In Vitro and in Vivo Studies on Protease-Activated Receptor-2

IN VITRO AND IN VIVO STUDIES ON PROTEASE-ACTIVATED RECEPTOR-2

Mark N. Adams Bachelor of Applied Science (Hons)

Queensland University of Technology Brisbane, Queensland, Australia

Submitted in fulfilment of the requirements for the degree of Doctor of Philosophy

2012

KEYWORDS

Protease-activated receptor-2 (PAR2), G-protein coupled receptor (GPCR), signal transduction, trafficking, palmitoylation, agonist, antagonist, prostate cancer, bone metastasis, SCID mice, intra-tibial injection.

i ABSTRACT

Protease-activated receptor-2 (PAR2) is a G protein coupled receptor (GPCR) that is activated by proteolytic cleavage of its amino terminal domain by trypsin-like serine proteases. Cleavage of this receptor exposes a neoepitope, termed the tethered ligand (TL), which binds intramolecularly within the receptor to stimulate signal transduction via coupled G proteins. PAR2-mediated signal transduction is also experimentally stimulated by hexapeptides (agonist peptides; APs) that are homologous to the TL sequence. Due to the irreversible nature of PAR2 proteolysis, downstream signal transduction is tightly regulated. Following activation, PAR2 is rapidly uncoupled from downstream signalling by the post-translational modifications phosphorylation and ubiquination which facilitate interactions with β- arrestin. This scaffolding protein couples PAR2 to the internalisation machinery initiating its desensitisation and trafficking through the early and late endosomes followed by receptor degradation.

PAR2 is widely expressed in mammalian tissues with key roles for this receptor in cardiovascular, respiratory, nervous and musculoskeletal systems. This receptor has also been linked to pathological states with aberrant expression and signalling noted in several cancers. In prostate cancer, PAR2 signalling induces migration and proliferation of tumour derived cell lines, while elevated receptor expression has been noted in malignant tissues. Importantly, a role for this receptor has also been suggested in prostate cancer bone metastasis as coexpression of PAR2 and a proteolytic activator has been demonstrated by immunohistochemical analysis.

Based on these data, the primary focus of this project has been on two aspects of PAR2 biology. The first is characterisation of cellular mechanisms that regulate PAR2 signalling and trafficking. The second aspect is the role of this receptor in prostate cancer bone metastasis. In addition, to permit these studies, it was first necessary to evaluate the specificity of the commercially available anti-PAR2 antibodies SAM11, C17, N19 and H99.

ii The evaluation of the four commercially available antibodies was assessed using four techniques: immunoprecipitation; Western blot analysis; immunofluorescence; and flow cytometry. These approaches demonstrated that three of the antibodies efficiently detect ectopically expressed PAR2 by each of these techniques. A significant finding from this study was that N19 was the only antibody able to specifically detect N-glycosylated endogenous PAR2 by Western blot analysis. This analysis was performed on lysates from prostate cancer derived cell lines and tissue derived from wildtype and PAR2 knockout mice. Importantly, further evaluation demonstrated that this antibody also efficiently detects endogenous PAR2 at the cell surface by flow cytometry.

The anti-PAR2 antibody N19 was used to explore the in vitro role of palmitoylation, the post-translational addition of palmitate, in PAR2 signalling, trafficking, cell surface expression and desensitization. Significantly, use of the palmitoylation inhibitor 2-bromopalmitate indicated that palmitate addition is important in trafficking of PAR2 endogenously expressed by prostate cancer cell lines. This was supported by palmitate labelling experiments using two approaches which showed that PAR2 stably expressed by CHO cells is palmitoylated and that palmitoylation occurs on cysteine 361. Another key finding from this study is that palmitoylation is required for optimal PAR2 signalling as Ca2+ flux assays indicated that in response to trypsin agonism, palmitoylation deficient PAR2 is ~9 fold less potent than wildtype receptor with a reduction of about 33% in the maximum signal induced via the mutant receptor. Confocal microscopy, flow cytometry and cell surface biotinylation analyses demonstrated that palmitoylation is required for efficient cell surface expression of PAR2. Importantly, this study also identified that palmitoylation of this receptor within the Golgi apparatus is required for efficient agonist-induced rab11a- mediated trafficking of PAR2 to the cell surface. Interestingly, palmitoylation is also required for receptor desensitization, as agonist-induced β-arrestin recruitment and receptor degradation were markedly reduced in CHO-PAR2-C361A cells compared with CHO-PAR2 cells. Collectively, these data provide new insights on the life cycle of PAR2 and demonstrate that palmitoylation is critical for efficient signalling, trafficking, cell surface localization and degradation of this receptor.

iii This project also evaluated PAR2 residues involved in ligand docking. Although the extracellular loop (ECL)2 of PAR2 is known to be required for agonist-induced signal transduction, the binding pocket for receptor agonists remains to be determined. In silico homology modelling, based on a crystal structure for the prototypical GPCR rhodopsin, and ligand docking were performed to identify PAR2 transmembrane (TM) amino acids potentially involved in agonist binding. These methods identified 12 candidate residues that were mutated to examine the binding site of the PAR2 TL, revealed by trypsin cleavage, as well as of the soluble ligands

2f-LIGRLO-NH2 and GB110, which are both structurally based on the AP SLIGRL-

NH2. Ligand binding was evaluated from the impact of the mutated residues on PAR2-mediated calcium mobilisation. An important finding from these experiments was that mutation of residues Y156 and Y326 significantly reduced 2f-LIGRLO-NH2 and GB110 agonist activity. L307 was also important for GB110 activity. Intriguingly, mutation of PAR2 residues did not alter trypsin-induced signalling to the same extent as for the soluble agonists. The reason for this difference remains to be further examined by in silico and in vitro experimentation and, potentially, crystal structure studies. However, these findings identified the importance of TM domains in PAR2 ligand docking and will enhance the design of both PAR2 agonists and potentially agents to inhibit signalling (antagonists).

The potential importance of PAR2 in prostate cancer bone metastasis was examined using a mouse model. In patients, prostate cancer bone metastases cause bone growth by disrupting bone homeostasis. In an attempt to mimic prostate cancer growth in bone, PAR2 responsive 22Rv1 prostate cancer cells, which form mixed osteoblastic and osteolytic lesions, were injected into the proximal aspect of mouse tibiae. A role for PAR2 was assessed by treating these mice with the recently developed PAR2 antagonist GB88. As controls, animals bearing intra-tibial tumours were also treated with vehicle (olive oil) or the prostate cancer chemotherapeutic docetaxel. The effect of these treatments on bone was examined radiographically and by micro-CT. Consistent with previous studies, 22Rv1 tumours caused osteoblastic periosteal spicule formation and concurrent osteolytic bone loss. Significantly, blockade of PAR2 signalling reduced the osteoblastic and osteolytic phenotype of 22Rv1 tumours in bone. No bone defects were detected in mice treated with docetaxel. These qualitative data will be followed in the future by quantitative micro-CT

iv analysis as well as histology and histomorphometry analysis of already collected tissues. Nonetheless, these preliminary experiments highlight a potential role for PAR2 in prostate cancer growth in bone.

In summary, in vitro studies have defined mechanisms regulating PAR2 activation, downstream signalling and trafficking and in vivo studies point to a potential role for this receptor in prostate cancer bone metastasis. The outcomes of this project are that a greater understanding of the biology of PAR2 may lead to the development of strategies to modulate the function of this receptor in disease.

v TABLE OF CONTENTS

KEYWORDS ...... i ABSTRACT ...... ii TABLE OF CONTENTS ...... vi LIST OF FIGURES ...... x LIST OF TABLES ...... xii LIST OF ABBREVIATIONS ...... xiii STATEMENT OF ORIGINAL AUTHORSHIP ...... xvi PUBLICATIONS ...... xvii ACKNOWLEDGEMENTS ...... xviii CHAPTER 1: INTRODUCTION ...... 1 1.1 G-protein coupled receptors ...... 3 1.2 Protease-activated receptors ...... 3 1.3 Protease-activated receptor-2 ...... 5 1.4 Agonists and antagonists of PAR2 ...... 7 1.4.1 Peptide and peptide mimetic agonists ...... 7 1.4.2 PAR2 antagonists ...... 10 1.5 Molecular Aspects of PAR2 Function ...... 11 1.5.1 Signal transduction ...... 12 1.5.2 Agonist-induced trafficking of PAR2 ...... 15 1.5.2.1 Mechanisms regulating PAR2 desensitisation ...... 15 1.5.2.2 Regulation of PAR2 by palmitoylation ...... 17 1.5.2.3 Mechanisms regulating PAR2 resensitisation ...... 18 1.6 PAR2 in cancer ...... 19 1.6.1 PAR2 in prostate cancer ...... 20 1.6.2 Prostate cancer metastases to bone ...... 21 1.7 PAR2 in the bone microenvironment ...... 22 1.8 Study hypothesis and aims ...... 22 CHAPTER 2: MATERIALS AND METHODS ...... 25 2.1 Materials ...... 27 2.1.1 General reagents ...... 27 2.1.2 Antibodies ...... 27 2.1.3 Enzymes and kits ...... 28 2.1.4 Cell culture reagents and cell lines ...... 28 2.1.5 Animals and housing ...... 28 2.1.6 Buffers ...... 29 2.1.7 Oligonucleotides ...... 30 2.1.8 Vectors ...... 30 2.1.9 Bacterial growth media and plates ...... 30

vi 2.2 Methods ...... 32 2.2.1 Cell culture ...... 32 2.2.2 RNA extraction ...... 32 2.2.3 Reverse transcription ...... 33 2.2.4 Quantitative real time PCR (qRT-PCR) ...... 33 2.2.5 Polymerase Chain Reaction (PCR) and agarose gel electrophoresis ...... 34 2.2.6 Restriction enzyme digests ...... 34 2.2.7 Ligation and bacterial transformation ...... 34 2.2.8 Colony screening, purification of plasmid DNA and DNA sequencing ...... 35 2.2.9 Transfections ...... 36 2.2.10 Cell membrane preparation ...... 36 2.2.11 Analysis of palmitoylation by acyl-biotinyl exchange chemistry ...... 36 2.2.12 Analysis of palmitoylation by metabolic labelling with an alkyne containing palmitate analogue followed by in vitro copper catalysed alkyne-azide cycloaddition (click) chemistry ...... 37 2.2.13 Cell surface biotinylation ...... 39 2.2.14 Receptor endocytosis ...... 39 2.2.15 Western blot analysis ...... 39 2.2.16 Flow cytometry ...... 40 2.2.17 FURA2 intracellular calcium mobilisation assay ...... 40 2.2.18 FLUO3 intracellular calcium mobilisation assay...... 41 2.2.19 Confocal microscopy ...... 41 2.2.20 Live cell Confocal microscopy ...... 42 2.2.21 Mouse intra-tibial injections of prostate cancer 22Rv1 cells ...... 42 2.2.22 Drug administration and tissue processing ...... 43 2.2.23 Micro-computed tomography (CT) of mouse bone ...... 43 CHAPTER 3: EVALUATION OF ANTI-PAR2 ANTIBODIES ...... 45 3.1 Introduction ...... 47 3.2 Methods ...... 51 3.2.1 Expression constructs ...... 51 3.3 Results ...... 52 3.3.1 Three anti-PAR2 antibodies are capable of specifically immunoprecipitating PAR2-FLAG ectopically expressed by CHO cells ...... 52 3.3.2 Ectopically expressed PAR2 is N-glycosylated ...... 52 3.3.3 Three anti-PAR2 antibodies are capable of detecting ectopically expressed PAR2-FLAG by Western blot analysis...... 54 3.3.4 Examination of the ability of three anti-PAR2 antibodies to detect endogenously expressed PAR2 by Western blot analysis ...... 56 3.3.5 Anti-PAR2 antibody N19 but not SAM11 detects an endogenously expressed N- glycosylated protein ...... 60 3.3.6 Further examination of anti-PAR2 antibody N19 by Western blot analysis ...... 62 3.3.7 Anti-PAR2 antibodies detect ectopically expressed PAR2 by confocal microscopy analysis ...... 64 3.3.8 Anti-PAR2 antibodies SAM11 and N19 detected ectopically and endogenously expressed PAR2 by flow cytometry analyses ...... 65 3.4 Discussion ...... 71 CHAPTER 4: THE ROLE OF PALMITOYLATION IN SIGNALLING, CELLULAR TRAFFICKING AND PLASMA MEMBRANE LOCALIZATION OF PAR2 ...... 75 4.1 Introduction ...... 77 4.2 Methods ...... 80 4.2.1 Expression constructs and mutagenesis ...... 80 4.3 Results ...... 81

vii 4.3.1 Palmitoylation is required for efficient cell surface localisation of endogenous PAR2 ...... 81 4.3.2 Generation of CHO cells stably expressing wildtype or C361A mutant PAR2...... 81 4.3.3 PAR2 is palmitoylated on C361 ...... 85 4.3.4 Palmitoylation of PAR2 is required for efficient downstream signal transduction ...... 88 4.3.5 Palmitoylation of PAR2 at C361 is required for efficient cell surface localization ...... 90 4.3.6 PAR2 agonism stimulates palmitate incorporation which occurs during secretory trafficking in pre-medial Golgi vesicles ...... 93 4.3.7 Palmitoylation of PAR2 is required for efficient rab11a-mediated receptor repopulation of the cell surface in response to agonist stimulation ...... 95 4.3.8 Mutation of PAR2 C361 alters agonist-induced recruitment of β-arrestin-1 and β- arrestin-2 ...... 97 4.3.9 PAR2 C361 is required for efficient agonist-induced receptor endocytosis and degradation ...... 97 4.4 Discussion ...... 101 CHAPTER 5: MUTAGENESIS STUDIES TO IDENTIFY PAR2 RESIDUES INVOLVED IN LIGAND ACTIVATED SIGNALLING ...... 107 5.1 Introduction ...... 109 5.2 Methods ...... 112 5.2.1 Expression constructs and mutagenesis ...... 112 5.2.2 Generation of CHO-flpin cells stably transfected with vector, wildtype PAR2 and mutant PAR2 ...... 112 5.2.3 Dose-response, EC50 and relative EC50 (REC) value calculation ...... 112 5.3 Results ...... 113 5.3.1 Computer homology modelling of PAR2 identifies residues potentially involved in ligand docking and receptor activation ...... 113 5.3.2 Characterisation of CHO-flpin cells stably expressing wildtype or mutant PAR2 ...... 117 5.3.3 Comparison of signalling induced via wildtype and mutant PAR2 ...... 122 5.3.3.1 Dose-response curves for vector and wildtype PAR2 ...... 122 5.3.3.2 Dose-response curves for TM domains II and III PAR2 mutants ...... 124 5.3.3.3 Dose-response curves for TM domains V and VI PAR2 mutants ...... 127 5.3.3.4 Dose-response curves for ECL3 and TM domain VII PAR2 mutants ...... 127 5.3.3.5 Comparison of Potency and REC values for trypsin, 2f-LIGRLO-NH2 and GB110 identifies residues involved in PAR2 agonist binding ...... 130 5.4 Discussion ...... 133 CHAPTER 6: EXAMINATION OF A ROLE FOR PAR2 IN PROSTATE CANCER BONE METASTASIS141 6.1 Introduction ...... 143 6.2 Results ...... 150 6.2.1 PAR2 expression in three osteoblastic prostate cancer cell lines ...... 150 6.2.2 PAR2 mediates signal transduction in 22Rv1 but not MDA-PCa-2b or C4-2B osteoblastic prostate cancer cells ...... 150 6.2.2.1 PAR2 activation induces ERK1/2 phosphorylation only in 22Rv1 cells ...... 150 6.2.2.2 PAR2 initiates calcium mobilisation in 22Rv1 prostate cancer cells but not C4- 2B or MDA-PCa-2b cells ...... 153 6.2.2.3 PAR2 antagonist GB88 blocks receptor-mediated calcium mobilisation in 22Rv1 prostate cancer cells ...... 153 6.2.3 Establishment of an animal model to examine the role of PAR2 in bone located prostate cancer tumours ...... 155 6.2.4 Qualitative radiographic analyses: intra-tibial prostatic bone lesions formed by 22Rv1 cells are reduced by GB88 treatment ...... 157 6.2.5 Quantitative micro-CT analysis: osteoblastic and osteolytic bone lesions formed by intra-tibial injection of 22Rv1 prostate cancer cells are reduced by GB88 ...... 161

viii 6.3 Discussion ...... 165 CHAPTER 7: CONCLUSIONS AND FUTURE DIRECTIONS ...... 171 7.1 Summary of findings ...... 173 7.2 Palmitoylation of PAR2 ...... 173 7.3 Anti-PAR2 antibodies ...... 174 7.4 Understanding how PAR2 is activated in prostate cancer bone metastasis ...... 175 7.5 Development of other PAR2 targeting pharmacological agents ...... 176 7.6 Possible dual activation of multiple PAR family members ...... 177 7.7 Possible dual activation of PAR2 and other cell surface receptor systems ...... 179 7.8 Final conclusion ...... 180 BIBLIOGRAPHY ...... 181

ix LIST OF FIGURES

Figure 1.1 Aspects of PAR-mediated signal transduction ...... 4 Figure 1.2 Structural features of PAR2 ...... 6 Figure 1.3 PAR2 desensitization, resensitization and trafficking...... 16 Figure 3.1 Epitopes recognised by anti-PAR2 antibodies...... 49 Figure 3.2 Three anti-PAR2 antibodies immunoprecipitate PAR2-FLAG ectopically expressed in CHO cells ...... 53 Figure 3.3 Ectopically expressed PAR2-FLAG is N-glycosylated ...... 53 Figure 3.4 Anti-PAR2 antibodies SAM11, C17 and N19 detect ectopically expressed PAR2 by Western blot analysis ...... 55 Figure 3.5 Anti-PAR2 antibodies SAM11, C17 and N19 detect ectopically expressed PAR2 in membrane preparations by Western blot analysis...... 57 Figure 3.6 PAR2 expression in prostate cancer cell lines and evaluation of anti-PAR2 antibodies by Western blot analysis...... 59 Figure 3.7 Antibody N19 but not SAM11 detects N-glycosylated PAR2 ...... 61 Figure 3.8 Anti-PAR2 antibody N19 Western blot analysis of wildtype and PAR2 knockout mouse tissue and LMF cells selectively expressing PAR family members...... 63 Figure 3.9 Anti-PAR2 antibodies detect ectopically expressed PAR2 by immunofluorescence and Confocal microscopy analysis ...... 66 Figure 3.10 Antibodies SAM11 and N19 detect ectopic and endogenous PAR2 by flow cytometry analysis...... 69 Figure 4.1 Inhibition of palmitoylation reduces cell surface expression of endogenous PAR2 ...... 83 Figure 4.2 Characterisation of CHO cells stably expressing wildtype or C361A mutant PAR2 ...... 84 Figure 4.3 PAR2 is palmitoylated on cysteine 361 ...... 87 Figure 4.4 Palmitoylation is required for PAR2 to efficiently stimulate calcium mobilisation ...... 89 Figure 4.5 Palmitoylation of PAR2 at C361 is required for efficient cell surface receptor localization ...... 91 Figure 4.6 Quantitative analysis of the effect of palmitoylation on PAR2 cell surface localisation ...... 92 Figure 4.7 PAR2 agonism stimulates palmitate incorporation which occurs during secretory trafficking in pre-medial Golgi vesicles ...... 94 Figure 4.8 Palmitoylation of PAR2 is required for efficient rab11a-mediated repopulation of the cell surface in response to agonist stimulation ...... 96 Figure 4.9 Mutation of PAR2 C361 alters agonist-induced recruitment of β-arrestin-1 delays β- arrestin-2 ...... 98 Figure 4.10 PAR2 C361 is required for efficient agonist-induced receptor endocytosis and degradation ...... 100 Figure 4.11 PAR2 palmitoylation is required for optimal receptor signalling, cellular trafficking and plasma membrane expression ...... 102 Figure 5.1 Computer homology model for PAR2 based on the crystal structure for bovine rhodopsin ...... 114 Figure 5.2 Diagram of PAR2 highlighting residues identified by homology modelling to potentially interact with PAR2 agonists ...... 116

x Figure 5.3 Protein expression of wildtype PAR2-FLAG and mutant PAR2-FLAG in stably expressing CHO-flpin cells ...... 118 Figure 5.4 Cell surface expression of mutant PAR2-FLAG compared with wildtype PAR2- FLAG in stably expressing CHO-flpin cells ...... 120

Figure 5.5 Dose-response curves of trypsin, 2f-LIGRLO-NH2 and GB110 on CHO-flpin cells stably transfected with vector or wildtype PAR2-FLAG ...... 123

Figure 5.8 Dose-response curves for trypsin, 2f-LIGRLO-NH2 and GB110 induced by treatment of CHO-flpin cells stably transfected with vector, wildtype PAR2-FLAG or PAR2-FLAG with ECL3 and TM VII mutants ...... 129 Figure 5.9 Three dimensional representation of PAR2 residues 156, 307 and 326 and structural features of 2f-LIGRLO-NH2 and GB110 ...... 136 Figure 6.1 Bone structure and site for intra-tibial injection of prostate cancer cells in mouse ...... 145 Figure 6.2 PAR2 expression in three osteoblastic prostate cancer cell lines...... 151 Figure 6.3 PAR2 activation induces ERK1/2 phosphorylation in 22Rv1 prostate cancer cells but not C4-2B or MDA-PCa-2b prostate cancer cell lines...... 152 Figure 6.4 PAR2 activation induces calcium mobilisation in 22Rv1 but not C4-2B or MDA- PCa-2b prostate cancer cells and signalling is blocked by the PAR2 antagonist GB88 ..... 154 Figure 6.5 Prostate cancer in bone metastasis experiment timeline and treatment course ...... 158 Figure 6.6 The PAR2 antagonist GB88 reduces qualitative indices of 22Rv1 prostatic bone lesions ...... 160 Figure 6.7 The PAR2 antagonist GB88 reduces the osteoblastic and osteolytic phenotype of 22Rv1 prostate cancer bone lesions ...... 163 Figure 7.1 Possible activation and modulation of PAR2 and other cell surface receptors ...... 178

xi LIST OF TABLES

Table 2.1 Oligonucleotide primer sequences used in cloning, site directed mutagenesis and qRT-PCR experiments ...... 31 Table 3.1 Summary of the examined anti-PAR2 antibodies...... 48 Table 5.1 PAR2 residues identified by computer aided homology modelling and ligand docking as potentially involved in ligand binding ...... 115 Table 5.2 Comparison of cell surface wildtype and mutant PAR2-FLAG stably expressed by CHO-flpin cells ...... 121

Table 5.3 Comparison of trypsin, 2f-LIGRLO-NH2 and GB110 potency values from dose- response curves of wildtype and mutant PAR2 stably expressing cells ...... 131 Table 6.1 Summary of prostate cancer cell lines and observed phenotype ...... 147 Table 6.2 Summary of animals used in in vivo prostate cancer bone metastasis model ...... 156

xii LIST OF ABBREVIATIONS

µg microgram µM micromolar µl microlitre µm micrometer 2BP 2-bromopalmitate 17-ODYA 17-octadecynoic acid ABE acyl-biotinyl exchange AP agonist peptide BCA bicinchoninic acid BFA Brefeldin A BSA bovine serum albumin oC degrees Celsius CAV1 caveolin-1 CCP clathrin coated pit CDCP1 cub domain containing protein 1 cDNA complementary DNA Cha cyclohexlalanine CHO chinese hamster ovary CIN chronophin CXCL chemokine (C-X-C motif) ligand DC dendritic cell DMEM Dulbecco’s modified eagle’s medium DMSO dimethylsulphoxide DNA deoxynucleic acid dNTP deoxynucleotide triphosphate EC50 half maximal effective concentration ECL extracellular loop EDTA ethylene diaminetetraacetate EE early endosome EGF(R) epidermal growth factor (receptor) ER endoplasmic reticulum ERK extracellular regulated kinase FAK focal adhesion kinase FCS fetal calf serum Flp flippase FRT flp recombination target FVIIa factor VIIa FXa factor Xa g gravitational force GDP guanosine diphosphate GFP green fluorescent protein GPCR G-protein coupled receptor GRK GPCR kinase GTP guanosine triphosphate h hour HBSS Hank’s buffered saline solution

xiii HER2 human epidermal growth factor receptor 2 HGF hepatocyte growth factor HRS hepatocyte growth factor-regulated tyrosine kinase substrate HSC haemopoietic stem cell ICL intracellular loop IGF insulin-like growth factor IgG immunoglobulin IL interleukin IKK inhibitory kinase kappaB IP3 inositol 1,4,5-triphosphate i.p. intraperitoneal injection I/R ischemia and reperfusion Jab1 Jun activation domain-binding protein 1 JNK c-Jun N-terminal kinase kDa kilodaltons kg kilogram KLK kallikrein KO knockout kVp peak kilovolts LB Luria Bertani LE late endosome LMF lung murine fibroblasts M molar MAPK mitogen-activated protein kinase MeSNA mercaptoethanesulfonic acid sodium salt MFI mean fluorescence intensity mg milligram mg/kg milligram per kilogram min minutes miRNA micro RNA ml millilitre MMP matrix metalloproteinase MMTV mouse mammary tumor virus MON monensin mRNA messenger RNA ms millisecond MSC mesenchymal stem cell NEM N-ethylmaleimide NF-κB nuclear factor-kappaB ng nanogram NOC nocodozole nm nanometre PAR protease-activated receptor PAT palmitoyl acyl transferase PBS phosphate buffered saline PCR polymerase chain reaction PH pulmonary hypertension PKC protein kinase C p.o. oral gavage PyMT polyoma middle T-antigen

xiv qRT-PCR quantitative reverse transcription PCR RANK(L) receptor for activation of nuclear factor kappaB (ligand) RNA ribonucleic acid rpm revolutions per minute RT room temperature SDS sodium dodecyl sulphate SDS-PAGE SDS polyacrylamide gel electrophoresis sec seconds shRNA short hairpin RNA TAE Tris-acetate EDTA TBTA t-butanol, tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl] amine TBS-T Tris buffered saline-Tween 20 TCEP Tris(2-carboxyethyl)phosphine TF tissue factor TGFβ transforming growth factor β TKI tyrosine kinase inhibitor TL tethered ligand TLR toll-like receptor TM transmembrane uPA(R) urokinase-type plasminogen activator (receptor) UV ultraviolet VEGF vascular endothelial growth factor WT wild type

xv STATEMENT OF ORIGINAL AUTHORSHIP

The work contained in this thesis has not been previously submitted to meet requirements for an award at this or any other higher education institution. To the best of my knowledge and belief, the thesis contains no material previously published or written by another person except where due reference is made.

Signature: ______Mark Adams

Date: ______

xvi PUBLICATIONS

Publications arising during this PhD program of study

1. Ramsay AJ, Reid JC, Adams MN, Samaratunga H, Dong Y, Clements JA, Hooper JD. Prostatic trypsin-like kallikrein-related peptidases (KLKs) and other prostate-expressed tryptic proteinases as regulators of signalling via proteinase-activated receptors (PARs). Biol Chem. 2008; 389(6): 653-668.

2. He Y, Wortmann A, Burke LJ, Reid JC, Adams MN, Abdul-Jabbar I, Quigley JP, Leduc R, Kirchhofer D, Hooper JD. Proteoloysis-induced N-terminal ectodomain shedding of the integral membrane glycoprotein CUB domain- containing protein 1 (CDCP1) is accompanied by tyrosine phosphorylation of its C-terminal domain and recruitment of Src and PKCδ. J. Biol. Chem. 2010; 285(34): 26162-26173.

3. Adams MN, Ramachandran R, Yau MK, Suen JY, Fairlie DP, Hollenberg MD, Hooper JD. Structure, function and pathophysiology of protease activated receptors. Pharmacol. Ther. 2011; 130(3): 248-282.

4. Adams MN, Christensen ME, He Y, Waterhouse NJ, Hooper JD. The role of palmitoylation in signalling, cellular trafficking and plasma membrane localisation of protease-activated receptor-2. PLoS One. 2011; 6(11):e28018.

xvii ACKNOWLEDGEMENTS

I would like to sincerely thank my principal supervisor Associate Professor John Hooper for giving me the opportunity to work in his laboratory (first as an Honours student and now as a PhD candidate) and for taking a significant amount of time over the years to support, encourage and mentor me. I would also like to thank my associate supervisors Professor Judith Clements and Professor David Fairlie for their assistance throughout my PhD.

I acknowledge the financial support of the Federal government for my PhD scholarship (APA Award) and the Mater Medical Research Institute for my top-up award. I would also like to acknowledge financial support from the Mater Medical Research Institute, Prostate Cancer Foundation Australia, Queensland Cancer Council and Queensland University of Technology for travel awards throughout my PhD enabling me to attend national and international conferences and laboratories.

Thank-you to Dr Jacky Suen, Dr Laura Gregory, Dr Hui He, Mrs Deborah Roche, Dr Yaowu He and Dr Roland Steck for teaching me or contributing to experiments in this thesis. Thank-you also to past and present members of the Hooper laboratory and the Hormone Dependent Cancer Program at Queensland University of Technology. Special thanks to Dr Andrew Ramsay (for being a model PhD student), Dr Yaowu He (for teaching me to dig for gold), Dr Lez Burke (for keeping me honest) and Mr Carson Stephens (for the giggles).

Finally, I am very grateful to my wife Danielle, for having an amazing level of patience and for supporting and encouraging me throughout my PhD. I am also very grateful to Mum, Dad, Nathan, Brent and other family and friends for their support and encouragement.

xviii

Chapter 1:

Introduction

2 1.1 G-protein coupled receptors G-protein coupled receptors (GPCRs) mediate homeostasis, physiological functions and progression of several diseases by transducing extracellular stimuli across the plasma membrane to induce a myriad of signal transduction pathways (Oldham and Hamm, 2008). Signal transduction in response to external stimuli most commonly occurs via coupling to specific G protein subunits (designated Gs, Gi/o, Gq/11, G12/13 (Simon et al., 1991)), although signalling via non-G protein dependent pathways has been reported (Sun et al., 2007). Elucidation of the mechanisms controlling these responses has assisted in the design of pharmacological agents to antagonise or agonise GPCRs with these drugs accounting for approximately 20-25% of available clinical pharmaceutical agents (Landry and Gies, 2008; Overington et al., 2006).

1.2 Protease-activated receptors Discovered in the 1990s, protease-activated receptors (PARs) are a sub-family of GPCRs comprising four members designated PAR1, PAR2, PAR3 and PAR4 (Hollenberg and Compton, 2002). In contrast to the wider GPCR superfamily, PARs are not activated in vivo by binding to a soluble ligand but, instead, are triggered by proteases which cleave extracellularly within the PAR amino terminus. PAR activation via proteolytic cleavage results in irreversible removal of the amino terminal pro-peptide and unmasking of a neoepitope termed the tethered ligand (TL; (Macfarlane et al., 2001; Vu et al., 1991) (Figure 1.1). These receptors are almost exclusively activated by trypsin fold serine proteases that have specificity for cleavage after Arg or Lys residues. In particular, PAR1, PAR3 and PAR4 are each cleaved by thrombin, while PAR2 and PAR4 are cleaved by trypsin. The TL is thought to interact intramolecularly with extracellular loop (ECL)2, which is glycoslyated on asparagine residues for PAR1 and PAR2 (Figure 1.2), and possibly also with other regions of the PAR, to initiate a conformational change in the receptor and downstream intracellular signaling (Adams et al., 2011; Macfarlane et al., 2001). By these mechanisms, PARs function as cell surface sensors of extracellular proteases, contributing to regulation of homeostasis, as well as to the dysfunctional responses that result in disease.

Figure 1.1 Aspects of PAR-mediated signal transduction

As shown in the left panel proteolysis of the PAR amino terminal, by protteases such as thrombin and trypsin, within a defined activation site reveals a neo-amino terminal termed the TL binds intramolecularly within the receptor enabling an allosteric change to elicit signal transduction from coupled heterotrimeric G-proteins (αβγ subunits). The PARs are known to couple with several Gα subunits inclluuding Gαi,

Gα12/13 and Gαq to activate downstream signaling messengers to induce a myriad of cellular and physiological effects. In addition, PAR2 is able to signal through the scaffolding protein β-arrestin (βarr) which also functions to desensitize PAR2 by uncoupling the receptor from G-proteinss. Although β-arrestin does not contribute to PAR1 signalling, it does participate in desensitization of this receptor. As shown in the right panel proteolytic cleavage of the PAR amino terminal outside the activation motif by proteases such as cathepsin G leeads to receptor inactivation or “diisarming”.

4 1.3 Protease-activated receptor-2 Following the identification of thrombin activated PAR1, the likelihood of another protease-activated receptor was postulated (Macfarlane et al., 2001; Scarborough et al., 1992; Vu et al., 1991). Initial cloning from a mouse genomic library using hybridisation probes homologous to regions of the bovine substance K receptor (Nystedt et al., 1994), identified a 3.7 kilobase nucleotide fragment with an open reading frame of 1191 nucleotides encoding PAR2. The human PAR2 gene, isolated from a genomic DNA library using a hybrisation probe derived from a portion of the 3’ mouse PAR2 exon (Nystedt et al., 1995), is localised to chromosome 5q13 and, like the PAR1 gene, consists of two exons (Schmidt et al., 1997).

As shown in Figure 1.2, PAR2 consists of 397 amino acids and contains seven putative transmembrane (TM) helices, an extracellular amino terminal domain encompassing a signal peptide of 25 residues and a pro-domain of 11 amino acids, three intracellular loops (ICL1-3), three extracellular loops (ECL1-3) and an intracellular carboxy terminal domain of 50 residues. As well as connecting TM4 and TM5, ECL2 also makes a crucial disulfide bond with TM3 thereby restricting access to the central cavity of PARs. These disulfide-forming cysteine residues, conserved amongst GPCRs, contribute to receptor structural stability (Hamm, 2001). Consensus motifs for post-translational modifications such as N-glycosylation, ubiquitination, phosphorylation and palmitoylation are also shown in Figure 1.2.

Human PAR2 is activated by cleavage after R36 revealing the TL sequence S37LIGKV ((Nystedt et al., 1995) Figure 1.2). In addition to being activated by trypsin (Nystedt et al., 1994), PAR2 is activated by mast cell tryptase (Molino et al., 1997), directly by TF/FVIIa and indirectly by TF/FVIIa-generated FXa (Camerer et al., 2000), acrosin (Smith et al., 2000), matriptase/MT-SP1 (Takeuchi et al., 2000), human airway trypsin-like protease (HAT (Miki et al., 2003)), trypsin IV (Cottrell et al., 2004), Granzyme A (Hansen et al., 2005), TMPRSS2 (Wilson et al., 2005), KLK2 (Mize et al., 2008), KLK4 (Mize et al., 2008; Ramsay et al., 2008a), KLK5, KLK6 and KLK14 (Oikonomopoulou et al., 2006) and the mold allergen serine protease Pen c 13 (Chiu et al., 2007). In a challenge to the dogma that PARs function exclusively as cell surface receptors for trypsin-like serine proteases and that cleavage outside the activation site recognized by these enzymes results in receptor

Figure 1.2 Structural features of PAR2

Sequence is shown for the signal peptide (green lettering) and pro-peptide (orange lettering) regions, the mature amino terminus, extracellular loop (ECL) regions 1, 2 and 3, intracellular loop (ICL) regions 1, 2 and 3, and the carboxy terminal domain of PAR2. This is overlayed on the seven transmembrane (TM) domain structure characteristic of GPCRs. Underlined amino terminal residues represent the TL of PAR2. Cysteine which forms a disulfide linkage between TM3 and ECL2 is highlighted in purple. Consensus sites for N-glycosylation are highlighted yellow. Red highlighted residues denote palmitoylation sites for PAR2. Highlighted grey and blue residues (Lysine (K), Serine (S) and Threonine (T)) represent post-translational modification sites for ubiquitination and phosphorylation.

6 disarming, it has also been reported that the bacterial cysteine gingipains (Uehara et al., 2005), house dust mite cysteine protease Der P1 (Adam et al., 2006), the mite serine proteases, Der P2 and Der P3 (Sun et al., 2001) and chitinase from Streptomyces griseus (Hong et al., 2008) induce signaling via PAR2.

Many proteases also act as negative regulators, preventing downstream signaling by proteolytically inactivating (or dis-arming; Figure 1.1) the PARs via one of two mechanisms. First, by cleavage of the receptor within the extracellular amino terminal domain C-terminal of the activation site to disarm the receptor by removal of the TL and, second, by cleaving elsewhere within the receptor to disable the receptor (Hansen et al., 2008; Kawabata et al., 1999; Ramachandran and Hollenberg, 2008). Examples include plasmin, protease 3 and calpain (Loew et al., 2000) and elastase and cathepsin G (Dulon et al., 2003; Dulon et al., 2005).

1.4 Agonists and antagonists of PAR2 PAR2 activation via proteolytic cleavage results in irreversible removal of the amino terminal pro-peptide and unmasking of the TL ((Macfarlane et al., 2001; Vu et al., 1991) Figure 1.1 and 1.2). Soon after PAR1 was cloned, it was established that, in addition to activation by intramolecular binding of the proteolytically exposed TL, this receptor could also be activated by a short synthetic peptide mimicking the sequence of the revealed TL (Scarborough et al., 1992; Vu et al., 1991). It is now known that PAR subtype-selective short synthetic peptides, called activating peptides (APs), can mimic the TL to activate downstream signalling. These short peptides have been valuable probes for understanding the roles that PARs play in physiology and disease (Macfarlane et al., 2001; Ramachandran and Hollenberg, 2008). This section summarises the progress that has been made in developing selective and potent reagents that can activate (agonism) or inhibit (antagonism) PAR2 in vitro and in vivo.

1.4.1 Peptide and peptide mimetic agonists PAR2 APs have been developed from the human (SLIGKV) and mouse (SLIGRL) receptors (Macfarlane et al., 2001). Interestingly, the AP derived from the murine receptor has greater potency (EC50 ~5 µM) than the human sequence (EC50 ~12 µM), but both sequences are selective for PAR2 over PAR1 (Maryanoff et al., 2001).

7 Structure-activity studies using synthetic PAR2 APs have identified key components that are important for potency and selectivity. These have shown that a minimum of

5 amino acids (e.g. SLIGK-NH2) are required to activate PAR2 (Hollenberg et al., 1996). However, pentapeptides of PAR2 AP show low µM agonist potency via PAR2 compared with the hexapeptides (Hollenberg et al., 1996). Moreover, substitution of serine at the N-terminal with basic, acidic or large hydrophobic residues also reduces agonist activity (Maryanoff et al., 2001). In addition, although introduction of charged or small amino acids (e.g. Gly, Ala) at Leu2 abolished activity, cyclohexlalanine (Cha) or Phe gave similar potency, but Phe was not selective for PAR2 over PAR1. Ile and Cha at position 3 gave optimal agonist activity, while substitution with Ala or Phe reduced the potency. At position 4, Gly could be replaced with Ala and Cha, but Cha also activated PAR1. Instead of having Arg or positively charged residues at position 5, it was surprising that this position also tolerated large hydrophobic groups, suggesting that a hydrophobic pocket with an acidic residue might be located at the corresponding position in the receptor. It was also demonstrated that Leu at position 6 of the mouse AP is less important, as substitution with other amino acids, including small/large aliphatic chain, aromatic, charged residues and amides, did not show a significant change in potency (Maryanoff et al., 2001).

Although these results support a requirement for only 5 amino acids in a PAR2 agonist, a significant decrease in potency was observed when Leu2, Ile3 or Gly4 was replaced with a D-amino acid, indicating that stereochemistry is also important for binding to the receptor (Maryanoff et al., 2001). Interestingly, Kawabata et al reported that a 2-furoyl moiety in place of serine at the N-terminal could substantially increase PAR2 AP potency by about 10-fold (2004). In addition, 2- furoyl-LIGRLO-NH2 (2f-LIGRLO-NH2) was 10-15 times more potent than

SLIGRL-NH2 at inducing intracellular calcium release in both cultured human and rat PAR2-expressing cells, some 300 times more potent in a murine-isolated blood vessel assay, and it was selective for PAR2 over PAR1 (McGuire et al., 2002; McGuire et al., 2004).

More recently, it has been reported that addition of a seventh and even eighth residue, such as isoleucine or leucine to the C-terminus of SLIGRL-NH2, increased

8 potency by ~6-8 fold, while replacing the sixth residue leucine with aromatic groups increased potency further (Barry et al., 2007; Devlin et al., 2007). In addition, replacement of the sixth residue of the heptapeptide with aromatic amino acids and peptides with 4-nitrophenylalanine, 3,4-dichlorophenylalanine or 3- benzothienylalanine at the sixth position generated APs with PAR2 responses equipotent with 2f-LIGRLO-NH2. Interestingly, structure-activity relationships suggested that other heterocycles and aromatic groups could be used as alternatives for 2-furoyl without much loss of potency (Barry et al., 2007), and that these structures were better than aliphatic groups, suggesting possible π-π interactions of the aromatic/heterocyclic ring with PAR2 (Barry et al., 2007).

In summary, these key potency and selectivity findings for PAR2 APs indicate that: S1 or small replacements like a small heterocycle enhance PAR2 AP potency; L2 is optimally replaced with Cha, while aromatic residues give similar potency but with diminished selectivity for PAR2 over PAR1; either Ile or Cha at the third position is optimal; G4 and L6 are not crucial for potency and can be substituted; R5 is possibly binding to a negatively-charged residue in PAR2, although substitution with large hydrophobic groups does not significantly reduce agonist potency, suggesting that this may be for affinity rather than agonist activity per se; addition of a seventh residue to the C-terminal may or may not improve potency, depending on the moiety at N-terminal.

Although some potent PAR2 APs have been developed, typically poor bioavailability of these molecules is still a major limitation for in vivo studies. Therefore, recent studies have attempted to develop non-peptidic PAR2 agonists. To improve drug-like properties, a new class of non-peptidic PAR2 agonists has been developed from hexapeptides through a heterocyclic replacement for the N-terminal serine and truncation from the C-terminal appending C-terminal non-peptidic fragments (Barry et al., 2010; Barry et al., 2007; Suen et al., 2011). Using intracellular Ca2+ release as a measure of PAR2 activation, agonist GB110 (compound 1, EC50 0.28 µM) selectively activated PAR2 on a range of cell lines (e.g. HT29, HEK293, HUVEC, A549, Panc1, MDA-MB-231, MKN1 and MKN45). It was equipotent with the best synthetic peptide agonists reported for PAR2 (compare with 2f-LIGRLO-NH2 EC50 0.24 µM) as well as being selective for PAR2 over PAR1, having no effect on PAR1

9 activation by thrombin or on intracellular Ca2+ release in cells desensitized to PAR2 activation (Barry et al., 2010; Suen et al., 2011).

O O O H N N N N H H N O O NH2

(1)

1.4.2 PAR2 antagonists Once it was established that PAR2 regulates various cellular responses, it became clear that antagonists of this receptor could potentially be useful in clinical settings. However, progress in development of potent PAR2 antagonists has so far been elusive. The first reported PAR2 antagonists, peptides FSLLRY-NH2 and LSIGRL-

NH2, block trypsin but not PAR2 AP activation of PAR2 (Al-Ani et al., 2002a). Plevin and colleagues have also reported a number of peptide antagonists of PAR2, inhibiting calcium signaling and nuclear factor-κB (NF-κB) activity induced by peptide agonists. The compounds were able to compete with a high affinity PAR2 radioligand with low micro molar Ki values (Kanke et al., 2009). One of the compounds, K14585 (Goh et al., 2009) inhibited intracellular calcium release induced by a PAR2 AP but failed to inhibit ERK signaling. The compound also stimulated p38 MAPK signaling through PAR2 that was independent of Gαq coupling. K14585 pretreatment inhibited p65 NF-κB phosphorylation and DNA binding but also stimulated a NF-κB reporter to the same level as PAR2 AP. Small molecule PAR2 antagonists have also been reported. For example, antagonist ENMD-1068, is reported to attenuate inflammation in a mouse model of arthritis (Kelso et al., 2006). However, although ENMD-1068 is selective for PAR2 over PAR1 in vitro, a practical issue surrounding the use of this antagonist is the fact that very large doses in mM concentrations are required to elicit an effect in vivo (Kelso et al., 2006; Lohman et al., 2012).

A recently developed novel PAR2 antagonist has shown promise. GB88 (compound 2) is a novel PAR2 antagonist derived from the PAR2 agonist GB110 (compound 1), with the same N-terminal but with a bulky C-terminal spiroindanepiperidine (Suen et al., 2011). It is the first PAR2 antagonist to reversibly inhibit PAR2 activation by

10 endogenous (e.g. trypsin), synthetic peptide (e.g. 2f-LIGRLO-NH2) and nonpeptide

(1) agonists at low concentrations (IC50 ~2 µM). It is also a surmountable

(competitive) antagonist of 2fLIGRLO-NH2, but an insurmountable antagonist of both trypsin and GB110, making it a valuable PAR2 silencing tool for mechanistic studies (Suen et al., 2011). Furthermore, GB88 is serum stable and orally active and has been investigated for PAR2 modulation in vivo in a range of animal models of human inflammatory and proliferative diseases. Given by oral gavage (p.o.) at 5 mg/kg this antagonist inhibits rat paw oedema illicited by PAR2 agonists but not PAR1 agonists, and substantially inhibits collagen-induced arthritis and experimentally-induced colitis in rats (Lohman et al., 2012; Suen et al., 2011). This compound may be of benefit in further characterizing PAR2 as a therapeutic target in inflammatory and other diseases.

O O H N N N H N O O

(2)

In summary, PAR2 is an attractive target for the development of agonists and antagonists. An aim of this project is to characterise the ligand binding pocket for PAR2 to aid future development of agonists and antagonists that target this receptor with greater potency.

1.5 Molecular Aspects of PAR2 Function PAR2-mediated signal transduction occurs not only via ‘conventional’ intracellular coupling to specific G protein subunits but also via G-protein independent pathways involving β-arrestin scaffolding (Defea, 2008; Sun et al., 2007). It is now apparent that these pathways are modulated at multiple levels including receptor translocation to, and localisation within, the plasma membrane, the properties of the agonist, molecular events following receptor activation including signal termination, internalisation and degradation, and mechanisms controlling recycling of uncleaved PAR to the plasma membrane (Adams et al., 2011). In this section the features of these signal transduction pathways including mechanisms modulating signaling and those governing receptor trafficking are described.

1.5.1 Signal transduction Signal transduction mediated by GPCRs such as PAR2 is sustained intracellularly via heterotrimeric G proteins consisting of an α-subunit (Gα) associated with a Gβγ dimer that couple directly with GPCR intracellular domains. Ligand activation of GPCRs induces GDP exchange for GTP within Gα and disassociation from the Gβγ dimer to evoke signal transduction by interacting with and regulating downstream effectors. It is apparent that downstream of the cell surface receptor, PAR-mediated signaling is similar to that of other GPCRs. For example, PAR2 induces classical signal transduction pathways by coupling directly with G proteins Gαq, Gαi and

Gα12/13 (Macfarlane et al., 2001; Soh et al., 2010; Steinhoff et al., 2005).

PAR2-initiated G protein-mediated signaling activates a complex array of downstream pathways that modulate cell growth, differentiation, inflammation and receptor transactivation (Figure 1.1). Coupling with Gαq activates phospholipase C which generates diacyl glycerol and inositol 1,4,5-triphosphate (IP3) from phophatidylinositol 4,5-bisphophate to mobilize Ca2+ from the endoplasmic reticulum into the cytosol. PAR2 activation generates IP3 and mobilizes intracellular Ca2+ stores in a range of cell types including platelets (Benka et al., 1995), kidney and intestinal epithelial cells (Bohm et al., 1996a; Bohm et al., 1996b), rat hippocampal neurons, astrocytes (Bushell et al., 2006), myocytes and fibroblasts (Ossovskaya and Bunnett, 2004).

Other effector pathways downstream of PAR2 activation are (1) the mitogen- activated protein kinase (MAPK) pathway that links these receptors with proliferative cellular responses, and (2) the NF-κB pathway which links PAR2 with inflammation. G protein dependent PAR2-mediated activation of extracellular regulated kinases 1/2 (ERK1/2) through the MAPK signaling cascade has been demonstrated in many cell types including smooth muscle and kidney epithelial cells (Belham et al., 1996; Yu et al., 1997), astrocytes (Wang et al., 2002) and dermal fibroblasts (Wang et al., 2006). In smooth muscle cells, trypsin and PAR2 APs stimulate components of the MAPK pathway, including MAPK kinase 1 and 2 and c- Raf-1, but do not activate the MAPK homologue p38 or c-Jun N-terminal kinase (JNK (Belham et al., 1996)). In contrast, in other cell-types such as keratinocytes and

12 primary cardiomyocytes, PAR2-mediated p38 MAPK and JNK activation is robust (Kanke et al., 2001; Sabri et al., 2000).

The NF-κB family of transcription factors is recognized as a primary regulator of pro-inflammatory mediator genes (Lenardo and Baltimore, 1989). Although not completely understood, a growing literature suggests important roles for PAR2 in this pathway. In a number of cell types, PAR2 agonists stimulate the DNA binding activity of NF-κB and activation of upstream inhibitory kinase κB (IKK) α and IKKβ (Bretschneider et al., 1999; Buddenkotte et al., 2005; Kanke et al., 2001; Shpacovitch et al., 2002). It appears from studies using selective protein kinase C (PKC) isoform inhibitors that PAR signaling via the NF-κB/IKK pathway is PKC dependent (Kanke et al., 2001). Interestingly, both PKC dependent and independent mechanisms regulate PAR2-mediated signaling via p38 and JNK (Kanke et al., 2001) and it has been shown that PAR2-mediated NF-κB p65 subunit phosphorylation is also PKC and Gαq/11 dependent (Goon Goh et al., 2008). Downstream of these events, NF-κB activation by PAR2 stimulation with trypsin and PAR2 AP mediates expression and secretion of many inflammatory mediators including chemokines CXCL1 and CXCL8 (Niu et al., 2008a; Rudack et al., 2007). NF-κB activation has also been demonstrated through the cooperative action of PAR2 and Toll-like receptor-4 (TLR4 (Nhu et al., 2010; Rallabhandi et al., 2008)). The ectopic co-expression of both receptors resulted in increased PAR2 AP induced NF-kB signaling culminating in enhanced interleukin (IL)-8 production and synergism between these receptors appears dependent on a direct PAR2 and TLR4 physical interaction (Rallabhandi et al., 2008). Interestingly, receptor cooperativity between PAR2 and the TLR4 related receptors, TLR2 and TLR3, resulting in NF-κB activation, has been shown in mucosal epithelial cells (Nhu et al., 2010).

It is also apparent that PAR2 can transactivate other receptor systems. For example, activation of PAR2 transactivates EGFR in the gastric cancer cell lines AGS and MKN28 (Caruso et al., 2006) and in human intestinal epithelial cells (Jarry et al., 2007; van der Merwe et al., 2008). Interestingly, PAR2-mediated EGFR transactivation in HaCaT keratinocytes induces ERK1/2 phosphorylation and leads to secretion of transforming growth factor-beta-1 (TGF-β1 (Rattenholl et al., 2007)).

13 Other G protein dependent pathways are also emerging through which PAR2 initiated signals are transduced. For example, activation of PAR2 stimulates RhoA in respiratory epithelial (Yagi et al., 2006), umbilical vein endothelial (Klarenbach et al., 2003) and prostate cancer cells (Greenberg et al., 2003) to induce FAK tyrosine phosphorylation and formation of stress fibers, contributing to cell adhesion and migration (Steinhoff et al., 2005). Particularly in endothelial cells, PAR2 AP stimulation concomitantly induced RhoA activation and transient phosphorylation of the Rac effector, p21-activated kinase, whereas treatment with PAR1 APs oppose this response by inhibiting Rac (Vouret-Craviari et al., 2003). PAR2 is also thought to regulate downstream transcriptional activation through cut-like homeobox-1 (CUX1), a transcription factor involved in tissue homeostatis (Wilson et al., 2009). Treatment of both epithelial and fibroblastic cells with trypsin induced a rapid increase in the DNA binding activity of CUX1 which specifically enhanced the expression of IL-1α, matrix metalloproteinase (MMP)-10 and cyclo-oxygenase-2 downstream of PAR2 (Nepveu, 2001; Sansregret and Nepveu, 2008).

In addition to these G protein dependent signaling events, it is clear that G protein independent signal transduction pathways are also initiated by proteolytic activation of PARs. The most completely characterized of these pathways involves PAR2 interactions with the scaffolding proteins β-arrestin 1 and 2, which bind to the phosphorylated carboxyl terminal of the receptor. In addition to mediating G protein independent signaling, these interactions uncouple PAR2 from downstream heterotrimeric G protein signal transduction and link the receptor with cellular internalisation machinery (Defea, 2008; DeFea et al., 2000; Dery et al., 1999). By this mechanism β-arrestins participate in G-protein signal termination and also mediate downstream signaling by acting as scaffolds permitting complex formation with multiple components including ERK1/2 to modulate G protein independent signal transduction. Interaction of PAR2 with β-arrestin 1 and 2 (Jacob et al., 2005b; Stalheim et al., 2005) prolongs intracellular receptor mediated signaling via ERK1/2 activation (Defea, 2008; DeFea et al., 2000; Ge et al., 2003; Stalheim et al., 2005). Interestingly, PAR2-mediated ERK1/2 activation is also possible in the absence (Stalheim et al., 2005) or prevention of β-arrestin binding (DeFea et al., 2000) but this activation is not prolonged and activity rapidly declines. Of pathophysiological relevance, prolonged ERK1/2 activation mediated by PAR2 interaction with β-

14 arrestin-1 and -2 elicits a direct cellular response by regulating intestinal paracellular permeability (Jacob et al., 2005b). In addition, in MDA-MB-231 breast cancer cell activation of this pathway requires β-arrestin1 and 2 induction of actin cytoskeletal reorganisation in conjunction with polarised pseudopodia extension and chemotaxis (Ge et al., 2003). PAR2 also mediates G-protein independent chemotaxis by another β-arrestin dependent pathway involving β-arrestin complex formation with cofilin and its activator chronophin (CIN) to modulate actin assembly within membrane protrusions (Zoudilova et al., 2007; Zoudilova et al., 2010).

