Complement C3a Receptor

Molecular Pharmacology of C3aR Modulators

MRS RANEE SINGH

D.V.M. (Doctor of Veterinary Medicine) M.Sc. (Pharmacology) M.Vet.Studies (Laboratory)

A thesis submitted for the degree of Doctor of Philosophy at The University of Queensland in 2012 Institute for Molecular Bioscience

i ABSTRACT

The complement system plays key roles in both the innate (non-specific) and acquired (specific) immune response. Complement activation is a tightly regulated process, which takes place in a sequential manner by at least three pathways. All of these pathways generate a small anaphylatoxin protein called C3a. C3a signals immune cells which infiltrate to sites of infection or injury and mount inflammatory responses leading to elimination of pathogens or damaged cells and induction of adaptive immunity. Genetic deficiency of C3 results in susceptibility to bacterial infection. However, in excess C3a can be harmful and catalyses the pathogenesis of many disease conditions. To date, there have been no truly selective and potent C3a receptor agonists/antagonists. Therefore, the purpose of this thesis was to explore the potential to develop a small molecule C3aR agonist/antagonist that can be a useful as a pharmacological probe or biomarker for studying the physiological and pharmacological roles of C3a. Chapter 1 briefly describes basic pharmacology terminology and reviews current knowledge of GPCRs, GPCR intracellular signalling pathways, C3a and C3a-C3aR related intracellular signalling pathways. Understanding the pro-inflammatory roles of C3a and its receptor in vitro and in vivo may provide a new therapeutic target for autoimmune diseases or inflammatory conditions. Chapter 2 reinvestigated reported C3a receptor ligands in a competitive binding assay and a functional assay. Compound affinity was measured by competitive binding against [125I]-C3a (80 pM). Compound functional potency (agonist or antagonist) was measured by fluorescent detection of intracellular calcium release in differentiated U937 cells. Receptor selectivity was measured by receptor desensitization using C3a or C5a, and by competitive binding with [125I]-C5a. C3aR superagonist peptide (WWGKKYRASKLGLAR) and newly reported pentapeptides WPLPR (15) and YPLPR (14) (oryzatensin derivatives) were found to have no selectivity or agonist potency for C3aR. Four decapeptides (8-11) reported as C5aR agonists were shown to be more selective for C3aR than C5aR in a competitive binding assay on human isolated monocytes. Structure-activity relationships (SAR) were studied for new hexapeptides by calcium release, FWTLAR (20) (EC50 350 nM) and FLTLAR (17) (EC50 320 nM).

They bound selectively to C3aR (IC50 82 nM, IC50 42 nM respectively) on human PBMCs with no binding to C5aR. Selectivity was confirmed by receptor desensitisation experiments. A new C3aR peptide antagonist (F1LTChaAR6) was obtained by modifying the leucine at position 4 with the bulkier group Cha, showing no agonist activity and binding tightly to C3aR (IC50 238 nM). Chapter 3 introduces an oxazole heterocycle into short peptides to restrict conformation freedom in an attempt to mimic the C3aR-binding conformation of the C-terminal activating domain of human ii C3a. This approach produced multiple potent C3aR agonists, including some with equal affinity to C3a itself and selectivity for C3aR over C5aR on isolated monocyte derived macrophage cells (HMDM). There was a linear correlation between the binding affinity of the compounds and their agonist potency. Chapter 4 describes two approaches for obtaining antagonists of C3aR and shows some structure-activity relationships. Firstly, a potent new C3aR agonist (56) from Chapter 3 was modified by incorporation of a phenyl group at the 5-position of oxazole ring, which improved C3aR binding affinity and led to a novel and potent nonpeptidic C3aR antagonist. Secondly, new non-peptidic agonists of C3aR were developed from the first reported C3aR antagonist by modification of either the linker or hydrophobic region of SB290157 with retention of the terminal arginine. Several potent C3aR antagonists were developed from an oxazole-containing compound by incorporating a phenyl group at the 5-position of oxazole ring. Two other compounds were derived by modifying SB290157 at the linker region through incorporation of a thiazole or oxazole ring. Chapter 5 explores the antagonist mechanism of non-peptidic C3aR antagonist, SB290157 and newly developed antagonist 146, as well as the peptidic antagonist FLTChaAR (25). It reports competition with three different classes of C3aR agonists: hC3a, two potent new hexapeptide analogues (17 and 20) and two potent new non-peptide agonists (63 and 80). SB290157 and 146 show competitive and surmountable antagonism against the hexapeptide agonists (17 and 20) and non-peptide agonist (63 and 80), supporting binding at the same site(s) on C3aR. Compound 25 showed insurmountable antagonism of all agonists (hC3a, 17, 20, 63 and 80), suggesting that 25 binds different sites from these compounds or it may be caused by C3aR internalisation or (pseudo) irreversible binding to C3aR. However, 25 is unlikely to be an allosteric antagonist due to competition with C3a radioligand in a competitive binding assay. In summary, this study has investigated and reported new C3aR agonists and antagonists that can be used to probe influences of C3a/C3aR in physiology and pharmacology.

iii DECLARATION BY AUTHOR

This thesis is composed of my original work, and contains no material previously published or written by another person except where due reference has been made in the text. I have clearly stated the contribution by others to jointly-authored works that I have included in my thesis.

I have clearly stated the contribution of others to my thesis as a whole, including statistical assistance, survey design, data analysis, significant technical procedures, professional editorial advice, and any other original research work used or reported in my thesis. The content of my thesis is the result of work I have carried out since the commencement of my research higher degree candidature and does not include a substantial part of work that has been submitted to qualify for the award of any other degree or diploma in any university or other tertiary institution. I have clearly stated which parts of my thesis, if any, have been submitted to qualify for another award.

I acknowledge that an electronic copy of my thesis must be lodged with the University Library and, subject to the General Award Rules of The University of Queensland, immediately made available for research and study in accordance with the Copyright Act 1968.

I acknowledge that copyright of all material contained in my thesis resides with the copyright holder(s) of that material. Where appropriate I have obtained copyright permission from the copyright holder to reproduce material in this thesis.

iv Publications during candidature Conor C. G. Scully; Jade S. Blackeney; Ranee Singh; Huy N. Hoang; Giovanni Abbenante; Robert C. Reid and David P. Fairlie. Selective Hexapeptide Agonists and Antagonists for Human Complement C3a Receptor. J. Med. Chem. 2010, 53, 4938-4948.

Conference Abstracts: Singh, R., Scully, C., Chu, P.,Blackeney, J.,Hoang,H., Abbenante, J., Reid, R. and Fairlie, D.P. (2010). Agonists and Antagonists of the human C3a Receptor. Proceedings of Chemistry and Structural Biology Division Symposium (IMB), UQ.

Singh, R., Reid, R., Chu P., Reed T. and Fairlie D.P. (2011) Towards human C3a Receptor Antagonists. Proceedings of Chemistry and Structural Biology Division Symposium (IMB), UQ.

Publications included in this thesis Conor C. G. Scully; Jade S. Blackeney; Ranee Singh; Huy N. Hoang; Giovanni Abbenante; Robert C. Reid and David P. Fairlie. Selective Hexapeptide Agonists and Antagonists for Human Complement C3a Receptor. J. Med. Chem. 2010, 53, 4938-4948. – Partially incorporated as paragraphs in Chapter 2.

Contributor Statement of contribution Singh, R. Designed and performed cell pharmacology experiments (90%) Analysis of the data (90 %) Fairlie, D.P. Intellectual input, project conception and planning, scientific guidance and review and analysis of results (70%) Reid, R.C. Intellectual input, project conception and planning, scientific guidance and review and analysis of results (30%) Scully, C.G.C. Structural design, synthesis and characterisation of compounds (80%) NMR studies (50%) Hoang, H.N. NMR studies (50%) Blakney, J.S. Structural design, synthesis and characterisation

v of compounds (20%) Performed experiments some compounds in calcium release assays (10%) Analysis of the data (10%)

Contributions by others to the thesis My supervisors, Prof. David Fairlie, Dr. Robert Reid and Dr. Conor Scully all contributed intellectually to this thesis through data analysis, discussion, scientific guidance and critical review of the project; Dr. Scully and Dr. Reid have contributed by manufacturing the majority of the compounds tested in this thesis.

Statement of parts of the thesis submitted to qualify for the award of another degree “None.”

vi ACKNOWLEDGEMENTS

I would like to acknowledge the assistance of a number of people who have helped me throughout my PhD. First, my supervisor who gave me opportunities to do the interesting projects. I gratefully acknowledge his wisdom, advice and support. I also would like to thank my co- supervisors Dr. Scully and Dr. Reid for their kind advice and assistance, without which I would never have made it to the end. I also would like to thank all the members of the Fairlie Lab, past and present for their advice, assistance and friendship. In particular, Jade Blackeney, Maria Halili, Gloria Ruiz Gomez, Praveen Kumar Madala and James Lim for helping out over the years and their advice when things were not working out. Thank you to Asa Anderson from Lewis Lab for being kind enough to teach me the radioligand binding techniques. I also would like to convey my thanks to Endeavour postgraduate scholarship for the financial support and my scholarship case manager Kelly Baum for her excellent support through out my candidature. Thank you to my friends and family who have provided me support and encouragement, especially my parents who have with their love and understanding always been beside me every step of the way. Lastly, thank you to my dear daughter and my husband Gurdeep, for all your love, endless patience and support over the years without which I would have never had the courage to make it through.

vii Keywords GPCR, complement, C3a, C3aR agonists, C3aR antagonists, macrophages, monocytes, inflammation

Australian and New Zealand Standard Research Classifications (ANZSRC) ANZSRC code: 060110, Receptors and Membrane Biology, 70%, ANZSRC code: 111501, Basic Pharmacology, 20%, ANZSRC code: 030406, Proteins and Peptides, 10%

Fields of Research (FoR) Classification FoR code: 0601,Biochemistry and Cll Biology, 80% FoR code: 1115, Pharmacology and Pharmaceutical Sciences, 20% FoR code: 0304, Medicinal and Biomolecular Chemistry, 10%

viii TABLE OF CONTENTS

ABSTRACT...... II

DECLARATION BY AUTHOR...... IV

ACKNOWLEDGEMENTS...... VII

TABLE OF CONTENTS ...... IX

LIST OF FIGURES ...... XIII

LIST OF TABLES ...... XIX

LIST OF ABBREVIATIONS ...... XXI

UNITS OF MEASUREMENT...... XXV

THREE AND ONE LETTER AMINO ACID CODES...... XXVI

TABLE OF COMPOUNDS ...... XXVII

CHAPTER 1 1.0 ABSTRACT...... 2 1.1 PHARMACOLOGY TERMINOLOGY...... 3 1.2 G-PROTEIN COUPLED RECEPTORS ...... 5

1.2.1 GPCR SIGNALLING ...... 8

1.2.2 CLASSIFICATION OF G-PROTEINS AND INTRACELLULAR PATHWAYS ...... 10 1.3 COMPLEMENT AND INFLAMMATION ...... 13 1.4 COMPLEMENT CASCADE ...... 13 1.5 REGULATION OF COMPLEMENT SYSTEM ...... 17 1.6 THERAPEUTIC POTENTIAL FOR COMPLEMENT INHIBITORS ...... 18 1.7 C3a ANAPHYLATOXIN...... 22

1.7.1 STRUCTURE OF ANAPHYLATOXIN C3a...... 22

1.7.2 PHYSIOLOGY OF ANAPHYLATOXIN C3a ...... 23

1.7.3 PATHOLOGY OF ANAPHYLATOXIN C3a ...... 24 1.8 RECEPTORS FOR C3a...... 31

1.8.1 C3a RECEPTORS...... 31

1.8.2 BINDING SITES OF C3a-C3aR ...... 32

1.8.3 SIGNALLING PATHWAY ASSOCIATED WITH C3a-C3aR ...... 33

ix 1.8.4 C3a RECEPTOR DISTRIBUTION ...... 35 1.9 CURRENT SMALL C3a PEPTIDE LIGANDS FOR C3aR...... 36

1.9.1 AGONISTS ...... 36

1.9.2 ANTAGONISTS...... 38 1.10 C5a ANAPHYLATOXIN...... 41 1.11 SUMMARY AND AIMS ...... 46 1.12 REFERENCES...... 49

CHAPTER 2 2.0 ABSTRACT...... 73 2.1 INTRODUCTION ...... 74 2.2 AIMS/OBJECTIVES...... 76 2.3 RESULTS ...... 78

2.3.1 SELECTIVITY AND POTENCY OF HUMAN C3a...... 78

2.3.2 SELECTIVITY AND POTENCY OF 15-RESIDUES OF HC3a AND HC5a ...... 82

2.3.3 SELECTIVITY AND POTENCY OF KNOWN TRUNCATED C3a PEPTIDES...... 85

2.3.4 C3a DES-ARG COMPOUNDS...... 91

2.3.5 C3a HEXAPEPTIDE ANALOGUES...... 92 2.4 DISCUSSION ...... 105 2.5 MATERIALS AND METHODS...... 108 2.6 REFERENCES...... 111

CHAPTER 3 3.0 ABSTRACT...... 121 3.1 INTRODUCTION ...... 122 3.2 AIMS/OBJECTIVES...... 126 3.3 RESULTS ...... 126

3.3.1 SERIES I: N-ACYL AMINO ACID OXAZOLE ARGININE DERIVATIVES...... 127

3.3.2 SERIES II: INCORPORATION OF A METHYL GROUP ON THE 5-POSITION ON

THE OXAZOLE RING...... 134

3.3.3 SELECTIVITY OF C3a NON-PEPTIDE LIGANDS FOR C5aR ...... 139

3.3.4 SELECTIVITY TEST: DESENSITISATION EXPERIMENTS...... 140

3.3.5 EFFECT OF C3a NON-PEPTIDE AGONISTS/ANTAGONISTS ON CALCIUM RELEASE FROM

HUMAN MONOCYTE DERIVED MACROPHAGES...... 143 x 3.4 DISCUSSION AND CONCLUSION...... 145 3.5 MATERIALS AND METHODS...... 148

3.5.1 MACROPHAGE CELL CULTURE AND DIFFERENTIATION...... 148

3.5.2 CALCIUM MOBILIZATION ASSAY ON HMDM ...... 148

3.5.3 RECEPTOR BINDING ASSAY OF HMDM...... 148

3.5.4 DATA ANALYSIS AND STATISTICS...... 149 3.6 REFERENCES...... 150

CHAPTER 4 4.0 ABSTRACT...... 158 4.1 INTRODUCTION ...... 160 4.2 AIMS/OBJECTIVES...... 162 4.3 RESULTS ...... 162

4.3.1 STRUCTURAL MODIFICATIONS OF OXAZOLE-CONTAINING C3aR AGONISTS ...... 162

CHAPTER 5 5.0 ABSTRACT...... 196 5.1 INTRODUCTION ...... 197 5.2 AIMS/OBJECTIVES...... 201 5.3 RESULTS ...... 202

5.3.1 COMPETITION OF NON-PEPTIDE ANTAGONIST SB290157 WITH C3aR AGONISTS ...... 203

5.3.2 COMPETITION OF PEPTIDE ANTAGONIST 25 (FLTCHAAR) WITH

C3aR AGONISTS ...... 212

5.3.3 COMPETITION OF NON-PEPTIDE ANTAGONIST 146 WITH C3aR AGONISTS...... 220 5.4 DISCUSSION ...... 224 5.5 MATERIALS AND METHODS...... 229

xi 5.5.1 CALCIUM MOBILIZATION ASSAY ON HMDM CELLS...... 229 5.6 REFERENCES...... 231

CHAPTER 6 SUMMARY...... 236

APPENDIX

SELECTIVE HEXAPEPTIDE AGONISTS AND ANTAGONISTS FOR HUMAN

COMPLEMENT C3a RECEPTOR ...... 241

xii LIST OF FIGURES

CHAPTER 1

Figure 1.1 Diagram of classes of ligands that regulate G-protein-coupled receptors 3

Figure 1.2 Schild Plot. A graph of log (dr-1) against log XB 4 Figure 1.3 (A). Representation of the crystal structure of Bovine Rhodopsin (pdb: 1U19). Rhodopsin. (B). Two-dimensional depiction of Bovine Rhodopsin 5 Figure 1.4 The function of G-protein 8 Figure 1.5 GPCR complex with potential activators and diverse biological functions 9 Figure 1.6 Signal transduction in G-protein-coupled receptors 12 Figure 1.7 Complement activation pathways 15 Figure 1.8 Cleavage of the alpha-chain of C3 by C3 convertase leads to the release of C3a anaphylatoxin 17 Figure 1.9 Crystal Structure of C3a 22 Figure 1.10 Complement anaphylatoxin sequences 23 Figure 1.11 C3aR with extracellular (EC) and intracellular (IC) loops and the helical transmembrane domains (TM) numbered 31 Figure 1.12 Interaction between C3a or C5a and their respective receptors 32 Figure 1.13 Homology modelling of C3aR 35 Figure 1.14 Comparison of the EC2 loop sequences of C3a receptors from human (Hu), guinea pig (Gp), rat (Rt), and mouse (Mo) 36

Figure 1.15 A. C-terminal pentapeptide sequence L73GLAR77 that is important for activation of C3aR. B. C3a Superagonist 38 Figure 1.16 Structures of C3aR antagonists 39 Figure 1.17 Amino acid sequence and NMR solution structure of C5a (pdb: 1KJS) 42 Figure 1.18 Chemical structure of the peptidic C5a antagonists PMX53 (3D53), PMX205 and JPE1375 45

CHAPTER 2 Figure 2.1 Affinities for C3a receptor of hC3a, SB290157 (3), hC5a and 3D53 (4) on isolated human monocytes 78 Figure 2.2 Affinities for C5a receptor of hC5a, 3D53 (4), hC3a and SB290157 (3) on isolated human monocytes 79 xiii Figure 2.3 Desensitisation experiments 80 Figure 2.4 Intracellular calcium release assay 81 Figure 2.5 Intracellular Ca2+ release induced by human C3a versus human C5a in differentiated human U937 cells 82 2+ Figure 2.6 Intracellular Ca release induced by C3a(63-77) (6) versus C5a(60-74) (7) in differentiated human U937 cells 83 Figure 2.7 Competitive [125I]-C3a binding assay of hC3a, 3, C5a, 4, 6 and 7 on isolated human monocytes 84 Figure 2.8 Competitive [125I]-C5a binding assay of hC5a, 4, C3a, 3, 6 and 7 on human monocytes 84 Figure 2.9 Competitive [125I]-C3a binding assay on human PBMCs 85 Figure 2.10 Competitive [125I]-C5a binding assay on human PBMCs 86 Figure 2.11 Competitive [125I]-C3a binding assay of C3a superagonist and C3a hexapeptides on human PBMCs 87 Figure 2.12 Competitive [125I]-C5a (B) binding assay on human PBMCs 87 Figure 2.13 Activity of hC5a, hC3a, 8, 5 and 12 in intracellular calcium mobilisation assay was measured in human dU937 cells 88 Figure 2.14 Competitive [125I]-C3a binding of FLTLA (26) and FISLA (27) on isolated human monocytes 91 Figure 2.15 Competitive [125I]-C3a binding of hC3a, SB290157, C3a and C5a 15 residues and C3a hexapeptide analogues on isolated human PBMCs 92 Figure 2.16 Competitive [125I]-C5a binding of hC5a, AcF-[OPdChaWR] (3D53) and C3a hexapeptide analogues on isolated human monocytes 93 Figure 2.17 Activity of putative C3a agonist peptides at a concentration of 100 µM 95 Figure 2.18 Activity of putative C3a agonist peptides at a concentration of 1 µM 96 Figure 2.19 Structure Activity Relationship studies of hexapeptide agonists

by substitution of F1WPLAR6 or F1LTLAR6 97-98 Figure 2.20 Structure of hexapeptides 16 – 20 101 Figure 2.21 Intracellular calcium release in differentiated U937 cells induced by hC3a, 20 and 46, 19 and 17 102 Figure 2.22 Intracellular Ca2+ release induced by variable concentration of C3a peptide analogues 102 Figure 2.23 Desensitisation assays of hexapeptide analogues and C3a superagonist peptide 103 xiv Figure 2.24 Correlation between binding affinity (pIC50) and agonist activity (pEC50) of hexapeptides agonists 104 Figure 2.25 NMR-derived solution structure for FLTLAR in DMSO-d6 (298 K) 105

CHAPTER 3 Figure 3.1 The 3D Crystal structure of C3a (77 amino acid residues) 125 Figure 3.2 Superimposition of an oxazole scaffold compound (56) and tetrapeptide C-terminal residues of hC3a (-GLAR) in a type II beta turn conformation 127 Figure 3.3 N-Acyl amino acid-oxazole-arginine scaffold 128 Figure 3.4 N-Acyl-modifications 130 Figure 3.5 Competitive binding between [125I]-C3a and C3a non-peptide ligands in human monocyte derived macrophages 132 Figure 3.6 Structure activity relationship studies of C3a non-peptidic agonists by substitution of AA with either isoleucine (A) or leucine (B) to oxazole scaffold 132 Figure 3.7 Competitive [125I]-C3a binding of hC3a, SB290157 and C3a non-peptide ligands on human monocyte derived macrophages 133 Figure 3.8 The comparisons in binding affinity of non-peptidic ligands 57, 56, 58, 59 and 60 133 Figure 3.9 The fifth position of oxazole was potentially a position for substitution with a methyl group which could mimic the side chain methyl of the alanine residue in C3a 134 Figure 3.10 Acyl-leucine-5-methyl-oxazole-arginine (compound 80) 135 Figure 3.11 A. Competitive binding between [125I]-C3a and non-peptide ligands in human monocyte derived macrophages. B. Non-peptide agonism screening in Ca2+ release assay 137-138 Figure 3.12 Competitive [125I]-C3a binding of hC3a, SB290157, 80 and 81 on human monocytes derived macrophage cells 138 Figure 3.13 Competitive binding between [125I]-C5a and C3a non-peptide ligands in human monocyte derived macrophages 139 Figure 3.14 Competitive [125I]-C5a binding of hC5a, 3D53 and C3a non-peptide ligands 63, 57, 75, 80, 81 on human monocyte derived macrophage cells 140

xv Figure 3.15 C3a receptor desensitisation calcium mobilisation plots performed on human monocyte derived macrophages (HMDM) for non-peptidic agonists 141-142 Figure 3.16 Non-peptide antagonism screening in Ca2+ release assay in human derived macrophage cells at a concentration of 10 µM 143 Figure 3.17 Intracellular Ca2+ mobilisation dose response induction by hC3a, SB290157 and variable concentration of C3a non-peptide agonist 144 Figure 3.18 Intracellular calcium release in HMDM cells induced by hC3a, 71, 65, 63, 56, 75, 57, 73 145

Figure 3.19 Correlation between binding affinity (pIC50) and agonist activity

(pEC50) of non-peptide agonists 147

CHAPTER 4 Figure 4.1 Oxazole-containing C3aR modulators 160 Figure 4.2 Structures for non-peptidic “antagonists” of C3aR reported in the literature 161 Figure 4.3 Overlay of 56 and GLAR (C-terminus of C3a) reveals the possibility of incorporation of the functional group into the 5th position of oxazole ring 163 Figure 4.4 Modification of the oxazole scaffold by incorporating a 5-methyl substituent 163 Figure 4.5 Modification of the oxazole scaffold by incorporating a 5-phenyl substituent 164 Figure 4.6 Competition with [125I]-C3a for binding to human monocyte-derived macrophages: hC3a, SB290157, 56, 80 and 105 165 Figure 4.7 Receptor affinities for C3aR measured by displacement of 80 pM [125I]-C3a from HMDM cells: hC3a, SB290157, 109 and 105 168 Figure 4.8 A. Competitive binding versus [125I]-C5a (20 pM) for C3aR-binding non-peptide ligands (20 µM) in human monocyte-derived macrophages 169 Figure 4.9 C3a receptor desensitisation in the calcium mobilisation assay on HMDM for 105 (B) and 109 (C) 170 Figure 4.10 Single point calcium release induced by Acyl-Leucine-5-Phenyl- Oxazole-Arginine compounds 171 Figure 4.11 Concentration dependent inhibition of SB290157, 105 and 109 172 xvi Figure 4.12 C3aR antagonist SB290157 174 Figure 4.13 R-glycine-arginine tripeptide/dipeptide surrogates 176 Figure 4.14 Cinnamoyl-glycine-arginine tripeptide surrogates 179 Figure 4.15 R-Tryptophan-arginine tripeptide surrogates 180 Figure 4.16 Structure activity relationship studies of C3a non-peptidic agonists by substitution of SB290157 in different regions 182 Figure 4.17 (a) SB290157 (b) Substitution of different heterocycles in the linker region 183 Figure 4.18 Competitive [125I]-C3a binding of hC3a, SB290157, and non-peptidic ligands with heterocycles in the linker region 143, 144, 145 and 146 on human monocytes derived macrophage cells 185 Figure 4.19 Intracellular calcium release from human monocyte derived macrophage cells induced by hC3a, 143, 144, 145, 146 186 Figure 4.20 Concentration dependent inhibition of C3a-inducd Ca2+ release in HMDM by SB290157, 145 and 146 187 Figure 4.21 Competitive binding between [125I]-C5a and heterocycle-containing ligands in human monocyte derived macrophage cells 188 Figure 4.22 C3a receptor desensitisation calcium release plots performed on HMDM for compounds with heterocycles at the linker region 189 Figure 4.23 Superimposition of SB290157 and furan compound (144) showing the location of the hydrogen bond acceptor atom 191

CHAPTER 5 Figure 5.1 Two-site model proposed for the interaction between C3a and its receptor 198 Figure 5.2 Structure of peptide agonists (17 and 20), peptide antagonist (25), non-peptide agonists (63 and 80) and non-peptide antagonists (105, 146 and SB290157) 200

Figure 5.3 Correlation between binding affinity (pIC50) and agonist activity (pEC50) of hC3a, peptide C3a agonists (17, 20), non-peptidic C3a agonists (80, 63) and C3a antagonists (SB290157, 25, 105, 146) 202 Figure 5.4 Antagonism by SB290157 against hC3a 203-204 Figure 5.5 SB290157 is a competitive antagonist against 17 on human monocyte derived macrophages 205-206 Figure 5.6 SB290157 is a competitive antagonist against 20 on human monocyte-derived macrophages 207

xvii Figure 5.7 SB290157 is a competitive and surmountable antagonist against non-peptide C3a agonist (63) 209 Figure 5.8 SB290157 is a competitive and surmountable antagonist against non-peptide C3a agonist (80) 211 Figure 5.9 Mechanism of C3aR antagonism 213 Figure 5.10 Compound 25 is a non-competitive and insurmountable antagonist against 17 214-215 Figure 5.11 Compound 25 is a non-competitive and insurmountable antagonist against 20 216 Figure 5.12 Compound 25 is a non-competitive and insurmountable antagonist against non-peptide agonist (63) 217-218 Figure 5.13 Compound 25 is a non-competitive and insurmountable antagonist against non-peptide agonist (80) 219 Figure 5.14 Compound 146 is a non-competitive and insurmountable antagonist against hC3a on human monocyte derived macrophages 220-221 Figure 5.15 Compound 146 is a competitive antagonist against 17 on human monocyte derived macrophages 222 Figure 5.16 Compound 146 is a competitive antagonist against 63 on human monocyte derived macrophages 223-224 Figure 5.17 Superimposition of the oxazole (green) (Boc-Leu-oxazole-Arg, 88) with the last four C-terminal residues of FLTChaAR (25) (NMR structure). B. SB290157 aligned with energy minimised compound 63 C. Alignment of furan and SB290157 228

xviii LIST OF TABLES

CHAPTER 1 Table 1.1 Family classification of known G-protein-coupled receptors 6 Table 1.2 Classification of Gα subunits and their effectors 11 Table 1.3 Complement inhibitors approved for intervention of complement cascade on the market or in clinical trials 20-21 Table 1.4 Functional activity of C3a on myeloid and non-myeloid cells 26-27 Table 1.5 Roles for C3a in diseases 28-30 Table 1.6 Summary of SB290157 used in animal models 40 Table 1.7 Examples of Synthetic Peptide Agonists and Antagonists of hC3a 41 Table 1.8 Functional activity of C5a on myeloid and non-myeloid cells 43

CHAPTER 2 Table 2.1 Affinity and Potency of Known Truncated Anaphylatoxin Agonists on dU937 cells 90 Table 2.2 Structure-Activity relationship (SAR) data for hexapeptide

analogues of selective agonist F1WPLAR6 94

Table 2.3 Structure-Activity of F1WPLAR6 and F1LTLAR6 analogues on human dU937 cells measured by calcium release assay 99-100

CHAPTER 3 Table 3.1 Competitive binding with [125I]-C3a and functional activity (Ca2+ mobilisation) of compounds on isolated human monocyte-derived macrophages (HMDM) 131 Table 3.2 Activity of acyl-leucine-5-methyl-oxazole-arginine (80) on competitive binding with [125I]-C3a and Ca2+ mobilisation from isolated human monocyte derived macrophage cells (HMDM) 135 Table 3.3 SAR data for series II: Acyl-leucine-5-Methyl-Oxazole-Arginine Scaffolds 136

xix CHAPTER 4 Table 4.1 SAR of acyl-leucine-5-phenyl-oxazole-arginine (105) comparison to boc-leucine-oxazole-arginine (56) and acyl-leucine-5-methyl-oxazole-arginine (80) on competitive binding with [125I]-C3a and Ca2+ mobilisation from HMDM cells 165 Table 4.2 SAR data of Acyl-Leucine-5-Phenyl-Oxazole-Arginine series 166 Table 4.3 Competitive binding of 109 with [125I]-C3a and Ca2+ mobilisation from human derived macrophage cells 167 Table 4.4 SAR data for series I: Alkylated SB290157 analogues 175 Table 4.5 SAR data for series II: R-Glycine-arginine tripeptide/dipeptide surrogates 176-177 Table 4.6 R-Glycine-arginine tripeptide surrogates as candidate C3aR ligand 178 Table 4.7 SAR data of Series III: Cinnamoyl-glycine-arginine tripeptide surrogates 179 Table 4.8 SAR data of Series IV: R-Tryptophan-arginine tripeptide surrogates 181 Table 4.9 C3aR binding and agonism: Effect of heterocycles in the linker region 184 Table 4.10 Comparison result between Agonism and Antagonism activity of 145 and 146 186

LIST OF ABBREVIATIONS

95% C.I. 95% confidence interval 3D53 Ac-cyclo-(2,6)-F[OP-(dCha)WR] Ab 2-aminobutyric acid AD Atopic dermatitis AHR Airway hyperresponsiveness aHUS Atypical haemolytic uremic syndrome Ahx 6-aminohexanoic acid Aib 2-aminoisobutyric acid AMD Age-related macular degeneration AMI Acute myocardial infarction AMP Adenosine-monophosphate ARDS Adult respiratory distress syndrome ASP Adenylation stimulating protein ATCC American Type Culture Collection BSA Bovine serum albumin

Bt2-cAMP 3’, 5’-dibutyryladenosine monophosphate ATP Adenosine triphosphate C-terminal/C-terminus Carboxy-terminal/us Ca2+ Divalent calcium C3a Complement factor 3a C3aR C3a receptor C4a Complement factor 4a C5a Complement factor 5a CABG Coronary artery bypass grafting C5aR C5a receptor CD Circular dichroism cDNA Complementary DNA C5L2 C5L2 receptor Cha Cyclohexylalanine CHO Chinese Hamster Ovary CNS Central Nervous System

xxi

CO2 Carbon Dioxide COPD Chronic obstructive pulmonary disease CR1 Complement receptor 1 CSF Cerebral spinal fluid CVF Cobra venom factor DAF Decay accelerating factor DAG Diacylglycerol DC Dendritic cell DIC Disseminated intravascular coagulation

DMSO-d6 Dimethyl sulfoxide-d6 or hexadeuterodimethyl sulfoxide dU937 Human monocytic cells from histiocytic lymphoma

differentiated with Bt2-cAMP EC2 Extracellular loop 2 of the C3a receptor

EC50 Molar concentration that produces 50% of the maximum response ECL Extracellular Loop ECP Eosinophil cationic protein ELISA Enzyme-linked immunosorbent assay

EM520 Emission wavelength at 520 nm ERK1/2 Extracellular signal - regulated kinases 1/2 FBS Fetal bovine serum FDA United States Food and Drug Administration Fmoc Fluorenylmethoxycarbonyl GEF Guanine nucleotide - exchange factor GM-CSF Granulocyte macrophage colony-stimulating factor G-protein Guanosine monophosphate protein GPCR G-protein coupled receptor GRK G-protein coupled receptor kinase HAE Hereditary angioedema HBSS Hanks’ balanced salt solution HCC Hepatitis C virus-related hepatocellular carcinoma HCV Hepatitis C virus HEK-293 Human embryonic kidney 293 cells HEPES 4-(2-hydroxyethyl)-1-piperazineethaesulfonic acid xxii hF Homophenylalanine HMC-1 Human mast cell line HMDM Human monocyte derived macrophage HUVEC Human umbilical vein endothelial cells 125I Iodine isotope 125 IBD Inflammatory bowel disease

IC50 Molar concentration of an unlabeled agonist/antagonist that inhibits 50% radioligand binding or molar concentration of antagonist that inhibits 50% of a known concentration of agonist activity Ig Immunoglobulin IL Interleukin IMDM Iscove’s modified Dulbucco’s medium iNos Inducible nitric oxide synthase

IP3 Inositol triphosphate I/R injury Ischemia/reperfusion injury LPS Lipopolysaccharide mAB Monoclonal antibody MAC Membrane attack complex (C5b - 9) mAChR Muscarinic acetylcholine receptor MAPK Mitogen-activated protein kinase MASP Mannose-binding lectin associated serine protease MBL Mannose binding lectin MCP Membrane cofactor protein MRL/lpr MRL/Mp-Tnfrsf lpr/lpr mouse model of human SLE mRNA Messenger ribonucleic acid MSC Mesenchymal stem cell Nle Norleucine N-terminal/N-terminus Amino-terminal/Amino-terminus NEAA Nonessential amino acids NMR Nuclear magnetic resonance PAF Platelet-activating factor PBMC Peripheral blood mononuclear cells PBS Phosphate buffer saline

xxiii

PDE Phosphodiesterase E pEC50 Negative logarithm of EC50 pIC50 Negative logarithm of IC50

PIP2 Phosphotidylinositol 4,5 -Biphosphate PK Protein kinase PLC Phospholipase C PLD Phospholipase D PMN Polymorphonuclear leukocyte PNH Paroxysmal nocturnal hemoglobinuria PTX Pertussis toxin RA Rheumatoid arthritis RANTES Regulated upon Activation, Normal T cell expressed and secreted RNA Ribonucleic acid SAR Structure-activity relationship S.E.M. Standard error of the mean SLE Systemic lupus erythematosus TM Transmembrane TNF-α Tumor necrosis factor-α U937 Human leukemic monocyte lymphoma cell line

xxiv

UNITS OF MEASUREMENT

p pico n nano µ micro m milli k kilo

Da Dalton oC Degree Celsius g Gram L Litre M Molar m metre min Minute(s) s Second(s) h Hour(s)

xxv

THREE AND ONE LETTER AMINO ACID CODES

Alanine Ala A Arginine Arg R Asparagine Asn N Aspartic acid Asp D Cyclohexylalanine Cha - Cysteine Cys C Glutamic acid Glu E Glutamine Gln Q Glycine Gly G Histidine His H Isoleucine Ile I Leucine Leu L Lysine Lys K Methionine Met M Norleucine Nle - Norvaline Nve - Ornithine Orn O Phenylalanine Phe F Proline Pro P Serine Ser S Threonine Thr T Tryptophan Trp W Tyrosine Tyr Y Valine Val V

‘d’ small d denoted D-stereochemistry of the amino acid.

xxvi LIST OF ALL COMPOUNDS

PEPTIDE COMPOUNDS

# Peptide # Peptide 1 C3a 41 FISLAR 2 C5a 42 FWALAR 3 SB290157 43 FWSLAR 4 3D53 44 FLPLAR 5 WWGKKYRASKLGLAR 45 FLALAR

6 C3a63-77 46 FLSLAR

7 C5a60-74 47 FLTIAR 8 YSFKPMPL(Me-a)R 48 FLTNleAR 9 YSFK(Me-D)MPLaR 49 FLT(dCha)AR 10 YSHKPMPLaR 50 FLTSAR 11 YSFKPMPLaR 51 FWTVAR 12 HLGLAR 52 FWSIAR 13 HLALAR 53 FLTLLR 14 YPLPR 54 FLTLPR

15 WPLPR 55 FLTLAR-NH2 16 FWPLAR 17 FLTLAR 18 FIPLAR 19 WWTLAR 20 FWTLAR 21 FYTLAR 22 FWGLAR 23 FLGLAR 24 FLTLAr 25 FLTChaAR 26 FLTLA 27 FISLA 28 hFLTLAR 29 Ac-FLTLAR 30 YLTLAR 31 RLTLAR 32 F(Cha)PLAR 33 F(Nap)PLAR 34 F(hF)PLAR 35 FOTLAR 36 FRTLAR 37 FYSLAR 38 FKTLAR 39 FNleTLAR 40 FVTLAR 41 FISLAR

xxvii NON-PEPTIDE COMPOUNDS (N-ACYL-AMINO ACID-OXAZOLE-ARGININE SCAFFOLD)

# R AA # R AA 56 Leu 68 Ile

57 Ile 69 Leu

58 Phe 70 Leu

59 Trp 71 Ile

60 Cha 72 Leu

61 Ile 73 Ile

62 Ile 74 Ile

63 Leu 75 Ile

64 Ile 76 Ile

65 Ile 77 Ile

66 Ile 78 Ile

67 Ile 79 Leu

xxviii NON-PEPTIDE AGONISTS: (ACYL-LEUCINE-5-METHYL-OXAZOLE-ARGININE SCAFFOLDS)

# R # R 80 93

81 94

82 95

83 96

84 97

85 98

86 99

87 100

88 101

89 102

90 103

91 104

92 - -

xxix C3aR ANTAGONISTS (ACYL-LEUCINE-5-PHENYL-OXAZOLE-ARGININE SCAFFOLDS)

# R # R 105 108

106 109

107 110

ALKYLATED SB290157 ANALOGUES

Structure # 111 H N CO H O 2 O

NH HN NH2 112 H N CO2H O O

NH HN NH2 113

H N CO H O 2 O NH

HN NH2 xxx R-GLYCINE-ARGININETRIPEPTIDE/DIPEPTIDE SURROGATES

# R # R 114 122

115 123

116 124

117 125

118 126

119 127

120 128

121 129

xxxi

# R # R 130 134

131 135

132 136

133 137

TRYPTOPHAN-ARGININE TRIPEPTIDE SURROGATES

# R # R 138 141

139 142

140

xxxii

SUBSTITUTION OF DIFFERENT HETEROCYCLES IN THE LINKER REGION

Position # X Y Z

143 N O CH

144 O CH CH

145 O N C-Me

146 S N C-Me

xxxiii CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

CHAPTER 1 PEPTIDE-ACTIVATED G-PROTEIN COUPLED RECEPTORS: - COMPLEMENT C3a RECEPTOR

1.0 ABSTRACT...... 2 1.1 PHARMACOLOGY TERMINOLOGY...... 3 1.2 G-PROTEIN COUPLED RECEPTORS...... 5

1.2.1 GPCR SIGNALLING ...... 8

1.2.2 CLASSIFICATION OF G-PROTEINS AND INTRACELLULAR PATHWAYS...... 10 1.3 COMPLEMENT AND INFLAMMATION ...... 13 1.4 COMPLEMENT CASCADE...... 13 1.5 REGULATION OF COMPLEMENT SYSTEM...... 17 1.6 THERAPEUTIC POTENTIAL FOR COMPLEMENT INHIBITORS ...... 18 1.7 C3a ANAPHYLATOXIN...... 22

1.7.1 STRUCTURE OF ANAPHYLATOXIN C3a ...... 22

1.7.2 PHYSIOLOGY OF ANAPHYLATOXIN C3a ...... 23

1.7.3 PATHOLOGY OF ANAPHYLATOXIN C3a...... 24 1.8 RECEPTORS FOR C3a...... 31

1.8.1 C3a RECEPTORS ...... 31

1.8.2 BINDING SITES OF C3a-C3aR...... 32

1.8.3 SIGNALLING PATHWAY ASSOCIATED WITH C3a-C3aR...... 33

1.8.4 C3a RECEPTOR DISTRIBUTION ...... 35 1.9 CURRENT SMALL C3a PEPTIDE LIGANDS FOR C3aR...... 36

1.9.1 AGONISTS...... 36

1.9.2 ANTAGONISTS ...... 38 1.10 C5a ANAPHYLATOXIN...... 41 1.11 SUMMARY AND AIMS ...... 46 1.12 REFERENCES...... 49

1 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

1.0 ABSTRACT

G-protein coupled receptors (GPCRs) or seven-transmembrane receptors are the most common class of receptors on cell surfaces, of which more than 800 in the human genome have been identified to date. They are also the most common drug targets for the pharmaceutical industry. The 7-transmembrane receptor binds to an agonist, which causes the receptor to change conformation and this results in intracellular binding to heterotrimeric G proteins and consequently signalling by second messenger systems. The GPCR kinases phosphorylate G-protein coupled receptor results in increased affinity for arrestins leading to termination of G-protein mediated signalling and densensitization of the GPCR to further activation for a period of time. The complement system plays an important role in both innate and acquired immunity and also acts as a primary mediator of inflammatory processes. Complement consists of over 30 soluble plasma proteins that are activated in tightly regulated sequence. One of the major products of this activation is the generation of small proteins, C3a and C5a, also known as ‘anaphylatoxins’. Anaphylatoxins play major roles in inflammation, priming and amplifying the immune response through chemoattraction of immune cells to inflammatory sites and promoting the secretion of proinflammatory agents, lysosomal enzymes, and reactive oxygen species (ROS). The biological effects of C3a and C5a are mediated through 7-transmembrane G-protein coupled receptor C3aR and C5aR respectively on specific target cells resulting in receptor phosphorylation and activation of relevant intracellular signal transduction pathways. C3aR are found in myeloid cells such as neutrophils, mast cells, eosinophils and basophils, monocytes/macrophages and non-myeloid cells for example epithelial cells and endothelial cells. C3a and its receptor have been implicated in various pathogeneses of autoimmune diseases and acute inflammatory conditions e.g. asthma, sepsis, allergies, lupus erythematosus (SLE), diabetes, psoriasis, arthritis, nephropathy, ischemia- reperfusion injury and others. The development of C3a potent and selective agonists or antagonists has potential for physiological probes or anti-inflammatory drugs. This chapter aims to describe the GPCRs and intracellular signalling cascades and complement pathways, which are implicated in inflammatory processes. The review focuses on C3a and C3aR on the basis of C3a and C3aR structure, the molecular interaction of C3a to its receptor and the signalling pathways in biological responses. In addition, the peptide/nonpeptide ligands have been developed for C3aR agonist/antagonists. C5a and C5aR are also reviewed in brief.

2 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

1.1 PHARMACOLOGY TERMINOLOGY

A Full agonist is a ligand that produces full receptor activation leading to production of the system maximal response. Partial agonists produce submaximal receptor activation leading to production of submaximal system response. Inverse agonists are agonists that can mediate the receptor activity of GPCR systems in two different systems. In non-constitutively active system it shows an antagonist effect whereas in constitutively activated GPCR systems (in absence of any ligand) it shows the effect of reducing the level of constitutive activation (negative efficacy). Allosteric agonists (Figure 1.1) function as agonists but activate the receptor through interaction at a site distinct from that of the endogenous agonist (common in non-peptide agonists for peptide receptor). In contrast, orthosteric agonists bind at the same site as the endogenous ligand. Allosteric modulator is a ligand that binds to an allosteric site on the GPCR to modulate the binding and/or signalling properties of the orthosteric site. Allosteric antagonist (inhibitor) is allosteric modulator that reduces orthosteric ligand affinity and/or orthosteric agonist efficacy. Allosteric enhancer (Potentiator) is allosteric modulators that potentiates orthosteric ligand affinitity on the receptor and/or agonists efficacy.

Figure 1.1 Diagram of classes of ligands that regulate G-protein-coupled receptors.

An Antagonist is a ligand that does not elicit a biological response itself upon binding to a receptor, but blocks or dampens agonist-mediated responses (zero efficacy). Reversible competitive antagonism is the most common and most important type, which has 2 main characteristics:

3 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

•In presence of antagonist, the agonist log concentration-effect curve is shifted to the right without change in slope or maximum (called surmountable), the extent of the shift being a measure of the dose ratio. •The dose ratio increases linearly with antagonist concentrations; the slope of this line is a measure of the affinity of the antagonist for the receptor.

A non-competitive antagonist mostly involves allosteric antagonism, which can decrease the affinity of an agonist for its receptor or prevent conformation changes for receptor activation. A dose response curve shows depression of the maximal response. No amount of agonist can completely overcome to get the same maximum response. Insurmountable antagonist The maximum effect of the agonist is reduced by either pretreatment or simultaneous treatment with the antagonist. This can occur in several distinct molecular mechanisms such as irreversible competitive antagonism or non-competitive antagonism. Schild analysis is the equation derived by Arunlakshana and Schild in 1959 to construct linear plots (Figure 1.2) designed to graphically estimate the affinity of competitive antagonists.

The Schild equation: log (dr-1)= log XB- log KB dr = concentration ratio

KB = equilibrium dissociation constant for the combination of antagonist B with the receptor

XB= conc. of reversible competitive antagonist B

Figure 1.2 Schild Plot. A graph of log (dr-1) against log XB.

Slope > 1 indicates inadequate equilibrium time. Slope < 1 indicates removal of agonist by a saturation uptake process or agonist is acting at more than one receptor. 4 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

As stated in the literature by Kenakin (1997), the Kb can be calculated for insurmountable antagonists. Kb is calculated from the equation Kb = [B]/Slope-1. By plotting 1/[A] against 1[A’] which A is the agonist concentration while A’ is the same concentration of agonist in the presence of antagonist. The value of calculated Kb will be more accurate when the maximal response to the agonist decreases less than 50% of the maximal response in the presence of antagonist.

1.2 G-PROTEIN COUPLED RECEPTORS

GPCRs are characterized by seven transmembrane alpha helices joined by intracellular and extracellular loop regions (Figure 1.3). GPCRs are membrane receptors that mediate cellular response by coupling to intracellular effector systems via G-proteins e.g. muscarinic acetylcholine receptor (mAChR), adrenergic receptors and chemokine receptors. The number of GPCRs encoded by the human genome has been estimated to be as great as ~800-1000. GPCRs are classified into three main families and many subfamilies on the basis of the pharmacological nature of their ligands and sequence similarity.

(A). (B).

N-terminus

EC1 EC2 EC3

TM1 TM2 TM3 TM4 TM5 TM6 TM7

IC1 IC2 IC3 C-terminus

Figure 1.3 (A). Representation of the crystal structure of Bovine Rhodopsin (pdb: 1U19). (B). Two-dimensional depiction of Bovine Rhodopsin, which composes of seven- transmembrane helices TM1 to TM7 connected by three extracellular (EC) and three intracellular (IC) loops.

5 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Rhodopsin like–receptor (Class I) is the largest and most diverse of these families that consists of rhodopsin, adenosine, melanocortin, neuropeptide, chemokine and others. The secretin- like family (Class II) consists of 25 members including the receptors for the GI peptide hormone family (eg. secretin, glucagon, vasoactive intestinal peptide) calcitonin and parathyroid hormone. The metabotropic glutamate receptor-like family (Class III) is the smallest family and includes the

GABAB receptor, calcium-sensing receptor and some taste receptors (Table 1.1). To date, GPCR-based drugs have become one of the most important in terms of drug discovery potential. However, clinically successful GPCR-based drugs have targeted a small number of known GPCRs (~30%) especially in the bioamine family, which is a subfamily of Class I GPCRs. The complete human genome sequencing reveals unknown GPCRs. GPCRs with unknown endogenous ligand are classified as “orphan”. The identification of new subtypes of known GPCRs and orphan receptor has opened opportunity for novel GPCR-based therapeutics and has drawn attention from the pharmaceutical industry.

Table 1.1 Family classification of known G-protein-coupled receptors (adapted from Chalmers et al., 2002).

Class I (A) Class II (B) Class III (C) Frizzled/ Rhodopsin-like Secretin-like Metabotropic- Smoothened glutamate- receptor-like Amine Acetylcholine Calcitonin Metabotropic Frizzled/ glutamate smoothened α and β Corticotropin- Extracellular adrenoceptors releasing factor calcium-sensing Dopamine Gastric inhibitory Putative peptide pheromone Histamine Glucagon GABAB Serotonin Growth-hormone- releasing hormone Octopamine Parathyroid hormone Trace amine PACAP Peptide Angiotensin Secretin Bombesin Vasoactive intestinal polypeptide Bradykinin Diuretic hormone C5a Latrophilin anaphylatoxin FMet-Leu-Phe Brain-specific angiogenesis inhibitor APJ-like Methuselah-like proteins

6 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR Interleukin-8 CC,CXC,CX3C, XC chemokine Chemokine- receptor-like Cholecystokinin Endothelin Melanocortin Neuropeptide Y Neurotensin Opioid Somatostatin Tachykinin Vasopressin-like Galanin-like Proteinase- actvated-like Orexin and neuropeptide FF Urotensin II Adrenomedulin Endothelin-B- like Neuromedin U Hormone protein Prosta- Prostaglandin noid Prostacyclin Thromboxane Nucleo- Adenosine tide-like Purinoceptor type U, Y, other Cannabinoid Platelet-activating factor Gonadotropin- releasing hormone Thyrotropin- releasing hormone Melatonin Viral Lysosphingolipid and lysophosphatidic acid Leukotriene B4

7 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

1.2.1 GPCR SIGNALLING

G-proteins consist of 3 subunits (α, β, γ), which are anchored to the membrane through attached lipid residues. Coupling of the alpha-subunit to an agonist-occupied receptor causes the bound GDP to exchange with intracellular GTP; the alpha-GTP complex is then dissociated from the receptor and from the βγ complex, and interacts with a target protein; for example it may be an enzyme such as adenylate cyclases, and phospholipases or ion channels. The regulation of these second messengers modulates the level of intracellular mediators for example cAMP, intracellular Ca2+, phosphoinositide and DAG, which in turn activate or inhibit effectors (PKA, PKC and PI- 3K), which lead to a biological response. The βγ complex may also activate a target protein (target 2). The GTPase activity of the α-subunit is increased when the target protein is bound, leading to hydrolysis of the bound GTP to GDP, whereupon the α subunit recombines with βγ (Figure 1.4).

1. Resting State

target target 1 2

! "#

Ligand GDP 2. Agonist binds to receptor

target target 1 2 "# ! GDP

3. Target proteins activated GTP

target target 1 2 ! "#

GTP

4. Hydrolysis of GTP

target target 1 2 ! "#

GDP + P Figure 1.4 The function of G-protein. G-protein consists of three subunits: α, β, γ that are anchored to the membrane. Binding of agonist to receptor causes the exchange of bound GDP for GTP the complex then dissociates from the receptor and from the βγ complex and interacts with a target protein (target 1) such as adenylate cyclase. The βγ complex also activates a target protein (target 2) for example GPCR kinase or mitogen-activated protein kinase. The activation is terminated by increasing of GTPase activity of α-subunit when it is bound to the target protein, which leads to hydrolysis from GTP to GDP. The α-subunit then recombines with βγ. (adapted from Rang et al., 2007). 8 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

A feedback mechanism occurs after the activation of the receptor by agonist that helps to decrease responsiveness and prevent the overstimulation of the receptor. There are three families of regulatory molecules that control the process of desensitization: second messenger-dependent protein kinases, GPCR kinases (GRKs) and arrestins (Ferguson et al., 1996; Lefkowitz, 1993; Zhang et al., 1997a). Desensitization is started by phosphorylation of activated G-protein coupled receptors by GRKs and/or second messenger-dependent protein kinases, which promotes the translocation of the cytosolic adaptor arrestin proteins to the membrane. Arrestin proteins then bind to the receptor and promote internalisation. GPCRs can be divided into two classes (A or B) according to their affinity for the β-arrestin. Class A receptors bind to β–arrestin 2 with higher affinity than β–arrestin 1 while Class B receptors bind to both with the same affinity. Class A receptors dissociate from β–arrestin soon after internalization. Class B receptors remain associated with β–arrestin throughout receptor internalization. Class A receptors can return to the cell surface rapidly, in approximately 30 mins. In contrast, class B receptors remain associated with β–arrestin and the endosome for around 1 hour after activation (Oakley et al., 2001; Oakley et al., 2000).

Biological Amino acids Lipids, amines and Ions Peptides & Proteins and Others !"#$%&'(()(%$

Biological Response eg. proliferation, di!erentiation etc.

a *+#$%&'(()(%$ a _ ` `

GDP Ion channels, G_ = 16 subunits PI3Ka, G` = 5 subunits PLC-` Ga = 12 subunit adenylyl cyclases Nucleus

_ i _q _s GTP GTP GTP

Adenylyl cyclases Adenylyl cyclases Inhibition of cAMP PLC-`, DAG, Increase cAMP production, concentration Ion channels Ca2+ Phosphodiesterases PKC Phospholipase

Figure 1.5 GPCR complex with potential activators and diverse biological functions (Figure adapted from Marinissen et al., 2001). 9 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

1.2.2 CLASSIFICATION OF G-PROTEINS AND INTRACELLULAR PATHWAYS

G-protein-mediated signal transduction results in many different biological functions in the human body. There are numerous subtypes of G-proteins, which are commonly classified by their alpha subunits. This subunit is divided into 4 families (Gαs, Gαi/Gαo, Gαq/Gα11, Gα12/Gα13)(Figure 1.5 and Table 1.2). Unlike alpha-subunits (16 subunits) there is only a small range of β and γ- subunits: five β-subunits and twelve γ-subunits.

(1) Gαs couple to amine receptors for catecholamines, histamine and serotonin as well as many other receptors and stimulates adenylyl cyclase, causing increased cAMP formation (Table 1.2).

(2) Gαi is the first subtype in the Gi family, which is designated with the “i” letter due to its ability to inhibit adenylyl cyclase, which decreases cAMP formation. Gαi is associated with opioid and cannabinoid receptor amongst others. It is blocked by pertussis toxin, which prevents dissociation of αβγ. Gαo has limited effects from the alpha subunit, it effects mainly from the βγ subunit. Like Gαi its effects are blocked by pertussis toxin, which occurs mainly in the nervous system.

(3) Gαq is associated with amine, peptide and prostanoid receptors. It activates phospholipase C, increasing production of the second messengers inositol (1,4,5) triphosphate (IP3) and diacylglycerol (DAG).

(4) Gα12/Gα13 is activated by certain GPCRs. The activation occurs by the exchange of GDP for GTP, which facilitates the activation of Rho-GDP. Rho is activated and in turn activates Rho kinase. Rho kinase can phosphorylate a variety of substrate proteins and can control a range of cellular functions such as smooth muscle cell contraction, proliferation, angiogenesis and synaptic remodelling. Gβγ subunit of G-protein coupled receptor associates with similar target protein for Gα subunits. It activates potassium channels, inhibits voltage-gated calcium channels, activates GPCR kinase, and activates mitogen-activated protein kinase cascades. Many Gβγ isoforms have been identified but specific functions are not yet known.

10 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Table 1.2 Classification of Gα subunits and their effectors

Family Subtype Effector

Gα Gαs ↑ Adenylyl cyclase

Gαolf ↑ Adenylyl cyclase

Gi Gαi1 ↓ Adenylyl cyclase

Gαi2 Rap 1 GAP Gαo ↑ GTPase of tubulin, ↑c-Src 2+, + Gαz ↓Ca ↑K channels

Gαt ↑ cGMP-PDE

Gαg unknown

Gq Gαq ↑ PLCβ Gα11 ↑ GRK 1-3

↑ Bruton’s tyrosine kinase (Gαq) ↑ p63-RhoGEF ↑ K+ channels + + G12 Gα12 ↑ Na /H exchaner ↑ PLD

Gα13 ↑ p115RhoGEF ↑ iNOS ↑ PLCe ↑ E-cadherin

PDE, phosphodiesterase E; iNos, inducible nitric oxide synthase, PLD, phospholipase D; GEF, guanine nucleotide-exchange factor. (Table adapted from Cabrera-Vera et al., 2003)

11 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

G-protein *+#$%&'(()(%$ dependent Desensitized signaling Receptor

Ca2+ Adenylate Channels cyclase

P Ca2+ --P PKA --P ATP --P--P cAMP + Gs Gi Arrestin _ PKC GRK PDE Internalisation via clathrin-mediated endocytosis

Recycling back to membrane G-protein independent ERK 1,2 signaling --P --P Arrestin MAP kinase Targeted for degradation pathway in Lysosomes !"#$%&'(()(%$ Gene expression

Figure 1.6 Signal transduction in G-protein-coupled receptors. The type 2 beta-adrenergic receptor regulated diverse signaling pathways. (Figure adapted from Rosenbaum et al., 2009).

Targets for G-proteins (Figure 1.6) • Adenylyl cyclase/cAMP: adenylyl cyclase catalyses formation of the intracellular messenger cAMP. cAMP activates various protein kinases that control cell function in many different ways by causing phosphorylation of various enzymes, carriers and other proteins.

• Phospholipase C/ inositol triphosphate (IP3)/diacylglycerol (DAG): Catalyses the

formation of second intracellular messengers, IP3 and DAG, from membrane 2+ 2+ phospholipid. IP3 acts to increase free cytosolic Ca by releasing Ca from intracellular compartments. Increased free Ca2+ initiates many events, including contraction, secretion, enzyme activation and membrane hyperpolarization. DAG activates protein kinase C, which controls many cellular functions by

12 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR phosphorylating a variety of proteins. Receptor-linked G-protein also regulates

Phospholipase A2 (formation of arachidonic acid, eicosanoids). • Ion channels (e.g. potassium and calcium channels) thus affecting membrane excitability and transmitter release etc.

