentration-dependent response to OT and the OTR-specific agonist TGOT in a collagen retraction assay that is indistinguishable from that obtained with myometrial strips from women at term [142, 144, 371].
Other human myometrial cell lines can also been used to study the biology of the myometrium and they include: the myometrial M11 cell line, the non- pregnant immortalized human myometrial cell line (ULTR), and the pregnant human myometrial PHM1 cell line. However, there are several different disadvantages associated with the use of each of these cell lines. M11 cells were derived from dispersed primary human myometrial cells by repeated passage without the use of any immortalizing or transforming agents [371]. Nonetheless,
M11 cells exhibits weaker OT-induced contractions and lower OTR receptor 69 density and smooth-muscle -actin levels than hTERT-C3 cells [371]. ULTR cells were derived from the uterus of a 49-year old woman undergoing a hysterectomy for multiple leiomyomas. These cells were immortalized using the human papilloma virus type 16 (HPV16) E6/E7 open reading frames in a defective retrovirus vector [143]. OTR properties such as receptor density or contractile responses have not been assessed in these cells. PHM1 cells are derived from the upper edge of the lower uterine segment at a time of a caesarean section from a pregnant woman and were immortalized using a defective adenovirus vector expressing HPV16 E6/E7 proteins [145]. These cells have been characterized in terms of OTR receptor density yet this is difficult to compare to OTR density in hTERT-C3 cells as the binding assays used were different, however, PHM1 cells exhibit weaker OT-induced contractions than hTERT-C3 cells [138, 145]. In addition, although these myometrial cell lines maintain some smooth muscle characteristics, such as elongated cell morphology and smooth muscle actin expression, the presence of more sensitive markers of smooth muscle differentiation like calponin, h-caldesmon and smoothelin, have not been examined. However, these markers are expressed in the parent cells of the hTERT-C3 cell line [372].
Moreover, the insertion of viral oncogenes to produce immortalized cells can result in the development of altered cellular signalling pathways and other characteristics associated with cancer cells [147, 372] in contrast with cell lines immortalized using the hTERT, which do not exhibit cellular changes associated with cancer and have normal cellular responses to DNA damage signals [134, 70
146, 148]. The expression of HPV16 E6 and E7 proteins may interfere with the normal action of cellular proteins involved in cell cycle control in cell lines immortalized using these proteins. E6 has been shown to bind and promote degradation of the wildtype p53 protein and E7 forms an inactivating complex with the retinoblastoma tumour suppressor gene product [132, 135].
There are also several advantages to using the immortalized clonal, untransformed hTERT-C3 cell line over a heterogenous cell population derived from isolated primary cells. First, the myometrium is composed of different cell types, each with a different level of OTR expression, resulting in a mixed primary cell population and lack of consistency and reproducibility [141, 371]. Second, immortalized clonal cell lines have nearly infinite growth capacity, thus producing a reproducible experimental system [371]. Thus, hTERT-C3 cells represent a valuable model for the study of human myometrial cell biology. It would still be interesting to validate my research findings in the setting of myometrial strips however.
1.14. PROJECT RATIONAL AND OBJECTIVES
GPCRs interact to form homo- and hetero-dimers and depending on the partner this interaction promotes changes in receptor function, which may influence ligand-binding affinity and specificity as well as downstream coupling, trafficking and desensitization of the receptors. Aside from receptor oligomerization, GPCRs also interact with receptor activity-modifying proteins, and bind various scaffolding proteins, such as -arrestin, to their intracellular 71
receptor domains. The OTR and the 2AR, as discussed above, are members of the GPCR superfamily and are both expressed in myometrial cells where their activation leads to uterine contraction or relaxation, respectively. These two receptors also couple to different G proteins. The OTR is functionally coupled mainly to Gq/11 as well as Gi, meanwhile the 2AR is predominantly understood as a Gs-coupled receptor. More importantly both OTR and 2AR are the pharmacological targets of the only two currently approved tocolytic agents used to control pre-term uterine contractions; the 2AR agonist ritodrine and the
OTR antagonist atosiban.
Although these two receptors display antagonistic effects on myometrial cells, they both activate ERK1/2, which has been implicated in uterine contraction and the onset of labour. Several studies, mostly performed in the rat myometrium, have identified activation of ERK1/2 as a component of the cascade of events leading to the development of labour. It has been shown that ERK1/2 can phosphorylate caldesmon (CaD) [373], an actin-binding protein which inhibits actin-activated ATPase activity in vitro and has been proposed as a regulator of smooth muscle actin-myosin interactions [374]. The onset of labour has been reported to be associated with basal activation of ERK2 and subsequent phosphorylation of CaD, which releases the inhibition on myometrial contraction
[375]. ERK1/2 signalling has been implicated in prostaglandin-mediated and oxytocin-mediated myometrial contraction [376, 377]. It has also been shown that ERK1/2 signalling is involved in inhibiting gap-junction-mediated cellular communication, which may have implications for contraction coordination [378]. 72
Further, ERK1/2 inhibition has been found to delay preterm labour induced by the antiprogesterone agent RU-486 [379]. ERK1/2 has also been shown to mediate stretch-induced c-fos expression in myometrial cells [380]. The c-fos gene encodes the Fos protein, which is an activator protein-1 (AP-1) transcription factor and is involved in activation of transcription of genes possessing AP-1 elements in their promoters [153]. c-fos is thought to play an important role in onset of labour because c-fos mRNA is elevated in the myometrium before the rise in expression of CAPs, which occurs with the onset of both term and preterm labour and “activates” the myometrium for labour [159, 381]. CAP proteins such as OTR and gap junction connexin43 (Cx43) both contain AP-1 sites in their promoters [152, 158]. Thus, phosphorylation of ERK1/2 was suggested to play a regulatory role in myometrial c-fos expression leading to CAP expression and initiation of labour by the mechanical strain imposed by the growing fetus [380].
The interplay of OTR and 2AR has been previously suggested. It was reported that 2AR signaling can have an effect on OTR signaling in the myometrium [382, 383]. However, the mechanisms underlying crosstalk between the two receptors have not been investigated in detail. Nonetheless, PKA has been shown to be involved in inhibiting OTR stimulated PLC in a human myometrial cell line [384]. The issue of second messenger crosstalk between the two receptors is discussed in more detail in Chapter 4. Further, our labs and others have shown that OTR and 2AR can dimerize with other GPCRs [156, 262,
267, 268, 270, 385-392]. However, it remained unclear how these two receptors activate ERK1/2 in the human myometrium or if they formed a heterodimer that 73 was functionally relevant.
This thesis aims to characterize the signalling pathways downstream of
OTR and 2AR to activate ERK1/2 in human myometrial cells and investigates the interactions between these receptors at functional and physical levels.
Understanding the mechanisms and consequences of the interactions between
OTR and 2AR is of significant pharmacological and physiological importance.
We hypothesized that both the OTR- and the 2AR-mediated ERK1/2 activation in human myometrial cells involve the activation of specific G proteins and their respective effectors. Further we hypothesized that each receptor is able to modulate signalling of the other and that the OTR and 2AR form a complex. In order to test these hypotheses, we established the following objectives:
1. To define the molecular mechanisms by which the OTR and the 2AR
mediate ERK1/2 activation in human myometrial hTERT-C3 cells
2. To investigate possible interactions between the signalling pathways of the
OTR and the 2AR, whether they are at the level of second messenger
crosstalk or allosteric in nature
3. To establish if the OTR and the 2AR form a complex
These objectives are addressed in the following two experimental chapters, which consist of two published manuscripts. The first manuscript
(Chapter 2) addresses objective 1 and seeks to determine the signalling pathways involved in mediating the activation of ERK1/2 by each receptor. Interestingly, 74
we found that the 2AR-mediated ERK1/2 activation occurs through a signalling pathway which involves Gi-PI3kinase-PKC and Src, and is dependent on the presence of OTR in myometrial cells and in HEK 293 cells. These results suggested a molecular interaction between the two receptors, and this issue is further pursued in the second manuscript (Chapter 3). 75
CHAPTER 2
2.1. PREFACE
The OTR and the 2AR represent important targets of tocolytic agents used to inhibit pre-term labour. Although, the OTR and the 2AR have opposing effects in the human myometrium of uterine contraction and relaxation, respectively, both receptors activate ERK1/2 which has been implicated in uterine contraction. Our initial objective was to determine the signalling pathways involved in mediating the activation of ERK1/2 by each receptor. Interestingly, we found that the 2AR-mediated ERK1/2 activation occurs through a signalling pathway which involves Gi-PI3kinase-PKC and Src, and is dependent on the presence of OTR in myometrial cells and in HEK 293 cells. The results presented in this manuscript suggest a molecular interaction between the two receptors.
Manuscript: Wrzal, P.K., Goupil, E., Laporte, S.A., Hébert, T.E., and Zingg,
H.H. Functional interactions between OTR and 2AR: implications for ERK1/2 activation in human myometrial cells. Cellular Signalling. January
2012;24(1):333-341. 76
2.2. MANUSCRIPT FUNCTIONAL INTERACTIONS BETWEEN THE OXYTOCIN RECEPTOR AND THE 2-ADRENERGIC RECEPTOR: IMPLICATIONS FOR ERK1/2 ACTIVATION IN HUMAN MYOMETRIAL CELLS Paulina K. Wrzala, Eugénie Goupila, Stéphane A. Laportea,b, Terence E. Héberta* and Hans H. Zingga,b,c* a Department of Pharmacology and Therapeutics, McGill University, Montréal, Québec, H3G 1Y6, Canada b Department of Medicine, McGill University, Montréal, Québec, H3A 1A1, Canada c Department of Obstetrics and Gynecology, McGill University, Montréal, Québec, H3A 1A1, Canada
*To whom correspondence should be addressed. Dr. Hans H. Zingg, MD., PhD., Department of Pharmacology and Therapeutics, McGill University, 3655 Promenade Sir-William-Osler, Room 1325 Montréal, Québec, H3G 1Y6, Canada Tel: (514) 398-3621; Fax: (514) 398-2045 E-mail: [email protected]
Dr. Terence E. Hébert, PhD., Department of Pharmacology and Therapeutics, McGill University, 3655 Promenade Sir-William-Osler, Room 1303 Montréal, Québec, H3G 1Y6, Canada Tel: (514) 398 1398; Fax: (514) 398-6690 E-mail: [email protected]
Running Title: ERK1/2 activation by 2AR and OTR in human myometrial cells
Keywords: Oxytocin receptor, 2-adrenergic receptor, ERK1/2, Myometrium, Dimerization 77
Abstract
The Gq-coupled oxytocin receptor (OTR) and the Gs-coupled 2-adrenergic receptor (2AR) are both expressed in myometrial cells and mediate uterine contraction and relaxation, respectively. The two receptors represent important pharmacological targets as OTR antagonists and 2AR agonists are used to control pre-term uterine contractions. Despite their physiologically antagonistic effects, both receptors activate the MAP kinases ERK1/2, which has been implicated in uterine contraction and the onset of labor. To determine the signalling pathways involved in mediating the ERK1/2 response, we assessed the effect of blockers of specific G protein-associated pathways. In human myometrial hTERT-C3 cells, inhibition of Gi as well as inhibition of the Gq/PKC pathway led to a reduction of both OTR- and 2AR-mediated
ERK1/2 activation. The involvement of Gq/PKC in 2AR-mediated ERK1/2 induction was unexpected. To test whether the emergence of this novel signalling mechanism was dependent on OTR expression in the same cell, we conducted experiments in HEK 293 cells that were transfected with the 2AR alone or co-transfected with the OTR. Using this approach, we found that
2AR-mediated ERK1/2 responses became sensitive to PKC inhibition only in cells co-transfected with the OTR. Inhibitor studies indicated the involvement of an atypical PKC isoform in this process. We confirmed the specific involvement of PKC in this pathway by assessing PKC translocation to the cell membrane. Consistent with our inhibitor studies, we found that 2AR- mediated PKC translocation was dependent on co-expression of OTR. The 78
present demonstration of a novel 2AR-coupled signalling pathway that is dependent on OTR co-expression is suggestive of a molecular interaction between the two receptors. 79
1. Introduction
The oxytocin receptor (OTR) and the 2-adrenergic receptor (2AR) are members of the large G protein-coupled receptor (GPCR) family. Both OTR and
2AR are co-expressed in myometrial cells, but relay opposing signals of contraction and relaxation in these cells, respectively. While the OTR, is functionally coupled mainly to Gq/11 as well as Gi in myometrial cells, the
2AR is predominantly a Gs-coupled receptor and mediates uterine relaxation via an increase in intracellular cAMP levels [393]. The two receptors represent important pharmacological targets, because OTR antagonists and 2AR agonists are used to control pre-term uterine contractions [359]. Despite their physiologically antagonistic effects, both receptors activate ERK1/2. Several studies, mostly performed in the rat myometrium, have identified the activation of
ERK1/2 as a component of the cascade of events leading to the development of labor [203, 380]. The onset of labor has been reported to be associated with basal activation of ERK1/2 [394].
The ERK1/2 signalling pathway constitutes one of the most ubiquitous signal transduction cascades. The mechanisms of ERK1/2 activation by GPCRs often depend on cell type and receptor involved, i.e. molecular context is an all- important determinant of the nature and duration of a particular ERK1/2 signal.
Although, both OTR and 2AR activate ERK1/2 the mechanisms by which they do so are different. The OTR has been shown to couple to both Gq and Gi in myometrial cells. OTR activation of Gq has been shown to be important for OT- 80 stimulated phospholipase C (PLC) activation and elevation of intracellular calcium in the myometrium and in the activation of ERK1/2 [80-82]. However, the OTR was also shown to activate Gi in several cell types and OTR-mediated
ERK1/2 phosphorylation has been reported to involve both PKC and Gi-Gβγ pathways as well as transactivation of the epidermal growth factor receptor
(EGFR) in PMH1 myometrial cells [82, 83, 310, 311]. It has also been shown that the OTR interacts with the scaffolding protein -arrestin2 [395] and that - arrestin2 is involved in OT-mediated ERK1/2 activation [312, 313]. Activation of
ERK1/2 has been shown to be important in oxytocin-mediated myometrial contractions [377].
To date, most studies investigating β2AR-mediated ERK1/2 activation have been performed in HEK 293 or COS-7 cells with transfected receptors.
Some studies have suggested that β2AR-mediated ERK1/2 activation involves
G subunits, rather than the Gs-cAMP-PKA pathway, which then act on a Ras- dependent pathway leading to activation of ERK1/2 [296]. Other studies show that β2AR activates ERK1/2 via Gs-PKA-dependent pathway [297]. Another G protein coupling mechanism relevant for ERK1/2 activation is the potential ability of β2AR to undergo PKA-dependent phosphorylation in response to agonist, which leads to a loss of Gs signalling and a switch to activation of Gi [85].
Further, β2AR-mediated ERK1/2 activation has also been reported to involve both
G protein and β-arrestin components [206]. Other reports suggest that Src activation plays a role in β2AR stimulation of ERK1/2 [299, 300]. 81
Previous studies have shown that 2AR signalling could affect OTR signalling in the myometrium [382, 383]. However, the mechanisms underlying crosstalk between the two receptors have not been investigated in detail.
Unraveling the nature and consequences of OTR/β2AR interactions and distinguishing direct physical interactions from indirect second messenger- mediated interactions, would be of significant pharmacological and physiological importance. The purpose of the present study was to define the signalling mechanisms involved in activating ERK1/2 by both the OTR and 2AR in myometrial cells. Our studies demonstrate the involvement of several G proteins in OTR- and 2AR-mediated ERK1/2 activation. We also identify a novel specific signaling pathway by which the 2AR activates ERK1/2 in myometrial cells. This pathway involves Gi-PI3K-PKC and is dependent on co-expression of the OTR and the 2AR in the same cell. Taken together our results define a novel functional interaction between the two receptors which may, in turn, be based on a physical interaction.
82
2. Materials and Methods
2.1. Reagents
Reagents were obtained from the following sources: DMEM/F12 tissue culture medium was from Invitrogen (Burlington, ON); fetal bovine serum (FBS) from Hyclone (Logan, UT); OT was from Sigma-Aldrich (St. Louis, MO); isoproterenol was from Tocris Bioscience (Bristol, UK); the inhibitors AG1478, wortmannin, PP2, PKC pseudosubstrate inhibitor, Gö6976 and Gö6983 were from Calbiochem (La Jolla, CA); Pertussis toxin (PTX) and Rö31-8220 and the protease inhibitors: leupeptin, benzamidine and trypsin were from Sigma-Aldrich
(St. Louis, MO). Primary anti-phospho-ERK1/2 (T202, Y204) antibody was from
Cell Signaling Technology (Beverly, MA); pan anti-ERK1/2 antibody was from
Stressgen (Ann Arbor, MI). Secondary antibody, horseradish peroxidase- conjugated anti-rabbit IgG, was from Sigma-Aldrich (St. Louis, MO).
Lipofectamine 2000 was from Invitrogen (Burlington, ON). All other analytical grade chemicals were obtained from Sigma-Aldrich, Fisher Scientific (Waltham,
MA), or VWR (West Chester, PA). -arrestin1/2 and control siRNA were obtained from Qiagen Inc. (Toronto, ON).
2.2. Constructs
OTR-YFP, an OTR construct with the coding sequence for YFP added in frame to the OTR C-terminus [396]. 2AR-HA, was used as described previously
[267]. PKC1-GFP and PKC-GFP constructs were a kind gift from Dr S.
Ferguson (Robarts Research Institute, London, ON). To generate mCherry-tagged 83
PKCs, the constructs mentioned above and a pcDNA3.1(+)-mCherry vector were digested with NheI and BsrGI to replace the GFP with mCherry. To generate the pcDNA3.1(+)-mCherry vector, pRSET-B-mCherry, a kind gift from Dr. Roger Y.
Tsien (University of California, San Diego, CA), was amplified by PCR to introduce a NheI site in 5' and KpnI site in 3' of the mCherry and cloned into pcDNA3.1(+). shRNA constructs targetting -arrestin1/2 were generous gifts from Dr. Marc Caron (Duke University, Durham, NC). All constructs were verified by bidirectional sequencing.
2.3. Cells culture and transfection
The myometrial human telomerase reverse transcriptase (hTERT)-HM cells were obtained from Dr. W. E. Rainey [372]. hTERT-C3 cells represent a selected subclone that we obtained by serial clonal dilution of hTERT-HM cells described previously [371]. hTERT-C3 cells were maintained in DMEM/F12 medium supplemented with 10% FBS and cultured at 37° C in 5% CO2. Cells close to confluency were passaged by trypsinization and plated in T175 flask at a one-quarter dilution every 4–5 d. HEK 293 cells were from Invitrogen (Carlsbad,
CA) and were grown in Dulbecco’s Modified Eagle’s Medium/high glucose
(DMEM) supplemented with 5% FBS and transfected using Lipofectamine 2000 as per manufacturer’s instructions. Experiments were carried out 48h after transfection.
2.4. Western blots
Immunoblotting was used to assess the expression of cellular proteins. 84
Briefly, cells were plated onto culture dishes and grown in appropriate media to near confluency. Cells were starved for 24 h in either DMEM/F12 supplemented with only 0.5% FBS (hTERT-C3 cells) or DMEM/high glucose supplemented with only 0.5% FBS (HEK 293 cells). Cells were stimulated with ligand(s) for different times at 37° C. Stimulation was stopped by two ice-cold PBS washes, and plates were flash frozen in liquid nitrogen. For inhibitor studies, cells were pre-treated with the inhibitor for 30min to 1h before stimulation with ligand(s).
Cells were lysed on ice with Laemmli buffer (50 mM Tris-HCl [pH 6.8], 2% sodium dodecyl sulfate, 10% glycerol, and 0.1M -mercaptoethanol) and were either homogenized using passage through a syringe and needle (hTERT-C3 cells) or by sonication (HEK 293 cells). Lysates were clarified by centrifugation at 15,
000 x g for 10 min at 4 C in a microcentrifuge. Proteins were denatured by boiling for 5 min, and subjected to SDS-PAGE and western blotting. Immunodetection involved different primary antibodies in conjunction with a secondary horseradish peroxidase-conjugated antibody and a chemiluminescence detection system
(Supersignal; Pierce). Quantification of band intensities was performed using
AlphaEase (Alpha Innotech Corp., San Leandro, CA).
2.5. Receptor quantification
Total receptor number for the β2AR was calculated from binding experiments using [125I]cyanopindolol (CYP) as the radioligand. Membranes were prepared and washed as previously described, with all steps of the process performed on ice [267]. Briefly, hTERT-C3 cells were washed twice with ice-cold
PBS. They were then incubated for 15 min with ice-cold lysis buffer containing 85
15 mM Tris–HCl (pH 7.4), 0.3 mM EDTA, 2 mM MgCl2, 5 μg/mL leupeptin, 10
μg/mL benzamidine and 5 μg/mL trypsin inhibitor. Subsequently, cells were collected and homogenized with a polytron. Lysates were centrifuged at 14500 rpm for 20 min at 4°C; the pellet was resuspended in ice-cold membrane buffer containing 50 mM Tris–HCl (pH 7.4), 3 mM MgCl2, 5 μg/mL leupeptin, 10
μg/mL benzamidine and 5 μg/mL trypsin inhibitor. Next, the pellet was homogenized using a 10-mL Potter-Elvehjem hand homogenizer and membranes were separated by centrifugation at 14500 rpm for 20 min at 4°C. The final pellet was resuspended in membrane buffer, protein amounts were determined and membranes were used for radioligand binding assay in which membrane preparations (20 μg of protein), in a total volume of 0.25 mL, were labelled with
230 pM [125I]-CYP. This represents a near saturating concentration of CYP.
Nonspecific binding was determined using 10 μM alprenolol.
2.6. PKC ζ-mCherry translocation
HEK 293 cells were maintained in DMEM supplemented with 10% (v/v) heat-inactivated fetal bovine serum and gentamycin (100 μg/ml). Cells expressing either 2AR-CFP alone or in combination with OTR-YFP were co-transfected with either PKC-mCherry or PKCβ1-mCherry. 24h post-transfection, cells were serum-starved for an additional 24h in DMEM containing 0.2% FBS (v/v). Next, cells were pretreated for 30 min with OTA (1M) or vehicle (water) followed by treatment with either oxytocin (OT) (100nM) or isoproterenol (ISO) (10µM) for 1 min. Cells were then fixed in 4% paraformaldehyde for 5 min at room temperature, and washed with PBS. Cover slips were mounted on the slides and 86 cells were visualized on a Zeiss LSM-510 Meta laser scanning microscope using excitation at 458 nm for CFP, 515 nm for YFP and 543 nm for mCherry.
Emission was measured with the following filter sets: BP475-525 nm for CFP,
BP530-600 nm for YFP and LP560 nm for mCherry.
2.7. Data analysis
All data are presented as means ± SEM. When not shown, error bars lie within symbols. Statistical differences between paired groups were calculated using Student’s t-test and comparisons between treated cells and untreated control cells was made using one-way ANOVA with Dunnett adjustment. One-sample t- test was used to compare treatment groups to the normalized values. Differences were considered statistically significant at P < 0.05. Data were represented graphically using GraphPad Prism version 4 (GraphPad Software, San Diego,
CA). 87
3. Results
3.1. Expression of 2AR in myometrial hTERT-C3 cells
125 The amount of 2AR in hTERT-C3 cells was assessed by measuring [ I]-
CYP binding and was determined to be 50+10 fmol/mg of membrane protein
(n=3). We have previously assessed the level of OTR expressed in human myometrial hTERT-C3 cells and determined it to be 705 fmol/mg of protein
[371]. Thus we confirmed that the 2AR is expressed in human myometrial hTERT-C3 cells but at levels 14-fold lower than the OTR. This difference in expression levels appears to mirror the situation in the intact human myometrium where similar differences in relative expression levels of the two receptors have been observed [397, 398].
3.2. Kinetics of OTR- and 2AR-mediated ERK1/2 activation in hTERT-C3 cells
We first assessed the ability of the two receptors to stimulate the ERK1/2
MAP kinase pathway in hTERT-C3 cells and determined the kinetics of ERK1/2 activation using agonists for each receptor. ERK1/2 activation patterns were distinct for OTR and 2AR (Fig. 2.1). The kinetics of OTR-mediated ERK1/2 activation were prolonged, with maximum activation at 5 min but sustained activation lasting up to 3h (Fig. 2.1a,b). In contrast, the kinetics of 2AR-induced
ERK1/2 activation were more transient, with a maximum activation at 5-10 min and a complete return to baseline within 30 min (Fig. 2.1c,d). Based on these initial findings, our subsequent investigations focused on time points between 2 min and 2h. 88
3.3. Identifying G proteins responsible for ERK1/2 activation in response to OTR or 2AR activation
In order to investigate the roles of the Gi and Gq pathways in activation of the ERK1/2 pathway via the two receptors, specific inhibitors were used.
Inhibition of Gi using PTX (100ng/ml) predominantly affected the early time points (2 and 5 min) of OTR-mediated ERK1/2 activation (Fig. 2.2a,b). The inhibition of 2AR-mediated ERK1/2 signalling, upon Gi inhibition, was essentially total, suggesting that this is the primary pathway coupling the 2AR to
ERK1/2 in these cells (Fig. 2.2c,d).
To determine if known downstream effectors of Gq were also involved in 2AR-mediated activation of ERK1/2, the involvement of PKC in the signalling cascade was assessed. Interestingly, upon treatment with the broad spectrum PKC inhibitor Gö6983 (1M), ERK1/2 activation mediated via either
OTR or 2AR was inhibited substantially (Fig. 2.3). Given the complete inhibition of 2AR-mediated ERK1/2 signalling by PTX, this suggested that 2AR-mediated
PKC activation may have been downstream of Gi, rather than Gq.
3.4. Role of PKC in 2AR-mediated ERK1/2 activation in hTERT-C3 cells
To investigate which PKC isoform(s) was responsible for mediating the effects of OTR and 2AR on activation of ERK1/2, we tested the involvement of different subsets of PKC isoforms using more selective inhibitors. Gö6983 inhibits all the major classes of PKC isoforms (classical, novel and atypical), 89 while Rö31-8220 inhibits only classical and novel PKC isoforms and Gö6976 only inhibits the classical isoforms. As mentioned above, the general PKC inhibitor Gö6983 attenuated both OTR- and 2AR-mediated ERK1/2 signalling
(Fig. 2.3). Although Rö31-8220, the inhibitor of classical and novel PKC isoforms, resulted in some attenuation of OT-mediated ERK1/2 signaling, it had no significant effect on 2AR-mediated ERK1/2 signalling (Fig. 2.4a,b).
Moreover, Gö6976, the inhibitor of classical PKC isoforms, had no effect on either OT- or 2AR-mediated ERK1/2 signalling (Fig. 2.4a,b). These results suggested that an atypical PKC isoform may be involved in 2AR-mediated
ERK1/2 signalling. Therefore, we examined the involvement of atypical PKCs using a specific pseudosubstrate inhibitor of PKCζ (20M) and noted that 2AR- mediated ERK1/2 activation was attenuated in hTERT-C3 cells (Fig 2.4c,d). We also observed an attenuation of the OTR-mediated ERK1/2 signal upon PKCζ inhibition, suggesting that Gi-dependent ERK/12 activation downstream of the
OTR might use a mechanism similar to that of the 2AR (Fig 2.4c,d).
3.5. Recapitulating 2AR-mediated ERK1/2 signalling in the presence of OTR in
HEK 293 cells
In order to investigate crosstalk between OTR and 2AR in regulation of
ERK1/2 activity in more detail, we reconstituted the system in a cell line that did not endogenously express OTR. HEK 293 cells were transfected with either 2AR or OTR alone or in combination. We again assessed the role of Gi in mediating
ERK1/2 activation via each receptor. We observed that 2AR-mediated activation 90 of ERK1/2 was sensitive to inhibition of Gi by PTX only when the two receptors were co-expressed (Fig. 2.5). Similarly, we also reassessed the role of PKC in mediating ERK1/2 activation by each receptor. Our results indicated that 2AR- mediated activation of ERK1/2 was only sensitive to inhibition of PKC with
Gö6983 when OTR was co-expressed (Fig. 2.6). This suggested that the presence of the OTR (activated or not) altered the pattern of 2AR signalling in HEK 293 cells, rendering it similar to what we observed in hTERT-C3 myometrial cells.
3.6. Role of PKCζ in 2AR-mediated ERK1/2 activation in HEK 293 cells
Results obtained in hTERT-C3 cells using the PKCζ inhibitor suggested the possibility that PKC or another atypical PKC was also involved in 2AR- mediated ERK1/2 activation in the presence of the OTR in HEK 293 cells. First, we directly assessed whether PKCζ was activated upon ligand stimulation of the
2AR. PKCs translocate to the plasma membrane when activated. Thus we investigated the translocation of PKCζ using confocal microscopy in cells expressing either the 2AR alone or in combination with OTR. As shown in Fig.
2.7a, treatment with isoproterenol did not stimulate translocation of PKCζ when
2AR was expressed alone. However, when both 2AR and OTR were co- expressed in the same cells, we found that isoproterenol induced membrane translocation of PKCζ (Fig. 2.7b). Subsequently, we tested whether blocking the
OTR with an antagonist, OTA, affected isoproterenol-induced translocation of
PKCζ in these cells. As shown in Fig. 2.7b, we found that this was indeed the case. Pretreatment with OTA also blocked the oxytocin-induced translocation of 91
PKCζ (Fig. 2.7b). In contrast, we did not observe translocation of PKC1 in response to isoproterenol in cells co-expressing the two receptors (Fig. 2.7c).
These experiments confirmed that 2AR-mediated membrane recruitment of
PKCζ, but not PKC1 was dependent on the presence of the OTR.
3.7. PI 3-kinase is involved in 2AR-mediated ERK1/2 activation in hTERT-C3 cells
We investigated the effect of wortmannin, an inhibitor of PI 3-kinase (PI3K), on
2AR-mediated ERK1/2 activation. Inhibition of PI3K using wortmannin (100 nM) dramatically reduced 2AR-mediated ERK1/2 activation; however, wortmannin did not have a significant effect on OTR-mediated ERK1/2 activation
(compare Fig. 2.8a,c with 2.8b,d). Also, we again noted that ERK1/2 activation was sensitive to inhibition of PI3K only when the OTR was co-expressed with the
2AR in HEK 293 cells (Fig. 2.9). Inhibition of Src using PP2 (5M) resulted in a reduction of both OTR- and 2AR-mediated ERK1/2 activation for both early and late phases, indicating that Src is downstream of the G protein/PKC pathway (Fig.
2.10). Thus, our results indicate that there is an interaction between OTR and
2AR and that the presence of the OTR enables the 2AR to signal through a pathway involving Gi-PI3K-PKC/src to mediate ERK1/2 activation in human myometrial cells and in HEK 293 cells.
92
3.8. Neither OTR- and 2AR-mediated ERK1/2 activation in hTERT-C3 cells require EGFR transactivation
It has been demonstrated that activation of ERK1/2 via OTR can occur through transactivation of a tyrosine kinase receptor such as EGF receptor
(EGFR) [83]. We confirmed that EGF stimulated ERK1/2 activation in hTERT-
C3 cells, and this response was blocked by treatment with AG-1478 (1 M), a specific inhibitor of the EGFR (Fig. 2.11a,b). We then assessed the role of the
EGFR in transactivating ERK1/2 following either OTR or 2AR activation. In contrast, we found that treatment with AG1478 had no effect on either OTR- mediated (Fig. 2.11c,d) or 2AR-mediated (Fig. 2.11e,f) ERK1/2 activation in myometrial hTERT-C3 cells, thus ruling out transactivation of EGFR as a possible signalling pathway linking the two receptors in hTERT-C3 cells. Finally, as it is known that both 2AR and OTR recruit -arrestin [175, 395], we assessed whether -arrestin was required for activation of ERK1/2 in hTERT-C3 cells.
Using either siRNA targetting both isoforms of -arrestin or shRNA targeting - arrestin2, we did not see any effect on stimulation of ERK1/2 by OTR (Fig.
2.12a,b). 93
4. Discussion
In the present study, we describe mechanisms by which OTR and 2AR activate ERK1/2 in human myometrial hTERT-C3 cells using pharmacological manipulation of downstream receptor signalling pathways (summarized in Fig.
2.13). Importantly, the experiments conducted here indicate that these two receptors interact at physiological levels, in the context of untransfected cells. We propose a model where OTR activates ERK1/2 via Gq in a PKC-dependent manner but which can also involve Gi. 2AR-mediated ERK1/2 activation in these cells was via a PTX-sensitive, PKC-dependent pathway, which, following our reconstitution experiments in HEK 293 cells, is only operative in the presence of the OTR. Our results suggest an interaction between endogenous OTR and
2AR in hTERT-C3 cells. There is an overwhelming amount of OTR compared to
2AR (14-fold higher as measured by ligand binding) in hTERT-C3 cells. This suggests that modulation of the 2AR, functional and/or physical, in the context of receptor heterodimers (as we show in the attached companion article), would be markedly influenced by the high levels of OTR. In this OTR-dominated system,
2AR-activated PKCs in turn trigger the sequential stimulation of Src, leading ultimately to ERK1/2 activation [399].
The activation of ERK1/2 by either receptor was PTX-sensitive. The attenuation of 2AR-mediated ERK1/2 by PTX was nearly complete. This suggested that a pathway involving Gi was the primary pathway by which 2AR coupled to activate ERK1/2 in hTERT-C3 cells. Less than 50% of OTR-mediated 94
ERK1/2 activation was abolished by PTX treatment, suggesting that individual
OTR complexes might be wired differently. We then investigated the involvement of PKC pathways in ERK1/2 activation. We used a broad spectrum inhibitor of
PKC, Gö6983, and found that for the OTR, inhibition of PKC lead to significant attenuation in ERK1/2 activation. The effects of inhibition of either Gi or the
PKC pathway in OTR-mediated ERK1/2 activation were primarily seen in the early phase of ERK1/2 activation. This indicated that the mechanism by which
OTR stimulation activated ERK1/2 involved, likely to an equal extent, two distinct G proteins, Gi and the Gq, in human myometrial cells. However, the mechanism of 2AR-mediated ERK1/2 activation was different. It appeared that the principal G protein mediating the effects of 2AR was through Gi; however, we observed that PKC was involved as well.
In order to inhibit the different subsets of PKC isoforms, we used the PKC inhibitors Gö6976, which inhibits classical PKC isoforms, and Rö-31-8220, which inhibits both the classical and novel subtypes but not the atypical PKC isoforms. The PKC inhibitor we used previously, Gö6983, inhibited all PKC isoforms. We found that inhibition with either Gö6976 or Rö-31-8220 did not attenuate the 2AR-mediated ERK1/2 activation in hTERT-C3 cells. Similarly, these inhibitors did not lead to a significant decrease in OTR-mediated ERK1/2 activation. Taken together, these results implied that atypical PKCs were responsible for the 2AR-mediated ERK1/2 activation. Inhibition of PKC also demonstrated its involvement in 2AR-mediated ERK1/2 activation in hTERT-C3 cells. Our results also suggest that PKC is involved in OTR-mediated ERK1/2 95 activation. These results are consistent with other reports which show activation of ERK1/2 by a mechanism involving Gi-PI3-kinase-PKC and Src [400-403].
Further, our results suggest that the involvement of PKC in 2AR-mediated
ERK1/2 activation depends on the presence of the OTR in the same cells.
Blocking the OTR with an antagonist (OTA) also blocked the 2AR-mediated membrane recruitment of PKC. This suggests an interaction between OTR and
2AR where the presence of the OTR is important for the signalling of 2AR to activate PKC.
Next we set out to recapitulate our mechanistic observations of ERK1/2 activation in hTERT-C3 by using HEK 293 cells, where we could control the expression of each receptor. We found that the pathway for 2AR-mediated
ERK1/2 activation that we observed in myometrial cells could only be seen in the presence of both receptors in HEK 293 cells. However, we noted that OTR- mediated ERK1/2 activation was independent of the presence of 2AR, consistent with their independent action in hTERT-C3 cells, even when expressed at similar levels.
Our studies also implicate PI3-kinase and Src in 2AR-mediated ERK1/2 activation. Although Src also appears to be involved in OTR-mediated activation of ERK 1/2, PI3-kinase does not, highlighting the unique mechanism involving both receptors, potentially implicating receptor heterodimerization (discussed in more detail in a companion article). Our results excluded the involvement of - arrestin in OTR-mediated ERK1/2 activation, while EGFR transactivation was not 96 involved ERK1/2 activation by either receptor in the human myometrial cells used here. This is in contrast to previous observations that show the involvement of
EGFR in OTR-mediated activation of ERK1/2 [83]. These differences may be due to inherent differences in the cell types used for the studies. We used an immortalized human hTERT-C3 myometrial cell line while others used a transformed human PHM1 myometrial cell line. A similar interpretation can be made regarding the role of -arrestin in ERK1/2 activation in hTERT-C3 cells.
Cellular context is likely to be a key feature underlying the architecture of specific signalling pathways. We did not pursue further studies with -arrestin in HEK
293 cells as we could recapitulate the signalling pathway simply by co-expressing the two receptors. We cannot exclude that the kinetics of -arrestin recruitment to either receptor might be altered when the two receptors are expressed together.
In summary, we present evidence that in human myometrial cells, endogenous 2AR and OTR, which are both important physiological modulators of uterine activity, interact to activate ERK1/2.The present findings suggest that the novel signalling pathway described here for 2AR-mediated ERK1/2 activation is a consequence of physical interactions between the two receptors, which we explore more deeply in a companion paper. Such physical interactions as hetero-oligomerization of receptors has been shown to influence many aspects of receptor function [267, 268, 270, 391, 404-406]. Exploring the nature of the interaction between OTR and 2AR is crucial as it may be of physiological relevance for the precise coordination of contraction in myometrial cells. 97
Acknowledgements
We thank Marc-André Sylvain and Roger Tsien for the pcDNA3.1(+)-mCherry construct, Dr. Marc Caron for the -arrestin shRNA constructs and Dr. Dominic
Devost for his helpful comments. This work was supported by grants from the
Canadian Institutes of Health Research to TEH (MOP-36279) and HHZ (MOP-
74675). TEH is a Chercheur National of the Fonds de la Recherche en Santé du
Québec (FRSQ). 98
99
Figures
Figure 2.1. Kinetics of ERK1/2 activation by OTR and 2AR in human myometrial hTERT-C3 cells. hTERT-C3 cells were grown in 0.05% serum for
24h and treated with either 100 nM OT (A,B) or 10 M of ISO (C,D) for different times at 37C. Cells were lysed and levels of phosphorylated ERK1/2
(pERK1/2) were assessed by western blotting using a specific phospho-ERK1/2 antibody (pERK1/2) and reprobed with a total ERK1/2 antibody (ERK1/2).
Representative blots are shown in the top panels (A, C). Blots from 3 independent experiments were quantitated using densitometry, values for phosphorylated
ERK1/2 were normalized with respect to total ERK1/2 levels and plotted as means +/- SEM as a percentage of maximal response (5 min, B,D). *, P < 0.05 vs. control. **, P < 0.01 vs. control.
100
101
Figure 2.2. Inhibition of Gi reduces both OT- and ISO-induced ERK1/2 activation in human myometrial hTERT-C3 cells. Cells were pretreated for
18h with PTX (100 ng/ml) or vehicle water (control) and subsequently exposed to either 100 nM OT (A,B) or 10 M of ISO (C,D) for different times at 37C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, plotted relative to total ERK1/2 levels and expressed as percentage of maximal response
(5 min) as in Fig.1 (n=3). Representative blots are shown in the top panels (A,C).
*, P < 0.05 vs. corresponding control value for each time point. **, P < 0.01 vs. corresponding control value for each time point.
102
103
Figure 2.3. Inhibition of PKC reduces OT- and ISO-induced ERK1/2 activation in human myometrial hTERT-C3 cells. Cells were pretreated for 30 min with Gö6983 (1 μM), a general PKC inhibitor, or vehicle DMSO (control) and subsequently exposed to either 100 nM OT (A,B) or 10 M of ISO (C,D) for different times at 37C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, plotted relative to total ERK1/2 levels and expressed as percentage of maximal response (5 min, n=3). Representative blots are shown in the top panels (A,C). *, P < 0.05 vs. corresponding control value for each time point. **, P < 0.01 vs. corresponding control value for each time point.
104
105
Figure 2.4. Specific inhibition of PKCζ isoform reduces OT- and ISO- induced ERK1/2 activation in human myometrial hTERT-C3 cells. (A) Cells were pretreated for 30 min with either inhibitor of classical/novel PKC isoforms,
Rö31-8220 (5 M), inhibitor of classical PKC isoforms, Gö6976 (1 M) or vehicle DMSO (control) and subsequently treated with 100 nM OT or 10 M of
ISO for 5 min at 37C. Phosphorylated ERK1/2 (pERK1/2) and total ERK1/2 levels were assessed by western blotting. (B) Quantitative evaluation of 3 independent experiments represented in (A) and expressed as percentage of OT response. (C) Cells were pretreated for 30 min with specific PKCζ inhibitor
(20μM) or vehicle water (control) and subsequently treated with 100 nM OT or 10
M of ISO for 5 min at 37C, (n=3). (D) Quantitative evaluation of (C) and expressed as percentage of OT response. *, P < 0.05 vs. corresponding control value for each treatment.
106
107
Figure 2.5. Effects of Gi inhibition on ISO-induced ERK1/2 activation in presence and absence of OTR in HEK293 cells. Cells were transfected with either OTR (A) or 2AR (B) alone or in combination (C). Cells were grown in
0.05% serum 24h prior to pretreatment with PTX (100 ng/ml) or vehicle water
(control) for 18h and subsequently treated with 100 nM OT or 10 M of ISO for different times at 37C. Phosphorylated ERK1/2 (pERK1/2) and total ERK1/2 levels were assessed by western blotting and expressed as percentage of maximal response (lower panel; n=3). Representative blots are shown in the top panels of
A, B, and C. *, P < 0.05 vs. corresponding control value for each time point. **, P
< 0.01 vs. corresponding control value for each time point.
108
109
Figure 2.6. Effects of PKC inhibition on ISO-induced ERK1/2 activation in absence and presence of OTR in HEK293 cells. Cells were transfected with either OTR (A) or 2AR (B) alone or in combination (C). Cells were grown in
0.05% serum 24h prior to pretreatment with Gö6983 (1μM) or vehicle DMSO
(control) for 30 min and subsequently treated with 100 nM OT or 10 M of ISO for different times at 37C. Phosphorylated ERK1/2 (pERK1/2) and total ERK1/2 levels were assessed by western blotting and expressed as percentage of maximal response (lower panel; n=3). Representative blots are shown in the top panels of
A, B, and C. *, P < 0.05 vs. corresponding control value for each time point.