In addition to β-arrestin mediated signaling, a recent report has identified Jun activation domain-binding protein 1 (Jab1), a component of the COP9 signalosome complex, as a PAR2 interacting and signal modulating protein. Post PAR2 activation, Jab1 dissociates from the receptor to bind and activate c-Jun and further activate the transcription factor activator protein-1. This novel signal transduction pathway highlights the diversity of cellular processes mediated by PAR2 and further links this receptor with regulation of gene expression (Luo et al., 2006).

1.5.2 Agonist-induced trafficking of PAR2 Following signal tranduction, GPCR desensitisation is generally induced by rapid phosphorylation, usually via GPCR kinases (GRKs (Krupnick and Benovic, 1998)), followed by arrestin binding, promoting G protein uncoupling. Receptor endocytosis ensues, removing the activated receptor from signaling effectors and the plasma membrane. For non-PAR GPCRs the internalized receptor is either recycled to repopulate the cell surface or targeted for lysosomal degradation (Ritter and Hall, 2009). However for PAR2, receptor recycling is prevented by the proteolysis inherent in the activation mechanism. In this section, the known mechanisms controlling PAR2 processing are described.

1.5.2.1 Mechanisms regulating PAR2 desensitisation Mechanisms regulating PAR2 signal termination, trafficking and degradation are summarized in Figure 1.3. Unlike unactivated PAR1 (Hein et al., 1994; Paing et al., 2006; Shapiro and Coughlin, 1998), unactivated PAR2 does not constitutively internalize. Also, proteolytically activated PAR2 is initially phosphorylated not by

Figure 1.3 PAR2 desensitization, resensitization and trafficking.

PAR2 internalises solely through agonist-triggered mechanisms. Activated PAR2 is phosphorylated and binds β-arrestin-1 and -2 (βarr) and internalises via clathrin coated pits (CCP) in a dynamin dependent mechanism. Endocytosed PAR2 is trafficked to the early endosome (EE) in a ubiquitination dependent process also requiring the GTPase rab5a. Continued sorting of PAR2 also requires hepatocyte growth factor-regulated tyrosine kinase substrate (HRS). PAR2 trafficked through the late endosome (LE) is deubiqutinated before lysosomal degradation. Resensitisation of the cell surface with nascent PAR2 occurs in a p24a, ARF-1 and rab11a dependent process from the Golgi.

16 GRKs, but rather by PKC isoforms (Bohm et al., 1996a) resulting in receptor desensitisation. Broad range inhibitors and PKC isoform-selective (α, β, γ, ε, η and θ) inhibitors block phosphorylation of PAR2 by PKC resulting in an increased Ca2+ mobilisation in response to trypsin and AP, whereas potent activators of PKC, particularly isoforms α and ε, decrease the PAR2-mediated Ca2+ response by causing receptor phosphorylation and enhancing the initiation of signal termination (Bohm et al., 1996a). Phosphorylation of PAR2 occurs on the 18 serine and threonine residues located within the receptor carboxy terminal domain (Figure 1.2, (Ricks and Trejo, 2009)), and this is critical for efficient receptor desensitisation and internalization through dynamin- and clathrin-dependent mechanisms (Ricks and Trejo, 2009). Furthermore, it is clear that phosphorylation of this receptor enhances its affinity for β-arrestin (Ricks and Trejo, 2009; Seatter et al., 2004). It is this interaction between β-arrestin and PAR2 that is essential for receptor uncoupling from G protein- dependent signaling cascades via a mechanism involving redistribution of β-arrestin from the cytosol and interaction with PAR2 at the plasma membrane. Following receptor activation, β-arrestin and PAR2 internalize into the early endosomes (Dery et al., 1999), an event that is mediated, at least in part, by the small GTPase Rab5a (Roosterman et al., 2003) a critical factor in endosome formation (Bucci et al., 1994).

PAR2 is also regulated by ubiquination. Mono-ubiquitination of activated PAR2 prevents its recycling to the cell surface. This modification is mediated by c-Cbl, a ubiquitin-protein isopeptide ligase, in a Src-dependent mechanism (Jacob et al., 2005a). In addition, continued trafficking of internalized PAR2 from endosomes to lysosomes is reliant on receptor deubiquitination by the endosomal deubiquitinating proteases AMSH and UBPY (Hasdemir et al., 2009) and the endosomal sorting protein hepatocyte growth factor-regulated tyrosine kinase substrate (Hasdemir et al., 2007).

1.5.2.2 Regulation of PAR2 by palmitoylation In GPCRs, palmitoylation (attachment of a 16 carbon saturated fatty acid) occurs on one or more cysteine residues via a thioester bond generally within the cytoplasmic tail 10 to 14 residues after the final TM domain (Probst et al., 1992). This modification is thought to act as a targeting signal for cell surface localisation and

17 formation of signaling complexes within plasma membrane microdomains (Chini and Parenti, 2004; Escriba et al., 2007). While the role of palmitoylation in regulation of PAR signaling is largely unexplored, it is known that PAR1 contains a di-cysteine motif and PAR2 a single cysteine residue located down stream of the last TM domain (Figure 1.2) at sites conserved in and functionally important for many palmitoylated GPCRs (Qanbar and Bouvier, 2003). Although palmitoylation of PAR1 and PAR2 at these cysteines had not been directly demonstrated at the commencement of this project, PAR1 mutants lacking the di-cysteine motif failed to couple efficiently with Gαq (Swift et al., 2006).

1.5.2.3 Mechanisms regulating PAR2 resensitisation It is clear from the above summary that proteolytic activation of PAR2 results in signal transduction as well as receptor internalisation and degradation to prevent persistent PAR2 signaling. A consequence of this mechanism is that large cytoplasmic stores of PAR2 are required to replenish cell surface localized receptor before PAR mediated signaling can recommence; a process known as resensitisation (Steinhoff et al., 2005). Therefore, in contrast to many GPCRs which show abundant cell surface expression and lower cytoplasmic levels, PAR2 is detected at high levels inside the cell with lower levels on the cell surface (Dery et al., 1999). Although the mechanisms involved in receptor resensitisation are poorly characterized, it is clear that another small GTPase, Rab11a, participates in the trafficking of cytoplasmic stores of nascent PAR2 from the Golgi apparatus to the plasma membrane (Figure 1.3). Rab11a co-localises with intracellular stores of PAR2 in vesicles during trafficking to the plasma membrane and overexpression of this GTPase accelerates replenishment of intact cell surface-localised PAR2 (Roosterman et al., 2003). Recently another Golgi localised protein, p24a, a type I TM protein, has also been shown to aid PAR2 exocytic trafficking. This protein interacts with the nascent PAR2 ECL2 at the Golgi apparatus in a mechanism regulated by ADP-ribosylation factor 1 (ARF-1 (Luo et al., 2007)).

In addition, N-glycosylation of this receptor is necessary for efficient PAR2 cell surface expression (Compton et al., 2002). While at the cell surface, N-glycosylation of PAR2 has been shown to impact receptor proteolytic activation. For example, mast cell tryptase displays variation in PAR2 activation dependent on glycosylation

18 at the two N-glycosylation sites of the receptor. Mutants lacking the N-glycoslytion site N30, located within the amino terminal of PAR2 (Figure 1.2) had increased tryptase-mediated PAR2 activation relative to the wildtype receptor. A receptor devoid of N222, located within ECL2, signaled as efficiently as wildtype PAR2 in response to tryptase (Compton et al., 2001). In contrast, trypsin signaling via PAR2 was unaffected by mutation of either site (Compton et al., 2001).

Although it is clear that mechanisms controlling PAR2 desensitisation and resensitisation are incompletely understood, it is thought that an equilibrium between these processes is essential for effective PAR2 signal transmission of extracellular stimuli across the plasma membrane (Adams et al., 2011; Grimsey et al., 2011; Soh et al., 2010). To investigate this further, an aim of this project is to examine mechanisms regulating PAR2 trafficking.

1.6 PAR2 in cancer Up-regulation of serine proteases with trypsin-like substrate specificity during malignant transformation has historically been proposed to facilitate cancer progression by cleavage of matrix components and activation of growth factors (Borgono and Diamandis, 2004; Fischer, 1946). However, with the identification of PARs it is now apparent that extracellular trypsin-like serine proteases can act on cancer and the tumour microenvironment via the PARs (Ossovskaya and Bunnett, 2004). In this regard, autocrine activation of PAR2 in a human breast cancer derived cell line by a secreted trypsin-like protease facilitates cell migration (Ge et al., 2004). Also, PAR2 is essential for breast cancer cell migration and invasion induced by the trypsin-like proteases factor VIIa and Xa (Morris et al., 2006). Importantly, upregulated PAR2 expression has also been demonstrated in tumour cells, macrophages and endothelial cells throughout the progression of breast cancer (D'Andrea et al., 2001). Similarly, aberrant PAR2 expression and signalling has been linked with lung (Jin et al., 2003; Nierodzik et al., 1998), colorectal (Darmoul et al., 2004a; Darmoul et al., 2004b; Darmoul et al., 2001; Heider et al., 2004), breast (Hernandez et al., 2009; Su et al., 2009; Versteeg et al., 2008), ovarian (Agarwal et al., 2008; Grisaru- Granovsky et al., 2005; Jahan et al., 2007) and prostate (Black et al., 2007; Ramsay et al., 2008a; Ramsay et al., 2008b) cancers.

19 Other in vitro and in vivo studies have further demonstrated a role for PAR2 in processes required for malignant progression. For example, a PAR2-deficient transgenic model of spontaneous murine breast adenomas (MMTV-PyMT) exhibited delayed onset and metastasis of breast tumours (Versteeg et al., 2008). Consistently, inhibition of TF-FVIIa-induced PAR2 signaling, but not PAR1, attenuated the growth of highly aggressive human breast cancer MDA-MB-231mfp cells in a xenograft model (Versteeg et al., 2008). Other studies have shown that PAR2 signaling also contributes to angiogenesis and breast cancer progression via TF-VIIa- and TF-Xa-mediated mechanisms (Ge et al., 2004; Hjortoe et al., 2004; Jiang et al., 2004; Morris et al., 2006) including the expression and secretion of growth factors, chemokines and cytokines known to promote angiogenesis and contribute to cancer progression (Albrektsen et al., 2007; Ruf et al., 2010). In particular, the cytoplasmic domain of TF negatively regulates PAR2-mediated angiogenesis and migration (Belting et al., 2004; Dorfleutner and Ruf, 2003). Interestingly, PAR2-mediated phosphorylation of this domain (Ahamed and Ruf, 2004) is suggested to prevent the negative regulatory function of TF and promote tumour progression (Belting et al., 2004; Ryden et al., 2010). Importantly, unlike WT mice, PAR2-deficient/MMTV- PyMT mice exhibited a lack of TF phosphorylation in breast tumours, supporting a functional role for PAR2 signaling and the TF cytoplasmic domain phosphorylation in breast cancer (Ryden et al., 2010). The relevance of these data to disease in humans is indicated by immunohistochemical analysis of patient samples showing that PAR2 signaling and TF cytoplasmic domain phosphorylation correlate with shorter recurrence free breast cancer survival (Ryden et al., 2010).

1.6.1 PAR2 in prostate cancer Prostate cancer, one of the most commonly occurring cancers in males of the western world, is found predominantly in men of older age groups and is responsible for 3% of deaths in males over 55, making it the second leading cause of cancer related morbidity (Siegel et al., 2012). A role for PAR2 in prostate cancer has been suggested from studies of cell lines and tissues. In cell based studies, activation of PAR2 with APs stimulated the migration of prostate cancer derived PC-3 and DU145 cells (Shi et al., 2004). PAR2 signalling also induces cell morphology changes in LNCaP prostate cancer cells by stimulating RhoA-dependent cytoskeletal reorganisation; a process crucial for cell migration (Greenberg et al., 2003). In these

20 cells, PAR2 signalling is also implicated in the amplification of MMP-2 and MMP-9 activity (Wilson et al., 2004), proteases involved in cancer metastasis (Curran and Murray, 2000; Lokeshwar, 1999). Activation of PAR2 with recombinant proteases that are endogenously expressed in the prostate has also implicated this receptor in disease progression with kallikrein (KLK)2 and KLK14 stimulating DU145 proliferation via PAR2 signalling (Mize et al., 2008). Immunohistochemistry analysis of tissue sections from 40 prostate cancer patients indicated that expression of PAR2 is elevated in cancerous cells compared with adjacent normal tissue (Black et al., 2007). In addition, elevated PAR2 expression was observed in tissue sections from prostate cancer patients with bone metastatic lesions (Ramsay et al., 2008a). In this study, co-expression of PAR2 and the agonist protease KLK4 was noted in cancer cells as well as osteoblasts (Ramsay et al., 2008a).

Prostate cancer progression is reliant upon a source of androgen (Feldman and Feldman, 2001). Hence, a common strategy to halt its progression is hormone deprivation by surgical castration (orchiectomy) or chemical castration with drugs inhibiting the production of testosterone (Tammela, 2004). Although initially responsive to treatment, prostate cancers eventually revert to castrate-resistant growth (Denis and Murphy, 1993). Progression of prostate cancer results in metastases to local and distant sites including the bladder, lung, lymph nodes and most commonly, the skeletal system (Bubendorf et al., 2000; Nieto et al., 2007; Saitoh et al., 1984). The high mortality rate associated with prostate cancer is attributed primarily to skeletal metastasis, which account for the majority of all prostate cancer related deaths (Coleman, 1997; Siegel et al., 2012).

1.6.2 Prostate cancer metastases to bone For prostate cancer bone metastases to form, cancer cells must first migrate from the primary site by haematogenous dissemination and colonise the bone (Kaplan et al., 2006a; Kaplan et al., 2006b). Once established in the bone, prostate tumours, unlike other tumour types, form osteoblastic lesions with the net production of bone (Guise et al., 2006; Roodman, 2004). The formation and progression of osteoblastic prostate cancer bone lesions is dependent on the reciprocal interaction between prostate cancer cells and the bone microenvironment (Ahmad et al., 2008; Baylink et al., 1993; Kaplan et al., 2006a; Kaplan et al., 2006b). However, the molecular

21 mechanisms regulating the interaction between prostate cancer cells and the bone microenvironment remain to be completely characterised.

1.7 PAR2 in the bone microenvironment A role for PAR2 in normal bone physiology has recently emerged that points to the potential relevance of this receptor to prostate cancer bone metastasis. PAR2 is expressed by a range of cell types in the mammalian skeletel system including bone marrow stromal cells, monocytes, chonodrocytes and in precursor and mature osteoblasts (Abraham et al., 2000; Mackie et al., 2008). PAR2 is required for regulation of osteoblast function; in particular osteoblast survival and the ability of these cells to mediate bone formation (Georgy et al., 2010). Also, in vitro studies indicate that PAR2 is required for osteoblast-mediated osteoclast differentiation (Smith et al., 2004). In vivo studies of PAR2 null mice have also indicated a role for PAR2 in bone processes. Although PAR2 null mice display no gross bone phenotype (Damiano et al., 1999; Georgy et al., 2011; Georgy et al., 2010; Lindner et al., 2000; Mackie et al., 2008; Schmidlin et al., 2002), the absence of PAR2 influences the skeletal architecture during growth and in response to bone injury (Georgy et al., 2011; O'Neill et al., 2012). For example, in the early stages of bone repair (day 7), PAR2 null mice exhibit significantly less bone of experimentally-induced bone fractures compared with wildtype littermates (Georgy et al., 2011).

The observed effects of PAR2 activation on PCa derived cell lines, the role of PAR2 in normal bone and the expression of this receptor in primary prostate tumours and bone metastatic lesions suggests a potential role for this receptor in prostate cancer bone metastasis. Therefore an aim of this project is to examine the role of PAR2 in prostate cancer bone lesions.

1.8 Study hypothesis and aims The hypothesis of this project is that a greater understanding of the biology of PAR2 will lead to new approaches to modulate the activity of this receptor in disease.

The overall goal of the project is to define mechanisms regulating PAR2 function and to examine the role of PAR2 in prostate cancer bone metastases. The specific aims are to:

22 1) Examine the role of palmitoylation in the regulation of PAR2 trafficking and signal transduction. 2) Identify amino acids required for agonist-stimulated PAR2 signal transduction. 3) Examine the effect of prostate cancer tumours on PAR2-mediated bone biology

Chapter 2:

Materials and Methods

26 2.1 Materials

2.1.1 General reagents

PAR2 activating peptide (AP; SLIGRL-NH2) was purchased from Auspep (Parkville, Australia). The PAR2 antagonist GB88 was a kind gift from Prof. David Fairlie (University of Queensland, Australia). The Golgi apparatus inhibitor brefeldin A, palmitoylation inhibitor 2-bromopalmitate, palmitate production inhibitor cerulenin,

CuSO4 III, protein synthesis inhibitor cycloheximide, hydroxylamine, lysosomal inhibitor MG132, medial Golgi inhibitor monensin, N-ethylmaleimide (NEM), endoplamisc reticulum inhibitor nocodazole, palmitate, sodium fluoride, sodium vanadate, t-butanol, tris[(1-benzyl-1H-1,2,3-triazol-4-yl)methyl] amine (TBTA) and tris(2-carboxyethyl)phosphine (TCEP) were from Sigma Aldrich (Sydney, Australia). 17-Octadecynoic acid (17-ODYA) was from Cayman Chemicals (Sapphire Bioscience, Waterloo, Australia) and biotin-azide and Fura-2 were purchased from Invitrogen (Mulgrave, Australia). EZ-link NHS-SS-biotin and EZ- link HPDP-biotin were from Pierce (Thermo Fisher Scientific, Scoresby, Australia). Protease inhibitor cocktail was purchased from Roche (Castle Hill, Australia). Live cell imaging 8 chamber µ-slides were obtained from Ibidi (In Vitro Technologies Pty Ltd, Noble Park, Australia).

2.1.2 Antibodies

Anti-PAR2 antibodies SAM11, H99, C17 and N19 and an anti-CAV1 antibody were from Santa Cruz Biotechnology (Quantum Scientific Pty Ltd, Murarrie, Australia). An anti-CDCP1 antibody against the last 13 amino acids of this protein was from Abcam (Sapphire Biosciences Pty Ltd., Waterloo, Australia). An anti-myc epitope (EQKLISEEDL) antibody, mouse monoclonal anti-phospho ERK1/2 and rabbit anti- ERK1/2 antibody were from Cell Signaling Technology (Genesearch Pty Ltd, Arundel, Australia). Anti-FLAG epitope (DYKDDDDK) antibody, anti-GAPDH antibody, anti-tubulin antibody and an anti-pan cadherin antibody were from Sigma Aldrich (Castle Hill, Australia). Alexa Fluor 680-conjugated streptavidin and donkey anti-mouse and donkey anti-goat Alexa Fluor 488-, donkey anti-goat Alexa Fluor 647- and goat anti-mouse, goat anti-rabbit and donkey anti-goat Alexa Fluor 680-conjugated secondaries were from Invitrogen and goat anti-mouse and goat anti-

27 rabbit IRdye 800-conjugated secondary antibodies were from LiCor (Millennium Science, Surrey Hills, Australia).

2.1.3 Enzymes and kits

All restriction enzymes were from New England Biolabs (Genesearch Pty Ltd, Arundel, Australia). Superscript III RNase H-cDNA synthesis kit and Platinum Taq DNA polymerase were from Invitrogen. Pfu Ultra DNA polymerase was from Agilent Technologies (Forest Hill, Australia). Expand High Fidelity DNA polymerase, T4 DNA ligase and the High Pure PCR clean up kit were from Roche. Active bovine trypsin was obtained from Worthington Biochemical (Scimar Pty Ltd, Templestowe, Australia). QIAprep miniprep kit was from Qiagen (Doncaster, Australia). Bicinchoninic acid (BCA) kit was from Pierce.

2.1.4 Cell culture reagents and cell lines

All cell culture media and reagents were from Invitrogen except for fetal calf serum (Sigma Aldrich), BRFF-HPC1 serum free medium (AthenaES, Sapphire Biosciences) and G418 (In Vitro Technologies). All cell lines were from ATCC (Manassas, VA).

2.1.5 Animals and housing

Five week old male wildtype severely combined immuno-deficient (SCID) mice were from the Australian Resource Centre (Perth, Western Australia). Animals were housed in specific pathogen free (SPC) conditions at the Biological Research Facility, Princess Alexandra Hospital, Brisbane, Queensland, Australia. All animal experimentation were approved by the Queensland University of Technology Ethics committee (1000000124) and the University of Queensland Ethics committee (MMRI/210/10).

28 2.1.6 Buffers

Agarose gel loading dye: 0.09% (w/v) bromophenol blue, 0.09% (w/v) xylene cyanol, 60% (v/v) glycerol, 60 mM EDTA, 1M Tris (pH 8.0).

TAE: 40 mM Tris-acetate, 1 mM EDTA (pH 8.0)

Membrane buffer: 5 mM Tris (pH 7.5), 0.5 mM EDTA, 1 x protease inhibitor cocktail (Roche), 1 mM sodium vanadate, 10 mM NaF.

Cell lysis buffer: 50 mM Tris (pH 7.4), 150 mM NaCl, 5 mM EDTA, 1% Triton X- 100 (v/v), 1 x protease inhibitor cocktail (Roche), 1 mM sodium vanadate, 10 mM NaF.

MeSNA reducing buffer: 50 mM 2-mercaptoethanesulfonic acid sodium salt (MeSNA), 100 mM Tris (pH 8.6), 100 mM NaCl.

6x Sodium Dodecyl Sulfate-Polyacrylamide Gel Electrophoresis (SDS-PAGE) loading dye: 430 mM Tris (pH 6.8), 18% (w/v) SDS, 0.2% (w/v) bromophenol blue, 30% (v/v) glycerol, 15% (v/v) β-mercaptoethanol, diluted to 1x solution in lysates.

10x SDS-PAGE running buffer: 250 mM Tris, 1.92 M glycine, 1% (w/v) SDS (pH

8.3), diluted to 1x solution in double distilled H20 (ddH2O).

Protein transfer buffer: 25 mM Tris, 192 mM glycine, 0.1% (w/v) SDS, 20% methanol (pH 8.3).

Hank’s buffered saline solution (HBSS): 25 mM HEPES (pH 7.4), 121 mM NaCl,

5.4 mM KCl, 0.8 mM MgCl2, 1.8 mM CaCl2, 5.5 mM glucose

Phosphate buffered saline (PBS): 137 mM NaCl, 10 mM Na2HPO4 ● 2 H2O, 2.7 mM

KCl, 2 mM KH2PO4 (pH 7.4).

Tris buffered saline with Tween-20 (TBS-T): 100 mM Tris, 137 mM NaCl, 0.1% (v/v) Tween-20 (pH 7.5).

FURA-2 loading buffer: HBSS including 2.5 mM probenecid and 0.01% (v/v) pluronic acid F-127.

29 FLUO3 loading buffer: HBSS including 2.5 mM probenecid, 0.01% (v/v) pluronic acid F-127 and 1% (v/v) FBS.

2.1.7 Oligonucleotides

All oligonucleotides used in DNA sequencing and polymerase chain reaction (PCR) were purchased from Sigma Aldrich. Table 2.1 lists the oligonucleotides used to clone, mutate or examine PAR2 mRNA expression throughout this PhD program of research.

2.1.8 Vectors

The pGEM-T easy vector system (Promega, Alexandria, New South Wales) was used in sub-cloning of PCR or digestion products. The expression vectors pcDNA 3.1 (neomycin) and pcDNA5 flpin vector system were from Invitrogen. pEGFP-N1 and pRIESneo2 were from Clontech (Mountain View, CA).

2.1.9 Bacterial growth media and plates

Bacterial cultures were propagated using 2x YT broth (1.6% (w/v) bacto tryptone, 1% (w/v) bacto yeast extract, 85.6 mM NaCl (pH 7.0)). Bacterial cultures or glycerol stocks were plated directly onto Luria-Bertani (LB) agar plates (1% (w/v) bacto tryptone, 0.5% (w/v) bacto yeast extract, 171.2 mM NaCl, 1.5% (w/v) agar (pH 7.0)) containing the appropriate antibiotic selection marker (100 µg/ml ampicillin or 30 µg/ml kanamycin). For blue-white colony selection of pGEM-T easy vectors, LB agar plates were supplemented with 20 µg/ml X-Gal and 100 mM isopropyl-β-D- thiogalactoside (IPTG).

30 Table 2.1 Oligonucleotide primer sequences used in cloning, site directed mutagenesis and qRT-PCR experiments Primer Sequence (5' to 3') gaattcCCAGGAGGATGGGGAGCCCCAG

PAR2‐ F myc R gcggccgcTTACAGGTCCTCCTCCGAGATAAGCTTCTGCTCATAGGAGGTCTTAACAGTGG F ggatccAGTCATGGTGAGCAAG Cloning mCherry R gcggccgcTTACTTGTACAG F TGCAAAGAACGCTCTCCTTGCCCGAAGTGTCCGCACTGTAAAG C361A R ACGTTTCTTGCGAGAGGAACGGGCTTCACAGGCGTGACATTTC F GACCTCCTCTCTGTCATCTGGGCCCCCTTGAAGATTGCCTATCAC F128A R GTGATAGGCAATCTTCAAGGGGGCCCAGATGACAGAGAGGAGGTC F GGGGAAGCTCTTTGTAATGTGGCTATTGGCTTTTTCTATGGCAAC L151A R GTTGCCATAGAAAAAGCCAATAGCCACATTACAAAGAGCTTCCCC F TGTAATGTGCTTATTGGCTTTGCCTATGGCAACATGTACTGTTCC F155A R GGAACAGTACATGTTGCCATAGGCAAAGCCAATAAGCACATTACA F ATGTGCTTATTGGCTTTTTCGCTGGCAACATGTACTGTTCCATT Y156A R AATGGAACAGTACATGTTGCCAGCGAAAAAGCCAATAAGCACAT F ATGTGCTTATTGGCTTTTTCTTGGGCAACATGTACTGTTCCATT

Y156L R AATGGAACAGTACATGTTGCCCAAGAAAAAGCCAATAAGCACAT F GGCTTTTTCTATGGCAACATGTTATGTTCCATTCTCTTCATGACC Y160L R GGTCATGAAGAGAATGGAACATAACATGTTGCCATAGAAAAAGCC F CTCTCTCTGGCCATTGGGGCCTTTCTGTTCCCCAGCCTTC mutagenesis V250A R GAAGGCTGGGGAACAGAAAGGCCCCAATGGCCAGAGAGAG F CCTGATCTGCTTCACTCCTAGTCTCGTTCTGCTTGTGGTGCATTA N304L directed R TAATGCACCACAAGCAGAACGAGACTAGGAGTGAAGCAGATCAGG

Site F CACTCCTAGTAACCTTCTGGCTGTGGTGCATTATTTTCTGA L307A R TCAGAAAATAATGCACCACAGCCAGAAGGTTACTAGGAGTG F CTCCTAGTAACCTTCTGCTTGCGGTGCATTATTTTCTGATTAA V308A R TTAATCAGAAAATAATGCACCGCAAGCAGAAGGTTACTAGGAG F AACCTTCTGCTTGTGGTGCATCTTTTTCTGATTAAGAGCCAGGGC Y311L R GCCCTGGCTCTTAATCAGAAAAAGATGCACCACAAGCAGAAGGTT F CAGAGCCATGTCTATGCCCTGGCCATTGTAGCCCTCTGCCTCTC Y326A R GAGAGGCAGAGGGCTACAATGGCCAGGGCATAGACATGGCTCTG F CAGAGCCATGTCTATGCCCTGCTCATTGTAGCCCCTCTGCCTCTC Y326L R GAGAGGCAGAGGGGCTACAATGAGCAGGGCATAGACATGGCTCTG F CTATGCCCTGTACATTGTAGCCGTCTGCCTCTCTACCCTTAACAG L330V R CTGTTAAGGGTAGAGAGGCAGACGGCTACAATGTACAGGGCATAG F CGTCGGGGCTTCCAGGAGGAT PAR2

R TATTGGTTCCTTGGATGGTGCCACTG GCAAATTCCATGGCACCGT

PCR F ‐ GAPDH R TCGCCCCCACTTGATTTTGG qRT F TGAACGTCTTGCTCGAGATGTG HPRT1 R CCAGCAGGTCAGCAAAGAATTT

31 2.2 Methods

2.2.1 Cell culture

Chinese hamster CHO-K1 (CHO) cells were grown in DMEM and prostate cancer lines PC3, DU145, 22Rv1 and LNCaP were grown in RPMI1640 medium, each supplemented with 10% fetal calf serum (FCS). Prostate cancer LNCaP-C42B cells were grown in T-medium supplemented with 10% FCS and MDA-PCa-2b cells were grown in BRFF-HPC1 medium supplemented with 20% FCS. Lung murine fibroblasts (LMF) from PAR1 null mice (Darrow et al., 1996) expressing human PAR1, PAR2 or PAR4 were from Johnson & Johnson Pharmaceutical Research and Development (Spring House, PA, USA) and cultured in DMEM supplemented with 10% FCS and 200 µg/ml hygromycin B. All cell lines were cultured in 100 units/ml of penicillin, and 100 units/ml of streptomycin in a 5% CO2 humidified atmosphere at 37°C. All cells were passaged using 0.5 mM EDTA in PBS. Cell numbers were calculated manually using a haemocytometer (ProSciTech, Thuringowa, Australia). All cell lines were regularly tested for mycoplasma contamination.

For cryopreservation, dissociated cells were resuspended in cell culture media containing 10% (v/v) dimethyl sulphoxide (DMSO), aliquoted into cyrovials and frozen to -80°C in a vessel containing isopropanol. For long term storage, frozen cells were transferred to gas phase nitrogen. To resuscitate frozen cell lines, cyrovials were thawed at 37°C and the cell suspension transferred to 10 ml of culture media. To remove residual DMSO, cells were centrifuged (370 g for 5 min) and the pellet was resuspended in culture media before transfer into T25 cm2 flasks.

2.2.2 RNA extraction

Total RNA was extracted from cells grown in a six-well plate using the TRIzol reagent (Invitrogen). TRIzol (500 µL) was added to adherent cells in each well and transferred to RNase-free tubes. Chloroform (200 µl per ml of TRIzol) was added to each sample and mixed vigorously for 30 sec before centrifugation (12,000 g for 15 min at 4°C). The aqueous phase was transferred to a new tube and mixed with

32 isopropanol (500 µl per ml of TRIzol) and stored at -20°C overnight. Precipitated RNA was centrifuged (14,000 g for 15 min at 4°C) and the pellet was washed with 70% ethanol (1 ml per ml of TRIzol). Samples were centrifuged and the supernatant was carefully removed. Pellets were allowed to air dry before resuspension in 30 µL of nuclease-free H2O (Invitrogen). RNA purity and concentration were assessed using a Nanodrop 1000 spectrophotometer (Thermo Fisher Scientific).

2.2.3 Reverse transcription

RNA (1 µg) was diluted to 10 µl with nuclease-free H2O and treated with 1 unit of DNase1 according to instructions from the manufacturer (Invitrogen). The reaction was stopped by incubating samples in 1x Stop Solution (Invitrogen) and heating at 65°C for 10 min. RNA was combined with 200 ng random hexamers (Invitrogen) and heated at 65°C for 5 min, followed by cooling on ice. First-strand synthesis was performed using 2 units of SuperScript III reverse transcriptase according to instructions from the manufacturer (Invitrogen).

2.2.4 Quantitative real time PCR (qRT-PCR) qRT-PCR reactions were performed in 96 well plates (Axygen, Quantum Scientific

Pty. Ltd.) and contained 2.5 µl diluted cDNA in nuclease-free H2O (1:5), 50 nM forward and reverse primer (see Table 2.1), 1x final concentration SYBR green PCR master mix (Applied Biosystems, Scoresby, Australia) and nuclease-free H2O (total volume of 20 µl). Reactions were performed using an ABI PRISM 7300 real-time PCR system (Applied Biosystems). Cycling conditions were 95°C for 10 min, 40 cycles of 95°C for 15 sec and 60°C for 1 min followed by a primer-template dissociation step. Threshold and baseline values were adjusted manually using the ABI PRISM 7300 SDS software (Applied Biosystems). Gene expression was calculated using the comparative CT (∆∆CT) method according to ABI User Bulletin #2 (Applied Biosystems).

33 2.2.5 Polymerase Chain Reaction (PCR) and agarose gel electrophoresis

PCR reaction mixtures contained 10-100 ng of DNA template, 200 nM of forward and reverse primer (see Table 2.1), 0.2 mM dNTP (Roche), 1x final concentration Taq DNA polymerase buffer (supplied with polymerase), 1.25 units of Taq DNA polymerase (see section 2.1.3 for DNA polymerases used in the PhD program of research) and nuclease-free H2O (total volume of 50 µl). Reactions were performed using a Bio-Rad S1000 Thermal Cycler (Bio-Rad, Gladesville, Australia). Cycling conditions were 95°C for 3 min, 20-35 cycles of 95°C for 30 sec, annealing temperatures of 50-65°C for 30 sec (see Table 2.1 for specific primer temperatures), 68-70°C for 30 sec (Platinum Taq, 72°C; Pfu Ultra, 68°C; Expand High Fidelity DNA polymerase, 72°C) and a final extension step of 68-72°C for 10 min. Reaction products in 50 µl were mixed with agarose gel loading dye (see section 2.1.5) and electrophoresed on 0.8-1% (w/v) agarose gels prepared with TAE (see section 2.1.5) containing SYBR Safe (1:10,000 (v/v); Invitrogen). DNA molecular weight was assessed using DNA Marker X (Roche). Agarose gels were visualised using a FLA- 5100 ultra-violet (UV) scanner (Fujifilm Australia Pty. Ltd.).

2.2.6 Restriction enzyme digests

Two µg of purified DNA from bacterial minipreps were digested with 5 units of the appropriate restriction endonuclease at 37°C for 1-2 h.

2.2.7 Ligation and bacterial transformation

PCR products or DNA restriction digestion products were excised from an agarose gel using a sterile scalpel blade and DNA purified using a High Pure PCR product purification kit according to instructions from the manufacturer (Roche). DNA purity was assessed by agarose gel electrophoresis. The purified DNA was ligated overnight at 16°C into pGEM-T easy (for PCR products) or directly into the expression vector (for re-cloning of DNA restriction digest products; Section 2.1.8) using 1x final concentration T4 DNA ligase buffer (supplied with ligase), T4 DNA ligase (Roche) and nuclease-free H2O (total volume 10 µl). Molar ratios of 1:4 or 1:7 of vector:insert were used.

Ligation products or constructs were transformed into XL-10 Gold ultracompetent E. coli cells (Agilent Technologies). Cells were thawed on ice. Plasmid DNA (50 ng) or 3-10 µl of ligation mixture was incubated with the cells for 30 min on ice before being heat-shocked at 42°C for 45 sec. Samples were chilled on ice for 2 min before incubation in 1 ml of 2x YT medium for 1 h at 37°C with shaking at 250 rpm. This culture (100-200 µl) was plated onto LB agar plates containing the appropriate selection antibiotic. Agar plates were incubated overnight at 37°C to allow colony growth.

2.2.8 Colony screening, purification of plasmid DNA and DNA sequencing

Blue-white screening was performed for pGEM-T easy vector sub-cloning; white colonies contained inserts, disrupting the LacZ gene. To screen colonies that contained inserts, an alkaline lysis method was used. To 200 µl of overnight bacterial culture, 200 µl of bacterial lysis buffer (see section 2.1.5) and 200 µl of 5 M potassium acetate (pH 5.0) was added sequentially and mixed vigorously. The mixture was centrifuged (16,000 g for 3 min) and the supernatant was mixed with an equal volume of isopropanol to precipitate DNA. The mixture was centrifuged

(16,000 g for 3 min) and the pellet resuspended in 30 µl of nuclease-free H2O. Samples were examined by agarose gel electrophoresis containing SYBR Safe and visualised by UV illumination. Plasmids were compared with a negative control (empty vector DNA) and plasmids of the expected molecular weight were selected for further DNA purification.

High quality plasmid DNA from individual colonies or frozen glycerol stocks, grown overnight in 2x YT medium at 37°C with shaking at 250 rpm, was purified using a QIAprep Spin miniprep kit according to instructions from the manufacturer (Qiagen). DNA purity and concentration were assessed using a Nanodrop 1000 spectrophotometer.

35 The sequence of all constructs was confirmed by DNA sequencing at the Australian Genome Research Facility (St. Lucia, Australia). Sequencing reactions contained 500 ng of double stranded plasmid DNA, 233 nM of primer and nuclease-free H2O (total volume of 12 µl).

2.2.9 Transfections

Transfections were performed using either Lipofectamine 2000 or Lipofectamine LTX according to instructions from the manufacturer (Invitrogen). For stable transfections, cells were transfected with an empty expression vector or vector containing the coding sequence of interest using Lipofectamine 2000 then selected in the appropriate antibiotic (G418 or hygromycin B) containing media for 10 days. Individual clones were selected and expanded before characterisation of protein expression.

2.2.10 Cell membrane preparation

Crude cell membrane extracts were collected as previously described (Compton et al., 2002) with some alterations. Briefly, cells at 50% confluence were washed with

PBS and distilled H2O for 30 sec to induce hyptonic cell shock. Swollen cells were mechanically resuspended in membrane buffer and disrupted by several passes through a 26 gauge needle. Cellular debris was removed by centrifugation (800 g for 10 min at 4°C). Crude membrane preparations were collected from the supernatant by ultracentrifugation using a Sorvall MX140 floor ultracentrifuge with a S100-AT4 rotor (100,000 g for 1 h at 4°C). Membrane pellets were then resuspended in lysis buffer before protein quantification using a BCA kit (Pierce).

2.2.11 Analysis of palmitoylation by acyl-biotinyl exchange chemistry

PAR2 palmitoylation was assessed by an acyl-biotinyl exchange approach (ABE; (Drisdel and Green, 2004)) described previously (Wan et al., 2007). Crude membrane extracts, isolated as above, were prepared from cells at 50% confluence. To assess the effect of PAR2 agonism on palmitoylation, before isolation of cell

36 membranes cells were treated with PAR2 AP (100 µM). To prevent protein palmitoylation, cells were incubated with 2-bromopalmitate for 16 h prior to collection of crude membrane. Membrane pellets were solubilised for 1 h at 4°C in lysis buffer containing 50 mM HEPES pH 7.4, 150 mM NaCl, 5 mM EDTA, 1% Triton X-100 (v/v), 1 x protease inhibitor cocktail (Roche) and, to block free thiols, 10 mM NEM. Solubilized protein (100 µg) was precipitated by chloroform-methanol (1:3:2 protein:methanol:chloroform; CM) and protein pellets were resuspended in a buffer containing 4% SDS (4SB; 4% SDS, 50 mM HEPES pH 7.4, 5 mM EDTA) and diluted five fold with lysis buffer containing 10 mM NEM and incubated at 4°C overnight with gentle agitation. Excess NEM was removed by three sequential CM precipitations followed by resuspension in 4SB (50 µl). Samples were divided into two equal portions. To cleave thioester bonds and allow incorporation of a biotin moiety at exposed sulphur atoms, one portion was diluted five fold in buffer (0.2% Triton X-100, 1 x protease inhibitor cocktail) containing hydroxylamine (0.7 M) and EZ-link HPDP-biotin (1 mM). As a control the other portion was diluted five fold in the same buffer in which hydroxylamine was replaced with Tris (50 mM Tris pH 7.4). In the absence of hydroxylamine, palmitate groups are not removed thereby preventing biotinylation mediated purification. Each portion was incubated at ambient temperature for 1 h with gentle agitation. Unreacted biotin was removed by three sequential CM precipitations and the protein pellet was resuspended in 4SB and diluted 10 fold in Tris containing lysis buffer. Biotinylated proteins were affinity purified using streptavidin beads (Pierce) by incubation at 4°C for 1 h. Bound proteins were eluted from washed beads with SDS sample buffer and subjected to denaturing polyacrylamide gel electrophoresis (SDS-PAGE) and palmitoylated PAR2 was detected by anti-myc Western blot analysis (section 2.2.15).

2.2.12 Analysis of palmitoylation by metabolic labelling with an alkyne containing palmitate analogue followed by in vitro copper catalysed alkyne-azide cycloaddition (click) chemistry

PAR2 palmitoylation was also examined using a modified version of a recently described method using an alkyne containing palmitate analog (17-ODYA) reacted to biotin-azide via click chemistry (Martin and Cravatt, 2009). To inhibit endogenous

37 palmitate synthesis cells were preincubated for 1 h at 37°C with cerulenin (5 µg/ml), before metabolic labelling with the palmitate analog 17-ODYA (25 µM) for 2 h at 37°C. To assess the effect of PAR2 agonism on palmitoylation, cells were labelled with 17-ODYA in the presence or absence of PAR2 AP (100 µM). To examine PAR2 palmitoylation in particular organelles, cells were labelled with 17-ODYA for 2 h in the presence of protein trafficking and organelle function inhibitors: Brefeldin A (5 µg/ml; blocks ER-to-Golgi protein traffic (Dinter and Berger, 1998)), nocodazole (20 µg/ml; blocks ER-to-pre-Golgi protein traffic (Storrie and Yang, 1998)), monensin (10 µM; blocks medial-Golgi-to-post-Golgi traffic (Dinter and Berger, 1998)), and 2-bromopalmitate (100 µM; inhibits protein palmitoylation (Jennings et al., 2009)). Nocodazole was added to cells 15 min before labelling with 17-ODYA, while Brefeldin A, monensin and MG132 were added 30 min prior to labelling with 17-ODYA. 2-Bromopalmitate was incubated with cells for 16 h before collection of membrane fractions. Crude membrane fractions, isolated as described above, were CM precipitated then resuspended in PBS containing 1.2% SDS by sonication and heating at 70°C for 10 min. Proteins (100 µg at 1 mg/ml) labelled with alkyne containing 17-ODYA were reacted with biotin-azide via standard click chemistry conditions (Martin and Cravatt, 2009); involving the addition of 100 µM biotin-azide (1 mM stock dissolved in DMSO), 1 mM TCEP (50 mM stock dissolved in H2O), 100 µM TBTA (1.7 mM stock dissolved in 20% DMSO/80% t-butanol) and

1 mM CuSO4 (50 mM stock dissolved in PBS) in order, followed by incubation in the dark at ambient temperature for 1 h. Samples were CM precipitated to remove excess biotin, resuspended in PBS containing 0.8% SDS and diluted eight fold in lysis buffer. Palmitoylated proteins modified with a biotin moiety were affinity purified using streptavidin beads by incubation at 4°C for 1 h with gentle agitation. Bound proteins were eluted from washed beads with SDS sample buffer and subjected to SDS-PAGE followed by anti-myc Western blot analysis to detect palmitoylated PAR2 (section 2.2.15). In experiments to assess whether palmitoylation was found exclusively on cysteines and not at amides, samples were boiled in 0.7 M hydroxylamine (to cleave thioester bonds) or 50 mM Tris pH 7.4 (control) prior to immobilisation on streptavidin beads.

38 2.2.13 Cell surface biotinylation

Cell surface proteins were isolated using cell impermeant EZ-link NHS-SS-biotin (1.22 mg/ml) as described previously (He et al., 2008). Briefly, cells at 50% confluence stably transfected with PAR2-myc, PAR2-C361A-myc or vector were washed with PBS and biotinylated for 1 h at 4°C. Cells were washed with PBS and whole cell lysates collected in lysis buffer. After removal of cellular debris by centrifugation (3000 rpm for 10 min at 4°C), lysates were incubated with streptavidin beads (Pierce) for 30 min at 4°C with gentle agitation. Biotinylated cell surface proteins immobilised on streptavidin beads were pelleted by centrifugation (3000 rpm for 5 min at 4°C) and together with intracellular proteins present in the supernatant were examined by Western blot analysis (section 2.2.15).

2.2.14 Receptor endocytosis

PAR2 endocytosis was assessed as previously described (Morimoto et al., 2005). Briefly, cell surface proteins were labelled with cell impermeant EZ-link NHS-SS- biotin (0.5 mg/ml) for 30 min at 4°C, then washed before incubation at 37°C in culture media containing PAR2 AP (100 µM) and MG132 (20 µM) for 5, 15 and 30 min to allow receptor internalisation. Surface biotin was removed by incubating cells at 4°C in MeSNA reducing buffer for 30 min before excess MeSNA was quenched with 60 mM iodoacetamide for 15 min in PBS. Whole cell lysates were collected in lysis buffer and incubated with streptavidin beads to isolate internalised PAR2 which was examined by Western blot analysis (section 2.2.15).

2.2.15 Western blot analysis Whole cell lysates, cell membrane preparations, proteins eluted from streptavidin beads, and proteins collected from the cell surface biotinylation protocol were separated with SDS-PAGE using 10% resolving layer and a 4% stacking layer and transferred to nitrocellulose membranes. After blocking with Odyssey blocking buffer (LiCor), membranes were incubated with primary antibodies overnight at 4°C, washed and then incubated with species-appropriate AlexaFluor 680 or IRdye 800- conjugated secondary antibodies for 45 min at ambient temperature (see figure

39 legends for antibody dilution). Following washing, membranes were scanned on an Odyssey infrared imaging system (LiCor). Consistent protein loading and transfer was determined by reprobing membranes with either an anti-tubulin antibody (Sigma Aldrich), anti-GAPDH antibody (Sigma Aldrich) or AlexaFluor-680-conjugated streptavidin (Invitrogen). Where relevant, signal intensity was determined by densitometry analysis using Odyssey software (LiCor).

2.2.16 Flow cytometry

Adherent cells at 50% confluence were lifted non-enzymatically, counted and 2.5 x105 cells were washed and stained with the mouse anti-PAR2 antibody SAM11 or the goat anti-PAR2 antibody N19 (2 µg/1x106 cells) in PBS containing 2% FCS for 30 min at 4ºC. After washing with PBS, cells were stained with an AlexaFluor 488- conjugated secondary antibody and 20,000 events were collected and analysed on a Beckman Coulter FC500 flow cytometer. For cell surface repopulation experiments, resuspended cells treated with 50 nM trypsin for 15 min at 37°C (to remove cell surface PAR2) were washed then incubated in trypsin-free DMEM for 0, 15 and 30 min and placed on ice, followed by staining with anti-PAR2 N19 antibody. For cells expressing GFP, a Alexa Fluor 647-conjugated secondary antibody was used to prevent spectral overlap and cells were analysed on a Beckman Coulter CyAN flow cytometer. For all experiments, mean fluorescence intensity (MFI) values were calculated by subtracting secondary only staining from specific anti-PAR2 staining.

2.2.17 FURA2 intracellular calcium mobilisation assay

Analysis of calcium mobilisation was performed as previously described (Ramsay et al., 2008a). Cells at 50% confluence lifted nonenzymatically were loaded with the fluorescence indicator FURA-2 acetoxymethyl ester (1 µM; Invitrogen) for 1 h at 37°C in FURA-2 loading buffer. Cells were washed with PBS and resuspended in the same buffer lacking Fura-2 and pluoronic acid. To initiate Ca2+ mobilisation cells were treated with trypsin (0.1–1000 nM trypsin for dose-response experiments) or PAR2 AP (100 µM). Calcium mobilisation was monitored using a Polarstar Optima fluorescent plate reader (BMG Labtech Pty Ltd, Mornington, Australia). All

40 experiments were performed in triplicate on three independent occasions with results displayed as mean ± SD normalised to maximal PAR2 response.

2.2.18 FLUO3 intracellular calcium mobilisation assay CHO-flpin cells stably transfected with empty vector, wildtype PAR2 or mutant PAR2 were seeded into 96-well black walled, glass bottomed plates (5 x 105 cells per well) and grown to 80% confluence. Growth medium was removed and cells were incubated for 1 hour at 37oC in FLUO3 loading buffer containing FLUO3 (dissolved in DMSO; 4 μM). Cells washed twice with HBSS were transferred to a FLIPR TETRA fluorescent plate reader (Molecular devices) for agonist injection and fluorescence measurements. PAR2 agonists trypsin, 2f-LIGRLO-NH2 and GB110 were added 10 s after reading commenced. For dose-response concentration curves, final concentrations of 0.1–1000 nM of trypsin and 0.01–100 µM of 2f-LIGRLO-

NH2 and GB110 were delivered to cells in triplicate wells per agonist concentration. Calcimycin (A23187) was used as a positive control for calcium efflux and to measure maximum fluorescence. Calcium mobilisation assays were performed on three independent occasions in triplicate with Dr Jacky Suen from the laboratory of my associate supervisor Prof. David Fairlie (Institute for Molecular Bioscience, University of Queensland).

2.2.19 Confocal microscopy

CHO cells seeded onto sterile coverslips at 70% confluence in a 24 well plate were transiently transfected with vector (pEGFP-N1) or PAR2-GFP (see Section 2.2.9). Twenty-four hours following transfection, the cells were washed in PBS and fixed with 2% formaldehyde for 20 minutes at room temperature. The cells were again washed in PBS before permeabilisation with 0.5% Triton X-100 in PBS for 5 minutes at room temperature. Cells were then blocked for 30 minutes with 3% BSA (Sigma Aldrich) in PBS containing 0.1% Triton X-100 before incubating the cells with anti-PAR2 antibodies SAM11, H99, C17 and N19 (1:70) in blocking buffer º overnight at 4 C. Fixed cells were again washed with PBS and incubated with species appropriate fluorescently tagged AlexaFluor 568 secondary antibodies (1:1,000) in

41 blocking buffer for 1 hour at room temperature. Following a final wash in PBS, coverslips were mounted onto slides using Immuno-Fluore mounting media (MP Biomedicals, Seven Hills, NSW, Australia) and sealed with nail polish. Immunofluoresence was examined using a Leica SP5 confocal microscope (Leica Microsystems, Gladesville, Australia) and images were processed using MetaMorph software and displayed using Corel Draw X5.

2.2.20 Live cell Confocal microscopy

Cells seeded in 8 chamber µ-slides (Ibidi) were transiently co-transfected with PAR2-GFP and PAR2-mCherry containing expression constructs for 12 h and ° imaged at 37 C in a 5% CO2 atmosphere using a Zeiss LSM510 confocal microscope (North Ryde, New South Wales). Images were processed using MetaMorph software and displayed using Corel Draw X5.

2.2.21 Mouse intra-tibial injections of prostate cancer 22Rv1 cells

Single cell suspensions of 22Rv1 prostate cancer cells were prepared according to section 2.2.1. Cells lifted non-enzymatically were counted and resuspended at a concentration of 2 x 107 cells/ml in PBS. Cells were stored on ice until injection. Five week old SCID mice were used for these experiments. Prior to injection, all animals were anaesthetised by intraperitoneal (i.p.) injection with 130 mg ketamine and 8.8 mg xylazine per kg of body weight. Core animal temperature was maintained by placing animals on a heat pad for the duration of the surgery. After positioning of the animals in dorsal recumbency, the knee joint of both hindlimbs were shaved and swabbed with 70% ethanol. A 26-guage needle was inserted ~3 mm into the proximal aspect of each tibia through the patellar ligament using a drilling motion. Single cell suspensions of 22Rv1 cells were implanted (10 µl) into the right tibia using a 25 µl syringe (Hamilton Co., Reno, Nevada, USA). PBS (10 µl) was injected into the contralateral tibia as a control to account for possible irritation and bone degradation during surgery. Following the procedure, an i.p. injection of buprenorphine (0.1 mg/kg body weight) was administered to animals as an analgesic. Animals were monitored weekly for body weight and ambulatory capacity.

2.2.22 Drug administration and tissue processing

Animals were allocated into four treatment groups consisting of 16 animals; (1) vehicle (olive oil as negative control; 8 µl/g), (2) docetaxel (as positive control; 10 mg/kg (Hung, 2007)), (3) PAR2 antagonist GB88 (5 mg/kg in olive oil (Adams et al., 2011; Suen et al., 2011) and (4) combined GB88 and docetaxel by i.p. injection. Treatment commenced one week following surgery and continued until 10 weeks post-surgery. Animals were treated daily with vehicle and GB88 by oral gavage (p.o.)(Lohman et al., 2011; Suen et al., 2011) and weekly with docetaxel by i.p. (Hung, 2007). Animals were also segregated into two analysis groups; (1) histology and (2) histomorphometry analysis. Accordingly, 32 animals were in each analysis group consisting of 8 animals per treatment group. For animals within the histomorphometry analysis group, the calcium binding fluorochrome label calcein (10 mg/kg of calcein at 1 mg/ml; Sigma Aldrich) was injected i.p. 7 days and 2 days before sacrifice.

Animals were sacrificed 10 weeks following surgery by carbon dioxide asphyxiation. Immediately following sacrifice, radiographic images were taken of both hindlimbs using an in vivo FX Kodak Image Station (Eastman Kodak Company, Rochester, NY, USA). After imaging, the tibia connected with the femur were harvested and the distal end of the tibia and proximal end of the femur were removed. Bones from histology analysis group animals were placed in PBS while bones from histomorphometry group animals were fixed with 4% paraformaldehyde for 24 hours. After fixation, these bones were placed in 70% ethanol.