1.3 COMPLEMENT AND INFLAMMATION

Inflammation is a local protective event. It causes a complex series of events including dilation of arteries, capillaries and venules, increased vascular permeability, increased blood flow and exudation of fluids and plasma proteins. These processes occur rapidly and are followed by adhesion of leukocytes to the vascular endothelium with subsequent influx of cells into the surrounding tissue. Complement is involved in inflammatory response. The direct effects of certain complement activation products, anaphylatoxins, can also stimulate the synthesis and/or secretion of many other inflammatory mediators. In 1920, the first four complement components were described (C1, C2, C3 and C4) and later in 1958 five more components had been identified. The letter C has been used to designate complement components along with number corresponding to the sequence in which the protein components are discovered, although, this number did not represent their activation sequence. The sequence of activation starts from C1 to C9 except for C4, which activated after C1 and before C2. C1 is composed of three subcomponents designated C1q, C1r and C1s. The cleavages of complement components resulting in fragments are designated by the name of the original component (e.g. C3) followed by a lowercase letter such as C3a and C3b. Inactive fragments have the lowercase letter “i” before the abbreviation such as iC3b.

1.4 COMPLEMENT CASCADE

The complement system is a key component of the immune defence against infectious organisms (bacteria, viruses, parasites) or tissue damage caused by chemicals, physical damage, radiation or neoplasia. The complement network is composed of over 30 soluble plasma proteins and receptors that play key roles in both innate (non-specific) and acquired (specific) immune responses (Sunyer et al., 2005). Activation of complement enables the host to distinguish between self and non-self, tagging the latter with complement proteins that facilitates lysis of bacteria and damaged cells as well as their elimination from the host. Complement activation is tightly regulated process in which it is activated in a sequential or cascade reaction. Complement activation is the result of the assemblage of individual components each of which activates the next component. 13 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

The cascade is completed when all nine in sequence of complement component (C1, C4, C2, C3 to C9) are activated. There are three distinct pathways called the classical pathway, the mannose-binding lectin pathway and the alternative pathways, which activate complement cascade as illustrated in Figure 1.7. The classical pathway is activated by antigens that bind to the antibodies (or antibody- dependent) IgM and certain IgG3 and IgG1 antibody isotypes which are the most efficient activators, resulting in rearrangment of Fc-conformation which increases affinity of the Fc domains of antibody for C1q - the first component of complement (Lachmann et al., 1984). The classical pathway has 3 phases: (1) the initiation phase (involving C1, C4 and C2) (2) the amplification phase (involves C3) and (3) the membrane attack phase (involves components C5b, C6, C7, C8 and C9). Once C1q is activated by binding to a specific antigen, the C1 binding sites are exposed to the Fc region of antibodies. C1 is a Ca2+-dependent complex of three polypeptides: (1) C1q, a protein with six immunoglobulin binding sites linked by collagen-like fibrils, which in turn activates through rearrangement the structure of (2) C1r and then (3) C1s, a serine protease that goes on to cleave C4 and then C2. C4 is cleaved by C1s into fragments C4a (anaphylatoxin) and C4b. In the presence of Mg2+ ions, the bound C4b interacts with the C-terminal C2a which renders C2 susceptible to cleavage by C1s. C2 is cleaved by C1s into two fragments C2b (small) and C2a (large). C2b diffuses while C2a remains bound to C4b. Joining of C4b and C2a and attachment to antigen-antibody complex leads eventually to the formation of C4b2a (or C3 convertase) that cleaves the component C3 into C3a and C3b. Some of the C3b is deposited on the membrane of the target cell (or called opsonises), where it acts as a site for the attachment of phagocytic cells and acts as an opsonin. In contrast, activation of the Mannose-binding lectin pathway is initiated independently of immunoglobulins, by mannose-binding lectin-associated serine proteases (MASP). The activation of serine protease MASP-1 and MASP-2 (structurally and functionally similar to C1r and C1s, respectively) are triggered by the binding of mannose-binding lectin (MBL; structurally similar to C1q) to mannose-containing carbohydrate moieties (sugar structures such as mannose, fructose, glucose and N-acetyl-glucosamine) on the microbial, or viral cell surfaces but not in mammalian molecules. The complex of MBL and MASP-1 and MASP-2 then cleaves and activates C4 and C2 in a similar way to the classical pathway (Monk et al., 2007; Yang, 2006). The lectin pathway is sometimes included as part of the classical pathway because it produces the same C3 convertase (C4b2a) following C3 being cleaved into C3a and C3b and then the C3b binding to the C3 convertase to form C5 convertase.

14 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Classical Pathway Lectin Pathway Alternative Pathway IgG and IgM containing Bacterial Surfaces Pathogen surface, LPS Ag-Ab complexes

MBL or Ficolin C3.(H2O) C1q and (C1r-C1s)2 MASPs Factor B

C4, C2 C3bB Factor D Ampli!cation loop C4b, C2a Ba C3 convertase C4b2a C3bBb

Anaphylatoxin Anaphylatoxin C3a C3 C3a

opsonin C3b C3b

C5 convertase C4bC2aC3b C5 C3bBb3b

Anaphylatoxin C5a C5b

C5bC6C7

C5bC6C7C8(C9)n Membrane Attack Complex

Figure 1.7 Complement activation pathways (Figure adapted from Sarma et al., 2011).

The alternative pathway is a more direct pathway that is activated in the absence of antibodies. The alternative pathway is initiated by many microbial cell wall components, such as zymosan (fungi), LPS (endotoxin) and teichoic acid, as well as other foreign surfaces and also protein A, C-reactive protein, cobra venom factor, polysaccharides and damaged tissue (Ganter et al., 2007; Gasque, 2004). The biological relevance of the alternate pathway is its ability to provide an amplification loop for the classical pathway. The mechanism of initiation of the pathway is not clearly understood, however it is thought to involve continuous low-rate hydrolysis (“tick-over”) of

15 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR a thioester bond within the complement protein C3 to produce a molecule, C3(H2O) which has structure and function that resembles C3b but is unstable unless it binds to the microbial cell membrane surface. In healthy cells, there are various control mechanisms that prevent the continuous activation of C3 and undesirable effects from complement activation. In the presence of factor B, D and properdin, C3(H2O) is converted to C3bBbP, which is the C3 convertase of the alternative pathway (Yang, 2006). Factor B is structurally and functionally similar to C2. Factor D cleaves factor B into two fragments, Ba and Bb. The biological role of Ba is unknown. Bb binds to C3b, which become susceptible to Factor D. Factor D acts similarly to C1s (serine protease). It has no activity on Factor B until the binding of Factor B to C3b, which forms the C3 convertase of the alternative pathway. Properdin is a gamma-globulin consisting of identical subunits which are held together by a covalent bond. There are two forms of properdin, native and activated, which differ from each other only by a small conformational change. The native properdin acts indirect by slowing the dissociation of Bb from C3b, which stabilizes C3bBb complex extending the half-life 10-fold (Ehrnthaller et al., 2011). The activated form of properdin binds to C3b (in the absence of Factor B) and promotes the assembly of C3bBb. After C3 is cleaved by C3 convertase, C3b binds to C3 convertase (either C4b2a3b or C(3b)nBbP) forming C5 convertase which cleaves C5 into C5a and C5b. C5b binds to C6, C7 and C8 and multiple C9s to form the membrane attack complex (MAC), C5b678(9)n, which perforates membranes of micro-organisms by generating a cylindrical structure. The membrane pores lead to osmotic imbalance resulting in the lysis of the target cell and subsequent elimination from the body. This tagging (opsonisation) by C3b and synthesis of convertases directly on foreign antigen surfaces enables the host to eliminate pathogens and damaged cells without creating too many potent pro-inflammatory complement proteins that might otherwise damage the host. The cleavage of C3 by C3 convertase or C5 by C5 convertase occurs at a site close to the N- termini of their α-chains (Figure 1.8) (Frank, 1997; Hugli, 1986; Muller-Eberhard, 1988; Volanakis et al., 1998) and leads to the generation of large fragments, C3b or C5b along with small activation fragments, C3a and C5a which are potent anaphylatoxins (Figure 1.7 and 1.8). These anaphylatoxins have the ability to trigger degranulation of mast cells, basophils and neutrophils, and therefore release vasoactive substances such as histamine, prostaglandins, kinins and serotonin. Also, they are potent chemotactic agents in the recruitment of immune cells to the sites of infection or damage. While C5a is the more potent mediator of leukocyte degranulation or phagocyte chemotaxis and many other functions, C3a is present at an approximately 20-fold higher concentration in plasma than C5a (Chenoweth et al., 1979; Ember et al., 1997; Gerard et al., 1994). Recently, there were reports that the complement acts not only in the “inflammation processes” but is also involved with the coagulation cascade (Amara et al., 2008; Laudes et al., 2002; Markiewski 16 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR et al., 2007b) as well as in the regulation of apoptosis (Guo et al., 2000; Markiewski et al., 2007a; Perianayagam et al., 2002; Riedemann et al., 2002) and cell growth (Markiewski et al., 2008).

C3 (195 kDa)

H2N COOH _-chain, 120 kDa) s s s s

HOOC NH2 `

C3a COOH (9kDa) s s s s C3b (185 kDa) Carboxypeptidase N NH2 C3a desArg

Figure 1.8 Cleavage of the alpha-chain of C3 by C3 convertase leads to the release of C3a anaphylatoxin (Figure adapted from Wagner et al., 2010).

1.5 REGULATION OF COMPLEMENT SYSTEM

Complement is an innate immunity of the body, which is non-specific, and it can attack both the pathogenic or damaged tissue and healthy host tissue. Thus, the body has developed the complement regulators to maintain the body homeostasis, which are divided into two categories: fluid phase and membrane-bound complement regulatory proteins. Fluid phase regulators: The most well known is C1 inhibitor (C1-INH or serpin1) which acts as a serine protease inhibitor of the classical and lectin pathways by inhibiting C1r/s and MASPs, respectively. In addition, there are other regulators in this group eg. Factor I, Factor H, C4b-binding protein (C4BP), Factor H-like protein 1 (FHL1) and carboxypeptidase N. Factor I is a protease which help to degrade C3b and C4b. Factor H recognizes self-surfaces, accelerates convertases decay and acts as a cofactor for Factor I. C4BP accelerates decay of lectin and classical convertases and also acts as cofactor of factor I in degradation of C3b. FHL -1 is a truncated homolog of Factor H which accelerates convertase decay and acts as a cofactor for factor I. Carboxypeptidase-N degrades C3a and C5a to their inactive forms (or desArg). Factor H, FHL-1 and C4BP are

17 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR important for complement regulation as they serve for self-recognition and prevention of self- damage. Membrane-bound regulators: Complement receptor 1 (CR1) is also known as CD35. The function of CD35 is to facilitate the decay of C3 and/or C5 convertase and also acts as cofactor for factor I and induces phagocytosis. Membrane cofactor protein (MCP; CD46) acts as a cofactor for factor I. Decay accelerating factor (DAF; CD55) increases the decay of C3 convertases in both classical and alternative pathways. CD59 (protectin) regulates of membrane attack complex (MAC) formation.

1.6 THERAPEUTIC POTENTIAL FOR COMPLEMENT INHIBITORS

Even though the complement proteins and regulators have been identified a long time ago, only a few complement inhibitors are developed or approved for clinical use. There is a summary of current complement inhibitors that approved for intervention of complement cascade on the market or in clinical trials (Table 1.3). The complement inhibitors are not important for therapeutic purpose but can be used as tools for physiological/pharmacological studies where the key important roles of complement is in in vivo models. Recently, the FDA approved human C1 inhibitor (Cinryze, Viro Pharma) for C1 inhibitor deficiency leading to hereditary angioedema (Cocchio et al., 2009; Davis, 2006) and Eculizumab (Alexion), which is a monoclonal antibody specific for C5. It effectively inhibits C5a and MAC formation. It was approved for the treatment of paroxysmal nocturnal hemoglobinuria (PNH) (Rother et al., 2007a). Serine protease inhibitor: As the complement activation pathway contains various serine proteases, it becomes one of the targets for complement-specific treatment. Nafamostat or FUT- 175 (Futhan) is a broad spectrum, non-specific serine protease inhibitor; it can inhibit pancreatic and coagulation enzymes as well as complement activation (classical and alternative pathways)(Schwertz et al., 2008). It is currently used for the treatment of acute pancreatitis and for prevention of thrombosis in disseminated intravascular coagulation (DIC) and extracorporal circulation. No other serine proteases have been developed because of the lack of selectivity and short half-life (Ricklin et al., 2007). Complement regulator: C1 inhibitor is an inhibitor for C1r, C1s and MASP2 as well as factor XIIa and kallikrein in the coagulation system. Currently, it is used in the clinic only for C1 inhibitor deficiency leading to hereditary angioedema (HAE)(Davis, 2006). HAE is an autosomal 18 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR disease due to the mutations and deficiency of C1 inhibitor resulting in dysregulation of kinin system activation. This leads to generation of bradykinin (potent vasoamine) and increased vascular permeability, therefore leading to the angioedema condition in HAE (Agostoni et al., 2004; Cicardi et al., 2005; Cugno et al., 2009). sCR1 or TP10 (Avant Immunotherapeutics, Needham, MA, USA) is a soluble complement regulator, it had a high potency in inhibiting the classical and alternative pathways which gave promising results in the treatment of ischemia/reperfusion injury and other conditions. TP10 was developed and used as a therapeutic after coronary artery bypass graft surgery. However, the development was discontinued in the clinical trial due to the lack of benefit in female patients (Li et al., 2006; Ricklin et al., 2007). A hybrid compound of complement regulators DAF and MCP has been developed (Sahu et al., 2000). The initial name was “complement activity blocker 2” (CAB2) but later entered to clinical trial under the name MLN-2222 which was used for the treatment of coronary artery bypass grafting (Sahu et al., 2000). Complement component inhibitor: Compstatin is a cyclic tridecapeptide, which was discovered by screening a phage peptide library. It’s binding to C3 and results in preventing the cleavage of C3 into C3a and C3b in both classical and alternative pathways. It has shown to be effective in various experimental disease models. Compstatin derivatives have been developed to improve the potency and have also shown safety and effectively in in vivo experiments. One compstatin derivative under the name POT-4 (Potentia Pharmacuticals, Inc.) has completed phase I clinical trial for treatment of AMD. C5 specific antibody: Eculizumab is a monoclonal antibody (mAb) against C5, which effectively inhibits C5a and membrane attack complex formation (Rother et al., 2007b). It was the first and only approved by FDA for the treatment of paroxysmal nocturnal hemoglobinuria (PNH), which is the life-threatening disorder, characterised by a chronic lysis of red blood cells (Parker, 2009; Rother et al., 2007b). Furthermore, it was investigated for the treatment of systemic lupus erythrematosus under clinical trial phase I (Robak et al., 2009). Pexelizumab is another C5-specific monoclonal antibody and has a short-acting effect. It is beneficial in reducing mortality rate after cardiopulmonary bypass surgery (Mahaffey et al., 2006). Neutrazumab (G2 Therapies, Darlinghurst, NSW, Australia) and TNX-558 bind to C5a and cause the inhibiton of C5a to its receptor. TNX-234 and TA 106 have been developed against factor D and factor B, respectively. One of the major drug targets in the complement cascade is C5a. C5a or anaphylatoxin is a potent pro-inflammatory which causes many complement–associated diseases when there is excessive formation. Thus, the inhibition of binding of C5a to its receptor is promising to reduce the 19 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR undesired inflammatory response without interfering with the defensive mechanism of the complement system. Complement target Anaphylatoxins: PMX-53 is a cyclic peptidomimetic C5aR antagonist, which has a small molecular size (under 1 kDa) and high oral bioavailability. However, it has a short-half life, which may limit its use. It exerts its beneficial effect in animal studies for I/R injury, rheumatoid arthritis, IBD and neurodegenerative disease (Arumugam et al., 2003; Woodruff et al., 2006; Woodruff et al., 2005).

Table 1.3 Complement inhibitors approved for intervention of complement cascade on the market or in clinical trials. Product (company) Description Mechanism of action Indication and status of development Cinryze (Viro Pharma) C1-INH, Control of bradykinin Approved in US for the serping1 generation prophylaxis of HAE in concentrate adolescents and adults Cetor (Sanquin) C1-INH Inhibit C1r/C1s, control Approved for HAE and of bradykinin generation preclinical or phase 1 for other eg. AMI, CABG Berinert (CSL Behring, C1-INH Inhibit C1r/C1s, control Approved for acute HAE in Lev Pharma of bradykinin generation Europe and US Rhucin (Pharming) rhC1-INH Controls bradykinin Clinical phase 3 for HAE generation sCR1/TP10 (Avant Soluble Decay accelerator, In clinical phase 2 for Immunotherapeutics) complement Factor I cofactor CABG regulator/ extracellular part of CR1 CAB2 or MLN-2222 Hybrid of Decay accelerator, In clinical phase I for (Millenium DAF and Factor I cofactor CABG Pharmaceuticals) MCP Eculizumab (Soliris, C5-specific Block C5 cleavage Approved for PNH and Alexion Pharmaceuticals) mAb of clinical trial for other eg. human (long aHUS and acute Ab- acting) mediated renal allograft rejection

20 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR Pexelizumab (Alexion mAb against Block C5 cleavage In clinical phase III for Pharmaceuticals) C5 (short- AMI and CABG acting) Ofatumumab (Genmab) mAb against Stimulation of In clinical trial phase II for CD20 complement-dependent treatment of RA, B-cell (human) cytotoxicity chronic lymphocytic leukemia and follicular lymphoma Compstatin/POT-4 Cyclic C3 inhibitor/ blocks the Completed clinical trial (compsatin derivative) peptide of 13 cleavage of C3 into C3a phase I for AMD (Potentia) residues and C3b PMX-53 (Arana) Cyclic C5aR antagonist Completed clinical phase II hexapeptide for RA and psoriasis TNX-234 (Genentech) Factor D- Block the alternative In clinical phase I for AMD specific Ab pathway rhMBL (Enzon) Recombinant Activated lectin In clinical phase Ib for human MBL pathway MBL deficiency, progenitor and stem cell transplant and liver transplant Nafamostat or FUT-175 Unspecific serine Used clinically for acute (Futhan) protease inhibitor pancreatitis and prevention (broad-spectrum) of thrombosis in DIC and extracorporal circulation. Available in Japan ARC1905 Aptamer- Block the cleavage of In clinical trial phase I for based C5 C5 into C5a/C5b AMD (for intravitreal inhibitor application) HAE, hereditary angioedema; AMI, acute myocardial infarction; AMD, age-related macular degeneration; CABG, coronary artery bypass grafting; PNH, paroxysmal nocturnal nemoglobinuria; RA, rheumatoid arthritis; DIC, disseminated intravascular coagulation, mAB, monoclonal antibody; MBL, mannose binding lectin, aHUS, atypical haemolytic uremic syndrome.

21 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

1.7 C3a ANAPHYLATOXIN

1.7.1 STRUCTURE OF ANAPHYLATOXIN C3a

The anaphylatoxins C3a, C4a and C5a are pro-inflammatory 74 - 77 residue bioactive peptide fragments (~9 kDa) cleaved from the serum proteins C3, C4, and C5 during activation of the complement cascade. The C3 convertase cleaves C3 into small C3a and large C3b fragments. C3a is a small protein of 77 amino acid residues. The amino acid sequences of C3a of human, pig and rat origin are known and show extensive homology (Huber et al., 1980; Hugli, 1990). Anaphylatoxins consist of a globular core of four anti-parallel α-helices stabilized by three intra-chain disulfide bonds (Figure 1.9 and 1.10). The N-terminus of the protein is responsible for conferring affinity to anaphylatoxins for their receptors, while the C-terminal pentapeptide (LGLAR in C3a and MQLGR in C5a) plays an important role in activating the receptors (Gerardyschahn et al., 1989; Hugli, 1986; Muller-Eberhard, 1988).

Figure 1.9 Crystal Structure of C3a. The human C3a consists an N-terminal domain intra- connected by three disulfide bonds (Cys22-Cys49, Cys23-Cys56 and Cys36-Cys57) and a helical C-terminal domain (Figure adapted from Huber et al., 1980).

22 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Figure 1.10 Complement anaphylatoxin sequences. Cysteines in the three-disulfide bonds are conserved and boxed above. Sequence homology of C3a, C4a and C5a shows common residues in C3a, C4a and C5a in black, common residues to C3a and C5a in blue, common residues to C3a and C4a in red and common residues to C4a and C5a in green (permission from Blakeney, 2007).

1.7.2 PHYSIOLOGY OF ANAPHYLATOXIN C3a

The lifetime of C3a and C5a in vivo is limited by carboxypeptidase-N (Bokisch et al., 1970; Campbell et al., 2002; Ember et al., 1997), which cleaves off the carboxy terminal arginine residue from both C3a and C5a leading to the desArg derivatives. C5a-desArg still binds the C5aR but retains only 1-10% of C5a’s anaphylactic activity, whereas C3a-desArg loses its ability to bind to C3aR and is thus inactive. In addition, C3a desArg and acylation-stimulating protein (ASP) are identical and also they are identical to C3a except for missing the arginine at the C-terminal. Recently, it was reported that ASP has potent anabolic effects on human adipose tissue where it stimulate glucose uptake, fatty acid storage (Cianflone et al., 1999; Cianflone et al., 2003; Murray et al., 1999; Xia et al., 2004) and triglyceride synthesis by binding to C5L2 receptor (Kalant et al., 2005). C3a, C4a and C5a anaphylatoxins share a high degree of homology as well as overlapping functions in the generation of an inflammatory response. C3a and C5a are expressed on the same cells, such as leukocytes and non-myeloid cells (endothelial and epithelial cells).

23 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

C3a and C5a are known for their stimulatory effects on leukocytes, including chemotaxis of neutrophils, eosinophils, monocytes/macrophage and mast cells (Daffern et al., 1995; Fernandez et al., 1978; Hugli, 1986; Hugli, 1990; Hugli, 1984). However, C5a is a more powerful chemoattractant than C3a; the predominance of C3a in the induction of chemotaxis is only on eosinophils and mast cells but not on neutrophils (Daffern et al., 1995; Hartmann et al., 1997). In a previous report, C3a activation caused release of effectors from eosinophils which in turn activate neutrophils (Daffern et al., 1995). C3a can stimulate the release of serotonin from guinea pig platelets (Fukuoka et al., 1988), modulate synthesis of IL-6 and TNF-α by B-lymphocytes and monocytes (Fischer et al., 1997; Fischer et al., 1999), and also stimulate the release of interleukin-1 (IL-1) from monocytes (Haeffnercavaillon et al., 1987) as summarised in Table 1.4. The anaphylatoxic effect is due to the activation of anaphylatoxin receptors on mast cells and basophils leading to the degranulation and release of histamine and other vasoactive or inflammatory mediators, such as arachidonic acid metabolites (prostaglandins, leukotrienes, lipoxins), lipid mediators (platelet-activating factor) and the nucleoside adenosine (Ember et al., 1998; Frank, 1997; Hugli, 1986; Muller-Eberhard, 1988; Volanakis et al., 1998). C3a also induces the release of reactive oxygen species in human eosinophils (Elsner et al., 1994a) as seen in Table 1.4. Furthermore, anaphylatoxins can cause smooth muscle contraction i.e. bronchial, intestinal and uterine smooth muscle (Hugli et al., 1975), increased vasopermeability (Cochrane et al., 1968; Vallota et al., 1973; Wuepper et al., 1972), vasodilation, release of cytokines and pro-inflammatory mediators and induce mucous secretion from globlet cells (Hartmann et al., 1997; Klos et al., 1992; Norgauer et al., 1993), bronchial, gastric and intestinal epithelial cells. When macrophages are exposed to C3a, they release interleukin-1 (Becker et al., 1978) and thromboxane A2 (Zanker et al., 1982).

1.7.3 PATHOLOGY OF ANAPHYLATOXIN C3a

C3a has been implicated in allergic bronchospasm and pulmonary inflammation when C3a is administered via intrabronchial instillation (Hoffmann et al., 1988; Stimler et al., 1980). There is some evidence for an implied role for C3a in allergic asthma: a) there is abolition of airway hyper- reactivity in mice lacking C3 or C3aR (Humbles et al., 2000); b) in ovalbumin-sensitized and – challenged mice, the C3 and C3aR proteins are up-regulated (Drouin, 2001); c) in bronchoalveolar lavage fluid and serum of asthmatics challenged by allergens there were increased levels of C3a. There are also some new roles for C3a and C5a, other than immune functions, such as increasing prostanoid formation and glucose output in perfused rat liver (Puschel et al., 1993;

24 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR Puschel et al., 1996). Recently, the study has shown that signalling through C3aR positively regulates basal neurogenesis and C3 difficient mice have impaired ischaemia-induced neurogenesis despite a large infarct area (Rahpeymai et al., 2006). In addition, C3a and C5a have been shown to protect neuronal cells against excitotoxicity (Osaka et al., 1999; Pasinetti et al., 1996; van Beek et al., 2001). Anaphylatoxins contribute to inflammatory responses that lead to elimination of invading pathogens or induction of adaptive immunity. Genetic deficiency of either C3 or C5 results in susceptibility to bacterial infection. The anaphylatoxins are associated with all types of inflammation. However, when in excess they can be harmful and play roles in pathogenesis of many disease conditions (Table 1.5), such as anaphylaxis, adult respiratory distress syndrome (ARDS)(Zilow et al., 1990), acute transplant rejection, ischemia reperfusion, sepsis (Drouin, 2001), brain inflammation and many autoimmune diseases; for example, psoriasis, atopic dermatitis (Kapp et al., 1985b), rheumatoid arthritis, and systemic lupus erythematosus (SLE) (Clancy, 2000; Volanakis et al., 1998), to name just a few. Therefore, effective antagonists for these anaphylatoxin receptors could be potential treatments for various inflammatory diseases.

25 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Table 1.4 Functional activity of C3a on myeloid and non-myeloid cells.

Cell type Activity Reference

Eosinophils Activates chemotaxis (Daffern et al., 1995) Degranulation (Takafuji et al., 1994) Activates reactive oxygen radical species (Elsner et al., 1994a) production Mobilisation of [Ca2+]i (Elsner et al., 1994a) Mast cells Chemotaxis (most effective) (Nilsson et al., 1996) Degranulation and histamine release (except (Ellati et al., 1994; Mousli et al., 1992; Nilsson for lung mast cells). et al., 1996) This is PTX sensitive (Zwirner et al., 1997) Neutrophils Induces release of lysosomal enzyme (Showell et al., 1982) Induces production of reactive oxygen species (Elsner et al., 1994b) Induces transient aggregation (Nagata et al., 1987a; Nagata et al., 1987b) Monocytes/ Induces release of PGE2 (Morgan, 1987) Macrophages Modulates pro-inflammatory cytokines (TNFα (Takabayashi et al., 1996) & IL-1β) in LPS-stimulated adherent monocytes at local inflammation sites (also C3adesArg) Modulates synthesis of IL-6 (as well as C3a- (Fischer et al., 1997; Fischer et al., 1999; desArg) and TNF-α by B-lymphocytes and Takabayashi et al., 1998; Zwirner et al., 1997) monocytes Releases [Ca2+] and is sensitive to PTX (but not increased intracellular [Ca2+] with stimulation by C3adesArg)

26 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Basophils Stimulates histamine and leukotriene release (Bischoff et al., 1990) dependent on presence of IL-3 T-cells Mediates T cell suppression, overcome by IL-2 (Morgan et al., 1985)

B-cells Suppresses IL-6 and TNF-α (Fischer et al., 1997) Platelets Induces aggregation and serotonin release (Damerau et al., 1986; Polley et al., 1983)

Natural killer cells Inhibits activity of NK cells (Charriaut et al., 1982)

Skin fibroblast and Increases triglyceride synthesis (Baldo et al., 1993; Esterbauer et al., 1999) adipocytes Glucose transport (C3adesArg) Epithelial cells Up-regulates interleukin-8 (IL-8) which can be (Monsinjon et al., 2001) inhibited by pertussis toxin

27 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Table 1.5 Roles for C3a in diseases.

Organ/ Condition/disease C3a Details Reference System levels Respiratory Asthma ↑ In asthma exacerbation, C3a and C5a were shown to promote smooth muscle cell contraction, (Bautsch et al., 2000; mucus hypersecretion and recruitment and degranulation of inflammatory cells. Genetic deletion Humbles et al., 2000; of C3aR mice was protected against the bronchoconstriction and airway hyper-responsiveness Krug et al., 2001; Mousli after allergen challenge. et al., 1992; Takafuji et In mild asthma a significant increase of C3a and C5a was found in BAL fluid after allergen al., 1994) challenge. C3a can stimulate the release of histamine and leukotrienes from basophils and mast cells as well as regulate synthesis of eosinophil cationic protein (ECP) by eosinophils. Ovalbumin-induced ↑ Up-regulation of C3aR on bronchial smooth muscle cells increases airway resistance. C3a is (Bautsch et al., 2000; asthma involved in early and late phases of asthma and is also involved in generating Th2 response in Drouin et al., 2002; asthma. Drouin et al., 2001; Humbles et al., 2000; Kohl, 2001) Adult respiratory distress ↑ The C3a: C3 ratio is a useful laboratory parameter for assessing and monitoring patients at risk (Zilow et al., 1990) syndrome for ARDS. C3a increase during the first few hours correlates with the development of ARDS in poly trauma patients. The complement activation in the early phase occurs predominantly via the alternative pathway.

Chronic Obstructive ↔ There was no significant difference in C3a/C3a desArg or C4a/C4a desArg measurements (Marc et al., 2004) Pulmonary disease between patients with COPD, asthma, and healthy. Skin Psoriatic lesion ↑ C3a levels were significantly increased which suggests that there is a continuous activation of the (Kapp et al., 1985b) complement system leading to the generation of inflammatory mediators. Atopic Dermatitis ↑ Circulating T cells from patients suffering from severe inflammatory skin diseases expressed (Kawamoto et al., 2004; C3aR whereas no expression of C3aR could be found in unstimulated T lymphocytes from mild Werfel et al., 2000) inflammatory skin disease patients or healthy. C3a-C3aR interactions play a role in AD. Leukocytoclastic ↑ Increase in C3a and C4a. (Terui et al., 1987) vasculitis Eczematous dermatitis ↑ Increase in C3a but not C4a indicated alternative pathway activation and delayed type (Terui et al., 1987) and papuloerythroderma hypersensitivity reactions are thought to be responsible for the induction of skin changes in these skin diseases.

28 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Cardio- Myocardial reperfusion Sequestration of C3a des-Arg in coronary circulation. (Semb et al., 1990) vascular Preeclampsia ↑ Elevated levels of C3a and C5a in preeclampsial pregnant women. (Haeger et al., 1989; Haeger et al., 1991) Atherosclerosis ↑ C3a induces changes in biophysical characteristics of fibrin networks, thinner fibres with (Oksjoki et al., 2007; increased tensile strength, which become less permeable. Complement activation (↑C3a) induces Shatstseytlina et al., major changes in fibrin structure, which in turn can induce further activation of the complement 1994) system. The positive feedback system may play an important role in establishing a fibrin infrastructure responsible for the progression of atherosclerosis. Endotoxemia ↑ Increase Up-regulation of C3aR (Drouin, 2001)

Sepsis ↑ Up-regulation of C3aR on bronchial epithelial, alveolar epithelial, pulmonary blood vessel and (Drouin, 2001; bronchial smooth muscles. Absence of the C3aR leads to increased susceptibility to LPS Kildsgaard et al., 2000; treatment thereby indicating anti-inflammatory effects of the C3a-C3aR interaction. Kohl, 2001)

Anaphylaxis ↑ C3a caused bronchoconstriction and hypotension, which occurs from release of histamine from (Regal, 1997) mast cells and basophils. In contrast, C5a caused bronchoconstriction, which was independent of histamine and dependent on products of arachidonate metabolism. Urinary Henoch-Schonlein ↑ C3a levels correlate with plasma creatinine level, therefore are sensitive indicators of the disease. (Abouragheb et al., system nephritis 1992) IgA nephropathy ↑ Elevated C3a levels (Janssen et al., 2000) Lupus nephritis ↑ C3aR was detected in glomerulus of lupus-nephritis patients and the intensity of C3aR correlated (Mizuno et al., 2007) with disease severity. C3aR may be used as a unique biomaker of diagnosis and progress of disease in lupus nephritis. Brain Eosinophil-derived C3a is a potent inducer. (Kohl, 2001) neurotoxin

Astocytoma cells ↑ Induced an increase in the IL-6 mRNA level which can be completely blocked by pertussis toxin (Sayah et al., 1999) stimulated by C3a or polyclonal anti-C3aR antibody. Experiment allergic Inhibition of complement activation is protective against EAE, an experimental model of MS. (Barnum, 2002; Barnum encephalomyelitis C3-derived biologically active fragments play important role in the pathophysiology of et al., 2006) complement in EAE, rather than C5 and the MAC.

29 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Lyme neuroborreliosis ↑ The intrathecal levels of C3a and C1q are increased in neuroborreliosis. C3a measurement in (Ekdahl et al., 2007) CFS could be used as a clinical marker to identify the patients at risk of developing chronic neuroborreliosis. Alzheimer’s disease ↑ Elevated levels of C3a in CSF (Loeffler et al., 1996)

Hepatic & Liver disease and ↑ Elevated C3a levels. There is significant correlation between high concentration of C3a and (Ronholm et al., 1994) pancreas orthotopic liver development of profound hypotension. transplant Acute pancreatitis ↑ The plasma level of complement C3a and C5b-9 measured daily during the first week after onset (Acioli et al., 1997; of symptoms represent highly specific and sensitive parameters for the prediction of severe acute Gloor et al., 2003) pancreatitis. C3a is a product from pancreatic enzyme, trypsin cleavage and causes the chemotaxis of neutrophils and subsequent lung sequestration. Chronic hepatitis C and ↑ C3a was elevated in patients with chronic hepatitis C and HCV-related HCC but not in HBV- (Lee et al., 2006) Hepatitis C virus-related related HCC. The complement C3a is a potential low-molecular weight serum marker associated hepatocellular carcinoma with chronic hepatitis C and HCV-related HCC. Joint & Bone Rheumatoid arthritis ↑ The plasma levels of C3a in patients with rheumatoid arthritis are twice those of the normal (Moxley et al., 1987) control, and the levels are positively correlated with disease activity.

Others Chronic otitis media ↑↑ Highly elevated C3a levels indicate ongoing complement activation and contribute to chemotactic (Narkio-Makela et al., and inflammatory potential in middle-ear effusion and correlate with the chronicity of the disease. 2000) Colorectal tumors ↑ Complement C3a desArg appears in significantly higher levels in the serum of patients with (Habermann et al., 2006) colorectal adenomas and carcinoma than those without. Active systemic lupus ↑ The elevated levels of the complement split product C3a desArg is significantly characterised in (Belmont et al., 1994) erythematosus patients with active disease. HIV ↑ Elevated C3a levels. Antifungal ↑ C3a exerts antimicrobial effects against the yeast Candida. Arginine residues were found to be (Sonesson et al., 2007) critical for the antifungal and membrane breaking activity of the C3a-derived antimicrobial peptide, CNY21(C3a; Cys57- Arg77).

30 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

1.8 RECEPTORS FOR C3a

1.8.1 C3a RECEPTORS

The receptor (C3aR) for the anaphylatoxin C3a was discovered in 1996 (Ames et al., 1996) through cDNA analysis of human neutrophils. It was also found that 37% of the nucleotides were identical to those of the C5a receptor (CD88) (Ames et al., 1996). C3aR exists on human chromosome band 12q13.2-3 (Paral, 1998) and belongs to the rhodopsin superfamily of G-protein coupled receptors (Gerard et al., 1994). It has seven transmembrane helices and 482 amino acids characterized by the presence of an extraordinarily large second extracellular loop (172 amino acids) between transmembrane helices 4 and 5 (Figure 1.11 and 1.13) (Ames et al., 1996; Crass et al., 1996; Roglic et al., 1996). The major unique feature of C3aR compared to other rhodopsin receptors is this very large extracellular loop (EC2), which has been postulated to contain some or all of the structural determinants for C3a binding (Ember et al., 1997). The cloning of C3a receptor from four species, human (Ames et al., 1996), mouse (Hsu et al., 1997; Tornetta et al., 1997), rat (Fukuoka et al., 1998) and guinea pig (Ember et al., 1998) confirmed the presence of a large second extracellular loop (Figures 1.14).

Figure 1.11 C3aR with extracellular (EC) and intracellular (IC) loops and the helical transmembrane domains (TM) numbered. (Figure adapted from Chao et al., 1999).

Another receptor C5L2 which binds to C5a and C5a desArg is also known to bind to human C3a desArg (Kalant et al., 2003). However, it is uncertain whether C3a and C3a desArg serve as ligands for C5L2 as conflicting data have been produced from different laboratories. There is a study showing that C5L2 negatively regulates inflammation response in a C5L2 knock-out mice model (Gao et al., 2005).

31 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

1.8.2 BINDING SITES OF C3a-C3aR

In a recent study (Chao et al., 1999; Sun et al., 1999) of chimeric C3aR/C5aR and loop deletion in C3aR, it was shown that the large second extracellular loop sequences adjacent to the transmembrane domain contain multiple aspartate residues which play an important role in non- effector ligand binding and also support a two-site C3a-C3aR interaction model, similar to that proposed (Chenoweth et al., 1980) for C5a-C5aR interaction. Furthermore, sulfation of a tyrosine at position 174 in the extracellular loop 2 plays an important role in the binding of C3a (Gao et al., 2003). The anionic residues near the N and C-termini of the C3aR second extracellular loop constitute a high affinity “non-effector site” (Site 1) interaction with cationic residues in the C- terminal helical region of C3a, whereas residues of the C3a C-terminal sequence LGLAR interact with the low affinity “effector site” (Site 2) in C3aR (as shown in Figure 1.12)(Chao et al., 1999).

(A). C3a/C3aR (B). C5a/C5aR

Figure 1.12 Interaction between C3a or C5a and their respective receptors. (A). C3a has a non-effector binding site (Site 1) along the C-terminal helical region that contacts the EC2 and the N-terminal region. Site 2 is the “effector binding site” which binds to the C-terminal LGLAR in the “pore” of C3aR. This model for C3a/C3aR interaction corresponds to a model originally proposed for multi-site binding of C5a with its receptor (Chenoweth et al., 1980) shown in (B) (Ember et al., 1998).

From a mutation study of C3aR, it was found that removal of the N-terminus of the receptor does not affect the affinity of C3a for its receptor (Crass et al., 1999), but substitution of the second extracellular loop (EC2) causes changes in its affinity and activity. However, studies show removal of 65% of the amino acids between 198-308 of human C3aR does not affect C3a affinity and activity (Chao et al., 1999). The removal of residues 174-183 of hC3aR EC2 was found to cause

32 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR loss of function in the receptor. When residues 184-198 and after residue 308 were deleted it was found the affinity of C3a for its receptor was significantly reduced (Chao et al., 1999) which implicates these receptor regions in C3a binding. Mutation of negatively charged residues aspartic acids 325, 326, and 367 by lysine in the upper transmembrane region of hC3aR resulted in a six- fold decrease in C3a affinity and a 14-fold decrease in calcium mobilisation, indicating a role in binding and activation. Aspartic acid mutations D183K/D186K caused a 4-fold decrease in affinity and a 40-fold decrease in calcium mobilisation (Chao et al., 1999).

1.8.3 SIGNALLING PATHWAY ASSOCIATED WITH C3a-C3aR

The binding of C3a to the C3a receptor results in an increase in cytosolic Ca2+ levels through the activation of pertussis toxin (PTX) sensitive G proteins (Elsner et al., 1994a; Klos et al., 1992; Norgauer et al., 1993; Zwirner et al., 1997), and activation of phospholipase C. In neutrophils, Klos et al. (1992) reported that C3a is similar to C5a in stimulating the formation of 1,4,5-triphosphates 2+ (IP3) and stimulates Ca release from intracellular stores of neutrophils. In contrast, the study by Norgauer et al. (1993) showed that C3a does not stimulate phosphatidylinositol biphosphate 3- kinase but stimulates different signalling pathways from C5a and it does not stimulate calcium mobilisation from intracellular stores like C5a. However, C3a activates pertussis toxin sensitive G- protein and triggers the influx of extracellular Ca2+. Nilsson et al.(1996) also reported that C3a and

C5a are coupled to Gi receptors in signal transduction pathways of human mast cell line HMC-1. Both C3a and C5a are potent chemotaxins in this cell line and showed that they activated mast cells through their specific receptor with no cross stimulation. Results from competitive binding and desensitisation studies are consistent with a unique receptor for C3a (and its analogues) that is distinct from the receptor for C5a (Burg et al., 1996; Klos et al., 1992; Zwirner et al., 1997). To date, there is limited information on the signalling from C3a-C3aR stimulation. In contrast, the signalling cascade of C5aR is well defined.

In leukocytes, C3aR and C5aR couple to Gαi (Buhl et al., 1995; Norgauer et al., 1993; Vanek et al., 1994; Zwirner et al., 1997) and Gα16, which are not inhibited by pertussis toxin (Crass et al.,

1996; Ha et al., 2000; Yang et al., 2001). Since there is abundant Gαi expressed in leukocytes and

Gα16 are found only expressed in human cells of hemopoietic lineage (Amatruda et al., 1991), G protein coupling appears to vary for different cell types, for example C3aR on endothelial cell was not coupling to Gαi (Schraufstatter et al., 2002). In astrocytes, the stimulation of C3aR and C5aR by the binding of C3a and C5a to their receptors lead to the activation of mitogen activated protein kinase (MAPK) pathway by phosphorylation of the p44 and p42 kinases, which mediated the signal through pertussis-sensitive

33 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

G-protein. The activation of MAPK pathway could induce the cytokines mRNA expression by C3a and C5a in astrocytes. MAPKs p42 and p44 function plays an important role in the regulation of cell growth and differentiation (Hill et al., 1995; Hunter, 1995; Marshall, 1995). This study also reported that the anaphylatoxins stimulate the activation of phospholipase C (PLC) and generation of IP3 which leads to the release of calcium from intracellular storage, by showing the result of significantly decreasing anaphylatoxin-induced IL-6 mRNA after pre-incubating of astrocytes with PLC-inhibitor (U-73122). Thus, the activated PLC pathway could lead to the activation of astrocytes by the stimulation of C3a and C5a. In contrast, the binding of C3a and C5a to their receptors on astrocyte cells showed inhibition of the adenylyl cyclase pathway by decreasing cAMP levels after activation of the cells with forskolin (Sayah et al., 2003). In the study of human umbilical vein endothelial cells (HUVEC), C3a caused a strong up- regulation of IL-8, IL-1β, and RANTES (Regulated upon Activation, Normal T cell Expressed and Secreted) mRNA in a time and dose-dependent manner. This was inhibited by pre-treatment with anti-receptor antibodies and pertussis toxin (PTX), supporting the involvement of a Gi-protein coupled signalling pathway in C3a receptor activation (Monsinjon et al., 2003). In contrast, a study by Schraufstatter and colleagues (2002) reported that actin reorganization induced by C3a on endothelial cells was not pertussis toxin sensitive but depended on Rho activation, which possibly involves signalling through Gα12 and/or Gα13 proteins. In monocytes or peripheral blood mononuclear cells (PBMC), C3a- and C5a- induced cytokine gene (including IL-1β, IL-6, IL-8 and TNFα) expression is likely the result of G-protein coupled NF-κB activation and in both C3a- and C5a-induced NF-κB activation in monocyte cells can be inhibited by pertussis toxin, indicating that

C5aR and C3aR may couple to Gi for this function (Pan, 1998). Recently Schraufstatter et al. (2009) in mesenchymal stem cells (MSCs), showed that C3a and C5a induce prolonged and powerful ERK1/2 and Akt phosphorylation. Subsequent to the stimulation, phospho-ERK1/2 was translocated to the nucleus causing the phosphorylation of the transcription factor Elk that has not been described for other cells. Additionally, the stimulation of C3aR by C3a in MSCs caused the translocation of C3aR to the nucleus. Nuclear translocation of C3aR has been suggested as the event that could lead to long-term effects such as nuclear ERK phosphorylation (Lu et al., 1998), transcriptional activation (Moughal et al., 2004), cell proliferation and differentiation (Gobeil et al., 2003; Gobeil et al., 2006; Goetzl, 2007). C3aR and C5aR are expressed on the cell surface of MSCs and both anaphylatoxins (C3a and C5a) are chemo-attractant for human bone marrow-derived MSCs. C3aR and C5aR mediate their chemotatic response in a guanine nucleotide-binding protein (Gi) activation-dependent fashion.

34 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Figure 1.13 Homology modelling of C3aR (generated by Dr Praveen Kumar Madala, Fairlie group, University of Queensland).

1.8.4 C3a RECEPTOR DISTRIBUTION

C3aR is expressed on a variety of cells in the periphery and central nervous system (Ames et al., 2001; Ames et al., 1996; Crass et al., 1996). By using antibodies, which bind the second extracellular domain of the C3aR, expression has been demonstrated on myeloid lineage such as neutrophils, eosinophils, basophils, mast cells and monocytes/macrophages (Ames et al., 1996; Crass et al., 1996; Hugli, 1986; Muller-Eberhard, 1988; Volanakis et al., 1998). In the study by Zwirner and colleagues (1997) using monoclonal antibodies it was shown that the numbers of C3aR molecules per cell in eosinophils and basophils are similar to the C5aR population. In monocytes, 6 times more C5aR than C3aR is expressed. On neutrophils the expression of C5aR is 20 times higher than the expression of C3aR, which suggests that eosinophils and basophils are the primary effector cells in peripheral blood for C3a stimulation. Furthermore, the myelomonocyte U937 and myeloblastic HL-60 cell lines express C3aRs when differentiated to a more mature phenotype by treatment with dibutyryl cyclic adenosine monophosphate (Klos et al., 1992). C3aR also exists on epithelial cells (Monsinjon et al., 2001), adrenal cortex, lung, human astrocyte cell line, human fetal astrocytes and microglia (Gasque et al., 1998; Ischenko et al., 1998). In the central nervous system (CNS), C3aR expression was found in neurons, especially in cortical and hippocampal neurons, and in glial cells (Davoust et al., 1999). In studies of C3aR in other species such as mice, an abundance of these receptors was found in heart and lung tissues (Tornetta et al., 1997). 35 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Figure 1.14 Comparison of the EC2 loop sequences of C3a receptors from human (Hu), guinea pig (Gp), rat (Rt), and mouse (Mo)(Modified from Chao et al., 1999).

1.9 CURRENT SMALL C3a PEPTIDE LIGANDS FOR C3aR

1.9.1 AGONISTS

The C-terminal sequence LGLAR (Figure 1.15 A) of C3a, including the guanidine and carboxyl groups of the arginine, is known to be important for biological activity (Caporale et al., 1980). This conclusion is supported by a previous study in which removal of the C-terminal amino acid arginine by carboxypeptidase B digestion leads to abolition of the tissue contractile activity (Bokisch et al., 1970). This was due to the lack of specific binding to receptors. A synthetic 21-residue C terminal fragment of C3a (C3a 57-77) was found to effect biological activities equipotent to that of C3a 1-77 (Caporale et al., 1980; Lu et al., 1984). This 21-residue C- terminal peptide of C3a was shown to exhibit a partial helical conformation in trifluoroethanol

36 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

(a helix favouring solvent) and when substituted with residues that disrupt helix formation, showed less potency than when substituted with helix-promoting residues (Hoeprich et al., 1986). This suggested that a helical conformation may contribute to the higher potency of the C3a 21-residue analogue (Lu et al., 1984). The pentapeptide LGLAR (C3a 73-77) was found to be the minimal sequence which shows C3a-specific activity (Caporale et al., 1980), albeit with a very low biological potency, much lower than for the native C3a protein itself (Caporale et al., 1980). This pentapeptide is highly conserved in human, cows, rat, mouse and guinea pig C3a (Hugli, 1990). When LGLAR was further shortened to LAR, it retained some activity if joined at the N-terminus to a synthetic group such as 9- fluorenylmethoxycarbonyl (Fmoc) or the 6-aminohexanoic acid (Ahx) (Kohl et al., 1990). This finding suggested that the hydrophobic region at the N-terminus helps to insert the effector sequence LAR into the transmembrane pocket (Kohl et al., 1990). The study by GerardySchahn and colleagues (1988) reports that LALAR had 3 times the activity of LGLAR. KÖhl and collegues discovered that the substitution of L-arginine by D-arginine at position 77 in C3a (or by other positively charged amino acids) terminated C3a activity, showing that both stereospecific and charge properties are important for binding (Kohl et al., 1990). In 1991, Ember and colleagues synthesised an analogue of C3a, consisting of a 15-residue peptide with a hydrophobic N-terminus, WWGKKYRASKLGLAR (the bold residues differ from human C3a, Figure 1.15 B and Table 1.7). This peptide, comparable to residues 63-77 of human C3a, was found to be 12-15 times more active than natural C3a and was thus called a ‘superagonist’. They suggested that the hydrophobic N-terminus interacts with a secondary binding site, while the effector unit (LGLAR) interacts with the primary binding site. This superagonist activity has since been discredited in the literature (Ames et al., 1997; Chao et al., 1999; Ischenko et al., 1998; Sun et al., 1999) and by our group (Blakeney, 2007). Jinsmaa and colleagues (2001) designed the C3a peptide agonist, Trp-Pro-Leu-Pro-Arg, which they claim showed anti-analgesic and anti-amnesic effects.

37 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

H2N NH NH

O O H H N N OH H2N N N H H O O O B.

HN NH2 H N NH NH2 2 NH NH NH OH O O O H H H H O H O H O H O N N N N N N N OH H2N N N N N N N N H H O O O H O H O H O H O H O

NH2 OH NH 2

Figure 1.15 A. C-terminal pentapeptide sequence L73GLAR77 that is important for activation of C3aR (Caporale et al., 1980; Sun et al., 1999). B. C3a Superagonist (Ember et al., 1991).

1.9.2 ANTAGONISTS

There is very little known about C3aR antagonism. Recently, Ames et al. (2001) reported N2-

[2,2-diphenylethoxy)acetyl]-L-arginine (SB290157, Figure 1.16 a) as a C3aR antagonist. This 125 arginine analogue was shown to compete with I-C3a radioligand binding in RBL cells (IC50 200 nM), as well as inhibiting C3a-induced calcium mobilisation in RBL cells and in human neutrophils

(IC50 ~ 28 nM). SB290157 also inhibits chemotaxis of HMC-1 cells and ATP-release from guinea pig platelets (Ames et al., 2001). It was also found to be active in two animal models of inflammation: firstly, a guinea pig LPS-induced lung neutrophilia model and secondly, a rat adjuvant-induced arthritis model (Ames et al., 2001). There are a number of studies both in vivo and in vitro in which SB290157 has been used (summarised in Table 1.6) (Baelder et al., 2005; Godau et al., 2004; Mollnes et al., 2002b; Proctor et al., 2004; Ratajczak et al., 2004). These studies report some unexpected results, which may be caused by non related-C3aR antagonism activities of this compound (Baelder et al., 2005; Proctor et al., 2004). Recently, Mathieu et al. (2005) showed that SB290157 has full agonist activity on C3aR in a variety of cell assays, such as calcium mobilisation

38 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR assay in transfected RBL cells, a beta-lactamase assay in CHO-NFAT-bla-Gα16 cells, and an enzyme release assay in differentiated U937 cells. However, the compound lacks agonist activity in guinea pig platelets, which have a low level of C3aR expression. A second non-peptidic C3aR antagonist was reported by Grant et al.(2001), 1,3-diiminoisoindoline (Figure 1.16 b) has low affinity for C3aR (IC50 1.7 µM). In addition, it showed antagonist activity in calcium mobilisation with IC50 2 µM. Denonne et al.(2007b) reported from SAR studies modification of SB290157 at the 125 linker region (Figure 1.16 c) which has a higher affinity for C3aR (pIC50 = 7.2 vs 20 pM [ I]-C3a) due to its rigid linker. However, SB290157 and the SB290157 modified compound, which both have an arginine group, have low availability and short half-life. Another SB290157 modified compound is reported from the same group (Denonne et al., 2007a) but without arginine at the C-terminal (Figure 1.16 d). This binds to C3aR with pIC50=5.8.

H2N NH

NH H N N N O O OH N H O

(b) 1,3-Diiminoisoindoline (a) SB290157

H2N NH Cl O NH O N N

O NH O OH N O H O Cl

(c) optimized derivative of SB290157 by (d) Amino-piperidine linker and a using 2,5 furyl with arginine moiety pyridine moiety (without arginine)

Figure 1.16 Structures of C3aR antagonists (Ames et al., 2001; Denonne et al., 2007a; Denonne et al., 2007b).

39 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Table 1.6 Summary of SB290157 used in animal models. Condition/disease Model Description Reference Lung inflammation a guinea pig LPS-induced (Ames et al., lung neutrophilia model 2001) Arthritis a rat adjuvant-induced (Ames et al., arthritis model 2001) I/R injury A rat model of intestine I/R SB290157 does not (Proctor et al., injury show the anti- 2004) inflammatory property in this model

I/R injury Mice were induced middle Administration of (Ducruet et al., cerebral artery occlusion SB290157 reduced 2008) inflammation and cerebral damage after I/R injury Lupus nephritis A MRL/lpr lupus mouse SB290157 reduces (Bao et al., 2005) model (share many the mortality rate in characteristics of human this model SLE) Allergic asthma A murine model of SB290157 does not (Baelder et al., Aspergillus fumigatus improved AHR 2005) extract induced pulmonary during the effector allergy phase of AHR Acute Respiratory Intravenous infusion of SB290157 inhibited (Proctor et al., distress syndrome CVF rat model the CVF-induced 2006) hypertensive response Hemodynamic effect Rat model to verify the SB290157 (Proctor et al., hemodynamic activities of significantly reduced 2009) C3a, C5a, C3a linear the hypertensive peptide, indomethacin and response of C3a SB290157 agonist stimulation SLE, systemic lupus erythematosus; I/R, ischemia-reperfusion; CVF, cobra venom factor; AHR, airway hyperresponsiveness; MRL/lpr is an abbreviation of MRL/Mp-Tnfrsf lpr/lpr (mouse model of human SLE).

Further studies are needed to identify more potent and selective C3aR antagonists for use in defining the physiological/pharmacological roles of C3a and to probe for therapeutic potential.

40 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Table 1.7 Examples of Synthetic Peptide Agonists and Antagonists of hC3a (Ember et al., 1998). Synthetic C3a analogues Potency Reference (agonists) Intact human C3a 2.0 - 4.3 nM (Kretzschmar et al., 3 nM ED50 1992; Lu et al., 1984)

1.9 - 4.1 nM (Lu et al., 1984) C-N-Y-I-T-E-L-R-R-Q-H-A-R-A-S-H- L-G-L-A-R (native sequence) 92 nM (Ember et al., 1991)

A-N-A-Aib-A-E-E-A-Aib-R-Q-A-Aib- 1.2 nM (Hugli, 1986) R-A-A-Aib-L-G-L-A-R(helix-enhancer sequence) A-N-A-Ab-A-P-A-Ab-R-Q-A-Ab-R-P- 900 nM A-Ab-L-G-L-A-R (helix-disrupted sequence) F-moc-Ahx-Y-R-R-G-R-A-A-A-L-G- 3 nM ED50 (Ambrosius et al., 1989; L-A-R Gerardyschahn et al., 1988) W-W-G-K-K-Y-R-A-S-K-L-G-L-A-R 6 nM (Ember et al., 1991)

C3a antagonists Y-R-R-G-R-Ahx-[C-G-G-L-C]-L-A-R* 0.5 Ki/ED50 (Kretzschmar et al., 1992) Y-R-R-G-R-Ahx-[C-G-A-L-C]-L-A-R* 0.4 Ki/ED50 (Kretzschmar et al., 1992) Aib = 2-aminoisobutyric acid; Ab = 2-aminobutyric acid; F-moc = fluorenyl methoxycarbonyl; Ahx = aminohexyl; * = cyclic peptides.

1.10 C5a ANAPHYLATOXIN

C5a is a pro-inflammatory small polypeptide which consists of 74 amino acids (Figure 1.17). Human C5a has complex anti-parallel 4-helix bundles stabilized by three disulfide bonds (cys21-47, cys22-44, cys34-55), which are connected by three peptide loops (Zhang et al., 1997b; Zuiderweg et al., 1989a; Zuiderweg et al., 1989b) (Figure 1.17). N-terminal of C5a is important for its binding affinity for C5aR while the agonistic C-terminal (M69QLGR74) is acting as receptor activating domain, which supports the two-site binding model of C5a to C5aR (Siciliano et al., 1994). C5a is generated by C5 convertase, which cleaves Component 5 (C5) at the terminal region of the alpha-chain.

41 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

C5a is the most potent inflammatory peptide and has various effects on many cell types (summarised in Table 1.8). C5a can modulate the response in several cell types with broad- spectrum biological activities eg. phagocytosis, degranulation, H2O2 production, granule enzyme release, delay or enhance of apoptosis, stimulate chemokine and cytokine production and chemotaxis. However, C5a is now known as a molecule which has far more effects not only in the immune system, but also it has been associated with developmental biology, CNS development and neurodegeneration, tissue regeneration and haematopoiesis (Haas et al., 2007; Klos et al., 2009; Monk et al., 2007).

MLQKKIEEIAAKYKHSVVKKCCYDGACVNNDETCEQ

1 10 20 30 RAARISLGPRCIKAFTECCVASQLRANISHKDMQLGR 40 50 60 70

Cys21 - 47

Helix 2 Helix 3 Helix 1

Cys 34 - 55 Cys 22-44 COOH NH2 Helix 4

Figure 1.17 Amino acid sequence and NMR solution structure of C5a (pdb: 1KJS). Human C5a is consisting of four-helix bundles, which arrange in antiparallel topology. They are stabilized by three disulfide bonds and connected by three peptide loops. The flexible carboxyl terminal (MQLGR) forms helical turns which are critical for receptor activation (modified from Manthey et al., 2009).