110
111
Figure 2.7. PKC recruitment to the plasma membrane following OTR and
2AR activation in HEK293 cells. Images obtained by confocal microscopy showing translocation of either PKCζ-mCherry or PKC1-mCherry to the cell membrane following ISO and OT treatment. Cells were transfected with either:
(A) 2AR-CFP and PKCζ-mCherry; or (B) 2AR-CFP, OTR-YFP and PKCζ- mCherry; or (C) 2AR-CFP, OTR-YFP and PKC1-mCherry. Cells were serum- starved for 30 min, where indicated cells were pretreated either with an OTR antagonist, OTA (100 nM), or vehicle water (control) for 15 min, and then treated with either OT (100 nM) or ISO (10 M) for 1 min. The merged panels represent
(A) YFP and mCherry, (B) and (C) CFP, YFP, and mCherry. The scale bar represents 10 μm. Results are representative of three independent experiments.
112
113
Figure 2.8. Effect of PI3K inhibition on OT- and ISO-induced ERK1/2 activity in hTERT-C3 cells. Cells were pretreated for 1h with Wortmannin (100 nM) or vehicle DMSO (control) and subsequently exposed to either 100 nM OT
(A,B) or 10 M of ISO (C,D) for different times at 37C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, plotted relative to total
ERK1/2 levels and expressed as percentage of maximal response (5 min, n=3).
Representative blots are shown in the top panels (A,C). *, P < 0.05 vs. corresponding control value for each time point. **, P < 0.01 vs. corresponding control value for each time point.
114
115
Figure 2.9. Effect of PI3K inhibition on ISO-induced ERK1/2activation in presence and absence of OTR in HEK293 cells. Cells were transfected with either OTR (A) or 2AR (B) alone or in combination (C). Cells were grown in
0.05% serum 24h prior to pretreatment for 1 h with Wortmannin (100 nM) or vehicle DMSO (control) and subsequently treated with 100 nM OT or 10 M of
ISO for different times at 37C. Phosphorylated ERK1/2 (pERK1/2) and total
ERK1/2 levels were assessed by western blotting and expressed as percentage of maximal response (5 min) (lower panel; n=3). Representative blots are shown in the top panels of A, B, and C. **, P < 0.01 vs. corresponding control value for each time point.
116
117
Figure 2.10. Effect of Src inhibition on OT- and ISO-induced ERK1/2 activity in hTERT-C3 cells. Cells were pretreated for 1h with PP2 (5 M) or vehicle DMSO (control) and subsequently exposed to either 100 nM OT (A,B) or
10 M of ISO (C,D) for different times at 37C. Cells were lysed and levels of pERK1/2 were assessed by western blotting and expressed as percentage of maximal response (5 min) (n=3). Representative blots are shown in the top panels
(A,C). *, P < 0.05 vs. corresponding control value for each time point. **, P <
0.01 vs. corresponding control value for each time point.
118
119
Figure 2.11. Effect of EGFR inhibition on OT- and ISO-induced ERK1/2 activity in hTERT-C3 cells. Cells were pretreated for 30 min with AG1478 (1
M) or vehicle DMSO (control) and subsequently exposed to either 100 nM OT
(A,B) or 10 M of ISO (C,D) for different times at 37C. (E) As a positive control for the effect of AG1478, cells were pretreated with AG1478 (1 M) or vehicle DMSO (control) for 30 min then treated with EGF (10 ng/ml) for 5 min, top panel shows a representative blot (n=3). Cells were lysed and levels of pERK1/2 were assessed by western blotting and expressed as percentage of maximal response (5 min) (n=3). Representative blots are shown in the top panels
(A,C). *, P < 0.05 vs. corresponding control value for each time point. **, P <
0.01 vs. corresponding control value for each time point. 120
121
Figure 2.12. Effect of -arrestin siRNA on OTR-mediated ERK1/2 activation in human myometrial hTERTC3 cells. Cells were transfected with either the indicated shRNAs (A) or siRNAs (B). Cells were grown in 0.05% serum 24h prior to treatment with 100 nM OT for 5 min at 37C. Phosphorylated ERK1/2
(pERK1/2) levels were assessed by western blotting (n=1).
122
123
Figure 2.13. Schematic representation of a model for the signaling pathway of ERK1/2 activation via β2AR and OTR in human myometrial hTERTC3 cells. Isoproterenol (ISO) stimulates β2AR and oxytocin (OT) stimulates OTR through their canonical Gs- and Gq-dependent pathways, respectively. In our native cell model, where both receptors are expressed endogenously, the overwhelming presence of OTR results in β2AR coupling to Gi to activate a pathway which includes PI3 kinase, and the atypical PKCζ pathway. PKCζ then stimulates the ERK1/2 cascade through a pathway involving sequential activation of Src. This β2AR-mediated signalling pathway is dependent on the presence of
OTR. The likely basis of this crosstalk is through receptor heterodimerization (the subject of a companion article). OTR activates ERK1/2 through the canonical
Gq signalling which involves PKC and Src, as well as through Gi.
124
125
CHAPTER 3
3.1. PREFACE
In the previous chapter, we described the signalling pathways involved in
OTR- and 2AR-mediated ERK1/2 activation in human hTERT-C3 myometrial cells. These studies can be most easily explained by the presence of a hetero- oligomeric complex between the two receptors. Our next step was thus to investigate potential allosteric interactions between OTR and 2AR in myometrial cells and to determine whether these two receptors did indeed form a complex. In our investigations, we found that the ability of each receptor to activate ERK1/2 in myometrial cells was modulated by different ligands for the putative partner receptor, consistent with the notion of receptor hetero-oligomers. We then characterized the putative OTR/2AR heterodimer in HEK 293 cells using several biochemical techniques.
Manuscript: Wrzal, P.K., Devost, D., Pétrin, D., Goupil, E., Iorio-Morin, C.,
Laporte, S.A., Zingg, H.H., and Hébert, T.E. Allosteric interactions between
OTR and 2AR modulate ERK1/2 activation in human myometrial cells. Cellular
Signalling. January 2012;24(1):342-350.
126
3.2. MANUSCRIPT ALLOSTERIC INTERACTIONS BETWEEN THE OXYTOCIN RECEPTOR AND THE β2-ADRENERGIC RECEPTOR IN THE MODULATION OF ERK1/2 ACTIVATION ARE MEDIATED BY HETERODIMERIZATION Paulina K. Wrzala, Dominic Devosta, Darlaine Pétrina, Eugénie Goupila, Christian Iorio-Morina, Stéphane A. Laportea,b, Hans H. Zingga,b,c* and Terence E. Héberta* a Department of Pharmacology and Therapeutics, McGill University, Montréal, Québec, H3G 1Y6, Canada b Department of Medicine, McGill University, Montréal, Québec, H3A 1A1, Canada c Department of Obstetrics and Gynecology, McGill University, Montréal, Québec, H3A 1A1, Canada
*To whom correspondence should be addressed. Dr. Terence E. Hébert, PhD., Department of Pharmacology and Therapeutics, McGill University, 3655 Promenade Sir-William-Osler, Room 1303 Montréal, Québec, H3G 1Y6, Canada Tel: (514) 398 1398; Fax: (514) 398-6690 E-mail: [email protected]
Dr. Hans H. Zingg, MD., PhD., Department of Pharmacology and Therapeutics, McGill University, 3655 Promenade Sir-William-Osler, Room 1325 Montréal, Québec, H3G 1Y6, Canada Tel: (514) 398-3621; Fax: (514) 398-2045 E-mail: [email protected]
Running Title: Heterodimer formation between the oxytocin and 2-adrenergic receptors.
Keywords: Oxytocin receptor, 2-adrenergic receptor, ERK1/2, allosteric modulation, signalling, dimerization 127
Abstract
The oxytocin receptor (OTR) and the 2-adrenergic receptor (2AR) are key regulators of uterine contraction. These two receptors are targets of tocolytic agents used to inhibit pre-term labour. Our recent study on the nature of OTR- and
2AR-mediated ERK1/2 activation in human hTERT-C3 myometrial cells suggested the presence of an OTR/2AR hetero-oligomeric complex (see companion article). The goal of this study was to investigate potential allosteric interactions between OTR and 2AR and establish the nature of the interactions between these receptors in myometrial cells. We found that OTR-mediated
ERK1/2 activation was attenuated significantly when cells were pretreated with the 2AR agonist isoproterenol or two antagonists, propranolol or timolol. In contrast, pretreatment of cells with a third 2AR antagonist, atenolol resulted in an increase in OTR-mediated ERK1/2 activation. Similarly, 2AR-mediated ERK1/2 activation was strongly attenuated by pretreatment with the OTR antagonists, atosiban and OTA. Physical interactions between OTR and 2AR were demonstrated using co-immunoprecipitation, bioluminescence resonance energy transfer (BRET) and protein-fragment complementation (PCA) assays in HEK
293 cells, the latter experiments indicating the interactions between the two receptors were direct. Our analyses suggest physical interactions between OTR and 2AR in the context of a new heterodimer pair lie at the heart of the allosteric effects.
128
1. Introduction
G protein-coupled receptors (GPCRs) make up the largest and most diverse family of transmembrane receptors encoded by the human genome, relaying information from diverse extracellular stimuli to the cell interior [1]. These receptors are also the most common targets of currently available drugs [2].
Initially, GPCRs were thought to exist only as monomers, however, during the past 20 years this view has been substantially altered. What has emerged is a view that the physiologically relevant GPCR signalling unit may consist of dimers or higher-structure oligomers of receptors [407].
Recent research suggests that GPCRs possess additional topographically distinct allosteric binding sites, recognized by structurally diverse allosteric modulators, which can regulate the activity of the endogenous (or orthosteric) ligand, changing agonist- and antagonist-mediated responses to affect signalling pathways selectively [294]. Allosteric ligands can either positively or negatively modulate receptor activity depending on the pathway being investigated, thus can act as biased ligands [294]. The existence of GPCRs as dimers conjures up the possibility of new functional properties distinct from the individual parent receptors. This combined with the fact that only about 40 GPCRs have been validated as true therapeutic targets (described until recently as agonists or antagonists at the orthosteric ligand-binding site) indicates a large potential for further drug discovery [294]. Understanding how different ligands for the parent receptors regulate a GPCR within a heterodimer, through allosteric effects 129 between the two protomers of the dimer or the protomer(s) and associated G proteins, is crucial in the development of new therapeutic strategies.
We have demonstrated a novel functional interaction between two GPCRs, the oxytocin receptor (OTR) and the 2-adrenergic receptor (2AR), in human myometrial cells (see companion article). In these studies the 2AR was coupled to a Gi/PI3-kinase-PKC-Src dependent pathway to mediate ERK1/2 activation in hTERT-C3 myometrial cells, but only in the presence of OTR. These receptors display an overlapping distribution in the human myometrium, where they have opposing effects on uterine contraction. Activation of the OTR leads to uterine contraction, while activation of the 2AR leads to uterine relaxation. In addition, the two receptors represent pharmacological targets of drugs currently used to control pre-term uterine contractions, OTR antagonists and 2AR agonists [359].
Both OTR and 2AR have been described to form homo- and heterodimers. Dimerization of OTR has actually been demonstrated ex vivo [392].
The OTR has also been shown to form heterodimers with vasopressin V1a and V2 receptors [388, 389]. The 2AR has been shown to homodimerize and to form heterodimers with several other GPCRs such as 1-and 3-adrenoceptors, the prostaglandin EP1 receptor, olfactory receptors as well as the -, - and -opioid receptors [156, 267, 268, 270, 385-387, 390, 391], among others. These studies suggest that GPCRs with multiple potential dimer partners can form distinct signalling hubs, about which unique hetero-oligomeric complexes can be built with distinct physiological partners and functions. Physical interactions between 130
GPCRs to form dimers promote changes in receptor function, which have been shown to influence ligand-binding affinity and specificity as well as downstream coupling, trafficking and desensitization of the receptors. Unraveling the nature of the interactions, which exist between OTR and 2AR, is of significant pharmacological and physiological importance. In the present study we have identified a novel physical interaction between OTR and 2AR when expressed in
HEK 293 cells, which could form the basis of unique signalling patterns seen when the receptors are co-expressed in native tissues. These interactions allowed for allosteric modulation of ERK1/2 activation mediated through one receptor partner by the other in human myometrial hTERT-C3 cells. 131
2. Materials and Methods
2.1. Reagents
Reagents were obtained from the following sources: DMEM/F12 tissue culture media was from Invitrogen (Burlington, ON); fetal bovine serum (FBS) from Hyclone (Logan, UT); OT was from Sigma-Aldrich (St. Louis, MO); isoproterenol and timolol were from Toctris Bioscience (Bristol, UK); atenolol was from Research Biochemicals International (Natick, MA). Propranolol was from Sigma-Aldrich (St. Louis, MO). OTA was obtained from Bachem (Torrence,
CA) and Atosiban was a kind gift from Ferring Research Institute Inc. (San
Diego, CA). Pertussis toxin (PTX) and protease inhibitors: leupeptin, benzamidine and trypsin were from Sigma-Aldrich (St. Louis, MO). Primary anti- phospho-ERK1/2 (T202, Y204) antibody was from Cell Signaling Technology
(Beverly, MA); pan anti- ERK1/2 antibody was from Stressgen (Ann Arbor, MI).
HA antibodies were from Covance. Flag antibody and secondary antibody, horseradish peroxidase-conjugated anti-rabbit IgG, were from Sigma-Aldrich (St.
Louis, MO). Lipofectamine 2000 was from Invitrogen (Burlington, ON). 50% slurry EZview Red Anti-Flag and Anti-mouse affinity gel beads and Flag peptide were from Sigma-Aldrich (St. Louis, MO); Coelenterazine h was from Molecular
Probes (Eugene, OR). All other analytical grade chemicals were obtained from
Sigma-Aldrich (St. Louis, MO), Fisher Scientific (Waltham, MA), and VWR
(West Chester, PA).
132
2.2. Constructs
The OTR-Rluc and 2AR-YFP constructs have been previously described [396,
408]. The GABAB2-YFP was a kind gift from Dr. Michel Bouvier (Université de
Montréal, Montreal, Canada). The HA-OTR construct was generated by PCR from human OTR using specific primers (forward: 5’-
TTATGCCTGCGGATCCGAGGGCGCGCTCGCAGCCAACT-3’, reverse 5’-
TTTAAACGCCGGATCTCACGCCGTGGATGGCTGGGA-3’). Using the In-
Fusion PCR cloning system (Clontech), the ends of the PCR fragment were fused to the homologous ends of a BamHI linearized pIRESpuro3 vector already containing the sequence for an HA tag located at the N-terminus of the receptor sequence. The Flag-β2AR construct was generated by PCR using HA-β2AR as template and the following primers; forward primer contained NheI, Kozak and
Flag sites (5’-
GCAGCTAGCGCCACCATGGATTATAAGGACGATGACGATAAGATGGG
GCAACCCGGGAAC-3’) and reverse primer contained BamHI site (5’-
CGTGGATCCTTACAGCAGTGAGTCATTTG -3’). The PCR product was digested with NheI/BamHI and subcloned into pIRESpuro3. The cloning strategy used to generate 2AR-Venus (1) and 2AR-Venus (2) has previously been described [409]. To construct plasmids coding for human OTR and rat Alk5 with a N- or C-terminal fragments of the GFP variant Venus, receptor coding sequences were amplified by PCR using the following specific primers (OTR forward : 5’-
GAATTGCGGCCGCCACCATGGAGGGCGCGCTCGCAGCCAACT-3’, 133 reverse: 5’- GCCACCATCGATCGCCGTGGATGGCTGGGAGC -3’; Alk5 forward: 5’- CTAAGCGGCCGCGCCACCATGGAGGCAGCATCGGCTGCTT
-3’, reverse 5’- GTAGCATCGATCTTGTCGTCGTCGTCCTT -3’). The resulting PCR fragments were cloned NotI and ClaI to replace the GCN4 leucine zipper from pcDNA3.1/Zeo (+)-GCN4 leucine zipper-Venus (1) and pcDNA3.1/Zeo (+)-GCN4 leucine zipper-Venus (2) cDNAs, generous gifts from
Dr. Stephen Michnick (Université de Montréal, Montreal, QC). All constructs were verified by restriction digestion and bidirectional sequencing. All constructs tagged for either BRET or the PCA were verified in functional assays to affirm that they retained the character of the untagged versions (data not shown).
2.3. Cell culture and transfection
Human myometrial cells immortalized by transfection with human telomerase reverse transcriptase (hTERT-HM cells) were obtained from W. E.
Rainey [372]. hTERT-C3 cells represent a selected subclone that we obtained by serial clonal dilution of hTERT-HM cells as described previously [371]. These cells were maintained in DMEM/F12 medium supplemented with 10% FBS and cultured at 37 C in 5% CO2. Cells close to confluency were passaged by trypsinization and plated in T175 flask at a one-quarter dilution every 4–5 d. HEK
293 cells were from Invitrogen (Burlington, ON) and were grown in Dulbecco’s modified Eagle’s medium high glucose (DMEM/high glucose) supplemented with
5% FBS and transfected using Lipofectamine 2000 as per the manufacturer’s instructions. Experiments were generally carried out 48h after transfection.
134
2.4. Western blots
Immunoblotting was used to assess the levels of cellular proteins. Briefly, cells were plated onto culture dishes and grown in appropriate media to near confluency. Cells were starved for 24h in either DMEM/F12 supplemented with only 0.5% FBS (hTERT-C3 cells) or DMEM/high glucose supplemented with only 0.5% FBS (HEK 293 cells). Cells were stimulated with ligand(s) for different times at 37° C. Stimulation was stopped by two ice-cold PBS washes, and plates were flash frozen in liquid nitrogen. For inhibitor studies, cells were pre-treated with the inhibitor in question for 30min - 1h before stimulation with ligand(s). Cells were lysed on ice with Laemmli buffer [50 mM Tris-HCl (pH
6.8), 2% sodium dodecyl sulfate, 10% glycerol, and 0.1 M -mercaptoethanol] and were either homogenized using syringe and needle (hTERT-C3 cells) or sonication (HEK 293 cells). Lysates were clarified by centrifugation at 15, 000 x g for 10 min at 4° C in a microcentrifuge. Proteins were denatured by boiling for 5 min, and subjected to SDS-PAGE and western blotting. Immunodetection involved different primary antibodies in conjunction with a second horseradish peroxidase-conjugated antibody and a chemiluminescence detection system
(Supersignal; Pierce). Quantification of band intensities was performed using
AlphaEase (Alpha Innotech Corp., San Leandro, CA).
2.5. Receptor co-immunoprecipitation
HEK 293 cells were grown and transfected as described above. All steps in membrane preparation and receptor co-immunoprecipitation procedures were done on ice and with ice-cold solutions. Membrane preparation and solubilization 135 was performed as described elsewhere with the following modifications [410].
48h post-transfection, cells were washed once with cold PBS 1X and resuspended in lysis buffer (5mM Tris-HCl pH 7.4, 2mM EDTA) containing the protease inhibitors leupeptin 5 g/ml, benzamide 10 g/ml and soybean trypsin inhibitor 5
g/ml. Cells were then homogenized using a Brinkman Polytron at 10,000 rpm for 2 x 10 s and cellular debris was removed by centrifugation for 5 min at 1000 rpm. The supernatant was transferred to 15 ml clear Sorval tubes and membranes were obtained by centrifugation for 20 min at 16,000 rpm. Membranes (pellet) were then solubilized at 4 °C for 30 min in solubilization buffer (75 mM Tris pH
8.0, 2 mM EDTA, 5 mM MgCl2, 0.5% n-Dodecyl -D-maltoside) containing protease inhibitors. The soluble and insoluble membrane fractions were then separated by centrifugation for 20 min at 18,300 rpm and supernatant was used for co-immunoprecipitation. 40 l/sample of either a 50% slurry of EZview Red
Anti-Flag or Anti-HA affinity gel beads were washed twice with 500 l of solubilization buffer (centrifuged for 2 min at 1600 rpm between washing steps).
Solubilized membranes (400-500 g/l) were then added to the beads and rocked overnight at 4° C. The next day, beads were pulled down by centrifugation for 2 min at 1600 rpm and washed twice with solubilization buffer and once with TBS
1X (rocking for 5 min, centrifugation 2 min at 1600 rpm). Then the appropriate elution peptide (Flag or HA peptide at 50 ng/l in TBS1X) was used to elute bound precipitates from beads for 30 min at 4° C. Beads were separated from elutes by centrifugation for 30s at 8200 x g. Eluants were subjected to western blot analysis. 136
2.6. Bioluminescence resonance energy transfer (BRET)
To detect and analyze interactions between 2AR and OTR, BRET was used as previously described [409, 411]. For this purpose, HEK 293 cells were seeded in 6-well plates and transfected with a fixed amount of an OTR construct tagged at its C terminus with Renilla luciferase (OTR-RLuc), and co-transfected with an increasing amount of plasmids encoding 2AR or GABA-B2 (the latter as a negative control), both tagged at their C-terminus with YFP (2AR-YFP,
GABAB2-YFP). Cells were collected 48h post-transfection. After the addition of the substrate coelenterazine h, emission was measured using an injector-equipped plate-reader spectrofluorometer (Fluostar Optima, BMG LabTechnologies) at the wavelengths of 475 nm and 535 nm, corresponding to the maxima of the emission spectra for RLuc and YFP, respectively. The BRET ratio was calculated as described [385]: BRET ratio = [(emission at 510–590 nm) − (emission at 440–500 nm) × Cf]/(emission at 440–500 nm), where Cf corresponded to (emission at 510–
590 nm)/(emission at 440–500 nm) for cells expressing OTR-RLuc alone. BRET ratios were plotted as means of 10 or 40 consecutive measurements taken at 0.4- to 0.5-sec intervals.
2.7. Protein-fragment complementation assays (PCA)
We also investigated interactions between 2AR and OTR with the use of a protein-fragment complementation assay (PCA), initially described in [409,
412]. For the purposes of our experiments, HEK 293 cells were seeded in 6- well plates and transfected with equal amounts of receptors of interest tagged with split Venus constructs in the following combinations (OTR-Venus1 and 137
OTR-Venus2, 2AR-Venus1 and 2AR-Venus2, OTR-Venus 1 and 2AR-
Venus 2, or OTR-Venus1 and Alk5-Venus2). For confocal microscopy, cells were maintained in DMEM supplemented with 10% (v/v) heat-inactivated fetal bovine serum and gentamycin (100 μg/ml) and grown on 35 mm plates. 48h post-transfection cells were starved in serum-free media for 30 min. Images were collected using live-cell microscopy at 37 °C on a Zeiss LSM-510 Meta laser scanning microscope equipped with XL-3 temperature chamber with a
63× glycerol/water immersion lens in single track mode using excitation at 515 nm for YFP and emission measured with the BP530-600 filter set. For direct measurement of reconstituted Venus levels, cells were collected 48h post- transfection, placed in 96-well white OptiPlate-96 plates(Perkin Elmer) and fluorescence from reconstituted Venus was measured directly using a Synergy2 microplate reader (BioTek), with excitation 485/20 nm and emission 528/20 nm filters.
2.8. Data analysis
All data are presented as means ± SEM. When not shown, error bars lie within symbols. Statistical differences between paired groups were calculated using Student’s t-test and comparisons between treated cells and untreated control cells was made using one-way ANOVA with Dunnett adjustment. One-sample t- test was used to compare treatment groups to the normalized maxima.
Differences were considered statistically significant at P < 0.05. Data were represented graphically using GraphPad Prism version 4 (GraphPad Software, San
Diego, CA). 138
3. Results
3.1. Molecular crosstalk between OTR and 2AR in hTERTC3 cells: role of PKA
In a companion paper, we showed results identifying functional interactions between the OTR and the 2AR in immortalized human myometrial cells. These studies suggested the possibility of physical interactions between the two receptors, although the molecular nature of these interactions was not fully elucidated. If indeed the two receptors interacted as part of a hetero-oligomeric complex, then it should be possible to detect allosteric interactions between the two distinct ligand binding sites which might translate into altered signalling with respect to a shared effector pathway, ERK1/2. To investigate possible allosteric interactions, we examined the effects of a 2AR agonist, isoproterenol, on the ability of the OTR to activate ERK1/2 in our myometrial cell model. We found that OTR-mediated ERK1/2 activation was strongly attenuated following pretreatment with isoproterenol (Fig. 3.1). We next wondered if the effects of isoproterenol on OTR-mediated signalling could be attributed to second messenger-based, crosstalk between the two receptors. PKA has already been implicated in inhibition of phospholipase C (PLC) downstream of Gαq-mediated signalling in myometrial cells [393]. Thus we assessed the putative role of PKA in crosstalk between OTR and 2AR. We noted that inhibition of PKA with PKI 6-
22A lead to a slight increase in OTR-mediated ERK1/2 activation; however, such treatment was not able to reverse the attenuation of OTR-mediated ERK1/2 activation following pretreatment with isoproterenol (Fig. 3.1). Further, direct 139 activation of adenylyl cyclase (AC) with forskolin, had an effect that partially mimicked the effect of isoproterenol on OTR-mediated ERK1/2 activation (Fig.
3.2). Taken together, these data suggest that canonical crosstalk between signalling pathways could only partially account for the effects of 2AR on OTR- mediated signalling.
3.2. Modulation of signalling by different classes of OTR and 2AR ligands in myometrial hTERTC3 cells
We wanted to further explore the effects of different classes of AR ligands on OTR-mediated ERK1/2 activation in hTERT-C3 cells and began by investigating the effects of the non-specific AR antagonist propranolol.
Surprisingly, ERK1/2 activation induced by oxytocin (OT) was again strongly attenuated following pretreatment with propranolol (Fig. 3.3). Interestingly, we noted that propranolol pretreatment did not attenuate the subsequent, yet more modest (compared with OT) isoproterenol-mediated ERK1/2 activation. It has been previously shown that although propranolol is an antagonist for 2AR- mediated activation of AC, it is a partial agonist for 2AR-mediated ERK1/2 activation [238, 413]. To exclude this potential confound due to the biased nature of propranolol effects on the 2AR, we next used timolol, which has been characterized as an antagonist for both AC and ERK1/2 downstream of 2AR activation [210]. Similar to propranolol, OTR-mediated ERK1/2 activation was attenuated following pretreatment with timolol (Fig. 3.4a,b). Interestingly, although there were alterations in the basal levels of ERK1/2 activation, timolol 140
also blocked 2AR-mediated activation of ERK1/2 (Fig. 3.4c,d). We next investigated the effects of a third ligand, in this case the 2AR inverse agonist, atenolol, on OTR-mediated ERK1/2 activation. Here, we found that pretreatment of cells with atenolol resulted in no significant decrease in OTR-mediated
ERK1/2 activation; in fact there was a trend toward an increased signal (Fig. 3.5).
Curiously, atenolol showed the same stimulatory trend for 2AR-stimulated
ERK1/2 activation, suggesting that it too acts as a biased agonist for this signalling pathway. These experiments demonstrated that different 2AR ligands, independent of their ability to modulate cAMP formation (and subsequently activate PKA) were capable of modulating signalling via the OTR, possibly through allosteric effects mediated through a 2AR/OTR hetero-oligomer.
To confirm and extend these experiments, we next investigated the effects of different OTR ligands on 2AR-mediated ERK1/2 activation. We did not use oxytocin here as it swamped out any effects of 2AR stimulation. However, we noted that 2AR-mediated ERK1/2 activation was attenuated following pretreatment with the OTR antagonist atosiban (Fig. 3.6a,c). A similar result was observed with the OTR antagonist OTA (Fig. 3.7a,c). As expected, both OTR antagonists also blocked OT-mediated ERK1/2 activation (Figs. 3.6b,d and
3.7b,d). The results with the different classes of receptor ligands cannot all be mediated through second messenger-mediated crosstalk between the two receptors. These results again strongly indicate that allosteric interactions occur between endogenous OTR and 2AR modulating their signalling capabilities, 141 likely through the agency of a receptor hetero-oligomer.
3.3. Physical interactions between 2AR and OTR in HEK 293 cells
In order to assess such potential physical interactions between 2AR and
OTR, we used several experimental strategies. First, we determined whether these two receptors could be co-immunoprecipitated when co-expressed in HEK 293 cells. We noted that HA-OTR and Flag-2AR co-immunoprecipitated only when these receptors were expressed in the same cells but not when membranes were mixed from cells expressing each receptor alone (Fig. 3.8). Several features of the putative 2AR/OTR complex are evident from these experiments. First, both mature and immature forms of the OTR were co-immunoprecipitated with the
2AR. This suggests that the interactions (as for most GPCR oligomers) occur in the ER during receptor biosynthesis. This is also indicated by the fact that mixtures of solubilized receptors from cells expressing each receptor alone, do not interact. Also, both monomeric and dimeric versions of HA-OTR were co- immunoprecipitated suggesting that these arrays may be multimeric in nature.
We next used bioluminescence resonance energy transfer (BRET) to confirm the results obtained by co-immunoprecipitation of 2AR and OTR in
HEK 293 cells. BRET was measured from cells co-transfected with OTR-Rluc and 2AR-YFP. The BRET signal was saturable in the presence of increasing amounts of acceptor and a fixed level of donor (Fig. 3.9a). This was in contrast to
GABAB2-YFP, which expressed at similar levels (as measured by measuring
YFP expression, data not shown) but did not saturate the OTR-Rluc donor, 142
suggesting that the OTR/2AR interaction was specific (Fig. 3.9a). Thus, BRET experiments also demonstrated that the two receptors were part of the same complex but could not confirm whether or not the interactions were direct.
Finally, a protein complementation assay (PCA), based on the reconstitution of Venus GFP, was used to confirm the results obtained using
BRET and to demonstrate that interactions between the receptors were direct.
Reconstitution of a correctly folded Venus moiety was detected when either OTR-
Venus1 and OTR-Venus2 (homodimer) or 2AR-Venus1 and 2AR-Venus2
(homodimer) were co-expressed either using confocal microscopy or in direct measurements of reconstituted fluorescence in a microplate reader (Fig. 3.9b,c).
Venus fluorescence was also detected from HEK 293 cells co-transfected with
OTR-Venus1 and 2AR-Venus2, whereas Alk5-Venus2 and OTR-Venus did not reconstitute Venus when co-expressed (Fig. 3.9b,c). Taken together, our results complement those obtained in the companion study showing a functional interaction between the two receptors by further demonstrating direct molecular association likely involving allosteric interactions between the two receptors when regulating a common effector molecule.
143
4. Discussion
In the present study, we demonstrated that the ligands of OTR and 2AR influence the ability of either partner receptor to activate ERK1/2 in human myometrial cells. We suggest that allosteric interactions exist between the endogenous OTR and 2AR, where one receptor modulates the signalling ability of the other. The allosteric nature of these interactions, under native conditions with endogenous receptors, led us to suppose that the two receptors interacted in the context of a heterooligomer. The ability of a 2AR agonist to reduce the OTR- mediated ERK1/2 activation was not, in and of itself, a novel finding. It has previously been shown that activation of cAMP-dependent protein kinase (PKA) can phosphorylate and inhibit phospholipase C, thus attenuating the signalling from Gq-coupled GPCRs [414, 415]. Consistent with previous observations, we found that activation of AC via forskolin partially attenuated OTR-mediated
ERK1/2 activation. However, using a specific inhibitor of PKA, we showed that this effect was not fully PKA-dependent. Two possible explanations exist for these observations: (1) an interaction between the two receptors, which cannot be explained by crosstalk at the level of second messengers, (2) crosstalk that is cAMP-dependent, but PKA-independent. Based on these results, and our study showing functional interactions between OTR and 2AR (see the associated companion article), we investigated the effects of 2AR antagonists on the ability of OTR to activate ERK1/2. Our findings showed that both propranolol and timolol also attenuated OTR-mediated ERK1/2 activation, but in a manner that clearly was independent of the generation of second messengers. Further, we 144
showed that a 2AR inverse agonist, atenolol, had no effect on the ability of OTR to activate ERK1/2. Further evidence for allosteric effects comes from the fact that two distinct OTR antagonists, which do not activate second messenger- dependent pathways, inhibited 2AR-mediated ERK1/2 signalling as well. This is particularly important as we demonstrated in the companion article that 2AR signals through a novel pathway which strictly depends on the presence of the
OTR. We also noted that the effects of the 2AR ligands used in this study on the signalling of the 2AR itself in human myometrial cells was different then what has been reported in HEK 293 cells [210, 238]. We found that timolol did not strongly block isoproterenol-mediated ERK1/2 activation [210]. Similarly, atenolol has been previously shown to be an inverse agonist for the ERK1/2 signalling pathway as it has been shown to block PMA-induced ERK1/2 activation [238], however, in our studies it did not attenuate isoproterenol- mediated ERK1/2 activation. This may be attributed to the cell type-specific differences which may have clear implications for future studies of allosterism and biased agonism. As we have evidence to suggest a heterodimer formation between OTR and 2AR, the effects of the 2AR ligands on 2AR-mediated
ERK1/2 activation we have seen here may be due to the unique pharmacology of this receptor when associated with OTR in the myometrium.
If we consider the existence of physical interactions between OTR and
2AR then perhaps the effects of the different OTR and 2AR ligands can be attributed to the stabilization of distinct conformations of one partner receptor by the ligand of the other. It has been shown that GPCRs can adopt several different 145 types of active states or conformations, with different ligands able to stabilize distinct states leading to the activation of unique signalling pathways via biased agonism [416, 417]. Our results demonstrate that different classes of ligands for a given receptor can affect partner receptors in different ways. A recent study has shown that the two-receptor equivalents in the context of a D2 dopamine receptor homodimer are organized asymmetrically with respect to their G protein partners such that occupation by ligand of one receptor activates the receptor and occupation of the other and thus modulates signalling allosterically [418]. In the context of a homodimer this may not be as important as either receptor can serve each role and such asymmetries might not be detected (see [419] for a discussion of this issue). However, for receptor hetero-oligomers, such asymmetries may have important consequences. Using the example, of a heterodimer between the
2AR and the -opioid receptor, or between the 2AR and OTR described here, an asymmetry with respect to how the complex is arranged may mean that in one case, depending on how the complex is formed, we might have a 2AR modulated by -opioid or OTR ligands and in another case a -opioid receptor or an OTR modulated by 2AR ligands [270, 386]. Thus, in one arrangement, the first receptor is the signalling receptor and second becomes an allosteric modulator and the converse is true when the system is organized the other way around. This greatly increases the potential organizational complexity of GPCR signalling and suggests that cell-specific determinants of signalling complex assembly will be of paramount importance in initially defining signalling specificity in a given tissue, cellular or subcellular compartment [244, 294]. This has important implications 146 for the formation of receptor heterodimers and hetero-oligomers, in that multiple asymmetrical arrangements are possible depending on the relative orientation of each monomer to the G protein and possibly to the effectors. More diversity is added when we consider heterotetramers which might have variable numbers of each component subunit or different potential arrangements of those subunits.
Receptor complexes can also contain multiple receptors, what some authors have termed as receptor mosaics [420]. For example, a number of recent studies have suggested that GPCRs can form higher order complexes. Protein complementation approaches have now been used to confirm and extend our knowledge regarding dimerization and oligomerization of GPCRs. Reconstitution of split luciferase (Gaussia or Renilla) and split GFP constructs have shown that dimers of 2AR and D2 dopamine receptors can be detected, complementing immunopurification and BRET approaches, and these approaches can be combined to detect and examine larger complexes [409, 421]. A number of investigators have used three partner PCA/RET to show that higher order complexes of GPCRs such as the A2A-adenosine receptor homo- and hetero- oligomers with CB1 cannabinoid/D2 dopamine receptors and CXCR4 multimers can be detected [422-426]. These larger arrays may in fact represent complicated allosteric machines which can be assembled by different cells in different ways, greatly expanding the combinatorial power of GPCR signalling. Our data presented here supports the notion that these complexes are formed during receptor biosynthesis. 147
The results presented in this study, taken together with the companion article, show that the formation of heterodimers between OTR and 2AR has functional consequences on the ability of each receptor to activate ERK1/2 in human myometrial cells. The fact that the signalling phenotype described in hTERT-C3 cells, where the two receptors are expressed endogenously, was recapitulated in HEK 293 cells expressing the receptors suggests that they interact in the same way, likely in the context of receptor heteroligomers. Though beyond the scope of the present article, it would certainly be of interest to demonstrate these interactions directly in hTERT-C3 cells using labelled ligands or antibodies by FRET. This suggests that the OTR/2AR hetero-oligomers have distinct signalling profiles and pharmacology which differ from either OTR or 2AR alone. In addition to providing novel lines of evidence for direct functional and allosteric interactions between two members of the GPCR family, these findings are of specific potential relevance for the development of novel tocolytic approaches involving these two receptor systems. 148
Acknowledgements
We thank Irina Glazkova for the Flag-2AR construct and Carlis Rejon for the
Alk5-Venus2 construct. This work was supported by grants from the Canadian
Institutes of Health Research (CIHR) to TEH (MOP-36279), to HHZ (MOP-
74675) and a CIHR Research Team Grant in GPCR Allosteric Regulation
(CTiGAR, CTP-79848). TEH is a Chercheur National of the Fonds de la
Recherche en Santé du Québec (FRSQ). 149
Figures
Figure 3.1. β2AR agonist isoproterenol (ISO) inhibits OTR-mediated ERK1/2 signalling in human myometrial cells. (A) hTERT-3 cells were grown in 0.05% serum for 24h then were pretreated with ISO (10 μM) or vehicle (control) for 15 min and subsequently exposed to OT (100 nM) for different times at 37C. (B)
Quantitative evaluation of 3 independent experiments as shown in (A). (C) Cells were pretreated for 1h with PKA inhibitor PKI 6-22A (5µM) or 15 min with ISO
(10 μM) or both or vehicle water (control) and subsequently treated with OT (100 nM) for different times at 37C. (D) Quantitative evaluation of 3 independent experiments as shown in (C). Cells were lysed and levels of phosphorylated
ERK1/2 (pERK1/2) and total ERK1/2 were assessed by western blotting. Values for phosphorylated ERK1/2 were normalized with respect to total ERK1/2 levels and plotted as means +/- SEM as a percentage of maximal response (5 min). *, P
< 0.05 vs. corresponding control value for each time point.
150
151
Figure 3.2. Forskolin treatment attenuates OTR-mediated ERK1/2 activation in human myometrial hTERTC3 cells. Cells were grown in 0.05% serum 24h prior to pretreatment with forskolin (5M) or vehicle (control) for 15 min and subsequently exposed to 100 nM OT for different times at 37C. Cells were lysed and levels of phosphorylated ERK1/2 (pERK1/2) and total ERK1/2 levels were assessed by western blotting (n=2).
152
153
Figure 3.3. β2AR antagonist propranolol (PRO) inhibits OTR-mediated
ERK1/2 signalling. hTERT-C3 cells were pretreated with PRO (10 μM) or vehicle (control) for 15 min and subsequently exposed to either 100 nM OT (A,B) or 1 M of ISO (C,D) for different times at 37C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, normalized to total ERK1/2 and expressed as percentage of maximal response as in Fig. 1 (n=3). Representative blots are shown in the top panels (A,C). *, P < 0.05 vs. corresponding control value for each time point. **, P < 0.01 vs. corresponding control value for each time point.
154
155
Figure 3.4. β2AR antagonist timolol (TIM) inhibits OTR-mediated ERK1/2 signalling. hTERT-C3 cells were pretreated with TIM (10 μM) or vehicle
(control) for 15 min and subsequently exposed to either 100 nM OT (A,B) or 1
M of ISO (C,D) for different times at 37C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, normalized to total ERK1/2 and expressed as percentage of maximal response (n=3). Representative blots are shown in the top panels (A,C). *, P < 0.05 vs. corresponding control value for each time point.
156
157
Figure 3.5. Effect of β2AR inverse agonist atenolol (AT) on OTR-mediated
ERK1/2 signalling. hTERT-C3 cells were pretreated with AT (10 μM) or vehicle
(control) for 15 min and subsequently exposed to 100 nM OT (A,B) or 1 M of
ISO (C,D) for different times at 37C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, normalized to total ERK1/2 and expressed as percentage of maximal response (n=3). Representative blots are shown in the top panels (A,C). *, P < 0.05 vs. corresponding control value for each time point.
158
159
Figure 3.6. OTR antagonist atosiban inhibits β2AR-mediated ERK1/2 signalling. hTERT-C3 cells were pretreated with atosiban (100 nM) or vehicle
(control) for 15 min and subsequently exposed to 10 M of ISO (A,B) or 1 nM
OT (C,D) for different times at 37C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, normalized to total ERK1/2 and expressed as percentage of maximal response (n=3). Representative blots are shown in the top panels (A,C). *, P < 0.05 vs. corresponding control value for each time point. **,
P < 0.01 vs. corresponding control value for each time point.
160
161
Figure 3.7. OTR antagonist OTA inhibits β2AR-mediated ERK1/2 signalling. hTERT-C3 cells were pretreated with either OTA (100 nM) or vehicle (control) for 15 min and subsequently exposed to 10 M of ISO (A,B) or 1 nM OT (C,D) for different times at 37C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, normalized to total ERK1/2 and expressed as percentage of maximal response (n=3). Representative blots are shown in the top panels (A,C). *, P < 0.05 vs. corresponding control value for each time point.
162
163
Figure 3.8. β2AR and OTR co-immunoprecipitation in HEK 293 cells.
Western blot analysis of immunoprecipitates of total lysates of HEK293 cells transiently transfected HA- and Flag-tagged receptors. Cells were transfected with either pcDNA vector (lanes 1,6 and 11), HA-OTR (lanes 2, 7 and 12) or Flag-
β2AR (lane 3, 8 and 13) alone; or with both receptors in combination (lanes 4, 9 and 14). As a control co-immunoprecipitation was assessed in a mixture of two cell populations transfected separately with either HA-OTR or Flag-β2AR (lanes
5, 10 and 15). Immunoprecipitation (IP) was performed using either an anti-Flag antibody, either immunoblotted (IB) with anti-HA (lanes 1-4) or anti-Flag (lanes
11-15). Total lysates are shown in lanes 6-10. Arrows indicate bands corresponding in molecular mass to the different forms of HA-OTR (left side) and
β2AR (right side). Molecular mass markers (on the left) are in kDa. Images are representative of three independent experiments.