2.2.23 Micro-computed tomography (CT) of mouse bone

The bones of mice within the histomorphometry analysis group were scanned in a µ- CT 40 micro computed tomography scanner (Scanco Medical, Brüttisellen, Switzerland), at an energy of 55 kVp and intensity of 145 µA with 200 ms integration time. The scans were reconstructed to three-dimensional datasets with an isotropic voxel size of 12 µm. For the proximal tibia, 175 slices were analysed up to

43 20 slices distal from the growth plate. An automatic algorithm was used to segment the trabecular compartment from the proximal tibia sections (Buie et al., 2007). The micro-CT scanner was operated by Dr Roland Steck (Institute of Health and Biomedical Innovation, Queensland University of Technology).

2.2.24 Statistics Results are mean ± SEM of at least 3 independent experiments unless otherwise stated. Statistical significance was assessed by Student’s t test. P < 0.05 was considered significant.

Chapter 3:

Evaluation of anti-PAR2 antibodies

46 3.1 Introduction As noted in Chapter 1, an aim of this PhD program of research is to examine the in vitro function of PAR2 in the initiation of signal transduction and trafficking of the receptor. In order to investigate these facets of PAR2, detection of the endogenous receptor with specific anti-receptor antibodies is critical. In this chapter, the specificity of commercially available anti-PAR2 antibodies is systematically investigated.

There are currently six commercially available anti-PAR2 antibodies all from Santa Cruz Biotechnology. At the time of commencing this study, four antibodies were available. These antibodies are summarised in Table 3.1 with the cognate receptor epitope illustrated in Figure 3.1. Anti-PAR2 antibodies SAM11, H99 and N19 recognise extracellular PAR2 epitopes whereas C17 recognises an intracellular epitope within the carboxyl terminal domain. Although these antibodies are from a commercial source, the specificity of these antibodies has not been systematically examined. Moreover of particular concern, the validity of commercially available anti-GPCR antibodies has been questioned (Michel et al., 2009). Analysis of 49 different antibodies designed to detect 19 receptor subtypes by Western blot analysis, immunohistochemistry and immunofluorescence indicated that all of these antibodies were non-specific (Bodei et al., 2009; Hamdani and van der Velden, 2009; Jensen et al., 2009; Jositsch et al., 2009; Lu and Bartfai, 2009; Pradidarcheep et al., 2009). Therefore, the authors encourage systematic testing of anti-GPCR antibodies to reduce the reporting of erroneous data (Michel et al., 2009). In this regard, a set of criteria has been proposed to experimentally demonstrate antibody specificity (Michel et al., 2009). These include: 1. Loss of staining in tissues from genetically modified animals lacking the receptor; 2. Reduction in staining in cells treated with a receptor specific knockdown construct e.g. siRNA; 3. Testing antibody selectivity against multiple receptor subtypes endogenously or exogenously expressed in the same cell line; 4. Testing multiple anti-receptor antibodies of different epitope and comparable banding patterns by Western blot analysis.

Table 3.1 Summary of the examined anti-PAR2 antibodies. h, human; m, mouse; r, rat. Antibody Species Epitope Reactivity SAM11 Mouse IgG2a Residues 37-50 h, m, r H99 Rabbit Residues 230-328 h, m, r C17 Goat C-terminus h, m, r N19 Goat N-terminus h

Figure 3.1 Epitopes recognised by anti-PAR2 antibodies. Schematic representation of the primary sequence and structure of human PAR2 including epitopes for the anti-PAR2 antibodies SAM11 (blue), H99 (green), C17 (purple) and N19 (orange). Arrowhead and arrow show respectively the cleavage sites for removal of the signal peptide and receptor activation (pro-peptide) to yield the TL sequence SLIGKV of human PAR2. Asterisks denote sites of N-linked glycosylation. Transmembrane (TM), extracellular (ECL) and intracellular (ICL) loop domains are indicated.

49 As there has been considerable doubt cast upon the specificity of anti-GPCR antibodies, in this chapter the specificity of the four anti-PAR2 antibodies commercially available at the start of this project (Table 3.1 and Figure 3.1) has been systematically examined. Each anti-PAR2 antibody was assessed by immunoprecipitation, Western blot analysis, immunofluorescence and flow cytometry. The utility of anti-PAR2 antibodies in these techniques was first verified by using cells ectopically expressing epitope tagged PAR2 to permit the detection of the overexpressed receptor and compare with anti-PAR2 antibody staining. Anti- PAR2 antibody specificity was further examined using tissue obtained from wildtype and PAR2 KO mice as well as lung murine fibroblasts (LMF) expressing PAR family members to determine antibody selectivity.

50 3.2 Methods 3.2.1 Expression constructs The human PAR2 open reading frame incorporating 3’ sequence encoding a carboxyl terminal FLAG (DYKDDDDK) was amplified by PCR using Expand High Fidelity polymerase mixture (Roche) and cloned into the pcDNA3.1neo vector (Invitrogen). A construct encoding PAR2 tagged at the carboxyl terminal with green fluorescent protein (GFP) was described previously (Ramsay et al., 2008a).

51 3.3 Results 3.3.1 Three anti-PAR2 antibodies are capable of specifically immunoprecipitating PAR2-FLAG ectopically expressed by CHO cells The specificity of anti-PAR2 antibodies SAM11, H99, C17 and N19 was examined by immunoprecipitation followed by Western blot analysis. First, whole cell lysates from CHO cells transiently expressing PAR2-FLAG were subjected to immunoprecipitation with each anti-PAR2 antibody and species appropriate control immunoglobulins (IgGs). Second, proteins immunoprecipitated by these antibodies were analysed by anti-FLAG Western blot anlysis. As shown in Figure 3.2, anti- PAR2 antibody SAM11 but not control mouse IgG immunoprecipitated transiently expressed PAR2-FLAG as a smear ranging from ~80 kDa to ~250 kDa. Similarly, immunoprecipitation with anti-PAR2 antibodies C17 and N19 also resulted in a smear ranging from ~80 kDa to ~250 kDa. In contrast, immunoprecipitation using anti-PAR2 antibody H99 yielded a similar banding pattern to the control rabbit IgG, except for a ~85 kDa band and a high molecular weight band which may potentially be ectopic PAR2. These data indicate that anti-PAR2 antibodies SAM11, C17 and N19 are capable of immunoprecipitating PAR2 transiently expressed by CHO cells.

3.3.2 Ectopically expressed PAR2 is N-glycosylated The predicted molecular weight of prepro-PAR2 (full length) is 44.1 kDa, while pro- PAR2 is 41.7 kDa and activated PAR2 is 40.5 kD. However, high molecular weight bands were detected by Western blot analysis of PAR2-FLAG immunoprecipitated from lysates by anti-PAR2 antibodies SAM11, C17 and N19. Ectopically expressed PAR2 is post-translationally modified by glycosylation (Compton et al., 2001; Compton et al., 2002) and ubiquitination (Jacob et al., 2005a) which is reported to contribute to high molecular weight protein species. To determine if N-glycosylation contributes to the high molecular weight bands in Figure 3.2, vector and PAR2- FLAG transfected CHO cells were treated in the absence or presence of the N- glycosylation inhibitor tunicamycin for 16 h. Anti-FLAG Western blot analysis was performed on whole cell lysates collected from untreated and tunicamycin treated cells. As shown in Figure 3.3, PAR2-FLAG from untreated cells migrated as a smear from ~70 kDa to ~200 kDa whereas tunicamycin treatment reduced the molecular weight of a portion of this high molecular weight PAR2 ranging from ~37 kDa to ~50 kDa.

Figure 3.2 Three anti-PAR2 antibodies immunoprecipitate PAR2-FLAG ectopically expressed in CHO cells

Lysates of CHO cells transiently expressing PAR2-FLAG were subjected to immunoprecipitation with anti-PAR2 antibodies SAM11 (mouse), C17 (goat), N19 (goat) and H99 (rabbit) or control mouse, rabbit or goat immunoglobulins (IgG). Immunoprecipitated proteins were subjected to anti-FLAG antibody Western blot analysis. The data are representative of three independent experiments. The heavy chain of antibody H99 and the rabbit IgG control immunoprecipitations are apparent because the anti-FLAG antibody was generated in rabbit.

Figure 3.3 Ectopically expressed PAR2-FLAG is N-glycosylated Anti-FLAG and anti-GAPDH Western blot analysis of lysates from CHO cells transiently expressing either vector or PAR2-FLAG. Before collection of lysates cells were either untreated or treated with the N-glycosylation inhibitor tunicamycin. The data are representative of three independent experiments.

3.3.3 Three anti-PAR2 antibodies are capable of detecting ectopically expressed PAR2-FLAG by Western blot analysis Having determined that the anti-PAR2 antibodies SAM11, C17 and N19 are able to efficiently enrich PAR2-FLAG from over-expressing cells, antibody specificity was further assessed by distinct Western blot analysis of whole cell lysates collected from CHO cells transiently transfected with either vector or PAR2-FLAG and subjected to anti-PAR2 SAM11, H99, C17 and N19 and anti-FLAG Western blot analysis. As shown in Figure 3.4, Western blot analysis with anti-PAR2 antibody SAM11 exhibited a banding pattern ranging from ~90 kDa to ~250 kDa, resembling the anti- FLAG antibody Western blot analysis. Minor cross reactivity was observed with a ~75 kDa band also detected in lysates from vector transfected cells. Anti-PAR2 antibody C17 also detected a PAR2-FLAG specific banding pattern. However, the smear observed with C17 ranged from ~37 kDa to ~250 kDa. Non-specific bands at ~70 kDa, ~100 kDa and ~150 kDa were detected with this antibody. Similar to anti- PAR2 antibody C17, N19 Western blot analysis detected a smear ranging from ~30 kDa to ~250 kDa specific to lysates collected from PAR2-FLAG expressing cells. Non-specific bands at ~50 kDa and ~200 kDa were detected with antibody N19. By contrast, Western blot analysis with anti-PAR2 antibody H99 did not detect any bands specific to lysates expressing PAR2-FLAG. Consistent with Western blot analysis of immunoprecipitated PAR2-FLAG, these data indicate that anti-PAR2 antibodies SAM11, C17 and N19 are capable of detecting PAR2-FLAG from whole cell lysates by Western blot analysis.

Although anti-PAR2 antibodies SAM11, C17 and N19 detected ectopic PAR2 from whole cell lysates, whether these antibodies specifically detect the receptor in the correct cellular compartment remains unknown. Therefore as PAR2 is a membrane spanning protein, to determine if anti-PAR2 antibodies detect ectopic PAR2 in cellular membranes, soluble and membrane components were collected from CHO cells transiently transfected with vector or PAR2-FLAG. Soluble and cell membrane fractions were collected by inducing osmotic stress with H2O followed by mechanical disruption; scraping and passing through a 26 gauge needle. Cell membranes were pelleted by ultracentrifugation at 100,000 g and the supernatant

Figure 3.4 Anti-PAR2 antibodies SAM11, C17 and N19 detect ectopically expressed PAR2 by Western blot analysis Western blot analysis of lysates from CHO cells transiently transfected with vector or PAR2-FLAG probed with anti-PAR2 antibodies SAM11, H99, C17 and N19 and anti-FLAG and anti-GAPDH antibodies. The data are representative of three independent experiments.

55 retained as the soluble fraction. Equal amounts of soluble and membrane fractions were analysed by SAM11, H99, C17, N19 and anti-FLAG antibody Western blot analysis. Western blot analysis to detect CAV1, a membrane protein, was also performed to assess for contamination of the soluble fraction with membrane proteins. As shown in Figure 3.5, the anti-PAR2 antibody SAM11 specifically recognised PAR2 ranging from ~65 kDa to ~250 kDa in the membrane fraction of CHO cells ectopically expressing PAR2-FLAG. No non-specific bands were detected from vector transfected CHO cells. Anti-PAR2 antibodies C17 and N19 also detected a prominent smear ranging from ~30 kDa to ~250 kDa in the membrane fraction from PAR2-FLAG expressing CHO cells. However, a less intense non- specific ladder of bands ranging from ~75 kDa to ~250 kDa was detected in the soluble fraction of vector and PAR2-FLAG transfected cells with anti-PAR2 antibody C17. A prominent ~250 kDa band was also detected in the membrane fraction of vector transfected cells by C17. Non-specific signals were also detected by N19 with a ladder of bands detected in the soluble fractions of vector and PAR2- FLAG transfected cells. Non-specific ~50 kDa, ~75 kDa and ~140 kDa bands were also detected in the membrane fraction of vector transfected cells by N19. However, unlike SAM11, C17 and N19, anti-PAR2 antibody H99 did not detect any PAR2 specific bands in the membrane fraction of PAR2 expressing cells, suggesting that this antibody is not suitable for the detection of PAR2 by Western blot analysis. Importantly, anti-CAV1 Western blot analysis showed no contamination of the cytosolic fraction with membrane, further highlighting that the signals detected in cytosolic fractions by C17 and N19 are non-specific. Moreover, the specific PAR2 reactive bands observed with SAM11, C17 and N19 are consistent with the Western blot analyses of lysates from transiently transfected CHO cells shown in Figure 3.4. These results indicate that anti-PAR2 antibodies SAM11, C17 and N19 are capable of detecting ectopically expressed PAR2. Due to the inability to detect ectopically expressed PAR2, H99 was excluded from further Western blot analysis.

3.3.4 Examination of the ability of three anti-PAR2 antibodies to detect endogenously expressed PAR2 by Western blot analysis Having determined that anti-PAR2 antibodies SAM11, C17 and N19 are able to detect ectopically expressed PAR2, the ability of these antibodies to detect the endogenously expressed receptor was next examined using PC-3 and DU145 prostate

Figure 3.5 Anti-PAR2 antibodies SAM11, C17 and N19 detect ectopically expressed PAR2 in membrane preparations by Western blot analysis.

Western blot analysis of soluble (S) and membrane (M) fractions from CHO cells transiently transfected with vector or PAR2-FLAG probed with anti-PAR2 antibodies SAM11, H99, C17 and N19 and an anti-FLAG antibody. Soluble fraction purity was assessed by anti-CAV1 Western blot analysis. The data are representative of three independent experiments.

57 cancer cell lines. Prior to performing anti-PAR2 antibody Western blot analysis, the expression of PAR2 in PC-3 and DU145 cells was examined by qRT-PCR analysis. Expression of PAR2 relative to HPRT1, determined using gene specific primers, was quantified using the ∆∆CT method (ABI User Bulletin #2; Applied Biosystems). As shown in Figure 3.6A, qRT-PCR expression analysis detected PAR2 transcripts in both PC-3 and DU145 prostate cancer cell lines. Relative to the housekeeping gene GAPDH, PAR2 expression in PC-3 cells was ~4 fold higher than in DU145 cells. These data are consistent with previous reports indicating that PC-3 and DU145 cells express functional PAR2 (Mize et al., 2008; Ramsay et al., 2008a).

The specificity of anti-PAR2 antibodies SAM11, C17 and N19 was next tested by Western blot analysis on soluble and membrane fractions collected from PC-3 and DU145 cells. Soluble and cellular membrane fractions were collected to enhance detection of the endogenous receptor by enriching for cellular membranes where PAR2 is expressed. Anti-PAR2 antibody SAM11, C17 and N19 Western blot analysis was performed on equal amounts of soluble and membrane fractions. Anti- CAV1 antibody Western blot analysis was also performed to assess purity of the soluble fractions. As shown in Figure 3.6B, although prominent non-specific signals were detected with SAM11 at ~75 kDa and ~160 kDa in soluble and membrane fractions from PC-3 cells, two membrane specific bands at ~45 kDa and ~250 kDa were detected with this antibody. In contrast, SAM11 Western blot analysis did not detect any membrane fraction specific bands from DU145 cells with non-specific signals detected at ~50 kDa, ~75 kDa and ~160 kDa. Similarly, Western blot analysis with the anti-PAR2 antibody C17 did not yield any membrane specific bands from either DU145 or PC-3 cells. Interestingly, anti-PAR2 antibody N19 Western blot analysis identified a prominent membrane specific band at ~50 kDa and smear ranging from ~50 kDa to ~80 kDa in both PC-3 and DU145 cells (note asterisk in Figure 3.6B). High molecular weight non-specific signals were detected with N19 in the soluble fraction of PC-3 and DU145 cells ranging from ~100 kDa to ~250 kDa. For comparative purposes, densitometric analysis was performed on the membrane specific signals detected with N19 to quantitatively assess protein levels. As shown in Figure 3.6C, this analysis indicated that the intensity of membrane specific bands were ~3 fold higher in PC-3 cells compared with DU145 cells. These protein levels

58 Figure 3.6 PAR2 expression in prostate cancer cell lines and evaluation of anti- PAR2 antibodies by Western blot analysis. A, Graphical representation of PAR2 mRNA expression in PC-3 and DU145 prostate cancer cell lines as assessed by qRT-PCR analysis. Comparative CT values normalised to housekeeping gene HPRT1 are presented as average fold expression relative to expression in PC-3 cells. Values were determined from 3 independent experiments performed in triplicate and are shown as mean ± SEM. B, Western blot analysis of soluble and membrane fractions from PC-3 and DU145 cells probed with anti-PAR2 antibodies SAM11, C17 and N19 and an anti-CAV1 antibody. *, denotes membrane fraction specific smear. S, soluble fraction; M, membrane fraction. C, Graphical representation of densitometry analysis of anti-PAR2 N19 antibody Western blot data comparing membrane fraction specific protein expression in PC-3 and DU145 cells. Values were determined from three independent experiments and are displayed as ± SEM.

59 correspond with the relative PAR2 mRNA expression levels (see Figure 3.6A), providing further support that the membrane specific signals detected by N19 are from endogenous PAR2. Therefore, these data indicate that anti-PAR2 antibody N19 is a suitable reagent to detect endogenous PAR2.

3.3.5 Anti-PAR2 antibody N19 but not SAM11 detects an endogenously expressed N-glycosylated protein Antibodies SAM11 and N19 both detected membrane specific PAR2 signals by Western blot analysis of PC-3 cells. To examine if, like ectopically expressed PAR2, N-glycosylation contributes to molecular weight of the protein species detected by SAM11 and N19, PC-3 cells were treated with N-glycosylation inhibitor tunicamycin for 16 h before collection of cellular components. Soluble and membrane fractions were collected from untreated and tunicamycin treated cells and subjected to SAM11 and N19 Western blot analysis. Western blot analysis for CUB domain containing protein 1 (CDCP1) was also performed to assess soluble fraction purity and ensure tunicamycin activity by serving as a deglycosylation positive control. CDCP1 is a TM cell surface protein and is heavily N-glycosylated with 14 N-glycosylation consensus sites (Hooper et al., 2003). CDCP1 is detected as two bands at ~70 kDa and ~140 kDa by Western blot analysis (He et al., 2010). As shown in Figure 3.7, Western blot analysis using SAM11 detected membrane specific signals at ~40 kDa, ~45 kDa, ~80 kDa, ~100 kDa and ~150 kDa with a non-specific band also found in the soluble fraction at ~75 kDa. However, unlike the tunicamycin-induced shift in molecular weight observed with ecoptically expressed PAR2 (see Figure 3.3), tunicamycin treatment of PC-3 cells did not cause a reduction in the molecular weight of any membrane specific bands detected with SAM11. Western blot analysis with antibody N19 detected a prominent membrane specific ~50 kDa band and a smear ranging from ~37 kDa to ~100 kDa. By contrast with SAM11, tunicamycin treatment of PC-3 cells induced a loss of the membrane specific smear detected by N19. Tunicamycin treatment did not reduce the molecular weight of the ~50 kDa band, suggesting this band is the unglycosylated protein. Importantly, consistent with previous reports, CDCP1 was detected as ~150 kDa and ~70 kDa bands in untreated PC-3 cells (Figure 3.7). Consistent with a previous report (Hooper et al., 2003), tunicamycin treatment of PC-3 cells reduced the molecular weight of full length CDCP1 from ~150 kDa to two bands at ~90 and ~95 kDa.

Figure 3.7 Antibody N19 but not SAM11 detects N-glycosylated PAR2

Soluble and membrane fractions from PC-3 cells treated in the presence or absence of the N-glycosylation inhibitor tunicamycin were analysed by Western blot analysis with anti-PAR2 antibodies SAM11 and N19 and an anti-CDCP1 antibody. S, soluble fraction; M, membrane fraction. The data are representative of three independent experiments.

61 These data confirmed that tunicamycin inhibited N-glycosylation. Therefore, antibody N19 but not SAM11 detects endogenous PAR2. The smear detected by N19 is characteristic of ecoptically expressed PAR2 and likely represents multiple N- glycoslyated forms of endogenous PAR2.

3.3.6 Further examination of anti-PAR2 antibody N19 by Western blot analysis Having determined that antibody N19, but not SAM11 or C17, is able to detect endogenous PAR2 that is sensitive to inhibition of N-glycosylation, further Western blot analysis was performed to examine the specificity of N19 for endogenous PAR2. Two approaches were used: (1) Western blot analysis of tissues from wildtype and PAR2 knockout mice and (2) Western blot analysis of soluble and membrane fractions collected from lung murine fibroblasts (LMF) cells isolated from PAR1 KO mice (referred to as NILF cells) and stably expressing human PAR1, PAR2 or PAR4 (Andrade-Gordon et al., 1999).

Antibody N19 Western blot analysis was performed on soluble and membrane fractions collected from the small intestine and liver of wildtype and PAR2 knockout mice. These tissues were chosen on the basis of previous reports indicating that PAR2 mRNA expression is highest in the small intestine and not detected in mouse liver (Nystedt et al., 1994). Soluble fraction purity was assessed with an anti-pan cadherin antibody. As shown in Figure 3.8A, Western blot analysis with N19 detected a smear ranging from ~40 kDa to greater than ~250 kDa in the membrane fraction of small intestine derived from wildtype mice. A smear ranging from ~60 kDa to greater than ~250 kDa was also detected with N19 in the soluble fraction of small intestine derived from wildtype mice. However, some of this signal will result from contaminating cellular membrane as indicated by the detection of cadherin protein in both soluble and membrane fractions. In the membrane fraction collected from the small intestine of PAR2 knockout mice, Western blot analysis with N19 detected weak signals ranging from ~150 kDa to greater than ~250 kDa. Signals ranging from ~80 kDa to greater than ~250 kDa were also detected in the soluble fraction of small intestine tissue from PAR2 knockout mice. Anti-pan cadherin Western blot analysis did not detect any signal in the soluble fraction from PAR2 knockout small intestine indicating that there was no membrane contamination.

Figure 3.8 Anti-PAR2 antibody N19 Western blot analysis of wildtype and PAR2 knockout mouse tissue and LMF cells selectively expressing PAR family members. A, Antibody N19 Western blot analysis of soluble and membrane fractions of small intestine and liver tissue collected from wildtype and PAR2 knockout mice. Soluble fraction purity was assessed by anti-Pan Cadherin Western blot analysis. B, Antibody N19 and anti-tubulin Western blot analysis of soluble and membrane fractions collected from LMF cells not expressing PARs (NILF) and cells selectively expressing PAR1, PAR2 and PAR4.

63 Interestingly, unlike small intestine tissue from wildtype mice, N19 Western blot analysis did not detect a prominent smear in the membrane fraction of wildtype mouse liver. However, weak signals ranging from ~150 kDa to greater than ~250 kDa were detected by N19 in soluble and membrane fractions from wildtype and PAR2 knockout mouse liver. Although high molecular weight non-specific signals exist for antibody N19, these data suggest that the smear ranging from ~40 kDa to at least ~80 kDa is endogenous PAR2.

The specificity of antibody N19 was next assessed using cells derived from lung fibroblasts from PAR1 KO mice (referred to as NILF cells (Andrade-Gordon et al., 1999)). Variants of NILF cells stably expressing human PAR1, PAR2 and PAR4 (Andrade-Gordon et al., 1999) were also used. Soluble and membrane fractions collected from these LMF cell lines were subjected to Western blot analysis with N19 and anti-tubulin Western blot analysis was performed to assess equal loading. As shown in Figure 3.8B, Western blot analysis with N19 detected bands at ~50 kDa and ranging from ~100 to ~150 kDa in the membrane fraction of NILF cells and cells expressing PAR1, PAR2 or PAR4. Signals were also detected by N19 in the soluble fraction of each cell line with bands ranging from ~60 kDa to greater than ~250 kDa. Importantly, a prominent smear detected in the membrane fraction of LMF-PAR2 cells ranging from ~40 kDa to ~150 kDa was not observed in the membrane fraction of the other lines. Moreover, this smear is consistent with Western blot analysis of ectopically expressed PAR2 (Figures 3.3 and 3.4) and similar to Western blot analysis to detect endogenous PAR2 in DU145 and PC-3 cells (Figures 3.5 and 3.6). These data provide additional support that antibody N19 is able to selectively detect PAR2 from other PAR family members.

3.3.7 Anti-PAR2 antibodies detect ectopically expressed PAR2 by confocal microscopy analysis Previous results sections have assessed anti-PAR2 antibodies by Western blot analysis. In this section, the specificity of antibodies SAM11, H99, C17 and N19 are examined by confocal microscopy analysis. To examine if these antibodies detect PAR2, CHO cells transiently expressing PAR2 tagged at the carboxyl terminal with GFP (Ramsay et al., 2008a) were fixed, permeabilised and stained with anti-PAR2 antibodies and fluorescently conjugated secondary antibodies as well as a DAPI

64 counterstain to observe the nucleus. Stained cells were then imaged on a Leica SP5 confocal microscope. Detection of GFP tagged PAR2 allows visualisation of ectopically expressed PAR2 in the green channel (488 nm excitation) while permitting concomitant assessment of the ability of SAM11, H99, C17 and N19 to detect PAR2 by immunofluorescence in the red channel (568 nm excitation). By this method, antibody specificity was assessed by the overlap between GFP (green) and secondary antibody signals (red) detected as yellow signal in merged images. Importantly, the GFP tag does not affect PAR2 signalling or receptor trafficking, allowing detection of an ectopically expressed PAR2 that localises to similar cellular compartments as the endogenous receptor (Dery et al., 1999; Roosterman et al., 2003). For this experiment, cells expressing PAR2-GFP were fixed, permeabilised and stained with anti-PAR2 antibodies and fluorescently conjugated secondary antibodies as well as a DAPI counterstain to observe the nucleus. Stained cells were then imaged on a Leica SP5 confocal microscope. Antibody specificity was assessed in each image by the overlap between GFP and secondary antibody signal detected as yellow signal in merged images. As shown in Figure 3.9, PAR2-GFP signal was present throughout transfected cells (second column). Similar signal was present in the red channel in cells expressing PAR2-GFP stained with the anti-PAR2 antibodies SAM11 (third column top row), C17 (third column third row) and N19 (third column fourth row). However in contrast, antibody H99 staining was restricted to perinuclear and distinct cytoplasmic structures that did not colocalise with much of the PAR2- GFP signal (third column second row). Merged images confirmed almost complete overlap between regions recognised by the anti-PAR2 antibodies and the PAR2-GFP signal. Importantly, cells not expressing PAR2-GFP (showing as blue nuclei in each image) were not stained by any of the four anti-PAR2 antibodies. Secondary only controls were clear of staining (the anti-mouse secondary is shown in the bottom row; data for anti-rabbit and anti-goat secondary antibodies are not shown). These data suggest that antibodies SAM11, C17 and N19 specifically detect ectopically expressed PAR2 by confocal microscopy analysis whereas H99 may detect only a subset of PAR2 localised within particular cellular compartments.

3.3.8 Anti-PAR2 antibodies SAM11 and N19 detected ectopically and endogenously expressed PAR2 by flow cytometry analyses The ability of antibodies SAM11 and N19 to detect cell surface PAR2 was also

Figure 3.9 Anti-PAR2 antibodies detect ectopically expressed PAR2 by immunofluorescence and Confocal microscopy analysis CHO cells transiently transfected with PAR2-GFP were fixed and permeabilised and then stained with DAPI (blue), the indicated anti-PAR2 and appropriate secondary antibodies (α-PAR2, red). Specificity of the primary antibodies is indicated by the yellow in the merged image, the level of staining observed in the secondary only row and in cells not expressing PAR2-GFP which are those that only have nuclei stained (blue). Scale bars in merged images represent 25 μm. Images are representative of data from four independent experiments.

66 examined by live cell flow cytometry analysis. These antibodies were chosen as SAM11 and N19 have extracellular PAR2 epitopes (see Figure 3.1). To first determine if SAM11 and N19 are capable of detecting cell surface PAR2, CHO cells ectopically expressing PAR2 were analysed. CHO-PAR2 cells were lifted non- enzymatically and stained with anti-PAR2 antibodies (2 µg per 1 x 106 cells) and fluorescently conjugated secondary antibodies. Cells stained with fluorescently conjugated secondary antibodies alone were used as controls. Flow cytometry analysis of CHO-PAR2 cells stained with SAM11 showed a fluorescence shift (grey dashed line) in comparison with cells transfected with vector (grey dotted line) and cells stained with secondary only (grey solid line), indicating detection of ectopically expressed PAR2 (Figure 3.10A). Similarly, flow cytometry analysis of CHO-PAR2 cells stained with N19 (black solid line) also showed a fluorescence shift compared with vector transfected cells (black dotted line) and secondary only stained cells (grey solid line). Importantly, no evidence of non-specific binding was apparent with either SAM11 or N19 from comparing staining of vector transfected cells and secondary only controls. Therefore antibodies SAM11 and N19 are capable of detecting ectopically expressed PAR2 at the cell surface.

The specificity of antibodies SAM11 and N19 for endogenous cell surface PAR2 was next examined using PC-3 cells. As shown in Figure 3.10B, SAM11 (black dashed line) and N19 (black solid line) both displayed fluorescence shifts compared with the secondary only control (grey solid line), indicating detection of an endogenously expressed protein. Notably, the signal with N19 was stronger compared with SAM11 (24.3 ± 0.5 MFI versus 12.3 ± 0.9), suggesting that N19 may be a more sensitive antibody.

As proteolytic activation of PAR2 induces rapid receptor internalisation (Bohm et al., 1996a; Ricks and Trejo, 2009), trypsin treatment was performed to remove PAR2 from the cell surface and determine whether the population of cells detected by flow cytometry with SAM11 or N19 is trypsin-sensitive. CHO-PAR2 and PC-3 cells were lifted non-enzymatically and either untreated or trypsin treated cells (50 nM for 5 min 37°C) were stained with SAM11 or N19 and subjected to flow cytometry analysis. As shown in Figure 3.10C, SAM11 staining of trypsin treated CHO-PAR2 cells (black dotted line) displayed a reduction in fluorescence compared with

Figure 3.10 legend, see over page

Figure 3.10 Antibodies SAM11 and N19 detect ectopic and endogenous PAR2 by flow cytometry analysis. A, Anti-PAR2 flow cytometry analysis of non-permeabilised CHO cells expressing either vector or PAR2 using antibodies N19 and SAM11. MFI values: N19 51.9 ± 2.5; SAM11 41.6 ± 3.2; signal from incubation of CHO-PAR2 cells with secondary antibody (2º), CHO-vector with antibody N19 (Vector N19) and CHO-vector with antibody SAM11 (Vector SAM11) were below 100. B, Flow cytometry analysis of cell surface PAR2 endogenously expressed by non-permeabilised PC-3 cells using N19 and SAM11 antibodies. MFI values: N19 24.3 ± 0.5; SAM11 12.3 ± 0.9; secondary antibody (2º) 0.60 ± 0.06. C, Antibody SAM11 flow cytometry analysis of non-permeabilised CHO-PAR2 (left) and PC-3 (right) cells treated in the presence or º absence of trypsin (50 nM) for 15 min at 37 C. MFI values: untreated CHO-PAR2 cells 44.2 ± 3.4; treated CHO-PAR2 cells 34.7 ± 1.9; untreated PC-3 cells 15.7 ± 2.3; treated PC-3 cells 8.4 ± 0.7. D, Antibody N19 flow cytometry analysis of non- permeabilised CHO cells expressing PAR2 and PC-3 cells treated in the presence or absence of trypsin (50 nM) for 5 min at 37°C. MFI values: untreated CHO-PAR2 cells 48.6 ± 0.7; treated CHO-PAR2 cells 23.1 ± 1.8; untreated PC-3 cells 26.8 ± 2.1; treated PC-3 cells 7.4 ± 0.8. Data in each panel are representative of 3 independent experiments.

69 untreated CHO-PAR2 cells stained with SAM11 (black solid line)(44.2 ± 3.4 MFI versus 34.7 ± 1.9 MFI), indicating less PAR2 at the cell surface following activation with trypsin. Similarly, SAM11 staining of trypsin treated PC-3 cells (black dotted line) also showed a population shift compared with untreated PC-3 cells stained with SAM11 (black solid line)(15.7 ± 2.3 MFI versus 8.4 ± 0.7), indicating that SAM11 also detects a trypsin-sensitive population of cells. Consistently, trypsin treated CHO-PAR2 and PC-3 cells (black solid line) also displayed reduced cell surface staining with antibody N19 compared with untreated CHO-PAR2 and PC-3 cells (black dotted line)(CHO-PAR2: 48.6 ± 0.7 MFI versus 23.1 ± 1.8; PC-3: 26.8 ± 2.1 MFI versus 7.4 ± 0.8; Figure 3.10D). These data indicate that SAM11 and N19 detect both ectopic and endogenous cell surface PAR2 by flow cytometry. Of note, although both antibodies detect cell surface PAR2, N19 staining of trypsin treated cell lines compared with untreated cells displayed a more pronounced reduction in cell surface protein than SAM11 stained cells (PC-3 treated cells: 53.5% reduction versus 27.6% reduction in surface PAR2).

70 3.4 Discussion To understand the role of PAR2, specific anti-PAR2 antibodies are required. This chapter has examined the ability of four commercially available antibodies to detect ectopically and endogenously expressed PAR2. The specificity of anti-PAR2 antibodies SAM11, H99, C17 and N19 was examined by immunoprecipitation, Western blot analysis, confocal microscopy and flow cytometry analysis. The key findings from this chapter are that anti-PAR2 antibodies SAM11, C17 and N19 are able to immunoprecipitate ectopically expressed PAR2 and efficiently detect ectopic PAR2 by Western blot analysis. Of these antibodies, only N19 is able to detect endogenous PAR2 by Western blot analysis and this receptor appears as a smear ranging from ~40 kDa to ~80 kDa that is sensitive to inhibition of N-glycosylation. Further evaluation of N19 specificity and selectivity using tissues from wildtype and PAR2 KO mice and LMF cells individually expressing PAR family members confirmed that the smear detected by this antibody is endogenous PAR2. In addition, anti-PAR2 antibodies SAM11, C17 and N19 were able to efficiently detect ectopic PAR2 by immunofluorescence. Importantly, anti-PAR2 antibodies SAM11 and N19 were also able to detect ectopic and endogenously expressed PAR2 by flow cytometry.

To first address whether anti-PAR2 antibodies were capable of detecting PAR2, cells ectopically expressing the receptor tagged (FLAG) at the carboxyl terminal were used. This approach enables comparison of staining obtained with validated anti-tag antibodies with staining achieved using anti-PAR2 antibodies. Using this approach, an important finding from this chapter is the ability of anti-PAR2 antibodies SAM11, C17 and N19 to detect ectopically expressed PAR2 by immunoprecipitation and Western blot analysis (Figure 3.2). Interestingly, for anti-PAR2 antibody H99, while much of the banding pattern was largely similar to the isotype control antibody, two PAR2-FLAG specific bands were detected (~85 kDa and a high molecular weight band), suggesting that this antibody may potentially detect ectopically expressed PAR2. Further analysis is required to determine if these bands are indeed ectopically expressed PAR2.

Importantly, Western blot analysis of ectopic PAR2 with antibodies SAM11, C17 and N19 detected the receptor as a smear at molecular weights greater than the

71 predicted molecular weight (44 kDa; Figures 3.2, 3.4 and 3.5). This is consistent with a previous report that also detected overexpressed PAR2 as a smear but ranging from ~55 kDa to ~100 kDa (Compton et al., 2002). This has also been reported for other GPCRs including the free fatty acid receptor GPR120 (Miyauchi et al., 2009) and endogenous PAR1 (Russo et al., 2009). Of interest, N-glycosylation is the dominant cause for high molecular weight species of PAR2 (Compton et al., 2002) as mutagenesis of the two PAR2 N-glycosylation sites (N30 and N222) reduces the PAR2 smear ranging from ~55-100 kDa to 33-48 kDa (Compton et al., 2002). Consistently, in this chapter inhibition of N-glycosylation by tunicamycin treatment induced the emergence of a prominent smear ranging from ~37-50 kDa (Figure 3.3). Interestingly, although tunicamycin treatment did not cause a complete reduction in the molecular weight for ectopically expressed PAR2, the remaining portion of unaffected high molecular weight PAR2 may result from other post-translational modifications such as ubiquitination. Although not examined in this chapter, ubiquitination has previously been reported to contribute to high molecular weight ectopically expressed PAR2 with smears ranging from ~80 kDa to ~300 kDa (Hasdemir et al., 2007; Hasdemir et al., 2009; Jacob et al., 2005a). Another possibility is that these high molecular weight PAR2 bands may also arise from receptor oligomerisation. In this regard, GPCRs are reported to form hetero- or homo-oligomers that can be identified by high molecular weight protein species in Western blot analysis (Ciruela et al., 2010). For example, Western blot analysis of the ectopically expressed 5HT1A and 5HT1D receptors (predicted monomer molecular weight of 46 kDa and 42 kDa respectively) identified oligomers migrating as ~80 kDa bands (Salim et al., 2002). Similarly, Western blot analysis for the histamine H4 receptor also identified high molecular weight bands by Western blot analysis that are due to hetero- and homo-oligomers (van Rijn et al., 2006). As PAR1 and PAR4 have also been shown to form receptor hetero-oligomers (Leger et al., 2006), it is possible that PAR2 forms receptor oligomers which may be observed via Western blot analysis.

Although antibodies SAM11, C17 and N19 were capable of detecting ectopic PAR2, N19 was the only reagent to detect the endogenous receptor by Western blot analysis (Figure 3.6, 3.7 and 3.8). Anti-PAR2 antibody C17 Western blot analysis for endogenous PAR2 in PC-3 and DU145 prostate cancer cell lines did not yield any

72 membrane fraction specific bands. Also, although membrane specific bands at ~45 kDa and ~250 kDa were identified by SAM11, these molecular weights differ to the species commonly reported for this antibody at ~65, ~85 and ~105 kDa (Bolton et al., 2003; Ge et al., 2003; Molino et al., 1997; Olianas et al., 2007). Ge and colleagues also report these molecular weight bands reduce to ~55 kDa following tunicamycin treatment (2004). In contrast, the molecular weight bands detected by SAM11 in this chapter were not sensitive to N-glycosylation inhibition (Figure 3.7). Unlike SAM11, anti-PAR2 antibody N19 detected a prominent tunicamycin sensitive smear in the membrane fraction from PC-3 cells (Figure 3.7). Importantly, the tunicamycin sensitive smear detected by N19 in PC-3 cells was also observed in the membrane fraction of wildtype but not PAR2 knockout mouse tissue (Figure 3.8).

Futhermore, this chapter also demonstrated that each anti-PAR2 antibody is able to detect ectopic PAR2 by immunofluorescence. Consistent with Western blot analysis of ectopic PAR2, immunofluorescence with SAM11, C17 and N19 indicated that these antibodies efficiently detect overexpressed PAR2 by confocal microscopy throughout all regions of the cell (Figure 3.9). In comparison, it appears that H99 only specifically detected the intracellular peri-nuclear fraction of ectopic PAR2. The reason why antibody H99 only detects a subset of ectopic PAR2 is not known but it is possible that the H99 epitope is masked during certain stages of cellular processing. In this regard, the selective detection of PAR2 within peri-nuclear cellular compartments such as the endoplasmic reticulum or Golgi apparatus during secretory trafficking may have experimental benefit. Future experiments are required to observe if these antibodies are also capable of detecting the endogenous receptor by immunofluorescence.

In addition, the specificity of antibodies SAM11 and N19 was examined by flow cytometry (Figure 3.10). This indicated that while both antibodies were capable of detecting natively folded cell surface receptor, N19 proved more sensitive than SAM11 at detecting ectopic and endogenous PAR2. Nevertheless, detection of cell surface localised PAR2 with SAM11 is reported in numerous cell types including prostate cancer cell lines (Mize et al., 2008; Ramsay et al., 2008a), dendritic cells (Csernok et al., 2006) and primary chondrocytes (Ferrell et al., 2010). Moreover, in line with the role for PAR2 in inflammation, increased cell surface PAR2 is also

73 reported using SAM11, in polymorphonuclear neutrophils following fungi exposure (Moretti et al., 2008) and in osteoarthritic osteoblasts compared with non-diseased osteoblasts (Amiable et al., 2009). Recently, anti-PAR2 antibody N19 has been used to follow variations in the level of cell surface receptor following PAR2 agonist treatment in HT29 colon cancer cells (Suen et al., 2011). Collectively, these reports further suggest that SAM11 and N19 are suitable reagents to detect PAR2 at the cell surface by flow cytometry.

In summary, the specificity of four anti-PAR2 antibodies has been examined. Data presented in this chapter demonstrates that while anti-PAR2 antibody N19 is suitable for the detection of endogenous PAR2 by Western blot analysis, SAM11 and N19 are both suitable antibodies to detect endogenous cell surface PAR2 by flow cytometry. As these antibodies are capable of detecting natively folded PAR2, future analyses of SAM11 and N19 are required to verify the ability of these reagents to detect this receptor by other techniques such as immunohistochemistry. Importantly, this chapter highlights the importance of systematic analysis to assess the specificity of antibodies.

Chapter 4:

The role of palmitoylation in signalling, cellular trafficking and plasma membrane localization of PAR2

76 4.1 Introduction In this chapter, the role of the post translational modification palmitoylation in regulating PAR2 initiated signal transduction and trafficking is examined. Importantly, having determined in Chapter 3 that anti-PAR2 antibody N19 is suitable for detection of endogenous PAR2, this antibody is used to examine the role of palmitoylation in the regulation of this receptor.

Post translational modifications, such as phosphorylation and palmitoylation, are key modulators of GPCR function (Qanbar and Bouvier, 2003; Tobin and Wheatley, 2004). For PAR2, post translational modifications contribute to the rapid arrest of receptor signalling and also initiate its internalisation following its irreversible activation (Grimsey et al., 2011). In particular, immediately following receptor activation, serine and threonine residues in the PAR2 carboxyl terminus are phosphorylated and intracellular lysine residues are modified by addition of ubiquitin (Jacob et al., 2005a; Ricks and Trejo, 2009). These modifications enable rapid recruitment of β-arrestin-1 and -2 from cytoplasmic stores to the cell surface. These scaffold proteins stably associate primarily with the intracellular carboxyl terminus of PAR2 at the plasma membrane to facilitate receptor desensitisation and internalisation (Jacob et al., 2005a; Oakley et al., 2001; Ricks and Trejo, 2009; Stalheim et al., 2005). Following rapid internalisation, β-arrestins redistribute with PAR2 to the early and late endosomes but do not associate with the receptor during transport to the lysosomes, where sustained PAR2 signalling is halted by degradation (Bohm et al., 1996a; Hasdemir et al., 2009).

A consequence of irreversible activation and rapid desensitisation and degradation is that large intracellular PAR2 stores are required to rapidly replenish the cell surface with nascent receptors thereby re-establishing the ability of cells to respond to proteolytic activity (Bohm et al., 1996a). During this process the GTPase rab11a participates in intracellular trafficking of PAR2 within the Golgi apparatus and toward the plasma membrane (Roosterman et al., 2003). Rab11a localises to recycling enodsomes and the Golgi apparatus to modulate the membrane traffic of cargo, including other GPCRs such as the human prostacyclin receptor (hIP)(Hamelin et al., 2005; Parent et al., 2009; Reid et al., 2010; Takahashi et al., 2007). Interestingly, the direct interaction of rab11a with hIP is regulated by receptor

77 palmitoylation (Reid et al., 2010). As such, palmitoylation is a key modification regulating the trafficking of GPCRs to and from the plasma membrane (Escriba et al., 2007; Qanbar and Bouvier, 2003; Tobin and Wheatley, 2004).

Palmitoylation is a dynamic and reversible modification that occurs commonly for GPCRs on one or more carboxyl terminal cysteines found 10 to 14 residues following the seventh TM domain (Probst et al., 1992). In addition to these thioester linkages (so called S-palmitoylation), there are a small number of examples where palmitate addition to cysteine is followed by structural rearrangement leading to palmitate modification of an amide (N-palmitoylation) (Magee and Courtneidge, 1985). Palmitoylation anchors the GPCR carboxyl domain to the inner leaflet of cell membranes and has diverse effects on GPCR function including regulation of cellular trafficking (Escriba et al., 2007; Qanbar and Bouvier, 2003). In this regard, palmitoylation for many GPCRs is required for efficient localisation and trafficking to the plasma membrane. Mutation of the consensus palmitoylation site in the chemokine CCR5 (Percherancier et al., 2001), vasopressin V2 (Sadeghi et al., 1997; Schulein et al., 1996), histamine H2 (Fukushima et al., 2001a) and δ-opioid receptors (Petaja-Repo et al., 2006) reduces the number of receptors reaching the cell surface. However the mechanisms regulating this process are not completely defined. Similarly, palmitoylation can also alter the endocytosis of some receptors, but the impact of this modification varies between GPCRs. For example, mutation of the palmitoylation sites in the vasopressin V2 receptor reduces agonist-induced internalisation resulting from less recruitment of β-arrestin-2 (Charest and Bouvier, 2003). In contrast, a palmitoylation-deficient mutant of the serotonin 4A receptor demonstrates enhanced β-arrestin-2-dependent endocytosis over the wildtype receptor (Ponimaskin et al., 2005).

Palmitoylation also regulates GPCR signalling by influencing G protein coupling. Palmitoylation stabilises the formation of an eighth helix termed the α-H8 (Palczewski et al., 2000). Formation of this domain by palmitoylation is required for efficient G protein binding and coupling with signal transduction pathways (Ahn et al., 2010; Delos Santos et al., 2006; Katragadda et al., 2004; Palczewski et al., 2000). For example, impaired signalling is reported for the palmitoylation-deficient mutant of the β2-adrenergic receptor (O'Dowd et al., 1989). In addition, loss of

78 palmitoylation sites and portions of the carboxyl terminal α-H8 of PAR1 (Swift et al., 2006), β1-adrenergic receptor (Delos Santos et al., 2006) and the CB1 cannabinoid receptor (Anavi-Goffer et al., 2007) also impairs G-protein coupling and reduces signal transduction.

Of relevance to this chapter, PAR2 has recently been shown to be palmitoylated at C361 which is found 14 residues following the seventh TM domain of this receptor (Botham et al., 2011). Palmitoylation of PAR2 is required for efficient signal transduction and trafficking from the cell surface (Botham et al., 2011). Perplexingly, Botham et al. further report that palmitoylation-deficient PAR2 is more highly expressed on the cell surface than wildtype receptor (2011). These data are in contrast with all other palmitoylated GPCRs (Escriba et al., 2007; Qanbar and Bouvier, 2003; Tobin and Wheatley, 2004).

In this chapter, palmitoylation of PAR2 is demonstrated by two in vitro methods: (i) acyl-biotinyl exchange (ABE; (Drisdel and Green, 2004)) and (ii) copper-catalysed azide-alkyne cycloaddition (click) chemistry (Martin and Cravatt, 2009). ABE enables the purification of palmitoylated proteins from lysates by removing endogenous palmitate moieties and replacing these with a thiol-specific biotin- containing reagent. In the second approach, palmitoylated proteins are examined by metabolic labelling of cells with the palmitate analogue 17-ODYA before reaction of a biotinylation reagent to 17-ODYA via click chemistry. Palmitoylated proteins are identified in both methods by capture of biotinylated proteins with streptavidin- agarose and detection by Western blot analysis. The impact of palmitoylation on PAR2 is also examined using site-directed mutagenesis and pharmacological inhibition of proteome palmitoylation by 2-bromo palmitate (Jennings et al., 2009). Using these approaches, the function of palmitoylation in regulating PAR2 trafficking, signalling and degradation is examined and these data are compared and contrasted with the recent paper from Botham et al. (2011).

79 4.2 Methods 4.2.1 Expression constructs and mutagenesis The human PAR2 open reading frame incorporating 3' sequence encoding a carboxyl terminal myc epitope (EQKLISEEDL) was amplified from a previously described construct (Ramsay et al., 2008a) by PCR using Expand High Fidelity polymerase mixture (Roche) and cloned into the pIRESneo2 vector (Clontech, Scientifix Pty Ltd, Clayton, Australia). Site-directed mutagenesis, to mutate this PAR2-myc construct at cysteine 361 to alanine (C361A) was performed using the primers listed in Table 2.1 and Pfu Ultra polymerase (Agilent Technologies, Forest Hill, Australia). An expression construct encoding PAR2 tagged with a carboxyl terminal monomeric Cherry (mCherry) was generated from the PAR2-GFP construct (Ramsay et al., 2008a). The GFP encoding fragment was removed from PAR2-GFP by restriction digestion using BamHI and NotI (New England Biolabs) and this was replaced with mCherry encoding DNA (Shaner et al., 2005). The sequence of all constructs was confirmed by DNA sequencing at the Australian Genome Research Facility (St. Lucia, Australia). Expression constructs encoding amino terminal GFP tagged wildtype rab11a and dominant negative mutant rab11a-S25N were from Dr Marci Scidmore (Cornell University) (Rzomp et al., 2003). Expression constructs encoding carboxyl terminal GFP tagged β-arrestin-1 & -2 were from Dr Robert Lefkowitz (Duke University Medical Center) (Wei et al., 2003).

80 4.3 Results 4.3.1 Palmitoylation is required for efficient cell surface localisation of endogenous PAR2 Shown in Figure 4.1A (upper panel), is an alignment of amino acids from TM domain 7 to the carboxyl terminal of two known palmitoylated GPCRs, vasopressin V2 receptor (V2R) (Sadeghi et al., 1997) and β2-adrenergic receptor (β2R) (Moffett et al., 1993), and the four members of the PAR family. The alignment shows that C361 of PAR2 aligns with the palmitoylation sites of V2R (a di-cysteine) and β2R (a mono-cysteine). Recently, Botham and co-workers have demonstrated that C361 of PAR2 (Figure 4.1A schematic), is palmitoylated in CHO-Pro5 cells stably expressing this receptor (2011).

As palmitoylation is known to impact on cell surface localization of other GPCRs (Qanbar and Bouvier, 2003), the effect of 2-bromopalmitate on the plasma membrane location of PAR2 was examined using prostate cancer PC-3, DU145 and 22Rv1 cells that are known to endogenously express this receptor (Mize et al., 2008; Ramsay et al., 2008a). Each prostate cancer cell line was treated with vehicle (DMSO) or 2-bromopalmitate (100 µM) for 16 h before flow cytometry analysis using anti-PAR2 antibody N19. As shown in Figure 4.1B, consistent with data from Botham et al., (2011) showing that ectopically expressed PAR2 is palmitoylated, 2- bromopalmitate reduced cell surface expression of PAR2 in PC-3, DU145 and 22Rv1 cells by ~50%.

4.3.2 Generation of CHO cells stably expressing wildtype or C361A mutant PAR2 To directly examine the mechanisms regulating palmitoylation of PAR2 and the role of this modification in receptor signalling and trafficking, CHO cells stably expressing either the wildtype receptor (PAR2-myc) or a mutant form lacking C361 (PAR2-C361A-myc) were generated. To assess the expression levels of wildtype and palmitoylation-deficient PAR2-myc, whole cell lysates were collected from each stably expressing cell line and subjected to Western blot analysis. As shown in Figure 4.2, anti-myc Western blot analysis detected wildtype and palmitoylation- deficient PAR2-myc as a smear ranging from ~30 kDa to ~250 kDa and indicated that mutant PAR2 is expressed at similar levels to the wildtype protein. Consistent

Figure 4.1 legend, see over page

Figure 4.1 Inhibition of palmitoylation reduces cell surface expression of endogenous PAR2 A, Upper panel; sequence alignment of the carboxyl terminal residues of arginine vasopressin receptor-2 (V2R; residues 329-371), β2-adrenergic receptor (β2R; residues 329-413), PAR1 (residues 375-425), PAR2 (residues 347-397), PAR3 (residues 362-374) and PAR4 (residues 344-385). The known palmitoylation sites of V2R, β2R, PAR2 and the conserved palmitoylation site of PAR1 are boxed. Lower panel; schematic representation of the structure of human PAR2 including the TL sequence (underline) within the amino terminal domain, C361 within a consensus palmitoylation motif present in the carboxyl terminal domain, seven TM (TM1 to TM7) domains, and a disulfide bond linking TM3 and the second extracellular loop (dotted line). B, Graphical representation of the effect of blocking palmitoylation on cell surface expression of PAR2. Plasma membrane levels of PAR2, expressed endogenously by PC3, DU145 and 22Rv1 cells, were determined by flow cytometry using the anti-PAR2 N19 antibody. Non-permeabilised cells were either treated with DMSO (negative control) or 2-bromopalmitate (+2BP) for 16 h. Secondary antibody only MFI values were subtracted from N19 values before calculation of the level of cell surface PAR2 present on 2BP treated cells relative to DMSO treated cells. Values were determined from 3 independent experiments and are shown as mean ± SD. **, P < 0.001.