42 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Table 1.8 Functional activity of C5a on myeloid and non-myeloid cells. Cell type Response to C5a

Neutrophils Increase adhesion molecules (Foreman et al., 1994) Chemoattractant (Snyderman et al., 1984) Release of superoxide (Mollnes et al., 2002a) Phagocytosis (Mollnes et al., 2002a) Stimulate the release of granule enzymes (Becker et al., 1985) Delayed apoptosis (Perianayagam et al., 2002) Stimulate the release of cytokines (Guo et al., 2005) Eosinophils Chemotattractant (DiScipio et al., 1999) Stimulate the release of granule enzymes (Egesten et al., 1998; Takafuji et al., 1994) Macrophages Enhance phagocytosis (Guo et al., 2005) Chemoattractant (Marder et al., 1985) Stimulate cytokine release (Okusawa et al., 1988) Monocytes Chemoattractant (Marder et al., 1985) Stimulate cytokine release (Okusawa et al., 1988) Mast cells Chemoattractant (Hartmann et al., 1997) Histamine secretion (Schulman et al., 1988) Basophils Stimulate histamine release (Schulman et al., 1988) T-lymphocytes Suppressed apoptosis (Strainic et al., 2008) Thymocytes Increase apoptosis (Guo et al., 2000) Plasmacytoid Chemoattractant (Gutzmer et al., 2006) dentritic cells Endothelial cells Vasodilation (Albrecht et al., 2004) Stimulate chemokine release (Albrecht et al., 2004) Enhance the adhesion molecule expressed (Monk et al., 2007) Hepatocytes Stimulate liver regeneration (Daveau et al., 2004) Microglia Chemoattractant (Yao et al., 1990) Vascular smooth Indirectly stimulate contraction (Monk et al., muscle cells 2007) Cardiomyocytes Decrease contractility (Monk et al., 2007)

C5a is metabolized by plasma enzyme carboxypeptidases, which remove the C-terminal arginine to form desArg derivative (C5a desArg)(Bokisch et al., 1970; Huey et al., 1983). The desArg form of C5a has a lower potency than C5a by approximately 10-1000 times depending on its function (Hugli et al., 1978). Both C5a and C5a desArg have binding affinity or Kd for C5aR at 1 and 412-600 nM, respectively (Marder et al., 1985; Okinaga et al., 2003).

43 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

C5a has been implicated in many pathogeneses of inflammatory conditions, autoimmune and neurodegenerative diseases when its level increases such as rheumatoid arthritis (Grant et al., 2002; Neumann et al., 2002; Weissmann, 2004), ARDS (Sarma et al., 2006), inflammatory bowel diseases (Woodruff et al., 2003), systemic lupus erythrematosus (Hopkins et al., 1988), I/R injury (Arumugam et al., 2006), glomerulonephritis (Welch, 2002), COPD (Marc et al., 2004), sepsis (Huber-Lang et al., 2002), multiple sclerosis (MullerLadner et al., 1996), asthma and allergy (Abe et al., 2001), psoriasis (Kapp et al., 1985a), atherosclerosis (Speidl et al., 2005), anaphylactic hemorrhagic shock (Harkin et al., 2004; Younger et al., 2001), gingivitis (Okada et al., 1979), peritonitis (Godau et al., 2004), tissue rejection (Gaca et al., 2006), meningitis, pancreatitis (Bhatia et al., 2000), burn (Piccolo et al., 1999), infection, antiphospholipid syndrome (Girardi et al., 2003), neurodegeneration (Van Beek et al., 2003) and macular degeneration (Kijlstra et al., 2005). C5a receptor (CD88) is a typical G-protein coupled receptor with 7TMs, which is classified as a rhodopsin-type receptor. It was the first anaphylatoxin receptor to be cloned in 1991 (Gerard et al., 1991). C5aR is located on chromosome 19 (Gerard et al., 1993). It is 42 kDa and consists of 350 amino acids. C5aR is widely expressed in both myeloid and non-myeloid cells such as vascular endothelial cells, cardiomyocytes, synoviocytes, astrocytes, microglia, neural stem cells, oligodendrocytes, articular chondrocytes, renal glomerular mesangial cells, hepatic kupfer cells, hepatocytes, bronchial epithelial and smooth muscle cells and inflamed keratinocytes (Lee et al., 2008; Monk et al., 2007). In myeloblastic cell lines, C5aR can be upregulated by using dibutyryl- cAMP, phorbol ester and interferon (Burg et al., 1996; Rubin et al., 1991). C5aR signalling depends on G-protein. C5aR coupled to pertussis toxin-sensitive (PTX)

Giα2, Giα3 (Rollins et al., 1991) or pertussis-insenstive Gα16 (Amatruda et al., 1993), which is expressed only in hematopoietic cells. The binding of C5a to C5aR activates signal transduction, which involves intracellular calcium mobilisation from intracellular stores as well as the extracellular medium. There are several activation pathways of C5aR signalling such as phosphatidylinositol-biphosphate-3-kinase/Akt (PKB) (Perianayagam et al., 2002), Ras/B- Raf/mitogen-activated protein kinase (Buhl et al., 1994; Mullmann et al., 1990), phospholipase D (PLD) (Mullmann et al., 1990), protein kinase C (PKC) (Buhl et al., 1994), sphingosine kinase (Melendez et al., 2004) and NF-κB (Kastl et al., 2006). Over the past decade, a few C5aR antagonists have been developed, for example nonpeptidic small molecules, C5a mutants, short peptides and cyclic peptides, mAb and antibody fragments.

44 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

Of the most useful C5aR antagonists are peptidomimetics, which are generated based on SAR studies. The modification of N-MePhe-Lys-Pro-D-cha-Trp-D-Arg or C-089, the first full C5aR antagonist (Konteatis et al., 1994), gives potent and selective C5a antagonists, cyclic PMX53 (3D53) (Ac-Phe-[Orn-Pro-DCha-Trp-Arg]) (Finch et al., 1999), cyclic PMX205 (HCin-[Orn-Pro- DCha-Trp-Arg])(March et al., 2004) and linear JPE1357 (Hoo-Phe-Orn-Pro-hle-Pff-Phe-

NH2)(Schnatbaum et al., 2006) (Figure 1.18). PMX53 is in clinical trial phase I for rheumatoid arthritis and psoriasis. Even though it has been shown to be safe and well tolerated, it fails to show a reduction of inflammation in synovial fluid of patients in clinical trial phase Ib (Vergunst et al., 2007).

NH2 NH2 H2N H2N NH NH

NH NH O O H H O N O N NH NH HN O HN O

NH NH O O N O N O NH NH O AcHN O

PMX53 PMX205 O NH O N O H

HN O O O O H N N NH N N 2 H H NH2 O O

JPE1375

Figure 1.18 Chemical structures of the peptidic C5a antagonists PMX53 (3D53), PMX205 and JPE1375.

45 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

1.11 SUMMARY AND AIMS

G-protein coupled receptor (GPCRs) are the most common drug targets for the pharmaceutical industry. Several hundred GPCRs are peptide-activated, such as complement C3a and C5a receptors. Complement is a non-specific defense mechanism of the body that also forms a bridge to adaptive immunity. One of the major products of complement activation is the generation of anaphylatoxins (C3a and C5a). Anaphylatoxins play a major role in inflammation, which prime and amplify the immune response through the chemotaxis of immune cells to inflammatory sites and induce the secretion of proinflammatory mediators, lysosomal enzymes and reactive oxygen species. C3a is involved in many inflammatory conditions and diseases such as anaphylaxis, adult respiratory distress syndrome (ARDS), acute transplant rejection, ischemia reperfusion injury, sepsis, brain inflammation and many autoimmune diseases eg. psoriasis, rheumatoid arthritis, and systemic lupus erythrematosus. Recently, it has been reported that C3a has a role in organ regeneration (Strey et al., 2003), in neuroprotection (Mukherjee et al., 2008) and in the release of progenitor haematopoietic stem cells (Wysoczynski et al., 2009). C3aR has been expressed in a variety of cells. However, there are few reported C3a agonists and antagonists. Ames et al.(2001) reported the C3aR antagonist, SB290157, which has recently shown agonist activities in some experiments 2+ (EC50 = 2.7± 1.7 in Ca release assay in RBL-hC3aR)(Mathieu et al., 2005). Up to date, there are no potent and selective C3a agonists or antagonists which can be used as a probe in pharmacological and physiological studies to understand the roles of C3aR. The aim of this thesis is to develop the new potent and selective peptidic and non-peptidic C3aR agonists or antagonists and test in competitive binding assay and functional assay. In addition, the structure-activity relationship also provides information for developing new compounds. Furthermore, the antagonism mechanism studied gives a better understanding of how the new ligands bind to C3a receptor.

Chapter 2 There are two parts of this study in this chapter. Firstly, to reinvestigate the known C3a peptides and four decapeptides for C5aR, which are reported in the literature in a competitive binding assay and also in a functional assay. In a functional assay the compounds were investigated by using measurement of the intracellular calcium mobilization on differentiated U937 cells.

46 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

The competitive binding assay was used to determine the affinity and selectivity of the test compounds by using constant radioligand [125I]-C3a at 80 pM or [125I]-C5a at 20 pM to compete with unlabelled test compound in PBMC. Secondly, a new series of hexapeptide compounds based on the C-terminus of hC3a has been developed and studied by using structure-activity relationships (SAR) to inform which amino acids confer potency and selectivity for the C3aR ligands. The selectivity of the compounds was studied by competitive binding assay and receptor desensitization experiments.

Chapter 3 From chapter 2 results, we found that hexapeptide C3aR agonist (FLTLAR) adopted a beta turn conformation at its C-terminus. We decided to introduce an oxazole heterocycle into the LAR region to impart conformational rigidity to the agonist peptides. This rigid linker acts as a beta turn inducing constraint and mimics the conformation found at the C-terminus of C3a in its crystal structure, which was found in the molecular modeling study. The new C3aR agonists were developed to give a library of this series. The competitive binding assay was used to determine the affinity and selectivity of the test compounds by using constant radioligand [125I]-C3a at 80 pM or [125I]-C5a at 20 pM to compete with unlabelled test compound in isolated human monocyte derived macrophage cells. The potency of the compounds was studied by using a calcium mobilization assay in HMDM.

Chapter 4 There were two approaches to developing the new C3aR antagonists in this chapter. Firstly, from the results of chapter 3 which showed the potent C3aR agonists when introducing an oxazole into the LAR region suggest that the rigid linker improves the binding and functional activity of the compounds. In this chapter we hypothesized that if we use a different substituent on the 5-position of the oxazole ring it may improve the binding affinity as well as improve the potency of the compounds. Secondly, using C3aR antagonist, SB290157 (N2-[2,2-diphenylethoxy)acetyl]-L-arginine), as a template, we developed new ligands. The previous results in the competitive binding assay (in Chapter 2) showed that SB290157 is binding tightly and specifically to C3aR without cross reactivity to C5aR. It also shows no agonist activity in the functional assay on human PBMCs. Hence, to explore the structure of SB290157 by dividing them into three parts, which are hydrophobic, linker and terminal arginine. Arginine was kept in place as it is important for biological activity of compounds to activate C3aR.

47 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

The compounds were modified in the first two regions; linker and hydrophobic by substitution with different amino acids and/or functional groups. New agonists and antagonists of C3aR were developed to contribute to the library of this series. Another series of compounds was explored by using the previous template, which was reported by Denonne et al.(2007b) with the substituent of furan at the linker which gives the higher affinity compounds.

Chapter 5 The objectives of this chapter were initially to study the most potent and selective available antagonists: peptide FLTChaAR (25) (from Chapter 2), peptidomimetic 146 (from Chapter 4) and reported nonpeptide SB290157 for mechanism of action against three different classes of C3aR agonists a) protein C3a, b) peptide FLTLAR (17), FWTLAR (20) and c) peptidomimetic 63 and 80). By identifying competitive/noncompetitive and surmountable/insurmountable behaviour in the Ca2+ mobilisation on HMDM, we could potentially learn about the binding location of the compounds.

48 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

1.12 REFERENCES

1. ABE, M, SHIBATA, K, AKATSU, H, SHIMIZU, N, SAKATA, N, KATSURAGI, T, OKADA, H (2001) Contribution of anaphylatoxin C5a to late airway responses after repeated exposure of antigen to allergic rats. J. Immunol. 167(8): 4651-4660.

2. ABOURAGHEB, HHA, WILLIAMS, AJ, BROWN, CB, MILFORDWARD, A (1992) Plasma-levels of the anapylatoxin-C3a and anaphylatoxin-C4a in patients with IGA nephropathy Henoch- Schonlein nephritis. Nephron 62(1): 22-26.

3. ACIOLI, JM, ISOBE, M, KAWASAKI, S (1997) Early complement system activation and neutrophil priming in acute pancreatitis: Participation of trypsin. Surgery 122(5): 909-917.

4. AGOSTONI, A, ET AL (2004) Hereditary and acquired angioedema: proceeding of the third C1 inhibitor deficiency workshop and beyond. J. Allergy Clin. Immunol. 114(supplement): S51- S131.

5. ALBRECHT, EA, CHINNAIYAN, AM, VARAMBALLY, S, KUMAR-SINHA, C, BARRETTE, TR, SARMA, JV, WARD, PA (2004) C5a-induced gene expression in human umbilical vein endothelial cells. Am. J. Pathol. 164(3): 849-859.

6. AMARA, U, RITTIRSCH, D, FLIERL, M, BRUCKNER, U, KLOS, A, GEBHARD, F, LAMBRIS, JD, HUBER-LANG, M (2008) Interaction Between the Coagulation and Complement System. In: Current Topics in Complement Ii, Lambris, JD (ed) Vol. 632, pp 71-79.

7. AMATRUDA, TT, GERARD, NP, GERARD, C, SIMON, MI (1993) Specific interactions of chemoattractant factor receptors with G-proteins. Journal of Biological Chemistry 268(14): 10139-10144.

8. AMATRUDA, TT, STEELE, DA, SLEPAK, VZ, SIMON, MI (1991) G-alpha-16, a G-protein alpha subunit specifically expressed in hematopoietic-cells. Proc. Natl. Acad. Sci. U. S. A. 88(13): 5587-5591.

9. AMBROSIUS, D, CASARETTO, M, GERARDYSCHAHN, R, SAUNDERS, D, BRANDENBURG, D, ZAHN, H (1989) Peptide analogs of the anaphylatoxin C3a-syntheses and properties Biol. Chem. Hoppe-Seyler 370(3): 217-227.

10. AMES, RS, LEE, D, FOLEY, JJ, JUREWICZ, AJ, TORNETTA, MA, BAUTSCH, W, SETTMACHER, B, KLOS, A, ERHARD, KF, COUSINS, RD, SULPIZIO, AC, HIEBLE, JP, MCCAFFERTY, G, WARD, KW, ADAMS, JL, BONDINELL, WE, UNDERWOOD, DC, OSBORN, RR, BADGER, AM, SARAU, HM (2001) Identification of a selective nonpeptide antagonist of the anaphylatoxin C3a receptor that demonstrates antiinflammatory activity in animal models. J. Immunol. 166(10): 6341-6348.

11. AMES, RS, LI, Y, SARAU, HM, NUTHULAGANTI, P, FOLEY, JJ, ELLIS, C, ZENG, Z, SU, K, JUREWICZ, AJ, HERTZBERG, RP, BERGSMA, DJ, KUMAR, C (1996) Molecular cloning and characterization of the human anaphylatoxin C3a receptor. Journal of Biological Chemistry 271(34): 20231-20234.

49 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

12. AMES, RS, TORNETTA, MA, FOLEY, JJ, HUGLI, TE, SARAU, HM (1997) Evidence that the receptor for C4a is distinct from the C3a receptor. Immunopharmacology 38(1-2): 87-92.

13. ARUMUGAM, TV, MAGNUS, T, WOODRUFF, TM, PROCTOR, LM, SHIELS, IA, TAYLOR, SM (2006) Complement mediators in ischemia-reperfusion injury. Clin. Chim. Acta 374(1-2): 33-45.

14. ARUMUGAM, TV, SHIELS, IA, STRACHAN, AJ, ABBENANTE, G, FAIRLIE, DP, TAYLOR, SM (2003) A small molecule C5a receptor antagonist protects kidneys from ischemia/reperfusion injury in rats. Kidney Int. 63(1): 134-142.

15. BAELDER, R, FUCHS, B, BAUTSCH, W, ZWIRNER, J, KOHL, J, HOYMANN, HG, GLAAB, T, ERPENBECK, VJ, KRUG, N, BRAUN, A (2005) Pharmacological targeting of anaphylatoxin receptors during the effector phase of allergic asthma suppresses airway hyperresponsiveness and airway inflammation. J. Immunol. 174(2): 783-789.

16. BALDO, A, SNIDERMAN, AD, STLUCE, S, AVRAMOGLU, RK, MASLOWSKA, M, HOANG, B, MONGE, JC, BELL, A, MULAY, S, CIANFLONE, K (1993) The adipsin acylation stimulating protein system and regulation of intracellular triglyceride synthesis. Journal of Clinical Investigation 92(3): 1543-1547.

17. BAO, LH, OSAWE, I, HAAS, M, QUIGG, RJ (2005) Signaling through up-regulated C3a receptor is key to the development of experimental lupus nephritis. J. Immunol. 175(3): 1947-1955.

18. BARNUM, SR (2002) Complement in central nervous system inflammation. Immunol. Res. 26(1-3): 7-13.

19. BARNUM, SR, SZALAI, AJ (2006) Complement and demyelinating disease: No MAC needed? Brain Res. Rev. 52(1): 58-68.

20. BAUTSCH, W, HOYMANN, HG, ZHANG, QW, MEIER-WIEDENBACH, I, RASCHKE, U, AMES, RS, SOHNS, B, FLEMME, N, ZU VILSENDORF, AM, GROVE, M, KLOS, A, KOHL, J (2000) Cutting edge: Guinea pigs with a natural C3a-receptor defect exhibit decreased bronchoconstriction in allergic airway disease: Evidence for an involvement of the C3a anaphylatoxin in the pathogenesis of asthma. J. Immunol. 165(10): 5401-5405.

21. BECKER, EL, KERMODE, JC, NACCACHE, PH, YASSIN, R, MARSH, ML, MUNOZ, JJ, SHAAFI, RI (1985) The inhibition of neutrophil granule enxyme-secretion and chemotaxis by pertussis toxin. J. Cell Biol. 100(5): 1641-1646.

22. BECKER, S, MEUER, S, HADDING, U, BITTERSUERMANN, D (1978) Scand. J. Immunol. 7: 173 - 178.

23. BELMONT, HM, BUYON, J, GIORNO, R, ABRAMSON, S (1994) Up-regulation of endothelial- cell adhesion molecules characterizes disease-activity in systemic lupus-erythematosus-the Shwartzman phenomenon revisited. Arthritis Rheum. 37(3): 376-383.

24. BHATIA, M, BRADY, M, SHOKUHI, S, CHRISTMAS, S, NEOPTOLEMOS, JP, SLAVIN, J (2000) Inflammatory mediators in acute pancreatitis. J. Pathol. 190(2): 117-125.

50 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

25. BISCHOFF, SC, DEWECK, AL, DAHINDEN, CA (1990) Interleukin-3 and granlocyte macrophage-colony-stimulating factor render human basophils responsive to low concentrations of complement component-C3a. Proc. Natl. Acad. Sci. U. S. A. 87(17): 6813- 6817.

26. BLAKENEY, JS Ligands for C3a and C5a receptors. PhD. Dissertation, The University of Queensland, Brisbane, 2007.

27. BOKISCH, VA, MULLER-EBERHARD, HJ (1970) Anaphylatoxin inactivator of human plasma: its isolation and characterization as a carboxypeptidas. J. Clin. Invest. 49: 2429-2436.

28. BUHL, AM, AVDI, N, WORTHEN, GS, JOHNSON, GL (1994) Mapping of the C5a receptor signal- transduction network in human neutrophils. Proc. Natl. Acad. Sci. U. S. A. 91(19): 9190-9194.

29. BUHL, AM, OSAWA, S, JOHNSON, GL (1995) Mitogen-activated protein-kinase activation requires 2 signal inputs from the human anaphylatoxin C5a receptor. Journal of Biological Chemistry 270(34): 19828-19832.

30. BURG, M, MARTIN, U, BOCK, D, RHEINHEIMER, C, KOHL, J, BAUTSCH, W, KLOS, A (1996) Differential regulation of the C3a and C5a receptors (CD88) by IFN-gamma and PMA in U937 cells and related myeloblastic cell lines. J. Immunol. 157(12): 5574-5581.

31. CABRERA-VERA, TM, VANHAUWE, J, THOMAS, TO, MEDKOVA, M, PREININGER, A, MAZZONI, MR, HAMM, HE (2003) Insights into G protein structure, function, and regulation. Endocr. Rev. 24(6): 765-781.

32. CAMPBELL, WD, LAZOURA, E, OKADA, N, OKADA, H (2002) Inactivation of C3a and C5a octapeptides by carboxypeptidase R and carboxypeptidase N. Microbiol. Immunol. 46(2): 131-134.

33. CAPORALE, LH, TIPPETT, PS, ERICKSON, BW, HUGLI, TE (1980) The active-site of C3a anaphylatoxin Journal of Biological Chemistry 255(22): 758-763.

34. CHALMERS, DT, BEHAN, DP (2002) The use of constitutively active GPCRs in drug discovery and functional genomics. Nat. Rev. Drug Discov. 1(8): 599-608.

35. CHAO, TH, EMBER, JA, EMBER, JA, WANG, MY, BAYON, Y, HUGLI, TE, YE, RD (1999) Role of the second extracellular loop of human C3a receptor in agonist binding and receptor function. Journal of Biological Chemistry 274(14): 9721-9728.

36. CHARRIAUT, C, SENIK, A, KOLB, JP, BAREL, M, FRADE, R (1982) Inhibition of in vitro natural-killer activity by the 3RD component of complement- role for the C3a fragment. Proceedings of the National Academy of Sciences of the United States of America- Biological Sciences 79(19): 6003-6007.

37. CHENOWETH, DE, ERICKSON, BW, HUGLI, TE (1979) Human C5a-related synthetic peptides as neutrophil chemotactic factors. Biochemical and Biophysical Research Communications 86(2): 227-234.

38. CHENOWETH, DE, HUGLI, TE (1980) Human C5a and C5a analogs as probes of the neutrophil C5a receptor. Mol. Immunol. 17(2): 151-161. 51 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

39. CIANFLONE, K, MASLOWSKA, M, SNIDERMAN, AD (1999) Acylation stimulating protein (ASP), an adipocyte autocrine: new directions. Seminars in Cell & Developmental Biology 10(1): 31-41.

40. CIANFLONE, K, XIA, ZN, CHEN, LY (2003) Critical review of acylation-stimulating protein physiology in humans and rodents. Biochim. Biophys. Acta-Biomembr. 1609(2): 127-143.

41. CICARDI, M, ZINGALE, L, ZANICHELLI, A, PAPPALARDO, E, CICARDI, B (2005) C1 inhibitor: molecular and clinical aspects. Springer Seminars in Immunopathology 27(3): 286-298.

42. CLANCY, R (2000) Circulating endothelial cells and vascular injury in systemic lupus erythematosus. Current Rheumatology Rep? 2: 39.

43. COCCHIO, C, MARZELLA, N (2009) Cinryze, a human plasma-derived C1 esterase inhibitor for prophylaxis of hereditary angioedema. P and T 34: 293-328.

44. COCHRANE, CG, MULLEREB.HJ (1968) Derivation of 2 distinct anaphylatoxin activities from third and fifth components of human complement Journal of Experimental Medicine 127(2): 371-&.

45. CRASS, T, AMES, RS, SARAU, HM, TORNETTA, MA, FOLEY, JJ, KOHL, J, KLOS, A, BAUTSCH, W (1999) Chimeric receptors of the human C3a receptor and C5a receptor (CD88). Journal of Biological Chemistry 274(13): 8367-8370.

46. CRASS, T, RAFFETSEDER, U, MARTIN, U, GROVE, M, KLOS, A, KOHL, J, BAUTSCH, W (1996) Expression cloning of the human C3a anaphylatoxin receptor (C3aR) from differentiated U- 937 cells. European Journal of Immunology 26(8): 1944-1950.

47. CUGNO, M, ZANICHELLI, A, FOIENI, F, CACCIA, S, CICARDI, M (2009) C1-inhibitor deficiency and angioedema: molecular mechanisms and clinical progress. Trends Mol. Med 15(2): 69-78.

48. DAFFERN, PJ, PFEIFER, PH, EMBER, JA, HUGLI, TE (1995) C3a is a chemotaxin for human eosinophils but not for neutrophils. I. C3a stimulation of neutrophils is secondary to eosinophil activation. Journal of Experimental Medicine 181(6): 2119-2127.

49. DAMERAU, B, ZIMMERMANN, B, CZORNIAK, K, WUSTEFELD, H, VOGT, W (1986) Role of the N-terminal regions of hog C3a, C5a and C5a- desArg in their biolgical-activities. Mol. Immunol. 23(4): 433-440.

50. DAVEAU, M, BENARD, M, SCOTTE, M, SCHOUFT, MT, HIRON, M, FRANCOIS, A, SALIER, JP, FONTAINE, M (2004) Expression of a functional c5a receptor in regenerating hepatocytes and its involvement in a proliferative signaling pathway in rat. J. Immunol. 173(5): 3418- 3424.

51. DAVIS, AE (2006) Mechanism of angioedema in first complement component inhibitor deficiency. Immunol. Allerg. Clin. North Am. 26(4): 633-+.

52. DAVOUST, N, JONES, J, STAHEL, PF, AMES, RS, BARNUM, SR (1999) Receptor for the C3a anaphylatoxin is expressed by neurons and glial cells. Glia 26(3): 201-211.

52 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

53. DENONNE, F, BINET, S, BURTON, M, COLLART, P, DEFAYS, S, DIPESA, A, ECKERT, M, GIANNARAS, A, KUMAR, S, LEVINE, B, NICOLAS, JM, PASAU, P, PEGURIER, C, PREDA, D, VAN HOUTVIN, N, VOLOSOV, A, ZOU, D (2007a) Discovery of new C3aR ligands. Part 2: Amino-piperidine derivatives. Bioorg. Med. Chem. Lett. 17(12): 3262-3265.

54. DENONNE, F, BINET, S, BURTON, M, COLLART, P, DIPESA, A, GANGULY, T, GIANNARAS, A, KUMAR, S, LEWIS, T, MAOUNIS, F, NICOLAS, JM, MANSLEY, T, PASAU, P, PREDA, D, STEBBINS, K, VOLOSOV, A, ZOU, D (2007b) Discovery of new C3aR ligands. Part 1: Arginine derivatives. Bioorg. Med. Chem. Lett. 17(12): 3258-3261.

55. DISCIPIO, RG, DAFFERN, PJ, JAGELS, MA, BROIDE, DN, SRIRAMARAO, P (1999) A comparison of C3a and C5a-mediated stable adhesion of rolling eosinophils in postcapillary venules and transendothelial migration in vitro and in vivo. J. Immunol. 162(2): 1127-1136.

56. DROUIN, SM (2001) Expression of the Complement Anaphylatoxin C3a and C5a receptors on Bronchial Epithelial and Smooth muscle cells in Models of Sepsis and Asthma. J. Immunol. 166: 2025-2032.

57. DROUIN, SM, CORRY, DB, HOLLMAN, TJ, KILDSGAARD, J, WETSEL, RA (2002) Absence of the complement anaphylatoxin C3a receptor suppresses Th2 effector functions in a murine model of pulmonary allergy. J. Immunol. 169(10): 5926-5933.

58. DROUIN, SM, CORRY, DB, KILDSGAARD, J, WETSEL, RA (2001) Cutting edge: The absence of C3 demonstrates a role for complement in Th2 effector functions in a murine model of pulmonary allergy. J. Immunol. 167(8): 4141-4145.

59. DUCRUET, AF, HASSID, BG, MACK, WJ, SOSUNOV, SA, OTTEN, ML, FUSCO, DJ, HICKMAN, ZL, KIM, GH, KOMOTAR, RJ, MOCCO, J, CONNOLLY, ES (2008) C3a receptor modulation of granulocyte infiltration after murine focal cerebral ischemia is reperfusion dependent. J. Cereb. Blood Flow Metab. 28(5): 1048-1058.

60. EGESTEN, A, MALM, J (1998) Eosinophil leukocyte degranulation in response to serum- opsonized beads: C5a and platelet-activating factor enhance ECP release, with roles for protein kinases A and C. Allergy 53(11): 1066-1073.

61. EHRNTHALLER, C, IGNATIUS, A, GEBHARD, F, HUBER-LANG, M (2011) New Insights of an Old Defense System: Structure, Function, and Clinical Relevance of the Complement System. Mol. Med. 17(3-4): 317-329.

62. EKDAHL, KN, SANDHOLM, K, ERNERUDH, J, CARLSSON, SA, GRANLUND, H, JANSSON, C, NYMAN, D, FORSBERG, P, HENNINGSSON, AJ (2007) Raised intrathecal levels of complement components C1q and C3a in lyme neuroborreliosis. Mol. Immunol. 44(1-3): 165-165.

63. ELLATI, SG, DAHINDEN, CA, CHURCH, MK (1994) Complement peptides C3a-induced and C5a-induced mediator release from dissociated human skin mast-cells. J. Invest. Dermatol. 102(5): 803-806.

64. ELSNER, J, OPPERMANN, M, CZECH, W, DOBOS, G, SCHOPF, E, NORGAUER, J, KAPP, A (1994a) C3a activates reactive oxygen radical species production and intracellular calcium transients in human eosinophils. European Journal of Immunology 24(3): 518-522.

53 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

65. ELSNER, J, OPPERMANN, M, CZECH, W, KAPP, A (1994b) C3a activates the respiratory burst in human polymorphonuclear neutrophilic leukocytes via pertussis-toxin-sensitive G- proteins. Blood 83(11): 3324-3331.

66. EMBER, JA, HUGLI, TE (1997) Complement factors and their receptors. Immunopharmacology 38(1-2): 3-15.

67. EMBER, JA, JAGELS, MA, HUGLI, TE (1998) Characterization of Complement Anaphylatoxins and their Biological Responses. In: The human complement system in health and disease, Volanakis, JE, Frank, MM (eds), pp 241-284. New York: Marcel Dekker.

68. EMBER, JA, JOHANSEN, NL, HUGLI, TE (1991) Designing synthetic superagonists of C3a anaphylatoxin. Biochemistry 30(15): 3603-3612.

69. ESTERBAUER, H, KREMPLER, F, OBERKOFLER, H, PATSCH, W (1999) The complement system: a pathway linking host defence and adipocyte biology. Eur. J. Clin. Invest. 29(8): 653-656.

70. FERGUSON, SSG, DOWNEY, WE, COLAPIETRO, AM, BARAK, LS, MENARD, L, CARON, MG (1996) Role of beta-arrestin in mediating agonist-promoted G protein-coupled receptor internalization. Science 271(5247): 363-366.

71. FERNANDEZ, HN, HENSON, PM, OTANI, A, HUGLI, TE (1978) Chemotactic response to human C3a and C5a anaphylatoxins. 1. Evaluation of C3a and C5a leukotaxis in vitro and under stimulated in vivo conditions. J. Immunol. 120(1): 109-115.

72. FINCH, AM, WONG, AK, PACZKOWSKI, NJ, WADI, SK, CRAIK, DJ, FAIRLIE, DP, TAYLOR, SM (1999) Low-molecular-weight peptidic and cyclic antagonists of the receptor for the complement factor C5a. Journal of Medicinal Chemistry 42(11): 1965-1974.

73. FISCHER, WH, HUGLI, TE (1997) Regulation of B cell functions by C3a and C3a(desArg) - Suppression of TNF-alpha, IL-6, and the polyclonal immune response. J. Immunol. 159(9): 4279-4286.

74. FISCHER, WH, JAGELS, MA, HUGLI, TE (1999) Regulation of IL-6 synthesis in human peripheral blood mononuclear cells by C3a and C3a(desArg). J. Immunol. 162(1): 453-459.

75. FOREMAN, KE, VAPORCIYAN, AA, BONISH, BK, JONES, ML, JOHNSON, KJ, GLOVSKY, MM, EDDY, SM, WARD, PA (1994) C5a-induced expression of P-selectin in endothelial cells. Journal of Clinical Investigation 94(3): 1147-1155.

76. FRANK, MM (1997) Complement & Kinin. In: Medical Immunology, Stites, DP, Terr, AI, Parslow, TG (eds), 9th edn, pp 169-181. Stamford: Appleton & Lange.

77. FUKUOKA, Y, EMBER, JA, HUGLI, TE (1998) Cloning and characterization of rat C3a receptor: Differential expression of rat C3a and C5a receptors by LPS stimulation. Biochemical and Biophysical Research Communications 242(3): 663-668.

78. FUKUOKA, Y, HUGLI, TE (1988) Demonstration of a specific C3a receptor on guinea pig platelets. J. Immunol. 140(10): 3496-3501.

54 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

79. GACA, JG, APPEL, JZ, LUKES, JG, GONZALEZ-STAWINSKI, GV, LESHER, A, PALESTRANT, D, LOGAN, JS, LOVE, SD, HOLZKNECHT, ZE, PLATT, JL, PARKER, W, DAVIS, RD (2006) Effect of an anti-C5a monoclonal antibody indicates a prominent role for anaphylatoxin in pulmonary xenograft dysfunction. Transplantation 81(12): 1686-1694.

80. GANTER, MT, BROHI, K, COHEN, MJ, SHAFFER, LA, WALSH, MC, STAHL, GL, PITTET, JF (2007) Role of the alternative pathway in the early complement activation following major trauma. Shock 28(1): 29-34.

81. GAO, HW, NEFF, TA, GUO, RF, SPEYER, CL, SARMA, JV, TOMLINS, S, MAN, YF, RIEDEMANN, NC, HOESEL, LM, YOUNKIN, E, ZETOUNE, FS, WARD, PA (2005) Evidence for a functional role of the second C5a receptor C5L2. Faseb J. 19(3): 1003-+.

82. GAO, JM, CHOE, H, BOTA, D, WRIGHT, PL, GERARD, C, GERARD, NP (2003) Sulfation of tyrosine 174 in the human C3a receptor is essential for binding of C3a anaphylatoxin. Journal of Biological Chemistry 278(39): 37902-37908.

83. GASQUE, P (2004) Complement: a unique innate immune sensor for danger signals. Mol. Immunol. 41(11): 1089-1098.

84. GASQUE, P, SINGHRAO, SK, NEAL, JW, WANG, P, SAYAH, S, FONTAINE, M, MORGAN, BP (1998) The receptor for complement anaphylatoxin C3a is expressed by myeloid cells and nonmyeloid cells in inflamed human central nervous system: Analysis in multiple sclerosis and bacterial meningitis. J. Immunol. 160(7): 3543-3554.

85. GERARD, C, GERARD, NP (1994) C5a Anaphylatoxin and its 7 transmembrane-segment receptor Annu. Rev. Immunol. 12: 775-808.

86. GERARD, NP, GERARD, C (1991) The chemotactic receptor for human-C5a anaphylatoxin Nature 349(6310): 614-617.

87. GERARD, NP, LU, B, HE, XP, EDDY, RL, SHOWS, TB, GERARD, C (1993) Human chemotaxis receptor genes cluster at 19Q13.3-13.4 - characterization of the human C5a receptor gene. Biochemistry 32(5): 1243-1250.

88. GERARDYSCHAHN, R, AMBROSIUS, D, CASARETTO, M, GROTZINGER, J, SAUNDERS, D, WOLLMER, A, BRANDENBURG, D, BITTERSUERMANN, D (1988) Design and biological- activity of a new generation of synthetic C3a analogs by combination of peptidic and non- peptidic elements. Biochem. J. 255(1): 209-216.

89. GERARDYSCHAHN, R, AMBROSIUS, D, SAUNDERS, D, CASARETTO, M, MITTLER, C, KARWARTH, G, GORGEN, S, BITTERSUERMANN, D (1989) Characterization of C3a receptor- proteins on quinea-pig platelets and human polymorphonuclear leukocytes. European Journal of Immunology 19(6): 1095-1102.

90. GIRARDI, G, BERMAN, J, REDECHA, P, SPRUCE, L, THURMAN, JM, KRAUS, D, HOLLMAN, TJ, CASALI, P, CAROLL, MC, WETSEL, RA, LAMBRIS, JD, HOLERS, VM, SALMON, JE (2003) Complement C5a receptors and neutrophils mediate fetal injury in the antiphospholipid syndrome. Journal of Clinical Investigation 112(11): 1644-1654.

91. GLOOR, B, STAHEL, PF, MULLER, CA, SCHMIDT, OI, BUCHLER, MW, UHL, W (2003) Predictive value of complement activation fragments C3a and sC5b-9 for development of 55 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR severe disease in patients with acute pancreatitis. Scand. J. Gastroenterol. 38(10): 1078- 1082.

92. GOBEIL, F, BERNIER, SG, VAZQUEZ-TELLO, A, BRAULT, S, BEAUCHAMP, MH, QUINIOU, C, MARRACHE, AM, CHECCHIN, D, SENNLAUB, F, HOU, X, NADER, M, BKAILY, G, RIBEIRO-DA- SILVA, A, GOETZL, EJ, CHEMTOB, S (2003) Modulation of pro-inflammatory gene expression by nuclear lysophosphatidic acid receptor type-1. Journal of Biological Chemistry 278(40): 38875-38883.

93. GOBEIL, F, FORTIER, A, ZHU, T, BOSSOLASCO, M, LEDUC, M, GRANDBOIS, M, HEVEKER, N, BKAILY, G, CHEMTOB, S, BARBAZ, D (2006) G-protein-coupled receptors signalling at the cell nucleus: an emerging paradigm. Can. J. Physiol. Pharmacol. 84(3-4): 287-297.

94. GODAU, J, HELLER, T, HAWLISCH, H, TRAPPE, M, HOWELLS, E, BEST, J, ZWIRNER, J, VERBEEK, JS, HOGARTH, PM, GERARD, C, VAN ROOIJEN, N, KLOS, A, GESSNER, JE, KOHL, J (2004) C5a initiates the inflammatory cascade in immune complex peritonitis. J. Immunol. 173(5): 3437-3445.

95. GOETZL, EJ (2007) Diverse pathways for nuclear signaling by G protein-coupled receptors and their ligands. Faseb J. 21(3): 638-642.

96. GRANT, EB, GUIADEEN, D, SINGER, M, ARGENTIERI, D, HLASTA, DJ, WACHTER, M (2001) Design, synthesis, and biological activity of diiminoisoindolines as complement component 3a antagonists. Bioorg. Med. Chem. Lett. 11(21): 2817-2820.

97. GRANT, EP, PICARELLA, D, BURWELL, T, DELANEY, T, CROCI, A, AVITAHL, N, HUMBLES, AA, GUTIERREZ-RAMOS, JC, BRISKIN, M, GERARD, C, COYLE, AJ (2002) Essential role for the C5a receptor in regulating the effector phase of synovial infiltration and joint destruction in experimental arthritis. Journal of Experimental Medicine 196(11): 1461-1471.

98. GUO, RF, HUBER-LANG, M, WANG, X, SARMA, V, PADGAONKAR, VA, CRAIG, RA, RIEDEMANN, NC, MCCLINTOCK, SD, HLAING, T, SHI, MM, WARD, PA (2000) Protective effects of anti-C5a in sepsis-induced thymocyte apoptosis. Journal of Clinical Investigation 106(10): 1271-1280.

99. GUO, RF, WARD, PA (2005) Role of C5A in inflammatory responses. Annu. Rev. Immunol. 23: 821-852.

100. GUTZMER, R, KOTHER, B, ZWIRNER, J, DIJKSTRA, D, PURWAR, R, WITTMANN, M, WERFEL, T (2006) Human plasmacytoid dendritic cells express receptors for anaphylatoxins C3a and C5a and are chemoattracted to C3a and C5a. J. Invest. Dermatol. 126(11): 2422- 2429.

101. HAAS, PJ, VAN STRIJP, J (2007) Anaphylatoxins - Their role in bacterial infection and inflammation. Immunol. Res. 37(3): 161-175.

102. HABERMANN, JK, ROBLICK, UJ, LUKE, BT, PRIETO, DA, FINLAY, WJJ, PODUST, VN, ROMAN, JM, OEVERMANN, E, SCHIEDECK, T, HOMANN, N, DUCHROW, M, CONRADS, TP, VEENSTRA, TD, BURT, SK, BRUCH, HP, AUER, G, RIED, T (2006) Increased serum levels of complement C3a anaphylatoxin indicate the presence of colorectal tumors. Gastroenterology 131(4): 1020-1029.

56 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

103. HAEFFNERCAVAILLON, N, CAVAILLON, JM, LAUDE, M, KAZATCHKINE, MD (1987) C3a(C3a desArg) induces production and release of interleukin-1 by cultured human- monocytes. J. Immunol. 139(3): 794-799.

104. HAEGER, M, BENGTSON, A, KARLSSON, K, HEIDEMAN, M (1989) Complement activation and anaphylatoxin (C3a and C5a) formation in preeclampsia and by amniotic- fluid. Obstet. Gynecol. 73(4): 551-556.

105. HAEGER, M, UNANDER, M, BENGTSSON, A (1991) Complement activation in relation to develpment of preeclampsia. Obstet. Gynecol. 78(1): 46-49.

106. HA, JH, DHANASEKARAN, N, KOH, HC, LEE, CH (2000) Single amino acid of G(alpha 16) (Ala(228)) is responsible for the ability of chemoattractant C5a receptor to induce G(alpha 16)-mediated inositol phosphate release. Biochemical and Biophysical Research Communications 278(2): 426-431.

107. HARKIN, DW, ROMASCHIN, A, TAYLOR, SM, RUBIN, BB, LINDSAY, TF (2004) Complement C5a receptor antagonist attenuates multiple organ injury in a model of ruptured abdominal aortic aneurysm. J. Vasc. Surg. 39(1): 196-206.

108. HARTMANN, K, HENZ, BM, KRU?GER-KRASAGAKES, S, KO?HL, J, BURGER, R, GURTL, S, HAASE, I, LIPPERT, U, ZUBERBIER, T (1997) C3a and C5a stimulate chemotaxis of human mast cells. Blood 89(8): 2863-2870.

109. HILL, CS, TREISMAN, R (1995) Transcriptional regulation by extracellular signals- mechanisms and specificity. Cell 80(2): 199-211.

110. HOEPRICH, PD, HUGLI, TE (1986) Helical conformation at the carboxy-terminal portion of human C3a is required for full activity. Biochemistry 25(8): 1945-1950.

111. HOFFMANN, T, BOTTGER, EC, BAUM, HP, MESSNER, M, HADDING, U, BITTERSUERMANN, D (1988) In vivo effects of C3a on neutrophils and its contribution to inflammatory lung processes in a guinea-pig model. Clin. Exp. Immunol. 71(3): 486-492.

112. HOPKINS, P, BELMONT, HM, BUYON, J, PHILIPS, M, WEISSMANN, G, ABRAMSON, SB (1988) Increased levels of plasma anaphylatoxins in systemic lupus-erythematosus predict flares of the disease and may elicit vascular injury in lupus cerebritis. Arthritis Rheum. 31(5): 632-641.

113. HSU, MH, EMBER, JA, WANG, MY, PROSSNITZ, ER, HUGLI, TE, YE, RD (1997) Cloning and functional characterization of the mouse C3a anaphylatoxin receptor gene. Immunogenetics 47(1): 64-72.

114. HUBER-LANG, MS, YOUNKIN, EM, SARMA, JV, MCGUIRE, SR, LU, KT, GUO, RF, PADGAONKAR, VA, CURNUTTE, JT, ERICKSON, R, WARD, PA (2002) Complement-induced impairment of innate immunity during sepsis. J. Immunol. 169(6): 3223-3231.

115. HUBER, R, SCHOLZE, H, PAQUES, EP, DEISENHOFER, J (1980) Crystal-structure analysis and molecular-model of human C3a anaphylatoxin. Hoppe-Seylers Zeitschrift Fur Physiologische Chemie 361(9): 1389-1399.

57 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

116. HUEY, R, BLOOR, CM, KAWAHARA, MS, HUGLI, TE (1983) Potentiation of the anaphylatoxins in vivo using an inhibitor of serum carboxypeptidase-N(SCPN). 1. Lethality and pathologic effects on pulmonay tissue. Am. J. Pathol. 112(1): 48-60.

117. HUGLI, TE (1986) Biochemistry and biology of anaphylatoxins. In: Complement Vol. 3, pp 111-127.

118. HUGLI, TE (1990) Structure and function of C3a anaphylatoxins. Current topics in Microbiology and Immunology 153: 181-208.

119. HUGLI, TE (1984) Structure and function of the anaphylatoxins. Springer Semin. Immunopathol. 7: 193.

120. HUGLI, TE, MULLER-EBERHARD, HJ (1978) Anaphylatoxins: C3a and C5a. Adv Immunol 26: 1-53.

121. HUGLI, TE, VALLOTA, EH, MULLEREBERHARD, HJ (1975) Purification and partial characterization of human and porcine C3a anaphylatoxin. Journal of Biological Chemistry 250(4): 1472-1478.

122. HUMBLES, AA, LU, B, NILSSON, CA, LILLY, C, ISRAEL, E, FUJIWARA, Y, GERARD, NP, GERARD, C (2000) A role for the C3a anaphylatoxin receptor in the effector phase of asthma. Nature 406(6799): 998-1001.

123. HUNTER, T (1995) Pprotein-kinases and phosphatases- the Yin and Yang of protein- phosphorylation and signaling. Cell 80(2): 225-236.

124. ISCHENKO, A, SAYAH, S, PATTE, C, ANDREEV, S, GASQUE, P, SCHOUFT, MT, VAUDRY, H, FONTAINE, M (1998) Expression of a functional anaphylatoxin C3a receptor by astrocytes. J. Neurochem. 71(6): 2487-2496.

125. JANSSEN, U, BAHLMANN, F, KOHL, J, ZWIRNER, J, HAUBITZ, M, FLOEGE, J (2000) Activation of the acute phase response and complement C3 in patients with IgA nephropathy. Am. J. Kidney Dis. 35(1): 21-28.

126. JINSMAA, Y, TAKENAKA, Y, YOSHIKAWA, M (2001) Designing of an orally active complement C3a agonist peptide with anti-analgesic and anti-amnesic activity. Peptides 22(1): 25-32.

127. KALANT, D, CAIN, SA, MASLOWSKA, M, SNIDERMAN, AD, CIANFLONE, K, MONK, PN (2003) The chemoattractant receptor-like protein C5L2 binds the C3a des- Arg(77)/acylation-stimulating protein. Journal of Biological Chemistry 278(13): 11123- 11129.

128. KALANT, D, MACLAREN, R, CUI, W, SAMANTA, R, MONK, PN, LAPORTE, SA, CIANFLONE, K (2005) C5L2 is a functional receptor for acylation-stimulating protein. Journal of Biological Chemistry 280(25): 23936-23944.

129. KAPP, A, SCHOPF, E (1985a) Involvement of complement in atopic-dermatitis. Acta Derm.-Venereol.: 152-154.

58 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

130. KAPP, A, WOKALEK, H, SCHOPF, E (1985b) Involvement of complement in psoriasis and atopic-dermatitis measurement of C3a and C5a,C-3, C-4 and C1 inactivator. Arch. Dermatol. Res. 277(5): 359-361.

131. KASTL, SP, SPEIDL, WS, KAUN, C, REGA, G, ASSADIAN, A, WEISS, TW, VALENT, P, HAGMUELLER, GW, MAURER, G, HUBER, K, WOJTA, J (2006) The complement component C5a induces the expression of plasminogen activator inhibitor-1 in human macrophages via NF-kappa B activation. J. Thromb. Haemost. 4(8): 1790-1797.

132. KAWAMOTO, S, YALCINDAG, A, LAOUINI, D, BRODEUR, S, BRYCE, P, LU, B, HUMBLES, AA, OETTGEN, H, GERARD, C, GEHA, RS (2004) The anaphylatoxin C3a downregulates the Th2 response to epicutaneously introduced antigen. Journal of Clinical Investigation 114(3): 399-407.

133. KENAKIN, TP (1997) Pharmacologic analysis of drug-receptor interaction. 3rd edn. Lippincott-Raven Publishers: Philadelphia

134. KIJLSTRA, A, LA HEIJ, EC, HENDRIKSE, F (2005) Immunological factors in the pathogenesis and treatment of age-related macular degeneration. Ocul. Immunol. Inflamm. 13(1): 3-11.

135. KILDSGAARD, J, HOLLMANN, TJ, MATTHEWS, KW, BIAN, K, MURAD, F, WETSEL, RA (2000) Cutting edge: Targeted disruption of the C3a receptor gene demonstrates a novel protective anti-inflammatory role for C3a in endotoxin-shock. J. Immunol. 165(10): 5406- 5409.

136. KLOS, A, BANK, S, GIETZ, C, BAUTSCH, W, KOHL, J, BURG, M, KRETZSCHMAR, T (1992) C3a receptor on Dibutyryl-cAMP-differentiated U937 cells and human neutrophils - the human C3a receptor characterized by funcitonal responses and I-125 C3a binding. Biochemistry 31(46): 11274-11282.

137. KLOS, A, TENNER, AJ, JOHSWICH, KO, AGER, RR, REIS, ES, KOHL, J (2009) The role of the anaphylatoxins in health and disease. Mol. Immunol. 46(14): 2753-2766.

138. KOHL, J (2001) Anaphylatoxins and infectious and non-infectious inflammatory diseases. Mol. Immunol. 38(2-3): 175-187.

139. KOHL, J, CASARETTO, M, GIER, M, KARWATH, G, GIETZ, C, BAUTSCH, W, SAUNDERS, D, BITTERSUERMANN, D (1990) Reevaluation of the C3a active-site using short synthetic C3a analogs. European Journal of Immunology 20(7): 1463-1468.

140. KONTEATIS, ZD, SICILIANO, SJ, VANRIPER, G, MOLINEAUX, CJ, PANDYA, S, FISCHER, P, ROSEN, H, MUMFORD, RA, SPRINGER, MS (1994) Development of C5a receptor antagonists- differential loss of functional responses. J. Immunol. 153(9): 4200-4205.

141. KRETZSCHMAR, T, POHL, M, CASARETTO, M, PRZEWOSNY, M, BAUTSCH, W, KLOS, A, SAUNDERS, D, KOHL, J (1992) Synthetic peptides as antagonists of the anaphylatoxin C3a. Eur. J. Biochem. 210(1): 185-191.

142. KRUG, N, TSCHERNIG, T, ERPENBECK, VJ, HOHLFELD, JM, KOHL, J (2001) Complement factors C3a and C5a are increased in bronchoalveolar lavage fluid after

59 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR segmental allergen provocation in subjects with asthma. Am. J. Respir. Crit. Care Med. 164(10): 1841-1843.

143. LACHMANN, PJ, HUGHES-JONES, NC (1984) Initiation of complement activation. In: Semin. Immunopathol. Vol. 7, pp 143-162: Springer.

144. LAUDES, IJ, CHU, JC, SIKRANTH, S, HUBER-LANG, M, GUO, RF, RIEDEMANN, N, SARMA, JV, SCHMAIER, AH, WARD, PA (2002) Anti-C5a ameliorates coagulation/fibrinolytic protein changes in a rat model of sepsis. Am. J. Pathol. 160(5): 1867-1875.

145. LEE, H, WHITFELD, PL, MACKAY, CR (2008) Receptors for complement C5a. The importance of C5aR and the enigmatic role of C5L2. Immunol. Cell Biol. 86(2): 153-160.

146. LEE, IN, CHEN, CH, SHEU, JC, LEE, HS, HUANG, GT, CHEN, DS, YU, CY, WEN, CL, LU, FJ, CHOW, LP (2006) Identification of complement C3a as a candidate biomarker in human chronic hepatitis C and HCV-related hepatocellular carcinoma using a proteomics approach. Proteomics 6(9): 2865-2873.

147. LEFKOWITZ, RJ (1993) G-protein -coupled receptor kinases Cell 74(3): 409-412.

148. LI, JS, JAGGERS, J, ANDERSON, PA (2006) The use of TP10, soluble complement receptor 1, in cardiopulmonary bypass. Expert Review of Cardiovascular Therapy 4(5): 649- 654.

149. LOEFFLER, DA, LEWITT, PA, JUNEAU, PL, PERRY, MF, BRICKMAN, CM (1996) Increased concentrations of complement component C3a in Alzheimer's disease CSF. Neurology 46(2): 5003-5003.

150. LU, D, YANG, H, SHAW, G, RAIZADA, MK (1998) Angiotensin II-induced nuclear targeting of the angiotensin type 1 (AT(1)) receptor in brain neurons. Endocrinology 139(1): 365-375.

151. LU, Z-X, FOK, K, ERICKSON, B, HUGLI, TE (1984) Conformation analysis of COOH- terminal segments of human C3a. Evidence of ordered conformation in an active 21-residue peptide. Journal of Biological Chemistry 259: 7367-7370.

152. MAHAFFEY, KW, VAN DE WERF, F, SHERNAN, SK, GRANGER, CB, VERRIER, ED, FILLOON, TG, TODARO, TG, ADAMS, PX, LEVY, JH, HASSELBLAD, V, ARMSTRONG, PW (2006) Effect of pexelizumab on mortality in patients with acute myocardial infarction or undergoing coronary artery bypass surgery: A systematic overview. Am. Heart J. 152(2): 291-296.

153. MANTHEY, HD, WOODRUFF, TM, TAYLOR, SM, MONK, PN (2009) Complement component 5a (C5a). Int. J. Biochem. Cell Biol. 41(11): 2114-2117.

154. MARCH, DR, PROCTOR, LM, STOERMER, MJ, SBAGLIA, R, ABBENANTE, G, REID, RC, WOODRUFF, TM, WADI, K, PACZKOWSKI, N, TYNDALL, JDA, TAYLOR, SM, FAIRLIE, DP (2004) Potent cyclic antagonists of the complement C5a receptor on human polymorphonuclear leukocytes. Relationships between structures and activity. Molecular Pharmacology 65(4): 868-879.

60 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

155. MARC, MM, KOROSEC, P, KOSNIK, M, KERN, I, FLEZAR, M, SUSKOVIC, S, SORLI, J (2004) Complement factors C3a, C4a, and C5a in chronic obstructive pulmonary disease and asthma. Am. J. Respir. Cell Mol. Biol. 31(2): 216-219.

156. MARDER, SR, CHENOWETH, DE, GOLDSTEIN, IM, PEREZ, HD (1985) Chemotactic responses of human peripheral-blood monocytes to the complement-derived peptides C5a and C5a desArg. J. Immunol. 134(5): 3325-3331.

157. MARINISSEN, MJ, GUTKIND, JS (2001) G-protein-coupled receptors and signaling networks: emerging paradigms. Trends Pharmacol. Sci. 22(7): 368-376.

158. MARKIEWSKI, MM, DEANGELIS, RA, BENENCIA, F, RICKLIN-LICHTSTEINER, SK, KOUTOULAKI, A, GERARD, C, COUKOS, G, LAMBRIS, JD (2008) Modulation of the antitumor immune response by complement. Nat. Immunol. 9(11): 1225-1235.

159. MARKIEWSKI, MM, DEANGELIS, RA, FOUKAS, PG, LAMBRIS, JD (2007a) Complement - A regulator of pro-survival pathways in liver regeneration. Mol. Immunol. 44(1-3): 151.

160. MARKIEWSKI, MM, NILSSON, B, EKDAHL, KN, MOLLNES, TE, LAMBRIS, JD (2007b) Complement and coagulation: strangers or partners in crime? Trends in Immunology 28(4): 184-192.

161. MARSHALL, CJ (1995) Specificity of receptor tyrosine kinase signaling-transient versus sustained extracellular signal-regulated kinase activation. Cell 80(2): 179-185.

162. MATHIEU, MC, SAWYER, N, GREIG, GM, HAMEL, M, KARGMAN, S, DUCHARME, Y, LAU, CK, FRIESEN, RW, O'NEILL, GP, GERVAIS, FG, THERIEN, AG (2005) The C3a receptor antagonist SB 290157 has agonist activity. Immunology Letters 100(2): 139-145.

163. MELENDEZ, AJ, IBRAHIM, FBM (2004) Antisense knockdown of sphingosine kinase 1 in human macrophages inhibits C-5a receptor-dependent signal transduction, Ca2+ signals, enzyme release, cytokine production, and chemotaxis. J. Immunol. 173(3): 1596- 1603.

164. MIZUNO, M, BLANCHIN, S, GASQUE, P, NISHIKAWA, K, MATSUO, S (2007) Complement C3a receptor was highly found in the glomeruli of inflamed human lupus nephritis. Mol. Immunol. 44(16): 3953-3953.

165. MOLLNES, TE, BREKKE, OL, FUNG, M, FURE, H, CHRISTIANSEN, D, BERGSETH, G, VIDEM, V, LAPPEGARD, KT, KOHL, J, LAMBRIS, JD (2002a) Essential role of the C5a receptor in E-coli-induced oxidative burst and phagocytosis revealed by a novel lepirudin- based human whole blood model of inflammation. Blood 100(5): 1869-1877.

166. MOLLNES, TE, SONG, WC, LAMBRIS, JD (2002b) Complement in inflammatory tissue damage and disease. Trends in Immunology 23(2): 61-64.

167. MONK, PN, SCOLA, AM, MADALA, P, FAIRLIE, DP (2007) Function, structure and therapeutic potential of complement C5a receptors. British Journal of Pharmacology 152(4): 429-448.

61 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

168. MONSINJON, T, GASQUE, P, CHAN, P, ISCHENKO, A, BRADY, JJ, FONTAINE, M (2003) Regulation by complement C3a and C5a anaphylatoxins of cytokine production in human umbilical vein endothelial cells. Faseb J. 17(9): 1003-1014.

169. MONSINJON, T, GASQUE, P, ISCHENKO, A, FONTAINE, M (2001) C3A binds to the seven transmembrane anaphylatoxin receptor expressed by epithelial cells and triggers the production of IL-8. FEBS Lett. 487(3): 339-346.

170. MORGAN, EL (1987) The role of prostaglandins in C3a-mediated suppression of human in vitro polyclonal antibody-responses. Clin. Immunol. Immunopathol. 44(1): 1-11.

171. MORGAN, EL, THOMAN, ML, HOBBS, MV, WEIGLE, WO, HUGLI, TE (1985) Human C3a-mediated suppression of the immune-response. 2. suppression of human in vitro polyclonal antibody-responses occurs through the generation of nonspecific- OKT8+suppressor T-cells. Clin. Immunol. Immunopathol. 37(1): 114-123.

172. MOUGHAL, NA, WATERS, C, SAMBI, B, PYNE, S, PYNE, NJ (2004) Nerve growth factor signaling involves interaction between the Trk A receptor and lysophosphatidate receptor 1 systems: nuclear translocation of the lysophosphatidate receptor 1 and Trk A receptors in pheochromocytoma 12 cells. Cell. Signal. 16(1): 127-136.