164
165
Figure 3.9. β2AR and OTR interactions shown using BRET and PCA in HEK
293 cells. (A) For BRET experiments, cells were transfected with a fixed amount of OTR-Rluc and co-transfected with an increasing amount of β2AR-YFP (). As a negative control cells were transfected with fixed amount of OTR-Rluc and increasing amounts of GABAB2-YFP (). Each point represents the mean +/-
SEM, n=3. (B,C) For PCA experiments, cells were transfected with the following pairs: OTR-Venus1 and β2AR-Venus2, OTR-Venus1 and OTR-Venus2, β2AR-
Venus1 and β2AR-Venus2. As a negative control cells were transfected with
OTR-Venus1 and Alk5-Venus2. Reconstitution of Venus was assessed by either confocal microscopy (B) and by direct measurement of Venus fluorescence (C).
Data are presented as mean +/- SEM (n = 3). *, P < 0.05 vs. control. 166
167
CHAPTER 4 – GENERAL DISCUSSION
4.1. SUMMARY
The overall objective of my thesis work was to characterize interactions between the OTR and the 2AR in the human myometrial cells. I have shown that both receptors activate ERK1/2 in human myometrial cells and that 2AR- mediated ERK1/2 activation occurs by a novel signalling mechanism which is dependent on the presence of OTR in the same cells. I have also shown that ligands for both OTR and 2AR were able to allosterically modulate each other at the level of ERK1/2 activation in human myometrial cells. Finally, I report heterodimer formation between OTR and 2AR. These findings are important in understanding how signalling complexes such as this can result in a unique pharmacology in cellulo which has implications for the coordination of contraction in the human myometrium as well as expanding our understanding regarding the complex and intricate signalling networks built around GPCRs.
The mechanisms by which the OTR and 2AR activate ERK1/2 have been previously investigated [82, 83, 85, 206, 296, 297, 300, 310-313, 427]. The OTR has been shown to activate ERK1/2 mainly via the Gq-PLC-PKC pathway [82,
83, 311]. However, this receptor was also found to activate Gi in CHO cells and pregnant rat myometrium [82, 310]. Moreover, in cultured myometrial cells,
OTR-mediated ERK1/2 activation was shown to be Gi-dependent [311]. Finally,
OTR-mediated ERK1/2 phosphorylation has been reported to involve both PKC and Gi-Gβγ pathways as well as transactivation of the epidermal growth factor 168 receptor (EGFR) in PMH1 myometrial cells [83]. Most recently, it has been shown that the scaffolding protein -arrestin2 is involved in OT-mediated
ERK1/2 activation in HEK 293 cells overexpressing OTR as well as in immortalized human myometrial ULTR cells [312, 313]. Studies investigating
β2AR-mediated ERK1/2 activation suggest the involvement of G subunits rather than the Gs subunit [296]. However, other reports show that β2AR activated ERK1/2 occurs via a Gs-PKA-dependent pathway [297]. Additional studies both in cultured cell lines and in vitro have demonstrated that, in response to agonist, the β2AR can undergo PKA-dependent phosphorylation leading to a loss of Gs signalling and a switch to activation of Gi [85]. β2AR-mediated
ERK1/2 activation was also reported to involve both G protein and β-arrestin components [206]. Other findings suggest that Src activation plays a role in β2AR stimulation of ERK1/2 [300, 427].
Although OTR-mediated activation of ERK1/2 has been studied in human myometrial cells, relatively little is known about how 2AR activates ERK1/2 in the human myometrium. In chapter 2, I describe a novel signalling pathway by which 2AR activates ERK1/2 in human myometrial cells. This signalling cascade involves Gi-PI3K-PKC and Src and depends on the presence of the
OTR and could be recapitulated in HEK 293 cells when the two receptors were co-expressed. It has previously been shown that 2AR is able to couple to Gi in
HEK 293 cells, following phosphorylation of the receptor by protein kinase A
(PKA) [85]. Typically, phosphorylation of the 2AR at the C-terminal domain by
G protein-coupled receptor kinase 2 (GRK2) leads to the recruitment of 169 scaffolding proteins such as -arrestin and targets the receptor for clathrin- mediated internalization [428, 429]. In addition to its role in receptor desensitization and internalization, -arrestins have also been reported to be involved in 2AR-mediated ERK1/2 activation [206]. GPCR stimulation can also activate phosphoinositide 3-kinases (PI3Ks) [430]. Both the protein and lipid kinase activities of PI3K, a class IB PI3K isoform, were shown to be important for 2AR internalization [408] Thus, stimulation of 2AR and subsequent Gi- mediated ERK1/2 activation may require internalization of the receptor into endosomes. Although, my preliminary results using knockdown of -arrestins revealed no change in the patterns of ERK1/2 activation, investigating the potential involvement of receptor internalization with respect to the spatiotemporal aspects of ERK1/2 activation might further our understanding of the different mechanisms used to achieve signal specificity and diversity.
The results presented in chapter 2 are consistent with other reports which show activation of ERK1/2 by a mechanism involving Gi-PI3-kinase-PKC and
Src by diverse GPCRs [400-403, 431]. For example, the endothelin ETA receptor
(ETAR) subtype has been shown to activate ERK1/2 via both PTX-sensitive and - insensitive G proteins activating atypical and classical/novel PKCs, respectively, in rat myometrial cells. Activation of Src, downstream of PKC was required to activate ERK1/2 in both cases. PKC , , and have been shown to directly interact with Src kinases [432-434]. Further, PKC, along with Ca2+, was shown to activate Src via Pyk2, a focal adhesion kinase, that recruits and activates Src directly [337, 435]. In addition to ERK1/2, activation of Ras by the ETAR was 170 found to be dependent on Src, as well as PKC activity, in rat myometrial cells
[403]. Interestingly, in myometrial cells from prepubertal rats, Gi- and Gq- dependent signalling pathways contribute approximately equally to the activation of ERK1/2. However, in rat puerperal myometrial cells at the end of gestation, the Gq pathway no longer contributes to ERK1/2 activation and instead is devoted to modulation of the contractile machinery [436]. These studies suggest that the signalling pathways leading to ERK1/2 activation change during gestation and the mechanisms regulating ERK1/2 are quite complex and may depend not only on the stimulus and the cell type but also on the physiological status of the tissue in question.
Further, activation of ERK1/2 by 2AR in myometrial cells represents a signalling pathway that could involve a unique complex of proteins which may have consequences which oppose the effects of 2AR-mediated Gs signalling and generation of cAMP in the human myometrium. Unlike Gs-mediated signalling, coupling to Gi can lead to activation of PKC. In chapter 2, I showed that 2AR was able to activate PKC, in particular, the atypical PKC. It remains to be determined if this mechanism, and the particular signalling components are unique to the myometrium, although my experiments in HEK 293 cells suggests this is not the case, as simply co-expressing the two receptors was sufficient to induce the 2AR to engage the Gi-mediated pathway. The molecular mechanisms which lead to the activation of ERK1/2 via OTR and 2AR may have important implications for uterine contractions and in other cells where they might be co- expressed. It is possible that a branch of the 2AR-mediated signalling pathway 171 activates a distinct pool of ERK1/2, and may function to suppress contractions or alternatively contribute to their production in an as yet unidentified manner. The mechanisms of OTR-mediated ERK1/2 activation may also reveal different branches of OTR signalling which may have unique effects on uterine contractions, either opposing or enhancing them. Understanding such mechanisms and how they relate to particular physiological consequences may present therapeutic opportunities for the treatment of pre-term labour. Targeting a unique signalling pathway, for instance, a 2AR-mediated signalling pathway which contributes to the initiation of uterine contractions, or an OTR-mediated signalling pathway which reduces 2AR-mediated uterine relaxation, may yield new and promising drug targets. Additionally, investigating, in parallel, other signalling pathways, for example by assessing levels of cAMP and IP3, calcium influx, and myometrial contraction, may reveal that each receptor activates several signalling pathways, each with its own specific physiological consequences, which may operate with distinct phenotypic outcomes, in some cases seemingly in opposition with each other.
It is worthwhile to note that the kinetics of ERK1/2 activation by the two receptors are distinct. OTR displayed early peak stimulation (2-5 min) and lower but sustained levels of ERK1/2 activation for up to 3h. In contrast, 2AR- mediated ERK1/2 activation was transient (2-5 min). In some instances, differential inhibition was observed between the different time points. As mentioned earlier, this may suggest that two distinct pathways operate downstream of the receptors, responsible for the rapid and sustained phases. 172
Studies investigating the kinetic and spatial characteristics of ERK1/2 activation by GPCRs have shown that the rapid and transient ERK1/2 activation was G protein-dependent and led to nuclear translocation of the activated ERK1/2, meanwhile, the slower more sustained ERK1/2 activation was arrestin2- dependent and targets were confined to the cytoplasm [203, 205, 206]. These different pools of activated ERK1/2 likely have consequences with respect to the substrates in the different compartments of the cell (nucleus vs. cytosol). An understanding of the substrates of ERK1/2 activation could provide new insights into the cellular responses controlled by each receptor, especially when we consider that these receptors have opposing effects on myometrial contraction, yet both activate ERK1/2. This might be achieved by phosphoproteomic profiling using high-performance liquid chromatography (HPLC) and mass spectroscopy
(MS) techniques. These techniques separate proteins based on their mass and charge and can be used to identify sites of phosphorylation on a protein from shifts in mass [437, 438]. A spatiotemporal analysis of ERK1/2 substrates activated by either OTR or 2AR (or both) may provide more clues about unique or combined roles of each receptor in maintaining a balance between uterine contraction and relaxation in the human myometrium.
In order to study the signalling mechanisms of ERK1/2 activation via OTR and 2AR in human myometrial cells, inhibitors of various signalling proteins were used. A problem associated with the use of inhibitors in the assessment of cellular signalling pathways is specificity. Some inhibitors are specific, while others may have significant off-target effects by altering the activity of additional 173 proteins. Our use of different cell lines and the collection of evidence using different methods confirmed results obtained with protein inhibitors. Knock- down of target signalling proteins, including each of the receptors in the myometrial cells by RNAi could represent another method to test the involvement of different signalling pathways in how these two receptors activate ERK1/2.
However, the uncertainty associated with off-target effects also has to be considered for RNA interference techniques. In order to ensure that RNAi experiments are informative, it is important to introduce proper controls, which should include the rescue of the expression of the target protein and a reversal in the observed phenotype associated with expression of target RNAi, as well as the validation of RNAi results using independent methods such as the use of small molecule inhibitors or antibodies. Other factors to consider when conducting
RNAi studies include the degree of RNAi achieved and the threshold at which the decrease in protein levels mediates a response [439]. Functional redundancy and the upregulation of compensatory proteins or pathways may influence the interpretation of results especially in studies where long-term, stable expression of
RNAi is used. Simply considering the challenges of RNAi and ensuring reproducibility in results can make this technique a powerful tool for the molecular analysis of cellular signalling pathways.
In chapter 3, I demonstrated physical interactions between 2AR and OTR in HEK 293 cells. Three different techniques were used to demonstrate heterodimer formation between these two receptors: co-immunoprecipitation, bioluminescence resonance energy transfer (BRET) and a GFP-based protein- 174 fragment complementation assay (PCA). The use of multiple techniques to test for the occurrence of physical interactions between receptors gives confidence to the results, even if there are limitations associated with each individual technique.
This set of experiments was performed in HEK 293 cells transfected with tagged
OTR and 2AR. Proper controls must be included in these experiments in order to ensure the observed interactions between receptors are specific and not simply due to molecular crowding. The tagged proteins were tested to ensure that they functioned the same way as the native proteins. The main limitation with the
GFP-based PCA is that once the reconstitution of the Venus protein takes place, upon the close physical proximity of the receptors, it is essentially permanently associated [440]. Thus, the dynamics of the complex cannot be measured.
However, this technique does report on the presence of an interaction between proteins and where particular complexes localize in the cell.
Collectively, my results suggested the existence of a physical interaction between OTR and 2AR and each technique provided a specific set of information about the receptor complex. Although the two receptors are present as a complex on the cell surface, they likely form a complex during their biosynthesis and may form higher order oligomers. It would be of interest to map out the domains of each receptor that are important for the formation of the OTR/2AR complex, which could be done using the same techniques with truncated and/or mutated versions of each receptor. Finally, assessment of the trafficking of the complex might provide important information on the lifecycle of such protein complexes and the signalling hubs they form. The 2AR has been shown to interact with a 175 number of other GPCRs [267-270, 385, 386, 441] as well as numerous non GPCR proteins such as adaptor protein Grb2 [442], tyrosine kinase Src [443, 444], eukaryotic initiation factor 2B (eIF-2B) [445], PKA-associated anchoring proteins
(AKAPs) [446-450], Na+/H+ exchanger regulatory factor 1 and 2 (NHERF-1/2)
[451, 452], N-ethylmaleimide-sensitive factor (NSF) [453] and several voltage- and ligand-gated ion channels [454-456]. The 2AR was also reported to interact with non GPCR receptors like the epidermal growth factor receptor (EGFR) which results in modulation of its activity [457-459]. The number of distinct
2AR interacting proteins and the variety of functional outcomes associated with them suggests that, much like scaffolding proteins, the 2AR may act as a platform, bringing together different signalling proteins under different conditions and in different cells, and thus acting as a signalling hub. Specificity in signalling may depend on the different proteins in a particular signalling complex as well as the localization of such complexes in distinct parts of the cell. The exact mechanisms which confer such specificity still need to be determined. However, the assembly, organization and trafficking of GPCR signalling complexes has been shown to involve a number of proteins including G subunits (reviewed in
[108]) or additional GPCR-interacting proteins (GIPs) (reviewed in [460]).
We also showed functional interactions between OTR and 2AR in human myometrial cells, whereby the ligands of each partner receptor influenced the signalling of the other in ways that were inconsistent with simple molecular crosstalk. Chronic exposure to oxytocin has been shown to desensitize pathways that stimulate adenylyl cyclase in rat myometrial cells and this was shown to 176 involve stimulation of PKC [461] and stimulation of different PKC isoforms has been shown to modulate 2AR signalling [462]. The reverse has also been
2+ reported, whereby 2AR-activated PKA inhibits OTR-induced increase in Ca and phosphoinisotol turnover as well as GTPS-stimulated PLC activity [384,
414, 463, 464]. In line with these observations, I show in chapter 3 that pretreatment of human myometrial cells with isoproterenol resulted in decreased
OTR-mediated activation of ERK1/2. However, I found that the inhibitory effect of isoproterenol was only partially PKA-mediated. OTR-mediated activation of
ERK1/2 was also attenuated when cells were pretreated with the 2AR antagonists propranolol and timolol. However, pretreatment with the 2AR antagonist, atenolol resulted in an increase in OTR-mediated ERK1/2 activation. Similarly,
2AR-mediated ERK1/2 activation was decreased by pretreatment with the OTR antagonists, atosiban and OTA. These findings indicate that the inhibitory effect on OTR and 2AR signalling that results from blocking the 2AR or the OTR respectively cannot be fully explained by crosstalk at the level of the second messengers, suggesting that additional mechanisms of interactions between the two receptors be considered. Nevertheless, it would complement our studies to investigate the role of PKA-independent cAMP in the inhibition of OTR-mediated
ERK1/2 activation when cells are pretreated with isoproterenol, as this cannot be excluded by my experiments presented here.
Based on the results presented in chapter 2 and the evidence in chapter 3 suggesting physical interactions between OTR and 2AR, I suggest that the functional interactions observed between the two receptors must be interpreted in 177 the context of a receptor heterodimer or higher order oligomer. There are several studies which suggest that heterodimerization of GPCRs results in new properties of the receptors [256-258, 261-263, 267-269, 272, 273]. These properties may include changes in ligand binding, as observed between chemokine receptors in a hetero-oligomeric complex [263]. They may also include changes in G protein coupling, as in the case of the heterodimer between dopamine D2 receptor (D2R) and cannabinoid CB1 receptor (CB1R), where a switch from CB1R coupling from
Gi to Gs has been observed upon co-activation of D2R [465, 466]; or G protein activation, which is increased for the angiotensin 1 receptor (AT1R) in the AT1R- bradykinin 2 receptor (B2R) heterodimer [258]. Alterations in internalization and signalling of receptors as a consequence of heterodimerization have also been reported. For example, 2AR association with 1AR and 3AR has been shown to reduce the rate of agonist-induced internalization as well as the ability of 2AR to activate ERK1/2 [267, 268]. Finally, changes in interactions with -arrestin were also seen in response to heterodimer formation. Heterodimerization of somatostatin receptor type 2 (SSTR2) and type 5 (SSTR5) was observed to decrease the interaction of SSTR2 with -arrestin [467]. My data provide information on the unique pharmacology of the OTR/2AR complex in human myometrial cells. Future studies should include testing the effects of the different
OTR and 2AR ligands on myometrial cell contraction. It would be also be informative to assess the receptor-ligand affinity of each receptor in the complex in the presence of the agonists and/or antagonists of the partner receptor using radioligand binding assays for each receptor in order to investigate allosteric 178 communication between the receptors at the level of the receptor-ligand interactions.
Lastly, in preliminary experiments, I also tested whether the implications of the functional studies with OTR and 2AR had broader application in the human myometrium. We investigated the functional interactions between either OTR or
2AR and other receptors present in the myometrium such as the prostaglandin
F2 receptor (FP) and the angiotensin II receptor (AT1R). I found that pretreatment with OTR and 2AR ligands influenced the ability of FP and AT1R to activate ERK1/2 in hTERT-C3 myometrial cells. I noted that pretreatment with
2AR ligands isoproterenol (agonist), propranolol (antagonist) and timolol
(antagonist) significantly attenuated both FP- and AT1R-mediated ERK1/2 activation (Fig. 4.1). Curiously, the 2AR inverse agonist atenolol had differential effects on the activation of ERK1/2 by FP and AT1R. Pretreatment with atenolol resulted in an attenuation of AT1R-mediated ERK1/2 activation, however, did not change the ability of FP to activate ERK1/2 (Figure 4.1). In chapter 3, I show that pretreatment with atenolol did not reduce the ability of OTR to activate ERK1/2.
Complex formation between the 2AR and AT1R has previously been shown
[271]. 179
Figure 4.1. Effects of β2AR ligands on FP- and AT1R-mediated ERK1/2 signalling. hTERT-C3 cells were grown in 0.05% serum for 24h then were pretreated with vehicle (control), ISO (10 μM), atenolol (AT, 10 μM), propranolol
(PRO, 10 μM), or timolol (TIM, 10 μM) for 15 min and subsequently exposed to
1 μM PGF2 (A,B) or 100 nM angiotensin II (C,D) for 5 min at 37C Cells were lysed and levels of phosphorylated (pERK1/2) and total ERK1/2 were assessed by western blotting. Values for phosphorylated ERK1/2 were normalized with respect to total ERK1/2 levels and plotted as means +/- SEM as a percentage of maximal response (5 min) (n=3). Representative blots are shown in the top panels
(A,C). *, P < 0.05 vs. corresponding control value for each time point. **, P <
0.01 vs. corresponding control value for each time point. 180
181
Figure 4.2. Effects of OTR antagonists on FP- and AT1R-mediated ERK1/2 signalling. hTERT-C3 cells were pretreated with vehicle (control), OTA (100 nM), or atosiban (100 nM) for 15 min and subsequently exposed to 1 μM PGF2
(A) or 100 nM angiotensin II (B) for 5 min at 37C. Cells were lysed and levels of pERK1/2 and total ERK1/2 were assessed by western blotting (n=2). 182
183
However, these preliminary results may suggest that perhaps, like the OTR and the AT1R, the FP also forms a complex with 2AR and if so, the 2AR may modulate different receptors in various ways, which may be important for their unique function in the cell. Further, we performed preliminary experiments (n=2) to assess the effect of OTR antagonists on FP- and AT1R-mediated ERK1/2 activation in human myometrial cells. We found that pretreatment with OTR antagonist atosiban attenuated FP-mediated ERK1/2 activation by approximately
20% while the OTR antagonist OTA had no effect (Figure 4.2). Our preliminary data also indicate that pretreatment with the OTR antagonists OTA or atosiban results in an attenuation of the AT1R-mediated ERK1/2 activation (Figure 4.2.).
These results suggest that interactions similar to those detected between OTR and
2AR may also be present between OTR or 2AR and both FP and the AT1R in the human myometrium. These observations should be investigated in more detail in order to determine the nature and implications of such functional interactions.
GPCR signalling has been proposed to involve large signalling complexes dependent on specific protein-protein interactions between signalling partners, some of which may begin early after their synthesis. By controlling protein interaction possibilities, the organization of a given signalling complex can influence the type of signal transduction it might produce. Therefore, one must consider the possibility of the presence of distinct populations of receptors, both with respect to different locations of receptors in the cell as well as different times 184 in the life of the cell, whereby in the nature of these complexes might impact signalling and thus engender different physiological consequences.
There is a significant amount of evidence which suggests that GPCR signalling complexes are assembled during their biosynthesis. Stable complexes between receptors, G proteins and effectors such as adenylyl cyclase isoforms,
Ca2+ activated K+ channels, L-type Ca2+ channels, and inwardly rectifying K+ channels, have been shown to exist as early as in the endoplasmic reticulum (ER)
[411, 455, 468-474]. For instance, the 2AR was found associate with both Kir3 ion channels and adenylyl cyclase [468, 475]. There are several proteins which have been shown to be involved in the organization, assembly and trafficking of
GPCR signalling complexes. G subunits represent one such “organizing factor”, that through its interactions with a wide variety of proteins, may facilitate the assembly of signalling complexes before their targeting to the plasma membrane [108]. In addition to interactions of receptor with the G subunits, several studies have shown that receptors can directly interact with the G subunits [476-478], and such interactions precede those with G and have been shown to initiate in the ER [475, 479, 480]. When the function of G is blocked using a membrane-localized version of the carboxy terminus of the G protein- coupled receptor kinase 2 (GRK2) (ARK-CT), formation of Kir3.1- G complexes in the ER is abolished, as is the formation of 2AR homodimers [108].
However, the roles that individual G subunits play in the expression and assembly of signalling complexes, and where in the cell these events take place, need to be clarified [108]. 185
Other key proteins involved in regulating signalling complexes are regulators of G protein signalling (RGS) proteins which also form interactions with receptors, G proteins and effector molecules [481-484]. For example, RGS4 has been shown to directly bind to G and effector proteins such as phospholipase C-1 or heteromeric Kir3.1/Kir3.4 channels [485, 486]. While,
RGS12 was found to complex with the 1B subunit of the G protein-regulated N- type Ca2+ channel [487]. Further, there are a number of scaffolding proteins that interact with GPCRs and may be involved in regulating GPCR signalling complex organization and/or localization [460, 488, 489]. For instance, the 2AR associates with the Na+-H+ exchanger regulatory factor proteins NHERF-1 and -2, a scaffold protein containing a PDZ domain, which links this receptor to signalling pathway regulating Na+-H+ exchanger [451, 452, 490]. Interactions between 2AR and the A-kinase anchoring proteins (AKAP)-250 (gravin) and
AKAP79 were shown to promote receptor association with PKA and receptor phosphorylation, respectively, regulating receptor desensitization and downstream signalling [446, 448-450]. AKAPs anchor PKA to specific locations in the cell, such as the plasma membrane, where it can activate distinct pools of cAMP [491,
492]. Finally, -arrestins have also been implicated in compartmentalization of
GPCR signalling complexes leading to the activation of specific signalling pathways discussed previously [162, 342].
Aside from interactions with “chaperone” proteins that enable the assembly and organization of GPCR signalling complexes, proteins involved in the trafficking of these complexes may be equally important in determining the 186 destination of the complex and thus the signalling pathways which may unfold.
Ras-related GTPases of the Rab family are known to regulate endocytosis, intracellular sorting, transport to lysosomes, and recycling to the plasma membrane [493, 494]. A recent study reported an interaction between Rab5 and
Go, which is important for planar cell polarity and Wingless signal transduction through the Frzzled (Fz) family of GPCRs in Drosophila. The recycling Rab4 and
Rab11 proteins functioned in Fz- and Go-mediated signaling to favor planar cell polarity over canonical Wingless signaling. The interplay between heterotrimeric
G proteins and Rab GTPases controlled receptor internalization, revealing a previously uncharacterized regulatory mechanism in GPCR signalling [495].
Additionally, the localization of specific signalling complexes in the plasma membrane has been proposed to involve lipid rafts. Lipid rafts are dynamic nanometer scale membrane domains and are formed by tightly packed cholesterol with saturated acyl chain lipids that leads to the formation of a liquid-ordered phase. These lipid raft domains separate the embedded proteins from the rest of the membrane. Several reports have proposed that cholesterol-enriched membrane lipid domains may influence both OTR and 2AR signal transduction properties by controlling the association with their signalling partners [76, 496].
Assessment of the role of cholesterol in the signalling of these two receptors, the interactions between the receptors and with their ligands would further our understanding about how signals are compartmentalized in myometrial cells.
Cholesterol may be important for protein folding and receptor transport to the plasma membrane. Populations of OTR and 2AR residing in cholesterol-rich 187 domains may transduce ligand binding via different signalling cascades as compared to receptors localized in domain with different cholesterol levels.
Aside from sorting signalling complexes in terms of location, the formation of such complexes may also be influenced by factors which modulate protein expression throughout the life of the cell. Modulation of receptor expression may represent a mechanism by which the receptor milieu in the cell changes in favour of particular signals, in addition to tissue and cell specific expression of other protein which are part of the signalling complex.
Investigating gene expression profiles of human myometrial cells with respect to the expression of receptors alone and in combination as well as in response to the different ligands of either OTR and/or 2AR may give insight into the regulation of receptor expression as well as the expression of other proteins in these cells.
Regulation of receptor expression may also be influenced by hormones acting on the myometrium. Several hormones have been shown to regulate OTR, including estrogen, progesterone and oxytocin itself [497-504]. The rat OTR promoter has also been shown to contain a cAMP response element and 2AR stimulation by isoproterenol was shown to increase rat myometrial OTR gene expression [505,
506]. Further, treatment of rats in labour with a 2AR antagonist, ICI-118,551, resulted in a decrease in OTR levels in isolated myometrial plasma membranes, as did pretreatment with an OTR antagonist, atosiban [383]. Further, the altered expression of cAMP response element binding protein (CREB), cAMP response element modulator protein (CREM) and activating transcription factor 2 (ATF2), which has been reported to change between non-pregnant, pregnant non-labouring 188 and spontaneously labouring women, might have substantial effects in terms of gene expression in human myometrial cells [507]. Thus, in addition to modulation of receptor function which might occur in the setting of a heterodimer, there may be other mechanisms, which may control the very existence of such heterodimers and in turn the signals they produce.
4.2. CONCLUSIONS
This study, which investigated the molecular mechanisms of ERK1/2 activation by the OTR and the 2AR in human myometrial cells as well as the functional and physical interactions between these receptors, has prompted a number of questions. It would be of interest to map out how the activation of
ERK1/2 by each receptor influences contractions in the human myometrial cells and if they can be allosterically modulated by the ligands of the partner receptor.
Understanding how the different signalling pathways activated by the OTR and the 2AR are sorted inside the cell and how signalling complexes such as the
OTR/2AR are formed would greatly benefit future design of novel therapeutics for pre-term labour as well as other GPCR combinations. The fact that the pharmacology of the oligomer is different from that of each receptor homodimer is quite significant and such complexes should be taken into consideration during drug development.
189
4.3. REFERENCES
1. Lagerstrom, M.C. and H.B. Schioth, Structural diversity of G protein- coupled receptors and significance for drug discovery. Nature Reviews Drug Discovery, 2008. 7(4): p. 339-357. 2. Overington, J.P., B. Al-Lazikani, and A.L. Hopkins, How many drug targets are there? Nature Reviews Drug Discovery, 2006. 5(12): p. 993- 996. 3. Palczewski, K., et al., Crystal structure of rhodopsin: a G protein-coupled receptor. Science, 2000. 289: p. 739-745. 4. Cherezov, V., et al., High-resolution crystal structure of an engineered human β2-adrenergic G protein–coupled receptor. Science, 2007. 318(5854): p. 1258-1265. 5. Rasmussen, S.G.F., et al., Crystal structure of the human β2-adrenergic G- protein-coupled receptor. Nature, 2007. 450(7168): p. 383-387. 6. Jaakola, V.-P., et al., The 2.6 Angstrom crystal structure of a human A2 adenosine receptor bound to an antagonist. Science, 2008. 322(5905): p. 1211-1217. 7. Warne, T., et al., Structure of a β1-adrenergic G-protein-coupled receptor. Nature, 2008. 454(7203): p. 486-491. 8. Chien, E.Y.T., et al., Structure of the human dopamine D3 receptor in complex with a D2/D3 selective antagonist. Science, 2010. 330(6007): p. 1091-1095. 9. Wu, B., et al., Structures of the CXCR4 chemokine GPCR with small- molecule and cyclic peptide antagonists. Science, 2010. 330(6007): p. 1066-1071. 10. Rasmussen, S.G.F., et al., Structure of a nanobody-stabilized active state of the β2 adrenoceptor. Nature, 2011. 469(7329): p. 175-180. 11. Rasmussen, S.G.F., et al., Crystal structure of the β2 adrenergic receptor- Gs protein complex. Nature, 2011. 477(7366): p.549-555. 12. Attwood, T. and J. Findlay, Fingerprinting G-protein-coupled receptors. Protein Engineering, 1994. 7(2): p. 195-203. 13. Kolakowski, L.J., GCRDb: a G-protein-coupled receptor database. Receptors Channels, 1994. 2(1): p. 1-7. 14. Fredriksson, R., et al., The G-protein-coupled receptors in the human genome form five main families. Phylogenetic analysis, paralogon groups, and fingerprints. Molecular Pharmacology, 2003. 63(6): p. 1256-1272. 15. Gether, U., Uncovering molecular mechanisms involved in activation of G protein-coupled receptors. Endocrine Reviews, 2000. 21(1): p. 90-113. 16. Rosendorff, A., B.J. Ebersole, and S.C. Sealfon, Conserved helix 7 tyrosine functions as an activation relay in the serotonin 5HT2C receptor. Molecular Brain Research, 2000. 84(1-2): p. 90-96. 17. Lu, Z.-L., J.W. Saldanha, and E.C. Hulme, Transmembrane domains 4 and 7 of the M1 muscarinic acetylcholine receptor are critical for ligand binding and the receptor activation switch. Journal of Biological Chemistry, 2001. 276(36): p. 34098-34104. 190
18. Rovati, G.E., V. Capra, and R.R. Neubig, The highly conserved DRY motif of class A G protein-coupled receptors: beyond the ground state. Molecular Pharmacology, 2007. 71(4): p. 959-964. 19. Wong, S., G protein selectivity is regulated by multiple intracellular regions of GPCRs. Neurosignals, 2003. 12(1): p. 1-12. 20. Favre, N., et al., The DRY motif as a molecular switch of the human oxytocin receptor. Biochemistry, 2005. 44(30): p. 9990-10008. 21. Jüppner, H., et al., The extracellular amino-terminal region of the parathyroid hormone (PTH)/PTH-related peptide receptor determines the binding affinity for carboxyl-terminal fragments of PTH-(1-34). Endocrinology, 1994. 134(2): p. 879-84. 22. Bergwitz, C., et al., Full activation of chimeric receptors by hybrids between parathyroid hormone and calcitonin. Journal of Biological Chemistry, 1996. 271(43): p. 26469-26472. 23. Grace, C.R.R., et al., NMR structure and peptide hormone binding site of the first extracellular domain of a type B1 G protein-coupled receptor. Proceedings of the National Academy of Sciences of the United States of America, 2004. 101(35): p. 12836-12841. 24. Lisenbee, C.S., M. Dong, and L.J. Miller, Paired cysteine mutagenesis to establish the pattern of disulfide bonds in the functional intact secretin receptor. Journal of Biological Chemistry, 2005. 280(13): p. 12330-12338. 25. Hoare, S., Mechanisms of peptide and nonpeptide ligand binding to class B G-protein-coupled receptors. Drug Discovery Today, 2005. 10(6): p. 417-427. 26. Parthier, C., et al., Passing the baton in class B GPCRs: peptide hormone activation via helix induction? Trends in Biochemical Sciences, 2009. 34(6): p. 303-310. 27. Gardella, T.J., et al., Determinants of [Arg2]PTH-(1-34) binding and signaling in the transmembrane region of the parathyroid hormone receptor. Endocrinology, 1994. 135(3): p. 1186-94. 28. Turner, P.R., T. Bambino, and R.A. Nissenson, Mutations of neighboring polar residues on the second transmembrane helix disrupt signaling by the parathyroid hormone receptor. Molecular Endocrinology, 1996. 10(2): p. 132-9. 29. Hoare, S.R.J., T.J. Gardella, and T.B. Usdin, Evaluating the signal transduction mechanism of the parathyroid hormone 1 receptor. Journal of Biological Chemistry, 2001. 276(11): p. 7741-7753. 30. Miller, L., et al., Structural basis of natural ligand binding and activation of the class II G-protein-coupled secretin receptor. Biochemical Society Transactions, 2007. 35(4): p. 709-712. 31. Masu, M., et al., Sequence and expression of a metabotropic glutamate receptor. Nature, 1991. 349(6312): p. 760-765. 32. Hermans, E. and R.A. Challiss, Structural, signalling and regulatory properties of the group I metabotropic glutamate receptors: prototypic family C G-protein-coupled receptors. Biochemical Journal, 2001. 359(3): p. 465-484. 191
33. Pollak, M.R., et al., Mutations in the human Ca2+-sensing receptor gene cause familial hypocalciuric hypercalcemia and neonatal severe hyperparathyroidism. Cell, 1993. 75(7): p. 1297-1303. 34. Francesconi, A. and R.M. Duvoisin, Role of the second and third intracellular loops of metabotropic glutamate receptors in mediating dual signal transduction activation. Journal of Biological Chemistry, 1998. 273(10): p. 5615-5624. 35. Mary, S., et al., A cluster of basic residues in the carboxyl-terminal tail of the short metabotropic glutamate receptor 1 variants impairs their coupling to phospholipase C. Journal of Biological Chemistry, 1998. 273(1): p. 425-432. 36. Chang, W., et al., Amino acids in the second and third intracellular loops of the parathyroid Ca2+-sensing receptor mediate efficient coupling to phospholipase C. Journal of Biological Chemistry, 2000. 275(26): p. 19955-19963. 37. McKnight, A. and S. Gordon, The EGF-TM7 family: unusual structures at the leukocyte surface. Journal of Leukocyte Biology, 1998. 63(3): p. 271- 280. 38. Stacey, M., et al., LNB-TM7, a group of seven-transmembrane proteins related to family-B G-protein-coupled receptors. Trends in Biochemical Sciences, 2000. 25(6): p. 284-289. 39. Robertson, H.M., Two large families of chemoreceptor genes in the nematodes Caenorhabditis elegans and Caenorhabditis briggsae reveal extensive gene duplication, diversification, movement, and intron loss. Genome Research, 1998. 8(5): p. 449-463. 40. Mombaerts, P., Seven-transmembrane proteins as odorant and chemosensory receptors. Science, 1999. 286(5440): p. 707-711. 41. Bockaert, J. and J. Philippe Pin, Molecular tinkering of G protein-coupled receptors: an evolutionary success. The EMBO Journal, 1999. 18(7): p. 1723-1729. 42. O'Dowd, B.F., et al., Palmitoylation of the human 2-adrenergic receptor. Mutation of Cys341 in the carboxyl tail leads to an uncoupled nonpalmitoylated form of the receptor. Journal of Biological Chemistry, 1989. 264(13): p. 7564-7569. 43. Tota, M.R. and C.D. Strader, Characterization of the binding domain of the -adrenergic receptor with the fluorescent antagonist carazolol. Evidence for a buried ligand binding site. Journal of Biological Chemistry, 1990. 265(28): p. 16891-16897. 44. Kim, J., et al., Glutamate residues in the second extracellular loop of the human A2 adenosine receptor are required for ligand recognition. Molecular Pharmacology, 1996. 49(4): p. 683-691. 45. Dixon, R., et al., Structural features required for ligand binding to the - adrenergic receptor. The EMBO Journal, 1987. 6(11): p. 3269-3275. 46. Kobilka, B., et al., Chimeric 2-,2-adrenergic receptors: delineation of domains involved in effector coupling and ligand binding specificity. Science, 1988. 240(4857): p. 1310-1316. 192
47. Rosenbaum, D.M., et al., GPCR engineering yields high-resolution structural insights into β2-adrenergic receptor function. Science, 2007. 318(5854): p. 1266-1273. 48. Rosenbaum, D.M., et al., Structure and function of an irreversible agonist- 2 adrenoceptor complex. Nature, 2011. 469(7329): p. 236-240. 49. Kolb, P., et al., Structure-based discovery of β2-adrenergic receptor ligands. Proceedings of the National Academy of Sciences of the United States of America, 2009. 106(16): p. 6843-6848. 50. Strader, C.D., et al., Structure and function of G protein-coupled receptors. Annual Review of Biochemistry, 1994. 63(1): p. 101-132. 51. Liapakis, G., et al., The forgotten serine. Journal of Biological Chemistry, 2000. 275(48): p. 37779-37788. 52. Suryanarayana, S. and B.K. Kobilka, Amino acid substitutions at position 312 in the seventh hydrophobic segment of the 2-adrenergic receptor modify ligand-binding specificity. Molecular Pharmacology, 1993. 44(1): p. 111-114. 53. Wieland, K., et al., Involvement of Asn-293 in stereospecific agonist recognition and in activation of the 2-adrenergic receptor. Proceedings of the National Academy of Sciences, 1996. 93(17): p. 9276-9281. 54. Strader, C.D., et al., Identification of residues required for ligand binding to the beta-adrenergic receptor. Proceedings of the National Academy of Sciences of the United States of America, 1987. 84(13): p. 4384-4388. 55. Sansuk, K., et al., A structural insight into the reorientation of transmembrane domains 3 and 5 during family A G protein-coupled receptor activation. Molecular Pharmacology, 2011. 79(2): p. 262-269. 56. Breitwieser, G.E., G protein–coupled receptor oligomerization. Circulation Research, 2004. 94(1): p. 17-27. 57. Jastrzebska, B., Y. Tsybovsky, and K. Palczewski, Complexes between photoactivated rhodopsin and transducin: progress and questions. Biochemical Journal, 2010. 428(1): p. 1-10. 58. Walker, P., et al., Acidic residues in extracellular loops of the human Y1 neuropeptide Y receptor are essential for ligand binding. Journal of Biological Chemistry, 1994. 269(4): p. 2863-2869. 59. Leong, S.R., R.C. Kabakoff, and C.A. Hébert, Complete mutagenesis of the extracellular domain of interleukin-8 (IL-8) type A receptor identifies charged residues mediating IL-8 binding and signal transduction. Journal of Biological Chemistry, 1994. 269(30): p. 19343-19348. 60. Liapakis, G., et al., Identification of ligand binding determinants in the somatostatin receptor subtypes 1 and 2. Journal of Biological Chemistry, 1996. 271(34): p. 20331-20339. 61. AbdAlla, S., et al., The N-terminal amino group of [Tyr8]bradykinin is bound adjacent to analogous amino acids of the human and rat B2 receptor. Journal of Biological Chemistry, 1996. 271(44): p. 27382-27387. 62. Marie-Pepin, C., et al., Novel “restoration of function” mutagenesis strategy to identify amino acids of the δ-opioid receptor involved in ligand binding. Journal of Biological Chemistry, 1997. 272(14): p. 9260-9267. 193