Figure 4.2 Characterisation of CHO cells stably expressing wildtype or C361A mutant PAR2 Anti-myc and anti-GAPDH Western blot analysis of lysates from CHO cells stably expressing either vector, PAR2-myc or PAR2-C361A-myc. Before collection of lysates cells were either untreated (-) or treated (+) with the N-glycosylation inhibitor tunicamycin. *, non-specific band.

84 with Chapter 3 (see Figure 3.2), Western blot analyses of lysates collected from these cells treated with the N-glycosylation inhibitor tunicamycin, indicated that wildtype PAR2 and the receptor mutated at C361 each carry similar levels of N-linked glycans (Figure 4.2).

4.3.3 PAR2 is palmitoylated on C361 Having determined that the CHO cells stably express equivalent levels of wildtype and PAR2-myc C361A, these cells were used in two approaches to assess palmitoylation of PAR2 at C361; ABE and click chemistry. In the first approach, using ABE chemistry, free cysteines on proteins from membrane preparations were first blocked using N-ethylmaleimide (NEM) (Figure 4.3A). Cysteine-palmitoyl thioester bonds were then specifically cleaved using the weak nucleophile hydroxylamine which breaks these bonds but not amide-palmitoyl linkages (Magee and Courtneidge, 1985). Liberated cysteines were then reacted with sulfhydryl reactive EZ-link HPDP-biotin to label proteins that had originally contained palmitoyl moieties (Figure 4.3A). Biotinylated proteins were isolated using streptavidin beads and the level of palmitoylated PAR2 present in this fraction was assessed by anti-myc Western blot analysis. In the second approach, using click chemistry, PAR2-myc and PAR2-C361A-myc CHO cells were treated with an inhibitor of endogenous palmitate synthesis, cerulenin, before metabolic labelling with the alkyne containing palmitate analog 17-ODYA (Figure 4.3B). Membrane preparations were collected and biotin-azide reacted to 17-ODYA labelled proteins via copper catalysed alkyne-azide cycloaddition (click) chemistry. Biotinylated proteins were isolated using streptavidin beads and PAR2 palmitoylation was detected from bead elutes by anti-myc Western blot analysis (Figure 4.3B)

As shown in Figure 4.3C, using ABE chemistry, PAR2-myc was detected as a smear ranging from ~30 kDa to ~250 kDa (same as Figure 4.2) indicating that this protein is palmitoylated. The effect of the palmitoylation inhibitor 2-bromopalmitate on PAR2 palmitoylation was next assessed. This inhibitor functions by inhibiting the enzymes required for palmitoylation, palmitoyl acyltransferases (PATs; (Jennings et al., 2009)). CHO cells stably expressing wildtype or palmitoylation-deficient PAR2- myc were treated with 2-bromopalmitate for 16 h prior to collection of lysates to assess PAR2 palmitoylation by the ABE approach. Consistently, this Western blot

Figure 4.3 legend, see over page.

86 Figure 4.3 PAR2 is palmitoylated on cysteine 361

A, Acyl-biotinyl exchange (ABE); free thiols on membrane proteins are first blocked with NEM (1) and endogenous palmitoyl groups are removed using hydroxylamine (2). A sulphydryl-reactive biotin moiety (HPDP-biotin) is reacted to the liberated cysteine followed by detection of biotinylated proteins by isolation with streptavidin beads and Western blot analysis of the eluted proteins (3). Figure adapted from Fukata and Fukata (2010). B, Click chemistry; production of endogenous cellular palmitate is inhibited by cerulenin to allow efficient metabolic labelling of cells with palmitate analogue, 17-ODYA. Following in situ incorporation of 17-ODYA into proteins (1), lysates are collected and biotin azide is reacted to the alkyne group of 17-ODYA via click chemistry (2). Biotinylated proteins are isolated with streptavidin beads followed by elution and detection by Western blot analysis (3). Figure adapted from Martin and Cravatt (2009). C, Analysis of PAR2 palmitoylation using ABE. Membrane fractions were collected from CHO-PAR2-myc and CHO-PAR2-C361A- myc cells either untreated (-) or treated (+) with 2-bromopalmitate (2BP) for 16 h. ABE was performed on the collected membrane fractions and recovered biotinlyated protein was subjected to anti-myc Western blot analysis to examine PAR2 palmitoylation. D, Analysis of PAR2 palmitoylation by examining cells metabolically labelled with an alkyne containing palmitate analogue (17-ODYA) followed by click chemistry. PAR2-myc and PAR2-C361A-myc CHO cells were incubated with 2-bromopalmitate for 16 h and cerulenin (5 µg/ml) for 1 h then labelled with 17-ODYA for 4 h, before collection of membrane preparations and reaction of biotin-azide to 17-ODYA labelled proteins via click chemistry. Biotinylated proteins were isolated using streptavidin beads and PAR2 palmitoylation was assessed by anti-myc Western blot analysis from bead elutes. E, Analysis of whether palmitoylation of PAR2 occurs exclusively on a cysteine residue and not via an amide linkage. Membrane preparations were collected from CHO- PAR2-myc and CHO-PAR2-C361A-myc 17-ODYA labelled cells. Labelled proteins were reacted with biotin-azide via click chemistry before incubation with 0.7 M hydroxylamine (H) to cleave thioester linkages or 50 mM tris pH 7.4 (T; control). Biotinylated proteins were isolated using streptavidin beads and PAR2 palmitoylation was assessed by anti-myc Western blot analysis from bead elutes.

87 analysis also showed that the palmitoylation inhibitor 2-bromopalmitate completely blocked palmitoylation of PAR2. In contrast with wildtype PAR2, the C361 mutant was not pamitoylated in the presence or absence of 2-bromopalmitate confirming, as demonstrated by Botham et al. (2011) that palmitoylation occurs at C361 (Figure 4.3).

As shown in Figure 4.3D, use of click chemistry to analyse PAR2 palmitoylation confirmed that wildtype but not C361A PAR2 is palmitoylated and that palmitoylation of wildtype PAR2 is blocked by the palmitoylation inhibitor 2- bromopalmitate. To confirm that palmitoylation of PAR2 occurs exclusively on a cysteine residue, and not via an amide linkage that can sometimes occur (Linder and Deschenes, 2007), biotinylated proteins were treated with the weak nucleophile hydroxylamine or, as a negative control, Tris, before isolation of proteins containing biotin using streptavidin beads. In contrast with the ABE chemistry approach, which is not capable of detecting amide-palmitate linkages, this modification to the click chemistry protocol permits the distinction between S-palmitoylation and N- palmitoylation. As shown in Figure 4.3E, hydroxylamine but not Tris treatment completely removed all palmitoylation of PAR2 indicating that this receptor is exclusively palmitoylated on a cysteine residue. Consistent with the report of Botham and colleagues (Botham et al., 2011), these data indicate that PAR2 is palmitoylated at C361. These data also indicate that this is the exclusive site at which PAR2 is palmitoylated.

4.3.4 Palmitoylation of PAR2 is required for efficient downstream signal transduction It has been reported for other GPCRs that palmitoylation regulates initiation of downstream signal transduction cascades (Anavi-Goffer et al., 2007; Delos Santos et al., 2006; O'Dowd et al., 1989; Swift et al., 2006). To directly examine the effect of the mutation of PAR2 C361 on signal transduction initiated via cleavage of this receptor by trypsin, calcium mobilisation assays were performed. As shown in Figure 4.4, in response to trypsin PAR2-C361-myc was ~9 fold less potent than PAR2-myc

(half-maximal response (EC50) of 44.3 versus 5.3 nM) with a reduction of about 33% in the maximum signal induced via the mutant receptor. These data are comparable with the findings of Botham et al. (PAR2 EC50 11.4 nM and PAR2- C361A EC50

Figure 4.4 Palmitoylation is required for PAR2 to efficiently stimulate calcium mobilisation CHO cells stably transfected with PAR2-myc and PAR2-C361A-myc were loaded with fluorescent calcium reporter FURA-2 and treated with increasing concentrations of trypsin (0.1-1000 nM). Fluorescence was measured using a BMG Polarstar fluorescent plate reader. Fluorescence at 510 nm was measured after sequential excitation at 340 nm (unbound FURA-2) and 380 nm (bound FURA-2). The ratio of unbound and bound FURA-2 is proportional to the intracellular concentrations of calcium ions. Comparison of dose-response curves for intracellular calcium mobilisation mediated by trypsin activation of PAR2-myc (○, EC50 5.3 ±0.5 nM) and PAR2-C361A-myc (□, EC50 44.3 ±0.8 nM). Experiments were performed in triplicate on 3 separate occasions and values are Mean +/- SD.

89 40.8 nM) and indicate that palmitoylation of C361 is essential for efficient PAR2 signal transduction (Botham et al., 2011).

4.3.5 Palmitoylation of PAR2 at C361 is required for efficient cell surface localization To examine whether the reduced signalling efficiency of PAR2-C361A is due to a reduction in the level of PAR2 expressed on the plasma membrane, the localization of wildtype and mutant PAR2 expressed by CHO cells was analysed. Confocal microscopy was performed on CHO cells transiently co-transfected with wildtype PAR2-mCherry and PAR2-GFP or wildtype PAR2-mCherry and PAR2-C361A- GFP. As indicated by the yellow signal from the merged images shown in Figure 4.5A, wildtype PAR2-GFP (green) and PAR2-mCherry (red) co-localised throughout subcellular compartments and the plasma membrane. In contrast, PAR2-C361A-GFP (green) co-localised with wildtype PAR2-mCherry (red) in large perinuclear compartments (yellow in merged image) but not at the cell surface where signal was predominantly from wildtype PAR2 (Figure 4.5B, red signal in merged image). These data contrast with the findings of Botham and co-workers who showed by flow cytometry that PAR2-C361A is expressed at higher levels than PAR2 on the surface of stably expressing by CHO-Pro5 cells (2011). However, the qualitative data shown in Figure 4.5 from confocal microscopy analysis were confirmed by quantitative analysis using flow cytometry and cell surface biotinylation assays. As shown in Figure 4.6A, flow cytometry analysis using anti-PAR2 antibody N19 indicated that PAR2-C361A levels on the cell surface were ~70% lower than wildtype PAR2 levels. The effect of C361 mutation on cell surface expression of PAR2 was about the same as 2-bromopalmitate inhibition of palmitoylation of PAR2 (Figure 4.6A). Consistent data were obtained by examining plasma membrane expression of PAR2 by cell surface biotinylation. As shown in Figure 4.6B (left), anti-myc Western blot analysis of intracellular (IC) and plasma membrane (PM) fractions collected from cell surface biotinylated CHO-vector, CHO-PAR2-myc and CHO-PAR2-C361A-myc cells showed much lower levels of PAR2-C361-myc in the PM fraction compared with PAR2-myc. Graphical representation of signals obtained from 3 independent experiments showed that cell surface levels of PAR2-C361-myc were 71.6 ± 4.4% lower than PAR2-myc (Figure 4.6B, right).

Figure 4.5 Palmitoylation of PAR2 at C361 is required for efficient cell surface receptor localization A, Images of CHO cells transiently co-transfected for 12 h with PAR2-mCherry (red) and PAR2-GFP (green). The merge image shows colocalization (yellow) of these two proteins. B, Images of CHO cells transiently co-transfected for 12 h with PAR2- mCherry and PAR2-C361A-GFP. The merge image shows colocalization (yellow) of these two proteins and regions where only PAR2-mCherry is expressed (red). Cells were analysed using a Zeiss LSM510 confocal microscope (63x oil immersion objective lens) and images were processed using MetaMorph software and displayed using CorelDraw X5. Images of cells are representative of three independent experiments. Scale bar, 10 μM. Confocal microscopy was performed together with Dr Melinda Christensen.

91 Figure 4.6 Quantitative analysis of the effect of palmitoylation on PAR2 cell surface localisation A, (Upper panel) Graphical representation of MFI values from N19 antibody flow cytometry analysis of CHO-PAR2-myc and -PAR2-C361A-myc cells treated with DMSO (negative control) or 2BP for 16 h. MFI values were calculated by subtracting secondary antibody only values from N19 values. (Lower panel) Anti- myc Western blot analysis of whole cell lysates to demonstrate expression of PAR2- myc. B, Examination of plasma membrane (PM) PAR2 levels by cell surface biotinylation. PM and intracellular (IC) fractions were collected from CHO-vector, - PAR2-myc and -PAR2-C361A-myc cells reacted with membrane impermeant NHS- EZ-link biotin. The PM containing fraction was collected using streptavidin beads. (Left panel) Anti-myc, -GAPDH (control for IC fraction) and -β1 integrin (control for PM fraction) Western blot analysis. (Right panel) Graphical representation of densitometric analysis of PM PAR2 normalised to β1 integrin. C, Graphical representation of fold change in cell surface PAR2 from N19 antibody flow cytometry analysis of DU145 cells transiently transfected with vector, PAR2-GFP or PAR2-C361A-GFP. Fold change was calculated from MFI values of PAR2 and PAR2-C361A cells relative to vector transfected cells. Data were from 3 independent experiments and are shown as mean ± SD. **, P < 0.05; ***, P < 0.001.

92 To examine if palmitoylation of PAR2 C361 is directly required for efficient cell surface expression of PAR2 in an endogenous expressing cell, DU145 cells were transiently transfected with expression constructs encoding wild type PAR2 or palmitoylation-deficient PAR2-C361A and surface expression was followed by flow cytometry using anti-PAR2 antibody N19. As shown in Figure 4.6C, after subtracting the level of endogenous cell surface PAR2 (identified from DU145- vector cells), DU145 cells transiently expressing PAR2-C361 exhibited 63.9 ± 1.4% less cell surface PAR2 compared with cells transiently expressing wild type receptor. Together, these data indicate that palmitoylation of PAR2 at C361 is required for efficient plasma membrane localisation of this receptor.

4.3.6 PAR2 agonism stimulates palmitate incorporation which occurs during secretory trafficking in pre-medial Golgi vesicles To understand mechanisms resulting in the reduced signalling efficiency and cell surface expression of PAR2-C361A-myc, the effect of agonist stimulation (PAR2 AP

SLIGRL-NH2) on palmitoylation of wildtype and C361A PAR2 was examined. The intracellular location where palmitoylation occurs was also investigated. As described above, metabolic labelling followed by click chemistry mediated biotin labelling of proteins was used to examine the effect of agonism on PAR2 palmitoylation. Membrane fractions were collected from cells metabolically labelled with 17-ODYA in the presence or absence of AP for 40 and 120 min. As shown in Figure 4.7A, anti-myc Western blot analysis indicated that incorporation of palmitate in untreated cells increased moderately between 40 and 120 min whereas PAR2 AP caused a significantly larger increase in palmitoylation over this period. These data suggest that PAR2 palmitoylation is induced by receptor agonism.

To examine the subcellular site at which PAR2 palmitoylation occurs, CHO-PAR2- myc cells were labelled with 17-ODYA for 2 h in the presence or absence of PAR2 AP and three different drugs known to reduce protein trafficking through the endoplasmic reticulum (ER) and Golgi organelles: brefeldin A treatment collapses the Golgi apparatus leading to protein accumulation in the ER (Dinter and Berger, 1998); nocodazole inhibits microtubule formation preventing ER to Golgi transport, leading to protein accumulation in late ER/early Golgi vesicles (Storrie and Yang, 1998); and monensin impedes protein trafficking leading to accumulation in the

Figure 4.7 PAR2 agonism stimulates palmitate incorporation which occurs during secretory trafficking in pre-medial Golgi vesicles A, CHO-PAR2-myc cells preincubated with cerulenin for 1 h were labelled with 17- ODYA for 1.5 h in the presence (+) or absence (-) of PAR2-AP (100 µM) for the indicated times. Membrane preparations were collected and biotin-azide reacted to 17-ODYA labelled proteins via click chemistry. Biotinylated proteins were isolated using streptavidin beads. PAR2 palmitoylation was assessed by anti-myc Western blot analysis of bead elutes. B, CHO-PAR2-myc cells preincubated with cerulenin were labelled with 17-ODYA for 2 h in the presence or absence of PAR2 AP (100 µM). Nocodazole (20 µg/ml; NOC) was added to medium 15 min prior to labelling with 17-ODYA and monensin (10 µM; MON) and brefeldin A (5 µg/ml; BFA) were added to medium 30 min prior to labelling with 17-ODYA. Labelled proteins were reacted to biotin-azide via click chemistry and biotinylated proteins isolated using streptavidin beads. Palmitoylated PAR2 was detected by anti-Myc Western blot analysis of bead elutes. Data shown in B is from cropped lanes from a single Western blot analysis. The data are representative of three independent experiments.

94 medial-Golgi apparatus (Dinter and Berger, 1998). Biotin-azide was reacted with equal amounts of 17-ODYA labelled membrane proteins via click chemistry and biotinylated proteins were isolated using streptavidin beads and PAR2 palmitoylation examined by anti-myc Western blot analysis from bead elutes. As shown in Figure 4.7B (vehicle), agonism induced robust PAR2 palmitoylation in vehicle treated cells and this was blocked in cells treated with brefeldin A (BFA) and nocodazole (NOC) indicating that palmitoylation of this receptor occurs after the ER. In contrast, PAR2 palmitoylation was still apparent in monensin treated cells (MON) indicating that this receptor is palmitoylated in pre-medial Golgi vesicles. These data suggest that PAR2 is palmitoylated in response to agonism during transport through sub-compartments of the Golgi apparatus prior to receptor transport to the cell surface.

4.3.7 Palmitoylation of PAR2 is required for efficient rab11a-mediated receptor repopulation of the cell surface in response to agonist stimulation Rab11a, a guanosine triphosphatase (GTPase), regulates exocytic membrane traffic and fusion (Chavrier et al., 1990; Wilcke et al., 2000) and enhances PAR2 trafficking to the cell surface (Roosterman et al., 2003). To examine whether palmitoylation is required for agonist stimulated rab11a-mediated PAR2 transport to the plasma membrane, flow cytometry analysis of CHO-PAR2-myc and PAR2- C361A-myc cells was performed using the anti-PAR2 antibody N19. These cells were transiently transfected with vector, rab11a-GFP or dominant negative mutant rab11a-S25N-GFP and treated in suspension with trypsin (50 nM) for 15 min before removal of trypsin and monitoring of cell surface expressed PAR2 after 15 and 30 min. As shown in Figure 4.8, levels of cell surface PAR2-myc in vector transfected cells steadily increased in response to trypsin agonism, indicating that the cell surface was being replenished with nascent receptor. Consistent with previous reports (Roosterman et al., 2003), wild type rab11a significantly increased cell surface localization of PAR2 and this effect was slowed by dominant negative rab11a-S25N. Consistent with data in Figures 4.4, 4.5 and 4.6, PAR2-C361A-myc trafficked to the cell surface but at a much lower rate that wildtype receptor (Figure 4.7, PAR2- C361A + Vector). In addition, replenishment of the cell surface with this mutant receptor was almost unaffected by rab11a-S25N. However, rab11a also induced a statistically significant increase (P < 0.05) in cell surface PAR2-C361A-myc in

Figure 4.8 Palmitoylation of PAR2 is required for efficient rab11a-mediated repopulation of the cell surface in response to agonist stimulation CHO-PAR2-myc and CHO-PAR2-C361A-myc cells transiently transfected with vector, rab11a-GFP or dominant negative rab11a-S25N-GFP were treated with 50 nM trypsin for 15 min to remove cell surface PAR2. Cells were then washed and incubated with trypsin-free DMEM for the indicated times to allow repopulation of PAR2 at the cell surface. Plasma membrane PAR2 was measured by flow cytometry using the anti-PAR2 N19 antibody. MFI values were used to calculate percentage of cell surface PAR2 relative to vector cells. Experiments were performed in triplicate on 3 independent occasions. Data are displayed as mean ± SD. Statistical significant differences - 30 min: PAR2 + vector 36.8 ± 1.6%, PAR2 + rab11a-GFP 65.5%, P < 0.0001; PAR2-C361A + vector 22.7 ± 3.5%, PAR2-C361A + rab11a-GFP 33.4 ± 4.0%, P < 0.05; PAR2-C361A + rab11a-GFP 33.4 ± 4.0%, PAR2 + rab11a-GFP 65.5%, P < 0.0001, compared with PAR2-C361A + rab11a-GFP.

96 response to receptor agonism (Figure 4.7). These data indicate that rab11a-mediated trafficking of PAR2 to the cell surface in response to agonism is largely palmitoylation dependent and that there is a smaller, but appreciable, palmitoylation independent component.

4.3.8 Mutation of PAR2 C361 alters agonist-induced recruitment of β-arrestin-1 and β-arrestin-2 β-arrestin-1 and -2 are rapidly recruited to the carboxyl terminus of PAR2 following receptor activation, leading to rapid internalisation (DeFea et al., 2000; Dery et al., 1999). To assess the impact of PAR2 palmitoylation on receptor desensitisation, PAR2 agonist-induced β-arrestin-1 and -2 recruitment was examined by confocal microscopy. CHO cells were transiently co-transfected with β-arrestin-1-GFP or β- arrestin-2-GFP with wildtype PAR2-mCherry or PAR2-C361A-mCherry. As shown in Figure 4.9A, β-arrestin-1-GFP and β-arrestin-2-GFP localise throughout the cytoplasm of untreated cells co-transfected with either wild type or palmitoylation- deficient PAR2-mCherry. Stimulation of wild type PAR2 with AP for 5 min induced rapid translocation of β-arrestin-1-GFP and β-arrestin-2-GFP to the plasma membrane and colocalization with PAR2 that was sustained for at least 15 min (Figure 4.9A, yellow in merged images). In contrast, β-arrestin-1-GFP remained localised to the cytoplasm of cells expressing PAR2-C361A-mCherry under agonist- induced conditions at both time points (Figure 4.9B, left). However, agonist stimulation of PAR2-C361A-myc induced a delayed relocation of β-arrestin-2-GFP to the plasma membrane (Figure 4.9B, right, yellow in merged image at 15 min). These data suggest that PAR2 palmitoylation is essential for agonist induced relocation of β-arrestin-1 to the cell surface but not for relocation of β-arrestin-2.

4.3.9 PAR2 C361 is required for efficient agonist-induced receptor endocytosis and degradation The data in Figures 4.4, 4.5 and 4.6 indicate that compared with wildtype receptor, PAR2-C361A displays reduced trafficking to the cell surface, lower levels of plasma membrane localization and diminished β-arrestin recruitment following receptor agonism. As it has previously been reported that after β-arrestin recruitment, agonism of PAR2 results in rapid receptor internalisation and sorting for lysosomal degradation (DeFea et al., 2000; Dery et al., 1999), the effect of loss of

Figure 4.9 Mutation of PAR2 C361 alters agonist-induced recruitment of β- arrestin-1 delays β-arrestin-2 Confocal microscopy analysis of untreated live cells and at 5 and 15 min after agonism with PAR2 AP (100 µM). A, CHO cells transiently co-expressing PAR2- mCherry (red) and β-arrestin-1-GFP (green) or β-arrestin-2-GFP (green). B, CHO cells transiently co-expressing PAR2-C361A-mCherry (red) and β-arrestin-1-GFP (green) or β-arrestin-2-GFP (green). Merged images in A and B highlight colocalization (yellow) of PAR2 and β-arrestins. Cells were analysed using a Zeiss LSM510 confocal microscope (63x oil immersion objective lens) and images were processed using MetaMorph software and displayed using CorelDraw X5. Images of cells are representative of three independent experiments. Scale bar, 10 μM. Confocal microscopy was performed together with Dr Melinda Christensen.

98 palmitoylation on the rate of endocytosis and degradation of this receptor was examined. First, to assess endocytosis, the rate of agonist-induced internalisation of wildtype and C361A PAR2 was investigated. CHO cells stably expressing wildtype PAR2-myc or PAR2-C361A-myc labelled with biotin were treated with PAR2 AP (100 µM) for 5, 15 and 30 min to allow endocytosis. After these periods of internalisation, residual cell surface biotin was removed by washing cells with the reducing agent MeSNA. Endocytosed proteins labelled with biotin were isolated from whole cell lysates using streptavidin beads and the level of PAR2 endocytosis examined from bead elutes by anti-myc Western blot analysis. As shown in Figure 4.10A, PAR2-myc endocytosis steadily increased over 30 min in response to PAR2 AP with the internalised receptor detected following 5 min and maximal PAR2 endocytosis detected at 15 min. In contrast, under the same conditions, internalisation of PAR2-C361A-myc was greatly reduced (Figure 4.10A). Graphical analysis of densitometric analysis of three independent Western blot analyses normalised to total cell surface PAR2 indicated that mutation of C361 reduced the level of PAR2 internalisation by approximately 50% (Figure 4.10A). These data indicate that palmitoylation is required for efficient agonist-induced PAR2 endocytosis.

To examine receptor degradation, CHO-PAR2-myc and CHO-PAR2-C361A-myc cells, treated with cycloheximide to inhibit de novo protein synthesis, were incubated with PAR2 AP. As shown in Figure 4.10B, anti-myc Western blot analysis indicated that PAR2-myc levels dropped markedly over a 4 h period following AP treatment. In contrast, under the same conditions the reduction in PAR2-C361A-myc levels was much less pronounced (Figure 4.9B). Graphical analysis of densitometric analysis of three independent Western blot analyses indicated that 4 hours after AP treatment levels of PAR2-myc dropped by 90 ± 4.5% whereas PAR2-C361A-myc reduced by 22.6 ± 2.4% (P < 0.0001, compared with PAR2-myc; Figure 4.9B). These data indicate that loss of palmitoylation reduces the rate of degradation of PAR2 in response to agonist stimulation.

99 Figure 4.10 PAR2 C361 is required for efficient agonist-induced receptor endocytosis and degradation A, CHO-PAR2-myc and CHO-PAR2-C361A-myc cells labelled with membrane impermeant NHS-EZ-link biotin were treated with PAR2 AP (100 µM) for the indicated times to induce receptor internalisation. Protein degradation was blocked by incubation of cells with the proteasome inhibitor MG132. Residual cell surface biotin was removed by washing with MeSNA and internalised biotin-labelled protein was isolated from whole cell lysates using streptavidin beads. Anti-myc Western blot analysis was performed on bead eluates and input lysates. Graphical representation of densitometry analysis of these data shown in the right panel. Percentage of internalised receptor was calculated by dividing the value for internalised PAR2 at each time point by the value for total surface PAR2 which was determined from an experiment performed in parallel in which cells were not treated with AP or MeSNA. B, CHO-PAR2-myc and CHO-PAR2-C361A-myc cells were treated with PAR2 AP (100 µM) for the indicated times in the presence of 70 µM cycloheximide, to prevent de novo protein synthesis. Lysates were examined by anti-myc and anti-GAPDH Western blot analysis to assess the levels of PAR2 remaining after receptor agonism. with graphical representation of densitometry analysis of these data shown in the right panel. PAR2 endocytosis and degradation were significant compared with PAR2-C361A; P < 0.001. Data in A and B are representative of 3 independent experiments with values displayed as Mean ± SD.

100 4.4 Discussion In this chapter, the role of palmitoylation in trafficking to and from the cell surface of the protease-activated GPCR PAR2 as well as the effect of this post-translational modification on receptor signalling was investigated. The data extend findings from a recent report by Botham and colleagues (Botham et al., 2011) demonstrating that palmitoylation of PAR2 can occur at C361. However, in several respects the data in this chapter are also inconsistent with a number of the key results of this group. Figure 4.1 shows for the first time that inhibition of palmitoylation reduces the cell surface localization of PAR2 by at least 50% in endogenous expressing cell lines (PC3, DU145 and 22Rv1). In addition, Figure 4.6 demonstrates that PAR2-C361A transiently expressed by one of these endogenous expressing cells lines (DU145) is inefficiently trafficked to the cell surface compared with wildtype PAR2. These data suggest that palmitoylation is required for efficient plasma membrane expression of endogenously expressed PAR2. This conclusion is the converse of the findings of Botham et al. who showed by flow cytometry that palmitoylation-deficient PAR2, stably expressed by CHO-Pro5 cells, is present on the plasma membrane at between ~150% (70% confluence) and ~175% (20% confluence) of the levels of wildtype PAR2 (Botham et al., 2011). However, data presented herein are supported by flow cytometry and cell surface biotinylation studies of wildtype and C361A PAR2 stably expressed by CHO cells as well as by confocal microscopy analysis of transiently transfected cells. In addition to demonstrating that PAR2 is exclusively palmitoylated at C361, these approaches showed, consistent with observations from endogenous expressing cells, that mutation of C361 results in at least a 60% reduction in the cell surface expression of PAR2.

As depicted in Figure 4.11, the findings from this chapter indicate that palmitoylation of PAR2 is required for optimal receptor signalling, cellular trafficking and plasma membrane expression. It is likely that mono-palmitoylation of PAR2 at C361 occurs in the Golgi apparatus prior to transit to the medial-Golgi and this is enhanced by receptor agonism at the cell surface. Efficient cell surface repopulation of PAR2 by the GTPase rab11a is also dependent on receptor palmitoylation and in its absence, nascent PAR2 accumulates in pre-Golgi vesicles or the early Golgi, preventing efficient trafficking toward the plasma membrane. Calcium flux data indicate that palmitoylation is also required for efficient PAR2 signalling via Gαq. However, the

101

Figure 4.11 PAR2 palmitoylation is required for optimal receptor signalling, cellular trafficking and plasma membrane expression Unpalmitoylated PAR2 located in Golgi vesicles (1). Palmitoylation occurs within the cis- to medial-Golgi (2). This anchors the carboxyl domain of PAR2 to the inner leaflet of cell membranes likely resulting in the formation of an intracellular eighth α-helix (H8) required for effective signalling (2). Interactions of palmitoylated PAR2 with the GTPase rab11a promotes agonist induced receptor repopulation of the cell surface (3). Activation of PAR2 (4) enhances palmitoylation of Golgi localised PAR2 (5) and, reciprocally, receptor palmitoylation is required for efficient PAR2 signal transduction (6). Receptor agonism induces β-arrestin translocation to the plasma membrane (7) which mediates non-G protein signal transduction and receptor internalisation (8). Internalised PAR2 is trafficked and sorted for lysosomal degradation via early and late endosomes (9).

102 observed reduction in Gαq signalling via PAR2-C361A compared with wildtype receptor is likely largely due to the reduced expression of PAR2 on the cell surface. It has previously been shown that receptor activation induces translocation of β- arrestin from the cytosol to the plasma membrane to facilitate non-G protein mediated signal transduction and PAR2 internalisation (DeFea et al., 2000; Dery et al., 1999). However, agonism of cells expressing palmitoylation-deficient PAR2 failed to induce β-arrestin-1 translocation while β-arrestin-2 translocation was delayed. As the confocal, flow cytometry and cell surface biotinylation assays indicated that palmitoylation-deficient PAR2 is capable of trafficking to the plasma membrane, albeit with reduced efficiency compared with wildtype receptor, the observed effects on β-arrestins suggest that palmitoylation is required for efficient internalisation of PAR2. This was supported by assays showing that endocytosis of this receptor is markedly reduced when C361 is mutated. Consistently, degradation of PAR2-C361A following receptor agonism occurs at much lower levels than wildtype PAR2, likely because lower levels of the mutated receptor are trafficked to the cell surface and available for activation. Thus, these data suggest that palmitoylation of PAR2, induced by cell surface agonism, is required for efficient trafficking from Golgi stores to the plasma membrane. At this location, palmitoylation is also required for efficient recruitment of proteins that are needed for signal transduction, receptor desensitisation and internalisation and sorting for lysosomal degradation. Accordingly, these data suggest that palmitoylation is an essential post-translational modification in the life cycle of PAR2.

The finding that palmitoylation of PAR2 occurs following ER export and during Golgi transit is consistent with the observation that 12 of the 23 known human PATs are located within the Golgi apparatus (Fernandez-Hernando et al., 2006; Ohno et al., 2006). In addition, several other proteins are palmitoylated at this site including endothelial nitric oxide synthase (Fernandez-Hernando et al., 2006), tetraspanins CD9 and CD151 (Sharma et al., 2008; Yang et al., 2002; Yang et al., 2004). Furthermore, several GPCRs are palmitoylated during transport from ER-Golgi intermediate compartments through to the medial Golgi including CCR5 (Blanpain et al., 2001; Percherancier et al., 2001), δ-opioid receptor (Petaja-Repo et al., 2006), thyrotropin receptor (Tanaka et al., 1998) and vasopressin V1a receptor (Hawtin et al., 2001). Further work is required to identify the PAT(s) that palmitoylate PAR2

103 within the Golgi apparatus. In addition, based on the diverse localization of PATs within cells and the dynamic nature of S-palmitoylation, it is possible that PAR2 will be palmitoylated at other cellular locations as occurs for the δ-opioid receptor

(Petaja-Repo et al., 2006) and the β1-adrenergic receptor (Zuckerman et al., 2011).

The palmitoylation of Golgi localised PAR2 in response to receptor agonism and prior to cell surface expression, suggests that this modification facilitates plasma membrane targeting. In support of this, the palmitoylation inhibitor 2- bromopalmitate reduced cell surface expression of both endogenous and overexpressed PAR2. Moreover, PAR2-C361A, although capable of trafficking to the cell surface, displayed greatly reduced plasma membrane localization. Consistent with these observations, other GPCRs and membrane receptors also display reduced cell surface expression following chemical inhibition of palmitoylation or mutation of carboxyl terminal palmitoylation sites. These include the chemokine CCR5 receptor (Blanpain et al., 2001; Percherancier et al., 2001), dopamine D1 receptor (Ng et al., 1994), follicle-stimulating hormone receptor (Ulloa-Aguirre et al., 2007), histamine H2 (Fukushima et al., 2001b), lutropin/choriogonadotropin receptor (Zhu et al., 1995), δ-opioid receptor (Petaja-Repo et al., 2006), V2R (Sadeghi et al., 1997; Schulein et al., 1996). In addition, there are examples of GPCRs (V1a vasopressin receptor, β1R and β2R) that show unaffected cell surface expression when palmitoylation is blocked (Hawtin et al., 2001; Moffett et al., 1993; Zuckerman et al., 2011). Accordingly, the proposal by Botham and colleagues that mutation of the PAR2 palmitoylation site results in elevated cell surface expression (Botham et al., 2011) should be viewed with caution. It is not clear why the data of these workers differ from the findings in this chapter, although it is possible that use of different CHO cell sub-lines may be a contributing factor. Whereas in this chapter CHO cells are used, Botham and colleagues employed the CHO-Pro5 line which is the parental cell line for several glycosylation defective sub-lines and an auxotroph requiring proline supplementation (Stanley et al., 1975).

Several years ago the crystal structure of the class A GPCR bovine rhodopsin revealed that a partially exposed α-helix, designated α-helix 8 (α-H8), is located between the final TM domain and the palmitoylated cysteines in the carboxyl terminus of this receptor (Palczewski et al., 2000). Since then this domain has been

104 identified in other GPCRs containing consensus carboxyl terminal palmitoylation sites including cannabinoid receptor 2, β2R, adrenoreceptor (Katragadda et al., 2004), cannabinoid receptor 1 (Ahn et al., 2010) and α1-adrenergic receptor (Delos Santos et al., 2006). It is clear from this chapter and the report of Botham et al. (Botham et al., 2011) that C361 of PAR2 is palmitoylated and this modification is at the carboxyl boundary of a putative PAR2 α-H8. These domains have been shown to be required for direct coupling of several GPCRs with G protein subunits (Ernst et al., 2000; Katragadda et al., 2004; Marin et al., 2000; Phillips and Cerione, 1992; Swift et al., 2006), and it is possible the α-H8 region functions by switches between inactive (helical) and active (non-helical) in response to ligand binding. Very recently the α-H8 of PAR1 was shown to be critical for Gq-dependent receptor signalling (Swift et al., 2006) and the direct relevance of this domain to PAR2 signalling is suggested by the observation that its deletion completely prevents Gαq signal transduction (Sevigny et al., 2011). Accordingly, it is possible that like other GPCRs, PAR2 palmitoylation contributes to α-H8 stability thereby enabling efficient G protein coupling. In addition, it is possible that PAR2 palmitoylation-induced stabilization of its α-H8 is required for rab11a-mediated receptor repopulation of the cell surface in response to agonist stimulation. This is based on the observation that palmitoylation stabilises the α-H8 of the prostacyclin receptor and this is required for its interactions with rab11a during receptor internalization and recycling (Reid et al., 2010). Consistently, whereas agonist-induced cell surface localization of wildtype PAR2 was significantly increased by rab11a, replenishment of the cell surface with mutant receptor was almost unaffected by this GTPase. Thus, although this possibility was not addressed experimentally, the PAR2 α-H8 may mediate any direct interactions between receptor and rab11a allowing efficient transport to the plasma membrane.

The results presented in this chapter indicate that mutation of the PAR2 palmitoylation site also negatively impacts on agonist-induced β-arrestin colocalization with this receptor at the plasma membrane. Similar effects on β- arrestin recruitment have been reported for the GPCRs V2R (Charest and Bouvier, 2003) and TRH receptor (Groarke et al., 2001). Interestingly, palmitoylation has been shown to alter phosphorylation of the GPCRs 5-hydroxytryptamine4a receptor (Ponimaskin et al., 2005), β2R (Moffett et al., 1993), CCR5 (Kraft et al., 2001),

105 TRH receptor (Gehret et al., 2010) and vasopressin V1a receptor (Hawtin et al., 2001) and of the AMPA membrane receptor GluR1 (Zhang et al., 2009a). Accordingly, as it is known that phosphorylation of the PAR2 carboxyl tail is a key regulator of β-arrestin recruitment to this receptor (Ricks and Trejo, 2009), it is possible that palmitoylation is a necessary prerequisite for the phosphorylation that is essential for β-arrestin recruitment to PAR2.

In summary, these data provide new insights on the role of palmitoylation in the life cycle of PAR2. The data indicate that this post-translational modification is required for maximal cell surface expression of endogenous and stably expressed PAR2 and is needed for both efficient signalling and agonist-induced rab11a-mediated PAR2 trafficking to the plasma membrane. At the plasma membrane agonist-induced β- arrestin recruitment is compromised by mutation of the PAR2 palmitoylation site as is receptor degradation in response to agonism. It was also noted that loss of palmitoylation does not completely block these PAR2 cellular processing events. Accordingly, further studies are required to examine other mechanisms regulating this receptor.

106

Chapter 5:

Mutagenesis studies to identify PAR2 residues involved in ligand activated signalling

107

108 5.1 Introduction In the previous chapter, the role of palmitoylation in modulating PAR2 signal transduction and trafficking was investigated. This modification was found to reduce the potency (EC50: effective concentration of agonist required to stimulate half of the maximal response) and efficacy (maximal response to agonist) of trypsin stimulated calcium mobilisation (Figure 4.4). In this chapter, PAR2-mediated signal transduction is investigated further with the evaluation of amino acids that are required for agonist-activated calcium mobilisation.

GPCRs are seven TM containing receptors and comprise the largest membrane protein family in the genome (Flower, 1999; Fredriksson et al., 2003). In response to a diverse array of ligands, GPCRs activate signal transduction contributing to normal physiology and disease (Oldham and Hamm, 2008). The initial event in GPCR activation is ligand binding which induces a conformational change within the receptor, permitting post-activation events such as phosphorylation, changes in palmitoylation status, second messenger generation and downstream signalling cascades (Oldham and Hamm, 2008; Simon et al., 1991). Because of the roles of aberrantly activated GPCRs in disease, small molecules that target these receptors to activate (agonists) or block (antagonists) signal transduction have therapeutic potential. In fact, 20-25% of all known therapeutic drugs target GPCRs (Landry and Gies, 2008; Overington et al., 2006). However, defining the precise ligand binding location is crucial for the design of efficacious and potent GPCR targeting drugs (Dorsam and Gutkind, 2007).

Crystal structures of GPCRs have led to identification of residues critical to ligand docking and also defined the ligand binding pocket for several GPCRs (Kobilka and Schertler, 2008). Currently, crystal structures exist for the turkey β1 adrenergic receptor (Warne et al., 2011; Warne et al., 2008), human β2 adrenergic receptor (Bokoch et al., 2010; Cherezov et al., 2007; Rasmussen et al., 2011a; Rasmussen et al., 2007; Rosenbaum et al., 2011), human adenosine A2 receptor (Jaakola et al., 2008; Lebon et al., 2011; Lebon and Tate, 2011; Xu et al., 2011), bovine (Palczewski et al., 2000; Teller et al., 2001) and squid (Murakami and Kouyama, 2011) rhodopsin, CXCR4 chemokine receptor (Wu et al., 2010) and histamine H1 receptor (Shimamura et al., 2011). In particular, crystal structures for the bovine

109 rhodopsin, β2 adrenergic and human adenosine A2 receptors bound to soluble agonists or antagonists have demonstrated that the TM regions of GPCRs are important for ligand docking (Topiol and Sabio, 2009). Based on these crystal structures, the seven GPCR TM α-helices arrange into a conical or cylindrical structure forming a central hydrophobic pocket (Blakeney et al., 2007; Rosenbaum et al., 2009). Receptor agonists and antagonists dock within this pocket binding to amino acids that protrude into the pocket milieu (Cherezov et al., 2007; Jaakola et al., 2008; Palczewski et al., 2000; Teller et al., 2001; Xu et al., 2011).

However, the generation of GPCR crystal structures is challenging due to receptor instability in detergent and high levels of hydrophobicity (Cherezov et al., 2001; Vaidehi et al., 2002). Other strategies to define receptor structural information include homology modelling and in silico ligand docking (Congreve et al., 2011; Radestock et al., 2008). Inferring regions of importance for ligand docking from homology modelling relies on using crystal structures as a template to derive a 3D structure (Moro et al., 2006; Patny et al., 2006). For GPCRs, the crystal structure for bovine rhodopsin is most commonly used in homology modelling and has provided valuable information for the adenosine A1 receptor (Gutierrez-de-Teran et al., 2004a; Gutierrez-de-Teran et al., 2004b) and dopamine D2-like receptor (Boeckler et al., 2005; Broer et al., 2003; Xie et al., 2003). Used in conjunction with homology modelling, in silico ligand docking can predict amino acids required for ligand binding by modelling the structure of an agonist or antagonist within the confines of the GPCR homology model (Bender et al., 2004). Combination of these in silico methods has been used successfully to identify lead compounds targeting the urotensin II receptor (Flohr et al., 2002) and metabotropic glutamate receptor 5 (Renner et al., 2005). However, experimental validation of the regions and amino acids identified in silico is required (Kristiansen, 2004). In this regard, combined in silico and site-directed mutagenesis studies have identified amino acids involved in ligand docking within the TM domains of several GPCRs including the adenosine A3 receptor (Jacobson et al., 2001; Moro et al., 2003), P2Y nucleotide receptor (Costanzi et al., 2004; Ivanov et al., 2005) and thyrotropin-releasing hormone receptor type 1 (Huang et al., 2005; Lu et al., 2004).

110 In this chapter, in silico experimental approaches were employed to characterise residues potentially involved in binding of the PAR2 TL induced by trypsin cleavage as well as the soluble agonists 2f-LIGRLO-NH2 and GB110. The last two molecules are PAR2 agonists with comparable potencies that are based on the hexapeptide

PAR2 AP, SLIGRL-NH2. Amino and carboxyl terminal changes to SLIGRL-NH2, yielding 2f-LIGRLO-NH2 and GB110, have improved potency over SLIGRL-NH2 while maintaining PAR2 selectivity over PAR1 (Barry et al., 2010; Barry et al., 2007; Kawabata et al., 2004; McGuire et al., 2004). As described in Chapter 1, although it is well established that PAR2 is activated by a proteolytic mechanism (Adams et al., 2011), the residues required for intramolecular docking of the newly exposed TL remains largely unknown. Previous reports have suggested a role for the

ECL2 of PAR2 in the docking of the TL as well as the PAR2 AP SLIGRL-NH2 (Al- Ani et al., 1999; Al-Ani et al., 2002b; Lerner et al., 1996). For example, substitution of the PAR1 ECL with the ECL for PAR2 renders the chimeric PAR1 receptor responsive to PAR2 APs (Lerner et al., 1996). Also, mutation of charged residues within the PAR2 ECL2 reduces the ability of PAR2 APs to efficiently stimulate calcium mobilisation (Al-Ani et al., 1999; Al-Ani et al., 2002b). As TM domains are critical for ligand binding for other GPCRs (Rosenbaum et al., 2009), it is likely that ligand binding to PAR2 will also involve residues within its membrane spanning regions.

This chapter describes use of a PAR2 homology model, based on the crystal structure of bovine rhodopsin, and in silico docking of the PAR2 agonist 2f-LIGRLO-NH2 to identify amino acids involved in ligand docking within the TM domains. Site- directed mutagenesis of these residues, to either Ala or Leu, was used to examine the roles of these amino acids in agonist docking.

111 5.2 Methods 5.2.1 Expression constructs and mutagenesis The human PAR2 open reading frame incorporating 3' sequence encoding a carboxyl terminal FLAG epitope (DYKDDDDK) was amplified by PCR using Expand High Fidelity polymerase mixture (Roche) from a previously described construct and cloned into the pcDNA5/Flippase (flp) recombination target (FRT) vector (Invitrogen) at the BamHI site. Site-directed mutagenesis was performed using Pfu Ultra polymerase (Agilent Technologies) and primers listed in Table 2.1 to mutate selected residues within the PAR2 coding sequence to an alanine or leucine. The sequence of all constructs was confirmed by DNA sequencing at the Australian Genome Research Facility (St. Lucia, Australia).

5.2.2 Generation of CHO-flpin cells stably transfected with vector, wildtype PAR2 and mutant PAR2 The Flpin system allows incorporation of a gene of interest into a specific location within the genome (O'Gorman et al., 1991). An FRT site has been introduced into CHO-flpin cells allowing direct integration of the pcDNA5/FRT vector which contains a corresponding FRT site (Bresson et al., 2004). CHO-flpin cells stably transfected with vector, wildtype or mutant PAR2 were generated as per section 2.2.8. To facilitate recombinant incorporation, CHO-flpin cells were co-transfected with pOG44 (Invitrogen) which expresses the Flp recombinase.

5.2.3 Dose-response, EC50 and relative EC50 (REC) value calculation The amplitude of calcium mobilisation for each PAR2 agonist was used to calculate agonist dose-response. Dose-response curves and EC50 values were generated and calculated in Graphpad Prism version 5.0 by Dr Jacky Suen as described previously

(Suen et al., 2011). REC values were calculated as a ratio of PAR2 mutant EC50 relative to wild type PAR2 EC50 for each receptor agonist.

112 5.3 Results 5.3.1 Computer homology modelling of PAR2 identifies residues potentially involved in ligand docking and receptor activation To investigate PAR2 ligand docking, a computer homology model of PAR2 was first generated. PAR2 was modelled on the inactive state of bovine rhodopsin, a prototypical and extensively studied GPCR (Filipek et al., 2003a; Filipek et al., 2003b; Piscitelli et al., 2006; Salom et al., 2006; Schertler, 2005; Szundi et al., 2006). At the time of commencing this study, bovine rhodopsin was the only GPCR crystal structure available. PAR2 homology modelling was performed using the Homology module within the InsightII molecular modelling suite by members of Prof. David Fairlie’s laboratory (Higginbottom et al., 2005). As shown in Figure 5.1, the seven helical TM domains in the 3D monomer model of PAR2 are arranged conically. The amino-terminus and three extracellular loops form the extracellular surface of PAR2, and three intracellular loops and the carboxyl-terminus form the intracellular surface of PAR2 (Figure 5.1). An eighth helix found perpendicular to the TM bundle is predicted to form part of the PAR2 intracellular surface (see Chapter 4; Figure 5.1).

To identify residues involved in PAR2 ligand docking, computer aided binding of the

PAR2 agonists SLIGRL-NH2 and 2f-LIRGLO-NH2 was performed using the flexible ligand docking program Gold version 3.1 and modelled using InsightII (Higginbottom et al., 2005). Agonist modelling and computer docking was also performed by members of Prof. David Fairlie’s laboratory. Table 5.1 lists 12 PAR2 residues identified by these approaches as being potentially involved in interacting with receptor agonists. These residues are: Phe 128 (F128), Leu 151 (L151), Phe 155 (F155), Tyr 156 (Y156), Tyr 160 (Y160), Val 250 (V250), Arg 304 (N304), Leu 307 (L307), Val 308 (V308), Tyr 311 (Y311), Tyr 326 (Y326) and Leu 330 (L330). As shown in Figure 5.2, these residues are located within TM domains, except for Y311 which is located within ECL3. (Table 5.1).

Having identified 12 residues potentially required for ligand docking, these amino acids were mutated to alanine, leucine or valine residues; each a non-polar alkyl containing amino acid (Table 5.1). Two mutants were generated for Y156 and Y326;

113

Figure 5.1 Computer homology model for PAR2 based on the crystal structure for bovine rhodopsin The human PAR2 protein sequence was aligned with the protein sequence for bovine rhodopsin and modelled on the bovine receptor crystal structure. The PAR2 homology model highlights the seven TM domains (green helix labelled I-VII) in a cylindrical bundle and an eighth intracellular helix (α-H8) found perpendicular to the TM bundle. The PAR2 homology model was generated by members of Prof. David Fairlie’s laboratory (Institute for Molecular Bioscience, University of Queensland).

114

Table 5.1 PAR2 residues identified by computer aided homology modelling and ligand docking as potentially involved in ligand binding The location of each residue within a TM domain or ECL is indicated as is the corresponding mutations. Residues mutated twice are boxed in red.

PAR2 TM helix Mutation mutations F128 II F → A L151 III L → A F155 III F → A Y156 III Y → A Y156 III Y → L Y160 III Y → L V250 V V → A N304 VI N → L L307 VI L → A V308 VI V → A Y311 ECL3 Y → L Y326 VII Y → A Y326 VII Y → L L330 VII L → V

115

Figure 5.2 Diagram of PAR2 highlighting residues identified by homology modelling to potentially interact with PAR2 agonists Residues identified by receptor modelling to potentially interact with PAR2 agonists are shown in red. Transmembrane (TM), extracellular (ECL) and intracellular (ICL) loop domains are indicated. The disulphide bridge between conserved cysteine residues in TM III and ECL2 is shown by the dotted blue line.

116 each was mutated to an Ala and Leu. PAR2-FLAG mutant constructs were generated by site directed mutagenesis using the primers listed in Table 2.1 and these plus wildtype PAR2 and vector constructs were stably transfected into CHO-flpin cells. These cells incorporate vectors containing the FRT sequence into a defined site within the genome, permitting the selection of populations of cells stably expressing similar levels of PAR2 (Bressen et al., 2004). CHO-flpin cells transfected with empty vector (pcDNA5), wildtype PAR2-FLAG or the 14 PAR2-FLAG mutants were selected by treating with hygromycin for 14 days. PAR2 expression levels were examined by anti-FLAG Western blot analysis of whole cell lysates from stably expressing CHO-flpin cells. Consistent with Western blot analysis from Chapter 3, wildtype PAR2 and each receptor mutant, except N304L, was detected as a smear ranging from ~30 kDa to ~250 kDa (Figure 5.3). These Western blot analyses also show that all PAR2-FLAG mutants, except for N304L, are expressed at similar levels to wildtype PAR-FLAG (Figure 5.3). Anti-GAPDH Western blot analysis confirmed equal loading. Due to time constraints, stable generation and further evaluation of PAR2-FLAG mutant N304L was discontinued.

5.3.2 Characterisation of CHO-flpin cells stably expressing wildtype or mutant PAR2 The level of cell surface expression for stably expressed wildtype PAR2 and receptor mutants was assessed by flow cytometry using the anti-PAR2 antibody N19 ((Suen et al., 2011); see Chapter 3.3.8). Compared with cells stably transfected with vector, flow cytometry analysis indicated that stably expressed wildtype PAR2-FLAG (black solid line) and each PAR2-FLAG mutant (grey dotted line) localised to the plasma membrane of CHO-flpin cells (Figure 5.4). With the exception of PAR2 mutants F128A, F155A and L307, MFI values for wildtype and each mutant PAR2-FLAG indicated similar levels of cell surface expression compared to the wildtype receptor (Table 5.2). Mutants F128A, F155A and L307 displayed increased cell surface expression with higher mutant MFI values compared with the wildtype PAR2 MFI value (compare 43.6 ± 3.9, 41.6 ± 5.8 and 47.3 ± 5.9 respectively with 34.3 ± 1.7).

Collectively, these Western blot and flow cytometry analyses indicate that the generated cell lines express similar levels of PAR2 with similar levels of receptor on the cell surface.

117

Figure 5.3 Protein expression of wildtype PAR2-FLAG and mutant PAR2- FLAG in stably expressing CHO-flpin cells Anti-FLAG and anti-GAPDH Western blot analysis of whole cell lysates collected from untransfected CHO-flpin cells and CHO-flpin cells stably expressing empty vector (pcDNA5), wildtype PAR2-FLAG or one of 14 PAR2-FLAG mutants. Data are representative of three independent experiments.

118

119

Figure 5.4 Cell surface expression of mutant PAR2-FLAG compared with wildtype PAR2-FLAG in stably expressing CHO-flpin cells Anti-PAR2 N19 flow cytometry analysis of non-permeabilised CHO-flpin cells stably transfected with vector (pcDNA5; grey filled population), wildtype PAR2- FLAG (solid black line) or mutant PAR2-FLAG (dotted grey line). Cells were incubated with anti-PAR2 antibody N19 followed by secondary antibody before analysis. MFI values for flow cytometry analyses are displayed in Table 5.2. The data are representative of three independent experiments.