173. MOUSLI, M, HUGLI, TE, LANDRY, Y, BRONNER, C (1992) A Mechanism of action for anaphylatoxin C3a stimulation of mast cells. J. Immunol. 148(8): 2456-2461.

174. MOXLEY, G, RUDDY, S (1987) Elevated plasma C-3 anaphylatoxin levels in rheumatoid-arthritis patients. Arthritis Rheum. 30(10): 1097-1104.

175. MUKHERJEE, P, THOMAS, S, PASINETTI, GM (2008) Complement anaphylatoxin C5a neuroprotects through regulation of glutamate receptor subunit 2 in vitro and in vivo. J. Neuroinflamm. 5.

176. MULLER-EBERHARD, HJ (1988) Molecular organization and function of the complement system. Annu Rev Biochem 57: 321-347.

177. MULLERLADNER, U, JONES, JL, WETSEL, RA, GAY, S, RAINE, CS, BARNUM, SR (1996) Enhanced expression of chemotactic receptors in multiple sclerosis lesions. J. Neurol. Sci. 144(1-2): 135-141.

178. MULLMANN, TJ, SIEGEL, MI, EGAN, RW, BILLAH, MM (1990) Complement C5a activation of phospholipase-D in human neutrophils- a major route to the production of phophatidates and diglycerides. J. Immunol. 144(5): 1901-1908.

179. MURRAY, I, KOHL, J, CIANFLONE, K (1999) Acylation-stimulating protein (ASP): structure-function determinants of cell surface binding and triacylglycerol synthetic activity. Biochem. J. 342: 41-48.

180. NAGATA, S, GLOVSKY, MM (1987a) Activation of human serum complement with allergens. 1. Generation of C3a,C4a and C5a and induction of human neutrophil aggregation. J. Allergy Clin. Immunol. 80(1): 24-32.

181. NAGATA, S, GLOVSKY, MM, KUNKEL, SL (1987b) Anaphylatoxin-induced neutrophil chemotaxis and aggregation - Limited aggregation and specific desensitization 62 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR induced by human C3a and synthetic C3a octapeptides. International Archives of Allergy and Applied Immunology 82(1): 4-9.

182. NARKIO-MAKELA, M, TEPPO, AM, MERI, S (2000) Complement C3 cleavage and cytokines interleukin-1 beta and tumor necrosis factor-alpha in otitis media with effusion. Laryngoscope 110(10): 1745-1749.

183. NEUMANN, E, BARNUM, SR, TARNER, IH, ECHOLS, J, FLECK, M, JUDEX, M, KULLMANN, F, MOUNTZ, JD, SCHOLMERICH, J, GAY, S, MULLER-LADNER, U (2002) Local production of complement proteins in rheumatoid arthritis synovium. Arthritis Rheum. 46(4): 934-945.

184. NILSSON, G, JOHNELL, M, HAMMER, CH, TIFFANY, HL, NILSSON, K, METCALFE, DD, SIEGBAHN, A, MURPHY, PM (1996) C3a and C5a are chemotaxins for human mast cells and act through distinct receptors via a pertussis toxin-sensitive signal transduction pathway. J. Immunol. 157(4): 1693-1698.

185. NORGAUER, J, DOBOS, G, KOWNATZKI, E, DAHINDEN, C, BURGER, R, KUPPER, R, GIERSCHIK, P (1993) Complement fragment C3a stimulates Ca2+ influx in neutrophils via a pertussis-toxin-sensitive G-protein. Eur. J. Biochem. 217(1): 289-294.

186. OAKLEY, RH, LAPORTE, SA, HOLT, JA, BARAK, LS, CARON, MG (2001) Molecular determinants underlying the formation of stable intracellular G protein-coupled receptor- beta-arrestin complexes after receptor endocytosis. Journal of Biological Chemistry 276(22): 19452-19460.

187. OAKLEY, RH, LAPORTE, SK, HOLT, JA, CARON, MG, BARAK, LS (2000) Differential affinities of visual arrestin, beta arrestin1, and beta arrestin2 for G protein-coupled receptors delineate two major classes of receptors. Journal of Biological Chemistry 275(22): 17201- 17210.

188. OKADA, H, SILVERMAN, MS (1979) Chemotactic activity in periodontal disease. 2. Generation of complement derived chemotactic factors. J. Periodont. Res. 14(2): 147-152.

189. OKINAGA, S, SLATTERY, D, HUMBLES, A, ZSENGELLER, Z, MORTEAU, O, KINRADE, MB, BRODBECK, RM, KRAUSE, JE, CHOE, HR, GERARD, NP, GERARD, C (2003) C5L2, a nonsignaling C5A binding protein. Biochemistry 42(31): 9406-9415.

190. OKSJOKI, R, LAINE, P, HELSKE, S, KREULA, PV, MAYRANPAA, MI, GASQUE, P, KOVANEN, PT, PENTIKAINEN, MO (2007) Receptors for the anaphylatoxins C3a and C5a are expressed in human atherosclerotic coronary plaques. Atherosclerosis 195(1): 90-99.

191. OKUSAWA, S, YANCEY, KB, VANDERMEER, JWM, ENDRES, S, LONNEMANN, G, HEFTER, K, FRANK, MM, BURKE, JF, DINARELLO, CA, GELFAND, JA (1988) C5a stimulates secretion of tumor necrosis factor from human mononuclear cells in vitro- comparison with secretion of interleukin-1beta and interleukin-1alpha. Journal of Experimental Medicine 168(1): 443-448.

192. OSAKA, H, MUKHERJEE, P, AISEN, PS, PASINETTI, GM (1999) Complement-derived anaphylatoxin C5a protects against glutamate-mediated neurotoxicity. J. Cell. Biochem. 73(3): 303-311.

63 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

193. PAN, ZXK (1998) Anaphylatoxins C5a and C3a induce nuclear factor kappa B activation in human peripheral blood monocytes. Biochim. Biophys. Acta-Gene Struct. Expression 1443(1-2): 90-98.

194. PARAL, D (1998) European Journal of Immunology 28: 2417-2423.

195. PARKER, C (2009) Eculizumab for paroxysmal nocturnal haemoglobinuria. Lancet 373(9665): 759-767.

196. PASINETTI, GM, TOCCO, G, SAKHI, S, MUSLEH, WD, DESIMONI, MG, MASCARUCCI, P, SCHREIBER, S, BAUDRY, M, FINCH, CE (1996) Hereditary deficiencies in complement C5 are associated with intensified neurodegenerative responses that implicate new roles for the C-system in neuronal and astrocytic functions. Neurobiol. Dis. 3(3): 197-204.

197. PERIANAYAGAM, MC, BALAKRISHNAN, VS, KING, AJ, PEREIRA, BJG, JABER, BL (2002) C5a delays apoptosis of human neutrophils by a phosphatidylinositol 3-kinase- signaling pathway. Kidney Int. 61(2): 456-463.

198. PICCOLO, MTS, WANG, Y, SANNOMIYA, P, PICCOLO, NS, PICCOLO, MS, HUGLI, TE, WARD, PA, TILL, GO (1999) Chemotactic mediator requirements in lung injury following skin burns in rats. Exp. Mol. Pathol. 66(3): 220-226.

199. POLLEY, MJ, NACHMAN, RL (1983) Human platelet activation by C3a and C3a des- Arg. Journal of Experimental Medicine 158(2): 603-615.

200. PROCTOR, LM, ARUMUGAM, TV, SHIELS, I, REID, RC, FAIRLIE, DP, TAYLOR, SM (2004) Comparative anti-inflammatory activities of antagonists to C3a and C5a receptors in a rat model of intestinal ischaemia/reperfusion injury. British Journal of Pharmacology 142(4): 756-764.

201. PROCTOR, LM, MOORE, TA, MONK, PN, SANDERSON, SD, TAYLOR, SM, WOODRUFF, TM (2009) Complement factors C3a and C5a have distinct hemodynamic effects in the rat. International Immunopharmacology 9(6): 800-806.

202. PROCTOR, LM, STRACHAN, AJ, WOODRUFF, TM, MAHADEVAN, IB, WILLIAMS, HM, SHIELS, IA, TAYLOR, SM (2006) Complement inhibitors selectively attenuate injury following administration of cobra venom factor to rats. International Immunopharmacology 6(8): 1224-1232.

203. PUSCHEL, GP, HESPELING, U, OPPERMANN, M, DIETER, P (1993) Increase in prostanoid formation in rat-liver macrophages (Kupffer cells) by human anaphylatoxin C3a. Hepatology 18(6): 1516-1521.

204. PUSCHEL, GP, NOLTE, A, SCHIEFERDECKER, HL, ROTHERMEL, E, GOTZE, O, JUNGERMANN, K (1996) Inhibition of anaphylatoxin C3a- and C5a- but not nerve stimulation- or noradrenaline-dependent increase in glucose output and reduction of flow in Kupffer cell-depleted perfused rat livers. Hepatology 24(3): 685-690.

205. RAHPEYMAI, Y, HIETALA, MA, WILHELMSSON, U, FOTHERINGHAM, A, DAVIES, I, NILSSON, AK, ZWIRNER, J, WETSEL, RA, GERARD, C, PEKNY, M, PEKNA, M (2006) Complement: a novel factor in basal and ischemia-induced neurogenesis. Embo J. 25(6): 1364-1374. 64 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

206. RANG, HP, DALE, MM (2007) Pharmacology. Vol. 6th. Churchill Livingstone/Elsevier: Philadelphia, Pa.

207. RATAJCZAK, J, RECA, R, KUCIA, M, MAJKA, M, ALLENDORF, DJ, BARAN, JT, JANOWSKA-WIECZOREK, A, WETSEL, RA, ROSS, GD, RATAJCZAK, MZ (2004) Mobilization studies in mice deficient in either C3 or C3a receptor (C3aR) reveal a novel role for complement in retention of hematopoietic stem/progenitor cells in bone marrow. Blood 103(6): 2071-2078.

208. REGAL, JF (1997) Role of the complement system in pulmonary disorders. Immunopharmacology 38(1-2): 17-25.

209. RICKLIN, D, LAMBRIS, JD (2007) Complement-targeted therapeutics. Nat Biotech 25(11): 1265-1275.

210. RIEDEMANN, NC, GUO, RF, LAUDES, IJ, KELLER, K, SARMA, VJ, PADGAONKAR, V, ZETOUNE, FS, WARD, PA (2002) C5a receptor and thymocyte apoptosis in sepsis. Faseb J. 16(6): 887-+.

211. ROBAK, E, ROBAK, T (2009) Monoclonal Antibodies in the Treatment of Systemic Lupus Erythematosus. Curr. Drug Targets 10(1): 26-37.

212. ROGLIC, A, PROSSNITZ, ER, CAVANAGH, SL, PAN, ZX, ZOU, AH, YE, RD (1996) cDNA cloning of a novel G protein-coupled receptor with a large extracellular loop structure. Biochim. Biophys. Acta-Gene Struct. Expression 1305(1-2): 39-43.

213. ROLLINS, TE, SICILIANO, S, KOBAYASHI, S, CIANCIARULO, DN, BONILLAARGUDO, V, COLLIER, K, SPRINGER, MS (1991) Purification of the active C5a receptor from human polymorphonuclear leukocytes as a receptor Gi complex. Proc. Natl. Acad. Sci. U. S. A. 88(3): 971-975.

214. RONHOLM, E, TOMASDOTTIR, H, RUNEBORG, J, BENGTSSON, A, BENGTSON, JP, STENQVIST, O, FRIMAN, S (1994) Complement system activation during orthotopic liver transplantation in man indications of peroperative complement system activation in the gut. Transplantation 57(11): 1594-1597.

215. ROSENBAUM, DM, RASMUSSEN, SGF, KOBILKA, BK (2009) The structure and function of G-protein-coupled receptors. Nature 459(7245): 356-363.

216. ROTHER, RP, ROLLINS, SA, MOJCIK, CF, BRODSKY, RA, BELL, L (2007a) Discovery and development of the complement inhibitor eculizumab for the treatment of paroxysmal nocturnal hemoglobinuria. Nat. Biotechnol. 25(11): 1256-1264.

217. ROTHER, RP, ROLLINS, SA, MOJCIK, CF, BRODSKY, RA, BELL, L (2007b) Discovery and development of the complement inhibitor eculizumab for the treatment of paroxysmal nocturnal hemoglobinuria (vol 25, pg 1256, 2007). Nat. Biotechnol. 25(12): 1488-1488.

218. RUBIN, J, TITUS, L, NANES, MS (1991) Regulation of complement 5a receptor expression in U937 cells by Phorbol ester. J. Leukoc. Biol. 50(5): 502-508.

65 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

219. SAHU, A, LAMBRIS, JD (2000) Complement inhibitors: a resurgent concept in anti- inflammatory therapeutics. Immunopharmacology 49(1-2): 133-148.

220. SARMA, JV, WARD, PA (2011) The complement system. Cell Tissue Res. 343(1): 227-235.

221. SARMA, VJ, HUBER-LANG, M, WARD, PA (2006) Complement in lung disease. Autoimmunity 39(5): 387-394.

222. SAYAH, S, ISCHENKO, AM, ZHAKHOV, A, BONNARD, AS, FONTAINE, M (1999) Expression of cytokines by human astrocytomas following stimulation by C3a and C5a anaphylatoxins: Specific increase in interleukin-6 mRNA expression. J. Neurochem. 72(6): 2426-2436.

223. SAYAH, S, JAUNEAU, AC, PATTE, C, TONON, MC, VAUDRY, H, FONTAINE, M (2003) Two different transduction pathways are activated by C3a and C5a anaphylatoxins on astrocytes. Mol. Brain Res. 112(1-2): 53-60.

224. SCHNATBAUM, K, LOCARDI, E, SCHARN, D, RICHTER, U, HAWLISCH, H, KNOLLE, J, POLAKOWSKI, T (2006) Peptidomimetic C5a receptor antagonists with hydrophobic substitutions at the C-terminus: Increased receptor specificity and in vivo activity. Bioorg. Med. Chem. Lett. 16(19): 5088-5092.

225. SCHRAUFSTATTER, IU, DISCIPIO, RG, ZHAO, M, KHALDOYANIDI, SK (2009) C3a and C5a Are Chemotactic Factors for Human Mesenchymal Stem Cells, Which Cause Prolonged ERK1/2 Phosphorylation. J. Immunol. 182(6): 3827-3836.

226. SCHRAUFSTATTER, IU, TRIEU, K, SIKORA, L, SRIRAMARAO, P, DISCIPIO, R (2002) Complement C3a and C5a induce different signal transduction cascades in endothelial cells. J. Immunol. 169(4): 2102-2110.

227. SCHULMAN, ES, POST, TJ, HENSON, PM, GICLAS, PC (1988) Differntial effects of the complement peptides, C5a and C5a desArg on human basophil and lung mast cell histamine release. Journal of Clinical Investigation 81(3): 918-923.

228. SCHWERTZ, H, CARTER, JM, RUSS, M, SCHUBERT, S, SCHLITT, A, BUERKE, U, SCHMIDT, M, HILLEN, H, WERDAN, K, BUERKE, M (2008) Serine protease inhibitor nafamostat given before reperfusion reduces inflammatory myocardial injury by complement and neutrophil inhibition. J. Cardiovasc. Pharmacol. 52(2): 151-160.

229. SEMB, AG, VAAGE, J, SORLIE, D, LIE, M, MJOS, OD (1990) Coronary trapping of a complement activation product (C3a des-Arg) during myocardial reperfusion in open-heart- surgery. Scand. J. Thorac. Cardiovasc. Surg. 24(3): 223-227.

230. SHATSTSEYTLINA, EA, NAIR, CH, DHALL, DP (1994) Complement activation- a new participant in the modulation of fibrin gel characteristics and the progression of atherosclerosis. Blood Coagul. Fibrinolysis 5(4): 529-535.

231. SHOWELL, HJ, GLOVSKY, MM, WARD, PA (1982) C3a-induced lysosomal enzyme- secretion from human-neutrophils lack of inhibition by F-Met-Leu-Phe antagonists and inhibition by arachidonic-acid antagonists. International Archives of Allergy and Applied Immunology 67(3): 227-232. 66 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

232. SICILIANO, SJ, ROLLINS, TE, DEMARTINO, J, KONTEATIS, Z, MALKOWITZ, L, VANRIPER, G, BONDY, S, ROSEN, H, SPRINGER, MS (1994) 2-Site binding of C5a by its receptor- an alternative binding paradigm for G-protein coupled receptors. Proc. Natl. Acad. Sci. U. S. A. 91(4): 1214-1218.

233. SNYDERMAN, R, PIKE, MC (1984) Chemoattractant receptors on phagocytic-cells. Annu. Rev. Immunol. 2: 257-281.

234. SONESSON, A, RINGSTAD, L, NORDAHL, EA, MALMSTEN, M, MORGELIN, M, SCHMIDTCHEN, A (2007) Antifungal activity of C3a and C3a-derived peptides against Candida. Biochim. Biophys. Acta-Biomembr. 1768(2): 346-353.

235. SPEIDL, WS, EXNER, M, AMIGHI, J, KASTL, SP, ZORN, G, MAURER, G, WAGNER, O, HUBER, K, MINAR, E, WOJTA, J, SCHILLINGER, M (2005) Complement component C5a predicts future cardiovascular events in patients with advanced atherosclerosis. Eur. Heart J. 26(21): 2294-2299.

236. STIMLER, NP, HUGLI, TE, BLOOR, CM (1980) Pulmonary injury induced by anaphylatoxins-C3a and C5a-anaphylatoxins. Am. J. Pathol. 100(2): 327-&.

237. STRAINIC, MG, LIU, JB, HUANG, DP, AN, F, LALLI, PN, MUQIM, N, SHAPIRO, VS, DUBYAK, GR, HEEGER, PS, MEDOF, ME (2008) Locally produced complement fragments C5a and C3a provide both costimulatory and survival signals to naive CD4(+) T cells. Immunity 28(3): 425-435.

238. STREY, CW, MARKIEWSKI, M, MASTELLOS, D, TUDORAN, R, SPRUCE, LA, GREENBAUM, LE, LAMBRIS, JD (2003) The proinflammatory mediators C3a and C5a are essential for liver regeneration. Journal of Experimental Medicine 198(6): 913-923.

239. SUN, JZ, EMBER, JA, CHAO, TH, FUKUOKA, Y, YE, RD, HUGLI, TE (1999) Identification of ligand effector binding sites in transmembrane regions of the human G protein-coupled C3a receptor. Protein Science 8(11): 2304-2311.

240. SUNYER, JO, BOSHRA, H, LI, J (2005) Evolution of anaphylatoxins, their diversity and novel roles in innate immunity: Insights from the study of fish complement. Vet. Immunol. Immunopathol. 108(1-2): 77-89.

241. TAKABAYASHI, T, VANNIER, E, BURKE, JF, TOMPKINS, RG, GELFAND, JA, CLARK, BD (1998) Both C3a and C3a(desArg) regulate interleukin-6 synthesis in human peripheral blood mononuclear cells. J. Infect. Dis. 177(6): 1622-1628.

242. TAKABAYASHI, T, VANNIER, E, CLARK, BD, MARGOLIS, NH, DINARELLO, CA, BURKE, JF, GELFAND, JA (1996) A new biologic role for C3a and C3a desArg - Regulation of TNF-alpha and IL-1 beta synthesis. J. Immunol. 156(9): 3455-3460.

243. TAKAFUJI, S, TADOKORO, K, ITO, K, DAHINDEN, CA (1994) Degranulation from human eosinophils stimulated with C3a and C5a. Int. Arch. Allergy Immunol. 104: 27-29.

244. TERUI, T, TAKEMATSU, H, KATO, T, OHKOHCHI, K, TAGAMI, H (1987) Plasma anaphylatoxin concentrations in inflammatory skin diseases. Tohoku J. Exp. Med. 151(2): 245-252. 67 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

245. TORNETTA, MA, FOLEY, JJ, SARAU, HM, AMES, RS (1997) The mouse anaphylatoxin C3a receptor: Molecular cloning, genomic organization, and functional expression. J. Immunol. 158(11): 5277-5282.

246. VALLOTA, EH, MULLEREB.HJ (1973) Formation of C3a and C5a anaphylatoxins in whole human-serum after inhibition of anaphylatoxin inactivator. Journal of Experimental Medicine 137(5): 1109-1123.

247. VAN BEEK, J, ELWARD, K, GASQUE, P (2003) Activation of complement in the central nervous system - Roles in neurodegeneration and neuroprotection. In: Neuroendocrine and Neural Regulation of Autoimmune and Inflammatory Disease: Molecular, Systems, and Clinical Insights, Sternberg, EM, Haour, FG, Smith, CC (eds) Vol. 992, pp 56-71.

248. VAN BEEK, J, NICOLE, O, ALI, C, ISCHENKO, A, MACKENZIE, ET, BUISSON, A, FONTAINE, M (2001) Complement anaphylatoxin C3a is selectively protective against NMDA-induced neuronal cell death. Neuroreport 12(2): 289-293.

249. VANEK, M, HAWKINS, LD, GUSOVSKY, F (1994) Coupling of the C5a receptor to G(i) in U-937 cells and in cells transfected with C5a receptor cDNA. Molecular Pharmacology 46(5): 832-839.

250. VERGUNST, CE, GERLAG, DM, DINANT, H, SCHULZ, L, VINKENOOG, M, SMEETS, TJM, SANDERS, ME, REEDQUIST, KA, TAK, PP (2007) Blocking the receptor for C5a in patients with rheumatoid arthritis does not reduce synovial inflammation. Rheumatology 46(12): 1773-1778.

251. VOLANAKIS, JE, FRANK, MM (1998) Human complement system in health and disease. Marcel Dekker: New York.

252. WAGNER, E, FRANK, MM (2010) Therapeutic potential of complement modulation. Nat. Rev. Drug Discov. 9(1): 43-56.

253. WEISSMANN, G (2004) Pathogenesis of rheumatoid arthritis. JCR-J. Clin. Rheumatol. 10(3): S26-S31.

254. WELCH, TR (2002) Complement in glomerulonephritis. Nature Genet. 31(4): 333- 334.

255. WERFEL, T, KIRCHHOFF, K, WITTMANN, M, BEGEMANN, G, KAPP, A, HEIDENREICH, F, GOTZE, O, ZWIRNER, J (2000) Activated human T lymphocytes express a functional C3a receptor. J. Immunol. 165(11): 6599-6605.

256. WOODRUFF, TM, ARUMUGAM, TV, SHIELS, IA, REID, RC, FAIRLIE, DP, TAYLOR, SM (2003) A potent human C5a receptor antagonist protects against disease pathology in a rat model of inflammatory bowel disease. J. Immunol. 171(10): 5514-5520.

257. WOODRUFF, TM, CRANE, JW, PROCTOR, LM, BULLER, KM, SHEK, AB, DE VOS, K, POLLITT, S, WILLIAMS, HM, SHIELS, IA, MONK, PN, TAYLOR, SM (2006) Therapeutic activity of C5a receptor antagonists in a rat model of neurodegeneration. Faseb J. 20(9): 1407-1417. 68 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

258. WOODRUFF, TM, POLLITT, S, PROCTOR, LM, STOCKS, SZ, MANTHEY, HD, WILLIAMS, HM, MAHADEVAN, IB, SHIELS, IA, TAYLOR, SM (2005) Increased potency of a novel complement factor 5a receptor antagonist in a rat model of inflammatory bowel disease. Journal of Pharmacology and Experimental Therapeutics 314(2): 811-817.

259. WUEPPER, K, BOKISHCH, V, MULLER-EBERHARD, HJ (1972) Clin. Exp. Immunol. 1: 13-20.

260. WYSOCZYNSKI, M, RECA, R, LEE, H, WU, W, RATAJCZAK, J, RATAJCZAK, MZ (2009) Defective engraftment of C3aR(-/-) hematopoietic stem progenitor cells shows a novel role of the C3a-C3aR axis in bone marrow homing. Leukemia 23(8): 1455-1461.

261. XIA, ZN, STANHOPE, KL, DIGITALE, E, SIMION, OM, CHEN, LY, HAVEL, P, CIANFLONE, K (2004) Acylation-stimulating protein (ASP)/complement C3adesArg deficiency results in increased energy expenditure in mice. Journal of Biological Chemistry 279(6): 4051-4057.

262. YANG, D (2006) Complement-derived Inflammatory Peptides: Anaphylatoxins. In: Handbook of Biologically Active Peptides, Kastin, AJ (ed), pp 553-558. Amsterdam: Elsevier.

263. YANG, M, SANG, HR, RAHMAN, A, WU, DQ, MALIK, AB, YE, RD (2001) G alpha(16) couples chemoattractant receptors to NF-kappa B activation. J. Immunol. 166(11): 6885-6892.

264. YAO, J, HARVATH, L, GILBERT, DL, COLTON, CA (1990) Chemotaxis by a CNS macrophage, the microglia. J. Neurosci. Res. 27(1): 36-42.

265. YOUNGER, JG, SASAKI, N, WAITE, MD, MURRAY, HN, SALEH, EF, RAVAGE, ZA, HIRSCHL, RB, WARD, PA, TILL, GO (2001) Detrimental effects of complement activation in hemorrhagic shock. J. Appl. Physiol. 90(2): 441-446.

266. ZANKER, B, RASOKAT, H, HADDING, U, BITTER-SUERMANN, D (1982) Agents Actions 11: 147-157.

267. ZHANG, J, BARAK, LS, WINKLER, KE, CARON, MG, FERGUSON, SSG (1997a) A central role for beta-arrestins and clathrin-coated vesicle-mediated endocytosis in beta(2)- adrenergic receptor resensitization - Differential regulation of receptor resensitization in two distinct cell types. Journal of Biological Chemistry 272(43): 27005-27014.

268. ZHANG, XL, BOYAR, W, TOTH, MJ, WENNOGLE, L, GONNELLA, NC (1997b) Structural definition of the C5a C terminus by two-dimensional nuclear magnetic resonance spectroscopy. Proteins 28(2): 261-267.

269. ZILOW, G, STURM, JA, ROTHER, U, KIRSCHFINK, M (1990) Complement activation and the prognostic value of C3a in patients at risk of adult respiratory - distress syndrome. Clin. Exp. Immunol. 79(2): 151-157.

270. ZUIDERWEG, ERP, FESIK, SW (1989a) Heteronuclear 3-dimensional NMR- spectroscopy of the inflammatory protein C5a. Biochemistry 28(6): 2387-2391.

69 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

271. ZUIDERWEG, ERP, NETTESHEIM, DG, MOLLISON, KW, CARTER, GW (1989b) Tertiary structure of human-complement component C5a in solution from nuclear magnetic- resonance data. Biochemistry 28(1): 172-185.

272. ZWIRNER, J, GOTZE, O, MOSER, A, SIEBER, A, BEGEMANN, G, KAPP, A, ELSNER, J, WERFEL, T (1997) Blood- and skin-derived monocytes/macrophages respond to C3a but not to C3a(desArg) with a transient release of calcium via a pertussis toxin-sensitive signal transduction pathway. European Journal of Immunology 27(9): 2317-2322.

70 CHAPTER 1 Peptide-Activated G-protein Coupled Receptors:- Complement C3aR

71 CHAPTER 2 Short Peptide Agonists for C3aR

CHAPTER 2 SHORT PEPTIDE AGONISTS FOR C3aR 2.0 ABSTRACT...... 73 2.1 INTRODUCTION...... 74 2.2 AIMS/OBJECTIVES...... 76 2.3 RESULTS ...... 78

2.3.1 SELECTIVITY AND POTENCY OF HUMAN C3a ...... 78 2.3.1.1 Affinity of hC3a and hC5a...... 78 2.3.1.2 Potency of hC3a and hC5a...... 80

2.3.2 SELECTIVITY AND POTENCY OF 15-RESIDUES OF HC3a AND HC5a ...... 82 2.3.2.1 Potencies of 15-residue components of hC3a and hC5a...... 82 2.3.2.2 Binding affinities of 15-residue components of hC3a and hC5a...... 83

2.3.3 SELECTIVITY AND POTENCY OF KNOWN TRUNCATED C3a PEPTIDES...... 85 2.3.3.1 Decapeptides ...... 85 2.3.3.1.1 Affinity of Known Truncated Anaphylatoxin Agonists: Decapeptides...... 85 2.3.3.2 Known Truncated C3a: C3a Superagonist and C3a hexapeptides...... 86 2.3.3.2.1 Affinity of C3a Superagonist and C3a hexapeptides ...... 87 2.3.3.2.2 Potency of Superagonist and known hexapeptides ...... 88 2.3.4 C3a DES-ARG COMPOUNDS...... 91 2.3.4.1 Binding affinities of des-Arg compounds to C3aR...... 91

2.3.5 C3a HEXAPEPTIDE ANALOGUES ...... 92 2.3.5.1 Binding affinity of hexapeptide analogues ...... 92 2.3.5.1.1 Affinity for C3aR of hexapeptide analogues ...... 92 2.3.5.1.2 Affinity for C5aR of hexapeptide analogues ...... 93 2.3.5.2 Potency of hexapeptide agonists...... 95 2.3.5.2.1 Structure-activity relationships (SAR) for hexapeptide agonist analogues of

F1WPLAR6: Single Point Agonist Potency Comparisons ...... 97 2.3.5.2.2 Dose dependence of hexapeptide analogues...... 102 2.3.5.2.3 Desensitisation of hexapeptide analogues...... 103 2.4 DISCUSSION ...... 105 2.5 MATERIALS AND METHODS ...... 108 2.6 REFERENCES...... 111

72 CHAPTER 2 Short Peptide Agonists for C3aR 2.0 ABSTRACT

Complement activation is one of the body’s most important responses against antigens and other foreign materials. The complement system includes more than 30 different proteins, which act together to form the membrane attack complex, which can cause lysis of bacteria, virus-infected or damaged cells. The complement component C3 plays a pivotal role in all three pathways (classical, lectin and alternative) of complement activation. Cleavage of C3 by C3-convertase (and possibly other proteases) generates the anaphylatoxin C3a, a 77 amino acid protein. C3a binds to its functional receptor, C3aR, which is a G protein – coupled receptor predominantly coupled to Gi. The agonist activity of C3a is due to the octapeptide of the C-terminal region, which triggers receptor activation. The truncated and modified 15 residues, with the incorporation of two tryptophans at the N-terminus, was reported to be 15-fold more potent than C3a in a guinea pig platelet aggregation assay. The superagonist activity has been disputed in the literature (Ames et al., 1997; Chao et al., 1999; Ischenko et al., 1998; Sun et al., 1999) and by our group in a previous study (Blakeney, 2007) since the compound showed less potency and no selectivity for C3aR. Therefore to date, there is no potent agonist of C3a composed of less than eight amino acids. In this chapter we study (i) the selectivity of compounds by first using a competitive radioligand binding assay with [125I]-labelled C3a and C5a on PBMCs, which were isolated from human plasma to determine the affinity of the compounds; and then the general selectivity by a C3a-induced receptor desensitisation experiment; and (ii) the potency of the compounds by using a calcium mobilisation assay which detects intracellular calcium trafficking by binding calcium to a cell permeable fluorophore. We found that previously reported compounds, such as the C3a “superagonist” (WWGKKYRASKLGLAR), were not selective for C3aR in a receptor desensitisation experiment. Four decapeptides, originally reported as C5aR agonists, also bound tightly to C3aR in PBMC cells. Two reported oryzatensin pentapeptide derivatives, WPLPR and YPLPR, also showed no selectivity for C3aR over C5aR in the competitive radioligand binding assay. A new series of hexapeptides based on the C-terminus of human C3a was designed and studied using structure- activity relationships (SAR). Two small peptide agonists were found to be highly potent agonists and selective for C3aR. FLTLAR and FWTLAR bind to C3aR tightly with IC50 ~ 40 and 80 nM, respectively on PBMC cells and in the calcium mobilisation assay had EC50 = 0.32 and 0.37 µM, respectively. Structure activity relationships and competitive binding results have led to a new

73 CHAPTER 2 Short Peptide Agonists for C3aR

2.1 INTRODUCTION

The complement system is an innate immune system consisting of more than 30 distinct plasma proteins that promote phagocytosis, lysis and removal of pathogens through opsonization (C3b) and immune cell stimulation (Gerard et al., 2002; Haas et al., 2007; Hugli, 1986; Hugli, 1981; Hugli, 1984; Hugli et al., 1978; Masters et al., 2009; Zipfel et al., 2009). The activation of the complement cascade generates products, which are the complement anaphylatoxins, C3a and C5a. Anaphylatoxins prime and amplify the immune response through chemoattraction of immune cells to inflammatory sites and also promote proinflammatory secretions, lysosomal enzymes, and reactive oxygen species (Gerard et al., 2002; Haas et al., 2007; Hugli, 1986; Hugli, 1981; Hugli, 1984; Hugli et al., 1978; Masters et al., 2009; Zipfel et al., 2009). C5a is the most potent proinflammatory peptide (Allegretti et al., 2005; Monk et al., 2007), which is generated via the complement cascade. C5a plays roles both within and beyond the immune system. C5a has been implicated in developmental biology, CNS development and neurodegeneration, tissue regeneration and haematopoiesis, in addition to its proinflammatory properties (Haas et al., 2007; Klos et al., 2009; Monk et al., 2007). C3a is 77 amino acid residues, which has proinflammatory properties but to a lesser extent than C5a. C3a induces smooth muscle cell contraction (Ember et al., 1998), chemotaxis of mast cells (Nilsson et al., 1996) and eosinophils (Daffern et al., 1995), increases vascular permeability, stimulates release of histamine in basophils (Kretzschmar et al., 1993) and mast cells (Ellati et al., 1994) and promotes the release of serotonin from guinea pig platelets (Fukuoka et al., 1988). C3a has been implicated in the pathogenesis and progression of inflammatory diseases like asthma, allergies, sepsis, lupus erythematosus, diabetes, arthritis, psoriasis, nephropathy and ischemia-reperfusion injury (Boos et al., 2004; Boos et al., 2005; Drouin et al., 2002; Drouin et al., 2001; Garrett et al., 2009; Humbles et al., 2000; Hutamekalin et al., 2010; Kawamoto et al., 2004; Kildsgaard et al., 2000; Kirschfink, 2001; Mamane et al., 2009; Mizutani et al., 2009; Mocco et al., 2006; Mueller-Ortiz et al., 2006; Rahpeymai et al., 2006; Rynkowski et al., 2009; Tang et al., 2009; Wenderfer et al., 2009). Recently there have been reported antimicrobial and antifungal properties for C3a derived peptides (Malmsten et al., 2007). The 21-residues of C3a and the 20-residue C-terminus of C3adesArg also show antibacterial properties against E. faecalis and P. aeruginosa (Nordahl et al., 2004), Escherichia coli and Staphylococcus aureus (Pasupuleti et al., 2008) and antifungal properties against Candida albicans (Sonesson et al., 2007). As more C3a and C3aR pharmacology has become established, interest has been renewed in C3a and there is a realisation of the need for

74 CHAPTER 2 Short Peptide Agonists for C3aR synthetic C3a agonists and antagonists for interrogating and potential treatment of C3a-mediated conditions in humans. Like C5a, it is the C-terminal octapeptide region of C3a that is implicated as the effector domain required for triggering receptor activation (Hoeprich et al., 1986). The biological activity of C3a is abolished by the cleavage off of C-terminal arginine by carboxypeptidase (Bokisch et al., 1970; Campbell et al., 2002; Ember et al., 1997) in the circulation and tissues. Desarginated forms of C5a and C3a are C5a desArg and C3a desArg (acylation-stimulating protein (ASP)) respectively. They are also less immunologically active. C5a desArg still retains some proinflammatory properties (1-10% of C5a) while C3a desArg has no activity and no binding to C3aR (Bokisch et al., 1970). To date, all synthetic C3a ligands contain the C-terminal arginine. Numerous peptides based on the C-terminus of C3a have been reported to compete with [125I]-C3a on cell surfaces (Jinsmaa et al., 2001; Takabayashi et al., 1996). There have been relatively few reports of new synthetic C3a ligands since the early 1990s. The development of short peptide agonists has focused on identifying the minimum of C-terminal peptide residues of C3a that is capable of activating C3aR and then modifying or adding to the pentapeptide (L73GLAR77)(Ambrosius et al., 1989; Caporale et al., 1980; Ember et al., 1991; Federwisch et al., 1992; Gerardyschahn et al., 1988; Hoeprich et al., 1986; Huey et al., 1984; Kohl et al., 1990; Kretzschmar et al., 1992; Mousli et al., 1992; Petering et al., 2000). Huey and colleagues (1984) reported the first peptide with activity comparable to C3a, a synthetic peptide comprising the C-terminal 21 residues of C3a which was equipotent with C3a in a lung strip contraction assay. This peptide was only 44% as effective as C3a in immunoglobulin (Ig) secretion by peripheral blood lymphocytes. By addition of hydrophobic moieties such as fluorenyl methoxy carbonyl (Fmoc) or aminohexanoic (Ahx) to the N-terminus of native C3a terminal 5-13 residue peptides increases the activity (Ambrosius et al., 1989; Gerardyschahn et al., 1988; Kohl et al., 1990; Kola et al., 1992). In the study by Hoeprich et al.(1986) which modified C3a by incorporating 21 residues of C3a with helix-inducing Ala and Aib residues at judicious positions resulted in a 250% improvement in activity of the peptide which correlated with increased helicity measured by circular dichroism (Hoeprich et al., 1986). The 15 residue peptides which were developed from searching minimal potent sequences of the C3a C-terminus in different species and modified with incorporation of a tryptophan dipeptide as the hydrophobic moiety at the N-terminus resulted in a superagonist ‘WWGKKYRASKLGLAR’ which is fifteen-fold more potent than C3a in a guinea pig platelet aggregation assay (Ember et al., 1991). However, it was equipotent with C3a in a calcium mobilisation assay using guinea pig or mouse C3aR transfected cells (Lienenklaus et al., 1998; Sun et al., 1999). Increasing the helicity of short synthetic ligands by addition or

75 CHAPTER 2 Short Peptide Agonists for C3aR substitution of residues with high helix propensity might increase the potency of designed ligand peptides (Hoeprich et al., 1986). Due to the reported activities of these synthetic C3a agonists using a variety of assay systems to evaluate the activities of these ligands from tissue organ bath for measurement of smooth muscle contraction (Caporale et al., 1980; Hoeprich et al., 1986; Huey et al., 1984; Lu et al., 1984; Unson et al., 1984), to mast cell histamine-release assays (Mousli et al., 1992) and ATP-release assays from guinea pig platelets (Ambrosius et al., 1989; Caporale et al., 1980; Federwisch et al., 1992; Gerardyschahn et al., 1988; Kohl et al., 1990; Kretzschmar et al., 1992; Pohl et al., 1993), which make it more difficult to compare the activity of these synthetic C3a peptides. Evaluation of the activities of synthetic ligands is further complicated by use of different species especially tissue and cells from guinea pigs, since guinea pigs have been found to have two functional C3a receptors (Fukuoka et al., 1998), in comparison to only one receptor found in humans. In this chapter, novel small synthetic C3aR agonists were developed from structure- activity relationships based on the hexapeptide FWPLAR, which was shown by previous studies in our group to have potency and selectivity for C3aR in calcium mobilisation and desensitisation assays on differentiated U937 cells respectively. Approximately 50 peptides were studied for their potential to bind and activate (agonists) or inhibit activation of human C3a receptor on human U937 cells and isolated human blood PBMCs. Two principal assays were used, (1) a competitive binding assay using 125I-labeled C3a (or C5a for selectivity comparisons) to evaluate selectivity of peptides and (2) a standard ligand-induced fluorescence intensity assay involving detection of intracellular calcium released into the cytoplasm from the endoplasmic reticulum by which were evaluated potency and selectivity (desensitisation assays) of peptide agonists. For the peptides evaluated herein there was a linear correlation between C3aR affinity and induced intracellular calcium release. The results have characterised the discovery of the most potent known small peptide agonists of C3aR (e.g. FLTLAR, EC50 320 nM in calcium mobilisation assay on dU937). Some compounds clearly bind to other receptors based on the results of an agonist-induced desensitisation assay, but key compounds appear to show selectivity and specificity for the C3a receptor.

2.2 AIMS/OBJECTIVES

To date, there have been no truly selective and potent agonists reported for probing the physiology and pharmacology roles of human C3a in vivo. Therefore, the development of small molecule agonists can be useful for understanding its role in vivo and for helping the development of antagonists for C3aR. 76 CHAPTER 2 Short Peptide Agonists for C3aR

The specific aims of this chapter were: 1. To determine the affinity and selectivity of hexapeptides for C3aR by using competitive [125I]-C3a and [125I]-C5a binding assays. 2. To evaluate the functionality of these compounds as C3aR agonists using an intracellular calcium mobilisation assay conducted on differentiated U937 cells. Structure-function relationships (SAR) were expected to inform which amino acids confer potency and selectivity for the ligands binding to C3aR over C5aR. 3. To further examine the selectivity of these compounds with receptor desensitisation experiments. 4. To re-examine the C5a decapeptides and native C3a on human monocytes for their binding affinity to C3aR, which can also provide more information for the ligand- receptor - binding conformation.

77 CHAPTER 2 Short Peptide Agonists for C3aR

2.3 RESULTS

2.3.1 SELECTIVITY AND POTENCY OF HUMAN C3a

2.3.1.1 Affinity of hC3a and hC5a

It is known that C3a and C5a are highly specific for their receptors with no detectable cross- reactivity (Burg et al., 1996; Crass et al., 1999). The selectivity was confirmed in the assays by two methods. Firstly, the competitive radioligands, [125I]-C3a and [125I]-C5a were used to check whether the ligands bind to C3aR and/or C5aR by competing with various concentrations of unlabelled ligands with the human C3a iodine labelled at a constant concentration. Secondly, a desensitisation experiment in which the administration of either C3a/C5a alone caused desensitisation of the C3aR or C5aR. The results of competitive radioligand binding studies confirm the selectivity of anaphylatoxin, showing that hC3a (IC50 0.1 nM, pIC50 9.99 ± 0.05, n=17) and the known C3aR antagonist, SB290157 (3) (IC50 143 nM, pIC50 6.84 ± 0.08, n=11) bind to C3aR (Figure 2.1) but not C5aR (Figure 2.2) on isolated human monocytes. A comparable result was found in [125I]-C5a competitive binding experiment. hC5a (IC50 1.0 nM, pIC50 8.87 ± 0.07, n=13) and the C5aR antagonist, AcF-[OPdChaWR] (3D53)(4) (IC50 124 nM, pIC50 6.85 ± 0.05, n=12) only bound with high affinity to C5aR (Figure 2.2) with no affinity for C3aR (Figure 2.1) on human isolated monocyte cells.

Figure 2.1 Affinities for C3a receptor of hC3a (l) n=17, SB290157 (3) (n) n=11, hC5a (Î) n=1 and 3D53 (4) (+) n=1 on isolated human monocytes. Affinities for C3aR measured by displacement of 80 pM [125I]-C3a on human PBMCs. It shows that human C5a and C5aR antagonist, 3D53 up to 1 µM and 100 µM respectively did not bind to C3aR. Each point represents the mean percentage specific binding ± S.E.M. 78 CHAPTER 2 Short Peptide Agonists for C3aR

Figure 2.2 Affinities for C5a receptor of hC5a (l) n=13, 3D53 (4) (n) n=12, hC3a (Î) n=3 and SB290157 (3) (+) n=3 on isolated human monocytes. Affinities for C5aR measured by displacement of [125I]-C5a on human PBMCs by unlabelled C5a, 3D53, hC3a or SB290157. It shows that human C3a and C3aR antagonist, SB290157 up to 1 µM and 100 µM respectively did not bind to C5aR. Each point represents the mean percentage specific binding ± S.E.M.

The second receptor selectivity measurement is based on receptor desensitisation. Receptor desensitisation, which usually follows exposure to agonists due to phosphorylation/internalisation of GPCRs, was used to measure the selectivity of C3a peptide analogues. Human C5a (300 nM) was administrated to differentiated U937 cells, which produces an efflux of intracellular Ca2+ that dissipates after 4-5 mins (Figure 2.3 A). A second addition of 300 nM C5a after 5 mins then has no effect (Figure 2.3 A), and this is attributed to either receptor (C5aR) phosphorylation or receptor internalisation from the cell surface. Either way the cell is no longer sensitive to C5a for at least 20 minutes. If instead the first administration of C5a (300 nM) is followed 5 minutes later by addition of 3 µM C3a (Figure 2.3 B), there is detectable fluorescence due to Ca2+ release, suggesting that C5a and C3a bind to different receptors and do not cross-react (Figure 2.3). When the experiment is reversed, and 3 µM C3a is administered to the cells first followed 5 minutes later by either 3 µM C3a (no signal, Figure 2.3 C) or 300 nM C5a (Ca2+ release, Figure 2.3 D), this shows that the cells have been desensitized only to C3a and not to C5a. The desensitisation experiment can therefore potentially confirm whether C3a agonists or C5a agonists are binding to the cell surface and inducing intracellular Ca2+ release.

79 CHAPTER 2 Short Peptide Agonists for C3aR

A. B.

C. D.

Figure 2.3 Desensitisation experiments. Anaphylatoxin receptor desensitisation experiments on differentiated U937 cells. Cells were first treated with hC5a (A and B) or hC3a (C and D), which induced a calcium response that dissipated over 4 mins prior to a second hC5a (A and D) or hC3a (B and C) injection.

2.3.1.2 Potency of hC3a and hC5a

Effect of Human C3a and Human C5a on Calcium Release Assay From U937 cells

U937 is a human leukemic monocyte lymphoma cell line. When C3a or a C3a analogue peptide binds to C3aR (a G-protein coupled receptor), intracellular G-proteins are activated through the exchange of GDP for GTP and the dissociation of the Gα and Gβγ subunits of G-proteins. When the receptor is activated, it leads to the release of intracellular calcium from the endoplasmic reticulum upon C3aR coupling to pertussis toxin-sensitive G-proteins (Gαi) subunits. C3aR may couple to Gα12/13 in endothelial cells (Schraufstatter et al., 2002). C3aR may also couple pertussis

80 CHAPTER 2 Short Peptide Agonists for C3aR toxin-insensitive Gα16 for signal transduction (Crass et al., 1996). The signal transduction of C3a includes activation of protein kinase C by phospholipase C and mitogen activated protein (MAP) kinases Erk1 and Erk2, in astrocytes (Langkabel et al., 1999; Sayah et al., 2003). To differentiate the monocyte-like phenotype of U937 cells with increased expression of C3aR on the cell surface, we exposed U937 cells to the membrane-permeable compound dibutyl- cAMP for 72 hours (Burg et al., 1996; Klos et al., 1992). This calcium mobilisation assay was used to measure the change in fluorescence induced by treatment of differentiated U937 cells with increasing C3a peptide concentrations (Figure 2.4).

Figure 2.4 Intracellular calcium release assay. 1. Fluo-3 is loaded onto the cells using a mild detergent, pluronic acid. 2. Fluo-3AM (red, outside the cells) is cleaved by intracellular esterases to release Fluo-3 (blue, inside the cell). 3. Activate GPCR with agonist. 4. Receptor activation leads to release of intracellular calcium release from the endoplasmic reticulum 5. Ca2+ binds to Fluo-3 in the cell and the Fluo-3 Ca2+ complex is excited at 485 nm light and emits at 520 nm light (adapted from Blakeney, 2007).

81 CHAPTER 2 Short Peptide Agonists for C3aR

In the calcium release assay, human C5a and human C3a stimulated release of intracellular calcium from differentiated U937 cells. hC5a is a more potent agonist (EC50 = 7 nM, pEC50 8.2 ±

0.09, n=3) than hC3a (EC50 = 52 nM, pEC50 = 7.3 ± 0.06, n=10). In this study it was shown that the upper plateau of C5a is higher than C3a-induced calcium release into the cytoplasm of differentiated U937 cells (Figure 2.5). There is a difference in maximum calcium response, which may be related to a difference in the calcium activation pathways between these receptors or the number of receptors that are expressed on the cell surface of differentiated U937 cells.

Figure 2.5 Intracellular Ca2+ release induced by human C3a (n) versus human C5a (l) in differentiated human U937 cells. C3a (n=10) and C5a (n=3) mediated calcium mobilisation from differentiated U937 cells expressed as percentage of maximum change in fluorescence induced by calcimycin (A23187) ± S.E.M.

2.3.2 SELECTIVITY AND POTENCY OF 15-RESIDUES OF hC3a AND hC5a

N-terminal truncation of C3a has been used to determine what constitutes the active portion of C3a but accessing short, potent peptides to mimic C3a function has not yet been successful.

2.3.2.1 Potency of 15-residue hC3a and hC5a

We next compared the C-termini of hC3a versus hC5a as agonists in the calcium 2+ mobilisation assay. At 100 µM concentration, C3a(63-77) (6) induced release of Ca into the cytoplasm of dU937 cells to about the same degree (EC50 ± 95% CI; 1.8 ± 0.55 µM) as 1 µM hC3a, whereas C5a(59-73)(7) at 100 µM was about ten-fold less effective (EC50 ± 95% CI; 27 ± 12.4 µM). However, there is a substantial reduction in functional potency upon truncation of C3a to its 15 82 CHAPTER 2 Short Peptide Agonists for C3aR residue C-terminal native peptide sequence which was measured by ability to induce intracellular calcium release from dU937 cells (Figure 2.6). The EC50 value of human C3a (EC50 52 nM) was reduced 35 fold when it was truncated to 15 residues (6) (EC50 1.8 µM). Compound 7 was around

3500 times less active than human C5a (hC5a EC50 = 7 nM; 7 EC50 = 24 µM).

2+ Figure 2.6 Intracellular Ca release induced by C3a(63-77) (6) (®) versus C5a(60-74) (7) (q) in differentiated human U937 cells. 6 and 7 mediated calcium mobilisation from differentiated U937 cells expressed as percentage of maximum change in fluorescence induced by human C3a (1 µM) and error bars represented S.E.M. with n=4.

2.3.2.2 Binding Affinities of 15-residue hC3a and hC5a

From the results of competitive binding studies, 7 has binding affinity for both C3aR (IC50

1.6 µM, Figure 2.7) and C5aR (IC50 1.7 µM, Figure 2.8) on human monocytes whereas the 6 has shown only affinity for C3aR (IC50 = 152 nM, Figure 2.7) but not C5aR (Figure 2.8) on human PBMCs. Thus the 15 residue peptide based on the C-terminal native sequences of C5a was not selective for its receptor so we decided to further study the binding of reported C5a decapeptide agonists: YSFKPMPL(Me-a)R (8), YSFK(Me-D)MPLaR (9), YSHKPMPLaR (10), YSFKPMPLaR (11) and C3aR superagonist –WWGKKYRASKLGLAR (5).

83 CHAPTER 2 Short Peptide Agonists for C3aR

Binding affinity for C3aR

Figure 2.7 Competitive [125I]-C3a binding assay of hC3a (l, n=17), 3 (n, n=11), C5a (Î n=1), 4 (+, n=1), 6 (p, n=4) and 7 (q, n=3) on isolated human monocytes. 80 pM [125I]- C3a was competed by increasing concentration of non-labelled hC3a (concentration 1 pM to 1 µM), 3, 6 and 7 (concentration 0.1 nM to 100 µM or 1 mM). 6 and 7 bind to C3aR. Each point represents the mean percentage specific binding ± S.E.M.

Binding affinity for C5aR

Figure 2.8 Competitive [125I]-C5a binding assay of hC5a (Î, n=13), 4 (+, n=12), C3a (l, n=3), 3 (n, n=3), 6 (p, n=3) and 7 (q, n=4) on human monocytes. Competitive [125I]-C5a binding of hC5a (concentration 1 pM to 1 µM), 4 and 7 (concentration 0.1 nM to 100 µM or 1 mM) on monocytes shows that hC3a up to 1 µM and 3, 6 up to100 µM respectively do not bind to C5aR. Each point represents the mean percentage specific binding ± S.E.M.

84 CHAPTER 2 Short Peptide Agonists for C3aR

2.3.3 SELECTIVITY AND POTENCY OF KNOWN TRUNCATED C3a PEPTIDES

2.3.3.1 Decapeptides

A set of decapeptides has been reported as C5aR agonists (Kawatsu et al., 1996; Taylor et al., 2001; Vogen et al., 2001), namely 8, 9, 10 and 11. From a previous study (Proctor et al., 2004)

C3aR binding affinities of 8 and 9 were IC50= 43 nM and IC50= 19 nM respectively when competing with 100 pM [125I]-C3a in polymorphonuclear cells (PMN), but bind to C5aR with lower 125 affinity (IC50 = 39 µM and 10 µM when competing with 50 pM [ I]-C5a on PMN). In our hands in competitive studies using human PBMCs competing with 80 pM [125I]-C3a or 20 pM [125]-C5a, these decapeptides (8, 9, 11) bind with greater affinity to C3aR than C5aR

(Table 2.1, Figure 2.9) especially 8 which has IC50 =14 nM for C3aR while binding to C5aR with lower affinity (IC50= 1.6 µM) which supports the result of previous studies using PMN (Proctor et al., 2004). In contrast, 10 shows no binding to C3aR (Figure 2.9) and C5aR (Figure 2.10) but it did stimulate the intracellular calcium release, which may cause stimulation of receptors other than C3aR and C5aR (Table 2.1).

2.3.3.1.1 Affinity of Known Truncated Anaphylatoxin Agonists: Decapeptides

Binding affinity for C3aR

Figure 2.9 Competitive [125I]-C3a binding assay on human PBMCs. Affinities for C3aR measured by displacement of 80 pM 125I-C3a by unlabelled hC3a (l) n=17, SB290157 (n) n=11 and 8 (q), 9 (p), 10 (♦) and 11 (¡) (n=3). It shows that 10 (♦) up to 100 µM did not bind to C3aR. Each point represents the mean percentage specific binding ± S.E.M.

85 CHAPTER 2 Short Peptide Agonists for C3aR

Binding affinity for C5aR

Figure 2.10 Competitive [125I]-C5a binding assay on human PBMCs. Affinities for C5aR measured by displacement of 20 pM 125I-C5a by unlabelled hC5a (l) (concentration 1 pM to 1 µM, n=13), 3D53 (n) n=12, 8 (q), 9 (p), 10 (♦) and 11 (¡) (n=3) (concentration 0.1 nM to 100 µM). Each point represents the mean percentage specific binding ± S.E.M.

2.3.3.2 Known Truncated C3a: C3a Superagonist and C3a hexapeptides

C3a 15-residue peptide WWGKKYRASKLGLAR (5) or “superagonist” was first reported by Ember and colleagues (1991) by adding a tryptophan dipeptide as the hydrophobic moiety and link to the C-terminal pentapeptide LGLAR that was reported to be fifteen-fold more potent than C3a in a guinea pig platelet aggregation assay. From the previous study by our group to reinvestigate the superagonist in a calcium mobilisation assay it was found that the compound 5 induced the release of intracellular calcium from dU937 cells (EC50 = 1.0 µM) but with nineteen-fold less potency than native hC3a (EC50 = 52 nM). Herein, we also further examined the selectivity of 5 by competing with 80 pM [125I]-C3a or 20 pM [125I]-C5a. The last six residues in the sequence of native hC3a, HLGLAR (12) and HLALAR (13) and the two pentapeptides YPLPR (14) and WPLPR (15) which were claimed to have an analgesic effect in mice when given intracerebroventricularly and be selective for C3aR (Jinsmaa et al., 2001) were also reinvestigated by radioligand competitive assays. 86 CHAPTER 2 Short Peptide Agonists for C3aR

2.3.3.2.1 Affinity of C3a Superagonist and C3a hexapeptides

From the results of this study we found that 5 binds with high affinity to C3aR (Figure

2.11) (IC50 = 2 nM) but with 20 times less affinity than native hC3a (IC50 = 0.1 nM) and does not bind to C5aR (Figure 2.12). However, the C-terminal hexapeptide sequences of human C3a

(HLGLAR, 12) binds to C3aR with lower affinity at IC50 = 5 µM which is similar to the binding affinity of HLALAR (13). In contrast, compound 15 has much weaker affinity for C3aR while 14 shows no affinity for C3aR (Table 2.1). Binding affinity for C3aR

Figure 2.11 Competitive [125I]-C3a binding assay of C3a superagonist and hexapeptides on human PBMCs. Affinities for C3aR measured by displacement of 80 pM 125I-C3a by unlabelled hC3a (l) n=17, 3 (n) n=11, 5 (p), 12 (q), 13 (♦), 14 (¡) and 15 (l) (n=3) for C3aR. Binding affinity for C5aR

Figure 2.12 Competitive [125I]-C5a binding assay on human PBMCs. Affinities for C3aR measured by displacement of 20 pM 125I-C5a from human PBMCs by unlabelled hC5a (l) n=13, 4 (n) n=12, 5 (p), 12 (q), 13 (♦), 14 (¡) and 15 (l) (n=3) for C5aR. Each point represents the mean percentage specific binding ± S.E.M. 87 CHAPTER 2 Short Peptide Agonists for C3aR

2.3.3.2.2 Potency of Superagonist and known hexapeptides

In the Calcium Release Assay (Figure 2.13), the agonist potency of the compounds correlated to the binding affinity of the compounds. For example, 12 which had lower affinity for

C3aR than 5 and 8 also showed lower potency than these compounds. Surprisingly, 5 (IC50 = 2

nM) which had six fold higher affinity for C3aR than 8 (IC50 = 13 nM) showed ten fold less

potency (EC50 = 1.3 µM) than 8 (EC50 = 0.13 µM) in the calcium mobilisation assay on dU937.

Figure 2.13 Activity of hC5a (n), hC3a (l), 8 (q), 5 (♦) and 12 (p) in the intracellular calcium mobilisation assay measured in human dU937 cells (n ≥ 3). Data is expressed as percentage of normalised maximum change in fluorescence ± S.E.M.

88 CHAPTER 2 Short Peptide Agonists for C3aR Table 2.1 Affinity and Potency of Known Truncated Anaphylatoxin Agonists on dU937 cells.