63. Heerding, J.N., et al., Mutational analysis of the angiotensin II type 2 receptor: contribution of conserved extracellular amino acids. Regulatory Peptides, 1997. 72(2-3): p. 97-103. 64. Cotte, N., et al., Identification of residues responsible for the selective binding of peptide antagonists and agonists in the V2 vasopressin receptor. Journal of Biological Chemistry, 1998. 273(45): p. 29462-29468. 65. Malitschek, B., et al., The N-terminal domain of γ-aminobutyric acid B receptors is sufficient to specify agonist and antagonist binding. Molecular Pharmacology, 1999. 56(2): p. 448-454. 66. Hawtin, S.R., H.C. Howard, and M. Wheatley, Identification of an extracellular segment of the oxytocin receptor providing agonist-specific binding epitopes. Biochemical Journal, 2001. 354(2): p. 465-472. 67. Wesley, V.J., et al., Agonist-specific, high-affinity binding epitopes are contributed by an arginine in the N-terminus of the human oxytocin receptor. Biochemistry, 2002. 41(16): p. 5086-5092. 68. Fanelli, F., et al., Activation mechanism of human oxytocin receptor: a combined study of experimental and computer-simulated mutagenesis. Molecular Pharmacology, 1999. 56(1): p. 214-225. 69. Elands, J., et al., 125I-labelled d(CH2)5[Tyr(Me)2,Thr4,Tyr-NH29]OVT: a selective oxytocin receptor ligand. European Journal of Pharmacology, 1988. 147(2): p. 197-207. 70. Postina, R., E. Kojro, and F. Fahrenholz, Separate agonist and peptide antagonist binding sites of the oxytocin receptor defined by their transfer into the V2 vasopressin receptor. Journal of Biological Chemistry, 1996. 271(49): p. 31593-31601. 71. Breton, C., et al., Direct identification of human oxytocin receptor-binding domains using a photoactivatable cyclic peptide antagonist. Journal of Biological Chemistry, 2001. 276(29): p. 26931-26941. 72. Gimpl, G., et al., Binding domains of the oxytocin receptor for the selective oxytocin receptor antagonist barusiban in comparison to the agonists oxytocin and carbetocin. European Journal of Pharmacology, 2005. 510(1-2): p. 9-16. 73. Yang, M., et al., Lysine 270 in the third intracellular domain of the oxytocin receptor is an important determinant for Gαq coupling specificity. Molecular Endocrinology, 2002. 16(4): p. 814-823. 74. Zhong, M., et al., Residues in the hydrophilic face of putative helix 8 of oxytocin receptor are important for receptor function. Biochemistry, 2004. 43(12): p. 3490-3498. 75. Zhong, M., et al., Amino acids in the COOH-terminal region of the oxytocin receptor third intracellular domain are important for receptor function. American Journal of Physiology - Endocrinology and Metabolism, 2007. 292(4): p. E977-E984. 76. Gimpl, G., et al., Oxytocin receptors: ligand binding, signalling and cholesterol dependence, in Progress in Brain Research, D.N. Inga and L. Rainer, Editors. 2008, Elsevier. p. 193-204. 77. Milligan, G. and E. Kostenis, Heterotrimeric G-proteins: a short history. British Journal of Pharmacology, 2006. 147(S1): p. S46-S55. 194
78. Hurowitz, E.H., et al., Genomic characterization of the human heterotrimeric G protein α, β, and γ subunit genes. DNA Research, 2000. 7(2): p. 111-120. 79. Hermans, E., Biochemical and pharmacological control of the multiplicity of coupling at G-protein-coupled receptors. Pharmacology & Therapeutics, 2003. 99(1): p. 25-44. 80. Arnaudeau, S., N. Lapretre, and J. Mironneau, Oxytocin mobilizes calcium from unique heparin-sensitive and thapsigargin-sensitive store in singla myometrial cells from pregnant rats. Pflügers Archiv, 1994. 428(1): p. 51- 59. 81. Ku, C., et al., Oxytocin stimulates myometrial guanosine triphosphatase and phospholipase-C activities via coupling to Gq/11. Endocrinology, 1995. 136(4): p. 1509-1515. 82. Strakova, Z., et al., ERK2 mediates oxytocin-stimulated PGE2 synthesis. American Journal of Physiology - Endocrinology and Metabolism, 1998. 274(4): p. E634-E641. 83. Zhong, M., M. Yang, and B.M. Sanborn, Extracellular signal-regulated kinase 1/2 activation by myometrial oxytocin receptor involves GqG and epidermal growth factor receptor tyrosine kinase activation. Endocrinology, 2003. 144(7): p. 2947-2956. 84. Daaka, Y., L.M. Luttrell, and R.J. Lefkowitz, Switching of the coupling of the 2-adrenergic receptor to different G proteins by protein kinase A. Nature, 1997. 390(6655): p. 88-91. 85. Lefkowitz, R.J., K.L. Pierce, and L.M. Luttrell, Dancing with different partners: protein kinase A phosphorylation of seven membrane-spanning receptors regulates their G protein-coupling specificity. Molecular Pharmacology, 2002. 62(5): p. 971-974. 86. Zamah, A.M., et al., Protein kinase A-mediated phosphorylation of the β2- adrenergic receptor regulates its coupling to Gs and Gi. Journal of Biological Chemistry, 2002. 277(34): p. 31249-31256. 87. Lawler, O.A., S.M. Miggin, and B.T. Kinsella, Protein kinase A-mediated phosphorylation of serine 357 of the mouse prostacyclin receptor regulates its coupling to Gs-, to Gi-, and to Gq-coupled effector signaling. Journal of Biological Chemistry, 2001. 276(36): p. 33596-33607. 88. Miggin, S.M. and B.T. Kinsella, Investigation of the mechanisms of G protein: effector coupling by the human and mouse prostacyclin receptors. Journal of Biological Chemistry, 2002. 277(30): p. 27053-27064. 89. Martin, N.P., et al., PKA-mediated phosphorylation of the 1-adrenergic receptor promotes Gs/Gi switching. Cellular Signalling, 2004. 16(12): p. 1397-1403. 90. Sato, M., et al., The β3-adrenoceptor agonist 4- [[(Hexylamino)carbonyl]amino]-N-[4-[2-[[(2S)-2-hydroxy-3-(4- hydroxyphenoxy)propyl]amino]ethyl]-phenyl]-benzenesulfonamide (L755507) and antagonist (S)-N-[4-[2-[[3-[3- (Acetamidomethyl)phenoxy]-2-hydroxypropyl]amino]- ethyl]phenyl]benzenesulfonamide (L748337) activate different signaling 195
pathways in chinese hamster ovary-K1 cells stably expressing the human β3-adrenoceptor. Molecular Pharmacology, 2008. 74(5): p. 1417-1428. 91. Cooper, D., Compartmentalization of adenylate cyclase and cAMP signalling. Biochemical Society Transactions, 2005. 33(6): p. 1319-1322. 92. Weber, I.T., et al., The cAMP-binding domains of the regulatory subunit of cAMP-dependent protein kinase and the catabolite gene activator protein are homologous. Proceedings of the National Academy of Sciences of the United States of America, 1982. 79(24): p. 7679-7683. 93. Ebrahimi, F.A.W. and A. Chess, Olfactory G proteins: simple and complex signal transduction. Current Biology, 1998. 8(12): p. R431-R433. 94. Avigan, J., et al., Pertussis toxin-catalyzed ADP-ribosylation of G(o) with mutations at the carboxyl terminus. Biochemistry, 1992. 31(33): p. 7736-7740. 95. Itoh, H., et al., Presence of three distinct molecular species of Gi protein subunit. Structure of rat cDNAs and human genomic DNAs. Journal of Biological Chemistry, 1988. 263(14): p. 6656-6664. 96. Bromberg, K., R. Iyengar, and J. He, Regulation of neurite outgrowth by G(i/o) signaling pathways. Frontiers in Bioscience, 2008. 13: p. 4544- 4557. 97. Leck, K.J., et al., Gz proteins are functionally coupled to dopamine D2- like receptors in vivo. Neuropharmacology, 2006. 51(3): p. 597-605. 98. Drin, G. and S. Scarlata, Stimulation of phospholipase C by membrane interactions, interdomain movement, and G protein binding - How many ways can you activate an enzyme? Cellular Signalling, 2007. 19(7): p. 1383-1392. 99. Fukami, K., et al., Phospholipase C is a key enzyme regulating intracellular calcium and modulating the phosphoinositide balance. Progress in Lipid Research, 2010. 49(4): p. 429-437. 100. Strathmann, M.P. and M.I. Simon, G12 and G13 subunits define a fourth class of G protein alpha subunits. Proceedings of the National Academy of Sciences of the United States of America, 1991. 88(13): p. 5582-5586. 101. Riobo, N.A. and D.R. Manning, Receptors coupled to heterotrimeric G proteins of the G12 family. Trends in Pharmacological Sciences, 2005. 26(3): p. 146-154. 102. Xu, N., T. Voyno-Yasenetskaya, and J.S. Gutkind, Potent transforming activity of the G13 subunit defines a novel family of oncogenes. Biochemical and Biophysical Research Communication, 1994. 201: p. 603-609. 103. Lopez, I., et al., A novel bifunctional phospholipase C that is regulated by Gα12 and stimulates the Ras/mitogen-activated protein kinase pathway. Journal of Biological Chemistry, 2001. 276(4): p. 2758-2765. 104. Radhika, V., et al., G13 stimulates cell migration through cortactin- interacting protein Hax-1. Journal of Biological Chemistry, 2004. 279: p. 49406-49413. 105. Radhika, V. and N. Dhanasekaran, Transforming G proteins. Oncogene, 2001. 20: p. 1607-1614. 196
106. Vogt, S., et al., Receptor-dependent RhoA activation in G12/G13-deficient cells. Journal of Biological Chemistry, 2003. 278(31): p. 28743-28749. 107. Moers, A., et al., G13 is an essential mediator of platelet activation in hemostasis and thrombosis. Nature Medicine, 2003. 9(11): p. 1418-1422. 108. Dupre, D., et al., The role of G subunits in the organization, assembly, and function of GPCR signaling complexes. Annual Review of Pharmacology and Toxicology, 2009. 49: p. 31-56. 109. Neer, E.J., Heterotrimeric C proteins: organizers of transmembrane signals. Cell, 1995. 80(2): p. 249-257. 110. Logothetis, D.E., et al., The subunits of GTP-binding proteins activate the muscarinic K+ channel in heart. Nature, 1987. 325(6102): p. 321-326. 111. Ulmsten U, Andersson KE, and W. L, Treatment of premature labor with the calcium antagonist nifedipine. Archives of Gynecology, 1980. 229(1): p. 1-5. 112. Taussig, R., et al., Distinct patterns of bidirectional regulation of mammalian adenylyl cyclases. Journal of Biological Chemistry, 1994. 269(8): p. 6093-6100. 113. Smith GN, Walker MC, and M. MJ, Randomised, double-blind, placebo controlled pilot study assessing nitroglycerin as a tocolytic. British Journal of Obstetrics and Gynaecology 1999. 106(7): p. 736-739. 114. Zhang, S., et al., A C-terminal mutant of the G protein β subunit deficient in the activation of phospholipase C-β. Journal of Biological Chemistry, 1996. 271(33): p. 20208-20212. 115. Zamponi, G.W., et al., Crosstalk between G proteins and protein kinase C mediated by the calcium channel 1 subunit. Nature, 1997. 385(6615): p. 442-446. 116. Lees CC, et al., Glyceryl trinitrate and ritodrine in tocolysis: an international multicenter randomized study. GTN preterm labour investigation group. Obstetrics and Gynecology 1999. 94(3): p. 403-408. 117. The worldwide atosiban versus -agonists study group participants are listed, i., Effectiveness and safety of the oxytocin antagonist atosiban versus -adrenergic agonists in the treatment of preterm labour. BJOG: An International Journal of Obstetrics & Gynaecology, 2001. 108(2): p. 133-142. 118. Sakai, M., et al., Evaluation of the tocolytic effect of a selective cyclooxygenase-2 inhibitor in a mouse model of lipopolysaccharide- induced preterm delivery. Molecular Human Reproduction, 2001. 7(6): p. 595-602. 119. Kurtzman, J.L., et al., Do nifedipine and verapamil potentiate the cardiac toxicity of magnesium sulfate? American Journal of Perinatology, 1993. 10(06): p. 450,452. 120. Watson, J.E., et al., Effects of poverty on home environment: an analysis of three-year outcome data for low birth weight premature infants. Journal of Pediatric Psychology, 1996. 21(3): p. 419-431. 197
121. Yallampalli C, et al., Role and regulation of nitric oxide in the uterus during pregnancy and parturition. Journal of the Society for Gynecological Investigation, 1998. 5(2): p. 58-67. 122. Sawdy R, et al., Use of a cyclo-oxygenase type-2-selective non-steroidal anti-inflammatory agent to prevent preterm delivery. Lancet, 1997. 350(9073): p. 265-266. 123. Crowther CA, Hiller JE, and D. LW, Magnesium sulphate for preventing preterm birth in threatened preterm labour. Cochrane Database of Systematic Reviews (Online), 2002. 4: p. CD001060. 124. King J, et al., Cyclo-oxygenase (COX) inhibitors for treating preterm labour. Cochrane Database of Systematic Reviews (Online), 2005. 2: p. CD001992. 125. Yousif MH and T. O, Tocolytic effect of the cyclooxygenase-2 inhibitor, meloxicam: studies on uterine contractions in the rat. The Journal of Pharmacy and Pharmacology 1998. 50(6): p. 681-685. 126. Gross, G., et al., Inhibition of cyclooxygenase-2 prevents inflammation- mediated preterm labor in the mouse. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology, 2000. 278(6): p. R1415-R1423. 127. Thorp JM Jr, et al., Nifedipine enhances the cardiac toxicity of magnesium sulfate in the isolated perfused Sprague-Dawley rat heart. American Journal of Obstetrics and Gynecology 1990. 163(2): p. 655-656. 128. Chavrier, P., et al., Localization of low molecular weight GTP binding proteins to exocytic and endocytic compartments. Cell, 1990. 62(2): p. 317-329. 129. Garfield RE, Sims S, and D. EE, Gap junctions: their presence and necessity in myometrium during parturition. Science, 1977. 198(4320): p. 958-960. 130. Schmidt, C.J., et al., Specificity of G protein and subunit interactions. Journal of Biological Chemistry, 1992. 267(20): p. 13807-13810. 131. Hendrix, E.M., et al., Steroid hormone regulation of rat myometrial gap junction formation: effects on cx43 levels and trafficking. Biology of Reproduction, 1995. 52(3): p. 547-560. 132. Scheffner, M., et al., The E6 oncoprotein encoded by human papillomavirus types 16 and 18 promotes the degradation of p53. Cell, 1990. 63(6): p. 1129-1136. 133. Geimonen, E., et al., Activation of protein kinase C in human uterine smooth muscle induces connexin-43 gene transcription through an AP-1 site in the promoter sequence. Journal of Biological Chemistry, 1996. 271(39): p. 23667-23674. 134. Dickson, M.A., et al., Human keratinocytes that express hTERT and also bypass a p16INK4a-enforced mechanism that limits life span become immortal yet retain normal growth and differentiation characteristics. Molecular and Cellular Biology, 2000. 20(4): p. 1436-1447. 135. Münger K, et al., Complex formation of human papillomavirus E7 proteins with the retinoblastoma tumor suppressor gene product. The EMBO Journal, 1989. 8(13): p. 4099-4105. 198
136. Grümmer, R., et al., Regulation of connexin26 and connexin43 expression in rat endometrium by ovarian steroid hormones. Biology of Reproduction, 1994. 51(6): p. 1109-1116. 137. Bernal, A.L., The regulation of uterine relaxation. Seminars in Cell & Developmental Biology, 2007. 18(3): p. 340-347. 138. Chevillard, G., et al., Identification of interleukin-1β regulated genes in uterine smooth muscle cells. Reproduction, 2007. 134(6): p. 811-822. 139. Bernal, A.L., Mechanisms of labour—biochemical aspects. BJOG: An International Journal of Obstetrics & Gynaecology, 2003. 110: p. 39-45. 140. Bossmar T, et al., Receptors for and myometrial responses to oxytocin and vasopressin in preterm and term human pregnancy: effects of the oxytocin antagonist atosiban. American Journal of Obstetrics and Gynecology, 1994. 171(6): p. 1634-1642. 141. Kimura, T., et al., Expression of oxytocin receptor in human pregnant myometrium. Endocrinology, 1996. 137(2): p. 780-5. 142. Elands, J., C. Barberis, and S. Jard, [3H]-[Thr4,Gly7]OT: a highly selective ligand for central and peripheral OT receptors. American Journal of Physiology - Endocrinology and Metabolism, 1988. 254(1): p. E31-E38. 143. Perez-Reyes, N., et al., Immortalization of primary human smooth muscle cells. Proceedings of the National Academy of Sciences of the United States of America, 1992. 89(4): p. 1224-1228. 144. Pierzynski, P., et al., Inhibitory effect of barusiban and atosiban on oxytocin-induced contractions of myometrium from preterm and term pregnant women. Journal of the Society for Gynecologic Investigation, 2004. 11(6): p. 384-387. 145. Monga, M., et al., Oxytocin-stimulated responses in a pregnant human immortalized myometrial cell line. Biology of Reproduction, 1996. 55(2): p. 427-432. 146. Wright WE, et al., Telomerase activity in human germline and embryonic tissues and cells. Developmental Genetics, 1996. 18(2): p. 173-179. 147. Fanning, E., Simian virus 40 large T antigen: the puzzle, the pieces, and the emerging picture. Journal of Virology, 1992. 66(3): p. 1289-1293. 148. Farwell DG, et al., Genetic and epigenetic changes in human epithelial cells immortalized by telomerase. The American Journal of Pathology, 2000. 156(5): p. 1537-1547. 149. Nunn, C., et al., RGS2 inhibits β-adrenergic receptor-induced cardiomyocyte hypertrophy. Cellular Signalling, 2010. 22(8): p. 1231- 1239. 150. Evanko, D.S., et al., Loss of association between activated Gq and G disrupts receptor-dependent and receptor-independent signaling. Cellular Signalling, 2005. 17(10): p. 1218-1228. 151. Conti, F., et al., Intracellular trafficking of the human oxytocin receptor: evidence of receptor recycling via a Rab4/Rab5 “short cycle”. American Journal of Physiology - Endocrinology and Metabolism, 2009. 296(3): p. E532-E542. 199
152. Inoue, T., et al., Structural organization of the human oxytocin receptor gene. Journal of Biological Chemistry, 1994. 269(51): p. 32451-6. 153. Whitmarsh AJ and D. RJ, Transcription factor AP-1 regulation by mitogen-activated protein kinase signal transduction pathways. Journal of Molecular Medicine (Berlin, Germany), 1996. 74(10): p. 589-607. 154. Luo, X., et al., Alternate Coupling of Receptors to Gs and Gi in Pancreatic and Submandibular Gland Cells. Journal of Biological Chemistry, 1999. 274(25): p. 17684-17690. 155. Sönnichsen, B., et al., Distinct membrane domains on endosomes in the recycling pathway visualized by multicolor imaging of Rab4, Rab5, and Rab11. The Journal of Cell Biology, 2000. 149(4): p. 901-914. 156. Hebert, T.E., et al., A peptide derived from a β2-adrenergic receptor transmembrane domain inhibits both receptor dimerization and activation. Journal of Biological Chemistry, 1996. 271(27): p. 16384-16392. 157. Okamoto, T., et al., Identification of a Gs activator region of the β2- adrenergic receptor that is autoregulated via protein kinase A-dependent phosphorylation. Cell, 1991. 67(4): p. 723-730. 158. Lefebvre DL, et al., Myometrial transcriptional regulation of the gap junction gene, connexin-43. Reproduction, Fertility, and Development, 1995. 7(3): p. 603-611. 159. Piersanti, M. and S.J. Lye, Increase in messenger ribonucleic acid encoding the myometrial gap junction protein, connexin-43, requires protein synthesis and is associated with increased expression of the activator protein-1, c-fos. Endocrinology, 1995. 136(8): p. 3571-8. 160. Chanrachakul, B., F.B. Pipkin, and R.N. Khan, Contribution of coupling between human myometrial β2-adrenoreceptor and the BKC channel to uterine quiescence. American Journal of Physiology - Cell Physiology, 2004. 287(6): p. C1747-C1752. 161. Shenoy, S.K. and R.J. Lefkowitz, Multifaceted roles of -arrestins in the regulation of seven-membrane-spanning receptor trafficking and signalling. Biochemical Journal, 2003. 375: p. 503-515. 162. Luttrell, L.M. and D. Gesty-Palmer, Beyond desensitization: physiological relevance of arrestin-dependent signaling. Pharmacological Reviews, 2010. 62(2): p. 305-330. 163. Ferguson, S.S.G., Evolving concepts in G-protein coupled receptor endocytosis: the role in receptor desensitization and signaling. Pharmacological Reviews, 2001. 53: p. 1-24. 164. Freedman, N.J. and R.J. Lefkowitz, Desensitization of G protein-coupled receptors. Recent Progress in Hormone Research, 1996. 51: p. 319-351. 165. Payne DM, et al., Identification of the regulatory phosphorylation sites in pp42/mitogen-activated protein kinase (MAP kinase). The EMBO Journal 1991. 10(4): p. 885-892. 166. Garfield, R.E., et al., Comparing uterine electromyography activity of antepartum patients versus term labor patients. American Journal of Obstetrics and Gynecology, 2005. 193(1): p. 23-29. 200
167. Smith, R., Parturition. New England Journal of Medicine, 2007. 356(3): p. 271-283. 168. Silverthorn, D., ed. Human physiology: an integrated approach. 2nd Ed ed. 2001, Prentice-Hall: New Jersey. 169. Robbins, D.J., et al., Regulation and properties of extracellular signal- regulated protein kinases 1 and 2 in vitro. Journal of Biological Chemistry, 1993. 268(7): p. 5097-5106. 170. Laporte, S.A., et al., The β2-adrenergic receptor/βarrestin complex recruits the clathrin adaptor AP-2 during endocytosis. Proceedings of the National Academy of Sciences of the United States of America, 1999. 96(7): p. 3712-3717. 171. Laporte, S.A., et al., The interaction of β-arrestin with the AP-2 adaptor is required for the clustering of β2-adrenergic receptor into clathrin-coated pits. Journal of Biological Chemistry, 2000. 275(30): p. 23120-23126. 172. Laporte, S.A., et al., β-arrestin/AP-2 interaction in G protein-coupled receptor internalization. Journal of Biological Chemistry, 2002. 277(11): p. 9247-9254. 173. Zhang, J., et al., Dynamin and β-arrestin reveal distinct mechanisms for G protein-coupled receptor internalization. Journal of Biological Chemistry, 1996. 271(31): p. 18302-18305. 174. Oakley, R.H., et al., Differential affinities of visual arrestin, βarrestin1, and βarrestin2 for G protein-coupled receptors delineate two major classes of receptors. Journal of Biological Chemistry, 2000. 275(22): p. 17201-17210. 175. Zhang, J., et al., A central role for β-arrestins and clathrin-coated vesicle- mediated endocytosis in β2-adrenergic receptor resensitization. Journal of Biological Chemistry, 1997. 272(43): p. 27005-27014. 176. Anborgh, P.H., et al., Receptor/β-arrestin complex formation and the differential trafficking and resensitization of β2-adrenergic and angiotensin II type 1A receptors. Molecular Endocrinology, 2000. 14(12): p. 2040-2053. 177. Seachrist, J.L., P.H. Anborgh, and S.S.G. Ferguson, β2-adrenergic receptor internalization, endosomal sorting, and plasma membrane recycling are regulated by Rab GTPases. Journal of Biological Chemistry, 2000. 275(35): p. 27221-27228. 178. Oakley, R.H., et al., Association of β-arrestin with G protein-coupled receptors during clathrin-mediated endocytosis dictates the profile of receptor resensitization. Journal of Biological Chemistry, 1999. 274(45): p. 32248-32257. 179. Pierce, K.L. and R.J. Lefkowitz, Classical and new roles of -arrestins in the regulation of G-protein-coupled receptors. Nature Reviews Neuroscience, 2001. 2(10): p. 727-733. 180. Innamorati, G., et al., The long and the short cycle. Journal of Biological Chemistry, 2001. 276(16): p. 13096-13103. 181. Bouley, R., et al., Downregulation of the vasopressin type 2 receptor after vasopressin-induced internalization: involvement of a lysosomal 201
degradation pathway. American Journal of Physiology - Cell Physiology, 2005. 288(6): p. C1390-C1401. 182. Oakley, R.H., et al., Molecular determinants underlying the formation of stable intracellular G protein-coupled receptor-β-arrestin complexes after receptor endocytosis. Journal of Biological Chemistry, 2001. 276(22): p. 19452-19460. 183. Moore, R.H., et al., Agonist-induced sorting of human 2-adrenergic receptors to lysosomes during downregulation. Journal of Cell Science, 1999. 112(3): p. 329-338. 184. Moore, R.H., et al., Specific changes in 2-adrenoceptor trafficking kinetics and intracellular sorting during downregulation. European Journal of Pharmacology, 1999. 369(1): p. 113-123. 185. Moore, R.H., et al., Rab11 regulates the recycling and lysosome targeting of β2-adrenergic receptors. Journal of Cell Science, 2004. 117(15): p. 3107-3117. 186. Phaneuf, S., et al., The desensitization of oxytocin receptors in human myometrial cells is accompanied by down-regulation of oxytocin receptor messenger RNA. Journal of Endocrinology, 1997. 154(1): p. 7-18. 187. Phaneuf, S., et al., Desensitization of oxytocin receptors in human myometrium. Human Reproduction Update, 1998. 4(5): p. 625-633. 188. Stenmark, H., Rab GTPases as coordinators of vesicle traffic. Nature Reviews Molecular Cell Biology, 2009. 10(8): p. 513-525. 189. Eswaran, H., et al., Prediction of labor in term and preterm pregnancies using non-invasive magnetomyographic recordings of uterine contractions. American Journal of Obstetrics and Gynecology, 2004. 190(6): p. 1598-1602. 190. Lowery CL, et al., First magnetomyographic recordings of uterine activity with spatial-temporal information using 151 channel sensor array (SARA). The Journal of the Arkansas Medical Society 2003. 100(3): p. 90-91. 191. Barlic, J., et al., Regulation of tyrosine kinase activation and granule release through -arrestin by CXCR1. Nature Immunology, 2000. 1(3): p. 227-233. 192. DeFea, K.A., et al., The proliferative and antiapoptotic effects of substance P are facilitated by formation of a β-arrestin-dependent scaffolding complex. Proceedings of the National Academy of Sciences of the United States of America, 2000. 97(20): p. 11086-11091. 193. DeFea, K.A., et al., β-arrestin–dependent endocytosis of proteinase- activated receptor 2 is required for intracellular targeting of activated ERK1/2. The Journal of Cell Biology, 2000. 148(6): p. 1267-1282. 194. McDonald, P.H., et al., β-arrestin 2: a receptor-regulated MAPK scaffold for the activation of JNK3. Science, 2000. 290(5496): p. 1574-1577. 195. Luttrell, L.M., et al., Activation and targeting of extracellular signal- regulated kinases by -arrestin scaffolds. Proceedings of the National Academy of Sciences of the United States of America, 2001. 98: p. 2449- 2454. 202
196. Shenoy, S.K., et al., Regulation of receptor fate by ubiquitination of activated β2-adrenergic receptor and β-arrestin. Science, 2001. 294(5545): p. 1307-1313. 197. Perry, S.J., et al., Targeting of cyclic AMP degradation to β2-adrenergic receptors by β-arrestins. Science, 2002. 298(5594): p. 834-836. 198. Nelson, C.D., et al., Targeting of diacylglycerol degradation to M1 muscarinic receptors by -arrestins. Science, 2007. 315(5812): p. 663- 666. 199. Witherow, D.S., et al., β-arrestin inhibits NF-κB activity by means of its interaction with the NF-κB inhibitor IκBα. Proceedings of the National Academy of Sciences of the United States of America, 2004. 101(23): p. 8603-8607. 200. Bhattacharya, M., et al., -arrestins regulate a Ral-GDS-Ral effector pathway that mediates cytoskeletal reorganization. Nature Cell Biology, 2002. 4(8): p. 547-555. 201. Beaulieu, J.-M., et al., An Akt/-arrestin 2/PP2A signaling complex mediates dopaminergic neurotransmission and behavior. Cell, 2005. 122(2): p. 261-273. 202. Xiao, K., et al., Functional specialization of β-arrestin interactions revealed by proteomic analysis. Proceedings of the National Academy of Sciences of the United States of America, 2007. 104(29): p. 12011-12016. 203. Ahn, S., et al., Differential kinetic and spatial patterns of β-arrestin and G protein-mediated ERK activation by the angiotensin II receptor. Journal of Biological Chemistry, 2004. 279(34): p. 35518-35525. 204. Gesty-Palmer, D., et al., β-arrestin 2 expression determines the transcriptional response to lysophosphatidic acid stimulation in murine embryo fibroblasts. Journal of Biological Chemistry, 2005. 280(37): p. 32157-32167. 205. Gesty-Palmer, D., et al., Distinct β-arrestin- and G protein-dependent pathways for parathyroid hormone receptor-stimulated ERK1/2 activation. Journal of Biological Chemistry, 2006. 281(16): p. 10856- 10864. 206. Shenoy, S.K., et al., β-arrestin-dependent, G protein-independent ERK1/2 activation by the β2 adrenergic receptor. Journal of Biological Chemistry, 2006. 281(2): p. 1261-1273. 207. Tohgo, A., et al., β-arrestin scaffolding of the ERK cascade enhances cytosolic ERK activity but inhibits ERK-mediated transcription following angiotensin AT1 receptor stimulation. Journal of Biological Chemistry, 2002. 277(11): p. 9429-9436. 208. Tohgo, A., et al., The stability of the G protein-coupled receptor-β- arrestin interaction determines the mechanism and functional consequence of ERK activation. Journal of Biological Chemistry, 2003. 278(8): p. 6258-6267. 209. Hara, M.R., et al., A stress response pathway regulates DNA damage through β2-adrenoreceptors and -arrestin-1. Nature, 2011. 477(7364): p.349-353. 203
210. Azzi, M., et al., β-arrestin-mediated activation of MAPK by inverse agonists reveals distinct active conformations for G protein-coupled receptors. Proceedings of the National Academy of Sciences of the United States of America, 2003. 100(20): p. 11406-11411. 211. Wei, H., et al., Independent β-arrestin 2 and G protein-mediated pathways for angiotensin II activation of extracellular signal-regulated kinases 1 and 2. Proceedings of the National Academy of Sciences of the United States of America, 2003. 100(19): p. 10782-10787. 212. Lee, M.-H., et al., Role of β-arrestin-mediated desensitization and signaling in the control of angiotensin AT1 receptor-stimulated transcription. Journal of Biological Chemistry, 2008. 283(4): p. 2088- 2097. 213. Kim, J., et al., Functional antagonism of different G protein-coupled receptor kinases for β-arrestin-mediated angiotensin II receptor signaling. Proceedings of the National Academy of Sciences of the United States of America, 2005. 102(5): p. 1442-1447. 214. Ren, X.-R., et al., Different G protein-coupled receptor kinases govern G protein and β-arrestin-mediated signaling of V2 vasopressin receptor. Proceedings of the National Academy of Sciences of the United States of America, 2005. 102(5): p. 1448-1453. 215. Jafri, F., et al., Constitutive ERK1/2 activation by a chimeric neurokinin 1 receptor-β-arrestin1 fusion protein. Journal of Biological Chemistry, 2006. 281(28): p. 19346-19357. 216. Conner, D.A., et al., -arrestin1 knockout mice appear normal but demonstrate altered cardiac responses to -adrenergic stimulation. Circulation Research, 1997. 81(6): p. 1021-1026. 217. Bohn, L.M., et al., Enhanced morphine analgesia in mice lacking β- arrestin 2. Science, 1999. 286(5449): p. 2495-2498. 218. Bohn, L.M., R.J. Lefkowitz, and M.G. Caron, Differential mechanisms of morphine antinociceptive tolerance revealed in βarrestin-2 knock-out mice. Journal of Neuroscience Methods, 2002. 22(23): p. 10494-10500. 219. Zhai, P., et al., Cardiac-specific overexpression of AT1 receptor mutant lacking Gαq/Gαi coupling causes hypertrophy and bradycardia in transgenic mice. The Journal of Clinical Investigation, 2005. 115(11): p. 3045-3056. 220. Noma, T., et al., β-arrestin–mediated β1-adrenergic receptor transactivation of the EGFR confers cardioprotection. The Journal of Clinical Investigation, 2007. 117(9): p. 2445-2458. 221. Wisler, J.W., et al., A unique mechanism of β-blocker action: carvedilol stimulates β-arrestin signaling. Proceedings of the National Academy of Sciences of the United States of America, 2007. 104(42): p. 16657-16662. 222. Drake, M.T., et al., β-arrestin-biased agonism at the β2-adrenergic receptor. Journal of Biological Chemistry, 2008. 283(9): p. 5669-5676. 223. Deupi, X. and B.K. Kobilka, Energy landscapes as a tool to integrate GPCR structure, dynamics, and function. Physiology, 2010. 25(5): p. 293- 303. 204
224. Bhattacharya, S., S.E. Hall, and N. Vaidehi, Agonist-induced conformational changes in bovine rhodopsin: insight into activation of G- protein-coupled receptors. Journal of Molecular Biology, 2008. 382(2): p. 539-555. 225. Bhattacharya, S., et al., Ligand-stabilized conformational states of human 2 adrenergic receptor: insight into G-protein-coupled receptor activation. Biophysical Journal, 2008. 94(6): p. 2027-2042. 226. Chapman, N.R., et al., Expression of the GTP-binding protein (Gαs) is repressed by the nuclear factor κB RelA subunit in human myometrium. Endocrinology, 2005. 146(11): p. 4994-5002. 227. MacDougall, M.W.J., G.N. Europe-Finner, and S.C. Robson, Human myometrial quiescence and activation during gestation and parturition involve dramatic changes in expression and activity of particulate type II (RIIα) protein kinase A holoenzyme. Journal of Clinical Endocrinology & Metabolism, 2003. 88(5): p. 2194-2205. 228. Lye, S.J., et al., The molecular basis of labour and tocolysis. Fetal and Maternal Medicine Review, 1998. 10(03): p. 121-136. 229. Chanrachakul, B., et al., Immunolocalization and protein expression of the subunit of the large-conductance calcium-activated potassium channel in human myometrium. Reproduction, 2003. 126(1): p. 43-48. 230. Brainard, A.M., et al., Maxi-K channels localize to caveolae in human myometrium: a role for an actin-channel-caveolin complex in the regulation of myometrial smooth muscle K+ current. American Journal of Physiology - Cell Physiology, 2005. 289(1): p. C49-C57. 231. Pei, G., et al., A constitutively active mutant 2-adrenergic receptor is constitutively desensitized and phosphorylated. Proceedings of the National Academy of Sciences of the United States of America, 1994. 91(7): p. 2699-2702. 232. Chini, B. and F. Fanelli, Molecular basis of ligand binding and receptor activation in the oxytocin and vasopressin receptor family. Experimental Physiology, 2000. 85(suppl 1): p. 59S-66S. 233. Vaidehi, N. and T. Kenakin, The role of conformational ensembles of seven transmembrane receptors in functional selectivity. Current Opinion in Pharmacology, 2010. 10(6): p. 775-781. 234. Boehr, D.D., R. Nussinov, and P.E. Wright, The role of dynamic conformational ensembles in biomolecular recognition. Nature Chemical Biology, 2009. 5(11): p. 789-796. 235. Yao, X., et al., Coupling ligand structure to specific conformational switches in the 2-adrenoceptor. Nature Chemical Biology, 2006. 2(8): p. 417-422. 236. Bokoch, M.P., et al., Ligand-specific regulation of the extracellular surface of a G-protein-coupled receptor. Nature, 2010. 463(7277): p. 108- 112. 237. Lohse, M.J., et al., Optical techniques to analyze real-time activation and signaling of G-protein-coupled receptors. Trends in Pharmacological Sciences, 2008. 29(3): p. 159-165. 205
238. Galandrin, S. and M. Bouvier, Distinct signaling profiles of β1 and β2 adrenergic receptor ligands toward adenylyl cyclase and mitogen- activated protein kinase reveals the pluridimensionality of efficacy. Molecular Pharmacology, 2006. 70(5): p. 1575-1584. 239. Kim, I.-M., et al., β-blockers alprenolol and carvedilol stimulate β- arrestin-mediated EGFR transactivation. Proceedings of the National Academy of Sciences of the United States of America, 2008. 105(38): p. 14555-14560. 240. Reversi, A., et al., The oxytocin receptor antagonist atosiban inhibits cell growth via a “biased agonist” mechanism. Journal of Biological Chemistry, 2005. 280(16): p. 16311-16318. 241. Challis JRG and L. SJ, Parturition., in The physiology of reproduction. , Knobil E and N. JD, Editors. 1994, Raven Press: New York. p. 985-1031. 242. Azzi, M., et al., β-arrestin-mediated activation of MAPK by inverse agonists reveals distinct active conformations for G protein-coupled receptors. Proceedings of the National Academy of Sciences of the United States of America, 2003. 100(20): p. 11406-11411. 243. Good, M.C., J.G. Zalatan, and W.A. Lim, Scaffold proteins: hubs for controlling the flow of cellular information. Science, 2011. 332(6030): p. 680-686. 244. Milligan, G., G protein-coupled receptor hetero-dimerization: contribution to pharmacology and function. British Journal of Pharmacology, 2009. 158(1): p. 5-14. 245. Milligan, G., The role of dimerisation in the cellular trafficking of G- protein-coupled receptors. Current Opinion in Pharmacology, 2010. 10(1): p. 23-29. 246. Birdsall, N.J.M., Class A GPCR heterodimers: evidence from binding studies. Trends in Pharmacological Sciences, 2010. 31(11): p. 499-508. 247. Whorton, M.R., et al., Efficient coupling of transducin to monomeric rhodopsin in a phospholipid bilayer. Journal of Biological Chemistry, 2008. 283(7): p. 4387-4394. 248. AbdAlla, S., et al., Increased AT1 receptor heterodimers in preeclampsia mediate enhanced angiotensin II responsiveness. Nature Medicine, 2001. 7(9): p. 1003-1009. 249. Vischer, H.F., et al., Viral hijacking of human receptors through heterodimerization. Biochemical and Biophysical Research Communications, 2008. 377(1): p. 93-97. 250. Rediger, A., et al., Heterodimerization of hypothalamic G-protein-coupled receptors involved in weight regulation. Obesity Facts, 2009. 2(2): p. 80- 86. 251. Gonzalez-Maeso, J., et al., Identification of a serotonin/glutamate receptor complex implicated in psychosis. Nature, 2008. 452(7183): p. 93-97. 252. Carrillo, J.J., J.F. López-Giménez, and G. Milligan, Multiple interactions between transmembrane helices generate the oligomeric α1b- adrenoceptor. Molecular Pharmacology, 2004. 66(5): p. 1123-1137. 253. Lopez-Gimenez, J.F., et al., The α1b-adrenoceptor exists as a higher- order oligomer: effective oligomerization is required for receptor 206
maturation, surface delivery, and function. Molecular Pharmacology, 2007. 71(4): p. 1015-1029. 254. Guo, W., et al., Dopamine D2 receptors form higher order oligomers at physiological expression levels. The EMBO Journal, 2008. 27(17): p. 2293-2304. 255. Salahpour, A., et al., Homodimerization of the β2-adrenergic receptor as a prerequisite for cell surface targeting. Journal of Biological Chemistry, 2004. 279(32): p. 33390-33397. 256. Gomes, I., et al., Heterodimerization of μ and δ opioid receptors: a role in opiate synergy. The Journal of Neuroscience, 2000. 20(22): p. RC110. 257. Rocheville, M., et al., Receptors for dopamine and somatostatin: formation of hetero-oligomers with enhanced functional activity. Science, 2000. 288(5463): p. 154-157. 258. AbdAlla, S., H. Lother, and U. Quitterer, AT1-receptor heterodimers show enhanced G-protein activation and altered receptor sequestration. Nature, 2000. 407(6800): p. 94-98. 259. Pfeiffer, M., et al., Homo- and heterodimerization of somatostatin receptor subtypes. Journal of Biological Chemistry, 2001. 276(17): p. 14027-14036. 260. George, S.R., et al., Oligomerization of μ- and δ-opioid receptors. Journal of Biological Chemistry, 2000. 275(34): p. 26128-26135. 261. Jordan, B.A., et al., Functional interactions between μ opioid and α2A- adrenergic receptors. Molecular Pharmacology, 2003. 64(6): p. 1317- 1324. 262. Zhu, W.-Z., et al., Heterodimerization of β1- and β2-adrenergic receptor subtypes optimizes β-adrenergic modulation of cardiac contractility. Circulation Research, 2005. 97(3): p. 244-251. 263. Sohy, D., et al., Hetero-oligomerization of CCR2, CCR5, and CXCR4 and the protean effects of “selective” antagonists. Journal of Biological Chemistry, 2009. 284(45): p. 31270-31279. 264. Vilardaga, J.-P., et al., Conformational cross-talk between 2A-adrenergic and -opioid receptors controls cell signaling. Nature Chemical Biology, 2008. 4(2): p. 126-131. 265. Breit, A., et al., Simultaneous activation of the δ opioid receptor (δOR)/sensory neuron-specific receptor-4 (SNSR-4) hetero-oligomer by the mixed bivalent agonist bovine adrenal medulla peptide 22 activates SNSR-4 but inhibits δOR signaling. Molecular Pharmacology, 2006. 70(2): p. 686-696. 266. Fan, T., et al., A role for the distal carboxyl tails in generating the novel pharmacology and G protein activation profile of μ and δ opioid receptor hetero-oligomers. Journal of Biological Chemistry, 2005. 280(46): p. 38478-38488. 267. Lavoie, C., et al., β1/β2-adrenergic receptor heterodimerization regulates β2-adrenergic receptor internalization and ERK signaling efficacy. Journal of Biological Chemistry, 2002. 277(38): p. 35402-35410. 207
268. Breit, A., M. Lagacé, and M. Bouvier, Hetero-oligomerization between β2- and β3-adrenergic receptors generates a β-adrenergic signaling unit with distinct functional properties. Journal of Biological Chemistry, 2004. 279(27): p. 28756-28765. 269. Xu, J., et al., Heterodimerization of α2A- and β1-adrenergic receptors. Journal of Biological Chemistry, 2003. 278(12): p. 10770-10777. 270. Jordan, B.A., et al., Oligomerization of opioid receptors with β2- adrenergic receptors: A role in trafficking and mitogen-activated protein kinase activation. Proceedings of the National Academy of Sciences of the United States of America, 2001. 98(1): p. 343-348. 271. Barki-Harrington, L., L.M. Luttrell, and H.A. Rockman, Dual inhibition of β-adrenergic and angiotensin II receptors by a single antagonist. Circulation, 2003. 108(13): p. 1611-1618. 272. Uberti, M.A., et al., Heterodimerization with β2-adrenergic receptors promotes surface expression and functional activity of α1D-adrenergic receptors. Journal of Pharmacology and Experimental Therapeutics, 2005. 313(1): p. 16-23. 273. Rozenfeld, R. and L.A. Devi, Receptor heterodimerization leads to a switch in signaling: -arrestin2-mediated ERK activation by µ-δ opioid receptor heterodimers. The FASEB Journal, 2007. 21(10): p. 2455-2465. 274. Waldhoer, M., et al., A heterodimer-selective agonist shows in vivo relevance of G protein-coupled receptor dimers. Proceedings of the National Academy of Sciences of the United States of America, 2005. 102(25): p. 9050-9055. 275. Valant, C., P. Sexton, and A. Christopoulos, Orthosteric/allosteric bitopic ligands: going hybrid at GPCRs. Molecular Interventions, 2009. 9(3): p. 125-135. 276. Kasai, R.S., et al., Full characterization of GPCR monomer–dimer dynamic equilibrium by single molecule imaging. The Journal of Cell Biology, 2011. 192(3): p. 463-480. 277. Hern, J.A., et al., Formation and dissociation of M1 muscarinic receptor dimers seen by total internal reflection fluorescence imaging of single molecules. Proceedings of the National Academy of Sciences of the United States of America, 2010. 107(6): p. 2693-2698. 278. Gurevich, V.V. and E.V. Gurevich, GPCR monomers and oligomers: it takes all kinds. Trends in Neurosciences, 2008. 31(2): p. 74-81. 279. Chun, H.J., et al., Apelin signaling antagonizes Ang II effects in mouse models of atherosclerosis. The Journal of Clinical Investigation, 2008. 118(10): p. 3343-3354. 280. Lourdes González-Hernández, M.d., et al., Angiotensin-II type 1 receptor (AT1R) and -1D adrenoceptor form a heterodimer during pregnancy- induced hypertension. Autonomic and Autacoid Pharmacology, 2010. 30(3): p. 167-172. 281. Rozenfeld, R., et al., AT1R-CB1R heteromerization reveals a new mechanism for the pathogenic properties of angiotensin II. The EMBO Journal, 2011. 30(12): p. 2350-2363. 208