120

Table 5.2 Comparison of cell surface wildtype and mutant PAR2-FLAG stably expressed by CHO-flpin cells MFI values from anti-PAR2 antibody N19 flow cytometry analysis were determined from 3 independent experiments and are shown as mean ± S.D. Fold change was calculated from MFI values (ratio of PAR2 mutant MFI to wildtype PAR2 MFI).

PAR2 MFI values Fold change mutations relative to wildtype PAR2 PAR2 34.3 ± 1.7 1.00 F128A 43.6 ± 3.9 1.27 L151A 35.6 ± 2.7 1.04 F155A 41.6 ± 5.8 1.21 Y156A 34.1 ± 2.3 0.99 Y156L 33.7 ± 2.7 0.98 Y160L 34.9 ± 3.2 1.02 V250A 36.2 ± 4.1 1.06 L307A 47.3 ± 5.9 1.38 V308A 34.9 ± 2.1 1.02 Y311L 36.3 ± 3.1 1.07 Y326A 34.9 ± 1.9 1.02 Y326L 32.6 ± 2.8 0.95 L330V 35.4 ± 2.3 1.03

121 5.3.3 Comparison of signalling induced via wildtype and mutant PAR2 The impact of PAR2 mutations on receptor signalling induced by PAR2 agonists, relative to wildtype PAR2, was assessed by dose-response curves obtained from calcium flux assays. This assay evaluates agonist-induced PAR2 signalling by a well characterised pathway involving the coupled G-protein Gαq. Following PAR2 agonism this G protein activates phospholipase C to generate diacyl glycerol and inositol 1,4,5-triphosphate (IP3) from phosphatidylinositol 4,5-bisphosphate resulting in efflux of Ca2+ from the endoplasmic reticulum into the cytosol. Data from this assay are used to determine the effect of each mutation on PAR2 agonist potency

(EC50; power of the agonist to activate 50% of the maximal response), and from this identify residues required for agonist induced signalling. Calcium flux assays were performed together with Dr Jacky Suen at the Institute for Molecular Biosciences, University of Queensland. Dose response curves were generated using Graphpad Prism version 5.0 by Dr Jacky Suen.

5.3.3.1 Dose-response curves for vector and wildtype PAR2 Dose response curves were obtained from calcium flux assays of CHO-flpin cells stably transfected with vector or wildtype PAR2 treated with three PAR2 agonists; trypsin (revealing endogenous TL) and the APs 2f-LIGRLO-NH2 and GB110. As shown in Figure 5.5, each agonist stimulates calcium mobilisation in a concentration- dependent manner, producing a characteristic sigmoidal curve. The calculated EC50 values for trypsin, 2f-LIGRLO-NH2 and GB110 were 3.5 nM, 89.0 nM and 145.9 nM respectively. These values were more potent than the reported EC50 values for trypsin (6 nM (Suen et al., 2011)), 2f-LIGRLO-NH2 (210 nM (Suen et al., 2011)) and GB110 (240 nM (Suen et al., 2011)) in endogenously expressing HT29 colon cancer cells. The difference in the EC50 values for each agonist may result from higher PAR2 expression in CHO-flpin cells compared with endogenous expression levels in HT29 cells. Nevertheless, consistent with previous studies (Suen et al., 2011), the potency orders for the three PAR2 agonists was trypsin » 2f-LIGRLO- NH2 > GB110. Importantly, CHO-flpin cells stably transfected with vector were not responsive to PAR2 agonists at each concentration tested.

122

Figure 5.5 Dose-response curves of trypsin, 2f-LIGRLO-NH2 and GB110 on CHO-flpin cells stably transfected with vector or wildtype PAR2-FLAG CHO-flpin cells stably expressing empty vector (black line) or wildtype PAR2 (red line) were loaded with fluorescent calcium reporter FLUO-3 and treated with increasing concentrations of trypsin (A), 2f-LIGRLO-NH2 (B) or GB110 (C). Calcium mobilisation was measured using a FLIPR TETRA fluorescent plate reader. The amplitude of calcium mobilisation, normalised to maximal PAR2 response, was plotted against the corresponding agonist concentration to produce a sigmoidal curve in Graphpad Prism version 5.0. EC50 values (nM) for each agonist are shown in red. Data are representative of four independent experiments for A and B, and three independent experiments for C.

123 5.3.3.2 Dose-response curves for TM domains II and III PAR2 mutants Calcium flux dose-response curves of cells stably expressing PAR2 with mutations in TMs II and III treated with each PAR2 agonist were compared with dose-response curves from cells stably transfected with vector or wildtype PAR2. As shown in Figure 5.6A, PAR2 mutants F128A, L151A, F155A, Y156A, Y156L and Y160L in TMs II and III retained the ability to trigger trypsin-dependent calcium flux.

However, compared with wildtype PAR2 (EC50 3.5 ± 0.2), each mutant displayed rightward shifts in mutant dose response curves indicating reductions in trypsin agonist potency (increased trypsin EC50 values over wildtype PAR2). Unlike mutants

L151A and F155A which displayed relatively minor reductions in EC50 (EC50 of 5.7 ± 0.1 and 5.8 ± 0.2 respectively), compared with wildtype PAR2, mutants F128A, Y156A, Y156L and Y160L showed the most pronounced reduction in trypsin potency with EC50 values of 7.4 ± 0.2, 13.4 ± 0.1, 11.2 ± 0.1 and 17.0 ± 0.1 respectively (Figure 5.6A).

Consistently, all PAR2 mutants retained the ability to signal in response to agonists

2f-LIGRLO-NH2 (Figure 5.6B) and GB110 (Figure 5.6C). With the exception of mutant L151A, each mutant displayed reductions in 2f-LIGRLO-NH2 and GB110 potency (compare with wildtype EC50 values of 89.0 ± 0.3 and 145.9 ± 0.3 for 2f-

LIGRLO-NH2 and GB110 respectively). 2f-LIGRLO-NH2 displayed a relatively small increase in potency (54.3 ± 0.2 EC50 compared with 89.0 ± 0.3 EC50 for wildtype PAR2) compared with the increase in GB110 potency induced by mutant

L151A (59.9 ± 0.3 EC50 compared with 145.9 ± 0.3 EC50 for wildtype PAR2). As shown in Figure 5.6B, PAR2 mutants F128A and F155A induced relatively small reductions in 2f-LIGRLO-NH2 potency (145.9 ± 0.2 and 103.7 ± 0.3 respectively compared with wildtype EC50 89.0 ± 0.3) whereas mutant Y160L displayed a pronounced reduction in 2f-LIGRLO-NH2 potency (EC50 239.1 ± 0.2 versus wildtype EC50 89.0 ± 0.3). By contrast, 2f-LIGRLO-NH2-induced signalling by PAR2 mutants Y156A and Y156L was markedly impaired compared with the wildtype receptor (EC50 values of 5596.0 ± 0.2 and 818.8 ± 0.1 respectively). As shown in Figure 5.6C, mutants F128A and Y160L induced relatively small reductions in GB110 potency (EC50 of 207.1 ± 0.6 and 235.3 ± 0.2 respectively versus wildtype EC50 145.9 ± 0.2) whereas mutant F155A induced a pronounced reduction in potency (EC50 331.8 ± 0.3). Similar to 2f-LIGRLO-NH2, GB110

124

125

Figure 5.6 Dose-response curves for trypsin, 2f-LIGRLO-NH2 and GB110 induced by treatment of CHO-flpin cells stably transfected with vector, wildtype PAR2-FLAG or PAR2-FLAG with TM II and III mutants CHO-flpin cells stably transfected with vector (black line; ■), wildtype PAR2 (red line; ●) or PAR2 mutants in TMs II or III (black line; F128A ▲, L151A ▼, F155A ♦, Y156A ▲, Y156L ○, Y160L □) were loaded with fluorescent calcium reporter

FLUO-3 and treated with increasing concentrations of trypsin (A), 2f-LIGRLO-NH2 (B) or GB110 (C). Calcium mobilisation was measured using a FLIPR TETRA fluorescent plate reader. The amplitude of calcium mobilisation, normalised to maximal PAR2 response, was plotted against the corresponding agonist concentration to produce a sigmoidal curve in Graphpad Prism version 5.0. PAR2 agonist EC50 values (nM) are shown for each mutant. Data are representative of four independent experiments for A and B, and three independent experiments for C.

126 potency was also markedly reduced by mutants Y156A and Y156L (EC50 values of 8468.0 ± 0.3 and 2259.0 ± 0.2 respectively).

5.3.3.3 Dose-response curves for TM domains V and VI PAR2 mutants The effect of PAR2 mutations within TMs V and VI were next examined by dose- response curves for receptor agonists trypsin, 2f-LIGRLO-NH2 and GB110. As shown in Figure 5.7A, compared with wildtype PAR2, mutants V250A, L307A and

V308A caused minor reductions in trypsin potency (EC50 of 4.6 ± 0.3, 5.3 ± 0.2 and 4.3 ± 0.3 respectively versus 3.5 ± 0.2). As shown in Figure 5.7B, mutants V250A and V308A induced minor increases in 2f-LIGRLO-NH2 potency (69.6 ± 0.4 and 66.9 ± 0.3 respectively compared with 89.0 ± 0.3) while mutant L307A caused a small reduction in 2f-LIGRLO-NH2 potency (EC50 of 162.8 ± 0.2 versus 89.0 ± 0.3). As shown in Figure 5.7C, compared with wildtype PAR2, mutants V250A and V308A induced a pronounced reduction and increase in GB110 activity respectively

(EC50 of 556.7 ± 0.2 and 43.9 ± 0.1 versus 145.9 ± 0.3). Unlike trypsin and 2f-

LIGRLO-NH2, GB110 activity was markedly reduced by mutant L307A with an

EC50 value of 2180.0 ± 0.2 (compare with wildtype PAR2 EC50 145.9 ± 0.3).

5.3.3.4 Dose-response curves for ECL3 and TM domain VII PAR2 mutants The dose-response curves from CHO-flpin cells transiently transfected with vector, wildtype PAR2 or PAR2 mutations in ECL3 and TMs VII treated with trypsin, 2f- LIGRLO-NH2 and GB110 were next examined. As shown in Figure 5.8A, with the exception of mutant L330V which caused a minor reduction in trypsin potency (EC50 5.9 ± 0.2 versus 3.5 ± 0.2), mutants Y311L, Y326A and Y326L induced pronounced reductions in trypsin potency (EC50 values of 9.8 ± 0.3, 16.0 ± 0.2 and 11.7 ± 0.2 respectively, versus 3.5 ± 0.2). As shown in Figure 5.8B, 2f-LIGRLO-NH2 was equally effective at initiating signalling via mutants Y311L and L330V as with the wildtype receptor (EC50 86.1 ± 0.3 and 83.1 ± 0.3 versus wildtype EC50 of 89.0 ±

0.3). Mutant Y326L induced a pronounced reduction in 2f-LIGRLO-NH2 potency

(EC50 222.5 ± 0.2 versus 89.0 ± 0.3). However, mutation of Y326 to an alanine residue markedly reduced 2f-LIGRLO-NH2 potency by 119.5 fold (EC50 10630.0 ±

0.4 versus 89.0 ± 0.3). Unlike typsin and 2f-LIGRLO-NH2, as shown in Figure 5.8C, mutations Y311L and L330V caused a respective minor and pronounced increase in

GB110 potency (EC50 114.4 ± 0.2 and 44.9 ± 0.3 versus wildtype EC50 145.9 ± 0.3).

127

Figure 5.7 Dose-response curves for trypsin, 2f-LIGRLO-NH2 and GB110 induced by treatment of CHO-flpin cells stably transfected with vector, wildtype PAR2-FLAG or PAR2-FLAG with TM V and VI mutants CHO-flpin cells stably transfected with vector (black line; ■), wildtype PAR2 (red line; ●) or PAR2 mutants in TMs V or VI (black line; V250A ∆, L307A , V308A ◊) were loaded with fluorescent calcium reporter FLUO-3 a.m. and treated with increasing concentrations of trypsin (A), 2f-LIGRLO-NH2 (B) or GB110 (C). Calcium mobilisation was measured using a FLIPR TETRA fluorescent plate reader. The amplitude of calcium mobilisation, normalised to maximal PAR2 response, was plotted against the corresponding agonist concentration to produce a sigmoidal curve in Graphpad Prism version 5.0. PAR2 agonist EC50 values (nM) are shown for each mutant. Data are representative of four independent experiments for A and B, and three independent experiments for C.

128 Figure 5.8 Dose-response cuurves for trypsin, 2f-LIGRLO-NH2 and GB110 induced bby treatment of CHO-flpin cells stably transfected with vector, wildtype PAR2-FLAG or PAR2-FLAG with ECL3 and TM VII mutants CHO-flpin cells stably transfected with vector (black line; ■), wildtype PAR2 (red line; ●) or PAR2 mutants in ECCL3 or TM VII (black line, Y311L , Y326A , Y326L +, L330V ) loaded with fluorescent calcium reporter FLUO-3 a.m. were treated with increasing concentrations of trypsin (A), 2f-LIGRLO-NH2 (B) or GB110 (C). Calcium mobilisation was measured using a FLIPR TETRA fluorescent plate reader. The amplitude in calcium mobilisation, normalised tto maximal PAR2 response, was plotted against the corresponding agonist concenttrration to produce a sigmoidal curve in Graphpad Prrism version 5.0. PAR2 agonist EC50 values (nM) are shown for each mutant. Data are representative of four independeent experiments for A and B, and three independent experiments for C.

129 In contrast, mutant Y326L induced a pronounced reduction in GB110 activity (EC50

733.7 ± 0.1 compared with the EC50 of 145.9 ± 0.3 for wildtype PAR2). Interestingly, consistent with 2f-LIGRLO-NH2, mutation of Y326 to an alanine resulted in a marked reduction in GB110 activity (EC50 5508.0 ± 0.1 versus 145.9 ± 0.3).

5.3.3.5 Comparison of Potency and REC values for trypsin, 2f-LIGRLO-NH2 and GB110 identifies residues involved in PAR2 agonist binding

The EC50 values calculated from the dose-response curves of CHO-flpin cells transiently expressing wildtype or mutant PAR2 treated with trypsin and both PAR2

APs (see Figures 5.6-5.8) are shown in Table 5.3. The fold change in EC50 for each

PAR2 mutant relative to the wildtype receptor (referred to as REC) was also calculated for the three PAR2 agonists (Al-Ani et al., 1999; Al-Ani et al., 2002b);

Table 5.3). For ease of analysis, REC values will be referred to for the remainder of this chapter. As shown in Table 5.3, relative to wildtype PAR2, all PAR2 mutations caused reductions in trypsin agonist REC values, indicated by values greater than 1.0. PAR2 mutations L151A, F155A, V250A, L307A, V308A and L330V produced lower reductions in trypsin potency with REC values less than 2.0, suggesting these residues may not be directly involved in trypsin-induced TL binding. The REC values for PAR2 mutants F128A (2.1), Y156A (3.8), Y156L (3.2), Y160L (4.9), Y311L (2.8), Y326A (4.6) and Y326L (3.3) demonstrated pronounced reductions in trypsin potency and point to a role in TL binding. However it is noted that changes in potency of trypsin-induced signalling were much lower than for the other 2 agonists, as described in the following paragraph.

REC values for 2f-LIGRLO-NH2 treated cells demonstrated small increases in agonist activity caused by PAR2 mutants L151A (0.6), V250A (0.8), V308A (0.8), Y311L (0.97) and L330V (0.9; Table 5.3). In contrast, PAR2 mutations F128A (1.6), F155A

(1.2) and L307A (1.8) resulted in small reductions in 2f-LIGRLO-NH2 activity. Of note, PAR2 mutations Y156A (62.9), Y156L (9.2) and Y326A (119.5) resulted in marked reductions in 2f-LIGRLO-NH2 activity, suggesting that residues Y156 and

Y326 are critical for 2f-LIGRLO-NH2 stimulated signal transduction. REC values for GB110 treated cells demonstrated a small increase in agonist activity caused by mutant Y311L (0.8) and pronounced increases in GB110 activity induced by

130

Table 5.3 Comparison of trypsin, 2f-LIGRLO-NH2 and GB110 potency values from dose-response curves of wildtype and mutant PAR2 stably expressing cells EC50 values were determined from dose-response curves of CHO-flpin cells stably transfected with wildtype and mutant PAR2 treating with increasing concentrations

of PAR2 agonists trypsin, 2f-LIGRLO-NH2 and GB110. Residues with the greatest

REC (ratio of PAR2 mutant EC50 to wildtype PAR2 EC50) for each PAR2 agonist are highlighted in bold.

Trypsin 2f‐LIGRLO‐NH2 GB110 PAR2 EC (nM) R EC (nM) R EC (nM) R mutations 50 EC 50 EC 50 EC PAR2 3.5 ± 0.2 1.0 89.0 ± 0.3 1.0 145.9 ± 0.3 1.0 F128A 7.4 ± 0.2 2.1 145.9 ± 0.2 1.6 207.1 ± 0.6 1.4 L151A 5.7 ± 0.1 1.6 54.3 ± 0.2 0.6 59.9 ± 0.4 0.4 F155A 5.8 ± 0.2 1.6 103.7 ± 0.3 1.2 331.8 ± 0.3 2.3 Y156A 13.4 ± 0.1 3.8 5596.0 ± 0.2 62.9 8468.0 ± 0.3 58.0 Y156L 11.2 ± 0.1 3.2 818.8 ± 0.1 9.2 2259.0 ± 0.2 15.5 Y160L 17.0 ± 0.1 4.9 239.1 ± 0.2 2.7 235.3 ± 0.2 1.6 V250A 4.6 ± 0.3 1.3 69.9 ± 0.4 0.8 556.7 ± 0.2 3.8 L307A 5.3 ± 0.2 1.5 162.8 ± 0.2 1.8 2180.0 ± 0.2 14.9 V308A 4.3 ± 0.3 1.2 66.9 ± 0.3 0.8 43.9 ± 0.1 0.3 Y311L 9.8 ± 0.3 2.8 86.1 ± 0.3 0.97 114.4 ± 0.1 0.8 Y326A 16.0 ± 0.2 4.6 10630.0 ± 0.2 119.5 5508.0 ± 0.1 37.8 Y326L 11.7 ± 0.2 3.3 222.5 ± 0.2 2.5 733.7 ± 0.1 5.0 L330V 5.9 ± 0.2 1.7 83.1 ± 0.3 0.9 44.9 ± 0.3 0.3

131 mutations L151A (0.4), V308A (0.3) and L330V (0.3; Table 5.3). In contrast, mutations F128A (1.4) and Y160L (1.6) yielded small reductions in GB110 activity whereas the reductions in GB110 activity induced by mutations F155A (2.3), V250A

(3.8) and Y326L (5.0) were more pronounced. Consistent with 2f-LIGRLO-NH2, the activity of GB110 was markedly reduced in PAR2 mutations Y156A (58.0), Y156L

(15.5) and Y326A (37.8). Interestingly, unlike 2f-LIGRLO-NH2 REC values for mutant L307A, GB110 activity was markedly reduced by L307A (14.9). These data indicate that residues Y156, L307 and Y326 are critical for GB110 stimulated signal transduction.

132 5.4 Discussion As indicated in Chapter 4, palmitoylation of a single PAR2 residue (C361) can regulate receptor signalling and trafficking. The goal of this chapter was to identify residues within PAR2, particularly within the hydrophobic TM domains, that are critical for ligand binding to activate receptor signalling. In this chapter, the binding of the endogenous TL (SLIGKV), revealed through trypsin cleavage, as well as the binding of two soluble PAR2 agonists, 2f-LIGRLO-NH2 and GB110, was examined. Twelve PAR2 residues potentially important in ligand binding were first identified using a homology model of PAR2 based on the prototypical GPCR rhodopsin and in silico ligand docking. These were mutated to examine the effect of altered amino acid side chains on ligand activated calcium mobilisation. The key findings of this chapter are that 1) relatively small reductions for trypsin ligand potency were caused by mutations F128A, Y156A, Y156L, Y160L, Y311L, Y326A and Y326L; 2) residues Y156 and Y326 were critical for full 2f-LIGRLO-NH2 agonist activity; and 3) residues Y156, L307 and Y326 were critical for full GB110 agonist activity.

One interesting finding from this chapter is the relatively small potency differences (Table 5.3) caused by PAR2 mutations on the ability of trypsin to stimulate signalling relative to the marked impact of several mutations on 2f-LIGRLO-NH2 and GB110 induced signalling. This finding is similar to those of two previous studies examining the role of ECL2 in PAR2 ligand binding and activation (Al-Ani et al., 1999; Al-Ani et al., 2002b). In these reports mutations in the TL and ECL2 of PAR2 had minimal impact on trypsin-induced calcium mobilisation. By contrast, calcium mobilisation induced by the PAR2 AP SLIGRL-NH2 was markedly reduced by these mutation. Al-ani and colleagues proposed that different structure-activity relationships exist for trypsin and the soluble PAR2 AP, perhaps indicating that other ligand docking sites exist (Al-ani et al., 2002). Interestingly, a more recent report identified that while the trypsin revealed mutated TL sequences LSIGRL- and AAIGRL- were capable of stimulating both calcium mobilisation and MAPK signal transduction pathways, soluble PAR2 APs of the same sequence, LSIGRL-NH2 and

AAIGRL-NH2, were unable to initiate signal transduction by either pathway (Ramachandran et al., 2009). The authors further hypothesised that the TL and soluble ligands used in these studies bind differentially or stabilise distinct receptor conformations to initiate individual signalling pathways (Ramachandran et al.,

133 2009). Thus in the same way, it is possible that the human PAR2 TL, SLIGKV, revealed by trypsin cleavage, docks in a binding pocket distinct from the soluble ligands tested in this chapter. It may also be possible that the residues tested herein do not bind directly with the TL, but participate in interactions between residues to stabilise certain receptor conformations. Therefore, loss of these residues may result in the small shifts in trypsin potency observed in this chapter by destabilising a receptor conformation.

Another possibility is that the relatively small potency shifts seen for trypsin result from an agonist bias. Biased agonism is the ability of an agonist to dock within a cognate receptor and activate certain signalling pathways (Andresen, 2011). In this way, a recent report identified TL residues that are important for triggering pathway specific signal transduction. In this study, wildtype PAR2 and SLIGRL-NH2 induced MAPK and calcium signalling whereas the mutated TL sequence SLAAAA- and soluble PAR2 AP, SLAAAA-NH2, induced MAPK signalling but failed to stimulate calcium mobilisation (Ramachandran et al., 2009). Interestingly, peptides AAIGRL-

NH2 and LSIGRL-NH2 failed to induce signal transduction (Ramachandran et al., 2009). In addition, mutation of TMII and ECL2 residues within the M2 muscarinic receptor (Y104A and Y177A) selectively prevented agonist-induced MAPK signalling but not calcium mobilisation (Gregory et al., 2010). These studies highlight the importance of receptor residues, within the TL and TM domains, to initiate pathway specific signal transduction. As such, it would be worthwhile to examine trypsin-induced MAPK signalling via the PAR2 mutants generated in this chapter to identify an impact on this signalling pathway.

This chapter has identified residues that merit further study. Two residues, Y160 and Y326, when mutated to leucine and alanine residues respectively, had the greatest impact on calcium mobilisation in response to trypsin (REC values of 4.9 (Y160L) and 4.6 (Y326A)). PAR2 residues Y156, mutated to alanine and leucine residues, and Y326, mutated to a leucine, also caused a reduction in responsiveness to trypsin

(REC values of 3.82 (Y156A), 3.19 (Y156L) and 3.34 (Y326L) respectively). Although mutation of these residues did not alter the responsiveness to trypsin to the same extent as to the two soluble agonists, the data suggest that residues Y156, Y160 and Y326 contribute, either directly or indirectly, to TL binding and PAR2

134 signalling. These data may also highlight that the regions around residues Y156, Y160 and Y326 are important for SLIGKV binding. To address this possibility, further mutations surrounding these residues are required.

To examine this possibility that other residues or regions of PAR2 are involved in TL docking, further modelling is also required. Although rhodopsin is a prototypical GPCR, difficulties exist in generating homology models based on the rhodopsin crystal structure. For example, due to relatively low sequence homology, there is inherent uncertainty in aligning PAR2 and rhodopsin sequences within TM helices (Topiol and Sabio, 2009). Also, the rhodopsin crystal structure is based on the inactive state of this receptor. Because multiple conformational states are likely to exist for receptors bound to different ligands (Reggio, 2006), it is difficult to predict regions involved in ligand docking based on this static model (Topiol and Sabio, 2009). Hence, it is possible that the PAR2 residues identified by in silico modelling in this chapter are not directly involved in TL binding. As other GPCR crystal structures have since been reported, use of these structures to generate PAR2 homology models based on the ligand activated adenosine A2 and β2-adrenergic receptors (Bokoch et al., 2010; Lebon and Tate, 2011; Rasmussen et al., 2007; Rasmussen et al., 2011b; Rosenbaum et al., 2011; Xu et al., 2011) would be beneficial. These models may also be used to verify the amino acids identified as important for soluble ligand binding.

In contrast with trypsin, the REC values for 2f-LIGRLO-NH2 and GB110 across PAR2 mutants varied greatly. A key finding from this chapter is that residues Y156 and Y326 are critical for full 2f-LIGRLO-NH2 and GB110 agonist activity. In addition, residue L307 was also required for full GB110 activity. It is important to note that these findings may suggest one of two possibilities: (1) these residues are required for supporting an allosteric conformation induced by docking of these soluble PAR2 agonists, or (2) these residues are required for direct interaction with both agonists. However, as shown in Figure 5.9A-B, these residues are clustered together and are located toward the top of the model of the PAR2 TM bundle. The side chains of residues Y156, L307 and Y326 point towards the central cavity of the TM bundle in this homology model of PAR2. Given the proximity and location of each identified residue (yellow), it is possible that these amino acids contribute to

135

Figure 5.9 Three dimensional representation of PAR2 residues 156, 307 and 326 and structural features of 2f-LIGRLO-NH2 and GB110 A, Side view of the PAR2 homology model TM bundle (cyan ribbon) with residues 156, 307 and 326 highlighted in yellow. Residues 201-240 of PAR2, pertaining to TM IV were removed to view the highlighted residues. B, Top view of the extracellular surface and TM bundle of the PAR2 homology model (cyan ribbon). Residues 156, 307 and 326 are highlighted in yellow. Comparison of the chemical structures for PAR2 agonists (C) 2f-LIGRLO-NH2 and (D) GB110. Regions common to 2f-LIGRLO-NH2 and GB110 are shown in black. Regions unique to 2f-

LIGRLO-NH2 are highlighted in blue while regions unique to GB110 are highlighted in red. *, highlights a carbonyl found in similar positions of C and D.

136 forming a binding pocket or docking site within the transmembrane domains for PAR2 agonists. In this way, the phenolic components of Y156 and Y326 may provide hydrogen bonding whereas L307 may provide an aliphatic interaction with

GB110. Figure 5.9C-D shows the structure of 2f-LIGRLO-NH2 and GB110 highlighting regions common and unique to both agonists. As residues Y156 and

Y326 affect both 2f-LIGRLO-NH2 and GB110 activity, it is likely these interact with regions shared between each agonist. Based on the preliminary agonist docking modelling, the 2-furanoyl cap of both agonists is possibly buried partway down the TM bundle, towards the intracellular surface of PAR2 (personal communication, Prof. David Fairlie). Speculatively, Y156, as the residue lowest down on the TM bundle, may interact with the 2-furanoyl moiety of each agonist, whereas Y326, as the residue located highest on the TM bundle, may participate in docking of the shared isoleucine or glycine residues between both agonists. Therefore, PAR2 residue L307, identified to exclusively impact GB110 activity, may interact with a region unique to this agonist (Red; Figure 5.9D). Of course, it is also possible that residues Y156 and Y326 may interact with regions outside of the regions shared by

2f-LIGRLO-NH2 and GB110. For example, Y326 may interact with the carbonyl group found in a similar position on both agonists (see asterisk in Figures 5.9C and 5.9D). However, further agonist modelling and computer docking are required to more accurately predict the points of interaction between both PAR2 agonists. Also, comparison of the affinity of the PAR2 agonists for wildtype and mutant receptors in the presence of a competitive antagonist may aid in defining the agonist binding pocket. In practice this would involve determining the concentration ratio for selected agonist/antagonist pairs. This ratio, which is a measure of agonist affinity, is the EC50 value for the agonist in the presence of saturating quantities of antagonist, divided by the EC50 value for agonist alone. Agonists could include the endogenous tethered ligand released by trypsin, or the soluble agonists 2f-LIGRLO-NH2 and GB110 and a suitable antagonist would be the recently described GB88 (Suen et al., 2011). Assuming a single binding site exists for these agonists within PAR2, differences in concentration ratios observed for wildtype and mutant receptors will assist in determining if agonist binding occurs at the same site as the antagonist and if each agonist binds at the same site as the other agonists.

137 It is noteworthy that the potential ligand docking site identified in this chapter is in keeping with the identified ligand docking sites found in other GPCRs. For example, receptor homology modelling and site directed mutagenesis studies of rhodopsin (Lu et al., 2002; Shi and Javitch, 2004; Strader et al., 1994; Wess, 1996), the neurotensin receptor 1 (Barroso et al., 2002; Labbe-Jullie et al., 1998), M1 (Huang et al., 1999; Hulme et al., 2003) and M3 muscarinic receptors (Wess, 1996) have defined receptor agonist and antagonist docking to within the hydrophobic TM helices of these GPCR surfaces (Blakeney et al., 2007). It will be interesting to assess whether the residues identified in these studies of other GPCRs align with the amino acids identified for PAR2 in this chapter.

Previous reports have suggested that the ECL2 of PAR2 is important for ligand docking (Al-ani et al., 1999; Al-ani et al., 2002; Lerner et al., 1996). In particular, substitution of the PAR1 ECL2 with the corresponding PAR2 ECL2 renders PAR1 responsive to PAR2 APs (Gerszten et al., 1994) while PAR2 ECL2 residues P231EE of rat PAR2 (residue sequence PEQ in human PAR2) affect receptor activation (Al- ani et al., 1999). In support of this proposal, residues within the ECL2 of the M3 muscarinic acetylcholine receptor (Scarselli et al., 2007), histamine H1(Strasser et al., 2008) and H4 receptors (Lim et al., 2008), dopamine D2 receptor (Shi and Javitch, 2004) and V1a vasopressin receptor (Conner et al., 2007) also directly participate in ligand docking. Importantly, the structure of rhodopsin also suggests that the ECL2 projects into the binding crevice between the TM domains (Palczewski et al., 2000). An important role for ECL2 in ligand docking is also suggested by the observation that while the structure of ECL2 for many GPCRs, including PAR2, is thought to be maintained by a disulphide bridge between conserved cysteine residues in TM III and ECL2 (Hamm et al., 2001; Palczewski et al., 2000; see blue dashed line in Figure 5.2), the ECL2 of the serotonin 5-HT(4a) receptor is also reported to undergo a conformational change upon receptor activation (Baneres et al., 2005). Therefore, although this study has identified the importance of several TM residues for initiation of PAR2 signalling, it is also likely that residues identified by Al-ani and colleagues (1999; 2002), as well as others within ECL2, contribute to PAR2 signalling either by direct ligand docking or stabilisation of a receptor conformation required for efficient signalling. Further

138 studies examining concurrent TM and ECL2 mutants are required to examine more closely the role of ECL2 in ligand docking.

Having identified several residues critical for initiation of PAR2 signal transduction, this chapter provides the basis for identifying other residues likely to participate in ligand docking as well as for studying the binding of other soluble PAR2 agonists and antagonists. These data, and future receptor:agonist computer modelling, will provide a starting point for the rational design of agonists and antagonists targeting PAR2 with greater efficacy and potency.

139

140

Chapter 6:

Examination of a role for PAR2 in prostate cancer bone metastasis

141

142 6.1 Introduction In previous chapters, the mechanisms surrounding PAR2 activation and signalling were investigated. In this chapter, the role of PAR2 in prostate cancer bone metastases is examined.

Prostate cancer is the most common soft tissue tumour afflicting men (Stanford et al., 1999). In late stage disease up to 90% of cases exhibit metastases to the skeletal system (Bubendorf et al., 2000; Kingsley et al., 2007; Mundy, 2002). Once established in bone, prostate tumours, unlike tumours from other primary sites (e.g. lung or breast cancer), form new woven bone that is accompanied by varying degrees of bone loss (osteolysis; (Koutsilieris, 1993). This osteolytic component of prostate cancer bone metastases permits the establishment of metastasised tumour cells within bone (Keller and Brown, 2004; Roudier et al., 2008). Patients with these lesions suffer debilitating bone pain (Mantyh, 2004; Mantyh and Hunt, 2004), bone fragility and hypercalcaemia (increased calcium levels in blood) (Mundy, 2002).

The chemotherapeutic drug docetaxel is currently used for the treatment of patients with prostate cancer bone metastasis and weekly treatment with this agent provides survival benefits of 2-3 months (Petrylak et al., 2004; Tannock et al., 2004). Docetaxel is a structural analogue of paclitaxel which is a natural extract from the plant Taxus brevifolia (Clarke and Rivory, 1999) and functions by stabilising microtubules and preventing actin depolymerisation (Eisenhauer and Vermorken, 1998). In this way docetaxel inhibits mitosis by preventing assembly of the mitotic spindle (Yvon et al., 1999). Because of the limited survival advantage provided by currently available anti-prostate cancer drugs, a greater understanding of the molecular mechanisms of bone metastatic prostate cancer is required to identify alternative or adjunctive therapies to improve patient care.

The proclivity to develop secondary prostate tumours in bone is not solely due to haematogenous dissemination, but also due to the fertile bone microenvironment (Ahmad et al., 2008; Baylink et al., 1993; Kaplan et al., 2006a; Kaplan et al., 2006b). This environment consists of two types of osseous tissue; cortical and trabecular bone. As shown in Figure 6.1, cortical bone forms the compact exterior of bone with spongy trabecular bone filling the interior of the epiphysis and metaphysis

143 but not the diaphysis of long bones (Seeman and Delmas, 2006). The interior surface of cortical bone in contact with trabecular bone is defined as the endosteal surface whereas the exterior of cortical bone delineates the periosteal surface (Figure 6.1). The structure and homeostasis of bones is maintained by cross-talk between the resident bone cells; osteoblasts which deposit bone matrix and osteoclasts which resorb this matrix (Seeman and Delmas, 2006). In this regard, osteoblasts regulate osteoclast differentiation through secretion of receptor for activation of nuclear factor kappa B ligand (RANKL) which binds to the RANK receptor on osteoclasts (Logothetis and Lin, 2005). RANKL is further regulated by the osteoblast secreted factor osteoprotegerin (OPG) which acts as a decoy receptor for RANKL (Logothetis and Lin, 2005). Increases in RANKL:OPG ratio result in osteoclast differentiation. Prostate cancer cells colonising the bone also interact with bone cells to promote a vicious cycle, disrupting the fine balance between osteoblasts and osteoclasts to allow tumour growth in bone (Guise et al., 2006). In the bone microenvironment, prostate tumour cells secrete factors such as RANKL to stimulate osteoclasts (Dougall et al., 1999) while also secreting factors to stimulate osteoblasts including Wnt proteins (Hall et al., 2005; Hall and Keller, 2006) and endothelin-1 (Nelson et al., 1999; Yin et al., 2003). The bone milieu also promotes a permissive environment for tumour growth by the secretion of transforming growth factor-β, insulin-like growth factor-I and -II, fibroblast growth factor-1 and -2 and vascular endothelial growth factor (Dai et al., 2004; Logothetis and Lin, 2005; Ye et al., 2008). Therefore, the reciprocal interaction of cancer cell and bone microenvironment leads to exaggerated stimulation of bone remodelling.

As discussed in chapter 1, a number of reports have implicated PAR2 in the regulation of both bone cells and tumour progression. For example, activation of PAR2 expressed in osteoblasts leads to the stimulation of collagen type I mRNA expression, important for bone formation, and also apoptosis, which regulates bone turnover (Abraham et al., 1998; Georgy et al., 2010; Smith et al., 2004). Moreover, PAR2 signalling also inhibits osteoblast-mediated osteoclast differentiation by reducing the RANKL:OPG ratio (Georgy et al., 2010).

Importantly, in addition to a role in bone, recent reports indicate that PAR2 is functionally important in cancer cells. Signalling via this receptor enhances

144

Figure 6.1 Bone structure and site for intra-tibial injection of prostate cancer cells in mouse A, Bone structure. Bone consists of two osseous tissues; compact cortical bone and spongy trabecular bone. Long bones (including tibia) can be compartmentalised into sections; the diaphysis consisting of cortical bone and bone marrow; proximal ephiphysis and proximal metaphysis, both consisting of cortical bone and trabecular bone. B, Intra-tibial injection. Prostate cancer cells are injected into the metaphysis on the proximal aspect of tibia of mice by drilling through the proximal ephiphysis with a 26-gauge needle.

145 proliferation, invasion or migration of cell lines derived from colon, melanoma, breast and gastric cancers (Caruso et al., 2006; Darmoul et al., 2004a; Ge et al., 2004; Hjortoe et al., 2004; Morris et al., 2006). PAR2 signalling also contributes to tumour growth of MDA-MB-231 breast cancer xenografts (Versteeg et al., 2008). Moreover, PAR2 activation in prostate cancer LNCaP cells induces cytoskeletal reorganisation via Rho family members (Greenberg et al., 2003), a mechanism crucial to cellular migration (Titus et al., 2005). Also, elevated expression of PAR2 is reported in benign prostatic hyperplasia (Mannowetz et al., 2010) and prostate cancer patient tissues (Black et al., 2007; Ramsay et al., 2008a). In addition to expression by normal osteoblasts (Abraham et al., 2000; Smith et al., 2004), PAR2 is expressed by osteoblasts proximal to skeletal prostate cancer tumours (Ramsay et al., 2008a). As these reports suggest a role for PAR2 in both bone cells and tumour cells, this chapter explores the possibility that this receptor regulates the interplay between these cells in prostate cancer bone metastasis.

A number of in vivo models have been previously reported to mimic human disease. These include (i) generation of a transgenic mouse that spontaneously forms metastatic prostate tumours (Gingrich et al., 1996); (ii) the tail vein injection of primary human prostate cancer cells into mice previously implanted with human fetal bone (Nemeth et al., 1999); and (iii) subcutaneous or intra-cardiac injection of prostate cancer cell lines (Thalmann et al., 1994). However, in animal models bone metastasis from a primary xenograft site is rare (Niu et al., 2008b; Niu et al., 2008c; Ren et al., 2008; Thalmann et al., 1994; Williams et al., 2007; Wu et al., 1998; Yang et al., 1999; Zhau et al., 2000). In this chapter, to examine the function of PAR2 in prostate cancer bone metastasis, prostate tumour cells were injected directly into mouse tibiae. The injection of prostate cancer cells into the proximal metaphysis of SCID mouse tibiae (Figure 6.1) has been used previously to study prostate cancer bone metastasis (Corey et al., 2002; Navone et al., 1998). However, a weakness of many of these models is the use of prostate cancer cell lines that do not form osteoblastic lesions (Geldof and Rao, 1990; Navone et al., 1998). As shown in Table 6.1, intra-tibial injection of PC-3 and DU145 prostate cancer derived cell lines results in osteolytic lesions, which as described above, does not adequately represent human disease. DU145 and PC-3 cells induce progressive osteoclast-induced degradation of cortical bone with PC-3 cells being the most aggressive

146

Table 6.1 Summary of prostate cancer cell lines and observed phenotype

Cell Source Phenotype in bone References

line (mouse model) PC‐3 Vertebral Osteolytic: osteoclast‐ (Corey et metastases from induced osteolysis of cortical al., 2002; hormone‐ bone Fritz et al.,

insensitive patient 2007; Kaighn et al., 1978) DU145 Human prostate Osteolytic: osteoclast‐ (Najy et al., cancer induced osteolysis of cortical 2011; Stone metastasised to bone et al., 1978) the brain

LNCaP Human prostate Osteoblastic & osteolytic: (Corey et cancer osteolysis of cortical bone al., 2002; metastasised to and osteoblastic trabecular Horoszewicz lymph nodes bone thickening et al., 1983) 22Rv1 Derived from an Osteoblastic & osteolytic (Henry et

outgrowth of the lesions: cortical bone al., 2005; castrate‐resistant osteolysis and periosteal Sramkoski CWR22R osteoblastic spicule et al., 1999) transplantable formation prostate cancer tumour C4‐2B Subline of LNCaP‐ Osteoblastic & osteolytic (Hall et al.,

C4 cells, derived lesions: concurrent cortical 2005; Jia et from intra‐cardiac bone osteolysis, thickening of al., 2008; xenograft of cells trabecular bone and Wu et al., metastasised to replacement of bone marrow 1998) bone with bone MDA‐ Paraspinal Osteoblastic: concurrent (Kundra et PCa‐2b meatstases from thickening of cortical and al., 2007;

castrate‐resistant trabecular bone Navone et patient al., 1997)

147 (Fritz et al., 2007; Najy et al., 2011). By contrast, intra-tibial injection of LNCaP, 22Rv1 and LNCaP C4-2B prostate cancer cells yield mixed osteolytic and osteoblastic lesions (Corey et al., 2002; Hall et al., 2005; Henry et al., 2005; Jia et al., 2008) whereas MDA-PCa-2b cells induce osteoblastic intra-tibial lesions (Kundra et al., 2007)(Table 6.1). Hence these cell lines more closely parallel human disease by forming either mixed osteolytic/osteoblastic or totally osteoblastic lesions. As such, the 22Rv1, LNCaP C4-2B subline and MDA-PCa-2b prostate cancer cell lines were evaluated for functional PAR2 in this study.

The impact of prostate cancer cells on bone is commonly assessed by qualitative radiographic analyses (Angelucci et al., 2004; Rosol et al., 2003). However, in order to more accurately assess the impact of prostate cancer cells on bone structure, quantitative micro-computed tomography (CT) is required. Unlike radiographic analyses, micro-CT imaging allows three-dimensional (3D) representations of bone and is a powerful non-destructive tool used to separately quantify cortical and trabecular bone morphology, identifying architectural changes in bones from small animals in tumour models (Arrington et al., 2006; Kurth and Muller, 2001; Ravoori et al., 2010). Analysis of cortical and trabecular bone is key to understanding the impact of prostate cancer cells on the skeletal system (Ravoori et al., 2010). Micro- CT provides accurate measurements of bone structure, closely correlating with measurements from destructive bone histomorphometry (Kapadia et al., 1998; Muller et al., 1998; Schmidt et al., 2003). These measurements allow for an accurate assessment of the impact of prostate cancer cells on bone morphology (Bouxsein et al., 2010).

Described in this chapter are experiments examining the role of PAR2 in bone located prostate tumours. A role for PAR2 was assessed by using animals bearing intra-tibial prostate tumours treated with the PAR2 antagonist GB88. As described in

Chapter 1, GB88 blocks PAR2-Gαq signalling in human HT29 colon cancer cells induced in vitro by several types of receptor agonists including protease (typsin), peptide (2f-LIGRLO-NH2) and nonpeptide (GB110) ligands (Adams et al., 2011; Suen et al., 2011). Importantly, in vivo administration of GB88 also reduces PAR2- mediated inflammation in animal models of oedema and experimentally-induced colitis (Suen et al., 2011; Lohman et al., 2011), suggesting that GB88 efficiently

148 blocks human and mouse PAR2 signalling. As controls, animals bearing intra-tibial tumours were also treated with vehicle (olive oil) or the prostate cancer chemotherapeutic docetaxel (Petrylak et al., 2004; Tannock et al., 2004). The effect of these treatments on bone was examined radiographically and by micro-CT. Importantly, the results presented in this chapter contribute to a larger study, where tumour-bearing tibia will also be subjected to histological and histomorphometry analyses. These additional analyses were not able to be performed due to time constraints but once completed will complement the results presented herein by defining the role of PAR2 in the prostate tumour (histology) and on in vivo osteoblast and osteoclast activity (histomorphometry).

149 6.2 Results 6.2.1 PAR2 expression in three osteoblastic prostate cancer cell lines To determine the most appropriate prostate cancer cell line for use in this chapter, three widely studied human prostate cancer cell lines with osteoblastic phenotypes were evaluated for PAR2 expression. These were 22Rv1, C4-2B and MDA-PCa-2b cells (summarised in Table 6.1). PAR2 mRNA expression was first examined by qRT-PCR relative to the housekeeping gene HPRT1. As shown in Figure 6.2A, C4- 2B cells express the highest level of PAR2 mRNA compared to MDA-PCa-2b and 22Rv1 cells. PAR2 expression was also examined at the protein level using flow cytometry with the anti-PAR2 antibody N19 (see Chapter 3.3.8; Suen et al., 2011). Cells lifted non-enzymatically were stained with the anti-PAR2 antibody N19 and fluorescently conjugated secondary antibodies. As shown in Figure 6.2B, PAR2 protein localises to the cell surface of each osteoblastic prostate cancer cell line. Comparing the secondary only signal with the signal for N19, higher levels of cell surface PAR2 were detected in C4-2B cells (12.34 ± 0.7) compared to MDA-PCa-2b (9.67 ± 0.4) and 22Rv1 (7.21 ± 0.8) cells, emulating PAR2 mRNA expression by these cells. These data indicate that PAR2 is expressed by each osteoblastic prostate cancer cell line.

6.2.2 PAR2 mediates signal transduction in 22Rv1 but not MDA-PCa-2b or C4-2B osteoblastic prostate cancer cells The ability of these cells to signal via PAR2 was next examined. Two signal transduction pathways induced by PAR2 activation were analysed: (1) ERK1/2 phosphorylation, triggered by PAR2 activation via G-protein-dependent and - independent pathways and (2) calcium mobilisation, triggered by PAR2 coupled to

Gαq (Soh et al., 2010).

6.2.2.1 PAR2 activation induces ERK1/2 phosphorylation only in 22Rv1 cells To assess PAR2 mediated induction of ERK1/2 phosphorylation, whole cell lysates were collected from the osteoblastic prostate cancer cell lines treated with 100 µM PAR2 AP or 50 ng/ml EGF (positive control for ERK1/2 phosphorylation). Lysates were subjected to anti-phosphorylated ERK1/2 and anti-total ERK1/2 Western blot analysis. As shown in Figure 6.3A (left panel), PAR2 AP induced ERK1/2 phosphorylation in 22Rv1 prostate cancer cells. Graphical analysis of densitometric

150

Figure 6.2 PAR2 expression in three osteoblastic prostate cancer cell lines. A, Quantitative RT-PCR analysis of PAR2 mRNA expression in 22Rv1, C4-2B and MDA-PCa-2b osteoblastic prostate cancer cell lines. PAR2 mRNA expression normalised, to housekeeping gene HPRT1, is represented graphically relative to receptor expression in 22Rv1 cells. B, Anti-PAR2 antibody N19 flow cytometry analysis of non-permeabilised 22Rv1, C4-2B and MDA-PCa-2b cell lines. Cells were either incubated with secondary only (grey population) or antibody N19 followed by secondary antibody (open population) before analysis. MFI values: 22Rv1 cells with N19 7.21 ± 0.8; 22Rv1 secondary only signal 0.48 ± 0.1; C4-2B cells with N19 12.34 ± 0.7; C4-2B secondary only signal 0.46 ± 0.1; MDA-PCa-2b cells with N19 9.67 ± 0.4; MDA-PCa-2b secondary only signal 0.51 ± 0.1. The data in panels A and B are representative of three independent experiments.

151

Figure 6.3 PAR2 activation induces ERK1/2 phosphorylation in 22Rv1 prostate cancer cells but not C4-2B or MDA-PCa-2b prostate cancer cell lines. A, Anti-phospho ERK (p-ERK) and anti-ERK Western blot analysis of whole cell lysates collected from 22Rv1 prostate cancer cells treated with PAR2 AP (100 µM) for the indicated times. Graphical representation of densitometric analysis of the ratio of p-ERK to ERK at each time point is shown in right panel. Data are representative of two independent experiments. Anti-p-ERK and anti-t-ERK Western blot analysis of (B) C4-2B and (C) MDA-PCa-2b prostate cancer cell line lysates treated with either PAR2 AP (100 µM) or EGF (50 ng/ml) for the indicated times. Data are representative of three independent experiments.

152 analysis of two independent Western blot analyses normalised to total ERK1/2 indicated that PAR2 AP induced maximal ERK1/2 phosphorylation at 5 minutes with activation reducing to background levels by 15 minutes (Figure 6.3A right panel). In contrast, C4-2B and MDA-PCa-2b prostate cancer cell lines treated with PAR2 AP did not induce ERK1/2 phosphorylation at any time point tested (Figure 6.3B and 6.3C respectively). Each cell line exhibited robust ERK1/2 phosphorylation in response to EGF treatment at 5 minutes indicating that signalling via the ERK pathway is intact in these cells (Figure 6.3A and 6.3B).

6.2.2.2 PAR2 initiates calcium mobilisation in 22Rv1 prostate cancer cells but not C4-2B or MDA-PCa-2b cells The ability of PAR2 AP to initiate calcium mobilisation in each prostate cancer cell line was next assessed by calcium flux assays. As shown in Figure 6.4A, treatment of 22Rv1 prostate cancer cells with PAR2 AP yielded robust calcium mobilisation. In contrast, PAR2 AP did not initiate calcium mobilisation in either C4-2B (Figure 6.4B) or MDA-PCa-2b prostate cancer cell lines (Figure 6.4C).

Collectively, these data demonstrate that PAR2 is capable of mediating signal transduction via G-protein dependent and independent pathways in 22Rv1 cells but not C4-2B or MDA-PCa-2b prostate cancer cell lines at the time points tested. Therefore, 22Rv1 cells were selected to examine the role of PAR2 in prostate cancer bone metastases.

6.2.2.3 PAR2 antagonist GB88 blocks receptor-mediated calcium mobilisation in 22Rv1 prostate cancer cells Having established that 22Rv1 prostate cancer cells are able to signal via PAR2, the ability of the PAR2 antagonist GB88 (see Chapter 1; Adams et al., 2011; Suen et al., 2011) to block receptor-mediated calcium mobilisation in this cell line was examined. 22Rv1 prostate cancer cells were pre-treated in the absence (DMSO) or presence of GB88 for 15 minutes before sequential challenge with PAR2 and PAR1 AP. As shown in Figure 6.4D (top panel), vehicle treated 22Rv1 cells activated with PAR2 AP and followed by PAR1 AP yielded robust calcium mobilisation induced by both receptor agonists. However, 22Rv1 cells pre-treated with GB88 and activated with PAR2 AP did not yield a response whereas PAR1 AP-induced a robust flux.

153

Figure 6.4 PAR2 activation induces calcium mobilisation in 22Rv1 but not C4- 2B or MDA-PCa-2b prostate cancer cells and signalling is blocked by the PAR2 antagonist GB88 (A) 22Rv1, (B) C4-2B and (C) MDA-PCa-2b prostate cancer cell lines loaded with fluorescent calcium reporter FURA-2 were treated with PAR2 AP (100 µM; filled in circle). Fluorescence was measured over three minutes using a BMG Polarstar fluorescent plate reader. Fluorescence at 510 nm was measured after sequential excitation at 340 nm (unbound FURA-2) and 380 nm (bound FURA-2). The ratio of unbound and bound FURA-2 is proportional to the intracellular concentrations of calcium ions (Ramsay et al., 2008a). D, 22Rv1 cells were preincubated with vehicle (DMSO; upper panel) or GB88 (50 µM; lower panel) for 15 minutes before sequential treatment with PAR2 AP (100 µM; filled circle) and PAR1 AP (100 µM; open circle). Experiments were performed in triplicate on three separate occasions.

154 These data confirm that GB88 selectively blocks PAR2 (Figure 6.4D, lower panel).

Therefore, GB88 is capable of preventing PAR2-Gαq mediated signal transduction in 22Rv1 osteoblastic prostate cancer cells.

6.2.3 Establishment of an animal model to examine the role of PAR2 in bone located prostate cancer tumours As described above, a role for PAR2 in prostate cancer bone metastases is suggested by previous studies highlighting the involvement of PAR2 in the bone microenvironment and receptor expression in bone lesions. To examine the role for PAR2 in prostate cancer bone metastasis, a mouse model was established in which osteoblastic 22Rv1 prostate cancer cells were injected into the tibia. The PAR2 antagonist GB88 (Suen et al., 2011) was used to block the function of this receptor in 22Rv1 cells in vivo.

As shown in Figure 6.1B, 22Rv1 cells (2 x 105 cells) were injected into the proximal aspect of right tibia of anaesthetised SCID mice using a 26 gauge needle. Intra-tibial injection of PBS alone into the proximal aspect of left tibia served as a contralateral limb control. As summarised in Table 6.2, intra-tibial injections were performed on 64 animals. Following surgery, animals were treated with an analgesic to aid recovery. However, three animals died during the recovery period (animals 42, 43 and 44; summarised in Table 6.2). To examine the role of PAR2 in the progression of prostate cancer bone lesions, a group of 16 animals were treated with the PAR2 antagonist GB88 (p.o. 5 mg/kg; Suen et al., 2011; Lohman et al., 2011). A group of 15 animals were treated with the prostate cancer chemotherapeutic docetaxel (i.p. 10 mg/kg; (Hung, 2007)). The cumulative effect of PAR2 blockade and anti-cancer therapy was examined by treating 15 mice with GB88 in combination with docetaxel. Another group of 15 animals were treated with vehicle alone (p.o. olive oil) to serve as an untreated control group (Table 6.2). As shown in Figure 6.5, animal treatments commenced 1 week after 22Rv1 cell injections and continued for a further 9 weeks. During this period, animals within the respective groups were treated either daily with vehicle and GB88 alone (Lohman et al., 2011) or weekly with docetaxel ((Hung, 2007); Figure 6.5). Animals remained ambulatory throughout the study with no significant weight variation observed between the treatment groups.