C3a Receptor Affinity C5a Receptor Affinity Apparent Agonist Activity

a b c # Peptide n pIC50 ± IC50 n pIC50 ± IC50 n pEC50 ± EC50 SE (nM) ± SES.E. (nM) SE (µM) 1 C3a 17 10 ± 0.05 0.1 3 ϕ - 10 7.3 ± 0.06Ð 0.05 2 C5a 1 ϕ - 13 9.0 ± 0.07 1.02 3 8.2 ± 0.09Ð 0.01 3 SB290157 11 6.8 ± 0.08 143 3 ϕ - 3 φd - 4 3D53 3 ϕ - 12 6.9 ± 0.05 124 3 φ - 5 WWGKKYRASKLGLAR 3 8.7 ± 0.20* 2.0 3 ϕ - 5 5.9 ± 0.16 1.3

6 C3a63-77 3 6.6 ± 0.23 250 3 ϕ - 4 5.8 ± 0.11 1.8 ë Ð 7 C5a60-74 3 5.8 ± 0.21* 1600 3 5.8 ± 0.17 1700 4 4.6 ± 0.11 24 8 YSFKPMPL(Me-a)R 3 7.9 ± 0.13* 13 3 5.7 ± 0.14ë 2000 5 6.9 ± 0.06Ð 0.13 9 YSFK(Me-D)MPLaR 3 6.1 ± 0.19 740 3 4.5 ± 0.45ë 29000 5 5.9 ± 0.05 1.4 10 YSHKPMPLaR 3 ϕ - 3 ϕ - 3 5.2 ± 0.03 5.7 11 YSFKPMPLaR 3 6.9 ± 0.10 140 3 5.3 ± 0.21ë 4700 2 6.7 ± 0.07Ð 0.22 12 HLGLAR 3 5.2 ± 0.21* 5900 3 ϕ - 3 5.4 ± 0.31 3.8 13 HLALAR 3 5.2 ± 0.29* 6300 3 ϕ - 3 φ 14 YPLPR 3 ϕ - 3 ϕ - 2 φ 15 WPLPR 3 4.5 ± 0.50* 27000 3 ϕ - 2 φ a Concentrations causing 50% inhibition of maximum binding of 125I-C3a to intact human PBMCs. bConcentration causing 50% inhibition of maximum binding of 125 c d I-C5a to intact human PBMCs; Concentration causing 50% maximum calcium mobilisation from Bt2-cAMP differentiated U937 cells. No agonist activity at 1

mM concentration, antagonist with pIC50 = 5.9 ± 0.1 (IC50 1.3 µM). ϕ No binding detected at 1 mM. φ No intracellular calcium release detected at 10 µM. pIC50

and pEC50 expressed as mean ± SE; n represents the number of separate experiments performed.; * p < 0.05 vs. 3 for 5, 7, 8, 12, 13 and 15; ë p < 0.05 vs. 4 for 7, 8, 9, 11 ; Ð p < 0.05 vs. 12 for 1, 2, 7, 8 and 11. 90 CHAPTER 2 Short Peptide Agonists for C3aR

2.3.4 C3a DES-ARG COMPOUNDS

2.3.4.1 Binding Affinities of des-Arg compounds to C3aR

The importance of arginine at the C-terminus of hC3a was confirmed by designing two compounds without arginine, FLTLA (26) and FISLA (27), which were examined in the competitive assay with [125I]-C3a. The results showed no binding of these desArg compounds (Figure 2.14). As we know the cleavage of C-terminal arginine by carboxypeptidase N converts C3a into C3a des-Arg, which lacks anaphylatoxic properties and deletes its ability to bind to C3aR. C3a desArg is also known as acylation-stimulating protein (ASP), which has potent anabolic effects on human adipose tissue where it increases glucose uptake and nonesterified fatty acid storage (Cianflone et al., 1999; Cianflone et al., 2003; Murray et al., 1999; Xia et al., 2004). Recently, it was found that ASP binds to C5L2 and stimulates triglyceride synthesis (Kalant et al., 2005).

Figure 2.14 Competitive [125I]-C3a binding of FLTLA (26) (Â) and FISLA (27) (n) on isolated human monocytes. Isolated human monocytes (15x106 cells/mL) were incubated for 1 h at room temperature with a constant concentration (80 pM; 2200 Ci/mmol) of 125I-C3a and increasing concentration of non-labelled hC3a (concentration 1 pM to 1 µM) and non-labelled hC5a (concentration 1 pM to 1 µM) or other compounds (0.1 nM to 100 µM). hC3a (l) n=17, 3 (n) n=11, 4 (+) n=1 and hC5a (x) n=1. Compound 26 and 27 up to 100 µM did not bind to C3aR. Each point represents the mean percentage specific binding ± S.E.M.

91 CHAPTER 2 Short Peptide Agonists for C3aR

2.3.5 C3a HEXAPEPTIDE ANALOGUES

Since the C-terminal Leu-Ala-Arg motif has been shown to be vital for activity at C3aR (Caporale et al., 1980; Kohl et al., 1990), these residues were incorporated with the shortest C5aR peptide agonist, FKP(dCha)(Cha)r-NH2 (Drapeau et al., 1993; Kawai et al., 1991; Konteatis et al.,

1994) to produce hexapeptide F6KPLAR1 which from previous studies in our group, was believed to have some activity and selectivity for C3aR (Proctor, 2004). The aim of the present study was to produce peptides to mimic the C-teminus of human C3a but deliver higher potency and selectivity as C3aR agonists for potential use as pharmacological probes of C3aR in vitro and ultimately in vivo.

2.3.5.1 Binding affinity of hexapeptide analogues

2.3.5.1.1 Affinity for C3aR of hexapeptide analogues

The most potent peptide analogues FLTLAR (17) and FWTLAR (20) bind to human isolated monocytes C3aR (Figure 2.15) with high affinities (IC50 = 42 nM, pIC50 ± SE = 7.38 ±

0.26, n=3 and IC50 = 82 nM, pIC50 = 7.09 ± 0.33, n=3, respectively), but not C5aR (IC50> 100 µM) (Figure 2.16). The D-Arg containing peptide FLTLAr (24) had 200-fold lower affinity than 17

(IC50 = 8 µM, pIC50 ± SE = 5.09 ± 0.5, n=3) (Table 2.2).

Figure 2.15 Competitive [125I]-C3a binding of hC3a, SB290157, C3a and C5a 15 residues and C3a hexapeptide analogues on isolated human PBMCs. Isolated human monocytes (15x106 cells/mL) were incubated for 1 h at room temperature with a constant concentration (80 pM; 2200 Ci/mmol) of 125I-C3a and increasing concentration of non-labelled C3a (concentration 1 pM to 1 µM) and non-labelled C5a (concentration 1 pM to 1 µM) or hexapeptide agonists (0.1 nM to 100 µM) or 7, 54 and 24 (0.1 nM to 1 mM). hC3a (l) n=17, 3 (n) n=11, 4 (+) and hC5a (x) , 6 (p), 7 (q) , 16 (♦), 17 (n), 20 (p), 18 (q), 22 (♦), 19 (l), 23 (∗), 21 (+), 54 (f), 24 (❢) n ≥ 3. Each point represents the mean percentage specific binding ± S.E.M. 92 CHAPTER 2 Short Peptide Agonists for C3aR

2.3.5.1.2 Affinity for C5aR of hexapeptide analogues

To determine whether compounds selectively bound to C3aR over C5aR, competition with C5a was also evaluated. By using competitive radioligand binding on human isolated monocytes 125 the IC50 of each ligand were determined by displacement of [ I] labelled C5a.

From these results, none of the C3a hexapeptide analogues (IC50 > 100 µM) bind to C5aR on human isolated monocytes, except for the C5a-terminal peptide (7) which did bind to C5aR (Figure 2.8) suggesting that all the hexapeptides are selective for C3aR over C5aR.

Figure 2.16 Competitive [125I]-C5a binding of hC5a, AcF-[OPdChaWR] (3D53) and C3a hexapeptide analogues on isolated human monocytes. Isolated human monocytes (15x106 cells/mL) were incubated for 1 h at room temperature with a constant concentration (20 pM; 2200 Ci/mmol) of 125I-C5a and increasing concentration of non-labelled hC5a (concentration 1 pM to 1 µM) and non-labelled hC3a (concentration 1 pM to 1 µM), 3D53(4) (0.01 nM to 1 µM) or hexapeptide agonists (0.1 nM to 100 µM) or 24 (0.1 nM to 1 mM). hC5a (l) n=13, 4 (n) n=12, 3 (+) and hC3a (x) , 25 (n), 16 (p), 17 (q), 18 (♦), 20 (l), 22 (∗), 19 (+) , 23 (f), 21 (n), 24 (q) , n ≥ 3. Each point represents the mean percentage specific binding ± S.E.M.

93 CHAPTER 2 Short Peptide Agonists for C3aR

Table 2.2 Structure-Activity relationship (SAR) data for hexapeptide analogues of selective agonist F1WPLAR6. C3a Receptor Affinity C5a Receptor Affinity Apparent Agonist Activity

a b c # Peptide n pIC50 ± IC50 n pIC50± IC50 n pEC50 ± EC50 SE (nM) SE (nM) SE (µM) 1 C3a 17 10 ± 0.05 0.10 3 ± S.E.ϕ - 10 7.3S.E. ± 0.06 0.05 2 C5a 1 ϕ - 13 9.0 ± 0.07 1.02 3 8.2 ± 0.09 0.01 3 SB290157 11 6.8 ± 0.08 143 3 ϕ - 3 φd 4 3D53 3 ϕ - 12 6.9 ± 0.05 124 3 φ 16 FWPLAR 4 6.6 ± 0.30 274 3 ϕ - 5 6.0 ± 0.15 0.95 17 FLTLAR 3 7.4 ± 0.26 42 3 ϕ - 5 6.5 ± 0.12 0.32 18 FIPLAR 3 6.6 ± 0.13 230 3 ϕ - 2 6.0 ± 0.23 0.98 19 WWTLAR 3 6.6 ± 0.12 244 3 ϕ - 3 6.0 ± 0.11 0.92 20 FWTLAR 3 7.1 ± 0.33 82 3 ϕ - 4 6.5 ± 0.16 0.35 21 FYTLAR 3 6.8 ± 0.17 166 3 ϕ - 2 5.3 ± 0.35Ð 4.8 22 FWGLAR 3 6.7 ± 0.14 204 3 ϕ - 2 6.3 ± 0.09 0.47 23 FLGLAR 3 6.8 ± 0.12 157 3 ϕ - 3 6.3 ± 0.33 0.52 24 FLTLAr 3 5.1 ± 0.50* 8200 3 ϕ - 2 4.2 ± 0.12Ð 57 25 FLTChaAR 2 6.6 ± 1.59 238 3 ϕ - 4 φ 26 FLTLA 2 ϕ - - 1 φ 27 FISLA 2 ϕ - - 1 φ

a Concentrations causing 50% inhibition of maximum binding of 125I-C3a to intact human PBMCs. bConcentration causing 50% inhibition of maximum binding 125 c d of I-C5a to intact human PBMCs; Concentration causing 50% maximum calcium mobilisation from Bt2-cAMP differentiated U937 cells. No agonist activity at

1 mM concentration, antagonist with pIC50 = 5.9 ± 0.1 (IC50 1.3 µM). ϕ No binding detected at 100 µM. φ No intracellular calcium release detected at 100 µM.

pIC50 and pEC50 expressed as mean ± SE; n represents the number of separate experiments performed.; pIC50 for C3aR: * p < 0.05 vs. 17 for 24. pEC50: Ð p < 0.05 vs. 17 for 21 and 24.

94 CHAPTER 2 Short Peptide Agonists for C3aR

2.3.5.2 Activity of hexapeptide agonists

Figure 2.17 Activity of putative C3a agonist peptides at a concentration of 100 µM. Calcium release from differentiated U937 cells induced by 100 µM of agonists is expressed as % maximum change in fluorescence induced by 1 µM of C3a. Error bars represented S.E.M. with n=2.

95 CHAPTER 2 Short Peptide Agonists for C3aR

Figure 2.18 Activity of putative C3a agonist peptides at a concentration of 1 µM. Calcium release from differentiated U937 cells induced by 1 µM of agonists is expressed as % maximum change in fluorescence induced by 1 µM of C3a. Error bars represented S.E.M. with n=4.

96 CHAPTER 2 Short Peptide Agonists for C3aR

2.3.5.2.1 Structure-activity relationships (SAR) for hexapeptide agonist analogues of F1WPLAR6: Single Point Agonist Potency Comparisons

A. Substitution at position 1 or 1 and 3, F1WP3LAR or B. Substitution at position 2, FW2PLAR or FL2TLAR

F1LT3LAR

97 CHAPTER 2 Short Peptide Agonists for C3aR C. Substitution at position 3, FWP3LAR or FLT3LAR D. Substitution at position 4 or 3 and 4, FWP3L4AR or

FLTL4AR

E. Substitution at position 5 and 6 FLTLA5R6 F. Cyclic peptides and others

Figure 2.19 Structure Activity Relationship studies of hexapeptide agonists by substitution of F1WPLAR6 or F1LTLAR6. Agonist activity for hexapeptides, which are compared by substitution at different positions, induced calcium release from human dU937 cells. 98 CHAPTER 2 Short Peptide Agonists for C3aR

Table 2.3 Structure-Activity of F1WPLAR6 and F1LTLAR6 analogues on human dU937 cells measured by calcium release assay. Apparent Agonist Activity c # Peptide n pEC50 ± S.E. EC50 (µM) 16 FWPLAR 5 6.02 ± 0.15 0.95 19 WWTLAR 3 6.04 ± 0.06 0.91 28 hFLTLAR 2 4.96 ± 0.08ëÐ 11 29 Ac-FLTLAR 2 6.28 ± 0.15 0.52 30 YLTLAR 2 5.61 ± 0.15 2.5 31 RLTLAR 4 ϕ - 32 F(Cha)PLAR 5 6.13 ± 0.22 0.75 33 F(Nap)PLAR 3 6.16± 0.22 0.70 34 F(hF)PLAR 3 5.98± 0.21 1.0 18 FIPLAR 2 6.01 ± 0.23 0.98 35 FOTLAR 2 ϕ - 36 FRTLAR 2 5.17 ± 0.36ëÐ 6.8 21 FYTLAR 2 5.32 ± 0.35ë 4.8 37 FYSLAR 2 4.82 ± 0.40*ëÐ 15 38 FKTLAR 2 5.36 ± 0.40 4.4 39 FNleTLAR 3 6.51 ± 0.30 0.31 40 FVTLAR 2 5.32 ± 0.15ë 4.8 41 FISLAR 3 5.95± 0.13 1.1 22 FWGLAR 2 6.33 ± 0.09 0.47 42 FWALAR 3 5.93 ± 0.12 1.2 43 FWSLAR 5 6.31 ± 0.09 0.49 20 FWTLAR 6 6.43 ± 0.09 0.37 17 FLTLAR 5 6.50 ± 0.12 0.32 23 FLGLAR 3 6.29 ± 0.33 0.52 44 FLPLAR 3 6.36± 0.31 0.44 45 FLALAR 3 5.96 ± 0.23 1.1 46 FLSLAR 4 6.45 ± 0.19 0.36 47 FLTlAR 2 ϕ - 48 FLTNleAR 2 5.58 ± 0.26 2.7 25 FLTChaAR 4 ϕ - 49 FLT(dCha)AR 2 ϕ -

99 CHAPTER 2 Short Peptide Agonists for C3aR 50 FLTSAR 3 ϕ - 51 FWTVAR 2 ϕ - 52 FWSIAR 2 5.38± 0.10 4.1 53 FLTLLR 4 ϕ - 54 FLTLPR 2 4.54 ± 0.19*ëÐ 29 24 FLTLAr 2 4.24 ± 0.12*ëÐ 57 ëÐ 55 FLTLAR-NH2 2 4.44 ± 0.19* 37 a Concentration causing 50% maximum calcium mobilisation from Bt2-cAMP differentiated human U937 cells. ϕ No calcium release detected at 1 mM. * p < 0.05 vs. 16 for 24, 37, 54, and 55; ë p < 0.05 vs. 17 for 21, 24, 28, 36, 37, 40, 54, 55; Ð p < 0.05 vs. 20 for 24, 28, 36, 37, 54 and 55.

Structure-activity relationships (Figure 2.17, 2.18 and 2.19) suggest the following points. The results of single point studies as well as dose concentration profiles suggest the following points. Position 1: Phenylalanine (F) at position 1 (corresponding to C3a residue 72) is an aromatic residue that lacks charge. When substituted by histidine (H) or arginine (R) agonist activity decreases, but is of comparable potency when replaced by tryptophan (W). Thus phenylalanine followed by tryptophan at this position appears to have the best size, aromaticity and lack of charge, required for C3a activity (Figure 2.19A and Table 2.3). Position 2: Substitution of tryptophan (W) in FWPLAR (16) by other hydrophobic amino acids such as Cha, Nap, hF, isoleucine (I), or leucine (L) did not show any significant change in nd EC50. However, substitute at the 2 position of FL2TLAR with ornithine (O), arginine (R), lysine (K), Tyrosine (Y) or valine (V) gave detrimental to agonist activity (Figure 2.19B and Table 2.3). Position 3: Substitution of proline (P) at position 3 with serine (FWSLAR) or threonine (T) or glycine (G) increased agonist potency, whereas alanine (A) at this position did not improve agonist activity (Figure 2.19C and Table 2.3).

Position 4: Substitution of leucine at position 4 with valine (V), Cha, D-Cha or D-L, diminishes agonist activity. Furthermore, substitution with isoleucine (I) (from FWSLAR to FWSIAR) results in 10 times less agonist potency (Figure 2.19D and Table 2.3). Position 5: Substituting alanine (A) at position 5 with leucine shows a decreased agonist activity (FLTLAR (17), EC50 ± 95% CI; 0.37 ± 0.18 µM; FLTLLR (53) no activity at 100 µM) (Figure 2.19E and Table 2.3).

Position 6: Replacing arginine (R) at position 6 with D-R decreased potency (Figure 2.19E and Table 2.3).

100 CHAPTER 2 Short Peptide Agonists for C3aR

In conclusion, position 1 requires a hydrophobic, aromatic residue such as phenylalanine or tryptophan (Figure 2.20, 2.21) for C3a agonist activity. All agonist potency is destroyed by insertion of polar, hydrophilic residues such as histidine and arginine at this position. For the second position, substitution with hydrophobic groups such as Cha, Nap or hF show similar agonist activity to FWPLAR. However, substitution by ornithine, lysine or arginine at this position lower potency of agonist activity (Table 2.3). At position 3 substitution by proline or alanine shows similar potency while substitution by serine, glycine or threonine showed an increase in potency (Figure 2.21). At positions 4, 5, 6 the substitution of leucine, alanine and arginine by any other residue was shown to be detrimental to agonist activity or had no improvement (Figure 2.22).

Therefore residues L4A5R6 are extremely important for the activity of these hexapeptides.

16 17

18 19

Figure 2.20 Structure of hexapeptides 16 - 20.

101 CHAPTER 2 Short Peptide Agonists for C3aR

2.3.5.2.2 Dose dependence of hexapeptide analogues

Figure 2.21 Intracellular calcium release in differentiated U937 cells induced by hC3a (l), 20 (l) and 46 (l), 19 (l) and 17 (l). The results are expressed as a percentage of maximum change in fluorescence ± S.E.M. (n=3).

Figure 2.22 Intracellular Ca2+ release induced by variable concentration of C3a peptide analogues. The dose response activity of 16 (n), 42 (p), 20 (q), 17 (®) 19 (l), 48 (¨), 21 (r), 45 (q), 18 (®), 25 (l), 53 (x), 31 (+), 41 (∗) and 46 (+) mediated intracellular Ca2+ mobilisation was measured in differentiated U937 cells (n ≥ 3). The results are expressed as a percentage of maximum change in fluorescence induced by hC3a (1 µM) ± S.E.M. 102 CHAPTER 2 Short Peptide Agonists for C3aR

The concentration response curve for differentiated U937 cells on some C3a peptide analogues are shown in Figure 2.22. From Table 2.3, the most potent hexapeptide agonists in this study were 17 (EC50 ± 95% CI; 0.37 ± 0.18 µM, pEC50 6.50 ± 0.12, n=5), FNleTLAR (39) (EC50 ±

95% CI; 0.71 ± 0.65 µM, pEC50 6.51 ± 0.3, n=3) and 20 (EC50 ± 95% CI; 0.45 ± 0.29 µM, pEC50 6.46 ± 0.16, n=4).

2.3.5.2.3 Desensitisation of hexapeptide analogues.

A. B.

C. D.

E. F.

103 CHAPTER 2 Short Peptide Agonists for C3aR Figure 2.23 Desensitisation assays of hexapeptide analogues and C3a superagonist peptide.

We next used a desensitisation experiment to test our C3a peptide analogues for indications of selectivity for the C3a receptor over the C5a receptor. Figure 2.23 shows that administration of 3 µM C3a to differentiated U937 cells causes an efflux of intracellular Ca2+ and successive addition of 100 µM 16 (Figure 2.23 B) and 300 µM 5 (Figure 2.23 F) still cause Ca2+ release. This suggests that 16 and 5 act on a different receptor to the desensitised C3a receptor. In contrast, 18, 20 and 17 (Figure 2.23 C, D and E, respectively) do not stimulate calcium release following C3a receptor desentisation indicating that they are selective for C3aR.

Figure 2.24 Correlation between binding affinity (pIC50) and agonist activity (pEC50) of hexapeptide agonists.

104 CHAPTER 2 Short Peptide Agonists for C3aR

(A).

(B).

Figure 2.25 NMR-derived solution structure for FLTLAR in DMSO-d6 (298 K). (A) Representative NMR structure of FLTLAR showing the H-bond formed between the side chain hydroxyl group of Thr4 and the backbone amide proton of Leu3. A distance of 2.8 Å separates the heavy atoms involved in the bond. (B) 15 of 20 lowest energy minimized NMR structures showing the backbone of FLTLAR. Residues are labelled with three-letter amino acid codes (These structures were produced by Dr. Scully, C.C.G. and Dr. Hoang, H.N. from Fairlie group).

2.4 DISCUSSION

Human C3a is implicated in the pathology of many inflammatory conditions and immune diseases. The finding of either an agonist or antagonist would have many beneficial uses. Unlike C5a, the pathological role of C3a has been difficult to assess due to the continuous degradation of C3 that releases into blood stream. C5a generally has an order of magnitude higher agonist potency than C3a in vitro and in vivo. However, the concentration of C3a in the plasma is significantly higher (between 10 to 15 fold) than C5a (Bitter 105 CHAPTER 2 Short Peptide Agonists for C3aR Suermann, 1988; Foreman et al., 1996). Some of the cellular and physiological responses associated with C3a include chemotaxis, induction of smooth muscle contraction, activation of leukocytes (primarily mast cells and eosinophils), secretion of proinflammatory cytokines and an increase in vascular permeability, which leads to migration of leukocytes into inflammatory sites (Gerard et al., 2002; Haas et al., 2007; Hawlisch et al., 2004). C3a and C5a generate their effects through specific receptors, C3aR and C5aR, which belong to the rhodopsin family of G-protein coupled receptors. C3aR and C5aR also share 37% amino acid sequence homology in the transmembrane region and the second intracellular loop (Haas et al., 2007). A two-site model, like C5aR, has been postulated for C3aR. Similar to C5aR, the C3a C-terminus synthetic peptide analogues can activate C3aR. The two-site binding model has been proposed for C5a binding to C5aR in which the N-terminal core of C5a binds to the N-terminus of the C5aR while the C-terminus binds to the active site which is located in the transmembrane region of C5aR (Chen et al., 1998; Siciliano et al., 1994). C3aR has a unique characteristic among G-protein coupled receptor (GPCR) of a large extracellular loop 2 which plays an important role in C3a binding (Gao et al., 2003). C3aR and C5aR are expressed in similar cells, especially on myeloid cells such as neutrophils, eosinophils, basophils, monocytes and mast cells. Recently, they have also been found in non-myeloid cells (epithelial and endothelial cells). In leukocytes, C3a and C5a mediate their effects by coupling

to pertussis toxin sensitive G-protein (Gαi) and pertussis toxin-insensitive G-protein (Gα16) (Buhl et al., 1995; Norgauer et al., 1993; Vanek et al., 1994; Zwirner et al., 1997). C3aR also

couples to Gα12 and Gα13 in endothelial cells (Schraufstatter et al., 2002). The coupling of

C3aR to Gαi which inhibits adenylate cyclase enzyme and the βγ subunit activates the MAP kinase pathway and phospholipase C (PLC). Cleavage of phospholipid membrane by PLC 2+ produces phosphoinositol 3 (IP3) which mobilizes intracellular Ca and diacylglycerol (DAG) which leads to the activation of proteinkinase C (PKC). After agonist activation of the GPCR the intracellular surface of the receptor becomes susceptible to phosphorylation by G- protein-coupled receptor kinases (GRKs), which promotes binding of arrestins to phosphorylation sites. In turn, this helps to prevent further agonist stimulation and uncoupling of G-protein (Haas et al., 2007). It is known that the residues responsible for activation of C3aR are localized within the C-terminal octapeptide region of C3a. There have been efforts to develop synthetic C3a ligands that functionally mimic C3a by creating short peptides based on the C3a C-terminus. Like C5a, the C-terminal arginine is vital for C3a biological activity. In the circulation, the enzyme carboxypeptidase N cleaves the C-terminal arginine of C3a and C5a forming the

106 CHAPTER 2 Short Peptide Agonists for C3aR degradation products C3a desArg and C5a desArg. The results of binding studies also shows that compounds FLTLA and FISLA do not bind to C3aR (Figure 2.14). There are difficulties in comparing the reported activities of small synthetic C3a peptides due to the variety of assay systems used to evaluate these ligands. Furthermore, guinea pig tissues and cells have been used, which have two functional C3aR in comparison to only one in human (Fukuoka et al., 1998). Therefore, known C3a agonists were reinvestigated for binding affinity against human C3aR as well as human C5aR and in a calcium mobilisation assay for function. From the competitive binding studies we found the decapeptide compounds which were originally reported as C5aR agonists bound much more tightly to C3aR than C5aR which was also consistent with the previous report of competitive binding in PMN (Proctor et al., 2004). While the C3a superagonist, WWGKKYRASKLGLAR was first reported as being 12-15 times more potent than hC3a in a guinea pig platelet aggregation assay, vascular permeability and ileum contraction (Ember et al., 1991) it was found by us to bind tightly to C3aR on human monocyte cells but had weak agonist activity in the calcium mobilization assay in dU937 cells. It also does not show selectivity for C3aR in a desensitization assay, in which induced calcium release from cells was measured after the C3aR was first desensitized by administration of hC3a (3 µM) to the cells. The cause of calcium release from the cells may be due to triggering of other receptors than C3aR. From the new series of hexapeptides obtained by modification of the C-terminus of

native human C3a, we found FWTLAR (20) and FLTLAR (17) are the most potent (EC50 ~ 0.3 µM) and also show selectivity for C3aR in competitive radioligand binding (in both [125I]- C3a and [125I]-C5a)) and a desensitization assay in which they do not induce the release of calcium after desensitization of the cells with C3a. There was a pseudo-linear correlation (Figure 2.24) between agonist potency (by measuring the release of Ca2+ into human dU937 cells) and the binding affinity for C3aR (measured by competition with [125I]- labeled C3a on PBMCs). From these data, we were able to find a new antagonist FLTChaAR (25) which has comparable binding affinity to

SB290157(3) for human C3aR on PBMCs with IC50 = 238 nM and IC50 = 143 nM respectively (Table 2.2). FLTChaAR (25) also does not bind to C5aR in a competitive binding assay with iodine labeled C5a. Unsurprisingly, 25 is an equipotent antagonist in the

calcium mobilization assay to inhibit C3a-induced calcium release from dU937 (IC50 = 1.5

µM vs IC50 1.3 µM)(Scully et al., 2010).

107 CHAPTER 2 Short Peptide Agonists for C3aR

An NMR-derived solution structure study from our group member for FLTLAR (17) (Figure 2.25) showed significant “turn” or “bend” propensity in DMSO. A feature of the C- terminus of C3a itself hints at a potential requirement for this structural feature for binding within the transmembrane region of human C3aR. This result is consistent with the finding that over 100 peptide-activated G-protein coupled receptors have shown a tendency to recognize protein and peptide ligands with turn structure (Fairlie et al., 2005; Tyndall et al., 2005) and that cyclic peptides that mimic turn conformations tend to be potent ligands for GPCR and other protein targets (Fairlie et al., 1995; Tyndall et al., 2005). The conclusions from the study of structure activity relationships (SAR) for potent and

selective hexapeptides are the following (1) the C terminal L4AR6 motif is important for agonist activity. Changing these motifs leads to significant reductions in agonist activity. (2) threonine at the third position gave the most potent compound although other amino acids (serine, glycine) were not much less potent. (3) at the second position, the most potent compounds had tryptophan, norleucine or leucine. (4) at the N-terminal or first position, ligands with phenylalanine or tryptophan were potent ligands. (5) Replacing leucine at the fourth position by a bulkier group leads to the antagonist (FLTChaAR).

2.5 MATERIALS AND METHODS

2.5.1 CELL ISOLATION AND CULTURE OF PBMCs

Human peripheral blood mononuclear cells (PBMC) were isolated from buffy coat (obtained from Australian Red Cross Blood Service, Kelvin Grove) using Ficoll-paque density centrifugation (GE Healthcare Bio-Science, Uppsala, Sweden) and cultured in complete media, consisting of

IMDM with 10% FBS, 10 U/mL penicillin, 10 U/mL streptomycin and 2mM L-glutamine

(Invitrogen) at 37 °C, with 5% CO2. Promonocytic U937 cells (cultured in RPMI media with 10%

FBS, L-Glu, PenStrep and NEAA) were pretreated 72 h prior to assay with membrane-permeable cAMP analogue Bt2-cAMP (0.5 mM), which induces cell differentiation to a monocyte/macrophage- or granulocyte-like phenotype. Cellular differentiation induces increased C3aR and C5aR expression on the cell membrane (Burg et al., 1996; Klos et al., 1992).

108 CHAPTER 2 Short Peptide Agonists for C3aR

2.5.2 RECEPTOR BINDING ASSAY ON PBMC CELLS Receptor binding was performed using [125I]-C3a, 80 pM (2200 Ci/mmol; Perkin Elmer, Torrance, CA, USA), PBMC cells (15x106 cells/mL and in the absence or presence of various concentrations of unlabeled C3a or C3a hexapeptide agonist for 60 min at room temperature with shaking in 50 mM Tris, 3 mM MgCl2, 0.1 mM CaCl2, 0.5% (w/v) Bovine serum albumin, pH 7.4. Unbound radioactivity was removed by filtration through glass microfiber filters GF/B (Whatman Iner. Ltd, England) which had been soaked in 0.6% polyethylenimine to reduce non-specific binding. The filter was washed 3 times with cold buffer (50 mM Tris-HCl) pH 7.4. Bound [125I]- C3a was then assessed by scintillation counting on a β-counter. Specific [125I]-C3a binding is defined as a difference between total binding and non-specific binding as determined in the presence of 1 µM unlabelled C3a. The IC50 value is the concentration of antagonist to inhibit the binding of labelled ligand by 50%. Nonlinear regression analysis (GraphPad Prism 5, USA) was performed on concentration response curves to determine IC50 and pIC50. The pIC50 for each compound was calculated for separate experiments and expressed as an arithmetic mean standard error (SE). IC50 values were expressed as a geometric mean.

2.5.3 INTRACELLULAR CALCIUM RELEASE ASSAY (FOR SUSPENSION CELLS, dU937 CELLS)

Pretreated U937 cells and culture medium were centrifuged (2500 rpm, 5 mins). The supernatant was removed, and the cells were resuspended in 2 mL wash buffer (1X HBSS, 20 mM HEPES, 2.5 mM Probenecid, pH 7.4) and cells were counted on a hemocytometer. One dye-loading buffer (12 mL wash buffer, 1% FBS, 25 µL Fluo-3 (final concentration 4 µM), 25 µL 20% Pluronic acid) was added per 5-7 million cells. Cells were suspended in dye loading buffer, incubated in a covered culture flask for 1 h at 37 °C, then centrifuged (2500 rpm, 5 mins). The supernatant was removed, cells resuspended in 3 mL wash buffer, centrifuged (2500 rpm, 5 mins) and the process repeated for cell density of 2x106 cells/mL (or 100,000 cells per well). 50 µL of test compound or buffer or control/well was plated out on sterile black-wall, clear-bottom 96-well plates (Corning Incorporated, NY, USA). The Plate was then loaded into the FLUOstar instrument (BMG LabTechnologies, Offenburg, Germany), where fluorescence was measured over one minute with excitation at 485 nm and emission at 520 nm of Fluo-3 bound Ca2+ complex, at 28 °C. Differentiated U937 cells were administered in situ 10 sec into the one-minute reading. Agonist responses were expressed as a percentage of calcimycin (A23187) or control compound activity,

109 CHAPTER 2 Short Peptide Agonists for C3aR measured as the maximum change in fluorescence through emission from the Fluo-3 bound calcium complex.

2.5.4 DATA ANALYSIS AND STATISTICS

For both receptor binding and agonist assays, non-linear regression was performed using Prism 5 (GraphPad Software, San Diego, CA). The statistical analysis has been utilized for receptor binding assays and functional assays.

In receptor binding assays, the pIC50 for each peptide was calculated for separate experiments and expressed as an arithmetic mean ± standard error (SE). A one-way ANOVA coupled with Tukey post test was performed on pIC50 values to compare compound affinities. P ≤ 0.05 was considered to indicate a statistically significant difference between receptor affinities. In Functional Assays, the concentration curves were analysed by non-linear regression

(GraphPad Prism 5, USA) and EC50 and pEC50 determined. A pEC50 for each compound was calculated from separate experiments and expressed as an arithmetic mean ± SE. A one-way

ANOVA coupled with a Tukey post test was performed on pEC50 values to examine statistical significance between agonist potencies of C3a analogue peptides. P ≤ 0.05 was considered to indicate a statistically significant difference.

110 CHAPTER 2 Short Peptide Agonists for C3aR

2.6 REFERENCES

1. ALLEGRETTI, M, MORICONI, A, BECCARI, AR, DI BITONDO, R, BIZZARRI, C, BERTINI, R, COLOTTA, F (2005) Targeting C5a: Recent advances in drug discovery. Current Medicinal Chemistry 12(2): 217-236.

2. AMBROSIUS, D, CASARETTO, M, GERARDYSCHAHN, R, SAUNDERS, D, BRANDENBURG, D, ZAHN, H (1989) Peptide analogs of the anaphylatoxin C3a-syntheses and properties Biol. Chem. Hoppe-Seyler 370(3): 217-227.

3. AMES, RS, TORNETTA, MA, FOLEY, JJ, HUGLI, TE, SARAU, HM (1997) Evidence that the receptor for C4a is distinct from the C3a receptor. Immunopharmacology 38(1-2): 87-92.

4. BITTER SUERMANN, D (1988) The complement system. Springer-Verlag: New York.

5. BLAKENEY, JS Ligands for C3a and C5a receptors. PhD. Dissertation, The University of Queensland, Brisbane, 2007.

6. BOKISCH, VA, MULLER-EBERHARD, HJ (1970) Anaphylatoxin inactivator of human plasma: its isolation and characterization as a carboxypeptidas. J. Clin. Invest. 49: 2429-2436.

7. BOOS, L, CAMPBELL, LL, AMES, R, WETSEL, RA, BARNUM, SR (2004) Deletion of the complement anaphylatoxin C3a receptor attenuates, whereas ectopic expression of C3a in the brain exacerbates, experimental autoimmune encephalomyelitis. J. Immunol. 173(7): 4708-4714.

8. BOOS, L, SZALAI, AJ, BARNUM, SR (2005) C3a expressed in the central nervous system protects against LPS-induced shock. Neurosci. Lett. 387(2): 68-71.

9. BUHL, AM, OSAWA, S, JOHNSON, GL (1995) Mitogen-activated protein-kinase activation requires 2 signal inputs from the human anaphylatoxin C5a receptor. Journal of Biological Chemistry 270(34): 19828-19832.

10. BURG, M, MARTIN, U, BOCK, D, RHEINHEIMER, C, KOHL, J, BAUTSCH, W, KLOS, A (1996) Differential regulation of the C3a and C5a receptors (CD88) by IFN-gamma and PMA in U937 cells and related myeloblastic cell lines. J. Immunol. 157(12): 5574-5581.

11. CAMPBELL, WD, LAZOURA, E, OKADA, N, OKADA, H (2002) Inactivation of C3a and C5a octapeptides by carboxypeptidase R and carboxypeptidase N. Microbiol. Immunol. 46(2): 131-134.

12. CAPORALE, LH, TIPPETT, PS, ERICKSON, BW, HUGLI, TE (1980) The active-site of C3a anaphylatoxin Journal of Biological Chemistry 255(22): 758-763.

13. CHAO, TH, EMBER, JA, EMBER, JA, WANG, MY, BAYON, Y, HUGLI, TE, YE, RD (1999) Role of the second extracellular loop of human C3a receptor in agonist binding and receptor function. Journal of Biological Chemistry 274(14): 9721-9728.

111 CHAPTER 2 Short Peptide Agonists for C3aR 14. CHEN, ZG, ZHANG, XL, GONNELLA, NC, PELLAS, TC, BOYAR, WC, NI, F (1998) Residues 21-30 within the extracellular N-terminal region of the C5a receptor represent a binding domain for the C5a anaphylatoxin. Journal of Biological Chemistry 273(17): 10411-10419.

15. CIANFLONE, K, MASLOWSKA, M, SNIDERMAN, AD (1999) Acylation stimulating protein (ASP), an adipocyte autocrine: new directions. Seminars in Cell & Developmental Biology 10(1): 31-41.

16. CIANFLONE, K, XIA, ZN, CHEN, LY (2003) Critical review of acylation-stimulating protein physiology in humans and rodents. Biochim. Biophys. Acta-Biomembr. 1609(2): 127-143.

17. CRASS, T, AMES, RS, SARAU, HM, TORNETTA, MA, FOLEY, JJ, KOHL, J, KLOS, A, BAUTSCH, W (1999) Chimeric receptors of the human C3a receptor and C5a receptor (CD88). Journal of Biological Chemistry 274(13): 8367-8370.

18. CRASS, T, RAFFETSEDER, U, MARTIN, U, GROVE, M, KLOS, A, KOHL, J, BAUTSCH, W (1996) Expression cloning of the human C3a anaphylatoxin receptor (C3aR) from differentiated U- 937 cells. European Journal of Immunology 26(8): 1944-1950.

19. DAFFERN, PJ, PFEIFER, PH, EMBER, JA, HUGLI, TE (1995) C3a is a chemotaxin for human eosinophils but not for neutrophils. I. C3a stimulation of neutrophils is secondary to eosinophil activation. Journal of Experimental Medicine 181(6): 2119-2127.

20. DRAPEAU, G, BROCHU, S, GODIN, D, LEVESQUE, L, RIOUX, F, MARCEAU, F (1993) Synthetic C5a-receptor agonists - pharmacology, methabolism and in vivo cardiovascular and hematologic effects. Biochemical Pharmacology 45(6): 1289-1299.

21. DROUIN, SM, CORRY, DB, HOLLMAN, TJ, KILDSGAARD, J, WETSEL, RA (2002) Absence of the complement anaphylatoxin C3a receptor suppresses Th2 effector functions in a murine model of pulmonary allergy. J. Immunol. 169(10): 5926-5933.

22. DROUIN, SM, CORRY, DB, KILDSGAARD, J, WETSEL, RA (2001) Cutting edge: The absence of C3 demonstrates a role for complement in Th2 effector functions in a murine model of pulmonary allergy. J. Immunol. 167(8): 4141-4145.

23. ELLATI, SG, DAHINDEN, CA, CHURCH, MK (1994) Complement peptides C3a-induced and C5a-induced mediator release from dissociated human skin mast-cells. J. Invest. Dermatol. 102(5): 803-806.

24. EMBER, JA, HUGLI, TE (1997) Complement factors and their receptors. Immunopharmacology 38(1-2): 3-15.

25. EMBER, JA, JAGELS, MA, HUGLI, TE (1998) Characterization of Complement Anaphylatoxins and their Biological Responses. In: The human complement system in health and disease, Volanakis, JE, Frank, MM (eds), pp 241-284. New York: Marcel Dekker.

26. EMBER, JA, JOHANSEN, NL, HUGLI, TE (1991) Designing synthetic superagonists of C3a anaphylatoxin. Biochemistry 30(15): 3603-3612.

27. FAIRLIE, DP, ABBENANTE, G, MARCH, DR (1995) Macrocyclic peptidomimetics- forcing peptides into bioactive conformations. Current Medicinal Chemistry 2(2): 654-686.

112 CHAPTER 2 Short Peptide Agonists for C3aR 28. FAIRLIE, DP, TYNDALL, JDA, ABBENANTE, G, PFEIFFER, B (2005) Over 100 peptide- activated GPCRS recognize turn motifs in ligands. Abstr. Pap. Am. Chem. Soc. 229: 010- MEDI.

29. FEDERWISCH, M, CASARETTO, M, GERARDYSCHAHN, R, BITTERSUERMANN, D, WOLLMER, A (1992) Enhanced biopotency of synthetic C3a-analogs by membrane-binding-a fluorescence anisotropy decay study. Biophysical Chemistry 44(3): 151-161.

30. FOREMAN, KE, GLOVSKY, MM, WARNER, RL, HORVATH, SJ, WARD, PA (1996) Comparative effect of C3a and C5a on adhesion molecule expression on neutrophils and endothelial cells. Inflammation 20(1): 1-9.

31. FUKUOKA, Y, EMBER, JA, HUGLI, TE (1998) Molecular cloning of two isoforms of the guinea pig C3a anaphylatoxin receptor: Alternative splicing in the large extracellular loop. J. Immunol. 161(6): 2977-2984.

32. FUKUOKA, Y, HUGLI, TE (1988) Demonstration of a specific C3a receptor on guinea pig platelets. J. Immunol. 140(10): 3496-3501.

33. GAO, JM, CHOE, H, BOTA, D, WRIGHT, PL, GERARD, C, GERARD, NP (2003) Sulfation of tyrosine 174 in the human C3a receptor is essential for binding of C3a anaphylatoxin. Journal of Biological Chemistry 278(39): 37902-37908.

34. GARRETT, MC, OTTEN, ML, STARKE, RM, KOMOTAR, RJ, MAGOTTI, P, LAMBRIS, JD, RYNKOWSKI, MA, CONNOLLY, ES (2009) Synergistic neuroprotective effects of C3a and C5a receptor blockade following intracerebral hemorrhage. Brain Res. 1298: 171-177.

35. GERARD, NP, GERARD, C (2002) Complement in allergy and asthma. Curr. Opin. Immunol. 14(6): 705-708.

36. GERARDYSCHAHN, R, AMBROSIUS, D, CASARETTO, M, GROTZINGER, J, SAUNDERS, D, WOLLMER, A, BRANDENBURG, D, BITTERSUERMANN, D (1988) Design and biological- activity of a new generation of synthetic C3a analogs by combination of peptidic and non- peptidic elements. Biochem. J. 255(1): 209-216.

37. HAAS, PJ, VAN STRIJP, J (2007) Anaphylatoxins - Their role in bacterial infection and inflammation. Immunol. Res. 37(3): 161-175.

38. HAWLISCH, H, WILLS-KARP, M, KARP, CL, KOHL, J (2004) The anaphylatoxins bridge innate and adaptive immune responses in allergic asthma. Mol. Immunol. 41(2-3): 123-131.

39. HOEPRICH, PD, HUGLI, TE (1986) Helical conformation at the carboxy-terminal portion of human C3a is required for full activity. Biochemistry 25(8): 1945-1950.

40. HUEY, R, ERICKSON, BW, BLOOR, CM, HUGLI, TE (1984) Contraction of guinea pig lung by synthetic oligopeptides related to human C3a. Immunopharmacology 8(1): 37-45.

41. HUGLI, TE (1986) Biochemistry and biology of anaphylatoxins. In: Complement Vol. 3, pp 111-127.

113 CHAPTER 2 Short Peptide Agonists for C3aR 42. HUGLI, TE (1981) The structural basis for anaphylatoxin and chemotactic functions of complement C-3a complement C-4a and complement C-5a. Critical Reviews in Immunology 1(4): 321-366.

43. HUGLI, TE (1984) Structure and function of the anaphylatoxins. Springer Semin. Immunopathol. 7: 193.

44. HUGLI, TE, MULLER-EBERHARD, HJ (1978) Anaphylatoxins: C3a and C5a. Adv Immunol 26: 1-53.

45. HUMBLES, AA, LU, B, NILSSON, CA, LILLY, C, ISRAEL, E, FUJIWARA, Y, GERARD, NP, GERARD, C (2000) A role for the C3a anaphylatoxin receptor in the effector phase of asthma. Nature 406(6799): 998-1001.

46. HUTAMEKALIN, P, TAKEDA, K, TANI, M, TSUGA, Y, OGAWA, N, MIZUTANI, N, YOSHINO, S (2010) Effect of the C3a-Receptor Antagonist SB 290157 on Anti-OVA Polyclonal Antibody-Induced Arthritis. J. Pharmacol. Sci. 112(1): 56-63.

47. ISCHENKO, A, SAYAH, S, PATTE, C, ANDREEV, S, GASQUE, P, SCHOUFT, MT, VAUDRY, H, FONTAINE, M (1998) Expression of a functional anaphylatoxin C3a receptor by astrocytes. J. Neurochem. 71(6): 2487-2496.

48. JINSMAA, Y, TAKENAKA, Y, YOSHIKAWA, M (2001) Designing of an orally active complement C3a agonist peptide with anti-analgesic and anti-amnesic activity. Peptides 22(1): 25-32.

49. KALANT, D, MACLAREN, R, CUI, W, SAMANTA, R, MONK, PN, LAPORTE, SA, CIANFLONE, K (2005) C5L2 is a functional receptor for acylation-stimulating protein. Journal of Biological Chemistry 280(25): 23936-23944.

50. KAWAI, M, QUINCY, DA, LANE, B, MOLLISON, KW, LULY, JR, CARTER, GW (1991) Identification and synthesis of a receptor-binding site of human anaphylatoxin C5a. Journal of Medicinal Chemistry 34(7): 2068-2071.

51. KAWAMOTO, S, YALCINDAG, A, LAOUINI, D, BRODEUR, S, BRYCE, P, LU, B, HUMBLES, AA, OETTGEN, H, GERARD, C, GEHA, RS (2004) The anaphylatoxin C3a downregulates the Th2 response to epicutaneously introduced antigen. Journal of Clinical Investigation 114(3): 399-407.

52. KAWATSU, R, SANDERSON, SD, BLANCO, I, KENDALL, N, FINCH, AM, TAYLOR, SM, COLCHER, D (1996) Conformationally biased analogs of human C5a mediate changes in vascular permeability. Journal of Pharmacology and Experimental Therapeutics 278(1): 432-440.

53. KILDSGAARD, J, HOLLMANN, TJ, MATTHEWS, KW, BIAN, K, MURAD, F, WETSEL, RA (2000) Cutting edge: Targeted disruption of the C3a receptor gene demonstrates a novel protective anti-inflammatory role for C3a in endotoxin-shock. J. Immunol. 165(10): 5406-5409.

54. KIRSCHFINK, M (2001) Targeting complement in therapy. Immunol. Rev. 180: 177-189.

55. KLOS, A, BANK, S, GIETZ, C, BAUTSCH, W, KOHL, J, BURG, M, KRETZSCHMAR, T (1992) C3a receptor on Dibutyryl-cAMP-differentiated U937 cells and human neutrophils - the 114 CHAPTER 2 Short Peptide Agonists for C3aR human C3a receptor characterized by funcitonal responses and I-125 C3a binding. Biochemistry 31(46): 11274-11282.

56. KLOS, A, TENNER, AJ, JOHSWICH, KO, AGER, RR, REIS, ES, KOHL, J (2009) The role of the anaphylatoxins in health and disease. Mol. Immunol. 46(14): 2753-2766.

57. KOHL, J, CASARETTO, M, GIER, M, KARWATH, G, GIETZ, C, BAUTSCH, W, SAUNDERS, D, BITTERSUERMANN, D (1990) Reevaluation of the C3a active-site using short synthetic C3a analogs. European Journal of Immunology 20(7): 1463-1468.

58. KOLA, A, KLOS, A, BAUTSCH, W, KRETZSCHMAR, T, KOHL, J (1992) Functional activities of synthetic anaphylatoxic peptides in widely used biological assays. Clin. Exp. Immunol. 88(2): 368-372.

59. KONTEATIS, ZD, SICILIANO, SJ, VANRIPER, G, MOLINEAUX, CJ, PANDYA, S, FISCHER, P, ROSEN, H, MUMFORD, RA, SPRINGER, MS (1994) Development of C5a receptor antagonists- differential loss of functional responses. J. Immunol. 153(9): 4200-4205.

60. KRETZSCHMAR, T, JEROMIN, A, GIETZ, C, BAUTSCH, W, KLOS, A, KOHL, J, RECHKEMMER, G, BITTERSUERMANN, D (1993) Chronic myelogenous leukemia-derived basophilic granulocytes express a functional active receptor for anaphylatoxin C3a. European Journal of Immunology 23(2): 558-561.

61. KRETZSCHMAR, T, POHL, M, CASARETTO, M, PRZEWOSNY, M, BAUTSCH, W, KLOS, A, SAUNDERS, D, KOHL, J (1992) Synthetic peptides as antagonists of the anaphylatoxin C3a. Eur. J. Biochem. 210(1): 185-191.

62. LANGKABEL, P, ZWIRNER, J, OPPERMANN, M (1999) Ligand-induced phosphorylation of anaphylatoxin receptors C3aR and C5aR is mediated by G protein-coupled receptor kinases. European Journal of Immunology 29(9): 3035-3046.

63. LIENENKLAUS, S, AMES, RS, TORNETTA, MA, SARAU, HM, FOLEY, JJ, CRASS, T, SOHNS, B, RAFFETSEDER, U, GROVE, M, HO?LZER, A, KLOS, A, KO?HL, J, BAUTSCH, W (1998) Cutting edge: Human anaphylatoxin C4a is a potent agonist of the guinea pig but not the human C3a receptor. J. Immunol. 161(5): 2089-2093.

64. LU, Z-X, FOK, K, ERICKSON, B, HUGLI, TE (1984) Conformation analysis of COOH- terminal segments of human C3a. Evidence of ordered conformation in an active 21-residue peptide. Journal of Biological Chemistry 259: 7367-7370.

65. MALMSTEN, M, SCHTMDTCHEN, A (2007) Antimicrobial C3a - Biology, biophysics, and evolution. Current Topics in Innate Immunity 598: 141-158.

66. MAMANE, Y, CHAN, CC, LAVALLEE, G, MORIN, N, XU, LJ, HUANG, JQ, GORDON, R, THOMAS, W, LAMB, J, SCHADT, EE, KENNEDY, BP, MANCINI, JA (2009) The C3a Anaphylatoxin Receptor Is a Key Mediator of Insulin Resistance and Functions by Modulating Adipose Tissue Macrophage Infiltration and Activation. Diabetes 58(9): 2006- 2017.

67. MASTERS, SL, SIMON, A, AKSENTIJEVICH, I, KASTNER, DL (2009) Horror Autoinflammaticus: The Molecular Pathophysiology of Autoinflammatory Disease. Annu. Rev. Immunol. 27: 621-668. 115 CHAPTER 2 Short Peptide Agonists for C3aR

68. MIZUTANI, N, NABE, T, YOSHINO, S (2009) Complement C3a Regulates Late Asthmatic Response and Airway Hyperresponsiveness in Mice. J. Immunol. 183(6): 4039-4046.

69. MOCCO, J, MACK, WJ, DUCRUET, AF, SOSUNOV, SA, SUGHRUE, ME, HASSID, BG, NAIR, MN, LAUFER, I, KOMOTAR, RJ, HOLLAND, MCH, PINSKY, DJ, CONNOLLY, ES (2006) Complement component C3 mediates inflammatory injury following focal cerebral ischemia. Circ.Res. 99(2): 209-217.

70. MONK, PN, SCOLA, AM, MADALA, P, FAIRLIE, DP (2007) Function, structure and therapeutic potential of complement C5a receptors. British Journal of Pharmacology 152(4): 429-448.

71. MOUSLI, M, HUGLI, TE, LANDRY, Y, BRONNER, C (1992) A Mechanism of action for anaphylatoxin C3a stimulation of mast cells. J. Immunol. 148(8): 2456-2461.

72. MUELLER-ORTIZ, SL, HOLLMANN, TJ, HAVILAND, DL, WETSEL, RA (2006) Ablation of the complement C3a anaphylatoxin receptor causes enhanced killing of Pseudomonas aeruginosa in a mouse model of pneumonia. Am. J. Physiol.-Lung Cell. Mol. Physiol. 291(2): L157-L165.

73. MURRAY, I, KOHL, J, CIANFLONE, K (1999) Acylation-stimulating protein (ASP): structure- function determinants of cell surface binding and triacylglycerol synthetic activity. Biochem. J. 342: 41-48.

74. NILSSON, G, JOHNELL, M, HAMMER, CH, TIFFANY, HL, NILSSON, K, METCALFE, DD, SIEGBAHN, A, MURPHY, PM (1996) C3a and C5a are chemotaxins for human mast cells and act through distinct receptors via a pertussis toxin-sensitive signal transduction pathway. J. Immunol. 157(4): 1693-1698.

75. NORDAHL, EA, RYDENGARD, V, NYBERG, P, NITSCHE, DP, MORGELIN, M, MALMSTEN, M, BJORCK, L, SCHMIDTCHEN, A (2004) Activation of the complement system generates antibacterial peptides. Proc. Natl. Acad. Sci. U. S. A. 101(48): 16879-16884.

76. NORGAUER, J, DOBOS, G, KOWNATZKI, E, DAHINDEN, C, BURGER, R, KUPPER, R, GIERSCHIK, P (1993) Complement fragment C3a stimulates Ca2+ influx in neutrophils via a pertussis-toxin-sensitive G-protein. Eur. J. Biochem. 217(1): 289-294.

77. PASUPULETI, M, WALSE, B, SVENSSON, B, MALMSTEN, M, SCHMIDTCHEN, A (2008) Rational design of antimicrobial C3a analogues with enhanced effects against staphylococci using an integrated structure and function-based approach. Biochemistry 47(35): 9057-9070.

78. PETERING, H, KO?HL, J, WEYERGRAF, A, DULKYS, Y, KIMMIG, D, SMOLARSKI, R, KAPP, A, ELSNER, J (2000) Characterization of synthetic C3a analog peptides on human eosinophils in comparison to the native complement component C3a. J. Immunol. 164(7): 3783-3789.

79. POHL, M, AMBROSIUS, D, GROTZINGER, J, KRETZSCHMAR, T, SAUNDERS, D, WOLLMER, A, BRANDENBURG, D, BITTERSUERMANN, D, HOCKER, H (1993) Cyclic disulfide anaplogs of the complement component C3a-synthsis and conformational investigations. International Journal of Peptide and Protein Research 41(4): 362-375.

80. PROCTOR, L. PhD Dissertation, University of Queensland, Brisbane, 2004. 116 CHAPTER 2 Short Peptide Agonists for C3aR

81. PROCTOR, LM, WOODRUFF, TM, PAZCKOWSKI, NP, MONK, PN, TAYLOR, SM (2004) Dual C3a/C5a receptor agonism by a C5a C-terminal analogue decapeptide. Mol. Immunol. 41(2- 3): 210.

82. RAHPEYMAI, Y, HIETALA, MA, WILHELMSSON, U, FOTHERINGHAM, A, DAVIES, I, NILSSON, AK, ZWIRNER, J, WETSEL, RA, GERARD, C, PEKNY, M, PEKNA, M (2006) Complement: a novel factor in basal and ischemia-induced neurogenesis. Embo J. 25(6): 1364-1374.

83. RYNKOWSKI, MA, KIM, GH, GARRETT, MC, ZACHARIA, BE, OTTEN, ML, SOSUNOV, SA, KOMOTAR, RJ, HASSID, BG, DUCRUET, AF, LAMBRIS, JD, CONNOLLY, ES (2009) C3a receptor antagonist attenuates brain injury after intracerebral hemorrhage. J. Cereb. Blood Flow Metab. 29(1): 98-107.

84. SAYAH, S, JAUNEAU, AC, PATTE, C, TONON, MC, VAUDRY, H, FONTAINE, M (2003) Two different transduction pathways are activated by C3a and C5a anaphylatoxins on astrocytes. Mol. Brain Res. 112(1-2): 53-60.

85. SCHRAUFSTATTER, IU, TRIEU, K, SIKORA, L, SRIRAMARAO, P, DISCIPIO, R (2002) Complement C3a and C5a induce different signal transduction cascades in endothelial cells. J. Immunol. 169(4): 2102-2110.

86. SCULLY, CCG, BLAKENEY, JS, SINGH, R, HOANG, HN, ABBENANTE, G, REID, RC, FAIRLIE, DP (2010) Selective Hexapeptide Agonists and Antagonists for Human Complement C3a Receptor. Journal of Medicinal Chemistry 53(13): 4938-4948.

87. SICILIANO, SJ, ROLLINS, TE, DEMARTINO, J, KONTEATIS, Z, MALKOWITZ, L, VANRIPER, G, BONDY, S, ROSEN, H, SPRINGER, MS (1994) 2-Site binding of C5a by its receptor- an alternative binding paradigm for G-protein coupled receptors. Proc. Natl. Acad. Sci. U. S. A. 91(4): 1214-1218.

88. SONESSON, A, RINGSTAD, L, NORDAHL, EA, MALMSTEN, M, MORGELIN, M, SCHMIDTCHEN, A (2007) Antifungal activity of C3a and C3a-derived peptides against Candida. Biochim. Biophys. Acta-Biomembr. 1768(2): 346-353.

89. SUN, JZ, EMBER, JA, CHAO, TH, FUKUOKA, Y, YE, RD, HUGLI, TE (1999) Identification of ligand effector binding sites in transmembrane regions of the human G protein-coupled C3a receptor. Protein Science 8(11): 2304-2311.

90. TAKABAYASHI, T, VANNIER, E, CLARK, BD, MARGOLIS, NH, DINARELLO, CA, BURKE, JF, GELFAND, JA (1996) A new biologic role for C3a and C3a desArg - Regulation of TNF- alpha and IL-1 beta synthesis. J. Immunol. 156(9): 3455-3460.

91. TANG, ZY, LU, B, HATCH, E, SACKS, SH, SHEERIN, NS (2009) C3a Mediates Epithelial-to- Mesenchymal Transition in Proteinuric Nephropathy. J. Am. Soc. Nephrol. 20(3): 593-603.