282. Nijmeijer, S., et al., The Epstein-Barr Virus-encoded G protein-coupled receptor BILF1 hetero-oligomerizes with human CXCR4, scavenges Gαi proteins, and constitutively impairs CXCR4 functioning. Journal of Biological Chemistry, 2010. 285(38): p. 29632-29641. 283. Rozenfeld, R. and L.A. Devi, Receptor heteromerization and drug discovery. Trends in Pharmacological Sciences, 2010. 31(3): p. 124-130. 284. Christopoulos, A. and T. Kenakin, G protein-coupled receptor allosterism and complexing. Pharmacological Reviews, 2002. 54(2): p. 323-374. 285. Brady, A.E. and L.E. Limbird, G protein-coupled receptor interacting proteins: emerging roles in localization and signal transduction. Cellular Signalling, 2002. 14(4): p. 297-309. 286. Bermeo, M.E., et al., Magnesium sulfate induces translocation of protein kinase C isoenzymes and in myometrial cells from pregnant women. American Journal of Obstetrics and Gynecology, 2004. 190(2): p. 522- 527. 287. Smith, N.J. and G. Milligan, Allostery at G protein-coupled receptor homo- and heteromers: uncharted pharmacological landscapes. Pharmacological Reviews, 2010. 62(4): p. 701-725. 288. Shlykov, S.G., et al., Capacitative cation entry in human myometrial cells and augmentation by hTrpC3 overexpression. Biology of Reproduction, 2003. 69(2): p. 647-655. 289. Keov, P., P.M. Sexton, and A. Christopoulos, Allosteric modulation of G protein-coupled receptors: a pharmacological perspective. Neuropharmacology, 2011. 60(1): p. 24-35. 290. Yang, M., et al., Multiple Trp isoforms implicated in capacitative calcium entry are expressed in human pregnant myometrium and myometrial cells. Biology of Reproduction, 2002. 67(3): p. 988-994. 291. Ehlert, F.J., Estimation of the affinities of allosteric ligands using radioligand binding and pharmacological null methods. Molecular Pharmacology, 1988. 33: p. 187-194. 292. Sanborn, B.M., Hormones and calcium: mechanisms controlling uterine smooth muscle contractile activity. Experimental Physiology, 2001. 86(02): p. 223-237. 293. Marcellino, D., et al., Identification of dopamine D1–D3 receptor heteromers. Journal of Biological Chemistry, 2008. 283(38): p. 26016- 26025. 294. Milligan, G. and N.J. Smith, Allosteric modulation of heterodimeric G- protein-coupled receptors. Trends in Pharmacological Sciences, 2007. 28(12): p. 615-620. 295. Rang, H., et al., eds. Pharmacology. 5th ed. 2003, Churchill Livingstone: Edinburgh. 296. Crespo, P., et al., Dual effect of -adrenergic receptors on mitogen- activated protein kinase. Evidence for a -dependent activation and a G s-cAMP-mediated inhibition. Journal of Biological Chemistry, 1995. 270(42): p. 25259-25265. 209
297. Schmitt, J. and P. Stork, β2-adrenergic receptor activates extracellular signal-regulated kinases (ERKs) via the small G protein Rap1 and the serine/threonine kinase B-Raf. Journal of Biological Chemistry, 2000. 275(August 18): p. 25342-25350. 298. Zou, Y., et al., Both Gs and Gi proteins are critically involved in isoproterenol-induced cardiomyocyte hypertrophy. Journal of Biological Chemistry, 1999. 274(14): p. 9760-9770. 299. Dhanasekaran, N., L. Heasley, and G. Johnson, G protein-coupled receptor systems involved in cell growth and oncogenesis. Endocrine Reviews, 1995. 16(3): p. 259-270. 300. Bueno, O.F. and J.D. Molkentin, Involvement of extracellular signal- regulated kinases 1/2 in cardiac hypertrophy and cell death. Circulation Research, 2002. 91(9): p. 776-781. 301. Ostrowski, N.L., Oxytocin receptor mRNA expression in rat brain: implications for behavioral integration and reproductive success. Psychoneuroendocrinology, 1998. 23(8): p. 989-1004. 302. Viero, C., et al., Review: oxytocin: crossing the bridge between basic science and pharmacotherapy. CNS Neuroscience & Therapeutics, 2010. 16(5): p. e138-e156. 303. Gimpl, G. and F. Fahrenholz, The oxytocin receptor system: structure, function, and regulation. Physiological Reviews, 2001. 81(2): p. 629-83. 304. Zingg, H.H., Oxytocin, in Endocrinology, 4th edition, L.J. De Groot and J.L. Jameson, Editors. 2000, W B Saunders Comp.: Philadelphia. p. 201- 208. 305. Zingg, H.H. and S.A. Laporte, The oxytocin receptor. Trends in Endocrinology and Metabolism, 2003. 14(5): p. 222-7. 306. Lim, M.M. and L.J. Young, Neuropeptidergic regulation of affiliative behavior and social bonding in animals. Hormones and Behaviour, 2006. 50(4): p. 506-17. 307. Cassoni, P., et al., Oxytocin and oxytocin receptors in cancer cells and proliferation. Journal of Neuroendocrinology, 2004. 16(4): p. 362-4. 308. Zhong, M., M. Yang, and B.M. Sanborn, Extracellular signal-regulated kinase 1/2 activation by myometrial oxytocin receptor involves G(q)G and epidermal growth factor receptor tyrosine kinase activation. Endocrinology, 2003. 144(7): p. 2947-56. 309. Ohmichi, M., et al., Oxytocin stimulates mitogen-activated protein kinase activity in cultured human puerperal uterine myometrial cells. Endocrinology, 1995. 136(5): p. 2082-7. 310. Strakova, Z. and M.S. Soloff, Coupling of oxytocin receptor to G proteins in rat myometrium during labor: Gi receptor interaction. American Journal of Physiology - Endocrinology and Metabolism, 1997. 272(5): p. E870-E876. 311. Ohmichi, M., et al., Oxytocin stimulates mitogen-activated protein kinase activity in cultured human puerperal uterine myometrial cells. Endocrinology, 1995. 136(5): p. 2082-2087. 312. Brighton, P.J., et al., Arrestins differentially regulate histamine- and oxytocin-evoked phospholipase C and mitogen-activated protein kinase 210
signalling in myometrial cells. British Journal of Pharmacology, 2011. 162(7): p. 1603-1617. 313. Grotegut, C.A., et al., β-arrestin mediates oxytocin receptor signaling, which regulates uterine contractility and cellular migration. American Journal of Physiology - Endocrinology and Metabolism, 2011. 300(3): p. E468-E477. 314. Cargnello, M. and P.P. Roux, Activation and function of the MAPKs and their substrates, the MAPK-activated protein kinases. Microbiology and Molecular Biology Reviews, 2011. 75(1): p. 50-83. 315. Raman, M., W. Chen, and M.H. Cobb, Differential regulation and properties of MAPKs. Oncogene, 2007. 26(22): p. 3100-3112. 316. Shaul, Y.D. and R. Seger, The MEK/ERK cascade: from signaling specificity to diverse functions. Biochimica et Biophysica Acta (BBA) - Molecular Cell Research, 2007. 1773(8): p. 1213-1226. 317. de Rooij, J., et al., Epac is a Rap1 guanine-nucleotide-exchange factor directly activated by cyclic AMP. Nature, 1998. 396: p. 474-477. 318. Quilliam, L.A., J.F. Rebhun, and A.F. Castro, A growing family of guanine nucleotide exchange factors is responsible for activation of Ras-family GTPases. Progress in Nucleic Acid Research and Molecular Biology, 2002. 71: p. 391-444. 319. Bos, J.L., Epac proteins: multi-purpose cAMP targets. Trends in Biochemichal Sciences, 2006. 31: p. 680-686. 320. Laroche-Joubert, N., et al., Protein kinase A-independent activation of ERK and H,K-ATPase by cAMP in native kidney cells: role of Epac I. Journal of Biological Chemistry, 2002. 277: p. 18598-18604. 321. Wan, Y. and X.Y. Huang, Analysis of the Gs//mitogen-activated protein kinase pathway in mutant S49 cells. Journal of Biological Chemistry, 1998. 273: p. 14533-14537. 322. Tanaka, S., et al., C3G, a guanine nucleotide-releasing protein expressed ubiquitously, binds to the Src homology 3 domains of CRK and GRB2//ASH proteins. Proceedings of the National Academy of Sciences of the United States of America, 1994. 91: p. 3443-3447. 323. Gotoh, T., et al., Identification of Rap1 as a target for the Crk SH3 domain-binding guanine nucleotide-releasing factor C3G. Molecular Cell Biology, 1995. 15: p. 6746-6753. 324. Schmitt, J.M. and P.J. Stork, G and G require distinct Src-dependent pathways to activate Rap1 and Ras. Journal of Biological Chemistry, 2002. 277: p. 43024-43032. 325. Wang, Z., et al., Rap1-mediated activation of extracellular signal- regulated kinases by cyclic AMP is dependent on the mode of Rap1 activation. Molecular Cell Biology, 2006. 26: p. 2130-2145. 326. Ambrosini, A., et al., cAMP cascade leads to Ras activation in cortical neurons. Molecular Brain Research, 2000. 75: p. 54-60. 327. Busca, R., et al., Ras mediates the cAMP-dependent activation of extracellular signal-regulated kinases (ERKs) in melanocytes. The EMBO Journal, 2000. 19: p. 2900-2910. 211
328. Norum, J.H., et al., Endogenous expression and protein kinase A- dependent phosphorylation of the guanine nucleotide exchange factor Ras-GRF1 in human embryonic kidney 293 cells. FEBS Journal, 2005. 272: p. 2304-2316. 329. Amsen, E.M., et al., The guanine nucleotide exchange factor CNrasGEF regulates melanogenesis and cell survival in melanoma cells. Journal of Biological Chemistry, 2006. 281: p. 121-128. 330. Norum, J.H., et al., Epac- and Rap-independent ERK1//2 phosphorylation induced by Gs-coupled receptor stimulation in HEK293 cells. FEBS Letters, 2007. 581: p. 15-20. 331. Wu, J., et al., Inhibition of the EGF-activated MAP kinase signaling pathway by adenosine 3', 5'-monophosphate. Science, 1993. 262: p. 1065- 1069. 332. Dhillon, A.S., et al., Cyclic AMP-dependent kinase regulates Raf-1 kinase mainly by phosphorylation of serine 259. Molecular Cell Biology, 2002. 22: p. 3237-3246. 333. Schmitt, J.M. and P.J. Stork, Cyclic AMP-mediated inhibition of cell growth requires the small G protein Rap1. Molecular Cell Biology, 2001. 21: p. 3671-3683. 334. Schmitt, J.M. and P.J. Stork, PKA phosphorylation of Src mediates cAMP's inhibition of cell growth via Rap1. Molecular Cell, 2002. 9: p. 85- 94. 335. Mochizuki, N., et al., Activation of the ERK//MAPK pathway by an isoform of rap1GAP associated with Gi. Nature, 1999. 400: p. 891-894. 336. Lev, S., et al., Protein tyrosine kinase PYK2 involved in Ca2+-induced regulation of ion channel and MAP kinase functions. Nature, 1995. 376: p. 737-745. 337. Dikic, I., et al., A role for Pyk2 and Src in linking G-protein-coupled receptors with MAP kinase activation. Nature, 1996. 383: p. 547-550. 338. Luttrell, L.M., et al., Role of c-Src tyrosine kinase in G protein-coupled receptor- and G subunit-mediated activation of mitogen-activated protein kinases. Journal of Biological Chemistry, 1996. 271: p. 19443- 19450. 339. Della Rocca, G.J., et al., Ras-dependent mitogen-activated protein kinase activation by G protein-coupled receptors. Convergence of Gi- and Gq- mediated pathways on calcium//calmodulin, Pyk2, and Src kinase. Journal of Biological Chemistry, 1997. 272: p. 19125-19132. 340. Kranenburg, O., I. Verlaan, and W.H. Moolenaar, Dynamin is required for the activation of mitogen-activated protein (MAP) kinase by MAP kinase kinase. Journal of Biological Chemistry, 1999. 274: p. 35301-35304. 341. Kranenburg, O., I. Verlaan, and W.H. Moolenaar, Gi-mediated tyrosine phosphorylation of Grb2 (growth-factor-receptor-bound protein 2)-bound dynamin-II by lysophosphatidic acid. Biochemical Journal, 1999. 339: p. 11-14. 342. Lefkowitz, R.J. and S.K. Shenoy, Transduction of receptor signals by - arrestins. Science, 2005. 308: p. 512-517. 212
343. Goldsmith, Z.G. and D.N. Dhanasekaran, G Protein regulation of MAPK networks. Oncogene, 0000. 26(22): p. 3122-3142. 344. Kolch, W., et al., Protein kinase C activates RAF-1 by direct phosphorylation. Nature, 1993. 364: p. 249-252. 345. Ueda, Y., et al., Protein kinase C activates the MEK-ERK pathway in a manner independent of Ras and dependent on Raf. Journal of Biological Chemistry, 1996. 271: p. 23512-23519. 346. Schonwasser, D.C., et al., Activation of the mitogen-activated protein kinase//extracellular signal-regulated kinase pathway by conventional, novel, and atypical protein kinase C isotypes. Molecular Cell Biology, 1998. 18: p. 790-798. 347. Guo, F.F., E. Kumahara, and D. Saffen, A CalDAG-GEFI//Rap1//B-Raf cassette couples M1 muscarinic acetylcholine receptors to the activation of ERK1//2. Journal of Biological Chemistry, 2001. 276: p. 25568-25581. 348. Zhong, M., M. Yang, and B.M. Sanborn, Extracellular signal-regulated kinase 1//2 activation by myometrial oxytocin receptor involves GqG and epidermal growth factor receptor tyrosine kinase activation. Endocrinology, 2003. 144: p. 2947-2956. 349. Voyno-Yasenetskaya, T.A., et al., G12 and G13 regulate extracellular signal-regulated kinase and c-Jun kinase pathways by different mechanisms in COS-7 cells. Journal of Biological Chemistry, 1996. 271: p. 21081-21087. 350. Luttrell, L.M., et al., Activation and targeting of extracellular signal- regulated kinases by β-arrestin scaffolds. Proceedings of the National Academy of Sciences of the United States of America, 2001. 98(5): p. 2449-2454. 351. Caunt, C.J., et al., Seven-transmembrane receptor signalling and ERK compartmentalization. Trends in Endocrinology & Metabolism, 2006. 17(7): p. 276-283. 352. Barbara M, S., Hormonal signaling and signal pathway crosstalk in the control of myometrial calcium dynamics. Seminars in Cell & Developmental Biology, 2007. 18(3): p. 305-314. 353. Mesonero, J.E., et al., Differential Expression of Inositol 1,4,5- Trisphosphate Receptor Types 1, 2, and 3 in Rat Myometrium and Endometrium During Gestation. Biology of Reproduction, 2000. 63(2): p. 532-537. 354. Monga M, Campbell DF, and S. BM, Oxytocin-stimulated capacitative calcium entry in human myometrial cells. American Journal of Obstetrics and Gynecology, 1999. 181(2): p. 424-429. 355. Sanborn BM, et al., Molecular signaling through G-protein-coupled receptors and the control of intracellular calcium in myometrium. Journal of the Society for Gynecologic Investigation, 2005. 12(7): p. 479-487. 356. Wray, S., T. Burdyga, and K. Noble, Calcium signalling in smooth muscle. Cell Calcium, 2005. 38(3–4): p. 397-407. 213
357. Meera, P., et al., Relaxin stimulates myometrial calcium-activated potassium channel activity via protein kinase A. American Journal of Physiology - Cell Physiology, 1995. 269(2): p. C312-C317. 358. Huang, J., et al., Inhibition of Gαq-dependent PLC-β1 activity by PKG and PKA is mediated by phosphorylation of RGS4 and GRK2. American Journal of Physiology - Cell Physiology, 2007. 292(1): p. C200-C208. 359. Simhan, H.N. and S.N. Caritis, Prevention of Preterm Delivery. New England Journal of Medicine, 2007. 357(5): p. 477-487. 360. Ledingham, M.-A., et al., Nitric oxide in parturition. BJOG: An International Journal of Obstetrics & Gynaecology, 2000. 107(5): p. 581- 593. 361. Chapman, N., G. Europe-Finner, and S. Robson, Expression and deoxyribonucleic acid-binding activity of the nuclear factor κB family in the human myometrium during pregnancy and labor. Journal of Clinical Endocrinology & Metabolism, 2004. 89(11): p. 5683 - 5693. 362. Doheny, H.C., et al., Functional coupling of β3-adrenoceptors and large conductance calcium-activated potassium channels in human uterine myocytes. Journal of Clinical Endocrinology & Metabolism, 2005. 90(10): p. 5786-5796. 363. Risek, B., et al., Modulation of gap junction transcript and protein expression during pregnancy in the rat. The Journal of Cell Biology, 1990. 110(2): p. 269-282. 364. Gyetvai K, et al., Tocolytics for preterm labor: a systematic review. Obstetrics and Gynecology, 1999. 94(5 Pt 2): p. 869-877. 365. Snyder SW and C. MS, Neuromuscular blockade with magnesium sulfate and nifedipine. American Journal of Obstetrics and Gynecology, 1989. 161(1): p. 35-36. 366. Read MD and W. DE, The use of a calcium antagonist (nifedipine) to suppress preterm labour. British Journal of Obstetrics and Gynaecology 1986. 93(9): p. 933-937. 367. Ferguson JE 2nd, et al., Cardiovascular and metabolic effects associated with nifedipine and ritodrine tocolysis. American Journal of Obstetrics and Gynecology, 1989. 161(3): p. 788-795. 368. El-Sayed YY, et al., Randomized comparison of intravenous nitroglycerin and magnesium sulfate for treatment of preterm labor. Obstetrics and Gynecology, 1999. 93(1): p. 79-83. 369. Anotayanonth S, et al., Betamimetics for inhibiting preterm labour. Cochrane Database of Systematic Reviews (Online), 2004. 4: p. CD004352. 370. Dodd JM, et al., Oral betamimetics for maintenance therapy after threatened preterm labour. Cochrane Database of Systematic Reviews (Online), 2006. 1: p. CD003927. 371. Devost, D. and H.H. Zingg, Novel in vitro system for functional assessment of oxytocin action. American Journal of Physiology - Endocrinology and Metabolism, 2007. 292(1): p. E1-E6. 372. Condon, J., et al., Telomerase immortalization of human myometrial cells. Biology of Reproduction, 2002. 67(2): p. 506-514. 214
373. Adam, L.P., et al., Activation of mitogen-activated protein kinase in porcine carotid arteries. Circulation Research, 1995. 76(2): p. 183-190. 374. Morgan, K.G. and S.S. Gangopadhyay, Invited review: cross-bridge regulation by thin filament-associated proteins. Journal of Applied Physiology, 2001. 91(2): p. 953-962. 375. Li, Y., et al., ERK1/2-mediated phosphorylation of myometrial caldesmon during pregnancy and labor. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology, 2003. 284(1): p. R192-R199. 376. Ohmichi, M., et al., Role of mitogen-activated protein kinase pathway in prostaglandin F2-induced rat puerperal uterine contraction. Endocrinology, 1997. 138(8): p. 3103-3111. 377. Nohara, A., et al., The role of mitogen-activated protein kinase in oxytocin-induced contraction of uterine smooth muscle in pregnant rat. Biochemical and Biophysical Research Communications, 1996. 229(3): p. 938-944. 378. Warn-Cramer, B.J., et al., Regulation of connexin-43 gap junctional intercellular communication by mitogen-activated protein kinase. Journal of Biological Chemistry, 1998. 273(15): p. 9188-9196. 379. Li, Y., et al., Role of ERK1/2 in uterine contractility and preterm labor in rats. American Journal of Physiology - Regulatory, Integrative and Comparative Physiology, 2004. 287(2): p. R328-R335. 380. Oldenhof, A.D., et al., Mitogen-activated protein kinases mediate stretch- induced c-fos mRNA expression in myometrial smooth muscle cells. American Journal of Physiology - Cell Physiology, 2002. 283(5): p. C1530-C1539. 381. Challis, J.R.G., et al., Endocrine and paracrine regulation of birth at term and preterm. Endocrine Reviews, 2000. 21(5): p. 514-550. 382. Engstrom, T., et al., Up-regulation of oxytocin receptors in non-pregnant rat myometrium by isoproterenol: effects of steroids. Journal of Endocrinology, 1999. 161(3): p. 403-411. 383. Engstrøm, T., et al., Effect of oxytocin receptor and β2-adrenoceptor blockade on myometrial oxytocin receptors in parturient rats. Biology of Reproduction, 1999. 60(2): p. 322-329. 384. Dodge, K.L. and B.M. Sanborn, Evidence for inhibition by protein kinase A of receptor/Gαq/phospholipase C (PLC) coupling by a mechanism not involving PLCβ2. Endocrinology, 1998. 139(5): p. 2265-2271. 385. Angers, S., et al., Detection of β2-adrenergic receptor dimerization in living cells using bioluminescence resonance energy transfer (BRET). Proceedings of the National Academy of Sciences of the United States of America, 2000. 97(7): p. 3684-3689. 386. McVey, M., et al., Monitoring receptor oligomerization using time- resolved fluorescence resonance energy transfer and bioluminescence resonance energy transfer. Journal of Biological Chemistry, 2001. 276(17): p. 14092-14099. 387. Ramsay, D., et al., Homo- and hetero-oligomeric interactions between G- protein-coupled receptors in living cells monitored by two variants of 215
bioluminescence resonance energy transfer (BRET): hetero-oligomers between receptor subtypes form more efficiently than between less closely related sequences. Biochemical Journal, 2002. 365(2): p. 429-440. 388. Terrillon, S., et al., Oxytocin and vasopressin V1a and V2 receptors form constitutive homo- and heterodimers during biosynthesis. Molecular Endocrinology, 2003. 17(4): p. 677-691. 389. Devost, D. and H.H. Zingg, Homo- and hetero-dimeric complex formations of the human oxytocin receptor. Journal of Neuroendocrinology, 2004. 16(4): p. 372-377. 390. Hague, C., et al., Olfactory receptor surface expression is driven by association with the β2-adrenergic receptor. Proceedings of the National Academy of Sciences of the United States of America, 2004. 101(37): p. 13672-13676. 391. McGraw, D.W., et al., Airway smooth muscle prostaglandin-EP1 receptors directly modulate β2–adrenergic receptors within a unique heterodimeric complex. The Journal of Clinical Investigation, 2006. 116(5): p. 1400-1409. 392. Albizu, L., et al., Time-resolved FRET between GPCR ligands reveals oligomers in native tissues. Nature Chemical Biology, 2010. 6(8): p. 587- 594. 393. Dodge, K.L. and B.M. Sanborn, Evidence for inhibition by protein kinase A of receptor/Gq/Phospholipase C (PLC) coupling by a mechanism not involving PLC2. Endocrinology, 1998. 139(5): p. 2265-2271. 394. Arai, K., et al., Differential requirement of G12, G13, Gq, and G for endothelin-1-induced c-Jun NH2-terminal kinase and extracellular signal- regulated kinase activation. Molecular Pharmacology, 2003. 63: p. 478- 488. 395. Hasbi, A., et al., Real-time detection of interactions between the human oxytocin receptor and G protein-coupled receptor kinase-2. Molecular Endocrinology, 2004. 18(5): p. 1277-1286. 396. Devost, D. and H. Zingg, Identification of dimeric and oligomeric complexes of the human oxytocin receptor by co-immunoprecipitation and bioluminescence resonance energy transfer. Journal of Molecular Endocrinology, 2003. 31(3): p. 461-471. 397. Smith, Y., et al., Decrease in myometrial -adrenergic receptors with prenatal cocaine use. Obstetrics Gynecology, 1995. 85(3): p. 357-360. 398. Rezapour, M., T. Bäckström, and U. Ulmsten, Myometrial steroid concentration and oxytocin receptor density in parturient women at term. Steroids, 1996. 61(6): p. 338-344. 399. Robin, P., et al., ET-1 stimulates ERK signaling pathway through sequential activation of PRKC and Src in rat myometrial cells. American Journal of Physiology - Cell Physiology, 2002. 283(1): p. C251 - 260. 400. Chou, M.M., et al., Regulation of protein kinase C by PI 3-kinase and PDK-1. Current Biology, 1998. 8(19): p. 1069-1078. 401. Takeda, H.M., T.; Takada, T.; Noguchi, T.; Yamao, T.; Tsuda, M.; Ochi, F.; Fukunaga, K.; Inagaki, K.; Kasuga, M., PI3-kinase and protein 216
kinase C- mediate RAS-independent activation of MAP kinase by a Gi protein-coupled receptor. The EMBO Journal, 1999. 18(2): p. 386-395. 402. Monick, M.M., et al., Protein kinase C ζ plays a central role in activation of the p42/44 mitogen-activated protein kinase by endotoxin in alveolar macrophages. The Journal of Immunology, 2000. 165(8): p. 4632-4639. 403. Robin, P., et al., ET-1 stimulates ERK signaling pathway through sequential activation of PKC and Src in rat myometrial cells. American Journal of Physiology - Cell Physiology, 2002. 283: p. C251-C260. 404. Terrillon, S. and M. Bouvier, Roles of G-protein-coupled receptor dimerization. EMBO Reports, 2004. 5(1): p. 30-34. 405. Prinster, S.C., C. Hague, and R.A. Hall, Heterodimerization of G protein- coupled receptors: specificity and functional significance. Pharmacological Reviews, 2005. 57(3): p. 289-298. 406. Lohse, M.J., Dimerization in GPCR mobility and signaling. Current Opinion in Pharmacology, 2010. 10(1): p. 53-58. 407. Palczewski, K., Oligomeric forms of G protein-coupled receptors (GPCRs). Trends in Biochemical Sciences, 2010. 35(11): p. 595-600. 408. Naga Prasad, S.V., et al., Phosphoinositide 3-kinase regulates β2- adrenergic receptor endocytosis by AP-2 recruitment to the receptor/β- arrestin complex. The Journal of Cell Biology, 2002. 158(3): p. 563-575. 409. Rebois, R.V., et al., Combining protein complementation assays with resonance energy transfer to detect multipartner protein complexes in living cells. Methods, 2008. 45(3): p. 214-218. 410. Daulat, A.M., et al., Purification and identification of G protein-coupled receptor protein complexes under native conditions. Molecular & Cellular Proteomics, 2007. 6(5): p. 835-844. 411. Rebois, R.V., et al., Heterotrimeric G proteins form stable complexes with adenylyl cyclase and Kir3.1 channels in living cells. Journal of Cell Science, 2006. 119(13): p. 2807-2818. 412. Remy, I. and S. Michnick, Application of protein-fragment complementation assays in cell biology. BioTechniques, 2007. 42(2): p. 137-145. 413. Azzi, M., et al., Allosteric effects of G protein overexpression on the binding of β-adrenergic ligands with distinct inverse efficacies. Molecular Pharmacology, 2001. 60(5): p. 999-1007. 414. Wen, Y., et al., Protein kinase-A inhibits phospholipase-C activity and alters protein phosphorylation in rat myometrial plasma membranes. Endocrinology, 1992. 131(3): p. 1377-82. 415. Liu, M. and M.I. Simon, Regulation by cAMP-dependent protein kinease of a G-protein-mediated phospholipase C. Nature, 1996. 382(6586): p. 83- 87. 416. Kenakin, T., Agonist-specific receptor conformations. Trends in Pharmacological Sciences, 1997. 18(11): p. 416-417. 417. Strange, P., G-protein coupled receptors: conformations and states. Biochemical Pharmacology, 1999. 58: p. 1081-1088. 217
418. Han, Y., et al., Allosteric communication between protomers of dopamine class A GPCR dimers modulates activation. Nature Chemical Biology, 2009. 5(9): p. 688-695. 419. Zylbergold, P. and T.E. Hebert, A division of labor: asymmetric roles for GPCR subunits in receptor dimers. Nature Chemical Biology, 2009. 5(9): p. 608-609. 420. Agnati, L.F., et al., On the expanding terminology in the GPCR field: the meaning of receptor mosaics and receptor heteromers. Journal of Receptors and Signal Transduction, 2010. 30(5): p. 287-303. 421. Guo, W., Dopamine D2 receptors form higher order oligomers at physiological expression levels. The EMBO Journal, 2008. 27: p. 2293- 2304. 422. Vidi, P.-A., et al., Ligand-dependent oligomerization of dopamine D2 and adenosine A2A receptors in living neuronal cells. Molecular Pharmacology, 2008. 74(3): p. 544-551. 423. Vidi, P.-A., et al., Adenosine A2A receptors assemble into higher-order oligomers at the plasma membrane. FEBS Letters, 2008. 582(29): p. 3985-3990. 424. Navarro, G., et al., Detection of heteromers formed by cannabinoid CB1, dopamine D2, and adenosine A2A G-protein-coupled receptors by combining bimolecular fluorescence complementation and bioluminescence energy transfer. ScientificWorldJournal, 2008. 8: p. 1088-1097. 425. Gandia, J., et al., Detection of higher-order G protein-coupled receptor oligomers by a combined BRET-BiFC technique. FEBS Letters, 2008. 582(20): p. 2979-2984. 426. Hamatake, M., et al., Ligand-independent higher-order multimerization of CXCR4, a G-protein-coupled chemokine receptor involved in targeted metastasis. Cancer Science, 2009. 100(1): p. 95-102. 427. Dhanasekaran, N., L.E. Heasley, and G.L. Johnson, G protein-coupled receptor systems involved in cell growth and oncogenesis. Endocrine Reviews, 1995. 16: p. 259-270. 428. Rockman, H.A., W.J. Koch, and R.J. Lefkowitz, Seven-transmembrane- spanning receptors and heart function. Nature, 2002. 415(6868): p. 206- 212. 429. Claing, A., et al., Endocytosis of G protein-coupled receptors: roles of G protein-coupled receptor kinases and -arrestin proteins. Progress in Neurobiology, 2002. 66(2): p. 61-79. 430. Vanhaesebroeck, B., et al., Phosphoinositide 3-kinases: A conserved family of signal transducers. Trends in Biochemical Sciences, 1997. 22(7): p. 267-272. 431. Hawes, B.E., et al., Phosphatidylinositol 3-kinase is an early intermediate in the Gβγ-mediated mitogen-activated protein kinase signaling pathway. Journal of Biological Chemistry, 1996. 271(21): p. 12133-12136. 432. Song, J., et al., Tyrosine phosphorylation-dependent and -independent associations of protein kinase C- with Src family kinases in the RBL-2H3 218
mast cell line: regulation of Src family kinase activity by protein kinase C- . Oncogene, 1998. 16(26): p. 3357-3368. 433. Seibenhener, M.L., et al., Identification of Src as a novel atypical protein kinase C-interacting protein. Molecular Cell Biology Research Communications, 1999. 2(1): p. 28-31. 434. Vondriska, T.M., et al., Protein kinase Cε– Src modules direct signal transduction in nitric oxide– induced cardioprotection: complex formation as a means for cardioprotective signaling. Circulation Research, 2001. 88(12): p. 1306-1313. 435. Sabri, A., et al., Calcium- and protein kinase C–dependent activation of the tyrosine kinase PYK2 by angiotensin II in vascular smooth muscle. Circulation Research, 1998. 83(8): p. 841-851. 436. Kimura, A., et al., Mitogen-activated protein kinase cascade is involved in endothelin-1-induced rat puerperal uterine contraction. Endocrinology, 1999. 140(2): p. 722-731. 437. Witze, E.S., et al., Mapping protein post-translational modifications with mass spectrometry. Nature Methods, 2007. 4(10): p. 798-806. 438. Kosako, H. and K. Nagano, Quantitative phosphoproteomics strategies for understanding protein kinase-mediated signal transduction pathways. Expert Review of Proteomics, 2011. 8(1): p. 81-94. 439. Huppi, K., S.E. Martin, and N.J. Caplen, Defining and assaying RNAi in mammalian cells. Molecular Cell, 2005. 17(1): p. 1-10. 440. Michnick, S.W., et al., Universal strategies in research and drug discovery based on protein-fragment complementation assays. Nature Reviews Drug Discovery, 2007. 6(7): p. 569-582. 441. Mercier, J.-F., et al., Quantitative assessment of β1- and β2-adrenergic receptor homo- and heterodimerization by bioluminescence resonance energy transfer. Journal of Biological Chemistry, 2002. 277(47): p. 44925- 44931. 442. Karoor, V., et al., Insulin stimulates sequestration of β-adrenergic receptors and enhanced association of β-adrenergic receptors with Grb2 via tyrosine 350. Journal of Biological Chemistry, 1998. 273(49): p. 33035-33041. 443. Fan, G.-f., et al., c-Src tyrosine kinase binds the β2-adrenergic receptor via phospho-Tyr-350, phosphorylates G-protein-linked receptor kinase 2, and mediates agonist-induced receptor desensitization. Journal of Biological Chemistry, 2001. 276(16): p. 13240-13247. 444. Shumay, E., et al., pp60Src Mediates insulin-stimulated sequestration of the β2-adrenergic receptor: insulin stimulates pp60Src phosphorylation and activation. Molecular Biology of the Cell, 2002. 13(11): p. 3943- 3954. 445. Klein, U., et al., A novel interaction between adrenergic receptors and the -subunit of eukaryotic initiation factor 2B. Journal of Biological Chemistry, 1997. 272(31): p. 19099-19102. 219
446. Shih, M., et al., Dynamic complexes of β2-adrenergic receptors with protein kinases and phosphatases and the role of gravin. Journal of Biological Chemistry, 1999. 274(3): p. 1588-1595. 447. Lin, F., H.-y. Wang, and C.C. Malbon, Gravin-mediated formation of signaling complexes in β2-adrenergic receptor desensitization and resensitization. Journal of Biological Chemistry, 2000. 275(25): p. 19025- 19034. 448. Fan, G.-f., et al., The scaffold protein gravin (cAMP-dependent protein kinase-anchoring protein 250) binds the β2-adrenergic receptor via the receptor cytoplasmic Arg-329 to Leu-413 domain and provides a mobile scaffold during desensitization. Journal of Biological Chemistry, 2001. 276(26): p. 24005-24014. 449. Fraser, I.D.C., et al., Assembly of an A kinase-anchoring protein–β2- adrenergic receptor complex facilitates receptor phosphorylation and signaling. Current Biology, 2000. 10(7): p. 409-412. 450. Cong, M., et al., Regulation of membrane targeting of the G protein- coupled receptor kinase 2 by protein kinase A and its anchoring protein AKAP79. Journal of Biological Chemistry, 2001. 276(18): p. 15192- 15199. + + 451. Hall, R.A., et al., The β2-adrenergic receptor interacts with the Na /H - exchanger regulatory factor to control Na+/H+ exchange. Nature, 1998. 392(6676): p. 626-630. 452. Hall, R.A., et al., A C-terminal motif found in the β2-adrenergic receptor, P2Y1 receptor and cystic fibrosis transmembrane conductance regulator determines binding to the Na+/H+ exchanger regulatory factor family of PDZ proteins. Proceedings of the National Academy of Sciences of the United States of America, 1998. 95(15): p. 8496-8501. 453. Cong, M., et al., Binding of the β2 adrenergic receptor to N- ethylmaleimide-sensitive factor regulates receptor recycling. Journal of Biological Chemistry, 2001. 276(48): p. 45145-45152. 454. Yamada, M., A. Inanobe, and Y. Kurachi, G protein regulation of potassium ion channels. Pharmacological Reviews, 1998. 50(4): p. 723- 757. 455. Davare, M.A., et al., A β2 adrenergic receptor signaling complex assembled with the Ca2+ channel Cav1.2. Science, 2001. 293(5527): p. 98- 101. 456. Raman, I.M., G. Tong, and C.E. Jahr, β-adrenergic regulation of synaptic NMDA receptors by cAMP-dependent protein kinase. Neuron, 1996. 16(2): p. 415-421. 457. Kim, J., et al., Regulation of epidermal growth factor receptor internalization by G protein-coupled receptors. Biochemistry, 2003. 42(10): p. 2887-2894. 458. Drube, S., et al., Ligand-independent and EGF receptor-supported transactivation: Lessons from β2-adrenergic receptor signalling. Cellular Signalling, 2006. 18(10): p. 1633-1646. 459. Kassel, K.M., N.A. Schulte, and M.L. Toews, Modulation of epidermal growth factor receptor binding to human airway smooth muscle cells by 220
glucocorticoids and β2-adrenergic receptor agonists. American Journal of Physiology - Lung Cellular and Molecular Physiology, 2009. 296(4): p. L693-L699. 460. Bockaert, J., et al., GPCR-interacting proteins (GIPs): nature and functions. Biochemical Society Transactions, 2004. 32(5): p. 851-855. 461. Lindeman, K.S., et al., Chronic oxytocin pretreatment inhibits adenylyl cyclase activity in cultured rat myometrial cells. Biology of Reproduction, 1998. 59(5): p. 1108-1115. 462. Guimond, J., et al., Role of specific protein kinase C isoforms in modulation of 1- and 2-adrenergic receptors. Cellular Signalling, 2005. 17(1): p. 49-58. 463. Anwer, K., J.A. Hovington, and B.M. Sanborn, Antagonism of contractants and relaxants at the level of intracellular calcium and phosphoinositide turnover in the rat uterus. Endocrinology, 1989. 124(6): p. 2995-3002. 464. Anwer, K., J.A. Hovington, and B.M. Sanborn, Involvement of protein kinase A in the regulation of intracellular free calcium and phosphoinositide turnover in rat myometrium. Biology of Reproduction, 1990. 43(5): p. 851-859. 465. Glass, M. and C.C. Felder, Concurrent stimulation of cannabinoid CB1 and dopamine D2 receptors augments cAMP accumulation in striatal neurons: evidence for a Gs linkage to the CB1 receptor. The Journal of Neuroscience, 1997. 17(14): p. 5327-5333. 466. Jarrahian, A., V.J. Watts, and E.L. Barker, D2 dopamine receptors modulate Gα-subunit coupling of the CB1 cannabinoid receptor. Journal of Pharmacology and Experimental Therapeutics, 2004. 308(3): p. 880- 886. 467. Grant, M., et al., Cell growth inhibition and functioning of human somatostatin receptor type 2 are modulated by receptor heterodimerization. Molecular Endocrinology, 2008. 22(10): p. 2278- 2292. 468. Lavine, N., et al., G protein-coupled receptors form stable complexes with inwardly rectifying potassium channels and adenylyl cyclase. Journal of Biological Chemistry, 2002. 277(48): p. 46010-46019. 469. Kitano, J., et al., Direct interaction and functional coupling between metabotropic glutamate receptor subtype 1 and voltage-sensitive Cav2.1 Ca2+ channel. Journal of Biological Chemistry, 2003. 278(27): p. 25101- 25108. 470. Liu, G., et al., Assembly of a Ca2+-dependent BK channel signaling complex by binding to 2 adrenergic receptor. The EMBO Journal, 2004. 23(11): p. 2196-2205. 471. Nikolov, E.N. and T.T. Ivanova-Nikolova, Coordination of membrane excitability through a GIRK1 signaling complex in the atria. Journal of Biological Chemistry, 2004. 279(22): p. 23630-23636. 472. Balijepalli, R.C., et al., Localization of cardiac L-type Ca2+ channels to a caveolar macromolecular signaling complex is required for β2-adrenergic 221
regulation. Proceedings of the National Academy of Sciences of the United States of America, 2006. 103(19): p. 7500-7505. 473. Philip, F., P. Sengupta, and S. Scarlata, Signaling through a G protein- coupled receptor and its corresponding G protein follows a stoichiometrically limited model. Journal of Biological Chemistry, 2007. 282(26): p. 19203-19216. 474. Robitaille, M., et al., Intracellular trafficking and assembly of specific Kir3 channel/G protein complexes. Cellular Signalling, 2009. 21(4): p. 488-501. 475. Dupré, D.J., et al., Signalling complexes associated with adenylyl cyclase II are assembled during their biosynthesis. Cellular Signalling, 2007. 19(3): p. 481-489. 476. Wu, G., et al., Receptor docking sites for G-protein βγ subunits. Journal of Biological Chemistry, 1998. 273(13): p. 7197-7200. 477. Wu, G., et al., Identification of Gβγ binding sites in the third intracellular loop of the M3-muscarinic receptor and their role in receptor regulation. Journal of Biological Chemistry, 2000. 275(12): p. 9026-9034. 478. Mahon, M.J., et al., A docking site for G protein βγ subunits on the parathyroid hormone 1 receptor supports signaling through multiple pathways. Molecular Endocrinology, 2006. 20(1): p. 136-146. 479. Dupré, D.J., et al., Seven transmembrane receptor core signaling complexes are assembled prior to plasma membrane trafficking. Journal of Biological Chemistry, 2006. 281(45): p. 34561-34573. 480. Dupré, D.J. and T.E. Hébert, Biosynthesis and trafficking of seven transmembrane receptor signalling complexes. Cellular Signalling, 2006. 18(10): p. 1549-1559. 481. Bernstein, L.S., et al., RGS2 binds directly and selectively to the M1 muscarinic acetylcholine receptor third intracellular loop to modulate Gq/11α signaling. Journal of Biological Chemistry, 2004. 279(20): p. 21248-21256. 482. Benians, A., et al., Regulators of G-protein signaling form a quaternary complex with the agonist, receptor, and G-protein. Journal of Biological Chemistry, 2005. 280(14): p. 13383-13394. 483. Roy, A.A., et al., RGS2 interacts with Gs and adenylyl cyclase in living cells. Cellular Signalling, 2006. 18(3): p. 336-348. 484. Jaén, C. and C.A. Doupnik, RGS3 and RGS4 differentially associate with G protein-coupled receptor-Kir3 channel signaling complexes revealing two modes of RGS modulation. Journal of Biological Chemistry, 2006. 281(45): p. 34549-34560. 485. Dowal, L., et al., Determination of the contact energies between a regulator of G protein signaling and G protein subunits and phospholipase Cβ1. Biochemistry, 2000. 40(2): p. 414-421. 486. Fujita, S., et al., A regulator of G protein signalling (RGS) protein confers agonist-dependent relaxation gating to a G protein-gated K+ channel. The Journal of Physiology, 2000. 526(2): p. 341-347. 487. Schiff, M.L., et al., Tyrosine-kinase-dependent recruitment of RGS12 to the N-type calcium channel. Nature, 2000. 408(6813): p. 723-727. 222
488. Hall, R.A. and R.J. Lefkowitz, Regulation of G protein–coupled receptor signaling by scaffold proteins. Circulation Research, 2002. 91(8): p. 672- 680. 489. Maurice, P., et al., GPCR-interacting proteins, major players of GPCR function. Advances in Pharmacology, 2011. 62: p. 349-380. 490. Cao, T.T., et al., A kinase-regulated PDZ-domain interaction controls endocytic sorting of the 2-adrenergic receptor. Nature, 1999. 401(6750): p. 286-290. 491. Kapiloff, M.S., Contributions of protein kinase A anchoring proteins to compartmentation of cAMP signaling in the heart. Molecular Pharmacology, 2002. 62(2): p. 193-199. 492. Dell’Acqua, M.L., et al., Regulation of neuronal PKA signaling through AKAP targeting dynamics. European Journal of Cell Biology, 2006. 85(7): p. 627-633. 493. Rosenfeld, J., B. Knoll, and R. Moore, Regulation of G-protein-coupled receptor activity by rab GTPases. Receptors Channels, 2002. 8(2): p. 87- 97. 494. Seachrist, J.L. and S.S.G. Ferguson, Regulation of G protein-coupled receptor endocytosis and trafficking by Rab GTPases. Life Sciences, 2003. 74(2-3): p. 225-235. 495. Purvanov, V., A. Koval, and V.L. Katanaev, A direct and functional interaction between Go and Rab5 during G protein-coupled receptor signaling. Science Signalling, 2010. 3(136): p. ra65-. 496. Pontier, S.M., et al., Cholesterol-dependent separation of the β2- adrenergic receptor from its partners determines signaling efficacy. Journal of Biological Chemistry, 2008. 283(36): p. 24659-24672. 497. Engstrom, T., A. Atke, and H. Vilhardt, Receptor-binding characteristics and contractile responsiveness of the myometrium following prolonged infusion of bradykinin and oxytocin in rats. Journal of Endocrinology, 1988. 118(1): p. 81-85. 498. Vallet, J.L., G.E. Lamming, and M. Batten, Control of endometrial oxytocin receptor and uterine response to oxytocin by progesterone and oestradiol in the ewe. Journal of Reproduction and Fertility, 1990. 90(2): p. 625-634. 499. Lau, T.M., et al., Increase in concentration of uterine oxytocin receptors and decrease in response to 13,14-dihydro-15-keto prostaglandin F2α in ewes after withdrawal of exogenous progesterone. Journal of Reproduction and Fertility, 1992. 95(3): p. 885-893. 500. Sheldrick, E.L. and H.C. Flick-Smith, Effect of ovarian hormones on oxytocin receptor concentrations in explants of uterus from ovariectomized ewes. Journal of Reproduction and Fertility, 1993. 97(1): p. 241-245. 501. Quinones-Jenab, V., et al., Effects of estrogen on oxytocin receptor messenger ribonucleic acid expression in the uterus, pituitary, and forebrain of the female rat. Neuroendocrinology, 1997. 65(1): p. 9-17. 223
502. Larcher, A., et al., Oxytocin receptor gene expression in the rat uterus during pregnancy and the estrous cycle and in response to gonadal steroid treatment. Endocrinology, 1995. 136(12): p. 5350-6. 503. Breton, C., et al., Oxytocin receptor messenger ribonucleic acid: characterization, regulation, and cellular localization in the rat pituitary gland. Endocrinology, 1995. 136(7): p. 2928-36. 504. Breton, C. and H.H. Zingg, Expression and region-specific regulation of the oxytocin receptor gene in rat brain. Endocrinology, 1997. 138(5): p. 1857-1862. 505. Bale, T. and D. Dorsa, NGF, cyclic AMP, and phorbol esters regulate oxytocin receptor gene transcription in SK-N-SH and MCF7 cells. Brain Research Molecular Brain Research, 1998. 53(1-2): p. 130-137. 506. Engstrom, T., et al., 2-adrenoceptor desensitization in non-pregnant estrogen-primed rat myometrium involves modulation of oxytocin receptor gene expression. Journal of Molecular Endocrinology, 1998. 20(2): p. 261- 270. 507. Bailey, J., et al., Expression of the cyclic AMP-dependent transcription factors, CREB, CREM and ATF2, in the human myometrium during pregnancy and labour. Molecular Human Reproduction, 2000. 6(7): p. 648-660.