155

Table 6.2 Summary of animals used in in vivo prostate cancer bone metastasis model Animals injected with 22Rv1 prostate cancer cells were numbered by ear clips and allocated into two analysis groups: (i) histology and micro-CT imaging or (ii) histomorphometry. Within each of the two analysis groups, eight animals were allocated into 4 treatment groups (GB88; docetaxel; GB88 in combination with docetaxel; and vehicle (olive oil)). During the study three animals died. Animal numbers displayed in red indicate those analysed in this chapter.

GB88 GB88 & Docetaxel Vehicle Docetaxel 1 2 3 4 9 10 15 16 17 18 19 20 25 26 27 28

33 34 35 36

41 42‐died 43‐died 44‐died 49 50 51 52

Histology analysis 55 56 57 58

5 6 7 8

& 11 12 13 14

21 22 23 24 29 30 31 32 37 38 39 40 analysis 45 46 47 48 CT ‐ 53 54 63 64

Histomorphometry 59 60 61 62

micro

156 To examine the effect of PAR2 blockade and docetaxel therapy on prostate cancer bone lesions, animals were also split into two analysis groups: (1) histology and (2) micro-CT imaging and histomorphometry analysis (Table 6.2). Thus, eight animals were categorised into each treatment group within the two analysis groups. However, due to animal deaths following surgery, seven animals were allocated to the deocetaxel, GB88 in combination with docetaxel, and vehicle treatment groups for bone histomorphometry analysis (Table 6.2). For all animals within the histomorphometry analysis group, calcein injections were performed at 7 days and 2 days prior to sacrifice (Figure 6.5). Calcein is a fluorescent label incorporated into bone and administration at two time points permits the measurement of bone metabolic processes by fluorescence microscopy (Erben and Glosmann, 2012).

6.2.4 Qualitative radiographic analyses: intra-tibial prostatic bone lesions formed by 22Rv1 cells are reduced by GB88 treatment At sacrifice, all animals were assessed radiographically to examine the impact of PAR2 blockade and docetaxel treatment on intra-tibial prostate cancer tumours and surrounding bone structures. Representative images are shown in Figure 6.6. This radiographic assessment indicated bone loss (arrowhead) and concurrent formation of bony periosteal spicules (arrow) in animals carrying vehicle treated 22Rv1 tumours with what appears to be swelling of the surrounding tissue (Figure 6.6A). Most radiographic images of vehicle treated animals displayed similar lesions with 13 of the 16 animals exhibiting concurrent bone loss and spicule formation. In comparison, no evidence of bone loss, spicule formation or tissue swelling was apparent from radiographic images of the PBS injected contralateral limbs from vehicle treated animals (Figure 6.6A). The observed dual osteolytic bone loss and osteoblastic spiculation after 9 weeks of treatment is consistent with a previous report of an in vivo model of intra-tibial injected 22Rv1 prostate cancer cells (Henry et al., 2005). These researchers observed osteolytic disruption of the cortical bone resulting in progression of the tumour to the periosteal surface of the tibia which was associated with osteoblastic spicule formation (Henry et al., 2005). Radiographic analysis of limbs from all 15 docetaxel treated animals revealed no bone loss and a lack of spicule formation and tissue swelling (Figure 6.6B). Radiographic analysis of animals treated with GB88 alone generally revealed some bone loss (arrowhead) but undetectable spicule formation in 22Rv1 injected limbs (Figure 6.6C). Interestingly,

157

Figure 6.5 Prostate cancer in bone metastasis experiment timeline and treatment course Male SCID mice were injected with PAR2 responsive 22Rv1 prostate cancer cells (2 x 105 cells in 10 µl) into the proximal aspect of the right tibia. Intra-tibial injection of PBS only (10 µl) into left tibia served as a contralateral control. After a week recovery period, animals were treated daily with vehicle (p.o. olive oil) or GB88 (p.o. 5 mg/kg) or weekly with docetaxel (10 mg/kg i.p. injection) or respective combined daily and weekly treatments of GB88 and docetaxel for 10 weeks. Animals allocated for histomorphometry analysis were injected i.p. with calcein 7 days and 2 days prior to sacrifice. Animals were sacrificed by carbon dioxide asphyxiation. Intra-tibial injections were performed with the help of Dr Hui He. Daily and weekly animal treatments were performed on rotation with the help of Dr Hui He and Dr Yaowu He.

158 (Figure 6.6 legend, see over page)

159

Figure 6.6 The PAR2 antagonist GB88 reduces qualitative indices of 22Rv1 prostatic bone lesions Male SCID mice challenged by intra-tibial injection of PBS (control; left column) or 22Rv1 cells (cancer; central column) and treated with vehicle (A), docetaxel (B), GB88 (C) or docetaxel and GB88 (D) for 9 weeks were sacrificed and X-rayed. Enlarged images of 22Rv1 injected limbs are shown to highlight the region containing the lesion. White arrowhead highlights osteolytic bone loss; White arrow indicates periosteal spicule formation.

160 animals treated with GB88 also exhibited a reduction in the amount of swelling observed in the tissue surround the bone. Of the 15 animals treated with GB88 alone, 10 exhibited subtle osteolytic lesions as shown in Figure 6.6C (arrowhead) while two animals displayed overt lesions similar to vehicle treated animals as shown in Figure 6.6A. The radiographic image of a PBS injected limb from a GB88 treated animal in Figure 6.6C is representative of all 15 mice in this group and showed no detectable bone defects. Consistent with docetaxel treatment alone, representative radiographic analysis of 22Rv1 injected limbs from combined GB88 and docetaxel treated animals revealed no bone loss and a lack of spicule formation and tissue swelling in 14 of the 15 mice in this group (Figure 6.6D). However, one animal displayed a subtle bone osteolytic lesion with some bone loss but undetectable spicule formation. As shown in Figure 6.6D, no bone defects were detected in any PBS injected contralateral limbs from combined GB88 and docetaxel treated animals. It is important to note that differences in the visualisation of the fibula in these representative radiographic images results from varied orientation of the hind limb during imaging. Accordingly, these data indicate that GB88 blocks the role of PAR2 in bone spicule formation induced by 22Rv1 prostate cancer cells. It is also likely that PAR2 is required for the regional osteolysis induced by these cells.

6.2.5 Quantitative micro-CT analysis: osteoblastic and osteolytic bone lesions formed by intra-tibial injection of 22Rv1 prostate cancer cells are reduced by GB88 To more accurately examined the effect of PAR2 blockade on bone remodelling, micro-CT analysis was performed. After sacrifice, the intact tibia and femur from hindlimbs of animals within the histomorphometry analysis group were excised and fixed with paraformaldehyde for 24 hours. The surrounding tissue was left on the tibia and femur to prevent damaging the bone and periosteal spicules observed in radiographic analysis (see Figure 6.6A). Fixed bones were imaged by micro-CT analysis using a µCT 40 scanner generating 175 2-dimensional (2D) slices of the tibial proximal metaphysis that were reconstructed to form a 3D image. As shown in Figure 6.7A, this analysis showed that 22Rv1 tumours from vehicle treated mice generally exhibited concurrent osteolytic cortical pitting (arrowhead) and overt osteoblastic periosteal bone growth resembling spicules (arrow). Of the eight animals within this analysis group (see Table 6.2), six vehicle treated animals displayed

161

162

Figure 6.7 The PAR2 antagonist GB88 reduces the osteoblastic and osteolytic phenotype of 22Rv1 prostate cancer bone lesions Intact tibia and femur from animals treated with vehicle (A), docetaxel (B), GB88 (C) or docetaxel and GB88 (D) were surgically removed immediately after sacrifice. Bones were fixed in paraformaldehyde and assessed by 3D micro-CT imaging. Arrowhead highlights osteolytic bone loss; Arrow indicates periosteal spicule formation. Bones were removed with the help of Dr Yaowu He, Dr Hui He and Mrs Deborah Roche. Micro-CT imaging was performed by Dr Roland Steck at the Institute for Health and Biomedical Innovation, Queensland University of Technology.

163 cortical pitting and periosteal growth in micro-CT images. These data are consistent with the concurrent osteolytic bone loss and osteoblastic periosteal spicule formation observed in radiographic images (see Figure 6.6A). In comparison, micro-CT imaging of PBS injected contralateral tibia revealed a lack of bone defects in all vehicle treated animals (Figure 6.7A). Micro-CT analysis of 22Rv1 injected tibia from docetaxel treated animals revealed that these limbs, similar to contralateral PBS injected tibia, generally had undetectable bone loss or spicule formation (Figure 6.7B). However, in this group of eight animals, one animal displayed some minimal cortical bone pitting but no periosteal bone growth in the 22Rv1 injected tibia. As shown in Figure 6.7C, micro-CT analysis revealed that 22Rv1 injected tibia from animals treated with GB88 alone exhibited subtle bone defects with some osteolytic cortical bone pitting (arrowhead) and some periosteal bone growth (arrow). Of this group of eight animals, five displayed altered bone phenotypes as shown in Figure 6.7C (right), two displayed undetectable bone defects while one animal exhibited overt bone defects, similar to the micro-CT analysis of vehicle treated animals as shown in Figure 6.7A (right). Analysis of contralateral PBS injected tibia by micro- CT did not detect any bone defects in any GB88 treated mice (Figure 6.7C). As shown in Figure 6.7D, micro-CT analysis revealed that 22Rv1 injected tibia from animals treated with GB88 in combination with docetaxel had no detectable bone loss or periosteal growth in any of the mice in this group (compare with Figure 6.7B). Similarly, no bone defects were detected in any contralateral PBS injected tibia from mice treated with GB88 in combination with docetaxel (Figure 6.7D). Consistent with the radiographic analysis, these results indicate that PAR2 blockade reduces 22Rv1-mediated spicule formation and also suggests that this antagonist reduces bone osteolysis.

164 6.3 Discussion Prostate cancer tumours disseminating into bone cause osteolysis that permits initial tumour cell engraftment, and the interplay between tumour and bone cells ultimately produces an increase in bone mass (Kingsley et al., 2007; Roudier et al., 2008). In order to effectively treat this facet of prostate cancer, a greater understanding of the molecular events underpinning interactions between bone and cancer cells is required. In this chapter, the role of PAR2 in prostate cancer bone metastasis has been examined using a mouse model. To mimic prostate cancer growth in bone, 22Rv1 prostate cancer cells were injected into the proximal aspect of mouse tibia. A role for PAR2 was assessed by treating animals carrying these intra-tibial prostate tumours with the receptor antagonist GB88. This antagonist has previously been used to show that inhibition of PAR2 prevents receptor agonist-induced inflammation (Suen et al., 2011) and experimentally-induced arthritis and colitis (Lohman et al., 2011). As controls, animals bearing intra-tibial tumours were treated with vehicle (olive oil) or the prostate cancer chemotherapeutic docetaxel. The key finding of this chapter is that the PAR2 antagonist GB88 reduced the impact of 22Rv1 cancer cells on bone. Radiographic analysis and micro-CT imaging demonstrated that GB88 reduced prostate cancer cell-mediated osteoblastic spicule formation and reduced osteolytic bone loss. These observations indicate that PAR2 signalling is required for 22Rv1 tumour-induced osteoblastic spicule formation and also suggests that this receptor participates in tumour-mediated osteolysis.

It is important to note that the data presented in this chapter will form part of a larger study which will further investigate the role for PAR2 in prostate cancer in bone metastasis. However, the analyses to date highlight difficulties in determining the impact of tumours on bone based solely on qualitative radiographic and micro-CT analyses. In this chapter, while both radiographic analysis and micro-CT imaging demonstrated mixed osteoblastic and osteolytic bone lesions, these analyses were unable to quantitatively analyse the impact of treatments on (i) the tumour and (ii) morphology and organisation of bone. As such, the histology and histomorphometry analyses that will be performed in the future will complement the analysis presented in this chapter by examining the tumour and the osteolytic and osteoblastic bone phenotype induced by 22Rv1 tumours. The histology analysis will be used to examine the role for PAR2 in the tumour and surrounding soft tissue by determining

165 tumour mass and the morphology of tumour cells. Histomorphometry analysis will also be performed to quantitatively measure the role of PAR2 signalling in tumour- mediated changes in osseous tissue (cortical and trabecular bone) structure and density. This analysis will also examine the dynamic remodelling of bone by measuring osteoblast-mediated bone formation, via the rate of calcein deposition (administered at two time points), and osteoclast-mediated bone resporption.

In our model, intra-tibial injection of 22Rv1 cells resulted in a mixed osteoblastic and osteolytic phenotype in vehicle treated animals with an appreciable loss in proximal metaphysis bone that was accompanied by formation of periosteal spicules (Figures 6.6 and 6.7). The 22Rv1-mediated mixed osteolytic and osteoblastic spiculated bone response is consistent with previous reports of intra-tibial injection of these cells into rats and mice (Andersen et al., 2003; Henry et al., 2005). Importantly, periosteal spicule formation is also seen in a subset of patients in clinical settings (Bloom et al., 1987; Hove and Gyldensted, 1990; Legier and Tauber, 1968; Lehrer et al., 1970; Ragsdale et al., 1981; Wyche and de Santos, 1978). This phenotype is thought to arise from perforation of the cortical bone (osteolysis) leading to protrusion of the tumour from interior endosteal surfaces and growth of the tumour on the periosteal bone surface, eventually resulting in osteoblastic spicule formation (Henry et al., 2005). However, it is much more common for prostate cancer patients to present with osteoblastic lesions exhibiting thickening of the trabecular and endosteal bone surfaces rather than with periosteal spicules (Charhon et al., 1983). In addition, prostate cancer bone lesions also present with a low osteolytic component (Keller and Brown, 2004; Roudier et al., 2008) that is not well represented by the 22Rv1 cell tumours observed in this study. In human disease osteolysis permits initial formation of bone lesions by creating a space within the bone matrix for prostate cancer cells to occupy (Schneider et al., 2005). In this regard, the qualitative and 3D analysis performed in this chapter indicates that the 22Rv1 tumour-mediated impact on bone does not completely mimic human disease.

Nonetheless, 22Rv1 cells were chosen for use in the current animal model for two reasons. First, these cells exhibit a mixed osteoblastic and osteolytic phenotype in bone (Andersen et al., 2003; Henry et al., 2005) that at least partially mimics human disease. Second, these cells express functional PAR2 as shown by ERK1/2

166 phosphorylation and calcium mobilisation assays. Although mixed osteoblastic/osteolytic C4-2B and osteoblastic MDA-PCa-2b cells express PAR2 mRNA and protein, these cell lines were non-responsive to PAR2 agonism and so were not suitable for this study. The inability for these cell lines to signal via the receptor at the selected time point (15 min) is intriguing. Possible explanations include that these cell lines may not express the G-protein effectors required to stimulate ERK phosphorylation or the efflux of calcium ions into the cytosol. It is also possible that mutations within PAR2, while not impacting trafficking (e.g. as observed by PAR2 cell surface expression in Figure 6.2), may prevent conformational changes within the receptor to induce signal transduction. Furthermore, although unlikely, it is possible that these lines respond more slowly than 22Rv1 cells and that signalling via ERK may have been apparent at times points after 15 min. Ideally, assays to assess PAR2 signalling at times points up to 24 hours should be performed. However, it is also possible that activation of PAR2 in these cell lines induces signalling pathways other than those examined in this chapter. Therefore further analysis is required to examine whether molecular mechanisms exist to prevent downstream signalling and whether PAR2 preferentially couples to other signalling pathways in C4-2B and MDA-PCa-2b cells.

It was shown in this chapter that antagonism of PAR2 with GB88 results in reduced periosteal spicule formation (Figure 6.6 and 6.7); this suggests a role for this receptor as a positive regulator of 22Rv1 cell-induced bone formation. These observations indicate that targeting PAR2 may provide a method to block the pathological bone formation that is characteristic of prostate cancer. Similarly, antagonism of PAR2 also appeared to reduce osteolysis, potentially indicating that this receptor is a positive regulator of 22Rv1 cell-induced bone loss. This suggests that targeting PAR2 may provide a method to block pathological bone loss that is seen in disorders such as osteoporosis (Boyle et al., 2003) and that accompanies breast cancer metastasis to bone (Guise et al., 2006; Roodman, 2004). Another interesting observation from this chapter is that antagonism of PAR2 potentially reduces swelling of tissue surrounding the bone. It is not known whether this swelling was due to an inflammatory response or the growth of the tumour outside of the bone. It is worth noting that SCID mice lack functional T cells and B cells and are only able to mount a complete inflammatory response when reconstituted with functional

167 blood mononuclear cells (containing T cells, B cells and macrophages)(Herz et al., 1998; Liu et al., 2000). Further histology analysis is required to determine whether this tissue, observed in radiographic images (Figure 6.6), results from an inflammatory response or is indicative of a tumour mass.

An interesting aspect of this study is that it has not been possible to determine whether it was PAR2 functioning in tumour cells or bone cells, or indeed each cell type, that was responsible for the observed phenotype. The following paragraphs speculate on the potential roles for this receptor in each of these cell types. First to be explored are reports supporting a role for tumour cell expressed PAR2 in the growth of prostate tumours in bone. Recent reports have shown that blockade of PAR2 signalling, initiated by TF-FVIIa, significantly reduced growth of subcutaneous MDA-MB-231mfp breast cancer cell mouse xenografts (Versteeg et al., 2008). Also, in prostate cancer derived cell lines, activation of PAR2 by the proteases KLK2 and KLK4 induced proliferation of DU145 cells (Mize et al., 2008). In addition, PAR2 activation is linked with LNCaP cell migration through RhoA-dependent cytoskeletal reorganisation of LNCaP cells (Greenberg et al., 2003). Furthermore, PAR2 signalling in LNCaP, PC-3 and DU145 cells also amplifies production of collagenase matrix metalloproteinases-2 and -9 (Wilson et al., 2004), proteases involved in matrix remodelling and linked with prostate cancer progression (Brehmer et al., 2003; Zhang et al., 2004). Hence, initiation of PAR2 signalling in tumour cells may contribute to the progression of prostate cancer tumours in bone. Further studies are required for a better understanding of how tumour cell expressed PAR2 impacts on prostate cancer bone lesions. These could include the use of 22Rv1 cells silenced for PAR2 in the intratibial injection model used in this chapter. Differences in the level of bone formation and bone lysis induced by PAR2 silenced and control cells could be used to infer the function of tumour cell expressed PAR2 in prostate cancer cell- induced bone changes.

PAR2 may also contribute to prostate cancer progression in bone by its function in osteoclasts and osteoblasts. In the bone microenvironment, osteoclasts are bone resorbing cells that differentiate from haemopoietic stem cells (HSCs) (Boyle et al., 2003) whereas osteoblasts differentiate from mesenchymal stem cells (MSCs) and form bone by synthesising and mineralising ECM (Kassem et al., 2008). Importantly,

168 PAR2 is expressed in each of these bone cells and is reported to regulate the function of each lineage (Abraham et al., 2000; Gregory et al., 2010; Georgy et al., 2011; Mackie et al., 2008; Smith et al., 2004). For example, in vitro activation of PAR2 in bone marrow cultures, which have abundant haemopoietic precursor cells, reduces parathyroid hormone, 1,25-dihydroxyvitamine D and IL-11 stimulated osteoclastogenesis (Smith et al., 2004). Also, PAR2 activation in osteoblasts stimulates expression of ECM components but does not enhance osteoblast proliferation or differentiation (Georgy et al., 2010). Accordingly, it is possible that inhibition of PAR2 signalling may lead to (i) reduced tumour-induced osteolysis by limiting osteoclastogenesis and also (ii) reducing tumour-induced spicule formation by limiting ECM expression by osteoblasts. To further examine this possibility PAR2 KO mice could be used. Intra-tibial injection of osteoblastic C4-2B or MDA- PCa-2b cells, which do not express functional PAR2, into these mice would be able to examine the role of bone expressed PAR2 on prostate cancer bone lesions.

As PAR2 functions in both tumour and bone cells, it is likely that PAR2-mediated reciprocal interactions occur between these cells. In humans, osteoblastic prostate tumour bone metastases are thought to arise from tumour-mediated subjugation of bone homeostasis leading to inhibition of osteoclast activity or osteoclastogenesis (osteoclast differentiation) allowing unimpeded osteoblast activity (Logothetis and Lin, 2005). In this regard, activation of PAR2 is known to induce secretion of osteoclastic stimulators, such as IL-6, by a variety of cell types including neutrophils, monocytes and lymphocytes which are present in the bone microenvironment (Colognato et al., 2003; Howells et al., 1997; Mari et al., 1996). However, whether PAR2 signalling stimulates secretion of osteoclastic or osteoblastic factors from prostate tumours cells remains unknown. In addition, whether activation of PAR2 stimulates the reciprocal secretion of tumour promoting factors from bone cells also remains to be investigated. To address these issues, in vitro analysis of prostate tumour cells co-incubated with bone cells derived from wildtype or PAR2 KO mice may elucidate a role for this receptor in bone cell function. Similarly, co-incubation of bone cells with prostate tumours silenced for PAR2 expression may demonstrate a reciprocal function for this receptor in the tumour.

169 In these studies the cancer chemotherapeutic docetaxel was used as a positive control for reduction in 22Rv1 cell tumour formation and alterations to bone structure. Consistent with this use, docetaxel treatment prevented 22Rv1 tumour-induced osteoblastic spicule formation and osteolytic bone loss. Mice carrying tumours that were treated with a combination of PAR2 antagonist GB88 and docetaxel also exhibited reduced osteolysis and periosteal spicule formation. This reversal in tumour-induced bone phenotype is consistent with the effects of docetaxel in other reports. For example, docetaxel treatment significantly reduced subcutaneous 22Rv1 mouse xenograft tumour weight (Festuccia et al., 2009) and PSA levels while inducing mitotic arrest of 22Rv1 renal capsule xenografts (Gan et al., 2009). Also, mice bearing intra-tibial LNCaP-C4-2B tumours treated with docetaxel combined with dasatinib, a Src family kinase inhibitor, exhibited significantly reduced serum PSA levels and prevented tumour-induced osteolysis (Koreckij et al., 2009). In addition, docetaxel combined with bisphosphonate therapy, which prevents osteoclastic bone resorption, also significantly reduced tumour burden in an animal model of breast cancer bone metastases with no tumour cells detected in bone (van Beek et al., 2009). Altogether, these studies, consistent with the data presented in this chapter, highlight that use of docetaxel has a marked effect on tumours in the bone microenvironment.

In summary, this in vivo study has shown for the first time, through the use of the recently described antagonist GB88 (Suen et al., 2011), that PAR2 participates in spicule bone formation and osteolysis induced by 22Rv1 prostate cancer cells. Although the in vivo model used in this chapter does not completely mimic the human disease, these data suggest that PAR2 is a positive regulator of both bone lysis (osteoclastogenesis) and bone formation. To address the potential role for PAR2 in prostate cancer bone metastasis more closely, further histology and histomorphometry analysis will be performed on the samples that have already been collected during this study. Additional in vivo models, that have been proposed herein, would complement these analyses and further examine the function of PAR2 in this disease. Collectively, these studies may demonstrate the significance of PAR2 signalling in prostate cancer bone metastasis and also address whether this receptor is a rational target for treatment of this disease.

170

Chapter 7:

Conclusions and Future Directions

171

172 7.1 Summary of findings This PhD program of study has examined mechanisms which regulate PAR2 at the cellular level and the role of this receptor in prostate cancer bone metastases. The major findings include: (1) characterisation of antibodies capable of detecting the endogenously expressed receptor; (2) delineation of a role for palmitoylation in receptor signalling and trafficking; (3) characterisation of residues involved in ligand docking; and (4) investigation of the role of PAR2 in the changes in bone that accompany the growth of skeletal prostate cancer tumours. The following sections highlight new research directions that stem from these findings.

7.2 Palmitoylation of PAR2 A significant finding from this study is that palmitoylation of PAR2 on C361 regulates downstream signal transduction by modulating receptor trafficking. It will be interesting to examine whether dysregulation of this modification contributes to aberrant PAR2 signalling in pathological states. There are already a number of examples where dysfunctional palmitoylation has been shown to contribute to disease. These have arisen because of defects in the enzymes PATs that palmitoylate key regulatory proteins. For example, a single nucleotide polymorphism (SNP) in the gene encoding the PAT DHHC8, which ablates its enzyme activity, is linked with increased risk of schizophrenia (Chen et al., 2004; Mukai et al., 2008). Similarly, reduced palmitoylation of β-secretase (BACE) alters amyloid precursor protein processing and may result in the formation of the protein aggregates that are characteristic of Alzheimer’s disease (Benjannet et al., 2001; Sidera et al., 2005). Also, reduced palmitoylation of the huntingtin protein by the PAT DHHC17 prevents trafficking of huntingtin causing inclusion body formation and neurotoxicity in a mouse model of Huntington’s disease (Singaraja et al., 2011; Yanai et al., 2006). Currently, there is no information on the PATs that palmitoylate PAR2 in cells and tissues. Characterisation of these enzymes will be an important area for future studies on PAR2.

To further evaluate the importance of PAR2 palmitoylation, several in vitro and in vivo studies may be used. For example, comparing the level of PAR2 palmitoylation in normal and diseased cells in vitro may be used to support the proposal that dysregulated receptor modification contributes to disease. Also, the availability of

173 cell lines that express palmitoylation-deficient PAR2 may also prove useful. For example, growth of these cells orthotopically or as xenografts in mice will be useful for evaluating the importance of PAR2 palmitoylation in cancer. In addition, the importance of PAR2 palmitoylation could be evaluated in vivo by a transgenic knock-in mouse expressing PAR2 with a mutation at C361. In support of this proposal, knock-in mice expressing palmitoylation-deficient rhodopsin have light- induced defects of the retina suggesting an important and previously unknown role for this GPCR in vision (Maeda et al., 2010).

7.3 Anti-PAR2 antibodies Specific anti-PAR2 antibodies are required to characterise the function of this protein. A concerning finding from this study was that of four anti-PAR2 antibodies examined, only one antibody (N19) was able to detect the endogenous receptor by Western blot analysis and flow cytometry. It is reasonable to propose that additional specific antibodies will be required to complete the characterisation of PAR2. As each of the antibodies examined in this study were raised against peptides it is likely that different approaches will need to be employed to generate these important reagents. Of relevance, recently a novel method of antigen design, whereby extracellular sequences are combined into a single antigen, has led to the generation of specific antibodies for the β2-adrenergic, glucagon-like peptide 1 and neuropeptide Y receptor 5 GPCRs (Larsson et al., 2011). Use of this technique may lead to the generation of specific anti-PAR2 antibodies.

Generating antibodies that target specific regions of PAR2, such as ECL2 or the TM ligand binding pocket, may also have therapeutic potential. In support of this proposal, the anti-PAR2 antibody SAM11 blocks receptor activation and protects against experimental osteoarthritis in murine models (Ferrell et al., 2010) and carrageenan/kaolin-induced joint inflammation in mice (Kelso et al., 2006). Also, blocking PAR2 proteolysis with SAM11 prevents protease-mediated allergic sensitisation and airway inflammation (Arizmendi et al., 2011). It is possible that other PAR2 function blocking antibodies may be useful in inhibiting cancer. Importantly, humanised monoclonal antibodies targeting other cell surface receptors are currently in clinical evaluation as therapeutic agents for cancer treatment. Farletuzumab, an antibody for the folate receptor alpha, has shown efficacy in a

174 clinical trial of ovarian cancer patients (Kalli, 2007; Konner et al., 2010; Smith-Jones et al., 2008). In addition, blockade of human epidermal growth factor 2 (HER2) dimerization with Pertuzumab inhibits in vivo prostate cancer growth (Adams et al., 2006; Agus et al., 2002) and prolongs survival of metastatic breast cancer patients (Baselga et al., 2012). In the same way, future development of potent humanised anti-PAR2 antibodies that activate or inhibit receptor function may have therapeutic potential.

7.4 Understanding how PAR2 is activated in prostate cancer bone metastasis Another important finding from this study was the potential role for PAR2 in prostate cancer bone metastasis. An important area for future research will be the identification of the proteases in vivo that promote disease via PAR2 activation. Of relevance to the tumour and bone microenvironment, candidate proteases include KLK4 (Ramsay et al., 2008a), trypsin IV, matriptase/MT-SP1 and acrosin (Georgy et al., 2011), HAT (Miki et al., 2003), TMPRSS2 (Wilson et al., 2005). Also, hepsin expressed by these cells may indirectly activate PAR2 via a proteolytic cascade involving activation of matriptase/MT-SP1 (Georgy et al., 2011).

An issue related to the identification of the proteases responsible for PAR2 activation in vivo, is the identification of the inhibitors that block aberrant activity of these proteases in normal settings. It is already known that the level of protease inhibiting metal ions reduce during prostate cancer progression (Zaichick et al., 1997). This may be directly relevant to any role for PAR2 in prostate cancer as the PAR2 activators KLK4 and KLK14 are inactive in concentrations of Zn2+ above 100 µM (Borgono et al., 2007; Debela et al., 2006) whereas concentrations of this metal ion in prostate cancer drop to 70-80% of those present in normal glands (Costello et al., 2004; Zaichick et al., 1997). Interestingly, KLK4, KLK6 and KLK14 are also inhibited by the serpin anti-thrombin III (Borgono et al., 2007; Magklara et al., 2003; Obiezu et al., 2006) which is down-regulated during prostate cancer progression (Cao et al., 2002; Frenette et al., 1997). Thus, it is possible that trypsin-like proteases that are active in the tumour and bone microenvironment may be able to promote cancer progression via activation of PAR2 because of a reduction in the concentration of proteinaceous and non-proteinaceous inhibitors.

175 Inhibition of these enzymes may prove to be a viable option for modulating PAR2 signalling in disease. In fact, several protease inhibitors have already shown promise in preclinical models for treatment of disease. For example, synthetic serine protease inhibitors of β-mast cell tryptase (Cairns, 2005) and neutrophil elastase (Wright et al., 2002) attenuate the acute response to asthma, emphysema and α1-anti-trypsin deficiency (Venkatasamy and Spina, 2007). However, it is not clear whether inhibition of these proteases prevents activation of cell surface receptors in these diseases. Interestingly, of relevance to PAR2, broad range inhibition of serine proteases with aprotinin prevented the autocrine stimulation of MDA-MB-231 breast cancer cell migration by an unidentified trypsin-like serine protease (Ge et al., 2004). Thus, it is possible that a protease inhibitor may be used to target PAR2 signalling in disease. In fact, as more than one PAR2 activator is likely to be present in the tumour microenvironment, it may also be necessary to provide a cocktail of selective protease inhibitors to prevent PAR2 activation.

7.5 Development of other PAR2 targeting pharmacological agents As an alternative to protease inhibition, PAR2 signalling in certain diseases may be modulated by pharmacological agents that directly agonise or antagonise this receptor. Although selective reagents for PAR2 exist, other improved agonists and antagonists are required with enhanced drug-like properties to more potently target this receptor. However, to generate these improved reagents, a greater understanding of the amino acids involved in PAR2 ligand docking is required. To this end, a significant finding from this study is the identification of three residues in TM III, VI and VII that are required for docking of two soluble ligands. Intriguingly, these mutagenesis studies were much less definitive in identifying residues required for responsiveness to trypsin. It is not clear why the residues that mediate responsiveness to soluble ligands were not also necessary for docking of the TL exposed by trypsin. However, as mentioned earlier, it is possible that either (i) separate docking sites exist for the TL and soluble ligands and/or (ii) residues required for TL docking were not identified due to limitations within the rhodopsin-based homology model. To clarify this issue, crystal structures of inactive (basal state) PAR2 and both the proteolytically activated and agonist bound receptor would be beneficial. Although challenging to generate, these structures would further define residues critical for ligand docking and aide rational design of more potent and efficacious PAR2

176 agonists and antagonists. Similar studies may also be performed on the other PAR family members to generate a suite of reagents to target these receptors.

7.6 Possible dual activation of multiple PAR family members It is intriguing that more than one PAR will signal in a particular disease setting because of the presence of proteases that activate more than one family member (Figure 7.1). For example both PAR1 and PAR2 are activated by KLK4 (Gratio et al., 2010; Ramsay et al., 2008a), thrombin (Shi et al., 2004), TF-FVIIa (Albrektsen et al., 2007) other KLKs (Oikonomopoulou et al., 2006). In normal settings and potentially in diseases such as cancer, a particular protease has the potential to initiate a complex array of intracellular signalling pathways. In fact it is already known that dual activation of PAR1 and PAR2 enhances melanoma cell migration and prostate cancer cell chemokinesis toward fibronectin (Shi et al., 2004). Hence, as shown in Figure 7.1, in certain disease settings it may be advantageous to selectively agonise or antagonise only one of the PARs.

Two therapeutic avenues exist to modulate PAR family signalling in disease: (i) inhibition of a common protease that initiates PAR signalling; and (ii) modulation of PARs with potent and selective pharmacological agents. As mentioned in section 7.4, protease inhibition is potentially an attractive therapeutic option to modulate of PAR signalling that initiated by one or a small number of proteases. Proteases that may be worthwhile investigating in prostate cancer progression include KLK2 (Mize et al., 2008) and KLK4 (Gratio et al., 2010; Ramsay et al., 2008a) that activate PAR1 and PAR2, and trypsin (Nystedt et al., 1995; Xu et al., 1998) and KLK14 (Oikonomopoulou et al., 2006) that activate PAR2 and PAR4. Importantly, each of these receptors is elevated in primary prostate tumours (Black et al., 2007) while PAR1 and PAR2 are also present in prostate cancer bone lesions (Chay et al., 2002; Ramsay et al., 2008a; Zhang et al., 2009b). Therefore, as mentioned in section 7.5, as well as targeting proteolytic activation of these receptors, agonism or antagonism of PAR-mediated signal transduction pathways with pharmacological agents may also be used to treat disease. Building on the mouse model used in this project, further insight on the roles of PAR family members in prostate cancer bone metastasis could be obtained by treating these animals with the respective PAR1, PAR2 and PAR4 antagonists SCH 530348 (Chackalamannil et al., 2008), GB88

177

Figure 7.1 Possible activation and modulation of PAR2 and other cell surface receptors Convergent activation of PAR1, PAR2, CDCP1 and c-Met (by cleaving pro-HGF) by an in vivo protease may induce signal transduction to promote disease. Possible mechanisms to modulate downstream signalling induced by activation of these cell surface receptors include (i) developing selective and potent inhibitors of protease agonists and (ii) direct antibody based or pharmacologically based targeting of the receptors. Mechanisms to modulate these receptors include the generation of humanised antibodies targeting the PARs as well as improved small molecule agonists or antagonists. Also, either antibody based or pharmacologically based inhibitors targeting CDCP1 or the HGF/c-Met system could be employed block downstream signalling. In addition, CDCP1 and c-MET-mediated signal transduction could be inhibited by preventing phosphorylation of these receptors with tyrosine kinase inhibitors (TKI).

178 (Suen et al., 2011) and P4pal-10 (Houle et al., 2005; McDougall et al., 2009). In addition, challenge of immune deficient mice lacking PAR1, PAR2 and/or PAR4 with prostate cancer cell lines may also elucidate a role for receptor cooperativity in progression of this disease.

7.7 Possible dual activation of PAR2 and other cell surface receptor systems Analogous to the dual PAR activation discussed in section 7.5, it is already known that PAR2 is activated by the same proteases that activate other non-PAR cell surface receptors (Figure 7.1). Thus, it is likely that dual activation of these receptors will contribute to both physiological and disease settings. For example, the PAR2 activators matriptase (Bhatt et al., 2005; He et al., 2010) and trypsin (Brown et al., 2004) also initiate intracellular signalling via cleavage of the TM glycoprotein CUB domain-containing protein 1 (CDCP1). Proteolysis of CDCP1 induces signal transduction via Src and PKCδ (Brown et al., 2005; He et al., 2010) and promotes cell survival in cancer (Uekita et al., 2007; Wong et al., 2009). Similarly, the PAR2 activator matriptase is also a potent activator of pro-hepatocyte growth factor (pro- HGF) which initiates Akt and mTOR signalling via its receptor, c-Met. Importantly, the HGF/c-Met receptor system is a well characterised promoter of tumorigenesis and metastasis (Betsunoh et al., 2007; Conway et al., 2007; Lee et al., 2000; Parr et al., 2004; Szabo et al., 2011; Uhland et al., 2009). Hence, as shown in Figure 7.1, one protease may initiate a myriad of intracellular signal transduction pathways via cleavage at PAR2 and these other receptors. Although blockade of matriptase proteolysis may prevent activation of each receptor, it may be advantageous, in certain diseases, to modulate specific signalling pathways by targeting the individual receptors (Figure 7.1). These agents could include function blocking anti-CDCP1 antibodies (Casar et al., 2011), anti-Src tyrosine kinase inhibitors (TKI) such as dasatinib (Lombardo et al., 2004) to prevent signalling downstream of CDCP1 cleavage together with c-Met receptor-specific inhibitors (such as PHA- 665752)(Christensen et al., 2003) and the PAR2 antagonist GB88 (Suen et al., 2011) to prevent signalling downstream of this receptor. Further examination of in vivo models, such as tumour bearing mice, treated with these pharmacological agents is required to investigate this possibility more closely.

179 7.8 Final conclusion In summary, work presented in this study has defined mechanisms leading to activation of PAR2 and regulation of PAR2 signalling and trafficking. Also, a potential role for this receptor in prostate cancer bone metastasis has been identified. Future investigation is required into PAR2 activation and trafficking and the approaches necessary to modulate this interesting receptor. Importantly, these studies may yield data that can be used to develop drugs that can be used in the clinic to treat disease.

180 Chapter 8: BIBLIOGRAPHY

Abraham, L.A., Chinni, C., Jenkins, A.L., Lourbakos, A., Ally, N., Pike, R.N., Mackie, E.J. (2000). Expression of protease-activated receptor-2 by osteoblasts. Bone, 26(1):7-14.

Abraham, L.A., Jenkins, A.L., Stone, S.R., Mackie, E.J. (1998). Expression of the thrombin receptor in developing bone and associated tissues. J Bone Miner Res, 13(5):818-827.

Adam, E., Hansen, K.K., Astudillo Fernandez, O., Coulon, L., Bex, F., Duhant, X., Jaumotte, E., Hollenberg, M.D., Jacquet, A. (2006). The house dust mite allergen Der p 1, unlike Der p 3, stimulates the expression of interleukin-8 in human airway epithelial cells via a proteinase-activated receptor-2- independent mechanism. J Biol Chem, 281(11):6910-6923.

Adams, C.W., Allison, D.E., Flagella, K., Presta, L., Clarke, J., Dybdal, N., McKeever, K., Sliwkowski, M.X. (2006). Humanization of a recombinant monoclonal antibody to produce a therapeutic HER dimerization inhibitor, pertuzumab. Cancer Immunol Immunother, 55(6):717-727.

Adams, M.N., Ramachandran, R., Yau, M.K., Suen, J.Y., Fairlie, D.P., Hollenberg, M.D., Hooper, J.D. (2011). Structure, function and pathophysiology of protease activated receptors. Pharmacol Ther, 130(3):248-282.

Agarwal, A., Covic, L., Sevigny, L.M., Kaneider, N.C., Lazarides, K., Azabdaftari, G., Sharifi, S., Kuliopulos, A. (2008). Targeting a metalloprotease-PAR1 signaling system with cell-penetrating pepducins inhibits angiogenesis, ascites, and progression of ovarian cancer. Mol Cancer Ther, 7(9):2746-2757.

Agus, D.B., Akita, R.W., Fox, W.D., Lewis, G.D., Higgins, B., Pisacane, P.I., Lofgren, J.A., Tindell, C., Evans, D.P., Maiese, K., Scher, H.I., Sliwkowski, M.X. (2002). Targeting ligand-activated ErbB2 signaling inhibits breast and prostate tumor growth. Cancer Cell, 2(2):127-137.

Ahamed, J., Ruf, W. (2004). Protease-activated receptor 2-dependent phosphorylation of the tissue factor cytoplasmic domain. J Biol Chem, 279(22):23038-23044.

Ahmad, I., Sansom, O.J., Leung, H.Y. (2008). Advances in mouse models of prostate cancer. Expert Rev Mol Med, 10(e16.

Ahn, K.H., Nishiyama, A., Mierke, D.F., Kendall, D.A. (2010). Hydrophobic residues in helix 8 of cannabinoid receptor 1 are critical for structural and functional properties. Biochemistry, 49(3):502-511.

181 Al-Ani, B., Saifeddine, M., Kawabata, A., Hollenberg, M.D. (1999). Proteinase activated receptor 2: Role of extracellular loop 2 for ligand-mediated activation. Br J Pharmacol, 128(5):1105-1113.

Al-Ani, B., Saifeddine, M., Wijesuriya, S.J., Hollenberg, M.D. (2002a). Modified proteinase-activated receptor-1 and -2 derived peptides inhibit proteinase- activated receptor-2 activation by trypsin. J Pharmacol Exp Ther, 300(2):702-708.

Al-Ani, B., Wijesuriya, S.J., Hollenberg, M.D. (2002b). Proteinase-activated receptor 2: differential activation of the receptor by tethered ligand and soluble peptide analogs. J Pharmacol Exp Ther, 302(3):1046-1054.

Albrektsen, T., Sorensen, B.B., Hjorto, G.M., Fleckner, J., Rao, L.V., Petersen, L.C. (2007). Transcriptional program induced by factor VIIa-tissue factor, PAR1 and PAR2 in MDA-MB-231 cells. J Thromb Haemost, 5(8):1588-1597.

Amiable, N., Tat, S.K., Lajeunesse, D., Duval, N., Pelletier, J.P., Martel-Pelletier, J., Boileau, C. (2009). Proteinase-activated receptor (PAR)-2 activation impacts bone resorptive properties of human osteoarthritic subchondral bone osteoblasts. Bone, 44(6):1143-1150.

Anavi-Goffer, S., Fleischer, D., Hurst, D.P., Lynch, D.L., Barnett-Norris, J., Shi, S., Lewis, D.L., Mukhopadhyay, S., Howlett, A.C., Reggio, P.H., Abood, M.E. (2007). Helix 8 Leu in the CB1 cannabinoid receptor contributes to selective signal transduction mechanisms. J Biol Chem, 282(34):25100-25113.

Andersen, C., Bagi, C.M., Adams, S.W. (2003). Intra-tibial injection of human prostate cancer cell line CWR22 elicits osteoblastic response in immunodeficient rats. J Musculoskelet Neuronal Interact, 3(2):148-155.

Andrade-Gordon, P., Maryanoff, B.E., Derian, C.K., Zhang, H.C., Addo, M.F., Darrow, A.L., Eckardt, A.J., Hoekstra, W.J., McComsey, D.F., Oksenberg, D., Reynolds, E.E., Santulli, R.J., Scarborough, R.M., Smith, C.E., White, K.B. (1999). Design, synthesis, and biological characterization of a peptide- mimetic antagonist for a tethered-ligand receptor. Proc Natl Acad Sci USA, 96(22):12257-12262.

Andresen, B.T. (2011). A pharmacological primer of biased agonism. Endocr Metab Immune Disord Drug Targets, 11(2):92-98.

Angelucci, A., Gravina, G.L., Rucci, N., Festuccia, C., Muzi, P., Vicentini, C., Teti, A., Bologna, M. (2004). Evaluation of metastatic potential in prostate carcinoma: an in vivo model. International journal of oncology, 25(6):1713- 1720.

Arizmendi, N.G., Abel, M., Mihara, K., Davidson, C., Polley, D., Nadeem, A., El Mays, T., Gilmore, B.F., Walker, B., Gordon, J.R., Hollenberg, M.D., Vliagoftis, H. (2011). Mucosal allergic sensitization to cockroach allergens is

182 dependent on proteinase activity and proteinase-activated receptor-2 activation. J Immunol, 186(5):3164-3172.

Arrington, S.A., Schoonmaker, J.E., Damron, T.A., Mann, K.A., Allen, M.J. (2006). Temporal changes in bone mass and mechanical properties in a murine model of tumor osteolysis. Bone, 38(3):359-367.

Baneres, J.L., Mesnier, D., Martin, A., Joubert, L., Dumuis, A., Bockaert, J. (2005). Molecular characterization of a purified 5-HT4 receptor: a structural basis for drug efficacy. J Biol Chem, 280(21):20253-20260.

Barroso, S., Richard, F., Nicolas-Etheve, D., Kitabgi, P., Labbe-Jullie, C. (2002). Constitutive activation of the neurotensin receptor 1 by mutation of Phe(358) in Helix seven. Br J Pharmacol, 135(4):997-1002.

Barry, G.D., Suen, J.Y., Le, G.T., Cotterell, A., Reid, R.C., Fairlie, D.P. (2010). Novel agonists and antagonists for human protease activated receptor 2. J Med Chem, 53(20):7428-7440.

Barry, G.D., Suen, J.Y., Low, H.B., Pfeiffer, B., Flanagan, B., Halili, M., Le, G.T., Fairlie, D.P. (2007). A refined agonist pharmacophore for protease activated receptor 2. Bioorg Med Chem Lett, 17(20):5552-5557.

Baselga, J., Cortes, J., Kim, S.B., Im, S.A., Hegg, R., Im, Y.H., Roman, L., Pedrini, J.L., Pienkowski, T., Knott, A., Clark, E., Benyunes, M.C., Ross, G., Swain, S.M. (2012). Pertuzumab plus trastuzumab plus docetaxel for metastatic breast cancer. N Engl J Med, 366(2):109-119.

Baylink, D.J., Finkelman, R.D., Mohan, S. (1993). Growth factors to stimulate bone formation. J Bone Miner Res, 8 Suppl 2(S565-572.

Belham, C.M., Tate, R.J., Scott, P.H., Pemberton, A.D., Miller, H.R., Wadsworth, R.M., Gould, G.W., Plevin, R. (1996). Trypsin stimulates proteinase- activated receptor-2-dependent and -independent activation of mitogen- activated protein kinases. Biochem J, 320 ( Pt 3)(939-946.

Belting, M., Dorrell, M.I., Sandgren, S., Aguilar, E., Ahamed, J., Dorfleutner, A., Carmeliet, P., Mueller, B.M., Friedlander, M., Ruf, W. (2004). Regulation of angiogenesis by tissue factor cytoplasmic domain signaling. Nat Med, 10(5):502-509.

Bender, A., Mussa, H.Y., Glen, R.C., Reiling, S. (2004). Similarity searching of chemical databases using atom environment descriptors (MOLPRINT 2D): evaluation of performance. J Chem Inf Comput Sci, 44(5):1708-1718.

Benjannet, S., Elagoz, A., Wickham, L., Mamarbachi, M., Munzer, J.S., Basak, A., Lazure, C., Cromlish, J.A., Sisodia, S., Checler, F., Chretien, M., Seidah, N.G. (2001). Post-translational processing of beta-secretase (beta-amyloid- converting enzyme) and its ectodomain shedding. The pro- and

183 transmembrane/cytosolic domains affect its cellular activity and amyloid-beta production. J Biol Chem, 276(14):10879-10887.

Benka, M.L., Lee, M., Wang, G.R., Buckman, S., Burlacu, A., Cole, L., DePina, A., Dias, P., Granger, A., Grant, B., et al. (1995). The thrombin receptor in human platelets is coupled to a GTP binding protein of the G alpha q family. FEBS Lett, 363(1-2):49-52.

Betsunoh, H., Mukai, S., Akiyama, Y., Fukushima, T., Minamiguchi, N., Hasui, Y., Osada, Y., Kataoka, H. (2007). Clinical relevance of hepsin and hepatocyte growth factor activator inhibitor type 2 expression in renal cell carcinoma. Cancer Sci, 98(4):491-498.

Bhatt, A.S., Erdjument-Bromage, H., Tempst, P., Craik, C.S., Moasser, M.M. (2005). Adhesion signaling by a novel mitotic substrate of src kinases. Oncogene, 24(34):5333-5343.

Black, P.C., Mize, G.J., Karlin, P., Greenberg, D.L., Hawley, S.J., True, L.D., Vessella, R.L., Takayama, T.K. (2007). Overexpression of protease-activated receptors-1,-2, and-4 (PAR-1, -2, and -4) in prostate cancer. Prostate, 67(7):743-756.

Blakeney, J.S., Reid, R.C., Le, G.T., Fairlie, D.P. (2007). Nonpeptidic ligands for peptide-activated G protein-coupled receptors. Chemical reviews, 107(7):2960-3041.

Blanpain, C., Wittamer, V., Vanderwinden, J.M., Boom, A., Renneboog, B., Lee, B., Le Poul, E., El Asmar, L., Govaerts, C., Vassart, G., Doms, R.W., Parmentier, M. (2001). Palmitoylation of CCR5 is critical for receptor trafficking and efficient activation of intracellular signaling pathways. J Biol Chem, 276(26):23795-23804.

Bloom, R.A., Libson, E., Husband, J.E., Stoker, D.J. (1987). The periosteal sunburst reaction to bone metastases. A literature review and report of 20 additional cases. Skeletal Radiol, 16(8):629-634.

Bodei, S., Arrighi, N., Spano, P., Sigala, S. (2009). Should we be cautious on the use of commercially available antibodies to dopamine receptors? Naunyn Schmiedebergs Arch Pharmacol, 379(4):413-415.

Boeckler, F., Lanig, H., Gmeiner, P. (2005). Modeling the similarity and divergence of dopamine D2-like receptors and identification of validated ligand-receptor complexes. J Med Chem, 48(3):694-709.

Bohm, S.K., Khitin, L.M., Grady, E.F., Aponte, G., Payan, D.G., Bunnett, N.W. (1996a). Mechanisms of desensitization and resensitization of proteinase- activated receptor-2. J Biol Chem, 271(36):22003-22016.

Bohm, S.K., Kong, W., Bromme, D., Smeekens, S.P., Anderson, D.C., Connolly, A., Kahn, M., Nelken, N.A., Coughlin, S.R., Payan, D.G., Bunnett, N.W.

184 (1996b). Molecular cloning, expression and potential functions of the human proteinase-activated receptor-2. Biochem J, 314 ( Pt 3)(1009-1016.

Bokoch, M.P., Zou, Y., Rasmussen, S.G., Liu, C.W., Nygaard, R., Rosenbaum, D.M., Fung, J.J., Choi, H.J., Thian, F.S., Kobilka, T.S., Puglisi, J.D., Weis, W.I., Pardo, L., Prosser, R.S., Mueller, L., Kobilka, B.K. (2010). Ligand- specific regulation of the extracellular surface of a G-protein-coupled receptor. Nature, 463(7277):108-112.

Bolton, S.J., McNulty, C.A., Thomas, R.J., Hewitt, C.R., Wardlaw, A.J. (2003). Expression of and functional responses to protease-activated receptors on human eosinophils. J Leukoc Biol, 74(1):60-68.

Borgono, C.A., Diamandis, E.P. (2004). The emerging roles of human tissue kallikreins in cancer. Nat Rev Cancer, 4(11):876-890.

Borgono, C.A., Michael, I.P., Shaw, J.L., Luo, L.Y., Ghosh, M.C., Soosaipillai, A., Grass, L., Katsaros, D., Diamandis, E.P. (2007). Expression and functional characterization of the cancer-related serine protease, human tissue kallikrein 14. J Biol Chem, 282(4):2405-2422.

Botham, A., Guo, X., Xiao, Y.P., Morice, A.H., Compton, S.J., Sadofsky, L.R. (2011). Palmitoylation of human proteinase-activated receptor-2 differentially regulates receptor-triggered ERK1/2 activation, calcium signalling and endocytosis. Biochem J, 438(2):359-367.

Bouxsein, M.L., Boyd, S.K., Christiansen, B.A., Guldberg, R.E., Jepsen, K.J., Muller, R. (2010). Guidelines for assessment of bone microstructure in rodents using micro-computed tomography. J Bone Miner Res, 25(7):1468- 1486.

Boyle, W.J., Simonet, W.S., Lacey, D.L. (2003). Osteoclast differentiation and activation. Nature, 423(6937):337-342.

Brehmer, B., Biesterfeld, S., Jakse, G. (2003). Expression of matrix metalloproteinases (MMP-2 and -9) and their inhibitors (TIMP-1 and -2) in prostate cancer tissue. Prostate Cancer Prostatic Dis, 6(3):217-222.

Bresson, D., Pugniere, M., Roquet, F., Rebuffat, S.A., B, N.G., Cerutti, M., Guo, J., McLachlan, S.M., Rapoport, B., Estienne, V., Ruf, J., Chardes, T., Peraldi- Roux, S. (2004). Directed mutagenesis in region 713-720 of human thyroperoxidase assigns 713KFPED717 residues as being involved in the B domain of the discontinuous immunodominant region recognized by human autoantibodies. J Biol Chem, 279(37):39058-39067.

Bretschneider, E., Kaufmann, R., Braun, M., Wittpoth, M., Glusa, E., Nowak, G., Schror, K. (1999). Evidence for proteinase-activated receptor-2 (PAR-2)- mediated mitogenesis in coronary artery smooth muscle cells. Br J Pharmacol, 126(8):1735-1740.

185 Broer, B.M., Gurrath, M., Holtje, H.D. (2003). Molecular modelling studies on the ORL1-receptor and ORL1-agonists. J Comput Aided Mol Des, 17(11):739- 754.