92. TAYLOR, SM, SHERMAN, SA, KIRNARSKY, L, SANDERSON, SD (2001) Development of response-selective agonists of human C5a anaphylatoxin: Conformational, biological, and therapeutic considerations. Current Medicinal Chemistry 8(6): 675-684.

117 CHAPTER 2 Short Peptide Agonists for C3aR 93. TYNDALL, JDA, PFEIFFER, B, ABBENANTE, G, FAIRLIE, DP (2005) Over one hundred peptide-activated G protein-coupled receptors recognize ligands with turn structure. Chem. Rev. 105(3): 793-826.

94. UNSON, CG, ERICKSON, BW, HUGLI, TE (1984) Active-site of C3a anaphylatoxin- contributions of the lipophilic and orienting residues. Biochemistry 23(4): 585-589.

95. VANEK, M, HAWKINS, LD, GUSOVSKY, F (1994) Coupling of the C5a receptor to G(i) in U- 937 cells and in cells transfected with C5a receptor cDNA. Molecular Pharmacology 46(5): 832-839.

96. VOGEN, SM, PACZKOWSKI, NJ, KIRNARSKY, L, SHORT, A, WHITMORE, JB, SHERMAN, SA, TAYLOR, SM, SANDERSON, SD (2001) Differential activities of decapeptide agonists of human C5a: the conformational effects of backbone N-methylation. International Immunopharmacology 1(12): 2151-2162.

97. WENDERFER, SE, WANG, HY, KE, BZ, WETSEL, RA, BRAUN, MC (2009) C3a receptor deficiency accelerates the onset of renal injury in the MRL/Ipr mouse. Mol. Immunol. 46(7): 1397-1404.

98. XIA, ZN, STANHOPE, KL, DIGITALE, E, SIMION, OM, CHEN, LY, HAVEL, P, CIANFLONE, K (2004) Acylation-stimulating protein (ASP)/complement C3adesArg deficiency results in increased energy expenditure in mice. Journal of Biological Chemistry 279(6): 4051-4057.

99. ZIPFEL, PF, SKERKA, C (2009) Complement regulators and inhibitory proteins. Nature Reviews Immunology 9(10): 729-740.

100. ZWIRNER, J, GOTZE, O, MOSER, A, SIEBER, A, BEGEMANN, G, KAPP, A, ELSNER, J, WERFEL, T (1997) Blood- and skin-derived monocytes/macrophages respond to C3a but not to C3a(desArg) with a transient release of calcium via a pertussis toxin-sensitive signal transduction pathway. European Journal of Immunology 27(9): 2317-2322.

118 CHAPTER 2 Short Peptide Agonists for C3aR

119 CHAPTER 3 Non-peptide agonists for C3a receptor

CHAPTER 3 NON-PEPTIDE AGONISTS FOR C3a RECEPTOR

3.0 ABSTRACT...... 121 3.1 INTRODUCTION...... 122 3.2 AIMS/OBJECTIVES...... 126 3.3 RESULTS ...... 126

3.3.1 SERIES I: N-ACYL AMINO ACID OXAZOLE ARGININE DERIVATIVES...... 127 3.3.1.1 Affinity of Non-Peptide Agonists for C3aR on HMDM...... 132 3.3.1.2 Affinities of C3a Non-Peptide Ligands ...... 133

3.3.2 SERIES II: INCORPORATION OF A METHYL GROUP ON THE 5-POSITION ON THE

OXAZOLE RING ...... 134

3.3.3 SELECTIVITY OF C3a NON-PEPTIDE LIGANDS FOR C5aR...... 139 3.3.3.1 Competitive Radio-ligand Binding Assays: [125I]-C5a Binding...... 139 3.3.3.2 The Affinity to C5aR of Non-Peptide C3a Ligands: Studies in Concentration Dependent Manner...... 140

3.3.4 SELECTIVITY TEST: DESENSITISATION EXPERIMENTS...... 140

3.3.5 EFFECT OF C3a NON-PEPTIDE AGONISTS/ANTAGONISTS ON CALCIUM RELEASE

FROM HUMAN MONOCYTE DERIVED MACROPHAGES...... 143 3.3.5.1 C3aR Non-Peptide Agonists ...... 144 3.4 DISCUSSION AND CONCLUSIONS ...... 145 3.5 MATERIALS AND METHODS ...... 148

3.5.1 MACROPHAGE CELL CULTURE AND DIFFERENTIATION...... 148

3.5.2 CALCIUM MOBILIZATION ASSAY ON HMDM (ADHESION CELLS)...... 148

3.5.3 RECEPTOR BINDING ASSAY OF HMDM...... 148

3.5.4 DATA ANALYSIS AND STATISTICS ...... 149 3.6 REFERENCES...... 150

120 CHAPTER 3 Non-peptide agonists for C3a receptor

3.0 ABSTRACT

In Chapter 2, peptides such as agonist FLTLAR and antagonist FLTChaAR were developed using competitive radio-ligand binding and structure-activity relationships to identify potent peptide agonists of human C3a receptor. However, C3a-derived peptides have significant disadvantages for use in biology and medicine in their poor bioavailability, poor stability to protease enzymes, high clearance rates, and poor membrane permeability. Therefore, it is highly desirable to develop nonpeptidic C3a ligands that are more suitable for use as pharmacological/physiological probes for interrogating the roles of hC3a in vivo. As a step towards this goal, Chapter 3 now summarizes the introduction of a heterocycle to restrict the conformational freedom of short peptides, in an attempt to mimic the receptor-binding conformation of the C-terminal activating domain of human C3a. A molecular modelling study (Figure 3.2) revealed structural similarity between the C3a carboxyl terminal residues (GLAR), modelled in a beta turn conformation, and the model of compound 56. This compound contains an oxazole heterocycle that acts as a beta turn-inducing constraint and mimics the conformation found for the C-terminus of C3a in its crystal structure. This result supports a large set of observations that GPCRs have a general tendency to recognise turn structural motifs in their protein and peptide ligands (Fairlie et al., 2005; Madala et al., 2010; Tyndall et al., 2005). In this study, multiple potent agonist compounds were discovered for C3aR. They competed strongly with 125I-labelled C3a in binding to, and activating intracellular calcium release in, human monocyte-derived macrophages (HMDM). Indeed, some of the compounds had equal or even greater affinity than C3a itself. Building on the discovery in Chapter 2 that hexapeptide agonists for C3aR agonist (e.g. FLTLAR) adopted a beta turn conformation at their C-terminus, it was decided to introduce an oxazole heterocycle into the LAR region to impart conformational rigidity to the agonist peptides (Figure 3.3). This modification produced new compounds with improved affinities for human C3aR. The oxazole has electronegative oxygen and nitrogen atoms that can function as hydrogen bond acceptors, while introducing a rigid turn-like conformation to the molecule. The most potent and selective compounds in this study were 56, 63 and 75 (EC50 = 19, 7 and 17 nM, respectively) in HMDM cells. There was a linear correlation between the binding affinity of the compounds and their agonist potency.

121 CHAPTER 3 Non-peptide agonists for C3a receptor

3.1 INTRODUCTION

C3a is a potent inflammatory mediator, which targets a wide range of immune and non- immune cells. Release of C3a regulates vasodilation, increases vascular permeability and induces smooth muscle contraction (Ember et al., 1998). C3a can also trigger oxidative burst in macrophages (Murakami et al., 1993), neutrophils (Elsner et al., 1994b), and eosinophils (Elsner et al., 1994a). Stimulation of C3aR in basophils (Kretzschmar et al., 1993) and mast cells (Ellati et al., 1994) causes the release of histamine. C3a also promotes the release of serotonin from guinea pig platelets (Fukuoka et al., 1988), modulates synthesis of IL-6 and TNF-α by B-lymphocytes and monocytes (Fischer et al., 1997; Fischer et al., 1999), and also stimulates the release of IL-1 from monocytes (Haeffnercavaillon et al., 1987). The functional receptor for C3a is C3aR, which binds C3a with Kd ~1 nM, but does not bind its desArg form or C5a (Crass et al., 1996; Wilken et al., 1999). C3aR is a 54 kDa protein and is comprised of 482 amino acids. The binding of C3a to C3aR causes an increase in cytosolic calcium levels through activation of pertussis-toxin-sensitive G- proteins (Norgauer et al., 1993; Zwirner et al., 1997). However, in endothelial cells, the signal transduction is through pertussis-toxin insensitive Gα12 and Gα13 (Schraufstatter et al., 2002). C3aR is also expressed on myeloid cell such as neutrophils, eosinophils, basophils, mast cells and monocytes/macrophages, dendritic cells (DC) and microglia (Ames et al., 1996; Crass et al., 1996; Daffern et al., 1995; Gutzmer et al., 2004; Hugli, 1986; Klos et al., 1992; Muller-Eberhard, 1988). C3aR also exists on non-myeloid cells such as astrocytes (Gasque et al., 1998), epithelial cells (Monsinjon et al., 2001), and endothelial cells (Monsinjon et al., 2003).

C3a has been implicated in the pathogenesis and progression of numerous inflammatory conditions like asthma, allergies, arthritis, psoriasis, ischemia-reperfusion injury, sepsis, systemic lupus erythematosus, diabetes, nephropathy and others (Boos et al., 2004; Boos et al., 2005; Drouin et al., 2002; Drouin et al., 2001; Garrett et al., 2009; Humbles et al., 2000; Hutamekalin et al., 2010; Kawamoto et al., 2004; Kildsgaard et al., 2000; Kirschfink, 2001; Mamane et al., 2009; Mizutani et al., 2009; Mocco et al., 2006; Mueller-Ortiz et al., 2006; Rahpeymai et al., 2006; Rynkowski et al., 2009; Tang et al., 2009; Wenderfer et al., 2009). In OVA-induced models of asthma in C3aR-/- mice (Drouin et al., 2002) and C3aR-/- guinea pigs (Bautsch et al., 2000), deletion of C3aR resulted in attenuation of asthma, as indicated by airway hyper-responsiveness, infiltration of eosinophils and decrease in levels of IL-5, IL-3 and IgE. Most studies indicate that C3a has a greater role in asthma progression than C5a. Therefore, the C3aR may be an important drug target for the development of anti-asthma drugs.

122 CHAPTER 3 Non-peptide agonists for C3a receptor Human C3a is a highly cationic protein, which is comprised of 77 amino acids (human and porcine) and 78 (rat and mouse) with a molecular weight ca. 9000 (Hugli, 1975). The primary structure of C3a was reported in 1975 (Hugli, 1975). It was revealed to be highly cationic at the N- terminus (amino acids 7-21) and C-terminus (amino acid 64-77)(Hugli et al., 1975b). There are several studies of the C3a structure with some conflicting results, which may be due to the utilisation of different techniques; for example X-ray crystallography (Figure 3.1) (Huber et al., 1980; Paques et al., 1980), 1H NMR studies in aqueous solution from C3a desArg (Muto et al., 1985), and circular dichroism of native C3a (Hugli, 1975; Hugli et al., 1975a; Lu et al., 1984). Despite differences in the techniques, there is a general consensus for the C3a structure in that it has a central globular helical core (residues 15-57), stabilised by three disulfide bonds with the C- terminus extending from the helical core. Initial studies by CD suggested that the C3a molecule had regular helical structure with little beta-structure (Hugli et al., 1975a) and this was confirmed by crystallographic analysis; hC3a having 56% alpha-helicity (Huber et al., 1980). Further study by 2D NMR analysis of C3a by Nettesheim and colleagues (1988) identified the presence of three antiparallel helical regions in C3a at residues 17-28, 36-43 and 47-66. A transitional helix, from 67- 70, has also been reported (Chazin et al., 1988; Nettesheim et al., 1988). The N-terminus of C3a (1- 15) exhibits a high degree of flexibility. The C-terminus or effector region of C3a (69-77) has no well defined structure, which suggests either it remains flexible or folds back in a helical portion with a pseudo-beta turn conformation (Chazin et al., 1988). It has been postulated that the turn structure of the C-terminus of C3a may be important for the biological activity of C3a (Hoeprich et al., 1986). Circular dichroism techniques were used to study three truncated C-terminal sequences of native human C3a: C3a(57-77), C3a(65-77) and C3a(73-77) and it was indicated that these peptides have random coil conformations in aqueous solution(Lu et al., 1984). The 21-residue C-terminus (C3a 57-77) of C3a has been synthesised and shown by two groups (Huey et al., 1984; Lu et al., 1984) that this peptide is nearly equivalent in potency as the 77 amino acids of native human C3a. CD data for C3a(57-77) indicated that the 21-residue peptide adopts an alpha-helical conformation (~ 41% of content) after addition of 25% (v/v) trifluoroethanol (TFE). Furthermore, a 21-residue analogue of native C3a, incorporating helix-promoting residues 2-aminobutyric acid and 2- aminoisobutyric acid, showed a 250-fold increase in biological activity over both native C3a and C3a (57-77) (Hoeprich et al., 1986). In contrast, the incorporation of proline residues, which can disrupt helices, was shown to significantly reduce activity for C3a (57-77)(Hoeprich et al., 1986). This indicated a relationship between a preference for a turn conformation and the biological activity of the C-terminal sequence of C3a and derived short peptides.

123 CHAPTER 3 Non-peptide agonists for C3a receptor The NMR and crystallographic studies of C3a were unable to assign the secondary structure of the C-terminal pentapeptide (C3a 73-77), which was shown to be highly flexible in aqueous media (Hoeprich et al., 1986). It has been suggested (Hoeprich et al., 1986; Lu et al., 1984) that the C-terminal pentapeptide may form a helical conformation under hydrophobic conditions such as in the membrane environment prior to binding and subsequent activation of C3aR. Parallel to these physicochemical approaches to characterise the C3a conformation, there are also many studies reported for biological activity of C3a and synthetic C3a peptide analogues. The synthetic C3a analogues and substituted peptides have been used to determine the important amino acids and structural requirements for C3a biological activity. LGLAR (C3a 73-77) was found to be the minimum sequence of C3a in the C3a mediated C3a-C3aR interaction to have anaphylatoxin-like activity (Caporale et al., 1980). LGLAR is highly conserved in many species e.g. human, bovine, rat, mouse and guinea pig (Corbin et al., 1976; Hugli, 1975; Hugli et al., 1975b). The activity of LGLAR is lower than the native C3a by approximately 1000 fold (Hugli et al., 1977). There are many studies of synthetic peptides that have shown that the activity is dependent on the peptide length. The C-terminal arginine was found to be essential for biological activity. The cleavage of the arginine residue by carboxypeptidease N (serum inactivator of the anaphylatoxin peptides)(Bokisch et al., 1970) abrogates the biological acitivities of C3a but not C5a. However, Hugli (1975) reported on the basis of a CD study that C3a and C3a desArg retain the same structure and this was also consistent with the results of an NMR study (Muto et al., 1985). Furthermore, there was a study showing the relationship between the tertiary structure of C3a and its biological activity by using high temperature or denaturing chemicals which causes extensive and irreversible unfolding of structure and results in loss of biological activity of C3a (Hugli et al., 1978). A shorter C3a peptide analogue was found with more potency in biological assays. This enhanced activity was achieved by incorporating a hydrophobic fluorenyl-methoxycarbonyl (Fmoc) or 2-nitro-4-azidophenyl (Nap) or 6-aminohexanoyl (Ahx) group at the N-terminus of 5 to 13 residues of C-terminal C3a analogue peptides (Ember et al., 1991; Gerardyschahn et al., 1988; Kohl et al., 1990) and even the C3a analogue peptide can be further shortened to the minimal sequence LAR with addition of hydrophobicity linked with a spacer (eg. Fmoc and Ahx)(Kohl et al., 1990). A two to ten fold increase in the biological potency was observed in a guinea pig platelet ATP- release assay for the various 5 to 13 residue analogues of C3a which incorporated either Fmoc or Nap groups. It was hypothesised that the hydrophobic group was inserted into the membrane and helped to increase the effective concentration of the peptide for receptor activation (Ember et al., 1991).

124 CHAPTER 3 Non-peptide agonists for C3a receptor A C3a “superagonist” was reported by Ember and colleagues (1991) in which the natural amino acid tryptophan was used as an N-terminal hydrophobic residue (analogous to Fmoc and Map C3a analogues) and greater potency than C3a was reported. This 15-residue peptide (WWGKKYRASKLGLAR) was approximately 12-15 times more potent than C3a in a guinea pig platelet aggregation, vascular permeability and ileum contraction assay (Ember et al., 1991). However, there are many other studies suggesting that the C3a superagonist has lower affinity and less potency than hC3a in a calcium mobilisation assay on human C3aR transfected cells (Ames et al., 1996; Ames et al., 1997; Chao et al., 1999), human neutrophils (Ames et al., 1997), rat astrocytes (Ischenko et al., 1998) and guinea pig peritoneal macrophage cells (Takahashi et al., 1996).

Figure 3.1 The 3D Crystal structure of C3a (77 amino acid residue). It has a central globular helical core and is stabilised by three disulfide linkages (Cys: 22-49, 23-56, 36-57) with the C- terminus extending from the helical core and adopting a dynamic random coil conformation (Huber et al., 1980).

125 CHAPTER 3 Non-peptide agonists for C3a receptor

3.2 AIMS/OBJECTIVES

The objective was to investigate the activity of new ligands designed from the most potent and selective hexapeptide C3a agonist (FLTLAR) from Chapter 2, since that peptide showed a significant preference for a beta turn conformation in DMSO. The aim of this study was to explore new compounds containing an oxazole ring that can potentially impart and stabilise a turn structural motif. By replacing the amino acids and the R group at the N-terminus of this scaffold, it was predicted that enhanced ligand affinity for C3aR might be obtained. Because the arginine residue at the C-terminus of the native agonist C3a is important for the biology of C3a, this residue was not changed. Structure-activity relationships (SAR) of these compounds were obtained through radioligand binding assays using [125I]-C3a and [125I]-C5a and an intracellular Ca2+ release assay for monitoring agonist/antagonist function.

3.3 RESULTS

The previous chapter reported the most potent and selective C3aR hexapeptide agonist known, FLTLAR. This had high affinity for human C3aR (IC50 42 nM vs 80 pM of iodine labelled human C3a) in a competitive binding assay on PBMCs and had full agonist activity (EC50 320 nM) in an intracellular calcium release assay on dU937 cells. The 1H NMR spectrum of FLTLAR in the aprotic solvent DMSO-d6 showed a beta-turn like conformation. The C-terminus of human C3a also showed a turn conformation in its crystal structure (Huber et al., 1980). By using molecular modelling of the last four C-terminal residues of native C3a (-GLAR) in a beta turn conformation and superimposing it upon a model of an oxazole-containing compound (56) which was synthesised by a chemist in the Fairlie group, we found a high degree of similarity in their turn conformations (Figure 3.2). This suggested that an oxazole might be useful for generating active compounds that bind tightly to human C3aR and give greater potency and selectivity for C3aR than the peptides in Chapter 2.

126 CHAPTER 3 Non-peptide agonists for C3a receptor

Figure 3.2 Superimposition of an oxazole scaffold compound (56, cyan) and tetrapeptide C- terminal residues of hC3a (-GLAR, green) in a type II beta turn conformation. The oxygen atoms are represented in red and nitrogens are shown in blue (Courtesy of Dr. Robert Reid).

3.3.1 Series I: N-Acyl Amino Acid Oxazole Arginine Derivatives.

Denonne et al. (2007) have reported an analogue of SB290157 containing a furan ring as linker between the arginine residue and the ether moiety and showed higher affinity for C3aR than SB290157. This compound (144, in Chapter 4 Table 4.11) was prepared and tested by us but was found to bind with lower affinity than SB290157 and also displayed partial agonist effects. In our binding affinity study we used a competitive binding assay against 80 pM [125I]-C3a on HMDM cells whilst Denonne et al.(2007) used a different concentration of [125I]-C3a on HMC-1 cell membranes. We rationalised that the furan oxygen and the oxygen atom of the ether group may both function as hydrogen bond acceptors with a H-bond donor from the receptor, but that the furan was less effective. Therefore replacement of the furan by better H-bond acceptors might afford greater affinity, as the nitrogen is a much better hydrogen bond acceptor. There are other advantages from incorporating oxazole in the linker region, as the ring can introduce rigidity to the molecule and it also potentially introduces a turn conformation. It is known that over one hundred peptide-activated G-protein coupled receptors have a tendency to recognise protein and peptide ligands with turn-like conformations (Fairlie et al., 2005; Madala et al., 2010; Tyndall et al., 2005). Together these features could generate potent C3a agonists and antagonists.

127 CHAPTER 3 Non-peptide agonists for C3a receptor

A library of compounds was synthesised by the chemists in our group and they were assayed by me to determine the binding affinity for C3aR and functionality in the calcium mobilisation assay. Structure activity relationship (SAR) information was then used to identify molecular features that contribute to potency and agonist or antagonist function. It quickly became apparent from the screening at single point concentrations that the presence of the oxazole moiety afforded compounds with high affinity. Initial optimization of the generic structure drawn in Figure 3.3 was performed on the substituent denoted as AA by systematic substitution of the amino acid linked between the oxazole moiety and the acyl group, leaving the R substituent as a t-butoxy group (compounds 56, 57, 58, 59, and 60). The results of the binding affinity study in a concentration-dependent manner for these compounds showed that either leucine (56) or isoleucine (57) was the optimal amino acid in the linker region of the molecule. In contrast, the substitution of phenylalanine (58) or tryptophan (59) in the linker dramatically reduced the affinity of the molecule for the receptor (Table 3.1 and Figure 3.8).

AA O

O O NH N R HN CO2H

HN NH2 N-Acyl Oxazole Arginine Amino acid

Figure 3.3 N-Acyl amino acid-oxazole-arginine scaffold.

Cyclohexyalanine (compound 60) has also been explored due to the previous antagonist results obtained for hexapeptide FLTChaAR (Chapter 2). However, the calcium mobilisation assay showed that all the compounds had agonist activity, while SB290157 did not exert any agonist activity under the same assay conditions. Surprisingly, compounds 56 and 57 became more potent agonists than hC3a (EC50 65 nM) with EC50 19 and 27 nM, respectively (Table 3.1). There was a linear correlation between the competitive binding and calcium release assay data in this study. In 128 CHAPTER 3 Non-peptide agonists for C3a receptor addition, compounds 56 and 57 bound to C3a receptor more tightly than SB290157 (IC50 20, 30 vs 38 nM respectively in competitive assays with 80 pM [125I]-C3a) (Table 3.1). These molecules were modified to two new series of compounds that retain the oxazole linker substituted with either of the amino acids leucine or isoleucine, and replacing the t-butoxy (Boc) group at the hydrophobic cap (N-terminus) with different fragments as shown in Figure 3.4, 3.5 and 3.6. The compounds were then assayed to investigate structure activity relationships (SAR). Compounds 57, 61, 62, 64, 65, 66, 67, 68, 71, 73, 74, 75, 76, 77 and 78 include the side chain of isoleucine, while the molecules 56, 63, 69, 70, 72 and 79 contain the side chain of leucine (Figure 3.6).

Compounds 65 and 71 with the isoleucine side chain show high affinity for C3aR: 65 (IC50

= 123 nM) and 71 (IC50 = 155 nM). Compound 65 has an isoquinoline ring in its structure, while 71 has a 4-biphenyl aromatic group. Both of these compounds were potent agonists with EC50 approximately 41 and 8 nM respectively. Interestingly, there was no correlation between the binding affinity and functional activity of these compounds. This could be due to the lack of selectivity of the compounds which led to stimulation of the intracellular calcium release through other receptors as well as C3a receptor. Further modifications to the structure of these new lead compounds have generated some better compounds, such as 75 and 73. Compound 75 has a bromine substituent on the pyridine ring. This significantly improved the binding affinity of the compound with an IC50 = 5 nM and became one of the most potent agonist compounds in this study

(EC50 = 17 nM). Conversely, substitution of the hydrophobic region with a phenoxyphenyl group

(compound 73) gave lower binding affinity to the receptor (IC50= 55 nM), but had full agonist activity in calcium mobilisation assays tested on HMDM cells (EC50 = 54 nM). On the other hand, two molecules 56 and 63 incorporating the leucine side chain have shown tight binding to the C3aR. Compound 56 is similar to 57 (with the Boc group at the hydrophobic region). The only difference concerns the side chain of the amino acid substitution (isoleucine versus leucine). This alteration gives a slight improvement in both binding affinity and agonist potency to compound 56. The substitution of 56 with the indole ring (compound 63) at the hydrophobic region gave the greatest improvement in affinity. 63 had IC50 approximately 9 nM and was the most potent compound in the calcium mobilisation assay in HMDM cells (EC50 = 7 nM). These two compounds bind more tightly to the receptor than SB290157 (2 and 4-fold higher). They also have agonist potency 3 to 9 fold greater than native hC3a itself. Interestingly, compound 79, which contains a fluoro substituted phenyl ring binds to the C3a receptor with IC50 = 102 nM.

However, it shows less agonist activity (EC50 = 527 nM), suggesting that modification of this compound could eventually lead to C3aR antagonists.

129 CHAPTER 3 Non-peptide agonists for C3a receptor

Figure 3.4 N-Acyl-modifications. The hydrophobic/aromatic region of the oxazole-arginine scaffold was substituted with these different fragments.

130 CHAPTER 3 Non-peptide agonists for C3a receptor

Table 3.1 Competitive binding with [125I]-C3a and functional activity (Ca2+ mobilisation) of compounds on isolated human monocyte-derived macrophages (HMDM).

[125I]-C3a Binding assay Ca2+ release assay

IC50 EC50 Compound R AA n pIC50 ± SE %a n pEC50±SE %b (nM) (nM)

hC3a 12 9.64 ± 0.04 0.23 0 3 7.19 ± 0.06 65 -

SB290157 11 7.42 ± 0.06 38 6 7 5.52 ± 0.08 3050(IC50) 53 56 a Leu 3 7.71 ± 0.08« 20 21 4 7.72 ± 0.27 19 - 57 a Ile 3 7.53 ± 0.10Ï 30 18 4 7.57 ± 0.11 27 - * 58 a Phe 2 6.29 ± 0.14Ï« 509 34† - - - - * 59 a Trp 2 5.64 ± 0.31Ï« 2286 76† - - - - * * 60 a Cha 3 7.09 ± 0.11Ï« 82 2 6.87 ± 0.18Ï 135 - 61 b Ile 2 7.58 ± 0.07Ï 26 59 2 5.30 ± 0.18Ï* 5000 -

* * 62 c Ile 2 6.86 ± 0.08Ï« 137 33 2 5.25 ± 0.31Ï 5600 - 63 d Leu 3 8.03 ± 0.05« 9.3 7 3 8.15 ± 0.10 7.2 - * 64 e Ile 4 6.82 ± 0.10Ï« 152 19 3 7.14 ± 0.19Ï 73 - * 65 f Ile 3 6.91 ± 0.09Ï« 123 10 7 7.39 ± 0.12Ï 41 - * 66 g Ile 2 6.10 ± 0.16Ï« 800 68 - - - 73 * 67 h Ile 2 6.80 ± 0.15Ï« 157 42 - - - 83 68 i Ile 2 7.14 ± 0.19Ï* 72 76 2 6.30 ± 0.90Ï* 498 - * 69 j Leu 2 5.99 ± 0.25Ï« 1031 27 - - - 83 * 70 k Leu 2 6.35 ± 0.17Ï« 451 19 - - - 68 * 71 l Ile 3 6.81 ± 0.07Ï« 155 13 7 8.12 ± 0.21 8 - * 72 m Leu 2 6.85 ± 0.15Ï« 142 49 - - - 100 73 n Ile 3 7.26 ± 0.09Ï* 55 11 3 7.26 ± 0.08Ï 54 -

* * 74 o Ile 2 6.12 ± 0.13Ï« 754 19 3 6.13 ± 0.11Ï 744 - * 75 p Ile 3 8.30 ± 0.10« 5 14 3 7.76 ± 0.36 17 -

* * 76 q Ile 5 6.98 ± 0.11Ï« 104 17 3 6.62 ± 0.28Ï 240 - * 77 r Ile 5 6.15 ± 0.14Ï« 712 25 2 ∞ ∞ -

* * 78 s Ile 5 6.06 ± 0.13Ï« 872 23 2 6.07 ± 0.13Ï 849 - * * 79 t Leu 6 6.99 ± 0.10Ï« 102 15 3 6.28 ± 0.16Ï 527 - a = % binding of [125I]-C3a at compound concentration of 20 µM ( or † = [10 µM]); compared to the binding of [125I]- C3a with cells in the absence of compound (or Buffer) as 100% ; b = % Response to 100 nM C3a activity. The compounds tested at [10 µM]. Buffer at 100%; pIC50 and pEC50 expressed as mean ± SE; n represents the number of separate experiments performed..; « p < 0.05 vs. SB290157; * p <0.05 vs. 56 and Ï p < 0.05 vs. 63.

131 CHAPTER 3 Non-peptide agonists for C3a receptor 3.3.1.1 Affinity of Non-Peptide Agonists for C3aR on HMDM.

Competitive Radioligand Binding Assay: [125I]-C3a Binding

The evaluation of twenty non-peptidic compounds, at a single concentration (20 µM) in a competitive radioligand binding assay against human C3a is shown in Figure 3.5.

3000 ) m p

c 2000 (

a 3 C - ] I

1000 5 2 1 [

0 r 7 6 7 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 5 5 6 6 6 6 6 6 6 6 6 7 7 7 7 7 7 7 7 7 7 ffe u B B29015 S [compounds 20 µM] Figure 3.5 Competitive binding between [125I]-C3a and C3a non-peptide ligands in human monocyte derived macrophages. Human monocyte derived macrophage cells (HMDM) (1.5x106 cells/mL) were incubated for 1 h at room temperature with constant concentration (80 pM) of 125I-C3a and non-peptide ligands and SB290157 at concentrations of 20 µM. The buffer represents the total binding of [125I]-C3a to HMDM. Data is expressed as mean [125I]-C3a binding ± S.E.M. with n=2.

The non-peptidic ligands were screened for selectivity towards C3aR using a [125I]-C3a competitive binding assay. From the results shown in Figure 3.7 and 3.8, there are some compounds (eg. 57, 64, 65, 71 and 63 to 79) that bind to C3aR on human monocyte derived macrophage cells. A full concentration response curve has been obtained for these compounds in the same assay.

4000 4000 ) ) m 3000 m 3000 p p c c ( (

a a 3 2000 3 2000 C C - - ] ] I I

5 5 2 2

1 1000

1 1000 [ [

0 0 r 7 7 1 2 4 5 6 7 8 1 3 5 r 7 6 3 9 0 2 9 5 6 6 6 6 6 6 6 7 7 7 5 6 6 7 7 7 ffe ffe u u B B B29015 B29015 S Ligand [20 µM] S Ligand [20 µM] A. Isoleucine-oxazole Arginine Derivative B. Leucine-oxazole Arginine Derivative

Figure 3.6 Structure activity relationship studies of C3a non-peptidic agonists by substitution of AA with either isoleucine (A) or leucine (B) to oxazole scaffold. The affinities of the ligands were tested using a competitive binding assay to compete between [125I]-C3a (80 pM) and C3a non- peptide ligands at concentration of 20 µM in human monocytes derived macrophages.

132 hC3a (n=12) TR4401-19(n=2) CHAPTER 3 Non-peptide agonists for C3a receptorS B290157 (n=11) TR4167-66(n=1) 3.3.1.2 Affinities of C3a Non-Peptide Ligands TR72(n=3) TR4167-71(n=2) TR75(n=3) TR4401-24(n=1) 120 TR4401-01(n=3) TR4401-18(n=2)

g 100 TR4401-08(n=3) TR4167-116(n=2) n i TR4167-118(n=4) TR4167-117(n=1) d a 3 n 80 i TR4401-16(n=3) TR4401-06(n=2) C b -

] TR4401-12(n=3) TR4401-05(n=1) I c i 5 60 f 2

i TR4401-25 (n=3) TR4401-02(n=1) 1 c [

e f TR4401-27(n=3) TR4401-15(n=1)

p 40 o

S TR4401-20(n=6)

% 20 TR4401-26(n=5) TR4401-28(n=5) 0 TR4401-29(n=5) -14 -12 -10 -8 -6 -4 -2

log [Ligand](M) Figure 3.7 Competitive [125I]-C3a binding of hC3a (l, n=12), SB290157 (n, n=11) and C3a non-peptide ligands (57(), 65(·), 71(), 64(£), 63(r), 56(s), 75(¯), 73(*), 79(é), 76 (+), 77(Î), 78(), 70(), 58(£), 74(¯), 69(«), 61(*), 62(«), 68(¯), 67(¯), 66(+), 72(·), n≥3) on human monocyte derived macrophages. Human monocyte derived macrophage cells (1.5x106 cells/mL) were incubated for 1 h at room temperature with 80 pM of 125I-C3a and increasing concentrations of non-labelled C3a (concentration 1 pM to 1 µM) or non-labelled non-peptidic ligands (concentration 0.1 nM to 100 µM). Each point represented the mean percentage specific binding ± S.E.M.

The Binding Affinity Studies - Concentration Dependent Curves of Boc-oxazole Scaffold Ligands with the Variation of Amino acid Substituents.

Figure 3.8 The comparisons in binding affinity of non-peptidic ligands 57 (p, n=3), 56 (·, n=3), 58 (Æ, n=1), 59 (Î, n=1) and 60 (q, n=3). These compounds have a Boc group at the hydrophobic/aromatic region and substituted with difference amino acids at the linker region of oxazole-arginine scaffold. It shows that leucine or isoleucine is the optimal amino acid in this scaffold. hC3a (l, n=12) and SB290157 (n, n=11). 133 CHAPTER 3 Non-peptide agonists for C3a receptor

3.3.2 Series II: Incorporation of a Methyl Group on the 5-Position on the Oxazole Ring.

Molecular modelling studies of the tetrapeptide GLAR from the C-terminus of the hC3a sequence and 56 (N-acyl-leucine-oxazole-arginine scaffold, Figure 3.9) have suggested that the 5- position of the oxazole ring might mimic the location of the side chain methyl group of the alanine- 76 residue in C3a, and could lead to a more potent and selective ligand for C3aR. Therefore, compound 80 was synthesized by Dr Reid. This molecule contains a methyl group at the 5-position of the oxazole scaffold (Figure 3.10). Binding studies of the new molecule designed through modelling suggestion (80) showed no improvement in the affinity for C3aR and decreased the potency two fold in the calcium mobilisation experiments compared to compound 56 (Table 3.2).

(a.) (b.)

Figure 3.9 The fifth position of oxazole was potentially a position for substitution with a methyl group which could mimic the side chain methyl of the alanine residue in C3a. (a.) Superimposition of an oxazole scaffold compound (56, cyan) and tetrapeptide C-terminal residues of hC3a, GLAR (green) (b.) chemdraw structure of compound 56 with the 5-position indicated. (courtesy of Dr Reid).

Further investigation has been performed on compound 80. Structural modifications include different hydrophobic groups linked to the oxazole scaffold by replacing the Boc group at the hydrophobic region with aromatic substituents (Table 3.3).

134 CHAPTER 3 Non-peptide agonists for C3a receptor

Me O O

O O O O N N NH NH HN COOH HN COOH O O

NH NH

HN NH HN NH2 2

56 80

Figure 3.10 Acyl-leucine-5-methyl-oxazole-arginine (compound 80). Compound 80 mediated intracellular calcium mobilisation from HMDM cells with EC50 = 45 nM (n=3).

Table 3.2 Activity of acyl-leucine-5-methyl-oxazole-arginine (80) on competitive binding with [125I]-C3a and Ca2+ mobilisation from isolated human monocyte derived macrophage cells (HMDM).

[125I]-C3a Binding assay Ca2+ release assay

EC50 IC50 Compound R1 R AA n pIC50 ±SE n pEC50±SE (nM) (nM)

56 - O Leu 3 7.71± 0.08 20 4 7.72 ± 0.27 19

80 Me O Leu 3 7.86 ± 0.08 14 3 7.35 ± 0.18 45

pIC50 and pEC50 expressed as mean ± SE, n represents the number of separate experiments performed.

ACYL-LEUCINE-5-METHYL-OXAZOLE-ARGININE

Me O O N R NH HN COOH

HN NH2 135 CHAPTER 3 Non-peptide agonists for C3a receptor

Table 3.3 SAR data for series II: Acyl-leucine-5-Methyl-Oxazole-Arginine scaffolds.

%[125I]- %[125I]- % % compound R C3a compound R C3a agonist agonist binding binding

O O O 81 19 84 93 41 87 MeO S O MeO 82 14 62 94 O 40 26

H O N O S 83 26 56 95 44 103 O

O O 84 31 77 96 44 44 N MeO O

85 29 68 97 MeO N/A 134 N O O MeO O 86 N/A 109 98 65 67 N MeO O OMe O 87 39 107 99 42 69 MeO

O 88 26 69 100 N/A 64 O N H

O H 89 44 70 101 N 61 50

O O N O 90 N/A 95 102 N 49 48

O O 91 61 50 103 32 68

F3C O O OMe Br 92 O 43 81 104 N/A 105

%[125I]-C3a binding = % binding of [125I]-C3a at compound concentration of 20 µM; compared to the binding of [125I]-C3a with cells in the absence of compound (or Buffer) as 100% . The percentage binding of [125I]-C3a of SB290157 = 12. % agonist = % Response to 100 nM C3a activity . The compounds tested at [10 µM]. Buffer at 100%

136 CHAPTER 3 Non-peptide agonists for C3a receptor

A small library of compounds where the oxazole moiety has been functionalised at the 5th- position with a methyl group, and the acyl fragment includes various hydrophobic groups, was investigated in binding assays and calcium mobilisation in HMDM cells. The binding studies suggest that compounds 81, 82, 83, 84, 85, 103 and 88 have some affinity similar to SB290157 (percentage of [125I]-C3a binding = 12) (Table 3.3). Compounds 91, 98 and 101 have less affinity for C3aR in the competitive binding assay (Figure 3.11 A and 3.12). The first three compounds in this series, 81, 82 and 83, with benzofused heterocyclic rings at the hydrophobic region gave the highest affinity (compound 82 (with a sulphur atom) > 81 (with an oxygen atom) > 83 (with a nitrogen atom)). Compound 82 with a benzothiophene in its structure has shown two-fold higher affinity compared to compound 83 bearing an indole ring. This indicates that there must be some space at this region of the receptor, which could be large enough for the bulky groups to fit into it. Compounds 87 and 88 have cyclohexyl and cyclohexylmethyl groups and show some affinity to C3aR. In addition, the cyclohexylmethyl group at this position seems to be more optimal than the cyclohexyl group. This suggests that the pocket of the receptor in this region is deep therefore a larger compound can fit better into the pocket. The presence of a bulkier group, such as in compound 91, showed less affinity and potency. Moreover, the presence of oxygen on the methoxy group in compounds 92, 93, 94, 95, 96, 97, 98 and 99 did not give any improvement in binding affinity. On the other hand, compound 98 showed lower binding affinity.

A. 1500 ) m p

c 1000 (

a 3 C - ] I

500 5 2 1 [

0 r 7 1 2 3 4 5 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9 9 ffe 10 10 10 10 u B B29015 S [compounds 20 µM]

137 CHAPTER 3 Non-peptide agonists for C3a receptor B.

)

0 150 2 a 5 3 C m

125 E ( M

n e

c 100 0 n 0 e 1

c 75 y s b e

r d

o 50 e u l c F u

25 d

n i

% 0 7 1 2 3 4 5 7 8 9 1 2 3 4 5 6 8 9 1 2 3 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9 10 10 10

B29015 S [compounds 10 µM] Figure 3.11 A. Competitive binding between [125I]-C3a and non-peptide ligands in human monocyte derived macrophages. Human monocyte derived macrophage cells (HMDM) (1.5x106 cells/mL) were incubated for 1 h at room temperature with constant concentration (80 pM) of 125I-C3a and non-peptide ligands and SB290157 at a concentration of 20 µM. The buffer represents the total binding of [125I]-C3a to HMDM. B. Non-peptide agonism screening in Ca2+ release assay in human derived macrophage cells at a concentration of 10 µM. Induction of calcium efflux from HMDM cells by compounds expressed as a percentage of cytosolic calcium efflux induced by 100 nM C3a ± S.E.M. with n ≥ 2.

120 hC3a

g SB290157

n 100 i a d BR3924_64 3 n

i 80 C PC95_1 - b

] I c 5 i 60 2 f i 1 [ c

f e 40 o p

S 20 %

0 -14 -12 -10 -8 -6 -4 -2

[log ligand]

Figure 3.12 Competitive [125I]-C3a binding of hC3a (l, n=12), SB290157 (n, n=11), 80 (, n=3) and 81 (r, n=3) on human monocyte derived macrophage cells. 80 pM of [125I]-C3a was competed with increasing concentration of non-labelled hC3a (concentration 1 pM to 1 µM), SB290157, 80 (concentration 0.1 nM to 100 µM) and 81 (concentration 0.01 nM to 10 µM). It shows that 80 binds with high affinity to C3aR ((IC50 ± 95% CI) = 14.7 ± 5.4 nM). Each point represents the mean percentage of specific binding ± S.E.M.

138 CHAPTER 3 Non-peptide agonists for C3a receptor

3.3.3 Selectivity of C3a Non-Peptide Ligands for C5aR

3.3.3.1 Competitive Radio-ligand Binding Assays: [125I]-C5a Binding The competitive radiolabelled C5a assay was used to evaluate whether the compounds bound selectively to C3aR over C5aR. A single concentration screen (ligands at 20 µM) for C5aR binding affinity shows none of these compounds have affinity for C5aR on human macrophages (HMDM)(Figure 3.13). The binding result of all compounds was compared to a control compound 3D53, which is a known as a selective C5aR antagonist with no cross reactivity to C3aR (Chapter 2). This result suggests that all non-peptide ligands are selective for C3aR over C5aR (Figure 3.13)..

6000 ) m p

c 4000 (

a 5 C - ]

I 2000 5 2 1 [

0 r 3 6 7 1 2 3 4 5 6 7 8 9 0 1 2 3 4 5 6 7 8 9 5 5 6 6 6 6 6 6 6 6 6 7 7 7 7 7 7 7 7 7 7 ffe u 3D5 B [compounds 20 µM] 5000 ) 4000 m p c (

3000 a 5 C

- 2000 ] I 5 2

1 1000 [

0 r 3 1 2 3 6 5 7 8 9 1 2 3 4 5 6 7 8 9 1 2 3 4 8 8 8 8 8 8 8 8 9 9 9 9 9 9 9 9 9 ffe 10 10 10 10 u 3D5 B [compounds 20 µM]

Figure 3.13 Competitive binding between [125I]-C5a and C3a non-peptide ligands in human monocyte derived macrophages. Human monocyte derived macrophage cells (1.5x106 cells/mL) were incubated for 1 h at room temperature with constant concentration (20 pM) of [125I]-C5a and non-peptide ligands and 3D53 at a concentration of 20 µM. The buffer represents the total binding of [125I]-C5a to HMDM (~ 4000 cpm). Data is expressed as mean [125I]-C5a binding ± S.E.M. with n=2.

139 CHAPTER 3 Non-peptide agonists for C3a receptor

3.3.3.2 The Affinity to C5aR of Non-Peptide C3a Ligands: Studies in Concentration Dependent Manner.

To confirm this result, the most potent oxazole analogues have been studied in a full concentration response curve in the same assay as shown in Figure 3.14.

6000 hC5a )

m 3D53

p 4000 c TR4401-25 (

a TR4167-72 5

C TR4401-16 - ] I 2000 PC95_1 5 2 1

[ BR64

0 -14 -12 -10 -8 -6 -4 -2

Log [ ligands] (M)

Figure 3.14 Competitive [125I]-C5a binding of hC5a (l, n=3), 3D53 (n, n= 3) and C3a non- peptide ligands 63 (w), 57 (q), 75 (p), 80 (n), 81 (l), n=3 on human monocyte derived macrophage cells. Human monocyte derived macrophage cells (1.5x106 cells/mL) were incubated for 1 h at room temperature with 20 pM of [125I]-C5a and increasing concentrations of non-labelled C3a (concentration 1 pM to 0.3 µM) and non-peptidic ligands (concentration 0.1 nM to 100 µM). Each point represents the mean percentage specific binding ± S.E.M.

3.3.4 Selectivity Test: Desensitisation Experiments

Desensitisation experiments were used to measure the selectivity of various ligands for C3aR, since this can help to clarify whether the non-peptidic ligands bind selectively to C3aR over other GPCRs that signal through intracellular calcium release. The hC3a (3 µM) was administered to human monocyte derived macrophage cells, which produces an efflux of intracellular Ca2+ that dissipates after 4-5 mins (Figure 3.15). A second addition of 3 µM hC3a after 5 mins has no effect, and this effect is attributed to desensitisation of the receptor through either receptor (C3aR) phosphorylation or receptor internalisation from the cell surface. Figure 3.15 shows how compound 73 caused a second efflux of intracellular Ca2+. This

140 CHAPTER 3 Non-peptide agonists for C3a receptor suggests that this non-peptidic ligand is not purely selective for C3aR. In contrast, compounds 57,

65, 71 and 56 (EC50= 27, 41, 8 and 19 nM, respectively) appear to be selective for C3aR as a second addition of 1 µM of these compounds, following desensitisation with C3a, did not produce calcium efflux from macrophage cells. The possibility that these compounds (63, 73 and 80) may bind to C5aR has already been discounted by the competitive [125I]-C5a binding studies that showed no binding to C5aR (Figure 3.13 and 3.14). This suggests that the second efflux of calcium induced by these compounds is caused by a calcium-associated GPCR other than C3aR or C5aR.

A. 250 seconds 3µM C3a 3µM C3a ) 0 ) 2

0 100 5 2

100 a 5 a 3 M 3 M E C

( 80 E C

( 80

M e M e µ c µ c 3 n 60

3 n 60 e

y e c y b

c s b

s 40 e d r 40 e e 3 M C3a

r µ 1 M 57 e o c µ o c u u l 20 u u d l 20 F

d n F i

n i

% 0

0 60 120 180 240 300 360 0 60 120 180 240 300 360 Time(sec) Time(sec) 3µM C3a 3µM C3a ) ) 0 0 2 100 2 100 5 a 5 a 3 M 3 M E C

( 80 E C

( 80

M e M e µ c µ c 3 n 60

3 n 60

e y e y c b c

b s

s d 40 e

d 40 e r e r e o 1 M 63 c 1µM 75 o c µ u u l u u 20 d l

20 F d

n F i

n i

% 0

0 60 120 180 240 300 360 0 60 120 180 240 300 360 Time(sec) Time(sec) 3µM C3a 3µM C3a ) 0

2 100 ) 5 0 a

2 100 3 M 5 a E C 3

( 80

E C M e

( 80

µ c M e 3 n 60

µ c e y 3 n 60

c b e

y s c d 40 b e

s r e

d 40 e o 10µM 79 c 1µM 80 r e u u l o c 20 d F u u

l 20 n i d F

n i

% 0

% 0 0 60 120 180 240 300 360 0 60 120 180 240 300 360 Time(sec) Time(sec)

141 CHAPTER 3 Non-peptide agonists for C3a receptor

B. 280 seconds

)

3µM C3a ) 3µM C3a m m n n 0 100 2 0 100 a 5 2 a 3 5 3 m C

80 m E C ( 80

E M ( e µ

M c 3 e 60 µ

n c 3 y 60

e n b y

c e d b s 40

c e e d s 40 r c 3 M C3a e

µ e o u r c

u 20

d 1 M 65

l µ o u n F i u 20 d l n

0 i

% 0 0 60 120 180 240 300 360 % Time(sec) 0 60 120 180 240 300 360 Time(sec)

Figure 3.15 C3a receptor desensitisation calcium mobilisation plots performed on human monocyte derived macrophages (HMDM) for non-peptidic agonists. HMDM were first treated with 3 µM of hC3a, which induced a calcium response that was allowed to dissipate over 250 (A) or 280 (B) seconds prior to a second agonist injection.

142 CHAPTER 3 Non-peptide agonists for C3a receptor

3.3.5 Effect of C3a Non-Peptide Agonists/Antagonists on Calcium Release from Human Monocyte Derived Macrophages.

C3a non-peptidic ligands were also examined in the calcium mobilisation experiment for antagonism, using SB290157 as a control compound (Figure 3.16). Compounds with greater antagonist activity than SB290157 were examined further for calcium mobilisation in a dose dependent manner. The antagonist screening was used to test all compounds at a single concentration of non-peptidic compound, which can provide a quick result and allow the relative activities of multiple compounds to be compared in a single experiment. However it also can give a false positive if compounds display any partial agonism activity for the receptor. For example, if an agonist ligand is present, it can cause desensitisation/internalisation of the receptor and therefore inhibit the ability of the cell to mobilise calcium in response to stimulation with C3a. So compounds must first be shown not to cause Ca2+ release before they can be tested for antagonism.

)

m 100 n a 0 3 2 5 C

80 m M E n (

0 e

0 60 c 1

n y e b c 40 s d e e r c o u

u 20 d l n F i

0 % 7 7 1 2 4 5 6 7 8 1 5 6 6 6 6 6 6 6 7

B29015 S [compounds 10µM]

)

0 125 2 a 5 3 C m

E 100 ( M

n e

c 0 n

0 75 e 1

c y s b e

50 r d o e u l c F

u 25

n i

% 0 7 6 3 9 0 2 3 4 5 6 7 8 9 5 6 6 7 7 7 7 7 7 7 7 7

B29015 [compounds 10 M] S µ Figure 3.16 Non-peptide antagonism screening in Ca2+ release assay in human derived macrophage cells at a concentration of 10 µM. Intracellular calcium release caused by 100 nM hC3a. Inhibition expressed as a percentage of maximum change in fluorescence induced by 100 nM C3a. Error bars represented S.E.M. with n=2.

143 CHAPTER 3 Non-peptide agonists for C3a receptor 3.3.5.1 C3aR Non-Peptide Agonists

From the results of calcium mobilisation experiments, the concentration response curve for C3a non-peptide agonists on human monocyte derived macrophage cells (Figure 3.17 and 3.18), the most potent non-peptidic agonists in this study were compounds 75 (EC50 = 17 nM, pEC50 =

7.76 ± 0.36, n=3), 56 (EC50= 19 nM, pEC50 7.72 ± 0.27, n = 4 and 63 (EC50= 7.2 nM; pEC50 8.15 ±

0.10, n=3) and 71 (EC50 = 8 nM, pEC50 8.12 ± 0.21, n=7). The competitive binding studies (Figure 3.7 and Table 3.1) also show that 75, 56, 63 have high affinity for C3aR on HMDM (IC50 < 20 nM), while compound 71 has lower affinity than those compounds. It still can stimulate calcium release from cells at very low concentrations. This observation indicates that this compound, though having limited binding affinity, still has good potency in terms of calcium release from cells, similar to those compounds having better binding.

TR4401-08 (n=7) TR4401-01 (n=7) ) 125 TR4401-16 (n=3) m n a

0 TR4401-24 (n=3) 3 2 100 5 C TR4401-61 (n=3)

m M

E TR4167-118 (n=3) n (

0 75 e

0 TR4401-12 (n=4) c 1

y TR4401-25 (n=3) e

b 50 c

s TR4167-72 (n=4) d e e r

c TR4167-75 (n=3) o

u 25 u l d SB290157 (n=6) F n

i TR4401-27(n=3) 0

% hC3a (n=11) TR117(n=2) -12 -10 -8 -6 -4 -2 TR4167-61(n=3) Log [Agonist] (M) TR4401-06(n=2)

BR64(n=4) 2+ Figure 3.17 Intracellular Ca mobilisation dose response induction by hC3a (l, n=11),B RTR20(n=3) SB290157 («, n=6) and variable concentration of C3a non-peptide agonists. 71 (, n=7), 65 (£, n=7), 56 (£, n=4), 57 (s, n=4), 80 («, n=4), 75 (r, n=3), 63 (p, n=3), 74 (q, n=3),T R28 agonist (n=3) 64 (, n=3), 73 (Î, n=3), 62 (¯, n=2), 68 (r, n=2), 79 (®, n=3), 77 (, n=3), 76 (p, n=3).T R26 agonist (n=3) The results are expressed as a percentage of maximum change in fluorescence induced by hC3a (100 nM) ± S.E.M.

144 CHAPTER 3 Non-peptide agonists for C3a receptor

TR4401-08 (n=7) 120 TR4401-01 (n=7) TR4401-16 (n=3)

x 100 TR4401-12 (n=4) a e M

c TR4401-25 (n=3) 80 n d

e TR4167-72 (n=4) e c s i

s 60 l TR4401-27(n=3) e a r m o

r 40 hC3a (n=11) u l o F N 20 % 0

-12 -10 -8 -6 -4 -2

Log [Agonist] (M) Figure 3.18 Intracellular calcium release in HMDM cells induced by hC3a (s, n=11), 71 (l, n=7), 65 (n, n=7), 63 (p, n=), 56 (q, n=4), 75 (®, n=3), 57 (, n=4), 73 (r, n=3). The results are expressed as a percentage of maximum change in fluorescence ± S.E.M.

3.4 DISCUSSION AND CONCLUSIONS

C3a peptides have many disadvantages for use as pharmacological/physiological tools for in vivo studies. For example, they have low bioavailability, poor stability to protease enzymes, high clearance rates from serum, and also low membrane permeability. Therefore, the development of non-peptidic C3a ligands that presumably will have improved bioavailability and stability are needed to interrogate the roles of C3a in vitro and in vivo and to serve as leads to new pharmaceuticals. C3a is involved in the pathogenesis and progression of numerous inflammatory conditions, including asthma, allergies, arthritis, psoriasis, ischemia-reperfusion injury, sepsis, systemic lupus erythematosus, diabetes, nephropathy and others. It may be a very attractive target for therapeutic intervention, however as yet there has been no successful development of a small potent and selective C3aR agonist or antagonist that can be used as a pharmacological/physiological probe in studying the roles of C3a in vivo. The first and most studied non-peptidic antagonist of C3aR was the compound SB290157, which was reported by Smithkline Beecham in 2001 (Ames et al., 2001). It binds to C3aR with IC50 = 200 nM in a competitive radioligand binding assay in RBL cells and showed antagonist activities in both in vitro and in vivo studies (Ames et al., 2001). However, the study by Mathieu and

145 CHAPTER 3 Non-peptide agonists for C3a receptor colleagues (2005) showed that it has potent agonist activities in a variety of assays e.g. calcium mobilisation assay in transfected RBL cells, a beta-lactamase assay in CHO-NFAT-bla-Gα16 (chinese hamster ovary cells) and an enzyme release assay in dU937 cells. It has been suggested that SB290157 is a partial agonist, which shows its agonist activity when there is a high level of C3aR expression on the cells. We chose FLTLAR as a peptidic lead compound and modified the N-terminus, incorporated a turn-inducing oxazole scaffold but retained the arginine, to produce new compounds for these studies. From the new series of C3aR non-peptidic agonists obtained by modification of FLTLAR, we found that the incorporation of an oxazole ring leads to improved binding affinity and potency for many of the new compounds generated. The incorporation of oxazole was also a dramatic improvement on the previous report by Denonne et al. (2007) where replacing the ether group with a more rigid furan was claimed to give improved affinity for C3aR. We found that the optimal amino acid side chain for attachment to the oxazole scaffold was leucine (56) or isoleucine (57), with only a slight difference in the binding affinity and potency of the respective compounds.

Compounds 75, 63, 56 and 80 have the highest affinity for HMDM (IC50 = 5, 9.3, 20 and 14 nM respectively in competitive binding studies), show selectivity for C3aR in competitive radioligand binding assays (both 125I labelled C3a and C5a) and also in a desensitisation assay, which confirmed the selectivity of the ligands by their inability to induce calcium release after desensitisation of the cells with C3a. In Figure 3.19 there was a pseudolinear correlation between agonist potency (intracellular release of calcium in human monocyte derived macrophage cells) and the binding affinity for C3aR (measured by competition with [125I] labelled C3a on HMDM). From these data, we found that all of these compounds have agonist activities. Surprisingly some of the compounds show even more potent agonist activity in calcium mobilisation assay than does hC3a. Molecular modelling (Figure 3.2) showed the overlay of the C3a carboxyl terminal residues, G73LAR76, modelled in a beta turn conformation, with compound 56 displaying structural similarity. The oxazole at the linker region could assist the compound to adopt a beta-turn conformation that mimics the conformation of the C-terminus of C3a. This result also supports the theory that GPCRs have a tendency to recognise protein and peptide ligands with turn structure (Fairlie et al., 2005; Madala et al., 2010; Tyndall et al., 2005). In conclusion, we can say that the introduction of a rigid turn-like conformation by using an oxazole heterocycle has led us to the discovery of multiple potent and selective C3aR non-peptidic agonists (63, 75 and 56, EC50 < 20 nM in HMDM cells). These compounds could be useful for probing the physiological/pharmacological roles of C3a in vitro studies however these compounds retained the C-terminus arginine that could have poor bioavailability (Denonne et al., 2007) for 146 CHAPTER 3 Non-peptide agonists for C3a receptor in vivo studies therefore the need for further modification to improve these compounds by using arginine mimetics have been suggested.

Figure 3.19 Correlation between binding affinity (pIC50) and agonist activity (pEC50) of non-peptide agonists (56, 57, 60, 61, 62, 63, 64, 65, 68, 71, 73, 74, 75, 80), nonpeptide antagonist (SB290157) and hC3a.

147 CHAPTER 3 Non-peptide agonists for C3a receptor

3.5 MATERIALS AND METHODS

3.5.1 Macrophage Cell Culture and Differentiation.

Human monocyte-derived macrophage (HMDM) cells were kept in complete media, consisting of IMDM with 10% FBS, 10 U/mL penicillin, 10 U/mL streptomycin and 2 mM L- glutamine (Invitrogen). Cells were cultured at 37°C, with 5% CO2. For HMDM, peripheral blood mononuclear cells were isolated from buffy coat (obtained from Australian Red Cross Blood Service, Kelvin Grove) using Ficoll-paque density centrifugation (GE Healthcare Bio-Science, Uppsala, Sweden). CD14+ monocytes were positively selected using CD14+ MACS magnetic beads (Miltenyi Biotech, Auburn, CA, USA). Monocytes were differentiated to HMDM in complete media containing 104 U/mL (100 ng/mL) recombinant human macrophage colony stimulating factor (M-CSF) (PeptroTech Inc, Rocky Hill, New Jersey, USA) at 1.5 x 106 monocytes/mL. HMDM were supplemented with 50% fresh complete medium containing CSF-1 on Day 5 after seeding. Cells were harvested by gentle scraping in saline solution and replated for use on Day 7.