224
Abstracts
1. Wrzal, P.K., Hébert, T.E., and H.H. Zingg. Co-expression with the Oxytocin receptor enables the 2-adrenergic receptor to signal through Gq and PKC : Possible effect of heterodimerization. 12th Annual Great Lakes GPCR Joint Meeting, King City, Ontario, Canada. October 21-23, 2010.
2. Wrzal, P.K., Hébert, T.E., and H.H. Zingg. Evidence for allosteric interactions between oxytocin and 2-adrenergic receptors in human myometrial cells. 92nd Annual Meeting US Endocrine Society, San Diego, California, USA. June 19-22, 2010.
3. Wrzal, P.K., Hébert, T.E., and H.H. Zingg. Evidence for allosteric interactions between oxytocin and 2-adrenergic receptors. Pharmacology Research Day, McGill University, Montreal, Quebec, Canada, 2010.
4. Wrzal, P.K., Hébert, T.E., and H.H. Zingg. Allosteric communication between th the oxytocin and 2-adrenergic receptors modulates ERK1/2 activation. 11 Annual Great Lakes GPCR Joint Meeting, Rochester, New York, USA. October15-19, 2009.
5. Wrzal, P.K., Hébert, T.E., and H.H. Zingg. Crosstalk between oxytocin and 2-adrenergic receptors in mediating ERK1/2 activation. Pharmacology Research Day, McGill University, Montreal, Quebec, Canada, 2009.
6. Wrzal, P.K., Hébert, T.E., and H.H. Zingg. The mechanism of ERK1/2 activation by the myometrial oxytocin receptor: A matter of time. 10th Annual Great Lakes GPCR Joint Meeting, Bromont, Quebec, Canada. October 16-18, 2008.
7. Wrzal, P.K., Devost, D., Hébert, T.E., and H.H. Zingg. Interactions between the oxytocin and 2-adrenergic receptors: structural and functional implications. Pharmacology Research Day, McGill University, Montreal, Quebec, Canada, 2008.
225
8. Wrzal, P.K., Devost, D., Hébert, T.E., and H.H. Zingg. Interactions between the oxytocin and 2-adrenergic receptors: structural and functional implications. 90th Annual Meeting US Endocrine Society, San Francisco, California, USA. June 15-18, 2008.
9. Wrzal, P.K., Devost, D., Hébert, T.E., and H.H. Zingg. Interactions between the oxytocin and 2-adrenergic receptors: structural and functional implications. 9th Annual Great Lakes GPCR Joint Meeting, London, Ontario, Canada. September 27-29, 2007.
Publications
1. Wrzal, P.K., Devost, D., Pétrin, D., Goupil, E., Iorio-Morin, C., Laporte, S.A., Zingg, H.H., and T.E. Hébert. Allosteric interactions between the oxytocin receptor and the β2-adrenergic receptor in the modulation of ERK1/2 activation are mediated by heterodimerization. Cellular Signalling, 2012. 24(1): p.342-350.
2. Wrzal, P.K., Goupil, E., Laporte, S.A., Hébert, T.E., and H.H. Zingg. Functional interactions between the oxytocin receptor and the β2-adrenergic receptor: Implications for ERK1/2 activation in human myometrial cells. Cellular Signalling, 2012. 24(1): p.333-341.
3. Devost, D., Wrzal, P., and H.H. Zingg. Oxytocin receptor signalling, in Progress in Brain Research, I.D. Neumann and R. Landgraf, Editors. 2008, Vol 170, Elsevier. p.167-176.
Cellular Signalling 24 (2012) 342–350
Contents lists available at SciVerse ScienceDirect
Cellular Signalling
journal homepage: www.elsevier.com/locate/cellsig
Allosteric interactions between the oxytocin receptor and the β2-adrenergic receptor in the modulation of ERK1/2 activation are mediated by heterodimerization
Paulina K. Wrzal a, Dominic Devost a, Darlaine Pétrin a, Eugénie Goupil a, Christian Iorio-Morin a, Stéphane A. Laporte a,b, Hans H. Zingg a,b,c,⁎, Terence E. Hébert a,⁎⁎ a Department of Pharmacology and Therapeutics, McGill University, Montréal, Québec, Canada H3G 1Y6 b Department of Medicine, McGill University, Montréal, Québec, Canada H3G 1Y6 c Department of Obstetrics and Gynecology, McGill University, Montréal, Québec, Canada H3G 1Y6 article info abstract
Article history: The oxytocin receptor (OTR) and the β2-adrenergic receptor (β2AR) are key regulators of uterine contraction. Received 9 August 2011 These two receptors are targets of tocolytic agents used to inhibit pre-term labor. Our recent study on the nature Received in revised form 9 September 2011 of OTR- and β2AR-mediated ERK1/2 activation in human hTERT-C3 myometrial cells suggested the presence of Accepted 11 September 2011 an OTR/β AR hetero-oligomeric complex (see companion article). The goal of this study was to investigate po- Available online 22 September 2011 2 tential allosteric interactions between OTR and β2AR and establish the nature of the interactions between fi Keywords: these receptors in myometrial cells. We found that OTR-mediated ERK1/2 activation was attenuated signi cantly β Oxytocin receptor when cells were pretreated with the 2AR agonist isoproterenol or two antagonists, propranolol or timolol. In β β2-adrenergic receptor contrast, pretreatment of cells with a third 2AR antagonist, atenolol resulted in an increase in OTR-mediated ERK1/2 ERK1/2 activation. Similarly, β2AR-mediated ERK1/2 activation was strongly attenuated by pretreatment with Allosteric modulation the OTR antagonists, atosiban and OTA. Physical interactions between OTR and β2AR were demonstrated using Signalling co-immunoprecipitation, bioluminescence resonance energy transfer (BRET) and protein-fragment complemen- Dimerization tation (PCA) assays in HEK 293 cells, the latter experiments indicating the interactions between the two recep-
tors were direct. Our analyses suggest physical interactions between OTR and β2AR in the context of a new heterodimer pair lie at the heart of the allosteric effects. © 2011 Elsevier Inc. All rights reserved.
1. Introduction stimuli to the cell interior [1]. These receptors are also the most com- mon targets of currently available drugs [2]. Initially, GPCRs were G protein-coupled receptors (GPCRs) make up the largest and thought to exist only as monomers, however, during the past most diverse family of transmembrane receptors encoded by the 20 years this view has been substantially altered. What has emerged human genome, relaying information from diverse extracellular is a view that the physiologically relevant GPCR signalling unit may consist of dimers or higher-structure oligomers of receptors [3]. Recent research suggests that GPCRs possess additional topo- graphically distinct allosteric binding sites, recognized by structurally Abbreviations: AC, adenylyl cyclase; AT, atenolol; β2AR, β2-adrenergic receptor; BRET, bioluminescence resonance energy transfer; cAMP, cyclic adenosine monophosphate; diverse allosteric modulators, which can regulate the activity of the DMEM, Dulbecco's modified Eagle's medium; DMSO, dimethylsulphoxide; DNA, endogenous (or orthosteric) ligand, changing agonist- and antagonist- deoxyribonucleic acid; ERK, extracellular signal-regulated kinase; FBS, fetal bovine mediated responses to affect signalling pathways selectively [4]. Alloste- serum; GFP, green fluorescent protein; GPCR, G protein-coupled receptor; HRP, ric ligands can either positively or negatively modulate receptor activity horseradish peroxide; hTERT-C3, myometrial human telomerase reverse transcriptase- clone3 cells; ISO, isoproterenol; OT, oxytocin; OTA, oxytocin receptor antagonist; OTR, depending on the pathway being investigated, thus can act as biased li- oxytocin receptor; PBS, phosphate buffered saline; PCA, protein-fragment complementation gands [4]. The existence of GPCRs as dimers conjures up the possibility assay; PCR, polymerase chain reaction; PKA, protein kinase A; PKC, protein kinase C; PRO, of new functional properties distinct from the individual parent recep- propranolol; Rluc, Renilla luciferase; RNA, ribonucleic acid; TIM, timolol; YFP, yellow tors. This combined with the fact that only about 40 GPCRs have been fluorescent protein. ⁎ Correspondence to: H.H. Zingg, Department of Pharmacology and Therapeutics, validated as true therapeutic targets (described until recently as agonists McGill University, 3655 Promenade Sir-William-Osler, Room 1325, Montréal, Québec, or antagonists at the orthosteric ligand-binding site) indicate a large po- Canada H3G 1Y6. Tel.: +1 514 398 3621; fax: +1 514 398 2045. tential for further drug discovery [4]. Understanding how different li- ⁎⁎ Correspondence to: T.E. Hébert, Department of Pharmacology and Therapeutics, gands for the parent receptors regulate a GPCR within a heterodimer, McGill University, 3655 Promenade Sir-William-Osler, Room 1303, Montréal, Québec, through allosteric effects between the two protomers of the dimer or H3G 1Y6, Canada. Tel.: +1 514 398 1398; fax: +1 514 398 6690. E-mail addresses: [email protected] (H.H. Zingg), [email protected] the protomer(s) and associated G proteins, is crucial in the development (T.E. Hébert). of new therapeutic strategies.
0898-6568/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.cellsig.2011.09.020 P.K. Wrzal et al. / Cellular Signalling 24 (2012) 342–350 343
We have demonstrated a novel interaction between two GPCRs, (Université de Montréal, Montreal, Canada). The HA-OTR construct was the oxytocin receptor (OTR) and the β2-adrenergic receptor (β2AR), generated by PCR from human OTR using specific primers (forward: in human myometrial cells (see companion article). In these studies 5′-TTATGCCTGCGGATCCGAGGGCGCGCTCGCAGCCAACT-3′,reverse5′- the β2AR was coupled to a Gi/PI3-kinase/PKCζ/Src-dependent path- TTTAAACGCCGGATCTCACGCCGTGGATGGCTGGGA-3′). Using the In- way to mediate ERK1/2 activation in hTERT-C3 myometrial cells, but Fusion PCR cloning system (Clontech), the ends of the PCR fragment only in the presence of OTR. These receptors display an overlapping were fused to the homologous ends of a BamHI linearized pIRESpuro3 distribution in the human myometrium, where they have opposing vector already containing the sequence for an HA tag located at the N- effects on uterine contraction. Activation of the OTR leads to uterine terminus of the receptor sequence. The Flag-β2AR construct was generat- contraction, while activation of the β2AR leads to uterine relaxation. ed by PCR using HA-β2AR as template and the following primers; forward In addition, the two receptors represent pharmacological targets of primer contained NheI, Kozak and Flag sites (5′-GCAGCTAGCGCCACC- drugs currently used to control pre-term uterine contractions, OTR ATGGATTATAAGGACGATGACGATAAGATGGGGCAACCCGGGAAC-3′)and antagonists and β2AR agonists [5]. reverse primer contained BamHI site (5′-CGTGGATCCTTACAGCAGT- Both OTR and β2AR have been described to form homo- and het- GAGTCATTTG-3′). The PCR product was digested with NheI/BamHI and erodimers. Dimerization of OTR has actually been demonstrated ex subcloned into pIRESpuro3. The cloning strategy used to generate β2AR- vivo [6]. The OTR has also been shown to form heterodimers with va- Venus (1) and β2AR-Venus (2) has previously been described [20].To sopressin V1a and V2 receptors [7,8]. The β2AR has been shown to construct plasmids coding for human OTR and rat Alk5 with a N- or C- homodimerize and to form heterodimers with several other GPCRs terminal fragments of the GFP variant Venus, receptor coding sequences such as β1-and β3-adrenoceptors, the prostaglandin EP1 receptor, ol- were amplified by PCR using the following specific primers (OTR for- factory receptors as well as the μ-, δ- and κ-opioid receptors [9–17], ward : 5′-GAATTGCGGCCGCCACCATGGAGGGCGCGCTCGCAGCCAACT-3′, among others. These studies suggest that GPCRs with multiple poten- reverse: 5′-GCCACCATCGATCGCCGTGGATGGCTGGGAGC-3′;Alk5for- tial dimer partners can form distinct signalling hubs, about which ward: 5′-CTAAGCGGCCGCGCCACCATGGAGGCAGCATCGGCTGCTT-3′,re- unique hetero-oligomeric complexes can be built with distinct phys- verse 5′-GTAGCATCGATCTTGTCGTCGTCGTCCTT-3′). The resulting PCR iological partners and functions. Physical interactions between GPCRs fragments were cloned NotI and ClaI to replace the GCN4 leucine to form dimers promote changes in receptor function, which have zipper from pcDNA3.1/Zeo (+)-GCN4 leucine zipper-Venus (1) and been shown to influence ligand-binding affinity and specificity as pcDNA3.1/Zeo (+)-GCN4 leucine zipper-Venus (2) cDNAs, generous well as downstream coupling, trafficking and desensitization of the gifts from Dr. Stephen Michnick (Université de Montréal, Montreal, QC). receptors. Unraveling the nature of the interactions, which exist be- All constructs were verified by restriction digestion and bidirectional se- tween OTR and β2AR, is of significant pharmacological and physiolog- quencing. All constructs tagged for either BRET or the PCA were verified ical importance. In the present study we have identified a novel in functional assays to affirm that they retained the character of the physical interaction between OTR and β2AR when expressed in HEK untagged versions (data not shown). 293 cells, which could form the basis of unique signalling patterns seen when the receptors are co-expressed in native tissues. These in- 2.3. Cell culture and transfection teractions allowed for allosteric modulation of ERK1/2 activation mediated through one receptor partner by the other in human Human myometrial cells immortalized by transfection with myometrial hTERT-C3 cells. human telomerase reverse transcriptase (hTERT-HM cells) were obtained from W. E. Rainey [21]. hTERT-C3 cells represent a selected 2. Materials and methods subclone that we obtained by serial clonal dilution of hTERT-HM cells as described previously [22]. These cells were maintained in 2.1. Reagents DMEM/F12 medium supplemented with 10% FBS and cultured at
37 C in 5% CO2. Cells close to confluency were passaged by trypsiniza- Reagents were obtained from the following sources: DMEM/F12 tion and plated in T175 flask at a one-quarter dilution every 4–5d. tissue culture media was from Invitrogen (Burlington, ON); fetal bo- HEK 293 cells were from Invitrogen (Burlington, ON) and were vine serum (FBS) from Hyclone (Logan, UT); OT was from Sigma- grown in Dulbecco's modified Eagle's medium high glucose (DMEM/ Aldrich (St. Louis, MO); isoproterenol and timolol were from Toctris high glucose) supplemented with 5% FBS and transfected using Bioscience (Bristol, UK); and atenolol was from Research Biochemi- Lipofectamine 2000 as per the manufacturer's instructions. Experi- cals International (Natick, MA). Propranolol was from Sigma-Aldrich ments were generally carried out 48 h after transfection. (St. Louis, MO). OTA was obtained from Bachem (Torrence, CA) and Atosiban was a kind gift from Ferring Research Institute Inc. (San 2.4. Western blots Diego, CA). Pertussis toxin (PTX) and protease inhibitors: leupeptin, benzamidine and trypsin were from Sigma-Aldrich (St. Louis, MO). Immunoblotting was used to assess the levels of cellular proteins. Primary anti-phospho-ERK1/2 (T202, Y204) antibody was from Cell Briefly, cells were plated onto culture dishes and grown in appropri- Signaling Technology (Beverly, MA); pan anti- ERK1/2 antibody was ate media to near confluency. Cells were starved for 24 h in either from Stressgen (Ann Arbor, MI). HA antibodies were from Covance. DMEM/F12 supplemented with only 0.5% FBS (hTERT-C3 cells) or Flag antibody and secondary antibody, horseradish peroxidase- DMEM/high glucose supplemented with only 0.5% FBS (HEK 293 conjugated anti-rabbit IgG, were from Sigma-Aldrich (St. Louis, cells). Cells were stimulated with ligand(s) for different times at MO). Lipofectamine 2000 was from Invitrogen (Burlington, ON). 37 °C. Stimulation was stopped by two ice-cold PBS washes, and 50% slurry EZview Red Anti-Flag and Anti-mouse affinity gel beads plates were flash frozen in liquid nitrogen. For inhibitor studies, and Flag peptide were from Sigma-Aldrich (St. Louis, MO); Coelenter- cells were pre-treated with the inhibitor in question for 30 min — azine h was from Molecular Probes (Eugene, OR). All other analytical 1 h before stimulation with ligand(s). Cells were lysed on ice with grade chemicals were obtained from Sigma-Aldrich (St. Louis, MO), Laemmli buffer [50 mM Tris–HCl (pH 6.8), 2% sodium dodecyl sulfate, Fisher Scientific (Waltham, MA), and VWR (West Chester, PA). 10% glycerol, and 0.1 M β-mercaptoethanol] and were either homog- enized using syringe and needle (hTERT-C3 cells) or sonication (HEK 2.2. Constructs 293 cells). Lysates were clarified by centrifugation at 15, 000×g for 10 min at 4 °C in a microcentrifuge. Proteins were denatured by boil-
The OTR-Rluc and β2AR-YFP constructs have been previously de- ing for 5 min, and subjected to SDS-PAGE and western blotting. scribed [18,19]. The GABAB2-YFP was a kind gift from Dr. Michel Bouvier Immunodetection involved different primary antibodies in conjunction 344 P.K. Wrzal et al. / Cellular Signalling 24 (2012) 342–350
with a second horseradish peroxidase-conjugated antibody and a β2AR-Venus2, OTR-Venus 1 and β2AR-Venus 2, or OTR-Venus1 and chemiluminescence detection system (Supersignal; Pierce). Quantifica- Alk5-Venus2). For confocal microscopy, cells were maintained in tion of band intensities was performed using AlphaEase (Alpha Innotech DMEM supplemented with 10% (v/v) heat-inactivated fetal bovine Corp., San Leandro, CA). serum and gentamycin (100 μg/ml) and grown on 35 mm plates. 48 h post-transfection cells were starved in serum-free media for 2.5. Receptor co-immunoprecipitation 30 min. Images were collected using live-cell microscopy at 37 °C on a Zeiss LSM-510 Meta laser scanning microscope equipped with XL- HEK 293 cells were grown and transfected as described above. All 3 temperature chamber with a 63× glycerol/water immersion lens steps in membrane preparation and receptor co-immunoprecipitation in single track mode using excitation at 515 nm for YFP and emis- procedures were done on ice and with ice-cold solutions. Membrane sion measured with the BP530-600 filter set. For direct measure- preparation and solubilization were performed as described elsewhere ment of reconstituted Venus levels, cells were collected 48 h post- with the following modifications [23]. 48 h post-transfection, cells were transfection, placed in 96-well white OptiPlate-96 plates(Perkin washed once with cold PBS 1X and resuspended in lysis buffer (5 mM Elmer) and fluorescence from reconstituted Venus was measured Tris–HCl pH 7.4, 2 mM EDTA) containing the protease inhibitors leu- directly using a Synergy2 microplate reader (BioTek), with excitation peptin 5 μg/ml, benzamide 10 μg/ml and soybean trypsin inhibitor 485/20 nm and emission 528/20 nm filters. 5 μg/ml. Cells were then homogenized using a Brinkman Polytron at 10,000 rpm for 2×10 s and cellular debris was removed by centrifuga- 2.8. Data analysis tion for 5 min at 1000 rpm. The supernatant was transferred to 15 ml clear Sorval tubes and membranes were obtained by centrifugation All data are presented as means±SEM. When not shown, error bars for 20 min at 16,000 rpm. Membranes (pellet) were then solubilized lie within symbols. Statistical differences between paired groups were at 4 °C for 30 min in solubilization buffer (75 mM Tris pH 8.0, 2 mM calculated using Student's t-test and comparisons between treated
EDTA, 5 mM MgCl2,0.5%n-Dodecyl β-D-maltoside) containing protease cells and untreated control cells were made using one-way ANOVA inhibitors. The soluble and insoluble membrane fractions were then with Dunnett adjustment. One-sample t-test was used to compare treat- separated by centrifugation for 20 min at 18,300 rpm and supernatant ment groups to the normalized maxima. Differences were considered was used for co-immunoprecipitation. 40 μl/sample of either a 50% slur- statistically significant at Pb0.05. Data were represented graphically ry of EZview Red Anti-Flag or Anti-HA affinity gel beads were washed using GraphPad Prism version 4 (GraphPad Software, San Diego, CA). twice with 500 μl of solubilization buffer (centrifuged for 2 min at 1600 rpm between washing steps). Solubilized membranes (400– 3. Results 500 μg/μl) were then added to the beads and rocked overnight at 4 °C.
The next day, beads were pulled down by centrifugation for 2 min at 3.1. Molecular crosstalk between OTR and β2AR in hTERTC3 cells: role 1600 rpm and washed twice with solubilization buffer and once with of PKA TBS 1X (rocking for 5 min, centrifugation 2 min at 1600 rpm). Then the appropriate elution peptide (Flag or HA peptide at 50 ng/μlin In a companion paper, we showed results identifying functional in-
TBS1X) was used to elute bound precipitates from beads for 30 min at teractions between the OTR and the β2AR in immortalized human myo- 4 °C. Beads were separated from elutes by centrifugation for 30 s at metrial cells. These studies suggested the possibility of physical 8200×g. Eluants were subjected to western blot analysis. interactions between the two receptors, although the molecular nature of these interactions was not fully elucidated. If indeed the two recep- 2.6. Bioluminescence resonance energy transfer (BRET) tors interacted as part of a hetero-oligomeric complex, then it should be possible to detect allosteric interactions between the two distinct li-
To detect and analyze interactions between β2AR and OTR, BRET gand binding sites which might translate into altered signalling with re- was used as previously described [20,24]. For this purpose, HEK 293 spect to a shared effector pathway, ERK1/2. To investigate possible cells were seeded in 6-well plates and transfected with a fixed allosteric interactions, we examined the effects of a β2AR agonist, iso- amount of an OTR construct tagged at its C terminus with Renilla lucif- proterenol, on the ability of the OTR to activate ERK1/2 in our myome- erase (OTR-RLuc), and co-transfected with an increasing amount of trial cell model. We found that OTR-mediated ERK1/2 activation plasmids encoding β2AR or GABA-B2 (the latter as a negative con- was strongly attenuated following pretreatment with isoproterenol trol), both tagged at their C-terminus with YFP (β2AR-YFP, GABAB2- (Fig. 1). We next wondered if the effects of isoproterenol on OTR- YFP). Cells were collected 48 h post-transfection. After the addition mediated signalling could be attributed to second messenger-based, of the substrate coelenterazine h, emission was measured using an crosstalk between the two receptors. PKA has already been implicated injector-equipped plate-reader spectrofluorometer (Fluostar Optima, in inhibition of phospholipase C (PLC) downstream of Gαq-mediated BMG LabTechnologies) at the wavelengths of 475 nm and 535 nm, signalling in myometrial cells [26]. Thus we assessed the putative role corresponding to the maxima of the emission spectra for RLuc and of PKA in crosstalk between OTR and β2AR. We noted that inhibition YFP, respectively. The BRET ratio was calculated as described [10]: of PKA with PKI 6-22A leads to a slight increase in OTR-mediated BRET ratio=[(emission at 510–590 nm)−(emission at 440– ERK1/2 activation; however, such treatment was not able to fully re- 500 nm)×Cf]/(emission at 440–500 nm), where Cf corresponded to verse the attenuation of OTR-mediated ERK1/2 activation following (emission at 510–590 nm)/(emission at 440–500 nm) for cells pretreatment with isoproterenol (Fig. 1). Further, direct activation of expressing OTR-RLuc alone. BRET ratios were plotted as means of 10 adenylyl cyclase (AC) with forskolin, had an effect that partially mim- or 40 consecutive measurements taken at 0.4- to 0.5-s intervals. icked the effect of isoproterenol on OTR-mediated ERK1/2 activation (data not shown). Taken together, these data suggest that canonical 2.7. Protein-fragment complementation assays (PCA) crosstalk between signalling pathways could only partially account for
the effects of β2AR on OTR-mediated signalling. We also investigated interactions between β2AR and OTR with the use of a protein-fragment complementation assay (PCA), initially de- 3.2. Modulation of signalling by different classes of OTR and β2AR ligands scribed in [20,25]. For the purposes of our experiments, HEK 293 cells in myometrial hTERTC3 cells were seeded in 6-well plates and transfected with equal amounts of receptors of interest tagged with split Venus constructs in the follow- We wanted to further explore the effects of different classes of ing combinations (OTR-Venus1 and OTR-Venus2, β2AR-Venus1 and βAR ligands on OTR-mediated ERK1/2 activation in hTERT-C3 cells P.K. Wrzal et al. / Cellular Signalling 24 (2012) 342–350 345
Fig. 1. β2AR agonist isoproterenol (ISO) inhibits OTR-mediated ERK1/2 signalling in human myometrial cells. (A) hTERT-3 cells were grown in 0.05% serum for 24 h then were pre- treated with ISO (10 μM) or vehicle (control) for 15 min and subsequently exposed to OT (100 nM) for different times at 37 °C. (B) Quantitative evaluation of 3 independent ex- periments as shown in (A). (C) Cells were pretreated for 1 h with PKA inhibitor PKI 6-22A (5 μM) or 15 min with ISO (10 μM) or both or vehicle water (control) and subsequently treated with OT (100 nM) for different times at 37 °C. (D) Quantitative evaluation of 3 independent experiments as shown in (C). Cells were lysed and levels of phos- phorylated ERK1/2 (pERK1/2) and total ERK1/2 were assessed by western blotting. Values for phosphorylated ERK1/2 were normalized with respect to total ERK1/2 levels and plot- ted as means +/− SEM as a percentage of maximal response (5 min). *, Pb0.05 vs. corresponding control value for each time point.
and began by investigating the effects of the non-specific βAR following pretreatment with timolol (Fig. 3). Interestingly, although antagonist propranolol. Surprisingly, ERK1/2 activation induced by there were alterations in the basal levels of ERK1/2 activation, timolol oxytocin (OT) was again strongly attenuated following pretreatment also blocked β2AR-mediated activation of ERK1/2 (Fig. 3a, right panel, with propranolol (Fig. 2). Interestingly, we noted that propranolol Fig. 3b). We next investigated the effects of a third ligand, in this case pretreatment did not attenuate the subsequent, yet more modest the β2AR inverse agonist, atenolol, on OTR-mediated ERK1/2 activation. (compared with OT) isoproterenol-mediated ERK1/2 activation. It Here, we found that pretreatment of cells with atenolol resulted in no has been previously shown that although propranolol is an antagonist significant decrease in OTR-mediated ERK1/2 activation; in fact there for β2AR-mediated activation of AC, it is a partial agonist for β2AR- was a trend toward an increased signal (Fig. 4). Curiously, atenolol mediated ERK1/2 activation [27,28]. To exclude this potential con- showed the same stimulatory trend for β2AR-stimulated ERK1/2 activa- found due to the biased nature of propranolol effects on the β2AR, tion, suggesting that it too acts as a biased agonist for this signalling we next used timolol, which has been characterized as an antagonist pathway. These experiments demonstrated that different β2AR ligands, for both AC and ERK1/2 downstream of β2AR activation [29]. Similar independent of their ability to modulate cAMP formation (and subse- to propranolol, OTR-mediated ERK1/2 activation was attenuated quently activate PKA) were capable of modulating signalling via the
Fig. 2. β2AR antagonist propranolol (PRO) inhibits OTR-mediated ERK1/2 signalling. hTERT-C3 cells were pretreated with PRO (10 μM) or vehicle (control) for 15 min and subse- quently exposed to either 100 nM OT (A,B) or 1 μM of ISO (C,D) for different times at 37 °C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, normalized to total ERK1/2 and expressed as percentage of maximal response as in Fig. 1 (n=3). Representative blots are shown in the top panels (A,C). *, Pb0.05 vs. corresponding control value for each time point. **, Pb0.01 vs. corresponding control value for each time point. 346 P.K. Wrzal et al. / Cellular Signalling 24 (2012) 342–350
Fig. 3. β2AR antagonist timolol (TIM) inhibits OTR-mediated ERK1/2 signalling. hTERT-C3 cells were pretreated with TIM (10 μM) or vehicle (control) for 15 min and subsequently exposed to either 100 nM OT (A,B) or 1 μM of ISO (C,D) for different times at 37 °C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, normalized to total ERK1/2 and expressed as percentage of maximal response (n=3). Representative blots are shown in the top panels (A,C). *, Pb0.05 vs. corresponding control value for each time point.
OTR, possibly through allosteric effects mediated through a β2AR/OTR 3.3. Physical interactions between β2AR and OTR in HEK 293 cells hetero-oligomer.
To confirm and extend these experiments, we next investigated the In order to assess such potential physical interactions between β2AR effects of different OTR ligands on β2AR-mediated ERK1/2 activation. and OTR, we used several experimental strategies. First, we determined We did not use oxytocin here as it swamped out any effects of β2AR whether these two receptors could be co-immunoprecipitated when stimulation. However, we noted that β2AR-mediated ERK1/2 activation co-expressed in HEK 293 cells. We noted that HA-OTR and Flag-β2AR was attenuated following pretreatment with the OTR antagonist atosi- co-immunoprecipitated only when these receptors were expressed in ban (Fig. 5a,c). A similar result was observed with the OTR antagonist the same cells but not when membranes were mixed from cells expres- OTA (Fig. 6a,c). As expected, both OTR antagonists also blocked OT- sing each receptor alone (Fig. 7). Several features of the putative mediated ERK1/2 activation (Figs. 5b,d and 6b,d). The results with the β2AR/OTR complex are evident from these experiments. First, both ma- different classes of receptor ligands cannot all be mediated through sec- ture and immature forms of the OTR were co-immunoprecipitated with ond messenger-mediated crosstalk between the two receptors. These the β2AR. This suggests that the interactions (as for most GPCR oligo- results again strongly indicate that allosteric interactions occur between mers) occur in the ER during receptor biosynthesis. This is also indicat- endogenous OTR and β2AR modulating their signalling capabilities, like- ed by the fact that mixtures of solubilized receptors from cells ly through the agency of a receptor hetero-oligomer. expressing each receptor alone, do not interact. Also, both monomeric
Fig. 4. Effect of β2AR inverse agonist atenolol (AT) on OTR-mediated ERK1/2 signalling. hTERT-C3 cells were pretreated with AT (10 μM) or vehicle (control) for 15 min and sub- sequently exposed to 100 nM OT (A,B) or 1 μM of ISO (C,D) for different times at 37 °C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, normalized to total ERK1/2 and expressed as percentage of maximal response (n=3). Representative blots are shown in the top panels (A,C). *, Pb0.05 vs. corresponding control value for each time point. P.K. Wrzal et al. / Cellular Signalling 24 (2012) 342–350 347
Fig. 5. OTR antagonist atosiban inhibits β2AR-mediated ERK1/2 signalling. hTERT-C3 cells were pretreated with atosiban (100 nM) or vehicle (control) for 15 min and subsequently exposed to 10 μM of ISO (A,B) or 1 nM OT (C,D) for different times at 37 °C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, normalized to total ERK1/2 and expressed as percentage of maximal response (n=3). Representative blots are shown in the top panels (A,C). *, Pb0.05 vs. corresponding control value for each time point. **, Pb0.01 vs. corresponding control value for each time point. and dimeric versions of HA-OTR were co-immunoprecipitated suggest- detected when either OTR-Venus1 and OTR-Venus2 (homodimer) or ing that these arrays may be multimeric in nature. β2AR-Venus1 and β2AR-Venus2 (homodimer) were co-expressed We next used bioluminescence resonance energy transfer (BRET) either using confocal microscopy or in direct measurements of reconsti- to confirm the results obtained by co-immunoprecipitation of β2AR tuted fluorescence in a microplate reader (Fig. 8b,c). Venus fluorescence and OTR in HEK 293 cells. BRET was measured from cells co- was also detected from HEK 293 cells co-transfected with OTR-Venus1 transfected with OTR-Rluc and β2AR-YFP. The BRET signal was satura- and β2AR-Venus2, whereas Alk5-Venus2 and OTR-Venus did not recon- ble in the presence of increasing amounts of acceptor and a fixed level stitute Venus when co-expressed (Fig. 8b,c). Taken together, our results of donor (Fig. 8a). This was in contrast to GABAB2-YFP, which complement those obtained in the companion study showing a func- expressed at similar levels (as measured by measuring YFP expres- tional interaction between the two receptors by further demonstrating sion, data not shown) but did not saturate the OTR-Rluc donor, sug- direct molecular association likely involving allosteric interactions gesting that the OTR/β2AR interaction was specific(Fig. 8a). Thus, between the two receptors when regulating a common effector BRET experiments also demonstrated that the two receptors were pathway. part of the same complex but could not confirm whether or not the interactions were direct. 4. Discussion Finally, a protein complementation assay (PCA), based on the recon- stitution of Venus GFP, was used to confirm the results obtained using In the present study, we demonstrated that the ligands of OTR
BRET and to demonstrate that interactions between the receptors and β2AR influence the ability of either partner receptor to acti- were direct. Reconstitution of a correctly folded Venus moiety was vate ERK1/2 in human myometrial cells. We suggest that allosteric
Fig. 6. OTR antagonist OTA inhibits β2AR-mediated ERK1/2 signalling. hTERT-C3 cells were pretreated with either OTA (100 nM) or vehicle (control) for 15 min and subsequently exposed to 10 μM of ISO (A,B) or 1 nM OT (C,D) for different times at 37 °C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, normalized to total ERK1/2 and expressed as percentage of maximal response (n=3). Representative blots are shown in the top panels (A,C). *, Pb0.05 vs. corresponding control value for each time point. 348 P.K. Wrzal et al. / Cellular Signalling 24 (2012) 342–350
Fig. 7. β2AR and OTR co-immunoprecipitation in HEK 293 cells. Western blot analysis of immunoprecipitates of total lysates of HEK293 cells transiently transfected HA- and Flag- tagged receptors. Cells were transfected with either pcDNA vector (lanes 1,6 and 11), HA-OTR (lanes 2, 7 and 12) or Flag-β2AR (lane 3, 8 and 13) alone; or with both receptors in combination (lanes 4, 9 and 14). As a control co-immunoprecipitation was assessed in a mixture of two cell populations transfected separately with either HA-OTR or Flag-β2AR (lanes 5, 10 and 15). Immunoprecipitation (IP) was performed using either an anti-Flag antibody, either immunoblotted (IB) with anti-HA (lanes 1–4) or anti-Flag (lanes 11–
15). Total lysates are shown in lanes 6–10. Arrows indicate bands corresponding in molecular mass to the different forms of HA-OTR (left side) and β2AR (right side). Molecular mass markers (on the left) are in kDa. Images are representative of three independent experiments.
interactions exist between the endogenous OTR and β2AR, where block PMA-induced ERK1/2 activation [27],however,inourstud- one receptor modulates the signalling ability of the other. The al- ies it did not attenuate isoproterenol-mediated ERK1/2 activation. losteric nature of these interactions, under native conditions with This may be attributed to the cell type-specific differences which endogenous receptors, led us to suppose that the two receptors may have clear implications for future studies of allosterism and interacted in the context of a heterooligomer. The ability of a biased agonism. As we have evidence to suggest a heterodimer
β2AR agonist to reduce the OTR-mediated ERK1/2 activation was formation between OTR and β2AR, the effects of the β2AR ligands not, in and of itself, a novel finding. It has previously been on β2AR-mediated ERK1/2 activation we have seen here may be shown that activation of cAMP-dependent protein kinase (PKA) due to the unique pharmacology of this receptor when associated can phosphorylate and inhibit phospholipase C, thus attenuating with OTR in the myometrium. the signalling from Gαq-coupled GPCRs [30,31].Consistentwith If we consider the existence of physical interactions between previous observations, we found that activation of AC via forskolin OTR and β2AR then perhaps the effects of the different OTR and partially attenuated OTR-mediated ERK1/2 activation. However, β2AR ligands can be attributed to the stabilization of distinct con- using a specific inhibitor of PKA, we showed that this effect was formations of one partner receptor by the ligand of the other. It not fully PKA-dependent. Two possible explanations exist for has been shown that GPCRs can adopt several different types of these observations: (1) an interaction between the two receptors, active states or conformations, with different ligands able to stabi- which cannot be explained by crosstalk at the level of second lize distinct states leading to the activation of unique signalling messengers, and/or (2) crosstalk that is cAMP-dependent, but pathways via biased agonism [32,33].Ourresultsdemonstrate PKA-independent. Based on these results, and our study showing that different classes of ligands for a given receptor can affect functional interactions between OTR and β2AR (see the associated partner receptors in different ways. A recent study has shown companion article), we investigated the effects of β2AR antago- that the two-receptor equivalents in the context of a D2 dopamine nists on the ability of OTR to activate ERK1/2. Our findings receptor homodimer are organized asymmetrically with respect to showed that both propranolol and timolol also attenuated OTR- their G protein partners such that occupation by ligand of one re- mediated ERK1/2 activation, but in a manner that clearly was in- ceptor activates the receptor and occupation of the other and thus dependent of the generation of second messengers. Further, we modulates signalling allosterically [34]. In the context of a homodi- showed that a β2AR inverse agonist, atenolol, had no effect on mer this may not be as important as either receptor can serve each the ability of OTR to activate ERK1/2. Further evidence for alloste- role and such asymmetries might not be detected (see [35] for a discus- ric effects comes from the fact that two distinct OTR antagonists, sion of this issue). However, for receptor hetero-oligomers, such asym- which do not activate second messenger-dependent pathways, metries may have important consequences. Using the example, of a inhibited β2AR-mediated ERK1/2 signalling as well. This is partic- heterodimer between the β2AR and the δ-opioid receptor, or between ularly important as we demonstrated in the companion article the β2AR and OTR described here, an asymmetry with respect to how that β2AR signals through a novel pathway which strictly depends the complex is arranged may mean that in one case, depending on on the presence of the OTR. We also noted that the effects of the how the complex is formed, we might have a β2AR modulated by δ- β2AR ligands used in this study on the signalling of the β2AR itself opioid or OTR ligands and in another case a δ-opioidreceptororan in human myometrial cells were different from what has been OTR modulated by β2AR ligands [11,12]. Thus, in one arrangement, reported in HEK 293 cells [27,29]. We found that timolol did not the first receptor is the signalling receptor and the second becomes strongly block isoproterenol-mediated ERK1/2 activation [29]. an allosteric modulator and the converse is true when the system is or- Similarly, atenolol has been previously shown to be an inverse ag- ganized the other way around. This greatly increases the potential orga- onist for the ERK1/2 signalling pathway as it has been shown to nizational complexity of GPCR signalling and suggests that cell-specific P.K. Wrzal et al. / Cellular Signalling 24 (2012) 342–350 349
receptor mosaics [37]. For example, a number of recent studies have suggested that GPCRs can form higher order complexes. Protein com- plementation approaches have now been used to confirm and extend our knowledge regarding dimerization and oligomerization of GPCRs. Reconstitution of split luciferase (Gaussia or Renilla) and split GFP con-
structs has shown that dimers of β2AR and D2 dopamine receptors can be detected, complementing immunopurification and BRET ap- proaches, and these approaches can be combined to detect and examine larger complexes [20,38]. A number of investigators have used three partner PCA/RET to show that higher order complexes of GPCRs such
as the A2A-adenosine receptor homo- and hetero-oligomers with CB1 cannabinoid/D2 dopamine receptors and CXCR4 multimers can be detected [39–43]. These larger arrays may in fact represent complicated allosteric machines which can be assembled by different cells in differ- ent ways, greatly expanding the combinatorial power of GPCR signal- ling. Our data presented here supports the notion that these complexes are formed during receptor biosynthesis. The results presented in this study, taken together with the compan- ion article, show that the formation of heterodimers between OTR and
β2AR has functional consequences on the ability of each receptor to ac- tivate ERK1/2 in human myometrial cells. The fact that the signalling phenotype described in hTERT-C3 cells, where the two receptors are expressed endogenously, was recapitulated in HEK 293 cells expressing the receptors suggests that they interact in the same way, likely in the context of receptor heteroligomers. Though beyond the scope of the present article, it would certainly be of interest to demonstrate these in- teractions directly in hTERT-C3 cells using labeled ligands or antibodies
by FRET. This suggests that the OTR/β2AR hetero-oligomers have dis- tinct signalling profiles and pharmacology which differ from either
OTR or β2AR alone. In addition to providing novel lines of evidence for direct functional and allosteric interactions between two members of the GPCR family, these findings are of specific potential relevance for the development of novel tocolytic approaches involving these two re- ceptor systems.