Brown, T.A., Yang, T.M., Zaitsevskaia, T., Xia, Y., Dunn, C.A., Sigle, R.O., Knudsen, B., Carter, W.G. (2004). Adhesion or plasmin regulates tyrosine phosphorylation of a novel membrane glycoprotein p80/gp140/CUB domain- containing protein 1 in epithelia. J Biol Chem, 279(15):14772-14783.

Bubendorf, L., Schopfer, A., Wagner, U., Sauter, G., Moch, H., Willi, N., Gasser, T.C., Mihatsch, M.J. (2000). Metastatic patterns of prostate cancer: an autopsy study of 1,589 patients. Hum Pathol, 31(5):578-583.

Bucci, C., Wandinger-Ness, A., Lutcke, A., Chiariello, M., Bruni, C.B., Zerial, M. (1994). Rab5a is a common component of the apical and basolateral endocytic machinery in polarized epithelial cells. Proc Natl Acad Sci U S A, 91(11):5061-5065.

Buddenkotte, J., Stroh, C., Engels, I.H., Moormann, C., Shpacovitch, V.M., Seeliger, S., Vergnolle, N., Vestweber, D., Luger, T.A., Schulze-Osthoff, K., Steinhoff, M. (2005). Agonists of proteinase-activated receptor-2 stimulate upregulation of intercellular cell adhesion molecule-1 in primary human keratinocytes via activation of NF-kappa B. J Invest Dermatol, 124(1):38-45.

Buie, H.R., Campbell, G.M., Klinck, R.J., MacNeil, J.A., Boyd, S.K. (2007). Automatic segmentation of cortical and trabecular compartments based on a dual threshold technique for in vivo micro-CT bone analysis. Bone, 41(4):505-515.

Bushell, T.J., Plevin, R., Cobb, S., Irving, A.J. (2006). Characterization of proteinase-activated receptor 2 signalling and expression in rat hippocampal neurons and astrocytes. Neuropharmacology, 50(6):714-725.

Cairns, J.A. (2005). Inhibitors of mast cell tryptase beta as therapeutics for the treatment of asthma and inflammatory disorders. Pulm Pharmacol Ther, 18(1):55-66.

Camerer, E., Huang, W., Coughlin, S.R. (2000). Tissue factor- and factor X- dependent activation of protease-activated receptor 2 by factor VIIa. Proc Natl Acad Sci U S A, 97(10):5255-5260.

Cao, Y., Lundwall, A., Gadaleanu, V., Lilja, H., Bjartell, A. (2002). Anti-thrombin is expressed in the benign prostatic epithelium and in prostate cancer and is capable of forming complexes with prostate-specific antigen and human glandular kallikrein 2. Am J Pathol, 161(6):2053-2063.

Caruso, R., Pallone, F., Fina, D., Gioia, V., Peluso, I., Caprioli, F., Stolfi, C., Perfetti, A., Spagnoli, L.G., Palmieri, G., Macdonald, T.T., Monteleone, G. (2006). Protease-activated receptor-2 activation in gastric cancer cells promotes

186 epidermal growth factor receptor trans-activation and proliferation. Am J Pathol, 169(1):268-278.

Casar, B., He, Y., Iconomou, M., Hooper, J.D., Quigley, J.P., Deryugina, E.I. (2011). Blocking of CDCP1 cleavage in vivo prevents Akt-dependent survival and inhibits metastatic colonization through PARP1-mediated apoptosis of cancer cells. Oncogene, In press.

Chackalamannil, S., Wang, Y., Greenlee, W.J., Hu, Z., Xia, Y., Ahn, H.S., Boykow, G., Hsieh, Y., Palamanda, J., Agans-Fantuzzi, J., Kurowski, S., Graziano, M., Chintala, M. (2008). Discovery of a novel, orally active himbacine-based thrombin receptor antagonist (SCH 530348) with potent antiplatelet activity. J Med Chem, 51(11):3061-3064.

Charest, P.G., Bouvier, M. (2003). Palmitoylation of the V2 vasopressin receptor carboxyl tail enhances beta-arrestin recruitment leading to efficient receptor endocytosis and ERK1/2 activation. J Biol Chem, 278(42):41541-41551.

Charhon, S.A., Chapuy, M.C., Delvin, E.E., Valentin-Opran, A., Edouard, C.M., Meunier, P.J. (1983). Histomorphometric analysis of sclerotic bone metastases from prostatic carcinoma special reference to osteomalacia. Cancer, 51(5):918-924.

Chavrier, P., Vingron, M., Sander, C., Simons, K., Zerial, M. (1990). Molecular cloning of YPT1/SEC4-related cDNAs from an epithelial cell line. Mol Cell Biol, 10(12):6578-6585.

Chay, C.H., Cooper, C.R., Gendernalik, J.D., Dhanasekaran, S.M., Chinnaiyan, A.M., Rubin, M.A., Schmaier, A.H., Pienta, K.J. (2002). A functional thrombin receptor (PAR1) is expressed on bone-derived prostate cancer cell lines. Urology, 60(5):760-765.

Chen, W.Y., Shi, Y.Y., Zheng, Y.L., Zhao, X.Z., Zhang, G.J., Chen, S.Q., Yang, P.D., He, L. (2004). Case-control study and transmission disequilibrium test provide consistent evidence for association between schizophrenia and genetic variation in the 22q11 gene ZDHHC8. Hum Mol Genet, 13(23):2991- 2995.

Cherezov, V., Fersi, H., Caffrey, M. (2001). Crystallization screens: compatibility with the lipidic cubic phase for in meso crystallization of membrane proteins. Biophys J, 81(1):225-242.

Cherezov, V., Rosenbaum, D.M., Hanson, M.A., Rasmussen, S.G., Thian, F.S., Kobilka, T.S., Choi, H.J., Kuhn, P., Weis, W.I., Kobilka, B.K., Stevens, R.C. (2007). High-resolution crystal structure of an engineered human beta2- adrenergic G protein-coupled receptor. Science, 318(5854):1258-1265.

Chini, B., Parenti, M. (2004). G-protein coupled receptors in lipid rafts and caveolae: how, when and why do they go there? Journal of molecular endocrinology, 32(2):325-338.

187

Chiu, L.L., Perng, D.W., Yu, C.H., Su, S.N., Chow, L.P. (2007). Mold allergen, pen C 13, induces IL-8 expression in human airway epithelial cells by activating protease-activated receptor 1 and 2. J Immunol, 178(8):5237-5244.

Christensen, J.G., Schreck, R., Burrows, J., Kuruganti, P., Chan, E., Le, P., Chen, J., Wang, X., Ruslim, L., Blake, R., Lipson, K.E., Ramphal, J., Do, S., Cui, J.J., Cherrington, J.M., Mendel, D.B. (2003). A selective small molecule inhibitor of c-Met kinase inhibits c-Met-dependent phenotypes in vitro and exhibits cytoreductive antitumor activity in vivo. Cancer research, 63(21):7345-7355.

Ciruela, F., Vallano, A., Arnau, J.M., Sanchez, S., Borroto-Escuela, D.O., Agnati, L.F., Fuxe, K., Fernandez-Duenas, V. (2010). G protein-coupled receptor oligomerization for what? J Recept Signal Transduct Res, 30(5):322-330.

Clarke, S.J., Rivory, L.P. (1999). Clinical pharmacokinetics of docetaxel. Clin Pharmacokinet, 36(2):99-114.

Coleman, R.E. (1997). Skeletal complications of malignancy. Cancer, 80(8 Suppl):1588-1594.

Colognato, R., Slupsky, J.R., Jendrach, M., Burysek, L., Syrovets, T., Simmet, T. (2003). Differential expression and regulation of protease-activated receptors in human peripheral monocytes and monocyte-derived antigen-presenting cells. Blood, 102(7):2645-2652.

Compton, S.J., Renaux, B., Wijesuriya, S.J., Hollenberg, M.D. (2001). Glycosylation and the activation of proteinase-activated receptor 2 (PAR(2)) by human mast cell tryptase. Br J Pharmacol, 134(4):705-718.

Compton, S.J., Sandhu, S., Wijesuriya, S.J., Hollenberg, M.D. (2002). Glycosylation of human proteinase-activated receptor-2 (hPAR2): role in cell surface expression and signalling. Biochem J, 368(Pt 2):495-505.

Congreve, M., Langmead, C.J., Mason, J.S., Marshall, F.H. (2011). Progress in structure based drug design for G protein-coupled receptors. J Med Chem, 54(13):4283-4311.

Conner, M., Hawtin, S.R., Simms, J., Wootten, D., Lawson, Z., Conner, A.C., Parslow, R.A., Wheatley, M. (2007). Systematic analysis of the entire second extracellular loop of the V(1a) vasopressin receptor: key residues, conserved throughout a G-protein-coupled receptor family, identified. J Biol Chem, 282(24):17405-17412.

Conway, K., Ruge, F., Price, P., Harding, K.G., Jiang, W.G. (2007). Hepatocyte growth factor regulation: an integral part of why wounds become chronic. Wound Repair Regen, 15(5):683-692.

Corey, E., Quinn, J.E., Bladou, F., Brown, L.G., Roudier, M.P., Brown, J.M., Buhler, K.R., Vessella, R.L. (2002). Establishment and characterization of osseous

188 prostate cancer models: intra-tibial injection of human prostate cancer cells. Prostate, 52(1):20-33.

Costanzi, S., Mamedova, L., Gao, Z.G., Jacobson, K.A. (2004). Architecture of P2Y nucleotide receptors: structural comparison based on sequence analysis, mutagenesis, and homology modeling. J Med Chem, 47(22):5393-5404.

Costello, L.C., Feng, P., Milon, B., Tan, M., Franklin, R.B. (2004). Role of zinc in the pathogenesis and treatment of prostate cancer: critical issues to resolve. Prostate Cancer Prostatic Dis, 7(2):111-117.

Cottrell, G.S., Amadesi, S., Grady, E.F., Bunnett, N.W. (2004). Trypsin IV, a novel agonist of protease-activated receptors 2 and 4. J Biol Chem, 279(14):13532- 13539.

Csernok, E., Ai, M., Gross, W.L., Wicklein, D., Petersen, A., Lindner, B., Lamprecht, P., Holle, J.U., Hellmich, B. (2006). Wegener autoantigen induces maturation of dendritic cells and licenses them for Th1 priming via the protease-activated receptor-2 pathway. Blood, 107(11):4440-4448.

Curran, S., Murray, G.I. (2000). Matrix metalloproteinases: molecular aspects of their roles in tumour invasion and metastasis. Eur J Cancer, 36(13 Spec No):1621-1630.

D'Andrea, M.R., Derian, C.K., Santulli, R.J., Andrade-Gordon, P. (2001). Differential expression of protease-activated receptors-1 and -2 in stromal fibroblasts of normal, benign, and malignant human tissues. Am J Pathol, 158(6):2031-2041.

Dai, Z., Huang, Y., Sadee, W. (2004). Growth factor signaling and resistance to cancer chemotherapy. Curr Top Med Chem, 4(13):1347-1356.

Damiano, B.P., Cheung, W.M., Santulli, R.J., Fung-Leung, W.P., Ngo, K., Ye, R.D., Darrow, A.L., Derian, C.K., de Garavilla, L., Andrade-Gordon, P. (1999). Cardiovascular responses mediated by protease-activated receptor-2 (PAR-2) and thrombin receptor (PAR-1) are distinguished in mice deficient in PAR-2 or PAR-1. J Pharmacol Exp Ther, 288(2):671-678.

Darmoul, D., Gratio, V., Devaud, H., Laburthe, M. (2004a). Protease-activated receptor 2 in colon cancer: trypsin-induced MAPK phosphorylation and cell proliferation are mediated by epidermal growth factor receptor transactivation. J Biol Chem, 279(20):20927-20934.

Darmoul, D., Gratio, V., Devaud, H., Peiretti, F., Laburthe, M. (2004b). Activation of proteinase-activated receptor 1 promotes human colon cancer cell proliferation through epidermal growth factor receptor transactivation. Mol Cancer Res, 2(9):514-522.

189 Darmoul, D., Marie, J.C., Devaud, H., Gratio, V., Laburthe, M. (2001). Initiation of human colon cancer cell proliferation by trypsin acting at protease-activated receptor-2. Br J Cancer, 85(5):772-779.

Darrow, A.L., Fung-Leung, W.P., Ye, R.D., Santulli, R.J., Cheung, W.M., Derian, C.K., Burns, C.L., Damiano, B.P., Zhou, L., Keenan, C.M., Peterson, P.A., Andrade-Gordon, P. (1996). Biological consequences of thrombin receptor deficiency in mice. Thromb Haemost, 76(6):860-866.

Debela, M., Magdolen, V., Grimminger, V., Sommerhoff, C., Messerschmidt, A., Huber, R., Friedrich, R., Bode, W., Goettig, P. (2006). Crystal structures of human tissue kallikrein 4: activity modulation by a specific zinc binding site. J Mol Biol, 362(5):1094-1107.

Defea, K. (2008). Beta-arrestins and heterotrimeric G-proteins: collaborators and competitors in signal transduction. Br J Pharmacol, 153 Suppl 1(S298-309.

DeFea, K.A., Zalevsky, J., Thoma, M.S., Dery, O., Mullins, R.D., Bunnett, N.W. (2000). beta-arrestin-dependent endocytosis of proteinase-activated receptor 2 is required for intracellular targeting of activated ERK1/2. J Cell Biol, 148(6):1267-1281.

Delos Santos, N.M., Gardner, L.A., White, S.W., Bahouth, S.W. (2006). Characterization of the residues in helix 8 of the human beta1-adrenergic receptor that are involved in coupling the receptor to G proteins. J Biol Chem, 281(18):12896-12907.

Denis, L., Murphy, G.P. (1993). Overview of phase III trials on combined androgen treatment in patients with metastatic prostate cancer. Cancer, 72(12 Suppl):3888-3895.

Dery, O., Thoma, M.S., Wong, H., Grady, E.F., Bunnett, N.W. (1999). Trafficking of proteinase-activated receptor-2 and beta-arrestin-1 tagged with green fluorescent protein. beta-Arrestin-dependent endocytosis of a proteinase receptor. J Biol Chem, 274(26):18524-18535.

Devlin, M.G., Pfeiffer, B., Flanagan, B., Beyer, R.L., Cocks, T.M., Fairlie, D.P. (2007). Hepta and octapeptide agonists of protease-activated receptor 2. J Pept Sci, 13(12):856-861.

Dinter, A., Berger, E.G. (1998). Golgi-disturbing agents. Histochem Cell Biol, 109(5- 6):571-590.

Dorfleutner, A., Ruf, W. (2003). Regulation of tissue factor cytoplasmic domain phosphorylation by palmitoylation. Blood, 102(12):3998-4005.

Dorsam, R.T., Gutkind, J.S. (2007). G-protein-coupled receptors and cancer. Nat Rev Cancer, 7(2):79-94.

190 Dougall, W.C., Glaccum, M., Charrier, K., Rohrbach, K., Brasel, K., De Smedt, T., Daro, E., Smith, J., Tometsko, M.E., Maliszewski, C.R., Armstrong, A., Shen, V., Bain, S., Cosman, D., Anderson, D., Morrissey, P.J., Peschon, J.J., Schuh, J. (1999). RANK is essential for osteoclast and lymph node development. Genes & development, 13(18):2412-2424.

Drisdel, R.C., Green, W.N. (2004). Labeling and quantifying sites of protein palmitoylation. Biotechniques, 36(2):276-285.

Dulon, S., Cande, C., Bunnett, N.W., Hollenberg, M.D., Chignard, M., Pidard, D. (2003). Proteinase-activated receptor-2 and human lung epithelial cells: disarming by neutrophil serine proteinases. Am J Respir Cell Mol Biol, 28(3):339-346.

Dulon, S., Leduc, D., Cottrell, G.S., D'Alayer, J., Hansen, K.K., Bunnett, N.W., Hollenberg, M.D., Pidard, D., Chignard, M. (2005). Pseudomonas aeruginosa elastase disables proteinase-activated receptor 2 in respiratory epithelial cells. Am J Respir Cell Mol Biol, 32(5):411-419.

Eisenhauer, E.A., Vermorken, J.B. (1998). The taxoids. Comparative clinical pharmacology and therapeutic potential. Drugs, 55(1):5-30.

Erben, R.G., Glosmann, M. (2012). Histomorphometry in rodents. Methods Mol Biol, 816(279-303.

Ernst, O.P., Meyer, C.K., Marin, E.P., Henklein, P., Fu, W.Y., Sakmar, T.P., Hofmann, K.P. (2000). Mutation of the fourth cytoplasmic loop of rhodopsin affects binding of transducin and peptides derived from the carboxyl-terminal sequences of transducin alpha and gamma subunits. J Biol Chem, 275(3):1937-1943.

Escriba, P.V., Wedegaertner, P.B., Goni, F.M., Vogler, O. (2007). Lipid-protein interactions in GPCR-associated signaling. Biochim Biophys Acta, 1768(4):836-852.

Feldman, B.J., Feldman, D. (2001). The development of androgen-independent prostate cancer. Nat Rev Cancer, 1(1):34-45.

Fernandez-Hernando, C., Fukata, M., Bernatchez, P.N., Fukata, Y., Lin, M.I., Bredt, D.S., Sessa, W.C. (2006). Identification of Golgi-localized acyl transferases that palmitoylate and regulate endothelial nitric oxide synthase. J Cell Biol, 174(3):369-377.

Ferrell, W.R., Kelso, E.B., Lockhart, J.C., Plevin, R., McInnes, I.B. (2010). Protease- activated receptor 2: a novel pathogenic pathway in a murine model of osteoarthritis. Ann Rheum Dis, 69(11):2051-2054.

Festuccia, C., Gravina, G.L., D'Alessandro, A.M., Muzi, P., Millimaggi, D., Dolo, V., Ricevuto, E., Vicentini, C., Bologna, M. (2009). Azacitidine improves

191 antitumor effects of docetaxel and cisplatin in aggressive prostate cancer models. Endocr Relat Cancer, 16(2):401-413.

Filipek, S., Stenkamp, R.E., Teller, D.C., Palczewski, K. (2003a). G protein-coupled receptor rhodopsin: a prospectus. Annu Rev Physiol, 65(851-879.

Filipek, S., Teller, D.C., Palczewski, K., Stenkamp, R. (2003b). The crystallographic model of rhodopsin and its use in studies of other G protein-coupled receptors. Annu Rev Biophys Biomol Struct, 32(375-397.

Fischer, A. (1946). Mechanism of the proteolytic activity of malignant tissue cells. Nature, 157(442.

Flohr, S., Kurz, M., Kostenis, E., Brkovich, A., Fournier, A., Klabunde, T. (2002). Identification of nonpeptidic urotensin II receptor antagonists by virtual screening based on a pharmacophore model derived from structure-activity relationships and nuclear magnetic resonance studies on urotensin II. J Med Chem, 45(9):1799-1805.

Flower, D.R. (1999). Modelling G-protein-coupled receptors for drug design. Biochimica et biophysica acta, 1422(3):207-234.

Fredriksson, R., Lagerstrom, M.C., Lundin, L.G., Schioth, H.B. (2003). The G- protein-coupled receptors in the human genome form five main families. Phylogenetic analysis, paralogon groups, and fingerprints. Mol Pharmacol, 63(6):1256-1272.

Frenette, G., Deperthes, D., Tremblay, R.R., Lazure, C., Dube, J.Y. (1997). Purification of enzymatically active kallikrein hK2 from human seminal plasma. Biochimica et biophysica acta, 1334(1):109-115.

Fritz, V., Louis-Plence, P., Apparailly, F., Noel, D., Voide, R., Pillon, A., Nicolas, J.C., Muller, R., Jorgensen, C. (2007). Micro-CT combined with bioluminescence imaging: a dynamic approach to detect early tumor-bone interaction in a tumor osteolysis murine model. Bone, 40(4):1032-1040.

Fukata, Y., Fukata, M. (2010). Protein palmitoylation in neuronal development and synaptic plasticity. Nat Rev Neurosci, 11(3):161-175.

Fukushima, Y., Otsuka, H., Ishikawa, M., Asano, T., Anai, M., Katsube, T., Ogawa, K., Kajiwara, T., Ohkawa, S., Ishikawa, T., Omata, M., Saitoh, T. (2001a). Potent and long-lasting action of lafutidine on the human histamine H(2) receptor. Digestion, 64(3):155-160.

Fukushima, Y., Saitoh, T., Anai, M., Ogihara, T., Inukai, K., Funaki, M., Sakoda, H., Onishi, Y., Ono, H., Fujishiro, M., Ishikawa, T., Takata, K., Nagai, R., Omata, M., Asano, T. (2001b). Palmitoylation of the canine histamine H2 receptor occurs at Cys(305) and is important for cell surface targeting. Biochim Biophys Acta, 1539(3):181-191.

192 Gan, L., Chen, S., Wang, Y., Watahiki, A., Bohrer, L., Sun, Z., Huang, H. (2009). Inhibition of the androgen receptor as a novel mechanism of taxol chemotherapy in prostate cancer. Cancer research, 69(21):8386-8394.

Ge, L., Ly, Y., Hollenberg, M., DeFea, K. (2003). A beta-arrestin-dependent scaffold is associated with prolonged MAPK activation in pseudopodia during protease-activated receptor-2-induced chemotaxis. J Biol Chem, 278(36):34418-34426.

Ge, L., Shenoy, S.K., Lefkowitz, R.J., DeFea, K. (2004). Constitutive protease- activated receptor-2-mediated migration of MDA MB-231 breast cancer cells requires both beta-arrestin-1 and -2. J Biol Chem, 279(53):55419-55424.

Gehret, A.U., Jones, B.W., Tran, P.N., Cook, L.B., Greuber, E.K., Hinkle, P.M. (2010). Role of helix 8 of the thyrotropin-releasing hormone receptor in phosphorylation by G protein-coupled receptor kinase. Mol Pharmacol, 77(2):288-297.

Geldof, A.A., Rao, B.R. (1990). Factors in prostate cancer metastasis. Anticancer Res, 10(5A):1303-1306.

Georgy, S.R., Pagel, C.N., Ghasem-Zadeh, A., Zebaze, R.M., Pike, R.N., Sims, N.A., Mackie, E.J. (2011). Proteinase-activated receptor-2 is required for normal osteoblast and osteoclast differentiation during skeletal growth and repair. Bone, 50(3):704-712.

Georgy, S.R., Pagel, C.N., Wong, D.M., Sivagurunathan, S., Loh, L.H., Myers, D.E., Hollenberg, M.D., Pike, R.N., Mackie, E.J. (2010). Proteinase-activated receptor-2 (PAR2) and mouse osteoblasts: Regulation of cell function and lack of specificity of PAR2-activating peptides. Clin Exp Pharmacol Physiol, 37(3):328-336.

Gerszten, R.E., Chen, J., Ishii, M., Ishii, K., Wang, L., Nanevicz, T., Turck, C.W., Vu, T.K., Coughlin, S.R. (1994). Specificity of the thrombin receptor for agonist peptide is defined by its extracellular surface. Nature, 368(6472):648- 651.

Gingrich, J.R., Barrios, R.J., Morton, R.A., Boyce, B.F., DeMayo, F.J., Finegold, M.J., Angelopoulou, R., Rosen, J.M., Greenberg, N.M. (1996). Metastatic prostate cancer in a transgenic mouse. Cancer research, 56(18):4096-4102.

Goh, F.G., Ng, P.Y., Nilsson, M., Kanke, T., Plevin, R. (2009). Dual effect of the novel peptide antagonist K-14585 on proteinase-activated receptor-2- mediated signalling. Br J Pharmacol, 158(7):1695-1704.

Goon Goh, F., Sloss, C.M., Cunningham, M.R., Nilsson, M., Cadalbert, L., Plevin, R. (2008). G-protein-dependent and -independent pathways regulate proteinase-activated receptor-2 mediated p65 NFkappaB serine 536 phosphorylation in human keratinocytes. Cell Signal, 20(7):1267-1274.

193 Gratio, V., Beaufort, N., Seiz, L., Maier, J., Virca, G.D., Debela, M., Grebenchtchikov, N., Magdolen, V., Darmoul, D. (2010). Kallikrein-related peptidase 4: a new activator of the aberrantly expressed protease-activated receptor 1 in colon cancer cells. Am J Pathol, 176(3):1452-1461.

Greenberg, D.L., Mize, G.J., Takayama, T.K. (2003). Protease-activated receptor mediated RhoA signaling and cytoskeletal reorganization in LNCaP cells. Biochemistry, 42(3):702-709.

Gregory, K.J., Hall, N.E., Tobin, A.B., Sexton, P.M., Christopoulos, A. (2010). Identification of orthosteric and allosteric site mutations in M2 muscarinic acetylcholine receptors that contribute to ligand-selective signaling bias. J Biol Chem, 285(10):7459-7474.

Grimsey, N., Soto, A.G., Trejo, J. (2011). Regulation of protease-activated receptor signaling by post-translational modifications. IUBMB Life, 63(6):403-411.

Grisaru-Granovsky, S., Salah, Z., Maoz, M., Pruss, D., Beller, U., Bar-Shavit, R. (2005). Differential expression of protease activated receptor 1 (Par1) and pY397FAK in benign and malignant human ovarian tissue samples. Int J Cancer, 113(3):372-378.

Groarke, D.A., Drmota, T., Bahia, D.S., Evans, N.A., Wilson, S., Milligan, G. (2001). Analysis of the C-terminal tail of the rat thyrotropin-releasing hormone receptor-1 in interactions and cointernalization with beta-arrestin 1- green fluorescent protein. Mol Pharmacol, 59(2):375-385.

Guise, T.A., Mohammad, K.S., Clines, G., Stebbins, E.G., Wong, D.H., Higgins, L.S., Vessella, R., Corey, E., Padalecki, S., Suva, L., Chirgwin, J.M. (2006). Basic mechanisms responsible for osteolytic and osteoblastic bone metastases. Clin Cancer Res, 12(20 Pt 2):6213s-6216s.

Gutierrez-de-Teran, H., Centeno, N.B., Pastor, M., Sanz, F. (2004a). Novel approaches for modeling of the A1 adenosine receptor and its agonist binding site. Proteins, 54(4):705-715.

Gutierrez-de-Teran, H., Pastor, M., Centeno, N.B., Aqvist, J., Sanz, F. (2004b). Comparative analysis of putative agonist-binding modes in the human A1 adenosine receptor. Chembiochem, 5(6):841-849.

Hall, C.L., Bafico, A., Dai, J., Aaronson, S.A., Keller, E.T. (2005). Prostate cancer cells promote osteoblastic bone metastases through Wnts. Cancer research, 65(17):7554-7560.

Hall, C.L., Keller, E.T. (2006). The role of Wnts in bone metastases. Cancer Metastasis Rev, 25(4):551-558.

Hamdani, N., van der Velden, J. (2009). Lack of specificity of antibodies directed against human beta-adrenergic receptors. Naunyn Schmiedebergs Arch Pharmacol, 379(4):403-407.

194

Hamelin, E., Theriault, C., Laroche, G., Parent, J.L. (2005). The intracellular trafficking of the G protein-coupled receptor TPbeta depends on a direct interaction with Rab11. J Biol Chem, 280(43):36195-36205.

Hamm, H.E. (2001). How activated receptors couple to G proteins. Proc Natl Acad Sci U S A, 98(9):4819-4821.

Hansen, K.K., Oikonomopoulou, K., Baruch, A., Ramachandran, R., Beck, P., Diamandis, E.P., Hollenberg, M.D. (2008). Proteinases as hormones: targets and mechanisms for proteolytic signaling. Biol Chem, 389(8):971-982.

Hansen, K.K., Sherman, P.M., Cellars, L., Andrade-Gordon, P., Pan, Z., Baruch, A., Wallace, J.L., Hollenberg, M.D., Vergnolle, N. (2005). A major role for proteolytic activity and proteinase-activated receptor-2 in the pathogenesis of infectious colitis. Proc Natl Acad Sci U S A, 102(23):8363-8368.

Hasdemir, B., Bunnett, N.W., Cottrell, G.S. (2007). Hepatocyte growth factor- regulated tyrosine kinase substrate (HRS) mediates post-endocytic trafficking of protease-activated receptor 2 and calcitonin receptor-like receptor. J Biol Chem, 282(40):29646-29657.

Hasdemir, B., Murphy, J.E., Cottrell, G.S., Bunnett, N.W. (2009). Endosomal deubiquitinating enzymes control ubiquitination and down-regulation of protease-activated receptor 2. J Biol Chem, 284(41):28453-28466.

Hawtin, S.R., Tobin, A.B., Patel, S., Wheatley, M. (2001). Palmitoylation of the vasopressin V1a receptor reveals different conformational requirements for signaling, agonist-induced receptor phosphorylation, and sequestration. J Biol Chem, 276(41):38139-38146.

He, Y., Hryciw, D.H., Carroll, M.L., Myers, S.A., Whitbread, A.K., Kumar, S., Poronnik, P., Hooper, J.D. (2008). The ubiquitin-protein ligase Nedd4-2 differentially interacts with and regulates members of the Tweety family of chloride ion channels. J Biol Chem, 283(35):24000-24010.

He, Y., Wortmann, A., Burke, L.J., Reid, J.C., Adams, M.N., Abdul-Jabbar, I., Quigley, J.P., Leduc, R., Kirchhofer, D., Hooper, J.D. (2010). Proteolysis- induced N-terminal ectodomain shedding of the integral membrane glycoprotein CUB domain-containing protein 1 (CDCP1) is accompanied by tyrosine phosphorylation of its C-terminal domain and recruitment of Src and PKCdelta. J Biol Chem, 285(34):26162-26173.

Heider, I., Schulze, B., Oswald, E., Henklein, P., Scheele, J., Kaufmann, R. (2004). PAR1-type thrombin receptor stimulates migration and matrix adhesion of human colon carcinoma cells by a PKCepsilon-dependent mechanism. Oncol Res, 14(10):475-482.

195 Hein, L., Ishii, K., Coughlin, S.R., Kobilka, B.K. (1994). Intracellular targeting and trafficking of thrombin receptors. A novel mechanism for resensitization of a G protein-coupled receptor. J Biol Chem, 269(44):27719-27726.

Henry, M.D., Silva, M.D., Wen, S., Siebert, E., Solin, E., Chandra, S., Worland, P.J. (2005). Spiculated periosteal response induced by intraosseous injection of 22Rv1 prostate cancer cells resembles subset of bone metastases in prostate cancer patients. Prostate, 65(4):347-354.

Hernandez, N.A., Correa, E., Avila, E.P., Vela, T.A., Perez, V.M. (2009). PAR1 is selectively over expressed in high grade breast cancer patients: a cohort study. J Transl Med, 7(47.

Herz, U., Schnoy, N., Borelli, S., Weigl, L., Kasbohrer, U., Daser, A., Wahn, U., Kottgen, E., Renz, H. (1998). A human-SCID mouse model for allergic immune response bacterial superantigen enhances skin inflammation and suppresses IgE production. J Invest Dermatol, 110(3):224-231.

Higginbottom, A., Cain, S.A., Woodruff, T.M., Proctor, L.M., Madala, P.K., Tyndall, J.D., Taylor, S.M., Fairlie, D.P., Monk, P.N. (2005). Comparative agonist/antagonist responses in mutant human C5a receptors define the ligand binding site. J Biol Chem, 280(18):17831-17840.

Hjortoe, G.M., Petersen, L.C., Albrektsen, T., Sorensen, B.B., Norby, P.L., Mandal, S.K., Pendurthi, U.R., Rao, L.V. (2004). Tissue factor-factor VIIa-specific up-regulation of IL-8 expression in MDA-MB-231 cells is mediated by PAR- 2 and results in increased cell migration. Blood, 103(8):3029-3037.

Hollenberg, M.D., Compton, S.J. (2002). International Union of Pharmacology. XXVIII. Proteinase-activated receptors. Pharmacol Rev, 54(2):203-217.

Hollenberg, M.D., Saifeddine, M., al-Ani, B. (1996). Proteinase-activated receptor-2 in rat aorta: structural requirements for agonist activity of receptor-activating peptides. Mol Pharmacol, 49(2):229-233.

Hong, J.H., Hong, J.Y., Park, B., Lee, S.I., Seo, J.T., Kim, K.E., Sohn, M.H., Shin, D.M. (2008). Chitinase activates protease-activated receptor-2 in human airway epithelial cells. Am J Respir Cell Mol Biol, 39(5):530-535.

Hooper, J.D., Zijlstra, A., Aimes, R.T., Liang, H., Claassen, G.F., Tarin, D., Testa, J.E., Quigley, J.P. (2003). Subtractive immunization using highly metastatic human tumor cells identifies SIMA135/CDCP1, a 135 kDa cell surface phosphorylated glycoprotein antigen. Oncogene, 22(12):1783-1794.

Horoszewicz, J.S., Leong, S.S., Kawinski, E., Karr, J.P., Rosenthal, H., Chu, T.M., Mirand, E.A., Murphy, G.P. (1983). LNCaP model of human prostatic carcinoma. Cancer research, 43(4):1809-1818.

196 Houle, S., Papez, M.D., Ferazzini, M., Hollenberg, M.D., Vergnolle, N. (2005). Neutrophils and the kallikrein-kinin system in proteinase-activated receptor 4-mediated inflammation in rodents. Br J Pharmacol, 146(5):670-678.

Hove, B., Gyldensted, C. (1990). Spiculated vertebral metastases from prostatic carcinoma. Report of first two cases. Neuroradiology, 32(4):337-339.

Howells, G.L., Macey, M.G., Chinni, C., Hou, L., Fox, M.T., Harriott, P., Stone, S.R. (1997). Proteinase-activated receptor-2: expression by human neutrophils. J Cell Sci, 110 ( Pt 7)(881-887.

Huang, W., Osman, R., Gershengorn, M.C. (2005). Agonist-induced conformational changes in thyrotropin-releasing hormone receptor type I: disulfide cross- linking and molecular modeling approaches. Biochemistry, 44(7):2419-2431.

Huang, X.P., Nagy, P.I., Williams, F.E., Peseckis, S.M., Messer, W.S., Jr. (1999). Roles of threonine 192 and asparagine 382 in agonist and antagonist interactions with M1 muscarinic receptors. Br J Pharmacol, 126(3):735-745.

Hulme, E.C., Lu, Z.L., Bee, M.S. (2003). Scanning mutagenesis studies of the M1 muscarinic acetylcholine receptor. Receptors Channels, 9(4):215-228.

Hung, H. (2007). Bevacizumab plus 5-fluorouracil induce growth suppression in the CWR-22 and CWR-22R prostate cancer xenografts. Mol Cancer Ther, 6(8):2149-2157.

Ivanov, A.A., Baskin, II, Palyulin, V.A., Piccagli, L., Baraldi, P.G., Zefirov, N.S. (2005). Molecular modeling and molecular dynamics simulation of the human A2B adenosine receptor. The study of the possible binding modes of the A2B receptor antagonists. J Med Chem, 48(22):6813-6820.

Jaakola, V.P., Griffith, M.T., Hanson, M.A., Cherezov, V., Chien, E.Y., Lane, J.R., Ijzerman, A.P., Stevens, R.C. (2008). The 2.6 angstrom crystal structure of a human A2A adenosine receptor bound to an antagonist. Science, 322(5905):1211-1217.

Jacob, C., Cottrell, G.S., Gehringer, D., Schmidlin, F., Grady, E.F., Bunnett, N.W. (2005a). c-Cbl mediates ubiquitination, degradation, and down-regulation of human protease-activated receptor 2. J Biol Chem, 280(16):16076-16087.

Jacob, C., Yang, P.C., Darmoul, D., Amadesi, S., Saito, T., Cottrell, G.S., Coelho, A.M., Singh, P., Grady, E.F., Perdue, M., Bunnett, N.W. (2005b). Mast cell tryptase controls paracellular permeability of the intestine. Role of protease- activated receptor 2 and beta-arrestins. J Biol Chem, 280(36):31936-31948.

Jacobson, K.A., Gao, Z.G., Chen, A., Barak, D., Kim, S.A., Lee, K., Link, A., Rompaey, P.V., van Calenbergh, S., Liang, B.T. (2001). Neoceptor concept based on molecular complementarity in GPCRs: a mutant adenosine A(3) receptor with selectively enhanced affinity for amine-modified nucleosides. J Med Chem, 44(24):4125-4136.

197

Jahan, I., Fujimoto, J., Alam, S.M., Sato, E., Sakaguchi, H., Tamaya, T. (2007). Role of protease activated receptor-2 in tumor advancement of ovarian cancers. Ann Oncol, 18(9):1506-1512.

Jarry, A., Dorso, L., Gratio, V., Forgue-Lafitte, M.E., Laburthe, M., Laboisse, C.L., Darmoul, D. (2007). PAR-2 activation increases human intestinal mucin secretion through EGFR transactivation. Biochem Biophys Res Commun, 364(3):689-694.

Jennings, B.C., Nadolski, M.J., Ling, Y., Baker, M.B., Harrison, M.L., Deschenes, R.J., Linder, M.E. (2009). 2-Bromopalmitate and 2-(2-hydroxy-5-nitro- benzylidene)-benzo[b]thiophen-3-one inhibit DHHC-mediated palmitoylation in vitro. J Lipid Res, 50(2):233-242.

Jensen, B.C., Swigart, P.M., Simpson, P.C. (2009). Ten commercial antibodies for alpha-1-adrenergic receptor subtypes are nonspecific. Naunyn Schmiedebergs Arch Pharmacol, 379(4):409-412.

Jia, L., Yu, W., Wang, P., Sanders, B.G., Kline, K. (2008). In vivo and in vitro studies of anticancer actions of alpha-TEA for human prostate cancer cells. Prostate, 68(8):849-860.

Jiang, X., Bailly, M.A., Panetti, T.S., Cappello, M., Konigsberg, W.H., Bromberg, M.E. (2004). Formation of tissue factor-factor VIIa-factor Xa complex promotes cellular signaling and migration of human breast cancer cells. J Thromb Haemost, 2(1):93-101.

Jin, E., Fujiwara, M., Pan, X., Ghazizadeh, M., Arai, S., Ohaki, Y., Kajiwara, K., Takemura, T., Kawanami, O. (2003). Protease-activated receptor (PAR)-1 and PAR-2 participate in the cell growth of alveolar capillary endothelium in primary lung adenocarcinomas. Cancer, 97(3):703-713.

Jositsch, G., Papadakis, T., Haberberger, R.V., Wolff, M., Wess, J., Kummer, W. (2009). Suitability of muscarinic acetylcholine receptor antibodies for immunohistochemistry evaluated on tissue sections of receptor gene-deficient mice. Naunyn Schmiedebergs Arch Pharmacol, 379(4):389-395.

Kaighn, M.E., Lechner, J.F., Narayan, K.S., Jones, L.W. (1978). Prostate carcinoma: tissue culture cell lines. Natl Cancer Inst Monogr, 49):17-21.

Kalli, K.R. (2007). MORAb-003, a fully humanized monoclonal antibody against the folate receptor alpha, for the potential treatment of epithelial ovarian cancer. Curr Opin Investig Drugs, 8(12):1067-1073.

Kanke, T., Kabeya, M., Kubo, S., Kondo, S., Yasuoka, K., Tagashira, J., Ishiwata, H., Saka, M., Furuyama, T., Nishiyama, T., Doi, T., Hattori, Y., Kawabata, A., Cunningham, M.R., Plevin, R. (2009). Novel antagonists for proteinase- activated receptor 2: inhibition of cellular and vascular responses in vitro and in vivo. Br J Pharmacol, 158(1):361-371.

198

Kanke, T., Macfarlane, S.R., Seatter, M.J., Davenport, E., Paul, A., McKenzie, R.C., Plevin, R. (2001). Proteinase-activated receptor-2-mediated activation of stress-activated protein kinases and inhibitory kappa B kinases in NCTC 2544 keratinocytes. J Biol Chem, 276(34):31657-31666.

Kapadia, R.D., Stroup, G.B., Badger, A.M., Koller, B., Levin, J.M., Coatney, R.W., Dodds, R.A., Liang, X., Lark, M.W., Gowen, M. (1998). Applications of micro-CT and MR microscopy to study pre-clinical models of osteoporosis and osteoarthritis. Technol Health Care, 6(5-6):361-372.

Kaplan, R.N., Psaila, B., Lyden, D. (2006a). Bone marrow cells in the 'pre-metastatic niche': within bone and beyond. Cancer Metastasis Rev, 25(4):521-529.

Kaplan, R.N., Rafii, S., Lyden, D. (2006b). Preparing the "soil": the premetastatic niche. Cancer research, 66(23):11089-11093.

Kassem, M., Abdallah, B.M., Saeed, H. (2008). Osteoblastic cells: differentiation and trans-differentiation. Arch Biochem Biophys, 473(2):183-187.

Katragadda, M., Maciejewski, M.W., Yeagle, P.L. (2004). Structural studies of the putative helix 8 in the human beta(2) adrenergic receptor: an NMR study. Biochim Biophys Acta, 1663(1-2):74-81.

Kawabata, A., Kanke, T., Yonezawa, D., Ishiki, T., Saka, M., Kabeya, M., Sekiguchi, F., Kubo, S., Kuroda, R., Iwaki, M., Katsura, K., Plevin, R. (2004). Potent and metabolically stable agonists for protease-activated receptor-2: evaluation of activity in multiple assay systems in vitro and in vivo. J Pharmacol Exp Ther, 309(3):1098-1107.

Kawabata, A., Kuroda, R., Hollenberg, M.D. (1999). [Physiology of protease- activated receptors (PARs): involvement of PARs in digestive functions]. Nihon Yakurigaku Zasshi, 114 Suppl 1(173P-179P).

Keller, E.T., Brown, J. (2004). Prostate cancer bone metastases promote both osteolytic and osteoblastic activity. J Cell Biochem, 91(4):718-729.

Kelso, E.B., Lockhart, J.C., Hembrough, T., Dunning, L., Plevin, R., Hollenberg, M.D., Sommerhoff, C.P., McLean, J.S., Ferrell, W.R. (2006). Therapeutic promise of proteinase-activated receptor-2 antagonism in joint inflammation. J Pharmacol Exp Ther, 316(3):1017-1024.

Kingsley, L.A., Fournier, P.G., Chirgwin, J.M., Guise, T.A. (2007). Molecular biology of bone metastasis. Mol Cancer Ther, 6(10):2609-2617.

Klarenbach, S.W., Chipiuk, A., Nelson, R.C., Hollenberg, M.D., Murray, A.G. (2003). Differential actions of PAR2 and PAR1 in stimulating human endothelial cell exocytosis and permeability: the role of Rho-GTPases. Circ Res, 92(3):272-278.

199 Kobilka, B., Schertler, G.F. (2008). New G-protein-coupled receptor crystal structures: insights and limitations. Trends Pharmacol Sci, 29(2):79-83.

Konner, J.A., Bell-McGuinn, K.M., Sabbatini, P., Hensley, M.L., Tew, W.P., Pandit- Taskar, N., Vander Els, N., Phillips, M.D., Schweizer, C., Weil, S.C., Larson, S.M., Old, L.J. (2010). Farletuzumab, a humanized monoclonal antibody against folate receptor alpha, in epithelial ovarian cancer: a phase I study. Clin Cancer Res, 16(21):5288-5295.

Koreckij, T., Nguyen, H., Brown, L.G., Yu, E.Y., Vessella, R.L., Corey, E. (2009). Dasatinib inhibits the growth of prostate cancer in bone and provides additional protection from osteolysis. Br J Cancer, 101(2):263-268.

Koutsilieris, M. (1993). Osteoblastic metastasis in advanced prostate cancer. Anticancer Res, 13(2):443-449.

Kraft, K., Olbrich, H., Majoul, I., Mack, M., Proudfoot, A., Oppermann, M. (2001). Characterization of sequence determinants within the carboxyl-terminal domain of chemokine receptor CCR5 that regulate signaling and receptor internalization. J Biol Chem, 276(37):34408-34418.

Kristiansen, K. (2004). Molecular mechanisms of ligand binding, signaling, and regulation within the superfamily of G-protein-coupled receptors: molecular modeling and mutagenesis approaches to receptor structure and function. Pharmacol Ther, 103(1):21-80.

Krupnick, J.G., Benovic, J.L. (1998). The role of receptor kinases and arrestins in G protein-coupled receptor regulation. Annu Rev Pharmacol Toxicol, 38(289- 319.

Kundra, V., Ng, C.S., Ma, J., Bankson, J.A., Price, R.E., Cody, D.D., Do, K.A., Han, L., Navone, N.M. (2007). In vivo imaging of prostate cancer involving bone in a mouse model. Prostate, 67(1):50-60.

Kurth, A.A., Muller, R. (2001). The effect of an osteolytic tumor on the three- dimensional trabecular bone morphology in an animal model. Skeletal Radiol, 30(2):94-98.

Labbe-Jullie, C., Barroso, S., Nicolas-Eteve, D., Reversat, J.L., Botto, J.M., Mazella, J., Bernassau, J.M., Kitabgi, P. (1998). Mutagenesis and modeling of the neurotensin receptor NTR1. Identification of residues that are critical for binding SR 48692, a nonpeptide neurotensin antagonist. J Biol Chem, 273(26):16351-16357.

Landry, Y., Gies, J.P. (2008). Drugs and their molecular targets: an updated overview. Fundam Clin Pharmacol, 22(1):1-18.

Larsson, K., Hofstrom, C., Lindskog, C., Hansson, M., Angelidou, P., Hokfelt, T., Uhlen, M., Wernerus, H., Graslund, T., Hober, S. (2011). Novel antigen

200 design for the generation of antibodies to G-protein-coupled receptors. J Immunol Meth, 370(1-2):14-23.

Lebon, G., Bennett, K., Jazayeri, A., Tate, C.G. (2011). Thermostabilisation of an agonist-bound conformation of the human adenosine A(2A) receptor. J Mol Biol, 409(3):298-310.

Lebon, G., Tate, C.G. (2011). [Structure of the adenosine-bound conformation of the human adenosine A(2A) receptor]. Med Sci (Paris), 27(11):926-928.

Lee, S.L., Dickson, R.B., Lin, C.Y. (2000). Activation of hepatocyte growth factor and urokinase/plasminogen activator by matriptase, an epithelial membrane serine protease. J Biol Chem, 275(47):36720-36725.

Leger, A.J., Jacques, S.L., Badar, J., Kaneider, N.C., Derian, C.K., Andrade-Gordon, P., Covic, L., Kuliopulos, A. (2006). Blocking the protease-activated receptor 1-4 heterodimer in platelet-mediated thrombosis. Circulation, 113(9):1244- 1254.

Legier, J.F., Tauber, L.N. (1968). Solitary metastasis of occult prostatic carcinoma simulating osteogenic sarcoma. Cancer, 22(1):168-172.

Lehrer, H.Z., Maxfield, W.S., Nice, C.M. (1970). The periosteal "sunburst" pattern in metastatic bone tumors. Am J Roentgenol Radium Ther Nucl Med, 108(1):154-161.

Lenardo, M.J., Baltimore, D. (1989). NF-kappa B: a pleiotropic mediator of inducible and tissue-specific gene control. Cell, 58(2):227-229.

Lerner, D.J., Chen, M., Tram, T., Coughlin, S.R. (1996). Agonist recognition by proteinase-activated receptor 2 and thrombin receptor. Importance of extracellular loop interactions for receptor function. J Biol Chem, 271(24):13943-13947.

Lim, H.D., Jongejan, A., Bakker, R.A., Haaksma, E., de Esch, I.J., Leurs, R. (2008). Phenylalanine 169 in the second extracellular loop of the human histamine H4 receptor is responsible for the difference in agonist binding between human and mouse H4 receptors. J Pharmacol Exp Ther, 327(1):88-96.

Linder, M.E., Deschenes, R.J. (2007). Palmitoylation: policing protein stability and traffic. Nat Rev Mol Cell Biol, 8(1):74-84.

Lindner, J.R., Kahn, M.L., Coughlin, S.R., Sambrano, G.R., Schauble, E., Bernstein, D., Foy, D., Hafezi-Moghadam, A., Ley, K. (2000). Delayed onset of inflammation in protease-activated receptor-2-deficient mice. J Immunol, 165(11):6504-6510.

Liu, Z., Geboes, K., Colpaert, S., Overbergh, L., Mathieu, C., Heremans, H., de Boer, M., Boon, L., D'Haens, G., Rutgeerts, P., Ceuppens, J.L. (2000). Prevention of experimental colitis in SCID mice reconstituted with

201 CD45RBhigh CD4+ T cells by blocking the CD40-CD154 interactions. J Immunol, 164(11):6005-6014.

Loew, D., Perrault, C., Morales, M., Moog, S., Ravanat, C., Schuhler, S., Arcone, R., Pietropaolo, C., Cazenave, J.P., van Dorsselaer, A., Lanza, F. (2000). Proteolysis of the exodomain of recombinant protease-activated receptors: prediction of receptor activation or inactivation by MALDI mass spectrometry. Biochemistry, 39(35):10812-10822.

Logothetis, C.J., Lin, S.H. (2005). Osteoblasts in prostate cancer metastasis to bone. Nat Rev Cancer, 5(1):21-28.

Lohman, R.J., Cotterell, A.J., Suen, J., Liu, L., Do, A.T., Vesey, D.A., Fairlie, D.P. (2012). Antagonism of Protease-Activated Receptor 2 Protects against Experimental Colitis. J Pharmacol Exp Ther, 340(2):256-265.

Lokeshwar, B.L. (1999). MMP inhibition in prostate cancer. Ann N Y Acad Sci, 878(271-289.

Lombardo, L.J., Lee, F.Y., Chen, P., Norris, D., Barrish, J.C., Behnia, K., Castaneda, S., Cornelius, L.A., Das, J., Doweyko, A.M., Fairchild, C., Hunt, J.T., Inigo, I., Johnston, K., Kamath, A., Kan, D., Klei, H., Marathe, P., Pang, S., Peterson, R., Pitt, S., Schieven, G.L., Schmidt, R.J., Tokarski, J., Wen, M.L., Wityak, J., Borzilleri, R.M. (2004). Discovery of N-(2-chloro-6-methyl- phenyl)-2-(6-(4-(2-hydroxyethyl)- piperazin-1-yl)-2-methylpyrimidin-4- ylamino)thiazole-5-carboxamide (BMS-354825), a dual Src/Abl kinase inhibitor with potent antitumor activity in preclinical assays. J Med Chem, 47(27):6658-6661.

Lu, X., Bartfai, T. (2009). Analyzing the validity of GalR1 and GalR2 antibodies using knockout mice. Naunyn Schmiedebergs Arch Pharmacol, 379(4):417- 420.

Lu, X., Huang, W., Worthington, S., Drabik, P., Osman, R., Gershengorn, M.C. (2004). A model of inverse agonist action at thyrotropin-releasing hormone receptor type 1: role of a conserved tryptophan in helix 6. Mol Pharmacol, 66(5):1192-1200.

Lu, Z.L., Saldanha, J.W., Hulme, E.C. (2002). Seven-transmembrane receptors: crystals clarify. Trends Pharmacol Sci, 23(3):140-146.

Luo, W., Wang, Y., Hanck, T., Stricker, R., Reiser, G. (2006). Jab1, a novel protease-activated receptor-2 (PAR-2)-interacting protein, is involved in PAR-2-induced activation of activator protein-1. J Biol Chem, 281(12):7927- 7936.

Luo, W., Wang, Y., Reiser, G. (2007). p24A, a type I transmembrane protein, controls ARF1-dependent resensitization of protease-activated receptor-2 by influence on receptor trafficking. J Biol Chem, 282(41):30246-30255.

202 Macfarlane, S.R., Seatter, M.J., Kanke, T., Hunter, G.D., Plevin, R. (2001). Proteinase-activated receptors. Pharmacol Rev, 53(2):245-282.

Mackie, E.J., Loh, L.H., Sivagurunathan, S., Uaesoontrachoon, K., Yoo, H.J., Wong, D., Georgy, S.R., Pagel, C.N. (2008). Protease-activated receptors in the musculoskeletal system. Int J Biochem Cell Biol, 40(6-7):1169-1184.

Maeda, A., Okano, K., Park, P.S., Lem, J., Crouch, R.K., Maeda, T., Palczewski, K. (2010). Palmitoylation stabilizes unliganded rod opsin. Proc Natl Acad Sci U S A, 107(18):8428-8433.

Magee, A.I., Courtneidge, S.A. (1985). Two classes of fatty acid acylated proteins exist in eukaryotic cells. EMBO J, 4(5):1137-1144.

Magklara, A., Mellati, A.A., Wasney, G.A., Little, S.P., Sotiropoulou, G., Becker, G.W., Diamandis, E.P. (2003). Characterization of the enzymatic activity of human kallikrein 6: Autoactivation, substrate specificity, and regulation by inhibitors. Biochem Biophys Res Commun, 307(4):948-955.

Mannowetz, N., Wurdinger, R., Zippel, A., Aumuller, G., Wennemuth, G. (2010). Expression of proteinase-activated receptor-2 (PAR2) is androgen-dependent in stromal cell line (hPCPs) from benign prostatic hyperplasia. Prostate, 70(12):1350-1358.

Mantyh, P.W. (2004). A mechanism-based understanding of bone cancer pain. Novartis Found Symp, 261(194-214; discussion 214-199, 256-161.

Mantyh, P.W., Hunt, S.P. (2004). Mechanisms that generate and maintain bone cancer pain. Novartis Found Symp, 260(221-238; discussion 238-240, 277- 229.

Mari, B., Guerin, S., Far, D.F., Breitmayer, J.P., Belhacene, N., Peyron, J.F., Rossi, B., Auberger, P. (1996). Thrombin and trypsin-induced Ca(2+) mobilization in human T cell lines through interaction with different protease-activated receptors. FASEB J, 10(2):309-316.