3.5.2 Calcium Mobilization Assay on HMDM (Adhesion cells)

In this experiment the concept is the same as for the calcium mobilisation assay as described in Chapter 2. However, the protocol in preparation of the HMDM cells (adhesion cells) for the assays is different from the dU937 cells (suspension cells). Harvested HMDM cells were washed with 0.9% NaCl solution by centrifugation at 2500 rpm for 5 min, followed by resuspension of the cell pellet with complete media. Cells were seeded at 5 x 104 cells/well in a 96-well black-wall, clear bottom plate (DKSH, Zurich), and left overnight to adhere at 37°C. On the day of the experiment, supernatant was removed and cells were incubated in dye loading buffer (Hank’s Balanced Salt Solution (HBSS) with 4µM Fluo-3, 25 µL pluronic acid, 1% fetal bovine serum (FBS) and 2.5 mM probenecid) for one hour at 37°C. Cells were then washed twice with HBSS and transferred to a Fluorostar spectrofluorimeter (BMG, Durham NC) for agonist injection and fluorescence measurements.

3.5.3 Receptor Binding Assay of HMDM

Receptor binding assays were performed using [125I]-C3a 80 pM (2200 Ci/mmol; Perkin Elmer, Torrance, CA, USA), HMDM cells (1.5x106 cells/mL) and in the absence or presence of various concentrations of unlabelled C3a or C3a hexapeptide agonist for 60 mins at room temperature with shaking in 50 mM Tris, 3 mM MgCl2, 0.1 mM CaCl2, 0.5% (w/v) Bovine serum albumin, pH 7.4. Unbound radioactivity was removed by filtration through glass microfiber filters 148 CHAPTER 3 Non-peptide agonists for C3a receptor GF/B (Whatman Iner. Ltd, England) which had been soaked in 0.6% polyethylenimine to reduce non-specific binding. The filter was washed 3 times with cold buffer (50 mM Tris-HCl) pH 7.4. Bound [125I]-C3a will then be assessed by scintillation counting on a β-counter. Specific [125I]-C3a binding is defined as a difference between total binding and non-specific binding as determined in the presence of 1 µM unlabeled C3a. The IC50 value is the concentration of antagonist to inhibit the binding of labelled ligand by 50 percent.

3.5.4 Data analysis and statistics

Receptor binding data was expressed as specific binding (i.e. total binding – non-specific binding and non-specific binding = binding with 1 µM C3a). Functional activity data is expressed as a percentage of the maximum response induced by C3a (100 nM). Nonlinear regression analysis

(GraphPad Prism 5, USA) will be performed on concentration response curves to determine EC50 or

IC50 and pEC50 or pIC50.The statistical analysis has been utilized for both receptor binding assays and functional assays.

In receptor binding assays, pIC50 for each compound will be calculated for separate experiments and express as an arithmetic mean standard error (SE). IC50 values will be expressed as a geometric mean. A one-way ANOVA coupled with Tukey post test was performed on pIC50 values to compare compound affinities. P ≤ 0.05 was considered to indicate a statistically significant difference between receptor affinities. In Functional Assays, the concentration curves were analysed by non-linear regression

(GraphPad Prism 5, USA) and EC50 and pEC50 determined. A pEC50 for each compound was calculated from separate experiments and expressed as an arithmetic mean ± SE. A one-way

ANOVA coupled with a Tukey post test was performed on pEC50 values to examine statistical significance between agonist potencies of C3a analogue peptides/non-peptides. P ≤ 0.05 was considered to indicate a statistically significant difference.

149 CHAPTER 3 Non-peptide agonists for C3a receptor

3.6 REFERENCES

1. AMES, RS, LEE, D, FOLEY, JJ, JUREWICZ, AJ, TORNETTA, MA, BAUTSCH, W, SETTMACHER, B, KLOS, A, ERHARD, KF, COUSINS, RD, SULPIZIO, AC, HIEBLE, JP, MCCAFFERTY, G, WARD, KW, ADAMS, JL, BONDINELL, WE, UNDERWOOD, DC, OSBORN, RR, BADGER, AM, SARAU, HM (2001) Identification of a selective nonpeptide antagonist of the anaphylatoxin C3a receptor that demonstrates antiinflammatory activity in animal models. J. Immunol. 166(10): 6341-6348.

2. AMES, RS, LI, Y, SARAU, HM, NUTHULAGANTI, P, FOLEY, JJ, ELLIS, C, ZENG, Z, SU, K, JUREWICZ, AJ, HERTZBERG, RP, BERGSMA, DJ, KUMAR, C (1996) Molecular cloning and characterization of the human anaphylatoxin C3a receptor. Journal of Biological Chemistry 271(34): 20231-20234.

3. AMES, RS, TORNETTA, MA, FOLEY, JJ, HUGLI, TE, SARAU, HM (1997) Evidence that the receptor for C4a is distinct from the C3a receptor. Immunopharmacology 38(1-2): 87-92.

4. BAUTSCH, W, HOYMANN, HG, ZHANG, QW, MEIER-WIEDENBACH, I, RASCHKE, U, AMES, RS, SOHNS, B, FLEMME, N, ZU VILSENDORF, AM, GROVE, M, KLOS, A, KOHL, J (2000) Cutting edge: Guinea pigs with a natural C3a-receptor defect exhibit decreased bronchoconstriction in allergic airway disease: Evidence for an involvement of the C3a anaphylatoxin in the pathogenesis of asthma. J. Immunol. 165(10): 5401-5405.

5. BOKISCH, VA, MULLER-EBERHARD, HJ (1970) Anaphylatoxin inactivator of human plasma: its isolation and characterization as a carboxypeptidas. J. Clin. Invest. 49: 2429-2436.

6. BOOS, L, CAMPBELL, LL, AMES, R, WETSEL, RA, BARNUM, SR (2004) Deletion of the complement anaphylatoxin C3a receptor attenuates, whereas ectopic expression of C3a in the brain exacerbates, experimental autoimmune encephalomyelitis. J. Immunol. 173(7): 4708-4714.

7. BOOS, L, SZALAI, AJ, BARNUM, SR (2005) C3a expressed in the central nervous system protects against LPS-induced shock. Neurosci. Lett. 387(2): 68-71.

8. CAPORALE, LH, TIPPETT, PS, ERICKSON, BW, HUGLI, TE (1980) The active-site of C3a anaphylatoxin Journal of Biological Chemistry 255(22): 758-763.

9. CHAO, TH, EMBER, JA, EMBER, JA, WANG, MY, BAYON, Y, HUGLI, TE, YE, RD (1999) Role of the second extracellular loop of human C3a receptor in agonist binding and receptor function. Journal of Biological Chemistry 274(14): 9721-9728.

10. CHAZIN, WJ, HUGLI, TE, WRIGHT, PE (1988) H-1-NMR studies of human C3a anaphylatoxin in solution-sequential resonance assignments, secondary structure, and global fold. Biochemistry 27(26): 9139-9148.

11. CORBIN, NC, HUGLI, TE (1976) Primary structure of porcine C3a anaphylatoxin. J. Immunol. 117(3): 990-995.

150 CHAPTER 3 Non-peptide agonists for C3a receptor

12. CRASS, T, RAFFETSEDER, U, MARTIN, U, GROVE, M, KLOS, A, KOHL, J, BAUTSCH, W (1996) Expression cloning of the human C3a anaphylatoxin receptor (C3aR) from differentiated U- 937 cells. European Journal of Immunology 26(8): 1944-1950.

13. DAFFERN, PJ, PFEIFER, PH, EMBER, JA, HUGLI, TE (1995) C3a is a chemotaxin for human eosinophils but not for neutrophils. I. C3a stimulation of neutrophils is secondary to eosinophil activation. Journal of Experimental Medicine 181(6): 2119-2127.

14. DENONNE, F, BINET, S, BURTON, M, COLLART, P, DIPESA, A, GANGULY, T, GIANNARAS, A, KUMAR, S, LEWIS, T, MAOUNIS, F, NICOLAS, JM, MANSLEY, T, PASAU, P, PREDA, D, STEBBINS, K, VOLOSOV, A, ZOU, D (2007) Discovery of new C3aR ligands. Part 1: Arginine derivatives. Bioorg. Med. Chem. Lett. 17(12): 3258-3261.

15. DROUIN, SM, CORRY, DB, HOLLMAN, TJ, KILDSGAARD, J, WETSEL, RA (2002) Absence of the complement anaphylatoxin C3a receptor suppresses Th2 effector functions in a murine model of pulmonary allergy. J. Immunol. 169(10): 5926-5933.

16. DROUIN, SM, CORRY, DB, KILDSGAARD, J, WETSEL, RA (2001) Cutting edge: The absence of C3 demonstrates a role for complement in Th2 effector functions in a murine model of pulmonary allergy. J. Immunol. 167(8): 4141-4145.

17. ELLATI, SG, DAHINDEN, CA, CHURCH, MK (1994) Complement peptides C3a-induced and C5a-induced mediator release from dissociated human skin mast-cells. J. Invest. Dermatol. 102(5): 803-806.

18. ELSNER, J, OPPERMANN, M, CZECH, W, DOBOS, G, SCHOPF, E, NORGAUER, J, KAPP, A (1994a) C3a activates reactive oxygen radical species production and intracellular calcium transients in human eosinophils. European Journal of Immunology 24(3): 518-522.

19. ELSNER, J, OPPERMANN, M, CZECH, W, KAPP, A (1994b) C3a activates the respiratory burst in human polymorphonuclear neutrophilic leukocytes via pertussis-toxin-sensitive G- proteins. Blood 83(11): 3324-3331.

20. EMBER, JA, JAGELS, MA, HUGLI, TE (1998) Characterization of Complement Anaphylatoxins and their Biological Responses. In: The human complement system in health and disease, Volanakis, JE, Frank, MM (eds), pp 241-284. New York: Marcel Dekker.

21. EMBER, JA, JOHANSEN, NL, HUGLI, TE (1991) Designing synthetic superagonists of C3a anaphylatoxin. Biochemistry 30(15): 3603-3612.

22. FAIRLIE, DP, TYNDALL, JDA, ABBENANTE, G, PFEIFFER, B (2005) Over 100 peptide- activated GPCRS recognize turn motifs in ligands. Abstr. Pap. Am. Chem. Soc. 229: 010- MEDI.

23. FISCHER, WH, HUGLI, TE (1997) Regulation of B cell functions by C3a and C3a(desArg) - Suppression of TNF-alpha, IL-6, and the polyclonal immune response. J. Immunol. 159(9): 4279-4286.

24. FISCHER, WH, JAGELS, MA, HUGLI, TE (1999) Regulation of IL-6 synthesis in human peripheral blood mononuclear cells by C3a and C3a(desArg). J. Immunol. 162(1): 453-459.

151 CHAPTER 3 Non-peptide agonists for C3a receptor

25. FUKUOKA, Y, HUGLI, TE (1988) Demonstration of a specific C3a receptor on guinea pig platelets. J. Immunol. 140(10): 3496-3501.

26. GARRETT, MC, OTTEN, ML, STARKE, RM, KOMOTAR, RJ, MAGOTTI, P, LAMBRIS, JD, RYNKOWSKI, MA, CONNOLLY, ES (2009) Synergistic neuroprotective effects of C3a and C5a receptor blockade following intracerebral hemorrhage. Brain Res. 1298: 171-177.

27. GASQUE, P, SINGHRAO, SK, NEAL, JW, WANG, P, SAYAH, S, FONTAINE, M, MORGAN, BP (1998) The receptor for complement anaphylatoxin C3a is expressed by myeloid cells and nonmyeloid cells in inflamed human central nervous system: Analysis in multiple sclerosis and bacterial meningitis. J. Immunol. 160(7): 3543-3554.

28. GERARDYSCHAHN, R, AMBROSIUS, D, CASARETTO, M, GROTZINGER, J, SAUNDERS, D, WOLLMER, A, BRANDENBURG, D, BITTERSUERMANN, D (1988) Design and biological- activity of a new generation of synthetic C3a analogs by combination of peptidic and non- peptidic elements. Biochem. J. 255(1): 209-216.

29. GUTZMER, R, LISEWSKI, M, ZWIRNER, J, MOMMERT, S, DIESEL, C, WITTMANN, M, KAPP, A, WERFEL, T (2004) Human monocyte-derived dendritic cells are chemoattracted to C3a after up-regulation of the C3a receptor with interferons. Immunology 111(4): 435-443.

30. HAEFFNERCAVAILLON, N, CAVAILLON, JM, LAUDE, M, KAZATCHKINE, MD (1987) C3a(C3a desArg) induces production and release of interleukin-1 by cultured human-monocytes. J. Immunol. 139(3): 794-799.

31. HOEPRICH, PD, HUGLI, TE (1986) Helical conformation at the carboxy-terminal portion of human C3a is required for full activity. Biochemistry 25(8): 1945-1950.

32. HUBER, R, SCHOLZE, H, PAQUES, EP, DEISENHOFER, J (1980) Crystal-structure analysis and molecular-model of human C3a anaphylatoxin. Hoppe-Seylers Zeitschrift Fur Physiologische Chemie 361(9): 1389-1399.

33. HUEY, R, ERICKSON, BW, BLOOR, CM, HUGLI, TE (1984) Contraction of guinea pig lung by synthetic oligopeptides related to human C3a. Immunopharmacology 8(1): 37-45.

34. HUGLI, TE (1986) Biochemistry and biology of anaphylatoxins. In: Complement Vol. 3, pp 111-127.

35. HUGLI, TE (1975) Human anaphylatoxin (C3a) from 3rd component of complement- primary structure. Journal of Biological Chemistry 250(21): 8293-8301.

36. HUGLI, TE, ERICKSON, BW (1977) Synthetic peptides with biological-activities and specificity of human C3a anaphylatoxin. Proc. Natl. Acad. Sci. U. S. A. 74(5): 1826-1830.

37. HUGLI, TE, MORGAN, WT, MULLEREBERHARD, HJ (1975a) Circular-Dichroism of C3a anaphylatoxin- effect of pH, Heat, Guanidium Chloride, and Mercaptoethanol on conformation and function. Journal of Biological Chemistry 250(4): 1479-1483.

38. HUGLI, TE, MULLER-EBERHARD, HJ (1978) Anaphylatoxins: C3a and C5a. Adv Immunol 26: 1-53.

152 CHAPTER 3 Non-peptide agonists for C3a receptor

39. HUGLI, TE, VALLOTA, EH, MULLEREBERHARD, HJ (1975b) Purification and partial characterization of human and porcine C3a anaphylatoxin. Journal of Biological Chemistry 250(4): 1472-1478.

40. HUMBLES, AA, LU, B, NILSSON, CA, LILLY, C, ISRAEL, E, FUJIWARA, Y, GERARD, NP, GERARD, C (2000) A role for the C3a anaphylatoxin receptor in the effector phase of asthma. Nature 406(6799): 998-1001.

41. HUTAMEKALIN, P, TAKEDA, K, TANI, M, TSUGA, Y, OGAWA, N, MIZUTANI, N, YOSHINO, S (2010) Effect of the C3a-Receptor Antagonist SB 290157 on Anti-OVA Polyclonal Antibody-Induced Arthritis. J. Pharmacol. Sci. 112(1): 56-63.

42. ISCHENKO, A, SAYAH, S, PATTE, C, ANDREEV, S, GASQUE, P, SCHOUFT, MT, VAUDRY, H, FONTAINE, M (1998) Expression of a functional anaphylatoxin C3a receptor by astrocytes. J. Neurochem. 71(6): 2487-2496.

43. KAWAMOTO, S, YALCINDAG, A, LAOUINI, D, BRODEUR, S, BRYCE, P, LU, B, HUMBLES, AA, OETTGEN, H, GERARD, C, GEHA, RS (2004) The anaphylatoxin C3a downregulates the Th2 response to epicutaneously introduced antigen. Journal of Clinical Investigation 114(3): 399-407.

44. KILDSGAARD, J, HOLLMANN, TJ, MATTHEWS, KW, BIAN, K, MURAD, F, WETSEL, RA (2000) Cutting edge: Targeted disruption of the C3a receptor gene demonstrates a novel protective anti-inflammatory role for C3a in endotoxin-shock. J. Immunol. 165(10): 5406-5409.

45. KIRSCHFINK, M (2001) Targeting complement in therapy. Immunol. Rev. 180: 177-189.

46. KLOS, A, BANK, S, GIETZ, C, BAUTSCH, W, KOHL, J, BURG, M, KRETZSCHMAR, T (1992) C3a receptor on Dibutyryl-cAMP-differentiated U937 cells and human neutrophils - the human C3a receptor characterized by funcitonal responses and I-125 C3a binding. Biochemistry 31(46): 11274-11282.

47. KOHL, J, CASARETTO, M, GIER, M, KARWATH, G, GIETZ, C, BAUTSCH, W, SAUNDERS, D, BITTERSUERMANN, D (1990) Reevaluation of the C3a active-site using short synthetic C3a analogs. European Journal of Immunology 20(7): 1463-1468.

48. KRETZSCHMAR, T, JEROMIN, A, GIETZ, C, BAUTSCH, W, KLOS, A, KOHL, J, RECHKEMMER, G, BITTERSUERMANN, D (1993) Chronic myelogenous leukemia-derived basophilic granulocytes express a functional active receptor for anaphylatoxin C3a. European Journal of Immunology 23(2): 558-561.

49. LU, Z-X, FOK, K, ERICKSON, B, HUGLI, TE (1984) Conformation analysis of COOH- terminal segments of human C3a. Evidence of ordered conformation in an active 21-residue peptide. Journal of Biological Chemistry 259: 7367-7370.

50. MADALA, PK, TYNDALL, JDA, NALL, T, FAIRLIE, DP (2010) Update 1 of: Proteases Universally Recognize Beta Strands In Their Active Sites. Chem. Rev. 110(6): PR1-PR31.

51. MAMANE, Y, CHAN, CC, LAVALLEE, G, MORIN, N, XU, LJ, HUANG, JQ, GORDON, R, THOMAS, W, LAMB, J, SCHADT, EE, KENNEDY, BP, MANCINI, JA (2009) The C3a Anaphylatoxin Receptor Is a Key Mediator of Insulin Resistance and Functions by

153 CHAPTER 3 Non-peptide agonists for C3a receptor Modulating Adipose Tissue Macrophage Infiltration and Activation. Diabetes 58(9): 2006- 2017.

52. MATHIEU, MC, SAWYER, N, GREIG, GM, HAMEL, M, KARGMAN, S, DUCHARME, Y, LAU, CK, FRIESEN, RW, O'NEILL, GP, GERVAIS, FG, THERIEN, AG (2005) The C3a receptor antagonist SB 290157 has agonist activity. Immunology Letters 100(2): 139-145.

53. MIZUTANI, N, NABE, T, YOSHINO, S (2009) Complement C3a Regulates Late Asthmatic Response and Airway Hyperresponsiveness in Mice. J. Immunol. 183(6): 4039-4046.

54. MOCCO, J, MACK, WJ, DUCRUET, AF, SOSUNOV, SA, SUGHRUE, ME, HASSID, BG, NAIR, MN, LAUFER, I, KOMOTAR, RJ, HOLLAND, MCH, PINSKY, DJ, CONNOLLY, ES (2006) Complement component C3 mediates inflammatory injury following focal cerebral ischemia. Circ.Res. 99(2): 209-217.

55. MONSINJON, T, GASQUE, P, CHAN, P, ISCHENKO, A, BRADY, JJ, FONTAINE, M (2003) Regulation by complement C3a and C5a anaphylatoxins of cytokine production in human umbilical vein endothelial cells. Faseb J. 17(9): 1003-1014.

56. MONSINJON, T, GASQUE, P, ISCHENKO, A, FONTAINE, M (2001) C3A binds to the seven transmembrane anaphylatoxin receptor expressed by epithelial cells and triggers the production of IL-8. FEBS Lett. 487(3): 339-346.

57. MUELLER-ORTIZ, SL, HOLLMANN, TJ, HAVILAND, DL, WETSEL, RA (2006) Ablation of the complement C3a anaphylatoxin receptor causes enhanced killing of Pseudomonas aeruginosa in a mouse model of pneumonia. Am. J. Physiol.-Lung Cell. Mol. Physiol. 291(2): L157-L165.

58. MULLER-EBERHARD, HJ (1988) Molecular organization and function of the complement system. Annu Rev Biochem 57: 321-347.

59. MURAKAMI, Y, IMAMICHI, T, NAGASAWA, S (1993) Characterization of C3a anaphylaxtoxin receptor on guinea-pig macrophages. Immunology 79(4): 633-638.

60. MUTO, Y, FUKUMOTO, Y, ARATA, Y (1985) Proton nuclear magnetic resonance study of the 3rd component of complement solution conformation of the carboxyl terminal segment of C3a fragment. Biochemistry 24(23): 6659-6665.

61. NETTESHEIM, DG, EDALJI, RP, MOLLISON, KW, GREER, J, ZUIDERWEG, ERP (1988) Secondary structure of complement component C3a anaphylatoxin in solution as determined by NMR-Spectroscopy- differences between crystal and solution conformations. Proc. Natl. Acad. Sci. U. S. A. 85(14): 5036-5040.

62. NORGAUER, J, DOBOS, G, KOWNATZKI, E, DAHINDEN, C, BURGER, R, KUPPER, R, GIERSCHIK, P (1993) Complement fragment C3a stimulates Ca2+ influx in neutrophils via a pertussis-toxin-sensitive G-protein. Eur. J. Biochem. 217(1): 289-294.

63. PAQUES, EP, SCHOLZE, H, HUBER, R (1980) Purification and crystallization of human anaphylatoxin C3a. Hoppe-Seylers Zeitschrift Fur Physiologische Chemie 361(6): 977-980.

154 CHAPTER 3 Non-peptide agonists for C3a receptor

64. RAHPEYMAI, Y, HIETALA, MA, WILHELMSSON, U, FOTHERINGHAM, A, DAVIES, I, NILSSON, AK, ZWIRNER, J, WETSEL, RA, GERARD, C, PEKNY, M, PEKNA, M (2006) Complement: a novel factor in basal and ischemia-induced neurogenesis. Embo J. 25(6): 1364-1374.

65. RYNKOWSKI, MA, KIM, GH, GARRETT, MC, ZACHARIA, BE, OTTEN, ML, SOSUNOV, SA, KOMOTAR, RJ, HASSID, BG, DUCRUET, AF, LAMBRIS, JD, CONNOLLY, ES (2009) C3a receptor antagonist attenuates brain injury after intracerebral hemorrhage. J. Cereb. Blood Flow Metab. 29(1): 98-107.

66. SCHRAUFSTATTER, IU, TRIEU, K, SIKORA, L, SRIRAMARAO, P, DISCIPIO, R (2002) Complement C3a and C5a induce different signal transduction cascades in endothelial cells. J. Immunol. 169(4): 2102-2110.

67. TAKAHASHI, M, MORIGUCHI, S, IKENO, M, KONO, S, OHATA, K, USUI, H, KURAHASHI, K, SASAKI, R, YOSHIKAWA, M (1996) Studies on the ileum-contracting mechanisms and identification as a complement C3a receptor agonist of oryzatensin, a bioactive peptide derived from rice albumin. Peptides 17(1): 5-12.

68. TANG, ZY, LU, B, HATCH, E, SACKS, SH, SHEERIN, NS (2009) C3a Mediates Epithelial-to- Mesenchymal Transition in Proteinuric Nephropathy. J. Am. Soc. Nephrol. 20(3): 593-603.

69. TYNDALL, JDA, PFEIFFER, B, ABBENANTE, G, FAIRLIE, DP (2005) Over one hundred peptide-activated G protein-coupled receptors recognize ligands with turn structure. Chem. Rev. 105(3): 793-826.

70. WENDERFER, SE, WANG, HY, KE, BZ, WETSEL, RA, BRAUN, MC (2009) C3a receptor deficiency accelerates the onset of renal injury in the MRL/Ipr mouse. Mol. Immunol. 46(7): 1397-1404.

71. WILKEN, HC, GOTZE, O, WERFEL, T, ZWIRNER, J (1999) C3a(desArg) does not bind to and signal through the human C3a receptor. Immunology Letters 67(2): 141-145.

72. ZWIRNER, J, GOTZE, O, MOSER, A, SIEBER, A, BEGEMANN, G, KAPP, A, ELSNER, J, WERFEL, T (1997) Blood- and skin-derived monocytes/macrophages respond to C3a but not to C3a(desArg) with a transient release of calcium via a pertussis toxin-sensitive signal transduction pathway. European Journal of Immunology 27(9): 2317-2322.

155 CHAPTER 3 Non-peptide agonists for C3a receptor

156 CHAPTER 4 Towards C3a receptor antagonists

CHAPTER 4 TOWARDS C3a RECEPTOR ANTAGONISTS

4.0 ABSTRACT...... 158 4.1 INTRODUCTION...... 160 4.2 AIMS/OBJECTIVES...... 162 4.3 RESULTS ...... 162

4.3.1 STRUCTURAL MODIFICATIONS OF OXAZOLE-CONTAINING C3aR AGONISTS...... 162 4.3.1.1 Incorporation of a Phenyl Group at the Fifth Position of the Oxazole Ring...... 164 4.3.1.2 Receptor Selectivity: Competitive Binding with [125I]-C5a...... 168 4.3.1.3 Receptor Selectivity: Desensitisation...... 169 4.3.1.4 Structure-Activity Relationships (SAR) ...... 171

4.3.2 STRUCTURAL MODIFICATION OF C3aR ANTAGONIST: SB290157 ...... 173 4.3.2.1 Series I: Linker Alkylated SB290157 Analogues...... 174 4.3.2.2 Series II: R-Glycine-Arginine Tripeptide/Dipeptide Surrogates...... 176 4.3.2.3 Series III : Cinnamoyl-Glycine-Arginine Tripeptide Surrogates...... 178 4.3.2.4 Series IV: R-Tryptophan-Arginine Tripeptide Surrogates...... 180 4.3.2.5 Series V: Substitution of Furan at the Linker Gives Predominant Agonists ...... 183 4.3.2.5.1 C3aR Agonists Containing a Heterocycle in the Linker...... 185 4.3.2.5.2 C3aR Antagonists Containing Heterocycles...... 186 4.3.2.5.3 Selectivity of Heterocycle-Containing Ligands for C5aR...... 187 4.3.2.5.4 Receptor Selectivity: Desensitisation...... 188 4.4 DISCUSSION ...... 190 4.5 REFERENCES...... 192

157 CHAPTER 4 Towards C3a receptor antagonists

4.0 ABSTRACT

C3a has been implicated in many physiological processes and in the pathogenesis of many disease conditions. Therefore, effective antagonists of C3aR are predicted to have therapeutic potential for treating inflammatory diseases such as anaphylaxis, adult respiratory distress syndrome (ARDS), acute transplant rejection, ischemia reperfusion injury, sepsis (Drouin, 2001), brain inflammation and many other autoimmune diseases for example psoriasis, rheumatoid arthritis, and systemic lupus erythematosus (SLE)(Clancy, 2000; Volanakis et al., 1998). In the previous chapter, agonists were developed by using a molecular modelling study, which suggested that the oxazole–arginine dipeptide surrogate might have a similar turn conformation to that for the last four C-terminal residues of C3a, namely GLAR. The turn conformation has been observed for ligands that bind to many other GPCRs (Madala et al., 2010; Tyndall et al., 2005). By modification of the oxazole scaffold with different hydrophobic groups (Figure 4.1 A), we found a more potent C3aR agonist (56) that we used in this study. The incorporation of a phenyl group at the 5-position of the oxazole ring gave an improvement in binding affinity and eventually led to the novel and potent nonpeptidic C3aR antagonist compound (105). Further study by modification of the Boc group at the N-terminus of compound 105 also generated another potent C3aR antagonist compound (109). A second approach to developing C3aR antagonists was to use the known, albeit weak, antagonist compound SB290157 as a lead compound for further modifications (Figure 4.1 B). SB290157 was the first reported “potent” C3a receptor antagonist in several in vitro assays and two in vivo models. In 2005, SB290157 was reported to also have “potent” agonist activities in a variety of in vitro and in vivo assays. We proposed to develop new non-peptidic antagonists using SB290157 as a lead compound, because it binds tightly to C3aR and, in our hands, it showed no agonist activity in the Ca2+ assay on human monocyte derived macrophages. Moreover, it also showed no cross reactivity at, or binding to, C5aR. The SB290157 molecule can be considered to comprise three different regions (a hydrophobic N-terminus, a linker, and an arginine at the C- terminus). Our development of new antagonists focussed on modifying each of these components separately. We discovered several potent C3aR antagonists, which bind tightly to C3aR and are equipotent with C3a itself in calcium mobilisation assays in HMDM cells. These compounds feature a 5-phenyl oxazole ring or derivations of SB290157 with linker region replaced by a furan ring. The heterocycle was modified with varying hydrogen bond acceptors. The most potent and selective compounds in this study were 105, 109 and 145 and 146 (IC50 17, 14, 460 and 534 nM

158 CHAPTER 4 Towards C3a receptor antagonists respectively) in the calcium mobilisation assays in HMDM cells. Binding affinity correlated with antagonist potency, and our compounds are by far the most potent C3aR antagonists known.

159 CHAPTER 4 Towards C3a receptor antagonists

4.1 INTRODUCTION

2 In 2001, N -[(2,2-diphenylethoxy)acetyl]-L-arginine or SB290157 (Figure 4.2 A) was developed as a C3aR antagonist (Ames et al., 2001). SB290157 had IC50 = 200 nM in competitive radioligand binding assays in rat basophilic leukemia cells (RBL), and it was also able to inhibit

C3a-induced calcium mobilization in C3aR expressing RBL cells and human neutrophils (IC50 28 nM), C3a-induced receptor internalization in human neutrophils, C3a-mediated ATP release from guinea pig platelets, and C3a-induced potentiation of the contractile response to field stimulation of perfuse rat caudal artery. There were a number of in vivo studies reported in which SB290157 was used, such as guinea pig LPS-induced lung neutrophilia (Ames et al., 2001) and a rat adjuvant- induced arthritis model (Ames et al., 2001). Another C3aR antagonist was reported in 2001 - a 1,3- diiminoisoindoline (Figure 4.2 B) with an IC50 of 1.7 µM in the binding assay and an IC50 of 2.0 µM in the calcium mobilisation functional assay (Grant et al., 2001).

R AA O O O NH N HN CO H R 2

HN NH2

N-Acyl Oxazole Arginine Amino acid

(A) N-Acyl amino acid-oxazole- (B) SB290157 (R=H) arginine Figure 4.1 (A) oxazole-containing C3aR modulators. (B) SB290157 is N2-[(2,2- diphenylethoxy)acetyl]-L-arginine which has three different regions: (i) a hydrophobic cap (ii) a linker region and (iii) an N-terminal arginine.

In 2005, (Mathieu et al.) it was reported that SB290157 had potent agonist activities in a variety of assays involving different cell types that overexpressed C3aR; for example, transfected rat basophilic leukemia (RBL) cells, CHO-NFAT-bla-Gα16 (chinese hamster ovary cells) and differentiated human U937 cells. These agonist activities were reduced in guinea pig platelets, cells which are known to express low levels of C3aR (albeit as two receptor sub-types). It was suggested that SB290157 was a partial agonist. In another pharmacological study in rat intestinal ischemia/reperfusion injury, the effects of SB290157 could also not be attributed to antagonism of C3aR (Proctor et al., 2004).

160 CHAPTER 4 Towards C3a receptor antagonists The pharmacology Nobel laureate James Black (1989) stated that partial agonists can often be partial antagonists and may contain the same structural components that can be modified to give full antagonists without any agonist activity. Thus, we chose SB290157 as a lead compound and modified its hydrophobic and linker regions while retaining the C-terminal arginine to produce new compounds for these studies. Denonne and colleagues (2007a) conducted a study of different non-peptidic compounds incorporating either a pyridine moiety or an amino-piperidine linker in place of the arginine residue

(Figure 4.2 C pIC50 = 7.5), which was suspected to cause the poor bioavailability associated with SB290157. Numerous analogues were prepared to improve the binding affinity, however all analogues were found to have full agonist activity in functional assays.

In the same year, another study (Denonne et al., 2007b) reported a new compound (pIC50 7.5 vs 20 pM [125I]-C3a) with higher affinity for C3aR than SB290157. This new compound was a constrained analogue of SB290157 with a rigid furan ring and an arginine residue (Figure 4.2 D). The compound was evaluated for DMPK (drug metabolism and pharmacokinetics) properties in rats and showed low bioavailability (< 1%) attributed to poor intestinal absorption. In vivo studies using a model of airway inflammatory disease found that the compound (30 mg/kg ip) reduced levels of eosinophils in broncho-alveolar fluid and also decreased concentrations of IL-13.

H2N NH NH H N N N O O OH N H O

(A) SB290157 (B) 1,3 diiminoisoindoline IC50 = 1.7 µM in N2-[(2,2-diphenylethoxy)acetyl]-L-arginine binding assay (Grant et al., 2001). (Ames et al., 2001)

Cl O H2N NH O N N NH NH O O O OH N Cl H O

(C) Amino-piperidines without an arginine (D) Optimized derivative of SB290157 by moiety pIC50= 5.8 (Denonne et al., 2007a) using 2,5- furyl with arginine moiety 125 (Binding pIC50 = 7.2 vs 20 pM I-C3a) (Denonne et al., 2007b) Figure 4.2 Structures for non-peptidic “antagonists” of C3aR reported in the literature. 161 CHAPTER 4 Towards C3a receptor antagonists

One of the difficulties in evaluating the physiological/pharmacological roles of C3a is the lack of potency and selectivity of C3aR agonists and/or antagonists. We proposed to develop new non-peptidic antagonists based either on (a) the potent agonist oxazole-containing compound (56) from Chapter 3, or (b) the SB290157 compound, which has antagonist activity at least in the Ca2+ assay against C3aR.

4.2 AIMS/OBJECTIVES

The objectives of this Chapter were to investigate the activity of new ligands designed from (a) the potent oxazole-containing agonist (56) described in Chapter 3, and (b) the known C3aR antagonist SB290157. The competitive radioligand binding assay with 125I-C3a showed that SB290157 binds tightly to C3aR without any detectable cross reactivity with C5aR (Chapter 2). It also demonstrated no agonist activity in functional (Ca2+ release) assays on human PBMCs.

The oxazole-containing compound (56) was modified by incorporating 5-phenyl substituted oxazole derivatives and different hydrophobic groups at the N-terminus. The SB290157 compound can be described in terms of three different regions: (i) an N-terminal hydrophobic/aromatic cap, (ii) a linker region and (iii) a C-terminal arginine. Because the arginine residue at the C-terminal position of the native peptide agonist C3a is important for activating C3aR (although probably a binding residue), only the first two regions were modified in this chapter. New antagonists were designed from SB290157 by modifying the N-terminal hydrophobic cap or by using differently constrained heterocyclic linkers incorporating different hydrogen-bond acceptors. All compounds were supplied by chemists in the Fairlie Group, especially Mr Anthony Reed (Hons 2009), Ms Peifei Chu (Hons 2010) and Dr. Robert Reid. Structure-activity relationships (SAR) for these compounds were determined by me using radioligand binding and a functional assay. [125I]-C3a and [125I]-C5a were used to measure receptor binding affinity of new ligands on whole human monocyte-derived macrophages, and an intracellular calcium mobilisation assay was used to probe function.

4.3 RESULTS 4.3.1 Structural Modifications of Oxazole-Containing C3aR Agonists

In Chapter 3, a new series of potent C3a receptor agonists were derived by using the oxazole scaffold that showed a turn conformation in molecular modelling of compound 56 (N-acyl-

162 CHAPTER 4 Towards C3a receptor antagonists leucine-oxazole-arginine scaffold). This is shown (Chapter 3, Figure 3.2) superimposed on the C- terminal four residues of hC3a, namely GLAR. Further investigation in this present chapter has been concentrated initially on a small library of compounds (80-104), in which the fifth position of the oxazole ring incorporated a methyl substituent to mimic the location of the side chain methyl group of the alanine residue in the second last C-terminal position of C3a (Figure 4.3 and 4.4). However, there was no improvement in the binding affinity, and only a decrease in potency in the calcium mobilisation assay (Table 3.2 and Table 3.3).

Figure 4.3 Overlay of 56 (cyan) and GLAR (sequence corresponding to the C-terminus of C3a, namely C3a73-76, green) reveals the possibility of incorporation the functional group into the 5th position of oxazole ring. The location of the side chain methyl group of the alanine residue in C3a is boxed in red (courtesy of Dr. Robert Reid).

O O O NH N O HN CO2H

HN NH2 56 80

Figure 4.4 Modification of the oxazole scaffold by incorporating a 5-methyl substituent. Boc-leucine-oxazole-arginine (56) and Acyl-leucine-5-methyl-oxazole-arginine (80). 163 CHAPTER 4 Towards C3a receptor antagonists

We therefore next decided to add bulk at the fifth position by incorporating a 5-phenyl substituent on the oxazole ring of 56 to see if that increased or decreased receptor binding affinity or function (Figure 4.5). That analogue of 56, namely compound 105, was tested in the competitive binding assay and in calcium mobilisation in human monocyte-derived macrophages.

4.3.1.1 Incorporation of a Phenyl Group at the Fifth Position of the Oxazole Ring.

The first compound in this study (105, Figure 4.5) was modified from the potent compound 56 (Boc-leucine-oxazole-arginine) in Chapter 3.

O O O O NH N O O NH N O HN CO2H O HN CO2H

NH NH HN NH2 HN NH2 56 105

Figure 4.5 Modification of the oxazole scaffold by incorporating a 5-phenyl substituent. Boc-leucine-oxazole-arginine (56) and Boc-leucine-5-phenyl-oxazole-arginine (105).

From the results for binding affinity, compound 105 was shown to bind tightly to hC3aR with IC50 ± 95% = 37 ± 16 nM (Table 4.1, Figure 4.6). However, there was no improvement in binding affinity over the methyl-substituted compound (80)(Figure 4.4), which had IC50 ± 95% = 15 ± 5 nM, indicating that the 5-phenyl substituent is not more effective in filling out space in the receptor that the methyl group is projecting into. This suggests that even larger substituents might be accommodated at this position with improved receptor affinity.

164 CHAPTER 4 Towards C3a receptor antagonists

Table 4.1 SAR of acyl-leucine-5-phenyl-oxazole-arginine (105) comparison to boc-leucine- oxazole-arginine (56) and acyl-leucine-5-methyl-oxazole-arginine (80) on competitive binding with [125I]-C3a and Ca2+ mobilisation from HMDM cells.

Receptor Binding Apparent Agonist Apparent Affinity Activity Antagonist Activity

Agonist/ n pIC50 ± SE IC50 n pEC50 ± SE EC50 n pIC50 ± SE IC50 Antagonist (nM) (nM) (nM) hC3a 12 9.64 ± 0.04 0.23 3 7.2 ± 0.06 65 - - - SB290157 11 7.42 ± 0.06 38 4 * * 9 5.6 ± 0.09 2700 56 3 7.71± 0.06é 20 4 7.7 ± 0.27Ê 19 - - - 80 3 7.86± 0.08é 15 2 7.4 ± 0.18 45 - - - 105 4 7.43± 0.10Î# 37 4 * * 4 7.8 ± 0.17é 16.5 The compounds show no agonist activity in calcium mobilisation assay up to 100 µM. é p < 0.05 vs. SB290157; Î p < 0.05 vs. 56; # p < 0.05 vs. 80 and * p < 0.05 vs. hC3a.

120 g

n 100 i hC3a (n=3) a d 3

n SB290157 (n=3)

i 80 C - b

] PC_leu (n=4) I c 5 i 60 2 f BR3924_64 (n=3) i 1 [ c

f e 40 TR4401-12 (n=3) o p

S 20 % 0 -14 -12 -10 -8 -6 -4 -2

log [Ligand](M) Figure 4.6 Competition with [125I]-C3a for binding to human monocyte-derived macrophages: hC3a (l), SB290157 (n), 56 (u), 80 (q) and 105 (p); n≥3. Human monocyte-derived macrophages (1.5x106 cell/mL) were incubated for 1 h at room temperature (22°C) with 80 pM of [125I]-C3a and increasing concentrations of unlabelled C3a (concentration 1 pM to 1 µM) or non-peptides (concentration = 0.1 nM to 100 µM). Each point represented the mean percentage specific binding ± S.E.M.

165 CHAPTER 4 Towards C3a receptor antagonists A new series of compounds was next investigated in which a phenyl group was maintained at the 5-position of the oxazole ring and coupled with a hydrophobic group in place of the Boc group. Similar to the previous study in Chapter 3 (Acyl-leucine-5-methyl-oxazole-arginine compounds), the replacement of the Boc group with different hydrophobic groups in compound 105 was thought to be potentially capable of improving binding affinities of the compounds for the human C3a receptor.

Table 4.2 SAR data of Acyl-Leucine-5-Phenyl-Oxazole-Arginine series.

O O O NH N R HN COOH

HN NH2

Compound R %[125I]-C3a bindinga % Agonistb 200 nM 20 µM Br 106 65 25 78 N

107 58 66 N/A F S 108 74 50 N/A

MeO 109 50 31 18

110 61 45 23

a: % Binding of [125I]-C3a at compound concentrations of 200 nM and 20 µM; compared to the binding of [125I]-C3a with cells in the absence of compound (or Buffer) as 100%; n = 2. b: % Response to 100 nM C3a activity. The compounds tested at [10 µM] and the response of 100 nM hC3a is equal to 100% response. n= 2.

166 CHAPTER 4 Towards C3a receptor antagonists

From this study of competitive binding on HMDM cells, we found that compounds 106 and 109 bound most tightly to the receptor in this series (Table 4.2). In the single point assay, at 200 nM 109 had an affinity for the receptor of approximately 50% compared to SB290157 which had an affinity of 39%. At the concentration 20 µM, 109 had an affinity around two fold lower than for SB290157 (12%). Compounds 108 and 110 have benzothianaphthene and cyclohexane groups respectively, which were also used in Chapter 3 to study 5-methyl-oxazole-arginine agonists and led to better affinity for the receptor in the series of 5-phenyl-oxazole-arginines here. In addition compound 107, which has a para fluoro substituted aromatic ring, that can act as a hydrogen bond acceptor, has poor affinity for the receptor. Compound 106, which has the bulkier substituent (Br), binds tightly to the receptor at the higher concentration, however in the calcium mobilisation assay the compound caused stimulation (78%) of calcium release from HMDM cells. In contrast, compounds 109 and 110, which have oxygen on the methoxy (4-methoxybenzoyl substituent) and a cyclohexane group respectively, showed higher affinity and lower agonist activity in the calcium release assay. Compound 109 was chosen for further study in the full dose response in the competitive binding assay (Figure 4.7) and calcium mobilisation assay because of the higher binding affinity to the receptor and the lowest agonist activity of the compound.

109 showed high affinity for the receptor with IC50 ± 95% = 18 ± 7 nM (Table 4.3) while the compound 105 had IC50 ± 95% = 38 ± 16 nM. This result suggests that replacement of the Boc group in compound 105 with the hydrophobic ring, and an oxygen to act as a hydrogen bond acceptor, leads to improved binding affinity for the human C3a receptor on HMDMs.

Table 4.3 Competitive binding of 109 with [125I]-C3a and Ca2+ mobilisation from human derived macrophage cells.

Receptor Binding Apparent Apparent Affinity Agonist Antagonist Activity Activity

Agonist/ n pIC50 ± SE IC50 n pEC50 ± SE EC50 n pIC50 ± SE IC50 Antagonist (nM) (nM) (nM) hC3a 12 9.64 ± 0.04 0.23 3 7.2 ± 0.06 65 - - - SB290157 11 7.42 ± 0.06 38 7 * * 9 5.6 ± 0.09 2700 109 3 7.75 ± 0.09é 18 4 * * 4 7.9 ± 0.14é 14 *The compounds show no agonist activity in calcium mobilisation assay up to 100 µM. é p < 0.05 compare with SB290157.

167 CHAPTER 4 Towards C3a receptor antagonists

100 hC3a (n=12) g n

i SB290157 (n=11) a d 75 3 n PC-leu (n=4) i C - b

]

I PC_31 c 5 i 50 2 f i 1 [ c

f e o p

25 S

% 0 -14 -12 -10 -8 -6 -4 -2

Log [Ligand](M)

Figure 4.7 Receptor affinities for C3aR measured by displacement of 80 pM [125I]-C3a from HMDM cells: hC3a (l), SB290157 (n), 109 (l) and 105 (p) n≥3. Human monocyte- derived macrophage cells (1.5x106 cell/mL) were incubated for 1 h at room temperature with 80 pM of [125I]-C3a and increasing concentration of non-labelled C3a (concentration 1 pM to 1 µM) and non-labelled of 105 and 109 (concentration 0.1 nM to 100 µM). Each point represented the mean percentage specific binding ± S.E.M.(n ≥ 3).

4.3.1.2 Receptor Selectivity: Competitive Binding with [125I]-C5a.

A limited receptor selectivity study was performed by using [125I]-C5a radioligand competitive binding to evaluate whether compounds bound selectively to C3aR over C5aR. Firstly, all the compounds in this series were tested at a single concentration (20 µM) for competition with radioligand labelled C5a (20 pM) for human monocyte-derived macrophages. The binding of all compounds was compared with 3D53, which has been shown to selectively bind C5aR but not C3aR on HMDMs (Chapter 3). From the single concentration screening results shown (Figure 4.8 A), none of the compounds bind to C5aR, which suggests that all agonist/antagonist ligands tested were selective for C3aR over C5aR (Figure 4.8 A). The two compounds with highest affinity in the C3aR binding assay, 105 and 109, were chosen for further study in a full concentration-response curve in the same assay (Figure 4.8 B).

168 CHAPTER 4 Towards C3a receptor antagonists

5000

) 4000 m p c (

3000 a 5 C

- 2000 ] I 5 2

1 1000 [

0 r 3 5 6 7 8 9 0 ffe 10 10 10 10 10 11 u 3D5 B [compounds 20 M] µ B.

5000

) 4000 m

p hC5a c ( 3000 3D53 a 5 PC_leu C

- 2000 ]

I BRPC_31 5 2 1

[ 1000

0 -14 -12 -10 -8 -6 -4 -2 0

Log [Ligand](M) Figure 4.8 A. Competitive binding versus [125I]-C5a (20 pM) for C3aR-binding non-peptide ligands (20 µM) in human monocyte-derived macrophages. B. Competitive concentration- dependent binding of hC5a (l), 3D53 (n), 105 (p) and 109 (q) versus [125I]-C5a binding affinity on HMDM. This shows that 105 and 109 have no binding for C5aR. Each point represented the mean percentage specific binding ± S.E.M. with n≥ 3.

4.3.1.3 Receptor Selectivity: Desensitisation

Another method for qualitatively estimating receptor selectivity is based on receptor desensitisation. Administration of 3 µM hC3a to isolated human macrophages produces an efflux of intracellular calcium release from cells. The basic concept of the calcium mobilisation is to detect the change in fluorescence that occurs when the cells release intracellular calcium from the endoplasmic reticulum after stimulation of the cells with an agonist. Fluo-3AM was used as a cell- permeable calcium indicator and it needs to pass through the cell membrane with the aid of a mild 169 CHAPTER 4 Towards C3a receptor antagonists detergent, pluoronic acid, then esterase enzymes in cells cleave the Fluo-3AM into Fluo-3. Fluo-3 will bind to the intracellular calcium to become the Fluo-3 calcium complex and it can be detected when excited by 485 nm light and emits 520 nm light. After stimulating the HMDM cells using 3 µM hC3a, which causes efflux of calcium that dissipates after 5-6 mins, then a second addition of 3 µM C3a has no effect due to the desensitisation of cell surface C3aR (Figure 4.9 A). A similar result was found when the second addition was either compound 105 (Figure 4.9 B) or 109 (Figure 4.9 C) at 10 µM instead of 3 µM C3a, which suggests that these compounds are selective for C3aR and do not activate some other receptor linked to intracellular Ca2+ release in HMDMs.

(A.) (B.)

3µM C3a 3µM C3a ) ) 0 0 2 2 100 100 5 5 a a 3 3 M M E C E C

( 80

M M e e µ µ c c 3 3 n n 60

e e y y c c b b

s s d d 40 40 e e r e r e o c o c 10µM 105 u u u u 3 M C3a l l 20 µ 20 d d F F

n n i i % % 0 0

0 60 120 180 240 300 360 0 60 120 180 240 300 360 Time (sec) Time (sec)

(C.)

3µM C3a ) 0

2 100 5 a 3 M E C

( 80

M e µ c 3 n 60

e y c b

d 40 e r e o c 10µM 109 u u l 20 d F

n i

% 0

0 60 120 180 240 300 360 Time (sec)

Figure 4.9 C3a receptor desensitisation in the calcium mobilisation assay on HMDM for 105 (B) and 109 (C). HMDM were first treated with 3 µM C3a, which induced intracellular calcium release that was allowed to dissipate over 310 seconds prior to a second agonist injection.

170 CHAPTER 4 Towards C3a receptor antagonists

4.3.1.4 Structure-Activity Relationships (SAR)

All the compounds in this series were tested for agonist activity in a single point assay using the intracellular calcium mobilisation assay. In leukocytes such as eosinophils, neutrophils and basophils, C3a is known to induce calcium mobilisation and degranulation (Elsner et al., 1994; Nagata et al., 1987; Norgauer et al., 1993; Takafuji et al., 1994). These effects are mediated by coupling to pertussis toxin sensitive G-protein (Gαi) (Norgauer et al., 1993). Here our new compounds were tested at 10 µM and the results of the agonist response are compared to 100 nM hC3a responses under the same conditions (Figure 4.10). From Table 4.2, compound 106 showed good agonist activity comparable to that of hC3a. In contrast, compound 109 and 110 showed less agonist activity when tested at 10 µM. These compounds have lower binding affinities for C3aR than 106 at 20 µM.

A. Agonism B. Antagonism

100 a 3 C

e 80 c M n n e 0

c 60 0 s 1

e r o t o 40 u e l s F n % o

p 20 s e r 0 a 7 5 6 9 0 C3 10 10 10 11 h B29015 S [compound 10 µM]

Figure 4.10 Single point calcium release induced by Acyl-Leucine-5-Phenyl-Oxazole- Arginine compounds. A. 105, 106, 109 and 110 were tested at 10 µM in calcium mobilisation assay for agonist activity. hC3a (100 nM) and SB290157 (10 µM) were used as reference compounds. B. compound 107, 108, 110 and 109 were tested at 1 µM and 10 µM for antagonist activity. hC3a was used to induce calcium release after incubating HMDM cells with compounds for 15 mins. Inhibition expressed as a percentage of maximum change in fluorescence induced by 100 nM hC3a. Error bars represented S.E.M. with n≥ 2.

171 CHAPTER 4 Towards C3a receptor antagonists The single point calcium mobilisation data provided quick results for comparison. Compound 109 had the best antagonist activity, better than for the reference compound SB290157 (Figure 4.10 B). Moreover, 109 also gave the least agonist activity in the agonist activity screening results (Figure 4.10 A). Additionally, concentration dependence studies of calcium mobilisation were performed for compounds 105 and 109.

The antagonist activity of SB290157 in this study was IC50 = 2.9 ± 1.1 µM against 100 nM hC3a on isolated human derived macrophage cells, while compound 105 had IC50 22 ± 15 nM and compound 109 had IC50 17 ± 10 nM (Figure 4.11). These results suggest that compounds 105 and 109 are potent antagonists for C3aR, much more potent than SB290157 in inhibiting C3a induced calcium release from HMDM cells.

)

0 120 2 a 5 3 m C

100 E ( M SB290157 (n=2) e n

c 80 0

0 PC_leu (n=4) n 1 e

c 60 PC_31(n=4) y s b e

r d

o 40 e u c l u F

20 d n

% 0 -12 -10 -8 -6 -4 -2

Log [Antagonist] (M) Figure 4.11 Concentration dependent inhibition of SB290157 (l), 105 (n) and 109 (p). Various concentrations of SB290157, 105 and 109 were incubated with HMDMs 15 mins prior to the experiment. The cells were then challenged by 100 nM hC3a. Differences in fluorescence induced by intracellular calcium mobilisation were measured and expressed as percentage of response induced by 100 nM hC3a on untreated HMDMs ± S.E.M. Results were plotted to corresponding concentrations by using prism 5.0. All the compounds were tested with n ≥ 3.

In conclusion, substitution by a phenyl group at the fifth position of the oxazole ring has turned C3aR agonists into antagonists, and bulkier substitution at the N-terminus also improved potency. From this study, we discovered two potent new C3aR antagonists with potency at nanomolar concentrations, about 100 fold more potent as antagonists than SB290157. In addition, there was a correlation between the binding affinity and calcium mobilisation assay of the compounds in this study (Figure 4.7 and 4.11). Compound 109 (IC50 18 nM) bound tightly to

172 CHAPTER 4 Towards C3a receptor antagonists C3aR, possibly more tightly than 105 (37 nM) and SB290157 (38 nM), and it also showed the highest agonist potency for inhibiting calcium release after stimulating cells with 100 nM hC3a. The next approach will be the modification of SB290157, the first non-peptidic ligand that was reported as a C3aR antagonist and showed some affinity for C3aR but not C5aR.

4.3.2 Structural Modification of C3aR Antagonist: SB290157

SB290157 was first reported by Ames et al. (2001) as a potent C3a receptor antagonist, which showed antagonist activity in several in vitro assays and in vivo studies, including calcium mobilisation in RBL and neutrophils, LPS-induced lung neutrophilia model in guinea pig and an adjuvant–induced arthritis rat model. This study also showed that SB290157 binds tightly to C3aR, with IC50 200 nM in a competitive radioligand binding assay using RBL-2H3 cells. The results of Chapter 2 also showed that SB290157 had no cross reactivity with C5a in competitive radioligand binding and desensitisation assays. Taking these considerations in mind, we decided to use SB290157 as a lead compound and focussed on optimising the different components of SB290157. SB290157 can be regarded as having three different regions: (i) a hydrophobic diphenylmethane moiety, (ii) an aliphatic linker region and (iii) a C-terminal arginine residue (Figure 4.12). The C-terminal arginine of new ligands was retained to maintain the potency of the ligands. It has been shown that the arginine at the C-terminus is important for recognition (binding) and thereby maintaining high agonist activity for C3a and all potent C3a peptide agonists. Furthermore, the cleavage of C3a at the arginine position by carboxypeptidases N leads to the complete loss of biological activity because C3a-des Arg does not bind to C3aR (Wetsel et al., 2000). The modification of the arginine residue by replacing with arginine mimetics may improve bioavailability, solubility and, metabolic stability of the compounds (Denonne et al., 2007b) and might be part of future studies. The structure of SB290157 has been modified in five different ways: (1) the modification of the linker region while leaving unchanged the diphenylmethylene rings and arginine moieties, (2) the modification of the aromatic and linker regions where the ether bond and the linker are replaced by an amide functional group (R-glycine-arginine scaffold), (3) the aromatic region can also be replaced by cinnamic acids and derivatives while the ether bond and the linker region are replaced with an amide group (cinnamic acids-glycine-arginine), (4) the linker region has also been investigated by introducing a tryptophan residue and analogues with different substituents at the aromatic region, and finally (5) the linker has been replaced by furan moieties, including functionalised analogues.

173 CHAPTER 4 Towards C3a receptor antagonists

H2N NH ether NH

O OH N H O R

Hydrophobic cap N -Terminal Arginine Linker

Figure 4.12 C3aR antagonist SB290157.

4.3.2.1 Series I: Linker Alkylated SB290157 Analogues.

Initially, the methylene group located in the linker component of SB290157 was functionalised with different alkyl groups to explore whether the substitution of the space filling group might improve the affinity or the potency of the new family of the compounds generated (Table 4.4). Competitive radio-ligand binding assays have been used for affinity measurement and the calcium mobilisation assay for a functional test, both using human monocyte-derived macrophages to derive the structure-activity relationships. In the competitive radioligand binding assay the new agonist candidates were tested in competition with 80 pM of [125I]-C3a (2200 Ci/mmol, Perkin Elmer) or 20 pM of [125I]-C5a (2200 Ci/mmol, Perkin Elmer) for C3aR or C5aR binding on HMDM cells. Unlabelled compounds with high affinity for C3aR compete with labelled C3a for binding to C3aR, and this was determined by scintillation counting on a Wallac Beta scintillation counter. The non-specific binding for C3aR was determined by using at 1 µM C3a in competitive binding of compounds against [125I]-C3a or at 300 nM of C5a in competitive binding of compounds against [125I]-C5a . The calcium mobilisation assay was performed on HMDM cells using Fluo-3 as a calcium indicator. C3aR on myeloid cells is coupled to Gαi2/3 (Mousli et al., 1992). The activation of C3aR induces the release of calcium from the endoplasmic reticulum through phospholipase C(β). The calcium released into the cytoplasm was detected by the cell-permeable calcium binding fluorophore (e.g. Fluo-3). The fluorescence is then measured with a spectrofluorimeter.

174 CHAPTER 4 Towards C3a receptor antagonists The results of affinity and functional assays of the compounds have been compared to SB290157 as a control.

Table 4.4 SAR data for series I: Alkylated SB290157 analogues.

Com [125I]-C3a Binding Ca2+ release assay -pound Structure n pIC50 ± SE IC50 %a n pIC50 ±SE IC50 %b (nM) (nM)

SB H 11 7.42 ± 0.06 38 6 7 5.52 ± 0.08 3050 53 N CO H 290157 O 2 O

HN NH2 111 H - 50 - 86 N CO H O 2 O

HN NH2 112 H - 57 - 100 N CO H O 2 O

HN NH2 113 2 5.39 ± 0.35 4000 21 2 5.27 ± 0.17 5320 87

H N CO H O 2 O

HN NH2 a = % of binding of [125I]-C3a at compound concentration of 20 µM; compared to the binding of [125I]-C3a (80 pM) with cells in the absence of the compound (or Buffer) is 100% or in the presence of hC3a (1 µM) is 0%. Thus SB290157 had the best binding affinity to C3aR in this series. b = % of response to 100 nM hC3a activity (Antagonism test) . The compounds were tested at [10 µM] which were incubated with the cells and then tested with 100 nM of hC3a, giving a buffer response of 100%. Thus SB compound was the best antagonist.