Acknowledgments
We thank Irina Glazkova for the Flag-β2AR construct and Carlis Rejon for the Alk5-Venus2 construct. This work was supported by grants from the Canadian Institutes of Health Research (CIHR) to TEH (MOP-36279), to HHZ (MOP-74675) and a CIHR Research Team Grant in GPCR Allosteric Regulation (CTiGAR, CTP-79848). TEH is a Chercheur National of the Fonds de la Recherche en Santé du Québec (FRSQ).
References Fig. 8. β2AR and OTR interactions shown using BRET and PCA in HEK 293 cells. (A) For BRET experiments, cells were transfected with a fixed amount of OTR-Rluc and co- [1] M.C. Lagerstrom, H.B. Schioth, Nature Reviews. Drug Discovery 7 (2008) 339–357. transfected with an increasing amount of β AR-YFP (■). As a negative control cells 2 [2] J.P. Overington, B. Al-Lazikani, A.L. Hopkins, Nature Reviews. Drug Discovery 5 fi were transfected with xed amount of OTR-Rluc and increasing amounts of GABAB2- (2006) 993–996. ▲ − YFP ( ). Each point represents the mean +/ SEM, n=3. (B,C) For PCA experiments, [3] K. Palczewski, Trends in Biochemical Sciences 35 (2010) 595–600. β cells were transfected with the following pairs: OTR-Venus1 and 2AR-Venus2, OTR- [4] G. Milligan, N.J. Smith, Trends in Pharmacological Sciences 28 (2007) β β Venus1 and OTR-Venus2, 2AR-Venus1 and 2AR-Venus2. As a negative control cells 615–620. were transfected with OTR-Venus1 and Alk5-Venus2. Reconstitution of Venus was [5] H.N. Simhan, S.N. Caritis, The New England Journal of Medicine 357 (2007) assessed by either confocal microscopy (B) and by direct measurement of Venus fluo- 477–487. rescence (C). Data are presented as mean +/− SEM (n=3). *, Pb0.05 vs. control. [6] L. Albizu, M. Cottet, M. Kralikova, S. Stoev, R. Seyer, I. Brabet, T. Roux, H. Bazin, E. Bourrier, L. Lamarque, C. Breton, M.-L. Rives, A. Newman, J. Javitch, E. Trinquet, M. Manning, J.-P. Pin, B. Mouillac, T. Durroux, Nature Chemical Biology 6 (2010) 587–594. determinants of signalling complex assembly will be of paramount im- [7] S. Terrillon, T. Durroux, B. Mouillac, A. Breit, M.A. Ayoub, M. Taulan, R. Jockers, C. – portance in initially defining signalling specificity in a given tissue, Barberis, M. Bouvier, Molecular Endocrinology 17 (2003) 677 691. [8] D. Devost, H.H. Zingg, Journal of Neuroendocrinology 16 (2004) 372–377. cellular or subcellular compartment [4,36]. This has important implica- [9] T.E. Hébert, S. Moffett, J.-P. Morello, T.P. Loisel, D.G. Bichet, C. Barret, M. Bouvier, tions for the formation of receptor heterodimers and hetero-oligomers, Journal of Biological Chemistry 271 (1996) 16384–16392. in that multiple asymmetrical arrangements are possible depending on [10] S. Angers, A. Salahpour, E. Joly, S. Hilairet, D. Chelsky, M. Dennis, M. Bouvier, Proceedings of the National Academy of Sciences of the United States of America the relative orientation of each monomer to the G protein and possibly 97 (2000) 3684–3689. to the effectors. More diversity is added when we consider heterotetra- [11] M. McVey, D. Ramsay, E. Kellett, S. Rees, S. Wilson, A.J. Pope, G. Milligan, Journal of mers which might have variable numbers of each component subunit or Biological Chemistry 276 (2001) 14092–14099. [12] B.A. Jordan, N. Trapaidze, I. Gomes, R. Nivarthi, L.A. Devi, Proceedings of the different potential arrangements of those subunits. Receptor complexes National Academy of Sciences of the United States of America 98 (2001) can also contain multiple receptors, what some authors have termed as 343–348. 350 P.K. Wrzal et al. / Cellular Signalling 24 (2012) 342–350
[13] D. Ramsay, E. Kellett, M. McVey, S. Rees, G. Milligan, Biochemical Journal 365 (2002) [27] S. Galandrin, M. Bouvier, Molecular Pharmacology 70 (2006) 1575–1584. 429–440. [28] M. Azzi, G. Piñeyro, S. Pontier, S. Parent, H. Ansanay, M. Bouvier, Molecular Phar- [14] C. Lavoie, J.-F. Mercier, A. Salahpour, D. Umapathy, A. Breit, L.-R. Villeneuve, W.-Z. macology 60 (2001) 999–1007. Zhu, R.-P. Xiao, E.G. Lakatta, M. Bouvier, T.E. Hébert, Journal of Biological Chemistry [29] M. Azzi, P.G. Charest, S. Angers, G. Rousseau, T. Kohout, M. Bouvier, G. Piñeyro, 277 (2002) 35402–35410. Proceedings of the National Academy of Sciences of the United States of America [15] A. Breit, M. Lagacé, M. Bouvier, Journal of Biological Chemistry 279 (2004) 100 (2003) 11406–11411. 28756–28765. [30] Y. Wen, K. Anwer, S.P. Singh, B.M. Sanborn, Endocrinology 131 (1992) [16] D.W. McGraw, K.A. Mihlbachler, M.R. Schwarb, F.F. Rahman, K.M. Small, K.F. 1377–1382. Almoosa, S.B. Liggett, The Journal of Clinical Investigation 116 (2006) [31] M. Liu, M.I. Simon, Nature 382 (1996) 83–87. 1400–1409. [32] T. Kenakin, Trends in Pharmacological Sciences 18 (1997) 416–417. [17] C. Hague, M.A. Uberti, Z. Chen, C.F. Bush, S.V. Jones, K.J. Ressler, R.A. Hall, K.P. [33] P. Strange, Biochemical Pharmacology 58 (1999) 1081–1088. Minneman, Proceedings of the National Academy of Sciences of the United [34] Y. Han, I.S. Moreira, E. Urizar, H. Weinstein, J.A. Javitch, Nature Chemical Biology 5 States of America 101 (2004) 13672–13676. (2009) 688–695. [18] D. Devost, H. Zingg, Journal of Molecular Endocrinology 31 (2003) 461–471. [35] P. Zylbergold, T.E. Hébert, Nature Chemical Biology 5 (2009) 608–609. [19] S.V. Naga Prasad, S.A. Laporte, D. Chamberlain, M.G. Caron, L. Barak, H.A. Rockman, [36] G. Milligan, British Journal of Pharmacology 158 (2009) 5–14. The Journal of Cell Biology 158 (2002) 563–575. [37] L.F. Agnati, D. Guidolin, J.P. Vilardaga, F. Ciruela, K. Fuxe, Journal of Receptor and [20] R.V. Rebois, M. Robitaille, D. Pétrin, P. Zylbergold, P. Trieu, T.E. Hébert, Methods 45 Signal Transduction Research 30 (2010) 287–303. (2008) 214–218. [38] W. Guo, EMBO Journal 27 (2008) 2293–2304. [21] J. Condon, S. Yin, B. Mayhew, R.A. Word, W.E. Wright, J.W. Shay, W.E. Rainey, Bi- [39] P.-A. Vidi, B.R. Chemel, C.-D. Hu, V.J. Watts, Molecular Pharmacology 74 (2008) ology of Reproduction 67 (2002) 506–514. 544–551. [22] D. Devost, H.H. Zingg, American Journal of Physiology 292 (2007) E1–E6. [40] P.-A. Vidi, J. Chen, J.M.K. Irudayaraj, V.J. Watts, FEBS Letters 582 (2008) 3985–3990. [23] A.M. Daulat, P. Maurice, C. Froment, J.-L. Guillaume, C. Broussard, B. Monsarrat, P. [41] G. Navarro, P. Carriba, J. Gandia, F. Ciruela, V. Casado, A. Cortes, J. Mallol, E.I. Delagrange, R. Jockers, Molecular & Cellular Proteomics 6 (2007) 835–844. Canela, C. Lluís, R. Franco, TheScientificWorldJournal 8 (2008) 1088–1097. [24] R.V. Rebois, M. Robitaille, C. Galés, D.J. Dupré, A. Baragli, P. Trieu, N. Ethier, M. [42] J. Gandia, J. Galino, O.B. Amaral, A. Soriano, C. Lluís, R. Franco, F. Ciruela, FEBS Bouvier, T.E. Hébert, Journal of Cell Science 119 (2006) 2807–2818. Letters 582 (2008) 2979–2984. [25] I. Remy, S. Michnick, Biotechniques 42 (2007) 137–145. [43] M. Hamatake, T. Aoki, Y. Futahashi, E. Urano, N. Yamamoto, J. Komano, Cancer Science [26] K.L. Dodge, B.M. Sanborn, Endocrinology 139 (1998) 2265–2271. 100 (2009) 95–102. Cellular Signalling 24 (2012) 333–341
Contents lists available at SciVerse ScienceDirect
Cellular Signalling
journal homepage: www.elsevier.com/locate/cellsig
Functional interactions between the oxytocin receptor and the β2-adrenergic receptor: Implications for ERK1/2 activation in human myometrial cells
Paulina K. Wrzal a, Eugénie Goupil a, Stéphane A. Laporte a,b, Terence E. Hébert a,⁎, Hans H. Zingg a,b,c,⁎⁎ a Department of Pharmacology and Therapeutics, McGill University, Montréal, Québec, Canada H3G 1Y6 b Department of Medicine, McGill University, Montréal, Québec, Canada H3G 1Y6 c Department of Obstetrics and Gynecology, McGill University, Montréal, Québec, Canada H3G 1Y6 article info abstract
Article history: The Gq-coupled oxytocin receptor (OTR) and the Gs-coupled β2-adrenergic receptor (β2AR) are both Received 9 August 2011 expressed in myometrial cells and mediate uterine contraction and relaxation, respectively. The two receptors Received in revised form 9 September 2011 represent important pharmacological targets as OTR antagonists and β2AR agonists are used to control pre- Accepted 11 September 2011 term uterine contractions. Despite their physiologically antagonistic effects, both receptors activate the MAP Available online 22 September 2011 kinases ERK1/2, which has been implicated in uterine contraction and the onset of labor. To determine the sig- nalling pathways involved in mediating the ERK1/2 response, we assessed the effect of blockers of specificG Keywords: α Oxytocin receptor protein-associated pathways. In human myometrial hTERT-C3 cells, inhibition of G i as well as inhibition of α β β2-adrenergic receptor the G q/PKC pathway led to a reduction of both OTR- and 2AR-mediated ERK1/2 activation. The involvement ERK1/2 of Gαq/PKC in β2AR-mediated ERK1/2 induction was unexpected. To test whether the emergence of this novel Myometrium signalling mechanism was dependent on OTR expression in the same cell, we conducted experiments in HEK Dimerization 293 cells that were transfected with the β2AR alone or co-transfected with the OTR. Using this approach, we found that β2AR-mediated ERK1/2 responses became sensitive to PKC inhibition only in cells co-transfected with the OTR. Inhibitor studies indicated the involvement of an atypical PKC isoform in this process. We con- firmed the specific involvement of PKCζ in this pathway by assessing PKCζ translocation to the cell membrane.
Consistent with our inhibitor studies, we found that β2AR-mediated PKCζ translocation was dependent on co- expression of OTR. The present demonstration of a novel β2AR-coupled signalling pathway that is dependent on OTR co-expression is suggestive of a molecular interaction between the two receptors. © 2011 Elsevier Inc. All rights reserved.
1. Introduction Both OTR and β2AR are co-expressed in myometrial cells, but relay opposing signals of contraction and relaxation in these cells, respective-
The oxytocin receptor (OTR) and the β2-adrenergic receptor (β2AR) ly. While the OTR, is functionally coupled mainly to Gαq/11 as well as are members of the large G protein-coupled receptor (GPCR) family. Gαiinmyometrialcells,theβ2AR is predominantly a Gαs-coupled receptor and mediates uterine relaxation via an increase in intracellular
β β cAMP levels [1]. The two receptors represent important pharmacologi- Abbreviations: AC, adenylyl cyclase; 2AR, 2-adrenergic receptor; cAMP, cyclic β adenosine monophosphate; CFP, cyanide fluorescent protein; CYP, cyanopindolol; cal targets, because OTR antagonists and 2AR agonists are used to con- DMEM, Dulbecco's modified Eagle's medium; DMSO, dimethylsulphoxide; DNA, trol pre-term uterine contractions [2]. Despite their physiologically deoxyribonucleic acid; EGF, epithelial growth factor; EGFR, epithelial growth factor antagonistic effects, both receptors activate ERK1/2. Several studies, receptor; ERK, extracellular signal-regulated kinase; ET-1R, endothelin-1 receptor; mostly performed in the rat myometrium, have identified the activation FBS, fetal bovine serum; GPCR, G protein-coupled receptor; HRP, horseradish peroxide; of ERK1/2 as a component of the cascade of events leading to the devel- hTERT-C3, myometrial human telomerase reverse transcriptase-clone3 cells; ISO, isopro- terenol; MAPK, mitogen activated protein kinase; OT, oxytocin; OTA, oxytocin receptor opment of labor [3,4]. The onset of labor has been reported to be associ- antagonist; OTR, oxytocin receptor; PBS, phosphate buffered saline; PCR, polymerase ated with basal activation of ERK1/2 [5]. chain reaction; PI3K, phosphoinositide 3-kinase; PKA, protein kinase A; PKC, protein The ERK1/2 signalling pathway constitutes one of the most ubiq- kinase C; PLC, phospholipase; PTX, Pertussis toxin; RNA, ribonucleic acid; YFP, yellow uitous signal transduction cascades. The mechanisms of ERK1/2 acti- fluorescent protein. ⁎ Correspondence to: T.E. Hébert, Department of Pharmacology and Therapeutics, vation by GPCRs often depend on cell type and receptor involved, i.e. McGill University, 3655 Promenade Sir-William-Osler, Room 1303, Montréal, Québec, molecular context is an all-important determinant of the nature and Canada H3G 1Y6. Tel.: ++1 514 398 1398; fax: +1 514 398 6690. duration of a particular ERK1/2 signal. Although both OTR and β2AR ⁎⁎ Correspondence to: H.H. Zingg, Department of Pharmacology and Therapeutics, activate ERK1/2, the mechanisms by which they do so are different. McGill University, 3655 Promenade Sir-William-Osler, Room 1325, Montréal, Québec, The OTR has been shown to couple to both GαqandGαiinmyome- Canada H3G 1Y6. Tel.: +1 514 398 3621; fax: +1 514 398 2045. α E-mail addresses: [email protected] (T.E. Hébert), [email protected] trial cells. OTR activation of G q has been shown to be important for (H.H. Zingg). OT-stimulated phospholipase C (PLC) activation and elevation of
0898-6568/$ – see front matter © 2011 Elsevier Inc. All rights reserved. doi:10.1016/j.cellsig.2011.09.019 334 P.K. Wrzal et al. / Cellular Signalling 24 (2012) 333–341 intracellular calcium in the myometrium and in the activation of previously [25]. PKC 1-GFP and PKCζ-GFP constructs were a kind gift ERK1/2 [6–8]. However, the OTR was also shown to activate Gαiin from Dr S. Ferguson (Robarts Research Institute, London, ON). To gener- several cell types and OTR-mediated ERK1/2 phosphorylation has ate mCherry-tagged PKCs, the constructs mentioned above and a been reported to involve both PKC and Gαi-Gβγ pathways as well pcDNA3.1(+)-mCherry vector were digested with NheIandBsrGI to re- as transactivation of the epidermal growth factor receptor (EGFR) place the GFP with mCherry. To generate the pcDNA3.1(+)-mCherry in PMH1 myometrial cells [8–11].Ithasalsobeenshownthatthe vector, pRSET-B-mCherry, a kind gift from Dr. Roger Y. Tsien (University OTR interacts with the scaffolding protein β-arrestin2 [12] and that of California, San Diego, CA), was amplified by PCR to introduce a β-arrestin2 is involved in OT-mediated ERK1/2 activation [13,14]. NheI site in 5′ and KpnI site in 3′ of the mCherry and cloned into Activation of ERK1/2 has been shown to be important in oxytocin- pcDNA3.1(+). shRNA constructs targetting β-arrestin1/2 were gener- mediated myometrial contractions [15]. ous gifts from Dr. Marc Caron (Duke University, Durham, NC). All con-
To date, most studies investigating β2AR-mediated ERK1/2 activation structs were verified by bidirectional sequencing. have been performed in HEK 293 or COS-7 cells with transfected recep- tors. Some studies have suggested that β2AR-mediated ERK1/2 2.3. Cells culture and transfection activation involves Gβγ subunits, rather than the Gαs-cAMP-PKA path- way, which then act on a Ras-dependent pathway leading to activation The myometrial human telomerase reverse transcriptase (hTERT)- of ERK1/2 [16]. Other studies show that β2AR activates ERK1/2 via HM cells were obtained from Dr. W. E. Rainey [26]. hTERT-C3 cells rep- Gαs-PKA-dependent pathway [17]. Another G protein coupling mecha- resent a selected subclone that we obtained by serial clonal dilution of nism relevant for ERK1/2 activation is the potential ability of β2AR to un- hTERT-HM cells described previously [27]. hTERT-C3 cells were main- dergo PKA-dependent phosphorylation in response to agonist, which tained in DMEM/F12 medium supplemented with 10% FBS and cultured leads to a loss of Gαs signalling and a switch to activation of Gαi [18]. at 37 °C in 5% CO2. Cells close to confluency were passaged by trypsini- Further, β2AR-mediated ERK1/2 activation has also been reported to in- zation and plated in T175 flask at a one-quarter dilution every 4–5days. volvebothGproteinandβ-arrestin components [19]. Other reports sug- HEK 293 cells were from Invitrogen (Carlsbad, CA) and were grown in gest that Src activation plays a role in β2AR stimulation of ERK1/2 [20,21]. Dulbecco's Modified Eagle's Medium/high glucose (DMEM) supple- Previous studies have shown that β2AR signalling could affect mented with 5% FBS and transfected using Lipofectamine 2000 as per OTR signalling in the myometrium [22,23]. However, the mecha- manufacturer's instructions. Experiments were carried out 48 h after nisms underlying crosstalk between the two receptors have not transfection. been investigated in detail. Unraveling the nature and consequences of OTR/β2AR interactions and distinguishing direct physical interac- 2.4. Western blots tions from indirect second messenger-mediated interactions, would be of significant pharmacological and physiological importance. The Immunoblotting was used to assess the expression of cellular pro- purpose of the present study was to define the signalling mecha- teins. Briefly, cells were plated onto culture dishes and grown in appro- nisms involved in activating ERK1/2 by both the OTR and β2AR in priate media to near confluency. Cells were starved for 24 h in either myometrial cells. Our studies demonstrate the involvement of sever- DMEM/F12 supplemented with only 0.5% FBS (hTERT-C3 cells) or al G proteins in OTR- and β2AR-mediated ERK1/2 activation. We also DMEM/high glucose supplemented with only 0.5% FBS (HEK 293 cells). identify a novel specific signaling pathway by which the β2AR acti- Cells were stimulated with ligand(s) for different times at 37 °C. Stimula- vates ERK1/2 in myometrial cells. This pathway involves Gαi-PI3K- tion was stopped by two ice-cold PBS washes, and plates were flash fro-
PKCζ and is dependent on co-expression of the OTR and the β2AR zen in liquid nitrogen. For inhibitor studies, cells were pre-treated with in the same cell. Taken together our results define a novel functional the inhibitor for 30 min to 1 h before stimulation with ligand(s). Cells interaction between the two receptors which may, in turn, be based were lysed on ice with Laemmli buffer (50 mM Tris–HCl [pH 6.8], 2% so- on a physical interaction. dium dodecyl sulfate, 10% glycerol, and 0.1 M β-mercaptoethanol) and were either homogenized using passage through a syringe and needle 2. Materials and methods (hTERT-C3 cells) or by sonication (HEK 293 cells). Lysates were clarified by centrifugation at 15,000×g for10minat4C ina microcentrifuge. 2.1. Reagents Proteins were denatured by boiling for 5 min, and subjected to SDS-PAGE and western blotting. Immunodetection involved differ- Reagents were obtained from the following sources: DMEM/F12 tis- ent primary antibodies in conjunction with a secondary horseradish sue culture medium was from Invitrogen (Burlington, ON); fetal bovine peroxidase-conjugated antibody and a chemiluminescence detection serum (FBS) from Hyclone (Logan, UT); OT was from Sigma-Aldrich (St. system (Supersignal; Pierce). Quantification of band intensities was Louis, MO); isoproterenol was from Tocris Bioscience (Bristol, UK); the performed using AlphaEase (Alpha Innotech Corp., San Leandro, CA). inhibitors AG1478, wortmannin, PP2, PKCζ pseudosubstrate inhibitor, Gö6976 and Gö6983 were from Calbiochem (La Jolla, CA); Pertussis 2.5. Receptor quantification toxin (PTX) and Rö31-8220 and the protease inhibitors: leupeptin, ben- zamidine and trypsin were from Sigma-Aldrich (St. Louis, MO). Primary Total receptor number for the β2AR was calculated from binding ex- anti-phospho-ERK1/2 (T202, Y204) antibody was from Cell Signaling periments using [125I]cyanopindolol (CYP) as the radioligand. Mem- Technology (Beverly, MA); pan anti-ERK1/2 antibody was from Stress- branes were prepared and washed as previously described, with all gen (Ann Arbor, MI). Secondary antibody, horseradish peroxidase- steps of the process performed on ice [25].Briefly, hTERT-C3 cells conjugated anti-rabbit IgG, was from Sigma-Aldrich (St. Louis, MO). were washed twice with ice-cold PBS. They were then incubated for Lipofectamine 2000 was from Invitrogen (Burlington, ON). All other an- 15 min with ice-cold lysis buffer containing 15 mM Tris–HCl (pH 7.4), alytical grade chemicals were obtained from Sigma-Aldrich, Fisher Sci- 0.3mMEDTA,2mMMgCl2,5μg/mL leupeptin, 10 μg/mL benzamidine entific (Waltham, MA), or VWR (West Chester, PA). β-arrestin1/2 and and 5 μg/mL trypsin inhibitor. Subsequently, cells were collected and control siRNA were obtained from Qiagen Inc. (Toronto, ON). homogenized with a polytron. Lysates were centrifuged at 14,500 rpm for 20 min at 4 °C; the pellet was resuspended in ice-cold membrane
2.2. Constructs buffer containing 50 mM Tris–HCl (pH 7.4), 3 mM MgCl2,5μg/mL leu- peptin, 10 μg/mL benzamidine and 5 μg/mL trypsin inhibitor. Next, the OTR-YFP, an OTR construct with the coding sequence for YFP added pellet was homogenized using a 10-mL Potter-Elvehjem hand homoge- in frame to the OTR C-terminus [24]. β2AR-HA, was used as described nizer and membranes were separated by centrifugation at 14,500 rpm P.K. Wrzal et al. / Cellular Signalling 24 (2012) 333–341 335 for 20 min at 4 °C. The final pellet was resuspended in membrane buff- membrane protein (n=3). We have previously assessed the level of er, protein amounts were determined and membranes were used for OTR expressed in human myometrial hTERT-C3 cells and determined radioligand binding assay in which membrane preparations (20 μgof it to be 705 fmol/mg of protein [27]. Thus we confirmed that the β2AR protein), in a total volume of 0.25 mL, were labelled with 230 pM is expressed in human myometrial hTERT-C3 cells but at levels 14- [125I]-CYP. This represents a near saturating concentration of CYP. Non- fold lower than the OTR. This difference in expression levels appears specific binding was determined using 10 μM alprenolol. to mirror the situation in the intact human myometrium where sim- ilar differences in relative expression levels of the two receptors have 2.6. PKC ζ-mCherry translocation been observed [28,29].
HEK 293 cells were maintained in DMEM supplemented with 3.2. Kinetics of OTR- and β2AR-mediated ERK1/2 activation in hTERT-C3 10% (v/v) heat-inactivated fetal bovine serum and gentamycin cells
(100 μg/mL). Cells expressing either β2AR-CFP alone or in combina- tion with OTR-YFP were co-transfected with either PKCζ-mCherry We first assessed the ability of the two receptors to stimulate the or PKCβ1-mCherry. 24 h post-transfection, cells were serum-starved ERK1/2 MAP kinase pathway in hTERT-C3 cells and determined the ki- for an additional 24 h in DMEM containing 0.2% FBS (v/v). Next, netics of ERK1/2 activation using agonists for each receptor. ERK1/2 cells were pretreated for 30 min with OTA (1 μM) or vehicle (water) activation patterns were distinct for OTR and β2AR (Fig. 1). The kinet- followed by treatment with either oxytocin (OT) (100 nM) or isopro- ics of OTR-mediated ERK1/2 activation were prolonged, with maxi- terenol (ISO) (10 μM) for 1 min. Cells were then fixed in 4% parafor- mum activation at 5 min but sustained activation lasting up to 3 h maldehyde for 5 min at room temperature, and washed with PBS. (Fig. 1A,B). In contrast, the kinetics of β2AR-induced ERK1/2 activation Cover slips were mounted on the slides and cells were visualized on were more transient, with a maximum activation at 5–10 min and a a Zeiss LSM-510 Meta laser scanning microscope using excitation at complete return to baseline within 30 min (Fig. 1C,D). Based on 458 nm for CFP, 515 nm for YFP and 543 nm for mCherry. Emission these initial findings, our subsequent investigations focused on time was measured with the following filter sets: BP475-525 nm for CFP, points between 2 min and 2 h. BP530-600 nm for YFP and LP560 nm for mCherry. 3.3. Identifying G proteins responsible for ERK1/2 activation in response
2.7. Data analysis to OTR or β2AR activation
All data are presented as means±SEM. When not shown, error In order to investigate the roles of the GαiandGαq pathways in bars lie within symbols. Statistical differences between paired groups activation of the ERK1/2 pathway via the two receptors, specificinhibi- were calculated using Student's t-test and comparisons between trea- tors were used. Inhibition of Gαi using PTX (100 ng/mL) predominantly ted cells and untreated control cells was made using one-way ANOVA affected the early time points (2 and 5 min) of OTR-mediated ERK1/2 with Dunnett adjustment. One-sample t-test was used to compare activation (Fig. 2A,B). The inhibition of β2AR-mediated ERK1/2 signal- treatment groups to the normalized values. Differences were consid- ling, upon Gαi inhibition, was essentially total, suggesting that this is ered statistically significant at Pb0.05. Data were represented graph- the primary pathway coupling the β2AR to ERK1/2 in these cells ically using GraphPad Prism version 4 (GraphPad Software, San (Fig. 2C,D). Diego, CA). To determine if known downstream effectors of Gαq were also
involved in β2AR-mediated activation of ERK1/2, the involvement of 3. Results PKC in the signalling cascade was assessed. Interestingly, upon treat- ment with the broad spectrum PKC inhibitor Gö6983 (1 μM), ERK1/2
3.1. Expression of β2AR in myometrial hTERT-C3 cells activation mediated via either OTR or β2AR was inhibited substantially (Fig. 3). Given the complete inhibition of β2AR-mediated ERK1/2 signal- The amount of β2AR in hTERT-C3 cells was assessed by measuring ling by PTX, this suggested that β2AR-mediated PKC activation may [125I]-CYP binding and was determined to be 50±10 fmol/mg of have been downstream of Gi, rather than Gq.
Fig. 1. Kinetics of ERK1/2 activation by OTR and β2AR in human myometrial hTERT-C3 cells. hTERT-C3 cells were grown in 0.05% serum for 24 h and treated with either 100 nM OT (A,B) or 10 μM ISO (C,D) for different times at 37 °C. Cells were lysed and levels of phosphorylated ERK1/2 (pERK1/2) were assessed by western blotting using a specific phospho- ERK1/2 antibody (pERK1/2) and reprobed with a total ERK1/2 antibody (ERK1/2). Representative blots are shown in the top panels (A, C). Blots from 3 independent experiments were quantitated using densitometry, values for phosphorylated ERK1/2 were normalized with respect to total ERK1/2 levels and plotted as means +/− SEM as a percentage of maximal response (5 min, B,D). *, Pb0.05 vs. control. **, Pb0.01 vs. control. 336 P.K. Wrzal et al. / Cellular Signalling 24 (2012) 333–341
Fig. 2. Inhibition of Gαi reduces both OT- and ISO-induced ERK1/2 activation in human myometrial hTERT-C3 cells. Cells were pretreated for 18 h with PTX (100 ng/mL) or vehicle water (control) and subsequently exposed to either 100 nM OT (A,B) or 10 μM ISO (C,D) for different times at 37 °C. Cells were lysed and levels of pERK1/2 were assessed by west- ern blotting, plotted relative to total ERK1/2 levels and expressed as percentage of maximal response (5 min) as in Fig. 1 (n=3). Representative blots are shown in the top panels (A,C). *, Pb0.05 vs. corresponding control value for each time point. **, Pb0.01 vs. corresponding control value for each time point.
3.4. Role of PKC in β2AR-mediated ERK1/2 activation in hTERT-C3 cells noted that β2AR-mediated ERK1/2 activation was attenuated in hTERT-C3 cells (Fig. 4C,D). We also observed partial attenuation of the To investigate which PKC isoform(s) was responsible for mediat- OTR-mediated ERK1/2 signal upon PKCζ inhibition, suggesting that Gi- ing the effects of OTR and β2AR on activation of ERK1/2, we tested dependent ERK/12 activation downstream of the OTR might use a the involvement of different subsets of PKC isoforms using more mechanism similar to that of the β2AR (Fig. 4C,D). selective inhibitors. Gö6983inhibits all the major classes of PKC iso- forms (classical, novel and atypical), while Rö31-8220 inhibits only 3.5. Recapitulating β2AR-mediated ERK1/2 signalling in the presence of classical and novel PKC isoforms and Gö6976 only inhibits the clas- OTR in HEK 293 cells sical isoforms. As mentioned above, the general PKC inhibitor
Gö6983 attenuated both OTR- and β2AR-mediated ERK1/2 signalling In order to investigate crosstalk between OTR and β2AR in regula- (Fig. 3). Although Rö31-8220, the inhibitor of classical and novel PKC tion of ERK1/2 activity in more detail, we reconstituted the system in isoforms, attenuated OT-mediated ERK1/2 signaling, it had no signifi- a cell line that did not endogenously express OTR. HEK 293 cells were cant effect on β2AR-mediated ERK1/2 signalling (Fig. 4A,B). Moreover, transfected with either β2AR or OTR alone or in combination. We Gö6976, the inhibitor of classical PKC isoforms, had no effect on either again assessed the role of Gαi in mediating ERK1/2 activation via
OT- or β2AR-mediated ERK1/2 signalling (Fig. 4A,B). These results sug- each receptor. We observed that β2AR-mediated activation of gested that an atypical PKC isoform may be involved in β2AR-mediated ERK1/2 was sensitive to inhibition of Gαi by PTX only when the two ERK1/2 signalling. Therefore, we examined the involvement of atypical receptors were co-expressed (Fig. 5). Similarly, we also reassessed PKCs using a specific pseudosubstrate inhibitor of PKCζ (20 μM). and the role of PKC in mediating ERK1/2 activation by each receptor.
Fig. 3. Inhibition of PKC reduces OT- and ISO-induced ERK1/2 activation in human myometrial hTERT-C3 cells. Cells were pretreated for 30 min with Gö6983 (1 μM), a general PKC inhibitor, or vehicle DMSO (control) and subsequently exposed to either 100 nM OT (A,B) or 10 μM ISO (C,D) for different times at 37 °C. Cells were lysed and levels of pERK1/2 were assessed by Western blotting, plotted relative to total ERK1/2 levels and expressed as percentage of maximal response (5 min, n=3). Representative blots are shown in the top panels (A,C). *, Pb0.05 vs. corresponding control value for each time point. **, Pb0.01 vs. corresponding control value for each time point. P.K. Wrzal et al. / Cellular Signalling 24 (2012) 333–341 337
Fig. 4. Specific inhibition of PKCζ isoform reduces OT- and ISO-induced ERK1/2 activa- tion in human myometrial hTERT-C3 cells. (A) Cells were pretreated for 30 min with either inhibitor of classical/novel PKC isoforms, Rö31-8220 (5 μM), inhibitor of classical PKC isoforms, Gö6976 (1 μM) or vehicle DMSO (control) and subsequently treated with 100 nM OT or 10 μM ISO for 5 min at 37 °C. Phosphorylated ERK1/2 (pERK1/2) and total ERK1/2 levels were assessed by western blotting. (B) Quantitative evaluation of 3 independent experiments represented in (A) and expressed as percentage of OT response. (C) Cells were pretreated for 30 min with specific PKCζ inhibitor (20 μM) or vehicle water (control) and subsequently treated with 100 nM OT or 10 μM of ISO for 5 min at 37 °C, (n=3). (D) Quantitative evaluation of (C) and expressed as percent- age of OT response. *, Pb0.05 vs. corresponding control value for each treatment.
Fig. 5. Effects of Gαi inhibition on ISO-induced ERK1/2 activation in presence and ab-
sence of OTR in HEK293 cells. Cells were transfected with either OTR (A) or β2AR Our results indicated that β2AR-mediated activation of ERK1/2 was (B) alone or in combination (C). Cells were grown in 0.05% serum 24 h prior to pretreat- only sensitive to inhibition of PKC with Gö6983 when OTR was co- ment with PTX (100 ng/mL) or vehicle water (control) for 18 h and subsequently trea- ted with 100 nM OT or 10 μM ISO for different times at 37 °C. Phosphorylated ERK1/2 expressed (Fig. 6). This suggested that the presence of the OTR (acti- (pERK1/2) and total ERK1/2 levels were assessed by western blotting and expressed β vated or not) altered the pattern of 2AR signalling in HEK 293 cells, as percentage of maximal response (lower panel; n=3). Representative blots are rendering it similar to what we observed in hTERT-C3 myometrial shown in the top panels of A, B, and C. *, Pb0.05 vs. corresponding control value for cells. each time point. **, Pb0.01 vs. corresponding control value for each time point.
ζ 3.6. Role of PKCζ in β2AR-mediated ERK1/2 activation in HEK 293 cells treatment with isoproterenol did not stimulate translocation of PKC when β2AR was expressed alone. However, when both β2AR and Results obtained in hTERT-C3 cells using the PKCζ inhibitor OTR were co-expressed in the same cells, we found that isoproterenol suggested the possibility that PKCζ or another atypical PKC was also induced membrane translocation of PKCζ (Fig. 7B). Subsequently, we involved in β2AR-mediated ERK1/2 activation in the presence of the tested whether blocking the OTR with an antagonist, OTA, affected iso- OTR in HEK 293 cells. First, we directly assessed whether PKCζ was proterenol-induced translocation of PKCζ in these cells. As shown in activated upon ligand stimulation of the β2AR. PKCs translocate to Fig. 7B, we found that this was indeed the case. Pretreatment with the plasma membrane when activated. Thus we investigated the OTA also blocked the oxytocin-induced translocation of PKCζ translocation of PKCζ using confocal microscopy in cells expressing ei- (Fig. 7B). In contrast, we did not observe translocation of PKCβ1 in re- ther the β2AR alone or in combination with OTR. As shown in Fig. 7A, sponse to isoproterenol in cells co-expressing the two receptors 338 P.K. Wrzal et al. / Cellular Signalling 24 (2012) 333–341
3.7. PI 3-kinase is involved in β2AR-mediated ERK1/2 activation in hTERT-C3 cells
We investigated the effect of wortmannin, an inhibitor of PI 3-kinase
(PI3K), on β2AR-mediated ERK1/2 activation. Inhibition of PI3K using wortmannin (100 nM) dramatically reduced β2AR-mediated ERK1/2 activation; however, wortmannin did not have a significant effect on OTR-mediated ERK1/2 activation (compare Fig. 8A,C with Fig. 8B,D). Also, we again noted that ERK1/2 activation was sensitive to inhibition
of PI3K only when the OTR was co-expressed with the β2AR in HEK 293 cells (Supplementary Fig. 1). Inhibition of Src using PP2 (5 μM) resulted
in a reduction of both OTR- and β2AR-mediated ERK1/2 activation for both early and late phases, indicating that Src is downstream of the G protein/PKC pathway (Supplementary Fig. 2). Thus, our results indicate
that there is an interaction between OTR and β2AR and that the pres- ence of the OTR enables the β2AR to signal through a pathway involving Gαi-PI3K-PKCζ/src to mediate ERK1/2 activation in human myometrial cells and in HEK 293 cells.
3.8. Neither OTR- and β2AR-mediated ERK1/2 activation in hTERT-C3 cells require EGFR transactivation
It has been demonstrated that activation of ERK1/2 via OTR can occur through transactivation of a tyrosine kinase receptor such as EGF recep- tor (EGFR) [11].Weconfirmed that EGF stimulated ERK1/2 activation in hTERT-C3 cells, and this response was blocked by treatment with AG- 1478 (1 μM), a specific inhibitor of the EGFR (Supplementary Fig. 3E). We then assessed the role of the EGFR in transactivating ERK1/2 follow-
ing either OTR or β2AR activation. In contrast, we found that treatment with AG1478 had no effect on either OTR-mediated (Supplementary
Fig. 3A,B) or β2AR-mediated (Supplementary Fig. 3C,D) ERK1/2 activa- tion in myometrial hTERT-C3 cells, thus ruling out transactivation of EGFR as a possible signalling pathway linking the two receptors in
hTERT-C3 cells. Finally, as it known that both β2AR and OTR recruit β- arrestin [12,30], we assessed whether β-arrestin was required for activa- tion of ERK1/2 in hTERT-C3 cells. Using either siRNA or shRNA targetting both isoforms of β-arrestin, we did not see any effect on stimulation of ERK1/2 by the OTR (data not shown).
4. Discussion
In the present study, we describe mechanisms by which OTR and
β2AR activate ERK1/2 in human myometrial hTERT-C3 cells using phar- macological manipulation of downstream receptor signalling pathways (summarized in Fig. 9). Importantly, the experiments conducted here indicate that these two receptors interact at physiological levels, in the context of untransfected cells. We propose a model where OTR acti- vates ERK1/2 via Gαq in a PKC-dependent manner but which can also
Fig. 6. Effects of PKC inhibition on ISO-induced ERK1/2 activation in absence and pres- involve Gαi. β2AR-mediated ERK1/2 activation in these cells was via a β ence of OTR in HEK293 cells. Cells were transfected with either OTR (A) or 2AR (B) PTX-sensitive, PKCζ-dependent pathway, which, following our recon- alone or in combination (C). Cells were grown in 0.05% serum 24 h prior to pretreat- stitution experiments in HEK 293 cells, is only operative in the presence ment with Gö6983 (1 μM) or vehicle DMSO (control) for 30 min and subsequently treated with 100 nM OT or 10 μM of ISO for different times at 37 °C. Phosphorylated of the OTR. Our results suggest an interaction between endogenous OTR ERK1/2 (pERK1/2) and total ERK1/2 levels were assessed by western blotting and and β2AR in hTERT-C3 cells. There is an overwhelming amount of OTR expressed as percentage of maximal response (lower panel; n=3). Representative compared to β2AR (14-fold higher as measured by ligand binding) in b blots are shown in the top panels of A, B, and C. *, P 0.05 vs. corresponding control hTERT-C3 cells. This suggests that modulation of the β AR, functional value for each time point. 2 and/or physical, in the context of receptor heterodimers (as we show in the attached companion article), would be markedly influenced by fi β (Fig. 7C). These experiments con rmed that 2AR-mediated mem- the high levels of OTR. In this OTR-dominated system, β2AR-activated brane recruitment of PKCζ, but not PKCβ1 was dependent on the pres- PKCs in turn trigger the sequential stimulation of Src, leading ultimately ence of the OTR. to ERK1/2 activation.