Marin, E.P., Krishna, A.G., Zvyaga, T.A., Isele, J., Siebert, F., Sakmar, T.P. (2000). The amino terminus of the fourth cytoplasmic loop of rhodopsin modulates rhodopsin-transducin interaction. J Biol Chem, 275(3):1930-1936.

Martin, B.R., Cravatt, B.F. (2009). Large-scale profiling of protein palmitoylation in mammalian cells. Nat Methods, 6(2):135-138.

Maryanoff, B.E., Santulli, R.J., McComsey, D.F., Hoekstra, W.J., Hoey, K., Smith, C.E., Addo, M., Darrow, A.L., Andrade-Gordon, P. (2001). Protease- activated receptor-2 (PAR-2): structure-function study of receptor activation by diverse peptides related to tethered-ligand epitopes. Arch Biochem Biophys, 386(2):195-204.

203 McDougall, J.J., Zhang, C., Cellars, L., Joubert, E., Dixon, C.M., Vergnolle, N. (2009). Triggering of proteinase-activated receptor 4 leads to joint pain and inflammation in mice. Arthritis Rheum, 60(3):728-737.

McGuire, J.J., Dai, J., Andrade-Gordon, P., Triggle, C.R., Hollenberg, M.D. (2002). Proteinase-activated receptor-2 (PAR2): vascular effects of a PAR2-derived activating peptide via a receptor different than PAR2. J Pharmacol Exp Ther, 303(3):985-992.

McGuire, J.J., Saifeddine, M., Triggle, C.R., Sun, K., Hollenberg, M.D. (2004). 2- furoyl-LIGRLO-amide: a potent and selective proteinase-activated receptor 2 agonist. J Pharmacol Exp Ther, 309(3):1124-1131.

Michel, M.C., Wieland, T., Tsujimoto, G. (2009). How reliable are G-protein- coupled receptor antibodies? Naunyn Schmiedebergs Arch Pharmacol, 379(4):385-388.

Miki, M., Nakamura, Y., Takahashi, A., Nakaya, Y., Eguchi, H., Masegi, T., Yoneda, K., Yasuoka, S., Sone, S. (2003). Effect of human airway trypsin- like protease on intracellular free Ca2+ concentration in human bronchial epithelial cells. J Med Invest, 50(1-2):95-107.

Miyauchi, S., Hirasawa, A., Iga, T., Liu, N., Itsubo, C., Sadakane, K., Hara, T., Tsujimoto, G. (2009). Distribution and regulation of protein expression of the free fatty acid receptor GPR120. Naunyn-Schmiedeberg's archives of pharmacology, 379(4):427-434.

Mize, G.J., Wang, W., Takayama, T.K. (2008). Prostate-specific kallikreins-2 and -4 enhance the proliferation of DU-145 prostate cancer cells through protease- activated receptors-1 and -2. Mol Cancer Res, 6(6):1043-1051.

Moffett, S., Mouillac, B., Bonin, H., Bouvier, M. (1993). Altered phosphorylation and desensitization patterns of a human beta 2-adrenergic receptor lacking the palmitoylated Cys341. EMBO J, 12(1):349-356.

Molino, M., Barnathan, E.S., Numerof, R., Clark, J., Dreyer, M., Cumashi, A., Hoxie, J.A., Schechter, N., Woolkalis, M., Brass, L.F. (1997). Interactions of mast cell tryptase with thrombin receptors and PAR-2. J Biol Chem, 272(7):4043-4049.

Moretti, S., Bellocchio, S., Bonifazi, P., Bozza, S., Zelante, T., Bistoni, F., Romani, L. (2008). The contribution of PARs to inflammation and immunity to fungi. Mucosal Immunol, 1(2):156-168.

Morimoto, S., Nishimura, N., Terai, T., Manabe, S., Yamamoto, Y., Shinahara, W., Miyake, H., Tashiro, S., Shimada, M., Sasaki, T. (2005). Rab13 mediates the continuous endocytic recycling of occludin to the cell surface. J Biol Chem, 280(3):2220-2228.

204 Moro, S., Deflorian, F., Bacilieri, M., Spalluto, G. (2006). Ligand-based homology modeling as attractive tool to inspect GPCR structural plasticity. Curr Pharm Des, 12(17):2175-2185.

Moro, S., Deflorian, F., Spalluto, G., Pastorin, G., Cacciari, B., Kim, S.K., Jacobson, K.A. (2003). Demystifying the three dimensional structure of G protein- coupled receptors (GPCRs) with the aid of molecular modeling. Chem Commun (Camb), 24):2949-2956.

Morris, D.R., Ding, Y., Ricks, T.K., Gullapalli, A., Wolfe, B.L., Trejo, J. (2006). Protease-activated receptor-2 is essential for factor VIIa and Xa-induced signaling, migration, and invasion of breast cancer cells. Cancer Res, 66(1):307-314.

Mukai, J., Dhilla, A., Drew, L.J., Stark, K.L., Cao, L., MacDermott, A.B., Karayiorgou, M., Gogos, J.A. (2008). Palmitoylation-dependent neurodevelopmental deficits in a mouse model of 22q11 microdeletion. Nat Neurosci, 11(11):1302-1310.

Muller, R., Van Campenhout, H., Van Damme, B., Van Der Perre, G., Dequeker, J., Hildebrand, T., Ruegsegger, P. (1998). Morphometric analysis of human bone biopsies: a quantitative structural comparison of histological sections and micro-computed tomography. Bone, 23(1):59-66.

Mundy, G.R. (2002). Metastasis to bone: causes, consequences and therapeutic opportunities. Nat Rev Cancer, 2(8):584-593.

Murakami, M., Kouyama, T. (2011). Crystallographic analysis of the primary photochemical reaction of squid rhodopsin. J Mol Biol, 413(3):615-627.

Najy, A.J., Jung, Y.S., Won, J.J., Conley-Lacomb, M.K., Saliganan, A., Kim, C.J., Heath, E., Cher, M.L., Bonfil, R.D., Kim, H.R. (2011). Cediranib inhibits both the intraosseous growth of PDGF D-Positive prostate cancer cells and the associated bone reaction. Prostate, In press.

Navone, N.M., Logothetis, C.J., von Eschenbach, A.C., Troncoso, P. (1998). Model systems of prostate cancer: uses and limitations. Cancer Metastasis Rev, 17(4):361-371.

Navone, N.M., Olive, M., Ozen, M., Davis, R., Troncoso, P., Tu, S.M., Johnston, D., Pollack, A., Pathak, S., von Eschenbach, A.C., Logothetis, C.J. (1997). Establishment of two human prostate cancer cell lines derived from a single bone metastasis. Clin Cancer Res, 3(12 Pt 1):2493-2500.

Nelson, J.B., Nguyen, S.H., Wu-Wong, J.R., Opgenorth, T.J., Dixon, D.B., Chung, L.W., Inoue, N. (1999). New bone formation in an osteoblastic tumor model is increased by endothelin-1 overexpression and decreased by endothelin A receptor blockade. Urology, 53(5):1063-1069.

205 Nemeth, J.A., Harb, J.F., Barroso, U., Jr., He, Z., Grignon, D.J., Cher, M.L. (1999). Severe combined immunodeficient-hu model of human prostate cancer metastasis to human bone. Cancer research, 59(8):1987-1993.

Nepveu, A. (2001). Role of the multifunctional CDP/Cut/Cux homeodomain transcription factor in regulating differentiation, cell growth and development. Gene, 270(1-2):1-15.

Ng, G.Y., O'Dowd, B.F., Caron, M., Dennis, M., Brann, M.R., George, S.R. (1994). Phosphorylation and palmitoylation of the human D2L dopamine receptor in Sf9 cells. J Neurochem, 63(5):1589-1595.

Nhu, Q.M., Shirey, K., Teijaro, J.R., Farber, D.L., Netzel-Arnett, S., Antalis, T.M., Fasano, A., Vogel, S.N. (2010). Novel signaling interactions between proteinase-activated receptor 2 and Toll-like receptors in vitro and in vivo. Mucosal Immunol, 3(1):29-39.

Nierodzik, M.L., Chen, K., Takeshita, K., Li, J.J., Huang, Y.Q., Feng, X.S., D'Andrea, M.R., Andrade-Gordon, P., Karpatkin, S. (1998). Protease- activated receptor 1 (PAR-1) is required and rate-limiting for thrombin- enhanced experimental pulmonary metastasis. Blood, 92(10):3694-3700.

Nieto, M., Finn, S., Loda, M., Hahn, W.C. (2007). Prostate cancer: Re-focusing on androgen receptor signaling. Int J Biochem Cell Biol, 39(9):1562-1568.

Niu, Q.X., Chen, H.Q., Chen, Z.Y., Fu, Y.L., Lin, J.L., He, S.H. (2008a). Induction of inflammatory cytokine release from human umbilical vein endothelial cells by agonists of proteinase-activated receptor-2. Clin Exp Pharmacol Physiol, 35(1):89-96.

Niu, Y., Altuwaijri, S., Lai, K.P., Wu, C.T., Ricke, W.A., Messing, E.M., Yao, J., Yeh, S., Chang, C. (2008b). Androgen receptor is a tumor suppressor and proliferator in prostate cancer. Proc Natl Acad Sci U S A, 105(34):12182- 12187.

Niu, Y., Yeh, S., Miyamoto, H., Li, G., Altuwaijri, S., Yuan, J., Han, R., Ma, T., Kuo, H.C., Chang, C. (2008c). Tissue prostate-specific antigen facilitates refractory prostate tumor progression via enhancing ARA70-regulated androgen receptor transactivation. Cancer research, 68(17):7110-7119.

Nystedt, S., Emilsson, K., Larsson, A.K., Strombeck, B., Sundelin, J. (1995). Molecular cloning and functional expression of the gene encoding the human proteinase-activated receptor 2. Eur J Biochem, 232(1):84-89.

Nystedt, S., Emilsson, K., Wahlestedt, C., Sundelin, J. (1994). Molecular cloning of a potential proteinase activated receptor. Proc Natl Acad Sci U S A, 91(20):9208-9212.

O'Dowd, B.F., Hnatowich, M., Caron, M.G., Lefkowitz, R.J., Bouvier, M. (1989). Palmitoylation of the human beta 2-adrenergic receptor. Mutation of Cys341

206 in the carboxyl tail leads to an uncoupled nonpalmitoylated form of the receptor. J Biol Chem, 264(13):7564-7569.

O'Gorman, S., Fox, D.T., Wahl, G.M. (1991). Recombinase-mediated gene activation and site-specific integration in mammalian cells. Science, 251(4999):1351-1355.

O'Neill, K.R., Stutz, C.M., Mignemi, N.A., Cole, H., Murry, M.R., Nyman, J.S., Hamm, H., Schoenecker, J.G. (2012). Fracture healing in protease-activated receptor-2 deficient mice. J Orthop Res, In press.

Oakley, R.H., Laporte, S.A., Holt, J.A., Barak, L.S., Caron, M.G. (2001). Molecular determinants underlying the formation of stable intracellular G protein- coupled receptor-beta-arrestin complexes after receptor endocytosis*. J Biol Chem, 276(22):19452-19460.

Obiezu, C.V., Michael, I.P., Levesque, M.A., Diamandis, E.P. (2006). Human kallikrein 4: enzymatic activity, inhibition, and degradation of extracellular matrix proteins. Biol Chem, 387(6):749-759.

Ohno, Y., Kihara, A., Sano, T., Igarashi, Y. (2006). Intracellular localization and tissue-specific distribution of human and yeast DHHC cysteine-rich domain- containing proteins. Biochim Biophys Acta, 1761(4):474-483.

Oikonomopoulou, K., Hansen, K.K., Saifeddine, M., Tea, I., Blaber, M., Blaber, S.I., Scarisbrick, I., Andrade-Gordon, P., Cottrell, G.S., Bunnett, N.W., Diamandis, E.P., Hollenberg, M.D. (2006). Proteinase-activated receptors, targets for kallikrein signaling. J Biol Chem, 281(43):32095-32112.

Oldham, W.M., Hamm, H.E. (2008). Heterotrimeric G protein activation by G- protein-coupled receptors. Nat Rev Mol Cell Biol, 9(1):60-71.

Olianas, M.C., Dedoni, S., Onali, P. (2007). Proteinase-activated receptors 1 and 2 in rat olfactory system: layer-specific regulation of multiple signaling pathways in the main olfactory bulb and induction of neurite retraction in olfactory sensory neurons. Neuroscience, 146(3):1289-1301.

Ossovskaya, V.S., Bunnett, N.W. (2004). Protease-activated receptors: contribution to physiology and disease. Physiol Rev, 84(2):579-621.

Overington, J.P., Al-Lazikani, B., Hopkins, A.L. (2006). How many drug targets are there? Nature reviews. Drug discovery, 5(12):993-996.

Paing, M.M., Johnston, C.A., Siderovski, D.P., Trejo, J. (2006). Clathrin adaptor AP2 regulates thrombin receptor constitutive internalization and endothelial cell resensitization. Mol Cell Biol, 26(8):3231-3242.

Palczewski, K., Kumasaka, T., Hori, T., Behnke, C.A., Motoshima, H., Fox, B.A., Le Trong, I., Teller, D.C., Okada, T., Stenkamp, R.E., Yamamoto, M., Miyano,

207 M. (2000). Crystal structure of rhodopsin: A G protein-coupled receptor. Science, 289(5480):739-745.

Parent, A., Hamelin, E., Germain, P., Parent, J.L. (2009). Rab11 regulates the recycling of the beta2-adrenergic receptor through a direct interaction. Biochem J, 418(1):163-172.

Parr, C., Watkins, G., Mansel, R.E., Jiang, W.G. (2004). The hepatocyte growth factor regulatory factors in human breast cancer. Clin Cancer Res, 10(1 Pt 1):202-211.

Patny, A., Desai, P.V., Avery, M.A. (2006). Ligand-supported homology modeling of the human angiotensin II type 1 (AT(1)) receptor: insights into the molecular determinants of telmisartan binding. Proteins, 65(4):824-842.

Percherancier, Y., Planchenault, T., Valenzuela-Fernandez, A., Virelizier, J.L., Arenzana-Seisdedos, F., Bachelerie, F. (2001). Palmitoylation-dependent control of degradation, life span, and membrane expression of the CCR5 receptor. J Biol Chem, 276(34):31936-31944.

Petaja-Repo, U.E., Hogue, M., Leskela, T.T., Markkanen, P.M., Tuusa, J.T., Bouvier, M. (2006). Distinct subcellular localization for constitutive and agonist-modulated palmitoylation of the human delta opioid receptor. J Biol Chem, 281(23):15780-15789.

Petrylak, D.P., Tangen, C.M., Hussain, M.H., Lara, P.N., Jr., Jones, J.A., Taplin, M.E., Burch, P.A., Berry, D., Moinpour, C., Kohli, M., Benson, M.C., Small, E.J., Raghavan, D., Crawford, E.D. (2004). Docetaxel and estramustine compared with mitoxantrone and prednisone for advanced refractory prostate cancer. N Engl J Med, 351(15):1513-1520.

Phillips, W.J., Cerione, R.A. (1992). Rhodopsin/transducin interactions. I. Characterization of the binding of the transducin-beta gamma subunit complex to rhodopsin using fluorescence spectroscopy. J Biol Chem, 267(24):17032-17039.

Piscitelli, C.L., Angel, T.E., Bailey, B.W., Hargrave, P., Dratz, E.A., Lawrence, C.M. (2006). Equilibrium between metarhodopsin-I and metarhodopsin-II is dependent on the conformation of the third cytoplasmic loop. J Biol Chem, 281(10):6813-6825.

Ponimaskin, E., Dumuis, A., Gaven, F., Barthet, G., Heine, M., Glebov, K., Richter, D.W., Oppermann, M. (2005). Palmitoylation of the 5-hydroxytryptamine4a receptor regulates receptor phosphorylation, desensitization, and beta- arrestin-mediated endocytosis. Mol Pharmacol, 67(5):1434-1443.

Pradidarcheep, W., Stallen, J., Labruyere, W.T., Dabhoiwala, N.F., Michel, M.C., Lamers, W.H. (2009). Lack of specificity of commercially available antisera against muscarinergic and adrenergic receptors. Naunyn Schmiedebergs Arch Pharmacol, 379(4):397-402.

208

Probst, W.C., Snyder, L.A., Schuster, D.I., Brosius, J., Sealfon, S.C. (1992). Sequence alignment of the G-protein coupled receptor superfamily. DNA Cell Biol, 11(1):1-20.

Qanbar, R., Bouvier, M. (2003). Role of palmitoylation/depalmitoylation reactions in G-protein-coupled receptor function. Pharmacol Ther, 97(1):1-33.

Radestock, S., Weil, T., Renner, S. (2008). Homology model-based virtual screening for GPCR ligands using docking and target-biased scoring. J Chem Inf Model, 48(5):1104-1117.

Ragsdale, B.D., Madewell, J.E., Sweet, D.E. (1981). Radiologic and pathologic analysis of solitary bone lesions. Part II: periosteal reactions. Radiol Clin North Am, 19(4):749-783.

Rallabhandi, P., Nhu, Q.M., Toshchakov, V.Y., Piao, W., Medvedev, A.E., Hollenberg, M.D., Fasano, A., Vogel, S.N. (2008). Analysis of proteinase- activated receptor 2 and TLR4 signal transduction: a novel paradigm for receptor cooperativity. J Biol Chem, 283(36):24314-24325.

Ramachandran, R., Hollenberg, M.D. (2008). Proteinases and signalling: pathophysiological and therapeutic implications via PARs and more. Br J Pharmacol, 153 Suppl 1(S263-282.

Ramachandran, R., Mihara, K., Mathur, M., Rochdi, M.D., Bouvier, M., Defea, K., Hollenberg, M.D. (2009). Agonist-biased signaling via proteinase activated receptor-2: differential activation of calcium and mitogen-activated protein kinase pathways. Mol Pharmacol, 76(4):791-801.

Ramsay, A.J., Dong, Y., Hunt, M.L., Linn, M., Samaratunga, H., Clements, J.A., Hooper, J.D. (2008a). Kallikrein-related peptidase 4 (KLK4) initiates intracellular signaling via protease-activated receptors (PARs). KLK4 and PAR-2 are co-expressed during prostate cancer progression. J Biol Chem, 283(18):12293-12304.

Ramsay, A.J., Reid, J.C., Adams, M.N., Samaratunga, H., Dong, Y., Clements, J.A., Hooper, J.D. (2008b). Prostatic trypsin-like kallikrein-related peptidases (KLKs) and other prostate-expressed tryptic proteinases as regulators of signalling via proteinase-activated receptors (PARs). Biol Chem, 389(6):653- 668.

Rasmussen, S.G., Choi, H.J., Fung, J.J., Pardon, E., Casarosa, P., Chae, P.S., Devree, B.T., Rosenbaum, D.M., Thian, F.S., Kobilka, T.S., Schnapp, A., Konetzki, I., Sunahara, R.K., Gellman, S.H., Pautsch, A., Steyaert, J., Weis, W.I., Kobilka, B.K. (2011a). Structure of a nanobody-stabilized active state of the beta(2) adrenoceptor. Nature, 469(7329):175-180.

Rasmussen, S.G., Choi, H.J., Rosenbaum, D.M., Kobilka, T.S., Thian, F.S., Edwards, P.C., Burghammer, M., Ratnala, V.R., Sanishvili, R., Fischetti, R.F.,

209 Schertler, G.F., Weis, W.I., Kobilka, B.K. (2007). Crystal structure of the human beta2 adrenergic G-protein-coupled receptor. Nature, 450(7168):383- 387.

Rasmussen, S.G., DeVree, B.T., Zou, Y., Kruse, A.C., Chung, K.Y., Kobilka, T.S., Thian, F.S., Chae, P.S., Pardon, E., Calinski, D., Mathiesen, J.M., Shah, S.T., Lyons, J.A., Caffrey, M., Gellman, S.H., Steyaert, J., Skiniotis, G., Weis, W.I., Sunahara, R.K., Kobilka, B.K. (2011b). Crystal structure of the beta2 adrenergic receptor-Gs protein complex. Nature, 477(7366):549-555.

Rattenholl, A., Seeliger, S., Buddenkotte, J., Schon, M., Schon, M.P., Stander, S., Vergnolle, N., Steinhoff, M. (2007). Proteinase-activated receptor-2 (PAR2): a tumor suppressor in skin carcinogenesis. J Invest Dermatol, 127(9):2245- 2252.

Ravoori, M., Czaplinska, A.J., Sikes, C., Han, L., Johnson, E.M., Qiao, W., Ng, C., Cody, D.D., Murphy, W.A., Do, K.A., Navone, N.M., Kundra, V. (2010). Quantification of mineralized bone response to prostate cancer by noninvasive in vivo microCT and non-destructive ex vivo microCT and DXA in a mouse model. PLoS One, 5(3):e9854.

Reggio, P.H. (2006). Computational methods in drug design: modeling G protein- coupled receptor monomers, dimers, and oligomers. Aaps J, 8(2):E322-336.

Reid, H.M., Mulvaney, E.P., Turner, E.C., Kinsella, B.T. (2010). Interaction of the human prostacyclin receptor with Rab11: characterization of a novel Rab11 binding domain within alpha-helix 8 that is regulated by palmitoylation. J Biol Chem, 285(24):18709-18726.

Ren, C., Kumar, S., Chanda, D., Kallman, L., Chen, J., Mountz, J.D., Ponnazhagan, S. (2008). Cancer gene therapy using mesenchymal stem cells expressing interferon-beta in a mouse prostate cancer lung metastasis model. Gene Ther, 15(21):1446-1453.

Renner, S., Noeske, T., Parsons, C.G., Schneider, P., Weil, T., Schneider, G. (2005). New allosteric modulators of metabotropic glutamate receptor 5 (mGluR5) found by ligand-based virtual screening. Chembiochem, 6(4):620-625.

Ricks, T.K., Trejo, J. (2009). Phosphorylation of protease-activated receptor-2 differentially regulates desensitization and internalization. J Biol Chem, 284(49):34444-34457.

Ritter, S.L., Hall, R.A. (2009). Fine-tuning of GPCR activity by receptor-interacting proteins. Nat Rev Mol Cell Biol, 10(12):819-830.

Roodman, G.D. (2004). Mechanisms of bone metastasis. N Engl J Med, 350(16):1655-1664.

210 Roosterman, D., Schmidlin, F., Bunnett, N.W. (2003). Rab5a and rab11a mediate agonist-induced trafficking of protease-activated receptor 2. Am J Physiol, 284(5):C1319-1329.

Rosenbaum, D.M., Rasmussen, S.G., Kobilka, B.K. (2009). The structure and function of G-protein-coupled receptors. Nature, 459(7245):356-363.

Rosenbaum, D.M., Zhang, C., Lyons, J.A., Holl, R., Aragao, D., Arlow, D.H., Rasmussen, S.G., Choi, H.J., Devree, B.T., Sunahara, R.K., Chae, P.S., Gellman, S.H., Dror, R.O., Shaw, D.E., Weis, W.I., Caffrey, M., Gmeiner, P., Kobilka, B.K. (2011). Structure and function of an irreversible agonist- beta(2) adrenoceptor complex. Nature, 469(7329):236-240.

Rosol, T.J., Tannehill-Gregg, S.H., LeRoy, B.E., Mandl, S., Contag, C.H. (2003). Animal models of bone metastasis. Cancer, 97(3 Suppl):748-757.

Roudier, M.P., Morrissey, C., True, L.D., Higano, C.S., Vessella, R.L., Ott, S.M. (2008). Histopathological assessment of prostate cancer bone osteoblastic metastases. J Urol, 180(3):1154-1160.

Rudack, C., Steinhoff, M., Mooren, F., Buddenkotte, J., Becker, K., von Eiff, C., Sachse, F. (2007). PAR-2 activation regulates IL-8 and GRO-alpha synthesis by NF-kappaB, but not RANTES, IL-6, eotaxin or TARC expression in nasal epithelium. Clin Exp Allergy, 37(7):1009-1022.

Ruf, W., Yokota, N., Schaffner, F. (2010). Tissue factor in cancer progression and angiogenesis. Thrombosis research, 125 Suppl 2(S36-38.

Russo, A., Soh, U.J., Paing, M.M., Arora, P., Trejo, J. (2009). Caveolae are required for protease-selective signaling by protease-activated receptor-1. Proc Natl Acad Sci U S A, 106(15):6393-6397.

Ryden, L., Grabau, D., Schaffner, F., Jonsson, P.E., Ruf, W., Belting, M. (2010). Evidence for tissue factor phosphorylation and its correlation with protease- activated receptor expression and the prognosis of primary breast cancer. Int J Cancer, 126(10):2330-2340.

Rzomp, K.A., Scholtes, L.D., Briggs, B.J., Whittaker, G.R., Scidmore, M.A. (2003). Rab GTPases are recruited to chlamydial inclusions in both a species- dependent and species-independent manner. Infect Immun, 71(10):5855-5870.

Sabri, A., Muske, G., Zhang, H., Pak, E., Darrow, A., Andrade-Gordon, P., Steinberg, S.F. (2000). Signaling properties and functions of two distinct cardiomyocyte protease-activated receptors. Circ Res, 86(10):1054-1061.

Sadeghi, H.M., Innamorati, G., Dagarag, M., Birnbaumer, M. (1997). Palmitoylation of the V2 vasopressin receptor. Mol Pharmacol, 52(1):21-29.

211 Saitoh, H., Hida, M., Shimbo, T., Nakamura, K., Yamagata, J., Satoh, T. (1984). Metastatic patterns of prostatic cancer. Correlation between sites and number of organs involved. Cancer, 54(12):3078-3084.

Salim, K., Fenton, T., Bacha, J., Urien-Rodriguez, H., Bonnert, T., Skynner, H.A., Watts, E., Kerby, J., Heald, A., Beer, M., McAllister, G., Guest, P.C. (2002). Oligomerization of G-protein-coupled receptors shown by selective co- immunoprecipitation. J Biol Chem, 277(18):15482-15485.

Salom, D., Lodowski, D.T., Stenkamp, R.E., Le Trong, I., Golczak, M., Jastrzebska, B., Harris, T., Ballesteros, J.A., Palczewski, K. (2006). Crystal structure of a photoactivated deprotonated intermediate of rhodopsin. Proc Natl Acad Sci U S A, 103(44):16123-16128.

Sansregret, L., Nepveu, A. (2008). The multiple roles of CUX1: insights from mouse models and cell-based assays. Gene, 412(1-2):84-94.

Scarborough, R.M., Naughton, M.A., Teng, W., Hung, D.T., Rose, J., Vu, T.K., Wheaton, V.I., Turck, C.W., Coughlin, S.R. (1992). Tethered ligand agonist peptides. Structural requirements for thrombin receptor activation reveal mechanism of proteolytic unmasking of agonist function. J Biol Chem, 267(19):13146-13149.

Scarselli, M., Li, B., Kim, S.K., Wess, J. (2007). Multiple residues in the second extracellular loop are critical for M3 muscarinic acetylcholine receptor activation. J Biol Chem, 282(10):7385-7396.

Schertler, G.F. (2005). Structure of rhodopsin and the metarhodopsin I photointermediate. Curr Opin Struct Biol, 15(4):408-415.

Schmidlin, F., Amadesi, S., Dabbagh, K., Lewis, D.E., Knott, P., Bunnett, N.W., Gater, P.R., Geppetti, P., Bertrand, C., Stevens, M.E. (2002). Protease- activated receptor 2 mediates eosinophil infiltration and hyperreactivity in allergic inflammation of the airway. J Immunol, 169(9):5315-5321.

Schmidt, C., Priemel, M., Kohler, T., Weusten, A., Muller, R., Amling, M., Eckstein, F. (2003). Precision and accuracy of peripheral quantitative computed tomography (pQCT) in the mouse skeleton compared with histology and microcomputed tomography (microCT). J Bone Miner Res, 18(8):1486-1496.

Schmidt, V.A., Nierman, W.C., Feldblyum, T.V., Maglott, D.R., Bahou, W.F. (1997). The human thrombin receptor and proteinase activated receptor-2 genes are tightly linked on chromosome 5q13. Br J Haematol, 97(3):523-529.

Schneider, A., Kalikin, L.M., Mattos, A.C., Keller, E.T., Allen, M.J., Pienta, K.J., McCauley, L.K. (2005). Bone turnover mediates preferential localization of prostate cancer in the skeleton. Endocrinology, 146(4):1727-1736.

Schulein, R., Liebenhoff, U., Muller, H., Birnbaumer, M., Rosenthal, W. (1996). Properties of the human arginine vasopressin V2 receptor after site-directed

212 mutagenesis of its putative palmitoylation site. Biochem J, 313 ( Pt 2)(611- 616.

Seatter, M.J., Drummond, R., Kanke, T., Macfarlane, S.R., Hollenberg, M.D., Plevin, R. (2004). The role of the C-terminal tail in protease-activated receptor-2- mediated Ca2+ signalling, proline-rich tyrosine kinase-2 activation, and mitogen-activated protein kinase activity. Cell Signal, 16(1):21-29.

Seeman, E., Delmas, P.D. (2006). Bone quality--the material and structural basis of bone strength and fragility. N Engl J Med, 354(21):2250-2261.

Sevigny, L.M., Zhang, P., Bohm, A., Lazarides, K., Perides, G., Covic, L., Kuliopulos, A. (2011). Interdicting protease-activated receptor-2-driven inflammation with cell-penetrating pepducins. Proc Natl Acad Sci U S A, 108(20):8491-8496.

Shaner, N.C., Steinbach, P.A., Tsien, R.Y. (2005). A guide to choosing fluorescent proteins. Nat Methods, 2(12):905-909.

Shapiro, M.J., Coughlin, S.R. (1998). Separate signals for agonist-independent and agonist-triggered trafficking of protease-activated receptor 1. J Biol Chem, 273(44):29009-29014.

Sharma, C., Yang, X.H., Hemler, M.E. (2008). DHHC2 affects palmitoylation, stability, and functions of tetraspanins CD9 and CD151. Mol Biol Cell, 19(8):3415-3425.

Shi, L., Javitch, J.A. (2004). The second extracellular loop of the dopamine D2 receptor lines the binding-site crevice. Proc Natl Acad Sci U S A, 101(2):440- 445.

Shi, X., Gangadharan, B., Brass, L.F., Ruf, W., Mueller, B.M. (2004). Protease- activated receptors (PAR1 and PAR2) contribute to tumor cell motility and metastasis. Molecular cancer research : MCR, 2(7):395-402.

Shimamura, T., Shiroishi, M., Weyand, S., Tsujimoto, H., Winter, G., Katritch, V., Abagyan, R., Cherezov, V., Liu, W., Han, G.W., Kobayashi, T., Stevens, R.C., Iwata, S. (2011). Structure of the human histamine H1 receptor complex with doxepin. Nature, 475(7354):65-70.

Shpacovitch, V.M., Brzoska, T., Buddenkotte, J., Stroh, C., Sommerhoff, C.P., Ansel, J.C., Schulze-Osthoff, K., Bunnett, N.W., Luger, T.A., Steinhoff, M. (2002). Agonists of proteinase-activated receptor 2 induce cytokine release and activation of nuclear transcription factor kappaB in human dermal microvascular endothelial cells. J Invest Dermatol, 118(2):380-385.

Sidera, C., Parsons, R., Austen, B. (2005). Post-translational processing of beta- secretase in Alzheimer's disease. Proteomics, 5(6):1533-1543.

213 Siegel, R., Naishadham, D., Jemal, A. (2012). Cancer statistics, 2012. CA Cancer J Clin, 62(1):10-29.

Simon, M.I., Strathmann, M.P., Gautam, N. (1991). Diversity of G proteins in signal transduction. Science, 252(5007):802-808.

Singaraja, R.R., Huang, K., Sanders, S.S., Milnerwood, A.J., Hines, R., Lerch, J.P., Franciosi, S., Drisdel, R.C., Vaid, K., Young, F.B., Doty, C., Wan, J., Bissada, N., Henkelman, R.M., Green, W.N., Davis, N.G., Raymond, L.A., Hayden, M.R. (2011). Altered palmitoylation and neuropathological deficits in mice lacking HIP14. Hum Mol Genet, 20(20):3899-3909.

Smith-Jones, P.M., Pandit-Taskar, N., Cao, W., O'Donoghue, J., Philips, M.D., Carrasquillo, J., Konner, J.A., Old, L.J., Larson, S.M. (2008). Preclinical radioimmunotargeting of folate receptor alpha using the monoclonal antibody conjugate DOTA-MORAb-003. Nucl Med Biol, 35(3):343-351.

Smith, R., Jenkins, A., Lourbakos, A., Thompson, P., Ramakrishnan, V., Tomlinson, J., Deshpande, U., Johnson, D.A., Jones, R., Mackie, E.J., Pike, R.N. (2000). Evidence for the activation of PAR-2 by the sperm protease, acrosin: expression of the receptor on oocytes. FEBS Lett, 484(3):285-290.

Smith, R., Ransjo, M., Tatarczuch, L., Song, S.J., Pagel, C., Morrison, J.R., Pike, R.N., Mackie, E.J. (2004). Activation of protease-activated receptor-2 leads to inhibition of osteoclast differentiation. J Bone Miner Res, 19(3):507-516.

Soh, U.J., Dores, M.R., Chen, B., Trejo, J. (2010). Signal transduction by protease- activated receptors. Br J Pharmacol, 160(2):191-203.

Sramkoski, R.M., Pretlow, T.G., 2nd, Giaconia, J.M., Pretlow, T.P., Schwartz, S., Sy, M.S., Marengo, S.R., Rhim, J.S., Zhang, D., Jacobberger, J.W. (1999). A new human prostate carcinoma cell line, 22Rv1. In Vitro Cell Dev Biol Anim, 35(7):403-409.

Stalheim, L., Ding, Y., Gullapalli, A., Paing, M.M., Wolfe, B.L., Morris, D.R., Trejo, J. (2005). Multiple independent functions of arrestins in the regulation of protease-activated receptor-2 signaling and trafficking. Mol Pharmacol, 67(1):78-87.

Stanford, J.L., Wicklund, K.G., McKnight, B., Daling, J.R., Brawer, M.K. (1999). Vasectomy and risk of prostate cancer. Cancer Epidemiol Biomarkers Prev, 8(10):881-886.

Stanley, P., Caillibot, V., Siminovitch, L. (1975). Selection and characterization of eight phenotypically distinct lines of lectin-resistant Chinese hamster ovary cell. Cell, 6(2):121-128.

Steinhoff, M., Buddenkotte, J., Shpacovitch, V., Rattenholl, A., Moormann, C., Vergnolle, N., Luger, T.A., Hollenberg, M.D. (2005). Proteinase-activated

214 receptors: transducers of proteinase-mediated signaling in inflammation and immune response. Endocrine Rev, 26(1):1-43.

Stone, K.R., Mickey, D.D., Wunderli, H., Mickey, G.H., Paulson, D.F. (1978). Isolation of a human prostate carcinoma cell line (DU 145). Int J Cancer, 21(3):274-281.

Storrie, B., Yang, W. (1998). Dynamics of the interphase mammalian Golgi complex as revealed through drugs producing reversible Golgi disassembly. Biochim Biophys Acta, 1404(1-2):127-137.

Strader, C.D., Fong, T.M., Tota, M.R., Underwood, D., Dixon, R.A. (1994). Structure and function of G protein-coupled receptors. Annu Rev Biochem, 63(101-132.

Strasser, A., Wittmann, H.J., Seifert, R. (2008). Ligand-specific contribution of the N terminus and E2-loop to pharmacological properties of the histamine H1- receptor. J Pharmacol Exp Ther, 326(3):783-791.

Su, S., Li, Y., Luo, Y., Sheng, Y., Su, Y., Padia, R.N., Pan, Z.K., Dong, Z., Huang, S. (2009). Proteinase-activated receptor 2 expression in breast cancer and its role in breast cancer cell migration. Oncogene, 28(34):3047-3057.

Suen, J.Y., Barry, G.D., Lohman, R.J., Halili, M.A., Cotterell, A.J., Le, G.T., Fairlie, D.P. (2011). Modulating Human Proteinase Activated Receptor 2 With A Novel Antagonist (GB88) and Agonist (GB110). Br J Pharmacol, 165(5):1413-1423.

Sun, G., Stacey, M.A., Schmidt, M., Mori, L., Mattoli, S. (2001). Interaction of mite allergens Der p3 and Der p9 with protease-activated receptor-2 expressed by lung epithelial cells. J Immunol, 167(2):1014-1021.

Sun, Y., McGarrigle, D., Huang, X.Y. (2007). When a G protein-coupled receptor does not couple to a G protein. Molecular bioSystems, 3(12):849-854.

Swift, S., Leger, A.J., Talavera, J., Zhang, L., Bohm, A., Kuliopulos, A. (2006). Role of the PAR1 receptor 8th helix in signaling: the 7-8-1 receptor activation mechanism. J Biol Chem, 281(7):4109-4116.

Szabo, R., Rasmussen, A.L., Moyer, A.B., Kosa, P., Schafer, J.M., Molinolo, A.A., Gutkind, J.S., Bugge, T.H. (2011). c-Met-induced epithelial carcinogenesis is initiated by the serine protease matriptase. Oncogene, 30(17):2003-2016.

Szundi, I., Ruprecht, J.J., Epps, J., Villa, C., Swartz, T.E., Lewis, J.W., Schertler, G.F., Kliger, D.S. (2006). Rhodopsin photointermediates in two-dimensional crystals at physiological temperatures. Biochemistry, 45(15):4974-4982.

Takahashi, M., Murate, M., Fukuda, M., Sato, S.B., Ohta, A., Kobayashi, T. (2007). Cholesterol controls lipid endocytosis through Rab11. Mol Biol Cell, 18(7):2667-2677.

215

Takeuchi, T., Harris, J.L., Huang, W., Yan, K.W., Coughlin, S.R., Craik, C.S. (2000). Cellular localization of membrane-type serine protease 1 and identification of protease-activated receptor-2 and single-chain urokinase- type plasminogen activator as substrates. J Biol Chem, 275(34):26333-26342.

Tammela, T. (2004). Endocrine treatment of prostate cancer. J Steroid Biochem Mol Biol, 92(4):287-295.

Tanaka, K., Nagayama, Y., Nishihara, E., Namba, H., Yamashita, S., Niwa, M. (1998). Palmitoylation of human thyrotropin receptor: slower intracellular trafficking of the palmitoylation-defective mutant. Endocrinology, 139(2):803-806.

Tannock, I.F., de Wit, R., Berry, W.R., Horti, J., Pluzanska, A., Chi, K.N., Oudard, S., Theodore, C., James, N.D., Turesson, I., Rosenthal, M.A., Eisenberger, M.A. (2004). Docetaxel plus prednisone or mitoxantrone plus prednisone for advanced prostate cancer. N Engl J Med, 351(15):1502-1512.

Teller, D.C., Okada, T., Behnke, C.A., Palczewski, K., Stenkamp, R.E. (2001). Advances in determination of a high-resolution three-dimensional structure of rhodopsin, a model of G-protein-coupled receptors (GPCRs). Biochemistry, 40(26):7761-7772.

Thalmann, G.N., Anezinis, P.E., Chang, S.M., Zhau, H.E., Kim, E.E., Hopwood, V.L., Pathak, S., von Eschenbach, A.C., Chung, L.W. (1994). Androgen- independent cancer progression and bone metastasis in the LNCaP model of human prostate cancer. Cancer Res, 54(10):2577-2581.

Titus, B., Schwartz, M.A., Theodorescu, D. (2005). Rho proteins in cell migration and metastasis. Crit Rev Eukaryot Gene Expr, 15(2):103-114.

Tobin, A.B., Wheatley, M. (2004). G-protein-coupled receptor phosphorylation and palmitoylation. Methods Mol Biol, 259(275-281.

Topiol, S., Sabio, M. (2009). X-ray structure breakthroughs in the GPCR transmembrane region. Biochem Pharmacol, 78(1):11-20.

Uehara, A., Muramoto, K., Imamura, T., Nakayama, K., Potempa, J., Travis, J., Sugawara, S., Takada, H. (2005). Arginine-specific gingipains from Porphyromonas gingivalis stimulate production of hepatocyte growth factor (scatter factor) through protease-activated receptors in human gingival fibroblasts in culture. J Immunol, 175(9):6076-6084.

Uekita, T., Jia, L., Narisawa-Saito, M., Yokota, J., Kiyono, T., Sakai, R. (2007). CUB domain-containing protein 1 is a novel regulator of anoikis resistance in lung adenocarcinoma. Mol Cell Biol, 27(21):7649-7660.

Uhland, K., Siphos, B., Arkona, C., Schuster, M., Petri, B., Steinmetzer, P., Mueller, F., Schweinitz, A., Steinmetzer, T., Van De Locht, A. (2009). Use of IHC and

216 newly designed matriptase inhibitors to elucidate the role of matriptase in pancreatic ductal adenocarcinoma. Int J Oncol, 35(2):347-357.

Ulloa-Aguirre, A., Uribe, A., Zarinan, T., Bustos-Jaimes, I., Perez-Solis, M.A., Dias, J.A. (2007). Role of the intracellular domains of the human FSH receptor in G(alphaS) protein coupling and receptor expression. Mol Cell Endocrinol, 260-262:153-162.

Vaidehi, N., Floriano, W.B., Trabanino, R., Hall, S.E., Freddolino, P., Choi, E.J., Zamanakos, G., Goddard, W.A., 3rd. (2002). Prediction of structure and function of G protein-coupled receptors. Proc Natl Acad Sci U S A, 99(20):12622-12627. van Beek, E.R., Lowik, C.W., van Wijngaarden, J., Ebetino, F.H., Papapoulos, S.E. (2009). Synergistic effect of bisphosphonate and docetaxel on the growth of bone metastasis in an animal model of established metastatic bone disease. Breast Cancer Res Treat, 118(2):307-313. van der Merwe, J.Q., Hollenberg, M.D., MacNaughton, W.K. (2008). EGF receptor transactivation and MAP kinase mediate proteinase-activated receptor-2- induced chloride secretion in intestinal epithelial cells. Am J Physiol Gastrointest Liver Physiol, 294(2):G441-451. van Rijn, R.M., Chazot, P.L., Shenton, F.C., Sansuk, K., Bakker, R.A., Leurs, R. (2006). Oligomerization of recombinant and endogenously expressed human histamine H(4) receptors. Mol Pharmacol, 70(2):604-615.

Venkatasamy, R., Spina, D. (2007). Protease inhibitors in respiratory disease: focus on asthma and chronic obstructive pulmonary disease. Expert Rev Clin Immunol, 3(3):365-381.

Versteeg, H.H., Schaffner, F., Kerver, M., Ellies, L.G., Andrade-Gordon, P., Mueller, B.M., Ruf, W. (2008). Protease-activated receptor (PAR) 2, but not PAR1, signaling promotes the development of mammary adenocarcinoma in polyoma middle T mice. Cancer Res, 68(17):7219-7227.

Vouret-Craviari, V., Grall, D., Van Obberghen-Schilling, E. (2003). Modulation of Rho GTPase activity in endothelial cells by selective proteinase-activated receptor (PAR) agonists. J Thromb Haemost, 1(5):1103-1111.

Vu, T.K., Hung, D.T., Wheaton, V.I., Coughlin, S.R. (1991). Molecular cloning of a functional thrombin receptor reveals a novel proteolytic mechanism of receptor activation. Cell, 64(6):1057-1068.

Wan, J., Roth, A.F., Bailey, A.O., Davis, N.G. (2007). Palmitoylated proteins: purification and identification. Nat Protoc, 2(7):1573-1584.

Wang, H., Ubl, J.J., Stricker, R., Reiser, G. (2002). Thrombin (PAR-1)-induced proliferation in astrocytes via MAPK involves multiple signaling pathways. Am J Physiol, 283(5):C1351-1364.

217

Wang, L., Luo, J., Fu, Y., He, S. (2006). Induction of interleukin-8 secretion and activation of ERK1/2, p38 MAPK signaling pathways by thrombin in dermal fibroblasts. Int J Biochem Cell Biol, 38(9):1571-1583.

Warne, T., Moukhametzianov, R., Baker, J.G., Nehme, R., Edwards, P.C., Leslie, A.G., Schertler, G.F., Tate, C.G. (2011). The structural basis for agonist and partial agonist action on a beta(1)-adrenergic receptor. Nature, 469(7329):241-244.

Warne, T., Serrano-Vega, M.J., Baker, J.G., Moukhametzianov, R., Edwards, P.C., Henderson, R., Leslie, A.G., Tate, C.G., Schertler, G.F. (2008). Structure of a beta1-adrenergic G-protein-coupled receptor. Nature, 454(7203):486-491.

Wei, H., Ahn, S., Shenoy, S.K., Karnik, S.S., Hunyady, L., Luttrell, L.M., Lefkowitz, R.J. (2003). Independent beta-arrestin 2 and G protein-mediated pathways for angiotensin II activation of extracellular signal-regulated kinases 1 and 2. Proc Natl Acad Sci U S A, 100(19):10782-10787.

Wess, J. (1996). Molecular biology of muscarinic acetylcholine receptors. Crit Rev Neurobiol, 10(1):69-99.

Wilcke, M., Johannes, L., Galli, T., Mayau, V., Goud, B., Salamero, J. (2000). Rab11 regulates the compartmentalization of early endosomes required for efficient transport from early endosomes to the trans-golgi network. J Cell Biol, 151(6):1207-1220.

Williams, C.R., Tabios, R., Linehan, W.M., Neckers, L. (2007). Intratumor injection of the Hsp90 inhibitor 17AAG decreases tumor growth and induces apoptosis in a prostate cancer xenograft model. J Urol, 178(4 Pt 1):1528-1532.

Wilson, B.J., Harada, R., LeDuy, L., Hollenberg, M.D., Nepveu, A. (2009). CUX1 transcription factor is a downstream effector of the proteinase-activated receptor 2 (PAR2). J Biol Chem, 284(1):36-45.

Wilson, S., Greer, B., Hooper, J., Zijlstra, A., Walker, B., Quigley, J., Hawthorne, S. (2005). The membrane-anchored serine protease, TMPRSS2, activates PAR- 2 in prostate cancer cells. Biochem J, 388(Pt 3):967-972.

Wilson, S.R., Gallagher, S., Warpeha, K., Hawthorne, S.J. (2004). Amplification of MMP-2 and MMP-9 production by prostate cancer cell lines via activation of protease-activated receptors. Prostate, 60(2):168-174.

Wong, C.H., Baehner, F.L., Spassov, D.S., Ahuja, D., Wang, D., Hann, B., Blair, J., Shokat, K., Welm, A.L., Moasser, M.M. (2009). Phosphorylation of the SRC epithelial substrate Trask is tightly regulated in normal epithelia but widespread in many human epithelial cancers. Clin Cancer Res, 15(7):2311- 2322.

218 Wright, J.L., Farmer, S.G., Churg, A. (2002). Synthetic serine elastase inhibitor reduces cigarette smoke-induced emphysema in guinea pigs. Am J Respir Crit Care Med, 166(7):954-960.

Wu, B., Chien, E.Y., Mol, C.D., Fenalti, G., Liu, W., Katritch, V., Abagyan, R., Brooun, A., Wells, P., Bi, F.C., Hamel, D.J., Kuhn, P., Handel, T.M., Cherezov, V., Stevens, R.C. (2010). Structures of the CXCR4 chemokine GPCR with small-molecule and cyclic peptide antagonists. Science, 330(6007):1066-1071.

Wu, X., Senechal, K., Neshat, M.S., Whang, Y.E., Sawyers, C.L. (1998). The PTEN/MMAC1 tumor suppressor phosphatase functions as a negative regulator of the phosphoinositide 3-kinase/Akt pathway. Proc Natl Acad Sci U S A, 95(26):15587-15591.

Wyche, L.D., de Santos, L.A. (1978). Spiculated periosteal reaction in metastatic disease resembling osteosarcoma. Orthopedics, 1(3):215-221.

Xie, X.Q., Chen, J.Z., Billings, E.M. (2003). 3D structural model of the G-protein- coupled cannabinoid CB2 receptor. Proteins, 53(2):307-319.

Xu, F., Wu, H., Katritch, V., Han, G.W., Jacobson, K.A., Gao, Z.G., Cherezov, V., Stevens, R.C. (2011). Structure of an agonist-bound human A2A adenosine receptor. Science, 332(6027):322-327.

Xu, W.F., Andersen, H., Whitmore, T.E., Presnell, S.R., Yee, D.P., Ching, A., Gilbert, T., Davie, E.W., Foster, D.C. (1998). Cloning and characterization of human protease-activated receptor 4. Proc Natl Acad Sci U S A, 95(12):6642- 6646.

Yagi, Y., Otani, H., Ando, S., Oshiro, A., Kawai, K., Nishikawa, H., Araki, H., Fukuhara, S., Inagaki, C. (2006). Involvement of Rho signaling in PAR2- mediated regulation of neutrophil adhesion to lung epithelial cells. Eur J Pharmacol, 536(1-2):19-27.

Yanai, A., Huang, K., Kang, R., Singaraja, R.R., Arstikaitis, P., Gan, L., Orban, P.C., Mullard, A., Cowan, C.M., Raymond, L.A., Drisdel, R.C., Green, W.N., Ravikumar, B., Rubinsztein, D.C., El-Husseini, A., Hayden, M.R. (2006). Palmitoylation of huntingtin by HIP14 is essential for its trafficking and function. Nat Neurosci, 9(6):824-831.

Yang, M., Jiang, P., Sun, F.X., Hasegawa, S., Baranov, E., Chishima, T., Shimada, H., Moossa, A.R., Hoffman, R.M. (1999). A fluorescent orthotopic bone metastasis model of human prostate cancer. Cancer Res, 59(4):781-786.

Yang, X., Claas, C., Kraeft, S.K., Chen, L.B., Wang, Z., Kreidberg, J.A., Hemler, M.E. (2002). Palmitoylation of tetraspanin proteins: modulation of CD151 lateral interactions, subcellular distribution, and integrin-dependent cell morphology. Mol Biol Cell, 13(3):767-781.

219 Yang, X., Kovalenko, O.V., Tang, W., Claas, C., Stipp, C.S., Hemler, M.E. (2004). Palmitoylation supports assembly and function of integrin-tetraspanin complexes. J Cell Biol, 167(6):1231-1240.

Ye, L., Lewis-Russell, J.M., Sanders, A.J., Kynaston, H., Jiang, W.G. (2008). HGF/SF up-regulates the expression of bone morphogenetic protein 7 in prostate cancer cells. Urol Oncol, 26(2):190-197.

Yin, J.J., Mohammad, K.S., Kakonen, S.M., Harris, S., Wu-Wong, J.R., Wessale, J.L., Padley, R.J., Garrett, I.R., Chirgwin, J.M., Guise, T.A. (2003). A causal role for endothelin-1 in the pathogenesis of osteoblastic bone metastases. Proc Natl Acad Sci U S A, 100(19):10954-10959.

Yu, Z., Ahmad, S., Schwartz, J.L., Banville, D., Shen, S.H. (1997). Protein-tyrosine phosphatase SHP2 is positively linked to proteinase-activated receptor 2- mediated mitogenic pathway. J Biol Chem, 272(11):7519-7524.

Yvon, A.M., Wadsworth, P., Jordan, M.A. (1999). Taxol suppresses dynamics of individual microtubules in living human tumor cells. Mol Biol Cell, 10(4):947-959.

Zaichick, V., Sviridova, T.V., Zaichick, S.V. (1997). Zinc in the human prostate gland: normal, hyperplastic and cancerous. Int Urol Nephrol, 29(5):565-574.

Zhang, D., Hou, Q., Wang, M., Lin, A., Jarzylo, L., Navis, A., Raissi, A., Liu, F., Man, H.Y. (2009a). Na,K-ATPase activity regulates AMPA receptor turnover through proteasome-mediated proteolysis. J Neurosci, 29(14):4498-4511.

Zhang, L., Shi, J., Feng, J., Klocker, H., Lee, C., Zhang, J. (2004). Type IV collagenase (matrix metalloproteinase-2 and -9) in prostate cancer. Prostate Cancer Prostatic Dis, 7(4):327-332.

Zhang, X., Wang, W., True, L.D., Vessella, R.L., Takayama, T.K. (2009b). Protease- activated receptor-1 is upregulated in reactive stroma of primary prostate cancer and bone metastasis. Prostate, 69(7):727-736.

Zhau, H.E., Li, C.L., Chung, L.W. (2000). Establishment of human prostate carcinoma skeletal metastasis models. Cancer, 88(12 Suppl):2995-3001.

Zhu, H., Wang, H., Ascoli, M. (1995). The lutropin/choriogonadotropin receptor is palmitoylated at intracellular cysteine residues. Mol Endocrinol, 9(2):141- 150.

Zoudilova, M., Kumar, P., Ge, L., Wang, P., Bokoch, G.M., DeFea, K.A. (2007). Beta-arrestin-dependent regulation of the cofilin pathway downstream of protease-activated receptor-2. J Biol Chem, 282(28):20634-20646.

Zoudilova, M., Min, J., Richards, H.L., Carter, D., Huang, T., DeFea, K.A. (2010). beta-Arrestins scaffold cofilin with chronophin to direct localized actin

220 filament severing and membrane protrusions downstream of protease- activated receptor-2. J Biol Chem, 285(19):14318-14329.

Zuckerman, D.M., Hicks, S.W., Charron, G., Hang, H.C., Machamer, C.E. (2011). Differential regulation of two palmitoylation sites in the cytoplasmic tail of the beta1-adrenergic receptor. J Biol Chem, 286(21):19014-19023.

221