Fig. 7. PKC recruitment to the plasma membrane following OTR and β2AR activation in HEK293 cells. Images obtained by confocal microscopy showing translocation of either PKCζ- mCherry or PKCβ1-mCherry to the cell membrane following ISO and OT treatment. Cells were transfected with either: (A) β2AR-CFP and PKCζ-mCherry; or (B) β2AR-CFP, OTR-YFP and PKCζ-mCherry; or (C) β2AR-CFP, OTR-YFP and PKCβ1-mCherry. Cells were serum-starved for 30 min, where indicated cells were pretreated either with an OTR antagonist, OTA (100 μM), or vehicle water (control) for 15 min, and then treated with either OT (100 nM) or ISO (10 μM) for 1 min. The merged panels represent (A) YFP and mCherry, (B) and (C) CFP, YFP, and mCherry. The scale bar represents 10 μm. Results are representative of three independent experiments. P.K. Wrzal et al. / Cellular Signalling 24 (2012) 333–341 339
The activation of ERK1/2 by either receptor was PTX sensitive. The pathway by which β2AR coupled to activate ERK1/2 in hTERT-C3 attenuation of β2AR-mediated ERK1/2 by PTX was nearly complete. cells. Less than 50% of OTR-mediated ERK1/2 activation was abolished This suggested that a pathway involving Gαi was the primary by PTX treatment, suggesting that individual OTR complexes might be 340 P.K. Wrzal et al. / Cellular Signalling 24 (2012) 333–341
Fig. 8. Effect of PI3K inhibition on OT- and ISO-induced ERK1/2 activity in hTERT-C3 cells. Cells were pretreated for 1 h with Wortmannin (100 nM) or vehicle DMSO (control) and subsequently exposed to either 100 nM OT (A,B) or 10 μM of ISO (C,D) for different times at 37 °C. Cells were lysed and levels of pERK1/2 were assessed by western blotting, plotted relative to total ERK1/2 levels and expressed as percentage of maximal response (5 min, n=3). Representative blots are shown in the top panels (A,C). *, Pb0.05 vs. corresponding control value for each time point. **, Pb0.01 vs. corresponding control value for each time point. wired differently. We then investigated the involvement of PKC path- suggest that PKCζ is involved in OTR-mediated ERK1/2 activation. ways in ERK1/2 activation. We used a broad spectrum inhibitor of These results are consistent with other reports which show activation PKC, Gö6983, and found that for the OTR, inhibition of PKC lead to sig- of ERK1/2 by a mechanism involving Gαi-PI3-kinase-PKCζ and Src nificant attenuation in ERK1/2 activation. The effects of inhibition of [31–34]. Further, our results suggest that the involvement of PKCζ in either Gαi or the PKC pathway in OTR-mediated ERK1/2 activation β2AR-mediated ERK1/2 activation depends on the presence of the OTR were primarily seen in the early phase of ERK1/2 activation. This indi- in the same cells. Blocking the OTR with an antagonist (OTA) also cated that the mechanism by which OTR stimulation activated blocked the β2AR-mediated membrane recruitment of PKCζ. This sug- ERK1/2 involved, likely to an equal extent, two distinct G proteins, gests an interaction between OTR and β2AR where the presence of the Gαi and the Gαq, in human myometrial cells. However, the mecha- OTR is important for the signalling of β2AR to activate PKCζ. nism of β2AR-mediated ERK1/2 activation was different. It appeared Next we set out to recapitulate our mechanistic observations of that the principal G protein mediating the effects of β2AR was ERK1/2 activation in hTERT-C3 by using HEK 293 cells, where we through Gαi; however, we observed that PKC was involved as well. could control the expression of each receptor. We found that the
In order to inhibit the different subsets of PKC isoforms, we used the pathway for β2AR-mediated ERK1/2 activation that we observed in PKC inhibitors Gö6976, which inhibits classical PKC isoforms, and Rö- myometrial cells could only be seen in the presence of both receptors 31-8220, which inhibits both the classical and novel subtypes but not in HEK 293 cells. However, we noted that OTR-mediated ERK1/2 acti- the atypical PKC isoforms. The PKC inhibitor we used previously, vation was independent of the presence of β2AR, consistent with their Gö6983, inhibited all PKC isoforms. We found that inhibition with either independent action in hTERT-C3 cells, even when expressed at similar
Gö6976 or Rö-31-8220 did not attenuate the β2AR-mediated ERK1/2 ac- levels. tivation in hTERT-C3 cells. Similarly, these inhibitors did not lead to a Our studies also implicate PI3-kinase and Src in β2AR-mediated decrease in OTR-mediated ERK1/2 activation. Taken together, these re- ERK1/2 activation. Although Src also appears to be involved in OTR- sults implied that atypical PKCs were responsible for the β2AR-mediated mediated activation of ERK 1/2, PI3-kinase does not, highlighting the ERK1/2 activation. Inhibition of PKCζ also demonstrated its involvement unique mechanism involving both receptors, potentially implicating in β2AR-mediated ERK1/2 activation in hTERT-C3 cells. Our results also receptor heterodimerization (discussed in more detail in a companion article). Our results excluded the involvement of β-arrestin in OTR- mediated ERK1/2 activation, while EGFR transactivation was not in- volved ERK1/2 activation by either receptor in the human myometrial cells used here. This is in contrast to previous observations that show the involvement of EGFR in OTR-mediated activation of ERK1/2 [11]. These differences may be due to inherent differences in the cell types used for the studies. We used an immortalized human hTERT-C3 myo- metrial cell line while others used a transformed human PHM1 myome- trial cell line. A similar interpretation can be made regarding the role of β-arrestin in ERK1/2 activation in hTERT-C3 cells. Cellular context is likely to be a key feature underlying the architecture of specific signal- ling pathways. We did not pursue further studies with β-arrestin in HEK 293 cells as we could recapitulate the signalling pathway simply Fig. 9. Schematic representation of a model for the signaling pathway of ERK1/2 activa- by co-expressing the two receptors. We cannot exclude that the kinetics β tion via 2AR and OTR in human myometrial hTERTC3 cells. Isoproterenol (ISO) stim- of β-arrestin recruitment to either receptor might be altered when the ulates β2AR and oxytocin (OT) stimulates OTR through their canonical Gs- and Gq- dependent pathways, respectively. In our native cell model, where both receptors are two receptors are expressed together. expressed endogenously, the overwhelming presence of OTR results in β2AR coupling In summary, we present evidence that in human myometrial cells, α ζ to G i to activate a pathway which includes PI3 kinase, and the atypical PKC pathway. endogenous β2AR and OTR, which are both important physiological PKCζ then stimulates the ERK1/2 cascade through a pathway involving sequential acti- modulators of uterine activity, interact to activate ERK1/2.The present vation of Src. This β AR-mediated signalling pathway is dependent on the presence of 2 findings suggest that the novel signalling pathway described here for OTR. The likely basis of this crosstalk is through receptor heterodimerization (the sub- β ject of a companion article). OTR activates ERK1/2 through the canonical Gαq signal- 2AR-mediated ERK1/2 activation is a consequence of physical interac- ling which involves PKC and Src, as well as through Gαi. tions between the two receptors, which we explore more deeply in a P.K. Wrzal et al. / Cellular Signalling 24 (2012) 333–341 341 companion paper. Such physical interactions as hetero-oligomerization [15] A. Nohara, M. Ohmichi, K. Koike, N. Masumoto, M. Kobayashi, M. Akahane, H. Ike- fl gami, K. Hirota, A. Miyake, Y. Murata, Biochemical and Biophysical Research Com- of receptors has been shown to in uence many aspects of receptor func- munications 229 (1996) 938–944. tion [25,35–40]. Exploring the nature of the interaction between OTR [16] P. Crespo, T. Cachero, X. Ningzhi, J. Gutkind, Journal of Biological Chemistry 270 (1995) 25259–25265. and β2AR is crucial as it may be of physiological relevance for the precise [17] J. Schmitt, P. Stork, Journal of Biological Chemistry 275 (2000) 25342–25350. coordinationofcontractioninmyometrialcells. [18] R.J. Lefkowitz, K.L. Pierce, L.M. Luttrell, Molecular Pharmacology 62 (2002) Supplementary materials related to this article can be found on- 971–974. line at doi:10.1016/j.cellsig.2011.09.019. [19] S.K. Shenoy, M.T. Drake, C.D. Nelson, D.A. Houtz, K. Xiao, S. Madabushi, E. Reiter, R.T. Premont, O. Lichtarge, R.J. Lefkowitz, Journal of Biological Chemistry 281 (2006) 1261–1273. Acknowledgements [20] N. Dhanasekaran, L. Heasley, G. Johnson, Endocrine Reviews 16 (1995) 259–270. [21] O.F. Bueno, J.D. Molkentin, Circulation Research 91 (2002) 776–781. [22] T. Engstrom, P. Bratholm, N. Christensen, H. Vilhardt, Journal of Endocrinology We thank Marc-André Sylvain and Roger Tsien for the pcDNA3.1 – β 161 (1999) 403 411. (+)-mCherry construct, Dr. Marc Caron for the -arrestin shRNA con- [23] T. Engstrøm, P. Bratholm, H. Vilhardt, N.J. Christensen, Biology of Reproduction 60 structs and Dr. Dominic Devost for his helpful comments. This work (1999) 322–329. was supported by grants from the Canadian Institutes of Health Re- [24] D. Devost, H. Zingg, Journal of Molecular Endocrinology 31 (2003) 461–471. [25] C. Lavoie, J.-F. Mercier, A. Salahpour, D. Umapathy, A. Breit, L.-R. Villeneuve, W.-Z. search to TEH (MOP-36279) and HHZ (MOP-74675). TEH is a Chercheur Zhu, R.-P. Xiao, E.G. Lakatta, M. Bouvier, T.E. Hébert, Journal of Biological Chemis- National of the Fonds de la Recherche en Santé du Québec (FRSQ). try 277 (2002) 35402–35410. [26] J. Condon, S. Yin, B. Mayhew, R.A. Word, W.E. Wright, J.W. Shay, W.E. Rainey, Bi- ology of Reproduction 67 (2002) 506–514. References [27] D. Devost, H.H. Zingg, American Journal of Physiology 292 (2007) E1–E6. [28] Y. Smith, M. Dombrowski, K. Leach, W. Hurd, Obstetrics and Gynecology 85 – [1] K.L. Dodge, B.M. Sanborn, Endocrinology 139 (1998) 2265 2271. (1995) 357–360. [2] H.N. Simhan, S.N. Caritis, The New England Journal of Medicine 357 (2007) [29] M. Rezapour, T. Bäckström, U. Ulmsten, Steroids 61 (1996) 338–344. – 477 487. [30] J. Zhang, L.S. Barak, K.E. Winkler, M.G. Caron, S.S.G. Ferguson, Journal of Biological [3] A.D. Oldenhof, O.P. Shynlova, M. Liu, B.L. Langille, S.J. Lye, American Journal of Chemistry 272 (1997) 27005–27014. – Physiology. Cell Physiology 283 (2002) C1530 C1539. [31] M.M. Chou, W. Hou, J. Johnson, L.K. Graham, M.H. Lee, C.-S. Chen, A.C. Newton, B.S. [4] S. Ahn, S.K. Shenoy, H. Wei, R.J. Lefkowitz, Journal of Biological Chemistry 279 Schaffhausen, A. Toker, Current Biology 8 (1998) 1069–1078. – (2004) 35518 35525. [32] T. Takeda HM, T. Takada, T. Noguchi, T. Yamao, M. Tsuda, F. Ochi, K. Fukunaga, K. [5] K. Arai, Y. Maruyama, M. Nishida, S. Tanabe, S. Takagahara, T. Kozasa, Molecular Inagaki, M. Kasuga, EMBO Journal 18 (1999) 386–395. – Pharmacology 63 (2003) 478 488. [33] M.M. Monick, A.B. Carter, D.M. Flaherty, M.W. Peterson, G.W. Hunninghake, Jour- fl – [6] S. Arnaudeau, N. Lapretre, J. Mironneau, P ügers Archiv 428 (1994) 51 59. nal of Immunology 165 (2000) 4632–4639. [7] C. Ku, A. Qian, Y. Wen, K. Anwer, B. Sanborn, Endocrinology 136 (1995) [34] P. Robin, I. Boulven, C. Desmyter, S. Harbon, D. Leiber, American Journal of Phys- – 1509 1515. iology. Cell Physiology 283 (2002) C251–C260. [8] Z. Strakova, J.A. Copland, S.J. Lolait, M.S. Soloff, American Journal of Physiology [35] B.A. Jordan, N. Trapaidze, I. Gomes, R. Nivarthi, L.A. Devi, Proceedings of the Na- – 274 (1998) E634 E641. tional Academy of Sciences of the United States of America 98 (2001) 343–348. [9] M. Ohmichi, K. Koike, A. Nohara, Y. Kanda, Y. Sakamoto, Z. Zhang, K. Hirota, A. [36] S. Terrillon, M. Bouvier, EMBO Reports 5 (2004) 30–34. – Miyake, Endocrinology 136 (1995) 2082 2087. [37] A. Breit, M. Lagacé, M. Bouvier, Journal of Biological Chemistry 279 (2004) – [10] Z. Strakova, M.S. Soloff, American Journal of Physiology 272 (1997) E870 E876. 28756–28765. – [11] M. Zhong, M. Yang, B.M. Sanborn, Endocrinology 144 (2003) 2947 2956. [38] S.C. Prinster, C. Hague, R.A. Hall, Pharmacological Reviews 57 (2005) 289–298. [12] A. Hasbi, D. Devost, S.A. Laporte, H.H. Zingg, Molecular Endocrinology 18 (2004) [39] D.W. McGraw, K.A. Mihlbachler, M.R. Schwarb, F.F. Rahman, K.M. Small, K.F. – 1277 1286. Almoosa, S.B. Liggett, The Journal of Clinical Investigation 116 (2006) 1400–1409. [13] P.J. Brighton, S. Rana, R.A.J. Challiss, J.C. Konje, J.M. Willets, British Journal of Phar- [40] M.J. Lohse, Current Opinion in Pharmacology 10 (2010) 53–58. macology 162 (2011) 1603–1617. [14] C.A. Grotegut, L. Feng, L. Mao, R.P. Heine, A.P. Murtha, H.A. Rockman, American Journal of Physiology 300 (2011) E468–E477. I.D. Neumann and R. Landgraf (Eds.) Progress in Brain Research, Vol. 170 ISSN 0079-6123 Copyright r 2008 Elsevier B.V. All rights reserved
CHAPTER 15
Oxytocin receptor signalling
Dominic Devost1,ÃÃ, Paulina Wrzal1 and Hans H. Zingg1,2,3,Ã
1Department of Pharmacology, McGill University, Montreal, Quebec, Canada 2Department of Medicine, McGill University, Montreal, Quebec, Canada 3Department of Obstetrics & Gynecology, McGill University, Montreal, Quebec, Canada
Abstract: The great diversity of the expression sites and proposed function of the oxytocin (OXT) receptor (OXTR) is paralleled by a diversity of its signalling pathways, many of which have still remained unexplored. We have used different approaches to discover novel pathways. By means of a phosphoproteomics approach, we have detected several distinct OXT-induced changes in tyrosine as well as threonine phosphorylation states of intracellular protein in myometrial cells. The most prominent change involved dephosphorylation of a 95-kDa phosphothreonine moiety. By N-terminal amino acid microsequence analysis, this moiety was shown to correspond to eukaryotic translation factor eEF2. This protein is a key regulator of protein synthesis and mediates, upon dephosphorylation, the translocation step of peptide chain elongation. These findings define a novel mechanism by which OXT assumes a so far unrecognized trophic function. We next elucidated the intracellular pathway(s) involved. We found that this effect is not mediated by any of the known pathways known to induce eEF2 dephosphorylation (mTOR, ERK1/2 or p38) but by protein kinase C. Consistent with this idea, we also found that direct stimulation of protein kinase C with a phorbol ester induced eEF2 dephosphorylation in myometrial cells. Using phosphoERK antibodies, we discovered by Western blotting that OXT induced phosphorylation of a higher molecular weight ERK-related protein. We were able to show that this band corresponded to ‘‘big MAP kinase1’’ or ERK5. ERK5 is part of a distinct MAPK cascade and promotes expression of the myosin light chain gene and plays an obligatory role in muscle cell development and differentiation. The role of ERK5 in myometrium has remained unexplored, but it is likely to represent an important novel pathway mediating OXT’s effects on smooth muscle function. Further elucidation of these novel signalling pathways will have significant relevance for the development of novel pathway-specific OXTR agonists and antagonists.
Keywords: oxytocin; oxytocin receptor; trophic action; MAP kinases; ERK5; elongation factor 2; contraction in vitro assay
General thoughts on oxytocin receptor signalling
The oxytocin (OXT) receptor (OXTR) is a member of the family of G protein-coupled ÃCorresponding author. Tel.: +1 (514) 398 3621; Fax: +1 (514) 398 2045; E-mail: [email protected] receptors (GPCRs) and, together with the three ÃÃCo-Corresponding author: Tel.: +1 (514) 398 4888; vasopressin receptor subtypes V1a, V1b and V2, Fax: +1 (514) 398 2045; E-mail: [email protected] the OXTR forms a subfamily of structurally
DOI: 10.1016/S0079-6123(08)00415-9 167 168 related GPCRs. The OXTR mediates a very wide action, barusiban, has also been developed spectrum of physiological actions: OXTRs are (Reinheimer, 2007). On the other hand, expressed in different specific brain regions where OXT is the strongest uterotonic agent they mediate behavioural functions, ranging from known and is used pharmacologically ante- maternal behaviour to specific sexual and social partum to induce or augment labour as well behaviours. In the periphery, OXTRs mediate as postpartum to control postpartum hemor- effects on uterine contractions, mammary gland rhage (Saito et al., 2007). From a pharma- milk ejection and differentiation, pituitary prolac- cological point of view, the OXTR is the tin secretion, sodium excretion, T-cell function, target of two opposing clinically relevant cardiovascular control, cardiomyocyte and osteo- pharmacological strategies aimed at control- blast differentiation and endothelial cell function ling uterine motility. From a physiological (Zingg, 2000; Gimpl and Fahrenholz, 2001; Zingg point of view, the efficiency of OXTR and Laporte, 2003; Lim and Young, 2006). antagonists to block preterm labour con- Finally, OXTRs are widely expressed in different tractions indicates that premature upregula- cancers. Over 80% of breast carcinomas express tion of the OXT/OXTR signalling system the OXTRs, and OXTRs are present in endome- may be causally involved in provoking trial adenocarcinomas, choriocarcinomas, glio- preterm labour contractions. blastomas, neuroblastomas and small cell lung carcinomas (Cassoni et al., 2004). The great diversity of the expression sites and The OXTR has several features that render its proposed function of the OXTR is paralleled by a study particularly interesting and relevant: diversity of its signalling pathways, many of which have still remained unexplored. A Gq/11-mediated (1) The OXTR undergoes an exceptionally pathway leading to stimulation of phospholipase C dramatic tissue-specific up- and down regu- (PLC) inducing increased intracellular calcium and lation. In the uterus, OXTR expression inositol trisphosphate production has been clearly undergoes in all mammalian species studied defined (Zhong et al., 2003); however, the precise a 10–100 fold upregulation during pregnancy mechanisms by which OXT exerts its multiple (Soloff et al., 1979; Fuchs et al., 1984; biological actions have not been fully established. Larcher et al., 1995). This striking regulation In addition to activation of PLC, the OXTR is involves mechanisms at the level of gene also able to activate the MAP kinases ERK1 and transcription, mRNA stability and at the ERK2 in myometrial cells (Fig. 1A, B)(Ohmichi et level of signal transduction and trafficking al., 1995; Zhong et al., 2003). With respect to (Zingg and Laporte, 2003). myometrial contractions, the OXTR and the b2AR (2) The mixed OXT/vasopressin V1a receptor mediate opposite effects; however both receptors antagonist atosiban has proven to be clini- activate the ERK1/2 pathway. Yet, the dynamics cally effective in inhibiting preterm myome- of ERK1/2 activations differ between the two trial contractions. In fact, atosiban is equally receptors. The activation mediated by the b2AR is effective in delaying preterm birth as treat- very transient and lasts o10 min, OXTR-mediated ment by ritrodrine, an agonist specific for the ERK1/2 activation is more prolonged and lasts b2 adrenergic receptor (b2AR). However, W1h (Fig. 1A, B). The MAP kinases ERK1/2 atosiban is accompanied with significantly mediate several different biological actions. A main less side effects (French/Australian Atosiban distinction exists between their nuclear and cyto- Investigators, 2001; The Worldwide Atosi- plasmic actions. At the level of the cell nucleus, ban versus Beta-agonists Study, 2001). As a ERK1/2 mediate proliferative effects via mecha- result, atosiban (Tractociles) is currently nisms that include induction of c-fos synthesis approved in 29 countries in Europe for the (Karin, 1996). At the level of the cytoplasm, treatment of preterm labour. A more specific ERK1/2 effects have been proposed to mediate OXT antagonist with a longer duration of contractions and prostaglandin synthesis (Strakova 169
A
pERK1/2
0 2 5 10 30 0 2 5 10 30 min
B
pERK1/2
0 5 30 60 120 180 0 5 30 60 120 180 min
oxytocin (100nM) isoproterenol (10 µM)
C
pERK5
0251030 min oxytocin (100nM)
Fig. 1. (A, B) ERK1/2 activation in response to stimulation of OXTR (left) or the b2 adrenergic receptor (right) in myometrial cells. Immortalized human myometrial hTERT-C3 cells (Devost and Zingg, 2007) were exposed to OXT or isoproterenol for the times indicated and phosphorylated ERK1/2 was immuno-detected by Western blotting using an anti-phosphoERK1/2 antibody (Cell Signalling Technology). (C) Kinetics of OXT-induced ERK5 induction in M11 myometrial cells. Cells were kept in serum-free medium for 48 h and OXT was added for the times indicated. Cell lysates were immunoblotted with an anti-phospho-ERK5 antibody (Cell Signalling Technology). et al., 1998; Li et al., 2003). A mechanism has been leading to activation of cytoplasmic targets which, proposed that accounts for nuclear versus cyto- for myometrial cells, may include prostaglandin plasmic trafficking of ERK1/2 (Ahn et al., 2004). production and contraction. The model further The model proposes two general pathways of suggests that the former pathway is G-protein ERK1/2 activation: (a) a very transient ERK dependent whereas the latter is b-arrestin mediated activation (maxB5 min) that involves nuclear and G protein independent. translocation of activated ERK leading to specific The prolonged kinetics of ERK1/2 activation gene activation and a proliferative response; and by the OXTR suggests that OXTR-mediated (b) a sustained ERK activation (W30 min) that is ERK1/2 activation may include both nuclear and associated with ERK retention in the cytoplasm cytoplasmic effects, and that the sustained 170 activation of ERK1/2 is mediated via b-arrestin. following activation of the OXTR, we have Indeed, we have observed by confocal microscopy determined, on a global level, to what extent that the OXTR co-localizes with b-arrestin following OXTR signalling involves phosphorylation or internalization, suggesting that it internalizes into dephosphorylation of specific signalling compo- endosomes together with arrestin (unpublished nents. We analyzed OXT-induced changes in the data). On the other hand, the very transient protein phosphorylation pattern in lysates of CHO ERK1/2 activation kinetics associated with the cells stably transfected with the rat OXTR (CHO- b2AR suggest that the ERK1/2 action may be OTR cells), using one- and two-dimensional limited to nuclear effects. Studies determining to polyacrylamide gel electrophoresis in conjunction what extent OXTR-induced effects on ERK1/2 are with specific phosphotyrosine and phosphothreo- b-arrestin mediated are currently in progress. nine antibodies. As a result, we have detected a Interestingly, Rimoldi et al. (2003) have recently number of distinct OXT-induced changes in the shown that, if localized in caveolae (a type of lipid phosphorylation patterns of cellular proteins. raft), the OXTR induces a transient ERK activa- These included pY100 and pY65, a 100 and a tion and transduces a mitogenic signal. However, 65 kDa moiety, respectively, both phosphorylated if localized outside caveolae, the OXTR induced a on tyrosine. In addition, pT65, a 65 kDa protein sustained ERK response. Based on this finding, it phosphorylated on threonine, and pT95, a 95 kDa appears that differential localization to membrane protein de-phosphorylated on threonine. Phosphory- subdomains is an important determinant for lation of pY100 and pY65 was very rapid and OXTR function and has to be taken into consid- occurred within 2 min of OXT addition. The eration for a complete understanding of OXTR nature of these moieties is under investigation. function. Dephosphorylation of pT95 occurred between Further exploration and characterization of 5 and 60 min and was accompanied by the con- different OXTR-dependent pathways is of major comitant de novo phosphorylation of pT65 (Fig. 2). pharmacological importance. Recent advances in Maximum dephosphorylation of pT95 was observed the field of drug action have made it clear that, for at 20 min after OXT addition. Concomitantly, the a given GPCRs coupled to different signalling pathways, a given receptor ligand may act as an antagonist for one pathway and as an agonist for another. Such a ligand is now referred to a ‘‘biased agonist’’ (Galandrin and Bouvier, 2006). It turns out that many ligands, initially classified as ‘‘antagonists’’ are in fact biased agonists, if a more comprehensive spectrum of signalling pathways is taken into consideration (Galandrin and Bouvier, 2006). As a case in point, the mixed OXTR/ vasopressin V1a receptor antagonist atosiban mentioned above, has now been shown to be a biased agonist as well: Whereas it acts as an OXTR antagonist with respect to Gaq-mediated Fig. 2. OXT-induced changes in the overall threonine-phos- phorylation pattern in CHO cells stably transfected with the rat effects, it is a partial activator of the ERK1/2 OXTR (CHO-OTR cells). CHO-OTR cells were serum-starved pathway (Reversi et al., 2005). for 24 h and treated for the times indicated with 100 nM OXT with or without prior treatment with the OXT-antagonist OTA at 1 mM. Proteins in cell lysates were separated by 7.5% SDS- Novel OXTR signalling target: eEF2 PAGE and analyzed by immunoblotting using a specific anti- phospho-threonine antibody (Cell Signalling Technology). The position of the major de-phosphorylated substrate pT95 and the In an attempt to characterize more completely the major phosphorylated substrate pT65 are indicated. Adapted spectrum of signalling pathways set into action with permission from Devost et al., 2005. 171
de novo phosphorylation of a band at Mr 65 kDa (termed pT65) was observed with a maximum of phosphorylation coincident with the maximal de- phosphorylation of pT95. Additional bands were observed at 110 and 120 kDa that were rapidly phosphorylated within the first 2 min after OXT addition (Fig. 2). We used ion exchange chromatography and one- and two-dimensional gel chromatography to purify pT95. By N-terminal microsequence analy- sis, the purified pT95 moiety was shown to correspond to eukaryotic translation factor eEF2. This protein is a key regulator of protein synthesis and mediates, upon dephosphorylation, the trans- location step of peptide chain elongation (Browne and Proud, 2002). These findings define a novel mechanism by which OXT assumes a trophic function. To confirm that pT95 corresponded to eEF2, we performed immunoblot analysis using a commer- cially available anti-phospho eEF2 antibody. If pT95 corresponded indeed to phospho-eEF2, then OXT should induce a decrease in phospho-eEF2 immunoreactivity that corresponded to the one observed for pT95. Moreover, we wished to determine whether this phenomenon was restricted to CHO-OTR cells or whether it could also be observed in untransformed myometrial cells. As Fig. 3. OXT-induced eEF2 dephosphorylation in myometrial shown in Fig. 3A, OXT induced a rapid decrease M11 cells. (A) time course of OXT-induced dephosphorylation. in phospho-eEF2 immunoreactivity in myometrial Cells were treated as in Fig. 1 with OXT for different times and eEF2 phosphorylation was assessed by immunoblotting using M11 cells. The time course of dephosphorylation an anti-phospho-eEF2 antibody (Cell Signalling Technologies). corresponded to the one observed for pT95 in Autoradiograms resulting from three independent experiments CHO-OTR cells. This finding provided further were analyzed by densitometric analysis using ImageQuant 5.1. confirmation that the pT95 band corresponded to The control values were set to 100%. Each point represents the 7 eEF2 and indicated that OXT-induced eEF2 mean s.e.m. A representative autoradiogram is shown in the top panel. (B) Dose/response curve of OXT-induced eEF2 dephosphorylation occurs in myometrial cells. dephosphorylation. eEF2 phosphorylation was assessed as in We next determined the dose–response relation- (A), and the means7s.e.m. from three independent experiments ship of OXT-induced eEF2 dephosphorylation. As were plotted against the OXT concentration used. A represen- shown in Fig. 3B, the maximum effective concen- tative autoradiogram is shown in the top panel. Adapted with tration of OXT was 10À8 M, and the efficiency of permission from Devost et al., 2005. OXT induced dephosphorylation decreased with concentrations above 10À7 M. The fact that the Because it is widely accepted that eEF2 dephos- same dephosphorylation was observed in the non- phorylation is accompanied by an increase in the transformed myometrial cells as in OXTR-trans- rate of peptide chain elongation and, as a result, of fected CHO cells indicated that the observed OXT protein synthesis, we wished next to determine to effect is physiologically relevant, since it can be what extent OXTR activation was leading to a mediated by the endogenous OXTR in a physio- measurable increase in overall protein synthesis. logically relevant cell type. To this end we determined the effect of 100 nM 172
OXT on the amount of [35S]methionine incorpora- application of the general PKC inhibitor Go¨6983 tion into proteins in myometrial cells. OXT completely blocked the effect of OXT on eEF2 induced a significant 29% increase in the rate of dephosphorylation. To confirm further the hypo- total protein synthesis over a 2 h period (Devost thesis of a PKC-mediated effect of OXT on eEF2 et al., 2005). This stimulatory effect was similar to dephosphorylation, we blocked PKC activation the one induced by insulin (32%). This finding functionally by pretreatment with the peptide indicated that the observed OXT-induced dephos- ‘‘myr-psi PKC’’, an N-myristoylated pseudosub- phorylation of eEF2 is functionally meaningful strate of PKC with specificity for the PKCa and and supports a novel role for OXT as a trophic PKCb isotypes. Pretreatment with this peptide agent in the myometrium. also abrogated the effect of OXT on eEF2 dephos- In an attempt to delineate the pathway by which phorylation. Consistent with a PKC-mediated OXT exerts its effect on eEF2 dephosphorylation, effect, we found that direct phorbol ester-induced we systematically explored each of the pathways stimulation of PKC also led to a significant know to stimulate eEF2 activity (Devost et al., decrease in eEF2 phosphorylation (Devost et al., 2008). This included (i) the serine/threonine 2008). protein kinase ‘‘mammalian target of rapamycin’’, As an indication for the physiological significance mTOR, a pathway that is specifically inhibited by of these findings, we observed that in hTERT C3 the bacterial natural product rapamycin (Browne cells, OXT led to a 38% increase in [35S]-methionine and Proud, 2002); (ii) the MAP kinases ERK1 and incorporation into nascent proteins and that this ERK2, a pathway mediating the trophic effects of effect was abrogated by preincubation of cells with several GPCR agonists; (iii) ‘‘stress-activated the PKC inhibitor Go¨6983, indicating that PKC protein kinase 4’’ (SAPK4 or p38MAPK). activation represents an obligatory link connecting First we analyzed the role of the mTOR OXTR activation to increased protein synthesis via pathway by using the selective mTOR inhibitor eEF2 dephosphorylation (Devost et al., 2008). rapamycin. Whereas rapamycin pretreatment was effective in blocking insulin action on eEF2 dephosphorylation, it had no effect on the action Novel OXTR-linked signalling pathway: ERK5 of OXT on eEF2 dephosphorylation. To deter- mine whether p38 activation was involved in eEF2 Using phosphoMAPK antibodies, we discovered dephosphorylation, we used the selective p38 by Western blotting that OXT induced phospho- kinase inhibitor SB203580. However, blockage of rylation of a higher molecular weight ERK-related this pathway also remained without effect on the protein. We were able to show that this band ability of OXT to induce eEF2 dephosphorylation. corresponded to ‘‘big MAP kinase1’’ or ERK5. We next blocked ERK1/2 activation with 1 mMof ERK5 is part of a distinct MAPK cascade and the specific MEK-1 inhibitor U0126. Although promotes expression of the myosin light chain gene ERK1/2 activation was efficiently inhibited, appli- and plays an obligatory role in muscle cell cation of this MEK-1 inhibitor, we did not observe development and differentiation. The role of any effect on the action of OXT on eEF2 de- ERK5 in myometrium has remained unexplored, phosphorylation. Thus, none of the classically but it is likely to represent an important novel established pathways known to induce eEF2 pathway mediating OXT’s effects on smooth dephosphorylation appeared to mediate the effect muscle function. of OXT. We have confirmed that OXT induces ERK5 Due to its coupling to Gaq/11, the OXTR is able activity by using a specific phosphoERK5 anti- to activate protein kinase C (PKC). Therefore we body (Fig. 1C) as well as by an in vitro tested the hypothesis that the process might be phosphorylation assay using the highly specific mediated by PKC, although a link between PKC ERK5 substrate MEF2-C (unpublished). Our time and eEF2 dephosphorylation has never been course analysis using a phoshoERK5 antibody formally demonstrated. Indeed we found that showed that OXT-induced ERK5 activation was 173
Fig. 4. Schematic diagram of novel and established OXTR-linked signalling pathways. This diagram illustrates some novel potential OXTR signalling pathways and their relationship to established pathways. Note also the bifurcation of ERK1/2 signalling into nuclear and cytoplasmic signalling. See text for further explanations. sustained for at least 30 min (Fig. 1C). Interest- in mediating OXT-induced contractions. In sup- ingly, U0126, an inhibitor of MAPK kinases 1 and port of that idea, we found a dramatic increase in 2, blocked OXTR-mediated ERK1/2 activation, ERK5 phosphorylation in vivo in rat myometrium but at 10 mM, it also blocked ERK5 activation, during pregnancy with a maximum at labour implying that certain effects classically attributed (unpublished). We hypothesize that the MEK5/ to ERK1/2 may in fact be mediated by ERK5 ERK5 signalling cascade may be an important (Mody et al., 2001). We also determined that the mediator of OXT actions, including prostaglandin PKC inhibitor Go¨6983, which blocked ERK1/2 synthesis, myometrial and myoepithelial cell pro- activation, increased ERK5 activation (unpub- liferation and/or differentiation. If indeed ERK5 is lished results). Thus, whereas PKC has been involved in mediating specifically certain impor- proposed to mediate OXT-induced ERK1/2 acti- tant uterine functions, the ERK5 signalling path- vity (Strakova et al., 1998), PKC has an inhibitory way could represent an attractive drug target. effect on OXT-induced ERK5 activation (Fig. 4). In this context, it is interesting to note that PKC activation has been reported to inhibit OXT- In vitro contraction assay induced myometrial contractions (Phillippe, 1994). This finding is compatible with the idea that ERK5 To assess the involvement of ERK in mediating rather than ERK1/2 activation could be involved OXT-induced contractions, we have now developed 174
Fig. 5. OXT-induced contractions of myometrial cells in vitro. Collagen lattices were prepared and covered either by a layer of hTERT-C3 (top) or M11 (bottom) human myometrial cells. OXT alone or in combination with the OXT antagonist OTA was added as indicated. Lattices are visible in the center of each well. Adapted with permission from Devost and Zingg, 2007.
successfully a sensitive in vitro myometrial cell assay for high throughput analysis is currently in contraction assay (Devost and Zingg, 2007). In progress. brief, myometrial hTERT-C3 cells were layered on top of collagen matrices in 24-well plates. Two hours after plating, the matrix was detached and Conclusions allowed to float. OXT was then added at different concentrations for 1–18 h. Following fixation, Further elucidation of novel signalling pathways collagen matrices were digitally scanned and the linked to the OXTR will have significant relevance surface area was measured using the program for the development for novel pathway-specific ImageQuant 5.1. In the absence of OXT, cells OXTR agonists and antagonists. The effect of new induced a basal contraction of 4873%. Addition drug candidates will have to be assessed with of 100 nM OXT induced a significant increase in respect to each of these additional pathways. This the contraction to 6373%. Addition of the is particularly relevant in the light of the recent OXTR antagonist OTA in conjunction with realization that receptor ligands can act in a OXT reduced the contraction to below the basal pathway-specific fashion and exert differential levels to 3972% (Fig. 5 and Devost and Zingg, agonistic or antagonistic effects on distinct path- 2007). These results indicate that we have devel- ways linked to the same receptor. oped a very sensitive and reproducible in vitro bio-assay to assess quantitatively the biological action of OXT on myometrial cell contraction. Abbreviations Since this assay is based on the use of myometrial cell lines, it offers not only full control of the b2AR beta-2 adrenergic receptor extracellular environment but allows also the eEF2 eukaryotic elongation factor 2 possibility of genetically manipulating the cells ERK 1/2 extracellular signal-regulated used in the assay system. Adaptation of the kinases 1 and 2 175
ERK 5 extracellular signal-regulated the pluridimensionality of efficacy. Mol. Pharmacol., 70: kinase 5 1575–1584. GPCR G protein-coupled receptor Gimpl, G. and Fahrenholz, F. (2001) The oxytocin receptor system: structure, function, and regulation. Physiol. Rev., 81: MAPK mitogen-activated protein kinase MEF2-C myogenic enhancer factor 2-C 629–683. Karin, M. (1996) The regulation of AP-1 activity by mitogen- mTOR mammalian target of rapamycin activated protein kinases. Philos. Trans. R. Soc. Lond. B OXT oxytocin Biol. Sci., 351: 127–134. OXTR oxytocin receptor Larcher, A., Neculcea, J., Breton, C., Arslan, A., Rozen, F., PKC protein kinase C Russo, C. and Zingg, H.H. (1995) Oxytocin receptor gene PLC phospholipase C expression in the rat uterus during pregnancy and the estrous cycle and in response to gonadal steroid treatment. Endocri- s.e.m. standard error of mean nology, 136: 5350–5356. Li, Y., Je, H.D., Malek, S. and Morgan, K.G. (2003) ERK1/2- mediated phosphorylation of myometrial caldesmon Acknowledgement during pregnancy and labor. Am. J. Physiol., 284: R192–R199. Lim, M.M. and Young, L.J. (2006) Neuropeptidergic regula- Supported by a grant from the Canadian Institutes tion of affiliative behavior and social bonding in animals. for Health Research (CIHR). Horm. Behav., 50: 506–517. Mody, N., Leitch, J., Armstrong, C., Dixon, J. and Cohen, P. (2001) Effects of MAP kinase cascade inhibitors on the MKK5/ERK5 pathway. FEBS Lett., 502: 21–24. References Ohmichi, M., Koike, K., Nohara, A., Kanda, Y., Sakamoto, Y., Zhang, Z.X., Hirota, K. and Miyake, A. (1995) Oxytocin Ahn, S., Shenoy, S.K., Wei, H. and Lefkowitz, R.J. (2004) stimulates mitogen-activated protein kinase activity in Differential kinetic and spatial patterns of beta-arrestin2 and cultured human puerperal uterine myometrial cells. Endocri- G protein-mediated ERK activation by the angiotensin II nology, 136: 2082–2087. receptor. J. Biol. Chem., 279: 35518–35525. Phillippe, M. (1994) Protein kinase C, an inhibitor of oxytocin- Browne, G.J. and Proud, C.G. (2002) Regulation of peptide- stimulated phasic myometrial contractions. Biol. Reprod., 50: chain elongation in mammalian cells. Eur. J. Biochem., 269: 855–859. 5360–5368. Reinheimer, T.M. (2007) Barusiban suppresses oxytocin- Cassoni, P., Sapino, A., Marrocco, T., Chini, B. and Bussolati, induced preterm labour in non-human primates. BMC G. (2004) Oxytocin and oxytocin receptors in cancer cells and Pregnancy Childbirth, 7(Suppl 1): p. S15. proliferation. J. Neuroendocrinol., 16: 362–364. Reversi, A., Rimoldi, V., Marrocco, T., Cassoni, P., Bussolati, Devost, D., Carrier, M.E. and Zingg, H.H. (2008). Oxytocin- G., Parenti, M. and Chini, B. (2005) The oxytocin induced activation of eukaryotic elongation factor eEF2 in receptor antagonist atosiban inhibits cell growth via a myometrial cells is mediated by protein kinase C. Endocrino- ‘‘biased agonist’’ mechanism. J. Biol. Chem., 280: logy 149: 131–138. 16311–16318. Devost, D., Girotti, M., Carrier, M.E., Russo, C. and Zingg, Rimoldi, V., Reversi, A., Taverna, E., Rosa, P., Francolini, M., H.H. (2005) Oxytocin induces dephosphorylation of eukar- Cassoni, P., Parenti, M. and Chini, B. (2003) Oxytocin yotic elongation factor 2 in human myometrial cells. receptor elicits different EGFR/MAPK activation patterns Endocrinology, 146: 2265–2270. depending on its localization in caveolin-1 enriched domains. Devost, D. and Zingg, H.H. (2007) Novel in vitro system for Oncogene, 22: 6054–6060. functional assessment of oxytocin action. Am. J. Physiol., Saito, K., Haruki, A., Ishikawa, H., Takahashi, T., Nagase, H., 292: E1–E6. Koyama, M., Endo, M. and Hirahara, F. (2007) French/Australian Atosiban Investigators Group. (2001) Treat- Prospective study of intramuscular ergometrine compared ment of preterm labor with the oxytocin antagonist atosiban: with intramuscular oxytocin for prevention of postpartum a double-blind, randomized, controlled comparison with hemorrhage. J. Obstet. Gynaecol. Res., 33: 254–258. salbutamol. Eur. J. Obstet. Gynaecol. Reprod. Biol., 98: Soloff, M.S., Alexandrova, M. and Fernstrom, M.J. (1979) 177–185. Oxytocin receptors: triggers for parturition and lactation? Fuchs, A.R., Fuchs, F., Husslein, P. and Soloff, M.S. (1984) Science, 204: 1313–1315. Oxytocin receptors in the human uterus during pregnancy Strakova, Z., Copland, J.A., Lolait, S.J. and Soloff, M.S. (1998) and parturition. Am. J. Obstet. Gynecol., 150: 734–741. ERK2 mediates oxytocin-stimulated PGE2 synthesis. Am. J. Galandrin, S. and Bouvier, M. (2006) Distinct signaling profiles Physiol., 274: E634–E641. of beta1 and beta2 adrenergic receptor ligands toward The Worldwide Atosiban versus Beta-agonists Study adenylyl cyclase and mitogen-activated protein kinase reveals Group. (2001) Effectiveness and safety of the oxytocin 176
antagonist atosiban versus beta-adrenergic agonists in the growth factor receptor tyrosine kinase activation. Endocri- treatment of preterm labour. The Worldwide Atosiban versus nology, 144: 2947–2956. Beta-agonists Study Group. Br. J. Obst. Gynaecol., 108: Zingg, H.H. (2000) Oxytocin. In: De Groot L.J. and Jameson 133–142. J.L. (Eds.), Endocrinology (4th ed.). W B Saunders Comp., Zhong, M., Yang, M. and Sanborn, B.M. (2003) Extracellular Philadelphia, pp. 201–208. signal-regulated kinase 1/2 activation by myometrial oxytocin Zingg, H.H. and Laporte, S.A. (2003) The oxytocin receptor. receptor involves Galpha(q)Gbetagamma and epidermal Trends Endocrinol. Metab., 14: 222–227.