All the compounds were tested in a C3a binding assay first. The results suggest that the substitution of the methylene moiety between the ether and the carbonyl groups of SB290157 leads to decreased affinity for C3aR binding (-hydrogen > -butyl > -methyl ≈ -ethyl). As for the affinity assay, the compounds were tested for function at just one concentration (10 µM) and reported as a percentage of the calcium release observed for hC3a at 100 nM. The results show that 112 is not an antagonist, while 111 and 113 show some inhibition of calcium efflux, which was stimulated by 100 nM of hC3a. Further examination of 113 in a full concentration response range gave IC50 5.3 µM that is higher than for SB290157 (IC50 ∼3 µM). Affinity of 113 at the single point concentration of 20 µM was similar to that found for SB290157 (Figure 4.16 A). 175 CHAPTER 4 Towards C3a receptor antagonists

4.3.2.2 Series II: R-Glycine-Arginine Tripeptide/Dipeptide Surrogates.

Previous studies conducted by Dr. Jade Blakeney (PhD dissertation, University of Queensland, 2007) showed that substitution of the ether group in SB290157 with a planar amide lead to active C3aR ligands. In that series, the compounds were developed as tripeptides and dipeptide surrogates. The replacement of the ether group in SB290157 by an amide bond, which has partial double bond character and hence imparts a more rigid conformation to the compounds (Figure 4.13), could potentially lead to improved binding affinity of new analogues.

H N NH2

NH O H R N OH N H O O

Amino acid Glycine Arginine surrogate

Figure 4.13 R-glycine-arginine tripeptide/dipeptide surrogates. The ether bond was substituted by an amide functional group, which gave more rigidity to the molecule.

Table 4.5 SAR data for series II: R-Glycine-arginine tripeptide/dipeptide surrogates.

Compound R [125I]-C3a Binding Ca2+ release assay assay %a %b 114 68 83

115 100 75

116 100 64

117 100 78

176 CHAPTER 4 Towards C3a receptor antagonists 118 98 72

119 88 84

120 100 96

121 77 97

122 59 N/A

123 53 N/A

124 100 N/A

125 100 N/A

126 88 N/A

127 100 N/A

128 71 N/A

129 65 N/A

130 93 N/A

131 46 N/A

a = % of binding of [125I]-C3a at compound concentration of 20 µM; compared to the binding of [125I]-C3a (80 pM) with cells in the absence of the compound (or Buffer) is 100% or in the presence of hC3a (1 µM) is 0%. Thus compound 131 was the best binding affinity to C3aR on the cells in this series. b = % of response to 100 nM hC3a activity (Antagonism test) . The compounds were tested at [10 µM] which were incubated with the cells and then tested with 100 nM of hC3a, giving a buffer response of 100%. Thus compound 116 was the best antagonist in this series. N/A not tested. 177 CHAPTER 4 Towards C3a receptor antagonists

Table 4.6 R-Glycine-arginine tripeptide surrogates as candidate C3aR ligand.

Com- R [125I]-C3a Binding assay Ca2+ release assay pound

n pIC50±SE IC50 %a n pEC50± SE EC50 %b (nM) (nM) 132 2 5.65 ± 0.35 2200 55 - - - -

a = % of binding of [125I]-C3a at compound concentration of 20 µM; compared to the binding of [125I]-C3a (80 pM) with cells in the absence of the compound (or Buffer) is 100% or in the presence of hC3a (1 µM) is 0%. b = % of response to 100 nM hC3a activity (Antagonism test) . The compounds were tested at [10 µM] which were incubated with the cells and then tested with 100 nM of hC3a, giving a buffer response of 100%.

Compounds 114 – 132, having the glycine-arginine unit and other functional groups R in the hydrophobic region, showed significantly lower binding affinity for C3aR compared to SB290157 (Table 4.5 and 4.6). Compound 122 is the amide analogue of SB290157, which bound poorly indicating that replacement of the ether bond with an amide reduced the binding affinity of the compound to C3aR. Compound 123, with the 2-naphthoic acid substituent and no glycine residue, showed a slight improvement in binding affinity compared to 122. Compound 131, with a 4- biphenylacetyl fragment connected to the amide bond had the best affinity of this series of analogues. This could be attributed to the lack of the glycine moiety as linker fragment and amide bond at the aromatic region of the molecule.

4.3.2.3 Series III : Cinnamoyl-Glycine-Arginine Tripeptide Surrogates.

Substitution of the hydrophobic fragment R on R-glycine-arginine by cinnamoyl and derivatives has been investigated. The cinnamic acids were coupled to glycine-arginine in this series (Figure 4.14). The new family of compounds generated showed less affinity for C3a receptor in the competitive radioligand binding assay (Figure 4.16 B) and also less antagonist potency in the calcium mobilisation assay compared to SB290157 (Table 4.7). The benzophenone derivative 137 was completely inactive. The low affinity of the cinnamoyl-glycine-arginine tripeptide surrogate compounds for C3aR can be attributed to the lower flexibility of the carbon-carbon double bond linking the aromatic ring, making it difficult to correctly position large hydrophobic groups in the pocket of the receptor.

178 CHAPTER 4 Towards C3a receptor antagonists

H N NH2

NH O H R N OH N H O O

Amino acid Glycine Arginine surrogate

Figure 4.14 Cinnamoyl-glycine-arginine tripeptide surrogates.

Table 4.7 SAR data of Series III: Cinnamoyl-glycine-arginine tripeptide surrogates.

Compound R [125I]-C3a Binding assay Ca2+ release assay

%a %b

133 93 100

134 67 99

135 70 89

136 74 85

137 100 100

a = % of binding of [125I]-C3a at compound concentration of 20 µM; compared to the binding of [125I]-C3a (80 pM) with cells in the absence of the compound (or Buffer) is 100% or in the presence of hC3a (1 µM) is 0%. Thus compound 134 had the best binding affinity to C3aR on the cells in this series. b = % of response to 100 nM hC3a activity (Antagonism test) . The compounds were tested at [10 µM] which were incubated with the cells and then tested with 100 nM of hC3a, giving a buffer response of 100%. Thus compound 136 was the best antagonist in this series.

179 CHAPTER 4 Towards C3a receptor antagonists

4.3.2.4 Series IV: R-Tryptophan-Arginine Tripeptide Surrogates.

The fourth series of compounds was generated incorporating the tryptophan-arginine dipeptide unit and with various other functional groups “R” in the hydrophobic region (Figure 4.15). The substitution of the linker region with a tryptophan residue provided a potential hydrogen bond donor on the indole ring and additionally, a bulky hydrophobic structure that could fill space that may be available in the receptor. Compounds 138 to 142 were designed and synthesized by chemists in our group. Binding results of these compounds at 20 µM concentration showed that they increase their affinity for C3aR (Figure 4.16 C). However, none of these compounds was a potent antagonist in inhibiting calcium release from cells after challenge with 100 nM hC3a in the HMDM cells (Table 4.8).

Compound 138 demonstrated a partial agonist activity with an EC50 907 nM. This was the first compound to give a submicromolar agonist potency of the first four series, confirming a result reported by Blakeney in her PhD thesis (2007). The hydrophobic group R linked to the amide bond was phenylmandelic acid, consisting of a quaternary carbon with two phenyl rings and a hydroxyl group as substituents.

H N NH2

NH O H R N OH N H H O N O

Tryptophan Arginine

Figure 4.15 R-Tryptophan-arginine tripeptide surrogates.

180 CHAPTER 4 Towards C3a receptor antagonists

Table 4.8 SAR data of Series IV: R-Tryptophan-arginine tripeptide surrogates.

Com R [125I]-C3a Binding assay Ca2+ release assay Pound n pIC50±SE IC50 %a n pEC50 ±SE EC50 %b (nM) (nM) 138 - - 42 2 6.04 ± 0.17 907 -

139 - - 46 - - 90

140 2 5.93 ± 0.31 1200 35 - - 88

141 - - 100 - - 88

142 - - 57 - - 100

a = % of binding of [125I]-C3a at a compound concentration of 20 µM; compared to the binding of [125I]-C3a (80 pM) with cells in the absence of the compound (or Buffer) is 100% or in the presence of hC3a (1 µM) is 0%. Thus compound 140 was the best binding to C3aR on the cells in this series. b = % of response to 100 nM hC3a activity (Antagonism test) . The compounds were tested at [10 µM] which were incubated with the cells and then tested with 100 nM of hC3a, giving a buffer response of 100%.

181 CHAPTER 4 Towards C3a receptor antagonists

4000 )

m 3000 p c (

3 2000 C - ] I

5 2

1 1000 [

0 r 7 2 3 ffe 11 11 u B B29015 S Ligand [20 µM]

A. Linker Region Modification 5000

) 4000 m p c (

3000 a 3 C -

] 2000 I

5 2 1 [ 1000

0 r 7 4 5 6 7 8 9 0 1 3 4 5 6 7 ffe 11 11 11 11 11 11 12 12 13 13 13 13 13 u B B29015 Ligand [20 µM] S B. Glycine-Arginine Derivative 5000

) 4000 m p c (

3000 a 3 C -

] 2000 I

5 2 1 [ 1000

0 r 7 8 9 0 1 2 ffe 13 13 14 14 14 u B B29015 S Ligand [20 µM] C. Tryptophan-Arginine Derivative

Figure 4.16 Structure activity relationship studies of C3a non-peptidic agonists by substitution of SB290157 in different regions. The affinities of the ligands were tested by competitive binding with [125I]-C3a (80 pM) using C3aR non-peptide ligands at concentrations of 20 µM in human monocyte-derived macrophages. Error bars represented S.E.M. with n≥ 2.

182 CHAPTER 4 Towards C3a receptor antagonists

4.3.2.5 Series V: Substitution of Furan at the Linker Predominantly Gives Agonists

Denonne and colleagues (2007b) reported a compound which had higher affinity for C3aR 125 (pIC50 7.5 vs 20 pM [ I]-C3a) than SB290157 by optimising the structure of SB290157 with a rigid furan ring and retaining the arginine and ether oxygen that is assumed to contribute binding interactions because replacing furan at the linker with a benzene or thiophene rings which are not capable of acting as hydrogen bond acceptors showed less affinity for C3aR, with pIC50 4-6 and 5.2 vs 20 pM [125I]-C3a respectively. However, in our hands this furan compound was a C3aR agonist with relatively poor affinity for C3aR. We were interested in modifying the linker region of SB290157 by incorporating other 5- membered heterocycles but retaining the diphenylmethane unit and the arginine residue (Figure 4.17). Structure-activity relationships from such a study could provide information about potential H-bond interactions between the compounds and the C3a receptor. Initially, SB290157 and the previously reported compound (144) from Denonne and colleagues (2007b) were synthesised by our group and I measured C3a receptor binding affinities and calcium mobilisation for agonism (induced calcium release) and antagonism (inhibition of calcium release by the induction of 100 nM hC3a) in human monocyte-derived macrophages.

H2N NH Y Z NH O X O HN CO2H O OH N H R O NH

HN NH2

(a) (b)

Figure 4.17 (a) SB290157 (b) Substitution of different heterocycles in the linker region.

183 CHAPTER 4 Towards C3a receptor antagonists

Table 4.9 C3aR binding and agonism: Effect of heterocycles in the linker region.

X Y Z Receptor Apparent Binding Agonist Affinity Activity Agonist n pIC50 ± SE IC50 n pEC50 ± SE EC50 (nM) (nM) 143 N O CH 4 7.43 ± 0.10 37 4 8.2 ± 0.15 6.6 144 O CH CH 3 5.88 ± 0.11* 1300 2 6.19 ± 0.2 645 145 O N C-Me 3 6.22 ± 0.15*é 606 2 ϕ ϕ 146 S N C-Me 3 7.04 ± 0.12*éÎ 91 2 4.13 ± 2.7* 74800 ϕ compound showed no agonist activity up to 100 µM. * p < 0.05 vs. 143; é p < 0.05 vs. 144 and Î p < 0.05 vs. 145.

The results of the competitive binding studies in HMDM cells showed that SB290157 had a higher binding affinity than 144 (IC50 38 nM vs 1.3 µM respectively)(Table 4.9 and Figure 4.18).

The calcium mobilisation assay indicated that SB290157 (IC50 2.7 µM) was an antagonist without agonist activity, whereas 144 displayed partial agonist activity (EC50 645 nM) (Table 4.9). The other compounds in this series (143, 145 and 146) were synthesised by modified heterocycle ring to explore the possibility of improvement of the affinity of the compounds to the receptor by enhanced hydrogen bond acceptors. The compounds were tested in the C3a binding assay first. The results showed compound 143 had the highest affinity for C3aR, similar to

SB290157 (IC50 38 versus 37 nM, respectively). As expected, the oxazole in 143 conferred greater affinity for C3aR than the furan in 144, and we attributed this to the fact that oxazole has a much stronger H-bond accepting nitrogen (at 3rd position) than the oxygen (at 1st position) of furan. Compound 145 had around 15-fold lower affinity for C3aR than 143, but 2 fold greater than for the furan-containing 144. A drop in the binding potency of 145 may be due to the weaker hydrogen st bond interaction of oxygen (at 1 position). Compound thiazole 146 (IC50 91 nM), which had a sulfur atom at the 1st position, showed improved binding affinity compared with when an oxygen atom was at the same position (compound 145). This was surprisingly as a sulfur atom is not recognised as a good hydrogen bond acceptor, but it did improve the binding affinity of the compound, possibly because its lone pair orbitals project further into a pocket than those of oxygen.

184 CHAPTER 4 Towards C3a receptor antagonists

120

g hC3a (n=3) n 100 i a d SB290157 (n=3) 3 n

i 80 C - b BR3924_87

] I c 5 i 60 BR3924_88 2 f i 1 [ c

BR3924_90 f e 40 o p

BR3924_91 S 20 %

0 -14 -12 -10 -8 -6 -4 -2

Log [Ligand](M)

Figure 4.18 Competitive [125I]-C3a binding of hC3a (l), SB290157 (n), and non-peptidic ligands with heterocycles in the linker region 143 (q), 144 (¡), 145 (£) and 146 (U) on human monocyte derived macrophage cells. Human monocyte derived macrophage cells (1.5 x 106 cells/mL) were incubated for 1 h at room temperature with 80 pM of 125I-C3a and increasing concentration of non-labelled C3a (concentration 1 pM to 1 µM) and non-labelled non-peptidic ligands (concentration 0.1 nM to 100 µM). All compounds were tested with n ≥ 3. Each point represented the mean percentage specific binding ± S.E.M.

4.3.2.5.1 C3aR Agonists Containing a Heterocycle in the Linker

The compounds were tested in the calcium mobilisation assay at different concentrations.

Compound 143 and 144 displayed partial agonist activity with EC50 7 nM and 645 nM respectively, while compounds 145 and 146 showed negligible agonist activity (EC50 ≥ 50 µM)(Figure 4.19).

From the results we also found that compound 143 (pEC50 = 8.12 ± 0.15) had greater agonist potency than hC3a (pEC50 = 7.2 ± 0.06), although the plots below should probably be viewed as just qualitatively indicating comparable agonist potency (see Table 4.9). 145 and 146 were chosen for further examination of their antagonist activity in the calcium mobilisation assay.

185 CHAPTER 4 Towards C3a receptor antagonists

hC3a (n=5)

) 100 0 a 2

3 BR87 (n=6) 5 C m

h BR88 (n=4) 80 E

(

e BR90 (n=2) n

c 0 n 60 0

e BR91 (n=2) 1 c

s y

e 40 b r

o d u e l c F

20 u d

n i % 0 -12 -10 -8 -6 -4 -2 Log [Agonist] (M)

Figure 4.19 Intracellular calcium release from human monocyte derived macrophage cells induced by hC3a (), 143 (l), 144 (n), 145 (l), 146 (♦). Difference in fluorescence was measured and expressed as a percentage of maximum response induced by 100 nM hC3a ± S.E.M. Results were plotted against corresponding concentration in Prism 5.0. All compounds were tested with n ≥ 3.

4.3.2.5.2 C3aR Antagonists Containing Heterocycles.

In the calcium mobilisation assay, compounds 145 and 146 showed higher antagonist activity than SB290157 (Table 4.10) without agonist activity (Figure 4.20), which suggests that these compounds might provide clues for further C3aR antagonist development. However, the advantage of the heterocycle over the ether oxygen in SB290157 was not as great as it was for the agonist potency of the related compound 143. Perhaps the hydrogen bond acceptor power supplied by the heteroatom of the 5-membered ring is more important for agonist than antagonist function.

Table 4.10 Comparison of results between Agonist and Antagonist activity of 145 and 146.

Apparent Apparent Agonist Antagonist Activity Activity Agonist/ n pIC50 ± SE IC50 (nM) n pEC50 ± SE EC50 (nM) Antagonist 145 3 6.34 ± 0.23 460 2 ϕ ϕ 146 3 6.27 ± 0.22 534 2 4.13 ± 2.7 74800 ϕ compound showed no agonist activity up to 100 µM. no statistic significant of pIC50 value between 145 and 146. The statistical differences between two compounds (145 and 146) were accessed using Student’s t-test between them.

186 CHAPTER 4 Towards C3a receptor antagonists

) 0 M

2 100 1 n 5

0 5

M Antagonism 0

E 80

1 6 (

e a

c 1 3 60 n

C Agonism e 5 h c

s 40 y 6 e b r

o d 20 e u l c F

u 0 d n % i

-10 -8 -6 -4 -2

Log [Ligand] (M)

Figure 4.20 Concentration dependent inhibition of C3a-inducd Ca2+ release in HMDM by SB290157 (l), 145 () and 146 (n). Various concentrations of SB290157, 145 and 146 were incubated with HMDMs 15 mins prior to the experiment. The cells were then challenged by 100 nM hC3a. Differences in fluorescence induced by intracellular calcium mobilisation were measured and expressed as percentage of response induced by 100 nM hC3a on untreated HMDMs ± S.E.M. Results were plotted to corresponding concentrations by using Prism 5.0. Compound SB290157 (♦), 145 (q) and 146 (♦) were also tested for agonist activity in the same assay. All the compounds were tested with n ≥ 3.

4.3.2.5.3 Selectivity of Heterocycle-Containing Ligands for C5aR

Competitive Radioligand Binding Assays: [125I]-C5a Binding. The compounds (143-146) were tested in the competitive binding assay to evaluate selectively for C3aR over C5aR. A single concentration screening (ligands at 20 µM) competition with radiolabelled C5a 20 pM was performed. The results showed that none of these compounds have affinity for C5aR on human macrophage cells (Figure 4.21). The results of binding were compared to 3D53, which was known as a C5aR antagonist with no cross reactivity with C3aR. It is therefore suggested that the compounds incorporating heterocycles are selective for C3aR over C5aR.

187 CHAPTER 4 Towards C3a receptor antagonists

5000

) 4000 m p c (

3000 a 5 C

- 2000 ] I 5 2

1 1000 [

0 r 3 3 4 5 6 ffe 14 14 14 14 u 3D5 B compounds [20µM]

Figure 4.21 Competitive binding between [125I]-C5a and heterocycle-containing ligands in human monocyte derived macrophage cells. Human monocyte derived macrophages (1.5x106 cells/mL) were incubated for 1 h at room temperature with constant concentration (20 pM) of [125I]-C5a and furan-incorporated ligands and 3D53 at a concentration of 20 µM. The buffer represents the total binding of [125I]-C5a to HMDM (approximately 4000 cpm). Error bars are means ± S.E.M.

4.3.2.5.4 Receptor Selectivity: Desensitisation

The desensitisation experiments were used to measure the selectivity of ligands for C3aR which helped to clarify that the ligands have selectivity for C3aR over other GPCRs that signal through intracellular calcium release. The hC3a (3 µM) was administered to HMDMs which produces an efflux of intracellular calcium that dissipates after 5-6 mins (Figure 4.22). A second addition of 3 µM hC3a after 5 mins has no effect, and this effect is due to desensitisation of the receptor through desensitisation/ internalisation of the receptor from the cell surface. Figure 4.22 shows all compounds in these series in the second addition after desensitised cells with hC3a did not cause release of calcium from HMDM cells which suggests these heterocyclic ligands are selective for C3aR.

188 CHAPTER 4 Towards C3a receptor antagonists

3µM C3a 3µM C3a ) ) 0 0

2 100

2 100 5 a 5 a 3 M 3 M E C

E C

( 80

M e M e µ c µ c 3 n 60

3 n 60

e y e y c b c

b s

d 40 e d 40 e 10 M 143

r µ e r e o c o c u u 3 M C3a l u 20 µ u l d 20 F d

F n

i n i

% 0

0 60 120 180 240 300 360 0 60 120 180 240 300 360 Time (sec) Time (sec)

3µM C3a 3µM C3a ) ) 0 0 2 100 2

100 5 a 5 a 3 M 3 M E C

( 80 E C

( 80

M e M e µ c µ c 3 n 60

3 n 60

e y e y c b c

b s

s d 40 e

d 40 e r e

r 100 M 145 e µ o c o c u 10µM 144 u l u u 20 l 20 d F d

F n

i n i

% 0

0 60 120 180 240 300 360 0 60 120 180 240 300 360 Time (sec) Time (sec) 3µM C3a ) 0

2 100 5 a 3 M E C

( 80

M e µ c 3 n 60

e y c b

d 40 e r e 100 µM 146 o c u u l 20 d F

n i

% 0

0 60 120 180 240 300 360 Time (sec)

Figure 4.22 C3a receptor desensitisation calcium release plots performed on HMDM for compounds with heterocycles at the linker region. HMDM were first treated with hC3a 3 µM that induced a calcium response that was then allowed to dissipate over 310 secs prior to a second agonist administration.

189 CHAPTER 4 Towards C3a receptor antagonists

4.4 DISCUSSION

From the previous study in Chapter 3, we found that the oxazole–containing compounds showed high similarity in their conformations to the last four C-terminal residues of C3a (GLAR in a beta turn conformation), suggesting that the oxazole scaffold favoured a conformation leading to potent C3aR agonists/antagonists. The compounds with an oxazole scaffold did indeed bind to C3aR with low nanomolar affinity and greater potency than hC3a itself. Next we incorporated a methyl group at the fifth position of the oxazole, but this did not improve the affinity for C3aR and showed only agonist activity. Further studies in this chapter involved incorporating a larger group (phenyl) at the fifth position, which led to higher affinity for C3aR. The first compound in this series (105) showed no improvement over the methyl-substituted analogue with respect to affinity for C3aR. However, the compound had no agonist activity up to 100 µM in the calcium mobilisation assay on HMDM. Moreover, 105 showed some potent antagonist activity, with IC50 17 nM, which is much more potent than SB290157 (IC50 2.7 µM). Substituting the Boc group in 105 by the 4-methoxybenzoyl substituent produced compound 109, which had greater affinity for C3aR and higher antagonist potency than SB290157 but similar activity to 105. The SAR study showed that modification of the oxazole compound at the 5-position with a phenyl group transformed C3aR agonists into antagonists. In another new series of C3aR non-peptidic agonists/antagonists obtained by modification of SB290157, we found that the substitution of the linker of SB290157 with an appropriate heterocyclic ring leads to improved binding affinity and potency of the new compounds in this series. The incorporation of oxazole (Figure 4.23) led to a dramatic improvement over a previous report by Denonne et al. (2007b), where replacing the ether oxygen with a furan oxygen was claimed to give improved affinity for C3aR.

Compound 143 was the most potent (IC50 = 37 nM in competitive binding studies) however it showed agonist activity in the calcium mobilisation EC50 = 6.6 nM, whereas compounds 145 and

146 had lower affinity for C3aR (IC50 606 and 91 nM) with less agonist activity. However, compound 144 (Figure 4.23) was the same compound reported by Denonne in 2007 (Denonne et 125 al., 2007b). In our hands, 144 had affinity for C3aR with pIC50 = 5.88 ± 0.11 vs 80 pM in [ I]-C3a 125 in whole HMDMs, while Denonne and colleagues reported pIC50 = 7.2 vs 20 pM [ I]-C3a in a cell membrane preparation from HMC-1 cells. The difference between these results may be due to differences between protocols, since isolated membrane receptors frequently show 10-100 fold higher affinities for ligands than whole cells (Monk et al., 2007).

190 CHAPTER 4 Towards C3a receptor antagonists

All compounds showed selectivity for C3aR over C5aR in competitive radioligand binding assays using 125I labelled C3a and C5a, and also in a receptor desensitisation assay, which confirmed the selectivity of the ligands by their inability to induce calcium release after desensitisation of the cells with C3a. In conclusion, from these studies of structure-activity relationships (SAR) for potent and selective non-peptide C3aR antagonists, we found that the fifth position of the oxazole is the critical position for transforming an agonist to an antagonist. From another series, where substitution at the linker region of SB290157 with furan-like rings, the thiazole compound (146) improved the binding affinity and had antagonist activity with IC50 lower than for SB290157.

Figure 4.23 Superimposition of SB290157 (green) and furan compound (144) showing the location of the hydrogen bond acceptor atom (arginine residue not shown) (courtesy of Dr. Reid).

191 CHAPTER 4 Towards C3a receptor antagonists

4.5 REFERENCES

2. BLACK, J (1989) Drugs from emasculated hormones- the principle of syntopic antagonism. Science 245(4917): 486-493.

3. CLANCY, R (2000) Circulating endothelial cells and vascular injury in systemic lupus erythematosus. Current Rheumatology Rep? 2: 39.

4. DENONNE, F, BINET, S, BURTON, M, COLLART, P, DEFAYS, S, DIPESA, A, ECKERT, M, GIANNARAS, A, KUMAR, S, LEVINE, B, NICOLAS, JM, PASAU, P, PEGURIER, C, PREDA, D, VAN HOUTVIN, N, VOLOSOV, A, ZOU, D (2007a) Discovery of new C3aR ligands. Part 2: Amino-piperidine derivatives. Bioorg. Med. Chem. Lett. 17(12): 3262-3265.

5. DENONNE, F, BINET, S, BURTON, M, COLLART, P, DIPESA, A, GANGULY, T, GIANNARAS, A, KUMAR, S, LEWIS, T, MAOUNIS, F, NICOLAS, JM, MANSLEY, T, PASAU, P, PREDA, D, STEBBINS, K, VOLOSOV, A, ZOU, D (2007b) Discovery of new C3aR ligands. Part 1: Arginine derivatives. Bioorg. Med. Chem. Lett. 17(12): 3258-3261.

6. DROUIN, SM (2001) Expression of the Complement Anaphylatoxin C3a and C5a receptors on Bronchial Epithelial and Smooth muscle cells in Models of Sepsis and Asthma. J. Immunol. 166: 2025-2032.

7. ELSNER, J, OPPERMANN, M, CZECH, W, KAPP, A (1994) C3a activates the respiratory burst in human polymorphonuclear neutrophilic leukocytes via pertussis-toxin-sensitive G- proteins. Blood 83(11): 3324-3331.

8. GRANT, EB, GUIADEEN, D, SINGER, M, ARGENTIERI, D, HLASTA, DJ, WACHTER, M (2001) Design, synthesis, and biological activity of diiminoisoindolines as complement component 3a antagonists. Bioorg. Med. Chem. Lett. 11(21): 2817-2820.

9. MADALA, PK, TYNDALL, JDA, NALL, T, FAIRLIE, DP (2010) Update 1 of: Proteases Universally Recognize Beta Strands In Their Active Sites. Chem. Rev. 110(6): PR1-PR31.

10. MATHIEU, MC, SAWYER, N, GREIG, GM, HAMEL, M, KARGMAN, S, DUCHARME, Y, LAU, CK, FRIESEN, RW, O'NEILL, GP, GERVAIS, FG, THERIEN, AG (2005) The C3a receptor antagonist SB 290157 has agonist activity. Immunology Letters 100(2): 139-145.

11. MONK, PN, SCOLA, AM, MADALA, P, FAIRLIE, DP (2007) Function, structure and therapeutic potential of complement C5a receptors. British Journal of Pharmacology 152(4): 429-448.

12. MOUSLI, M, HUGLI, TE, LANDRY, Y, BRONNER, C (1992) A Mechanism of action for anaphylatoxin C3a stimulation of mast cells. J. Immunol. 148(8): 2456-2461. 192 CHAPTER 4 Towards C3a receptor antagonists

13. NAGATA, S, GLOVSKY, MM, KUNKEL, SL (1987) Anaphylatoxin-induced neutrophil chemotaxis and aggregation - Limited aggregation and specific desensitization induced by human C3a and synthetic C3a octapeptides. International Archives of Allergy and Applied Immunology 82(1): 4-9.

14. NORGAUER, J, DOBOS, G, KOWNATZKI, E, DAHINDEN, C, BURGER, R, KUPPER, R, GIERSCHIK, P (1993) Complement fragment C3a stimulates Ca2+ influx in neutrophils via a pertussis-toxin-sensitive G-protein. Eur. J. Biochem. 217(1): 289-294.

15. PROCTOR, LM, ARUMUGAM, TV, SHIELS, I, REID, RC, FAIRLIE, DP, TAYLOR, SM (2004) Comparative anti-inflammatory activities of antagonists to C3a and C5a receptors in a rat model of intestinal ischaemia/reperfusion injury. British Journal of Pharmacology 142(4): 756-764.

16. TAKAFUJI, S, TADOKORO, K, ITO, K, DAHINDEN, CA (1994) Degranulation from human eosinophils stimulated with C3a and C5a. Int. Arch. Allergy Immunol. 104: 27-29.

17. TYNDALL, JDA, PFEIFFER, B, ABBENANTE, G, FAIRLIE, DP (2005) Over one hundred peptide-activated G protein-coupled receptors recognize ligands with turn structure. Chem. Rev. 105(3): 793-826.

18. VOLANAKIS, JE, FRANK, MM (1998) Human complement system in health and disease. Marcel Dekker: New York.

19. WETSEL, RA, KILDSGAARD, J, HAVILAND, DL (2000) Complement Anaphylatoxins (C3a,C4a,C5a) and Their Receptors (C3aR, C5aR/CD88) as Therapeutic Targets in Inflammation. In: Therapeutic Interventions in the Complement System, Lambris, JD, Holers, VM (eds), p 269. Totowa,NJ: Humana Press Inc.

193 CHAPTER 4 Towards C3a receptor antagonists

194 CHAPTER 5 Antagonism of C3aR Agonists

CHAPTER 5 ANTAGONISM OF C3aR AGONISTS

5.0 ABSTRACT...... 196 5.1 INTRODUCTION...... 197 5.2 AIMS/OBJECTIVES...... 201 5.3 RESULTS ...... 202

5.3.1 COMPETITION OF NON-PEPTIDE ANTAGONIST SB290157 WITH C3aR AGONISTS...203 5.3.1.1 SB290157 is a Non-competitive and Insurmountable Antagonist of hC3a...... 203 5.3.1.2 SB290157 is a Competitive and Surmountable Antagonist of Agonists 17 and 20...... 204 5.3.1.3 SB290157 is a Competitive and Surmountable Antagonist of Agonists 63 and 80...... 208

5.3.2 COMPETITION OF PEPTIDE ANTAGONIST 25 (FLTCHAAR) WITH

C3aR AGONISTS ...... 212 5.3.2.1 Compound 25 is a Non-competitive and Insurmountable Antagonist of hC3a ....212 5.3.2.2 Compound 25 is a Non-competitive and Insurmountable Antagonist of Agonists 17 and 20...... 214 5.3.2.3 Compound 25 is a Non-competitive and Insurmountable Antagonist of Agonists 63 and 80...... 217

5.3.3 COMPETITION OF NON-PEPTIDE ANTAGONIST 146 WITH C3aR AGONISTS...... 220 5.3.3.1 Compound 146 is a Non-competitive and Insurmountable Antagonist of hC3a ..220 5.3.3.2 Compound 146 is a Competitive and Surmountable Antagonist of Agonist 17...221 5.3.3.3 Compound 146 is a Competitive and Surmountable Antagonist of Non-Peptide Agonist 63...... 223 5.4 DISCUSSION ...... 224 5.5 MATERIALS AND METHODS ...... 229

5.5.1 CALCIUM MOBILIZATION ASSAY ON HMDM CELLS...... 229 5.5.1.1 Antagonist Surmountability/Insurmountability: ...... 229 5.5.1.2 Data Analysis ...... 229 5.6 REFERENCES...... 231

195 CHAPTER 5 Antagonism of C3aR Agonists

5.0 ABSTRACT

Potent and selective peptide antagonist 25 and non-peptide antagonist 146 have been compared in this chapter with SB290157 for their capacity to inhibit intracellular calcium release in HMDM stimulated by up to 100 µM of three different classes of C3aR agonists: human C3a, peptide agonists (17 and 20), and non-peptide agonists (63 and 80). The antagonists SB290157, 25 and 146 all displayed insurmountable and non-competitive antagonism against C3a for C3aR. This is attributed to either a possible allosteric binding site for the compounds on C3aR or possibly irreversible or pseudo-irreversible binding to the C3a receptor. By contrast, SB290157 was a surmountable and competitive antagonist against peptide agonists (17 and 20) and non-peptide agonists (63 and 80). Compound 25 was an insurmountable antagonist for all C3a agonists (hC3a, peptide and non-peptide agonists). Compound 146 behaved similarly to SB290157 in exhibiting surmountable and competitive antagonism of peptide and non-peptide C3aR agonists. These results point to a mechanism of insurmountable antagonism for compound 25 against all agonists (hC3a, 17, 20, 63 and 80) involving binding to C3aR that is unlikely to be at an allosteric site since it competed with [125I]-C3a in the competitive binding assay. It is more likely that compound 25 binds pseudo-irreversibly to C3aR. The competitive antagonism study of C3aRA (SB290157) against all peptide and non- peptide agonists showed a concentration of C3aRA (SB290157) that caused a two-fold rightward shift of agonists in the following order: 20 (63 nM) < 17 (100 nM) < 63 (126 nM) < 80 (158 nM). This indicates that the non-peptide agonists (63 and 80) bind to C3aR more tightly than peptide agonists (17 and 20), thus requiring higher concentrations of C3aRA (SB290157) to shift the EC50 value of agonists two-fold to the right.

196 CHAPTER 5 Antagonism of C3aR Agonists

5.1 INTRODUCTION

The peptide inflammatory mediator C3a is released after activation of the complement system by the cleavage of Component 3 by C3 convertases in either of the three known complement activation pathways. C3a causes chemotaxis and degranulation of eosinophils (Daffern et al., 1995) basophils and mast cells (Hartmann et al., 1997), stimulates cytokines release from macrophages and monocytes (Fischer et al., 1999; Takabayashi et al., 1998; Takabayashi et al., 1996), induces smooth muscle contraction (Hugli, 1975) and increased vascular permeability (Vallota et al., 1973). C3a and its receptor have been implicated in various immuno-inflammatory and autoimmune diseases. Studies of C3a in vitro and in vivo show that C3a is involved in asthma (Humbles et al., 2000), allergies (Drouin et al., 2002; Drouin et al., 2001), sepsis (Drouin, 2001; Kildsgaard et al., 2000), allergic rhinitis (Andersson et al., 1994; Jun et al., 2008), systemic lupus erythematosus (Bao et al., 2005), diabetes (Mamane et al., 2009), arthritis (Hutamekalin et al., 2010), psoriasis (Kapp et al., 1985), nephropathy (Tang et al., 2009; Wenderfer et al., 2009) and ischemia reperfusion injury (Mocco et al., 2006; Proctor et al., 2004). C3a mediates its inflammatory functions by interacting with its receptor, C3aR that belongs to the rhodopsin family of 7- transmembrane GPCR (Kretzschmar et al., 1993; Martin et al., 1997; Zwirner et al., 1997). Like C5a, the carboxy-terminal sequence of C3a (LGLAR) is important to activate its receptor (Caporale et al., 1980; Chazin et al., 1988; Hoeprich et al., 1986; Kawai et al., 1991; Kohl et al., 1990; Zhang et al., 1997). The minimum sequence is LAR which was reported by Kohl and colleagues (1990) is vital for biological activity of C3a. There are many groups that focus on developing the therapeutic potential of selective C3aR agonists/antagonists by modification of the C-terminal region of the native peptide C3a. From the results of chimeric and deletion mutagenesis of C3aR, a two-site binding model was proposed (Figure 5.1) for the C3a - C3aR interaction which was similar to that proposed for the C5a-C5aR interaction (Chao et al., 1999; Chenoweth et al., 1980; Ember et al., 1997). The “non-effector” site or “recognition” binding site (site 1) is hypothesized to involve an interaction between the anionic residues near the N and C-termini of extracellular loop two of C3aR and cationic residues in the C-terminal helical region of C3a. The second binding site, called the “effector” site (site 2), is thought to be responsible for activation of the receptor. The C-terminal sequence of C3a (LGLAR) interacts in a binding pocket, which is formed by the transmembrane domains of the receptor (Chao et al., 1999; Ember et al., 1997). Unlike C5aR, the C3aR does not have a highly anionic N-terminal region, which constitutes a non-effector or recognition site in C5aR for C5a-C5aR interaction. The replacement of the C3aR N-terminal sequence with that from the hC5aR, or truncated up to 16 amino acids, had minimal effect on C3a binding (Chao et al.,

197 CHAPTER 5 Antagonism of C3aR Agonists 1999; Crass et al., 1999). Thus, the N-terminal region of C3aR has a minor role in the C3a - C3aR interaction while the substitution of the 2nd extracellular loop causes changes in its affinity and activity.

Figure 5.1 Two-site model proposed for the interaction between C3a and its receptor. The proposed model for the C3a - C3aR interaction corresponds to a model originally proposed for multi-site binding of C5a-C5aR interaction (Chenoweth et al., 1980; Siciliano et al., 1994). The non-effector site (site 1) is hypothesized to involve the interaction of the negatively charged regions of the 2nd extracellular loop with the positively charged residues in the C3a molecule close to the C-terminal effector region (site 2) (Ember et al., 1997).

The crucial role of C3a in activation of complement pathways makes the C3a/C3aR interaction a highly desirable drug target. However, there are currently no drug-like molecules available that are potent and selective agonists or antagonists for C3aR.

198 CHAPTER 5 Antagonism of C3aR Agonists

Recently, our group (in Chapter 2) reported potent and selective hexapeptide C3aR agonists (17 and 20) and an antagonist (25)(Scully et al., 2010)(Figure 5.2). These compounds may bind to the receptor in a turn conformation. The hexapeptide agonists (17 and 20) had high affinities for the 125 human C3aR (IC50 40 and 82 nM against 80 pM I-C3a) on peripheral blood monocytes, and were potent agonists of C3a-stimulated intracellular calcium release (EC50 0.32 µM, pEC50 = 6.50 ± 0.12; n=5 and 0.37 µM, pEC50 = 6.43 ± 0.09; n=6), respectively) in dU937 cells. The new hexapeptide antagonist (25) showed no agonist activity up to 100 µM, bound tightly to C3aR (IC50 238 nM 125 against 80 pM I-C3a) on PBMCs, and exhibited similar antagonist potency to SB290157 (IC50 =1.3 µM versus 100 nM of hC3a on dU937 cells.) Incorporation of a heterocycle into C-terminal tripeptide sequences resulted in Chapter 3 and Chapter 4 reporting new ‘non-peptidic’ C3aR agonists (63, 80; Figure 5.2) and antagonists

(105, 146; Figure 5.2), which bound with high affinity to C3aR (IC50 = 9 nM, 14 nM, 37 nM, 91 nM respectively) on HMDMs, as revealed by radioligand competition assay using 125I-C3a (80 pM).

Agonists 63 (EC50 = 7 nM, pEC50 = 8.15 ± 0.10; n=3) and 80 (EC50 = 45 nM, pEC50 = 7.35 ± 0.18; n=2) were of comparable potency to C3a in inducing intracellular calcium release in isolated HMDM. The substitution of oxazole in Chapter 4 by 5-phenyl oxazole transformed agonists into potent antagonists like 105 (IC50 = 16.5 nM against 100 nM hC3a) and by thiophene into antagonist

146 (IC50 = 534 nM) on HMDM. In this Chapter, we now examine more closely the nature of the antagonism for three compounds (SB290157, peptide antagonist 25, non-peptide antagonist (146) against individual agonists of three classes: human C3a, potent peptide agonists (17, 20) and potent non-peptide agonists (63 and 80). The results from this study could lead us to a better understanding of the binding site of these compounds on C3aR and whether they are orthosteric or allosteric antagonists.

199 CHAPTER 5 Antagonism of C3aR Agonists

HN NH2 HN NH2 NH NH NH

O O O O O O H H H H H N N N OH H N N N OH 2 N N N 2 N N N H H H H H H O O O O O O OH OH

17 20

HN NH2

NH O O N O O O H H H NH2 H N N N OH N N HN 2 N N N H H H NH O O O OH O HO C 2 25 63

Me O O O O N O O NH NH N HN COOH O O HN CO2H

NH NH HN NH HN NH2 2 80 105

H2N NH N O H NH N S OH O O O OH N NH H O HN NH2 146 SB290157

Figure 5.2 Structures of peptide agonists (17 and 20), peptide antagonist (25), non-peptide agonists (63 and 80) and non-peptide antagonists (105, 146 and SB290157).

200 CHAPTER 5 Antagonism of C3aR Agonists

5.2 AIMS/OBJECTIVES

The objectives of this chapter were initially to study the most potent and selective currently available antagonists (peptide FLTChaAR (25), nonpeptide SB290157, and SB290157 analogue 146) for their mechanism(s) of action against three different classes of agonists of C3aR (protein C3a, peptide 17 and 20, peptidomimetics 63 and 80). By identifying competitive/noncompetitive and surmountable/insurmountable behaviour in the Ca2+ release assay, we could potentially learn about the relative binding locations of the compounds and their different mechanisms of action. There have been no reports of human C3aR subtypes so far, with only one C3aR in the human genome having been described (Crass et al., 1996). This is in contrast to C3aR in guinea pigs for which there were 2 C3aR subtypes reported, but only after a great deal of research in the 1980s and 1990s on guinea pig tissues (Fukuoka et al., 1998). This new study using a variety of new C3aR antagonists may give confidence to other researchers wanting to further classify the human C3aR and its pharmacology in vitro and in vivo. In this study we decided to compare the pharmacological profiles of SB290157 and the new peptide antagonist (25) and non-peptide antagonist (146) in the calcium mobilisation assay on human monocyte-derived macrophages.

201 CHAPTER 5 Antagonism of C3aR Agonists

5.3 RESULTS

Results from the previous chapters can be plotted to examine any correlation between receptor binding affinity at human C3aR and receptor activation as defined by intracellular calcium release (Figure 5.3). The plot below shows a stark correlation that separates the agonists (17, 20, 63, 80) from the antagonists (25, 105, 146, SB290157). At first, we thought that the higher than predicted agonist potencies for synthetic compounds might be reflecting off-target agonist action via some other receptor(s) that could be contributing to intracellular calcium release. However, our receptor desensitization experiments do not support that conclusion. It is more likely, in our opinion, that the reason our compounds 17, 20, 63, 80 are not precisely on the line defined by C3a is probably more to do with our data for C3a, which does tend to vary from batch to batch and depending upon the supplier. Currently, commercial source of C3a are isolated from serum and are supplied in a phosphate buffer solution. The batches of C3a show variability decrease in the ability to induce calcium release, which shows the change in EC50 and reduces the maximum response of calcium release at the EC100 of C3a. It may be a result of serum/ C3a variability at the source, concentration variations or freeze/thaw cycles. However, even the small amounts of C3a obtained from the source of variation showed its high potency with EC50 = 18.6 nM (pEC50 = 7.7 ± 0.08, n=4) for calcium release in HMDM.

Figure 5.3 Correlation between binding affinity (pIC50) and agonist activity (pEC50) of hC3a, peptide C3a agonists (17, 20), non-peptidic C3a agonists (80, 63) and C3a antagonists (SB290157, 25, 105, 146). There were no agonist activities for SB290157, 25, 105 and 146 in the calcium mobilisation assay test on HMDM cells. Error bars represented S.E. 202 CHAPTER 5 Antagonism of C3aR Agonists

5.3.1 Competition of Non-peptide Antagonist SB290157 with C3aR Agonists

5.3.1.1 SB290157 is a Non-competitive and Insurmountable Antagonist of hC3a

C3a stimulated calcium release in a concentration-dependent manner in HMDM derived from differentiating monocytes that isolated from fresh, whole, human blood. It was found that human C3a mediated calcium release that was non-competitively and insurmountably inhibited by the C3aR antagonist, SB290157. This was indicated by a vertical reduction in the C3a-induced response without any shift of IC50, even at the higher concentrations of agonist. This means that human C3a was not able to attain full receptor activation in the presence of SB290157 which was unable to be completely displaced from C3aR. This is consistent with irreversible or pseudo- irreversible binding or with it binding at an allosteric site on the receptor. The maximal hC3a (1 µM) mediated Ca2+ release was inhibited to 97.6 ± 5.0 %, 84.2 ± 3.0 %, 85.8 ± 2.5 %, 72.0 ± 2.2 % by C3aR antagonist (SB290157) at concentration of 200 nM, 1, 2 and 8 µM, respectively (Figure 5.4). The EC50 of hC3a in the absence of C3a antagonist

(SB290157) was 18.6 nM (pEC50 = 7.7 ± 0.08, n = 4) and EC50 was higher in the presence of increasing concentrations of SB290157. EC50 (pEC50) was 31.0 nM (7.5 ± 0.07), 40.8 nM (7.4 ± 0.06), 53.5 nM (7.3 ± 0.05) and 75.7 nM (7.1 ± 0.04) in the presence of 0.2, 1, 2 and 8 µM of SB290157. Schild plot for C3a antagonist SB290157 against hC3a (slope <1).

(A).

+ "$$ 3 $ 3 # ? 2

+ %$ > + " = 1 1 + / 9 .

; &$ - < D 9 + ; C

: '$ B + 9 8 A ) 9 7 ; 6 #$ 7 5 + A < + @

4 $

!"# !"$ !% !& !' ()*+,-./0123

203 CHAPTER 5 Antagonism of C3aR Agonists

(B).

)*+!,$%-'.% -'-( $#% ( 4 #

; / %$%-'00 ! , * : $#' 9

0 !" $%&'(

8 # ,

7 $#( 6 * 5 2 , $#$ +

* !"#$ !%#& !%#$ !&#& !&#$ !'#& ) !$#(

)*+,-./01,234

Figure 5.4 Antagonism by SB290157 against hC3a. (A) SB290157 is a non-competitive and insurmountable antagonist against hC3a, showing concentration-dependent (zero (l), 200 nM (n), 1 µM (), 2 µM (), 8 µM (®) SB290157) inhibition of iCa2+ release in human monocyte derive macrophage cells (HMDM) induced by varying concentrations (0.01 nM to 1 µM) of hC3a. Data is expressed as % of maximal change in fluorescence induced by hC3a (1 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for C3a antagonist SB290157 against hC3a (slope <1).

5.3.1.2 SB290157 is a Competitive and Surmountable Antagonist of Agonists 17 and 20

The peptidic C3a agonists 17 and 20 were the first two potent and selective C3a agonists reported from our group in 2010 (Scully et al., 2010). 17 and 20 at 100 µM showed full agonist efficacy on human C3aR like hC3a at 100 nM in stimulating the release of intracellular calcium in dU937 cells (Chapter 2). Concentration response curves for 17 and 20 were generated on HMDM cells in the absence of SB290157 and presence of 200 nM, 300 nM, 1, 2, 4, 8 and 10 µM that pre-treated cells 15 mins before the calcium mobilisation assays were performed. These curves were then compared with the control curve (in the absence of SB290157) as shown in Figure 5.5 A and 5.6 A.

SB290157 is a competitive and surmountable antagonist against compound 17

SB290157 or C3aR antagonist inhibited the concentration response of 17 and 20 in the calcium mobilisation assay (Figure 5.5 A and 5.6 A) by shifting the concentration-response curves

204 CHAPTER 5 Antagonism of C3aR Agonists to the right or horizontally. However, the maximal response for agonist activities of these compounds can be achieved by increasing the concentration of agonist compounds. These are characteristics of a competitive and surmountable antagonist. The EC50 of 17 in the absence of C3a antagonist (SB290157) was 435 nM (pEC50 = 6.4 ± 0.11, n=4) and EC50 was higher in the presence of increasing concentrations of SB290157. EC50 was 1.3 µM (5.9 ± 0.08), 2.5 µM (5.6 ± 0.07), 8.6 µM (5.1 ± 0.07), 33 µM (4.5 ± 0.25) and 17.4 µM (4.8 ± 0.02) in the presence of 300 nM, 1, 2, 8 and 10 µM of SB290157 on HMDM. The results (Figure 5.5 A) were converted to a Schild plot (Figure 5.5 B) to calculate the

Schild slope factor and pA2 value of SB290157 against 17. The Schild slope is an indication of the competitive relationship between agonist and antagonist. The pA2 value for the antagonist represents a negative logarithm of the concentration of antagonist (SB290157) required to cause a two-fold rightward shift of agonist EC50 (17 or 20). The Schild plot suggests that SB290157 is a competitive and surmountable antagonist of C3aR against compound 17. The slope factor of approximately 1.0 (0.9 ± 0.1) suggests a fully competitive relationship between these two compounds (i.e. ratio of 1:1). The calculated pA2 value indicates that the concentration of SB290157 required to shift the compound 17 curve by two fold is

~158 nM (pA2 = 6.8, Figure 5.5 B). We conclude that SB290157 likely binds at the orthosteric site for agonist 17; in other words they bind at the same site on human C3aR on HMDMs.

(A).

205 CHAPTER 5 Antagonism of C3aR Agonists

(B).

Figure 5.5 SB290157 is a competitive antagonist against 17 on human monocyte derived macrophages, (A) indicated by shifting of EC50 values due to competition at various concentrations of SB290157 (zero (l), 0.3 (o), 1 (r), 2(), 8 (®), 10 () µM) as measured by the inhibition of iCa2+ release in HMDM which was induced by varying concentrations of C3a peptide agonist 17 (3 nM to 100 µM). Data is expressed as % of maximal change in fluorescence induced by 17 (100 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for SB290157 against 17 (slope = 0.9 ± 0.1). Calculated pA2 value for SB290157 against 17 was 6.80 (or A2 = 158 nM).

SB290157 is a competitive and surmountable antagonist against compound 20

SB290157 is also a competitive antagonist (reversible and surmountable) against C3a peptidic agonist, 20. Increasing concentrations of SB290157 resulted in a horizontal shift in the agonist concentration-response curves (Figure 5.6 A). SB290157 is a surmountable antagonist of 20, which can be fully displaced by 20 at high concentrations. In Figure 5.6 A, the EC50 of 20 in the absence of C3a antagonist (SB290157) was 11.7 nM (pEC50 = 7.9 ± 0.6, n=4) and EC50 was higher in the presence of increasing concentrations of SB290157. EC50 was 48.9 nM (7.3 ± 0.29), 289.5 nM (6.5 ± 0.14), 627.3 nM (6.2 ± 0.22), and 1.4 µM (5.9 ± 0.12) in the presence of 200 nM, 2, 4 and 8 µM of SB290157 on HMDM. The data reveal that SB290157 and agonist 20 likely bind at the same site on human C3aR on HMDMs since the Schild plot has a slope of very close to 1.0.

206 CHAPTER 5 Antagonism of C3aR Agonists (A). + 0 # # ' ' +

< "## + ? ; ) + :

. $# / + + 6 + 8 # %# 9 # 6 " + 8

) &# 7 > + 6 5 6 ) 7 '# 4 9 3 ) 2 + = # 7 + 6 5 1

!"# !$ !% !& !' ()*+,'#-+./0

(B).

)#$ 5 ()*!+%$,'-&$(,',.

( # ! / $%$,'-- -

+ )#% < ; : 9

- (#$ 8

7 !"#$%$&'#

+ (#% 6 3 -

, %#$ + * %#% !"#$ !"#% !&#$ !&#% !$#$ !$#% !'#$ *+,-./0)1%($"2-345

Figure 5.6 SB290157 is a competitive antagonist against 20 on human monocyte-derived macrophages, which (A) showing shifting of EC50 values due to competition at varies concentrations of SB290157 (zero (l), 0.2 (n), 2 (), 4 (), 8 (®) µM) as measured by the inhibition of iCa2+ release in HMDM which was induced by varying concentrations of C3a peptide agonist FWTLAR (1 nM to 100 µM). Data is expressed as % of maximal change in fluorescence induced by 20 (100 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for SB290157 against 20 (slope = 1). Calculated pA2 value for SB290157 against 20 was 7.2.

207 CHAPTER 5 Antagonism of C3aR Agonists

The Schild plot suggests that SB290157 is a surmountable antagonist of C3aR against 20. The slope factor of approximately 1.0 (0.97 ± 0.05), suggests the competitive relationship between these two compounds is equal to 1:1. The calculated pA2 value indicates that the concentration of

SB290157 required to shift the compound 20 curve by two fold is ~ 63 nM (pA2 = 7.2, Figure 5.6 B).

5.3.1.3 SB290157 is a Competitive and Surmountable Antagonist of Agonists 63 and 80

SB290157 is a competitive and surmountable antagonist against compound 63

Compound 63 is an oxazole-containing compound that shows a beta-turn structure from the molecular modelling and may potentially mimic the receptor bound conformation of the C-terminal tetrapeptide sequence (GLAR) of human C3a. From Chapter 3 in competitive binding studies, 63 binds tightly to C3aR with IC50 = 9.3 nM (pIC50 = 8.03 ± 0.05) and in the functional assay of calcium release it showed high potency in stimulating calcium from macrophage cells comparable to human C3a (EC50 = 7.2 nM, pEC50 = 8.15 ± 0.10). In selectivity studies by using competitive radioligand of C5a, 63 had no binding affinity to C5aR. Desensitisation assays also confirmed the selectivity of the non-peptidic ligand in stimulation of calcium release through C3aR only. The mechanism of antagonism of SB290157 against 63 study showed that SB290157 (C3aRA) is a surmountable and reversible antagonist against non-peptide agonist (63).

Results were plotted for concentration-dependent response curves in Figure 5.7 A, the EC50 of 63 in the absence of C3a antagonist (SB290157) was 8.4 nM (pEC50 = 8.08 ± 0.16, n=4) and

EC50 was higher in the presence of increasing concentrations of SB290157. EC50 was 23 nM (pEC50 = 7.64 ± 0.13), 162 nM (6.79 ± 0.2), 501 nM (6.3 ± 0.18), and 767 nM (6.12 ± 0.11) in the presence of 200 nM, 2, 4 and 8 µM of SB290157 respectively. This clearly showed that increasing concentrations of SB290157 cause significantly increased EC50 values for 63 in a concentration- dependent manner, while having no significant reduction effect on the maximal response of the compound (Figure 5.7 A). This is further supported by a Schild plot (Figure 5.7 B) with a linear slope of 1.1± 0.08 (slope ~1), which indicates the relationship between agonist and antagonist as competitive. The pA2 value is the concentration of SB290157 required to shift the agonist curve of

63 by two-fold, and is ~ 126 nM (or pA2 = 6.9).

208 CHAPTER 5 Antagonism of C3aR Agonists

(A). + 1 $ - # & < + "$$ ? 0 ) + ; +

/ %$ + 0 7 + + 9

$ &$ : " 7 + 9 ) >

8 '$ + 7 7 6 8

) #$ : 5 4 ) 3 = + $ 8 + 7 6 2 !"# !"$ !% !& !' ()*+,&-.+/01

(B).

)#$ 5 )*+!,$%$-'-$($.'./ ( ! - 0#$%$.'(($

+ )#% < ; : 9

- (#$

8 !" $%$&'(

7 #

+ (#% 6 3 -

, %#$ + * %#% !"#$ !"#% !&#$ !&#% !$#$ !$#% !'#$ *+,-./0)1%($"2-345

Figure 5.7 SB290157 is a competitive and surmountable antagonist against non-peptide C3a agonist (63), (A) shows concentration - dependent (zero (l), 200 nM (n), 1 µM (), 2 µM (), 8 µM (®), n≥4) inhibition of iCa2+ release in human monocyte derived macrophage cells (HMDM) induced by varying concentrations (0.1 nM to 10 µM) of 63. Data is expressed as % of maximal change in fluorescence induced by 63 (10 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for C3a antagonist SB290157 against 63 (slope = 1). Calculated pA2 value for SB290157 against 63 was 6.9. 209 CHAPTER 5 Antagonism of C3aR Agonists

SB290157 is a competitive and surmountable antagonist against compound 80

Compound 80 is the acyl-leucine-5-methyl-oxazole-arginine compound. It binds tightly to

C3aR with IC50 = 14 nM (pIC50 = 7.86 ± 0.08, n=3, Chapter 4). However it had lower agonist potency than 63 (EC50 = 45 nM versus 7 nM respectively). It also has selectivity for C3aR, which was shown in the results of a competitive binding assay of [125I]-C5a and in the C3aR desensitisation experiment. The effect of SB290157 was determined in HMDM cells using 80 as the stimulus to release calcium inside cells. Concentration response curves of 80 were measured following incubation of the HMDMs with increasing concentrations of C3aR antagonist (SB290157). The results showed that SB290157 inhibited compound 80 in a competitive and surmountable fashion, since increasing concentrations of SB290157 increased the EC50 value of 80 and also made the dose response curve for the agonist shift rightward. However, the curves of 80 in presence and absence of antagonist did not show a clear result and the number of experiments need to be increased (Figure 5.8 A).

Results were plotted in a dose-dependent response curve in Figure 5.8 A, the EC50 of 80 in the absence of C3a antagonist (SB290157) was 59.4 nM (pEC50 = 7.23 ± 0.08, n=6) and EC50 was higher in the presence of increasing concentrations of SB290157. EC50 was 76.1 nM (7.12 ± 0.16), 285.6 nM (6.54 ± 0.11), 375.9 nM (6.43 ± 0.08), and 2.75 µM (5.56 ± 0.05) in the presence of 200 nM, 1, 2 and 8 µM of SB290157 (n=2) respectively. The results from dose response curve of SB290157 against 80 is further supported by Schild plot (Figure 5.8 B) with a linear slope of 0.95 ± 0.10 (slope ~1). The Schild plot suggests that

SB290157 is a competitive and surmountable antagonist of C3aR against 80. The pA2 value is the concentration of SB290157 that is required to shift the agonist curve of 80 by two-fold is ~ 158 nM

(or pA2 = 6.8).

210 CHAPTER 5 Antagonism of C3aR Agonists (A). $ + ' ' 8 &

7 +)) $ < 6 " $ 5 #) ) * $ + 1 ' 3 %) ' 4 ; 1 $ 3 "

: ') 2 $ 1 1 0 2 " *) 4 / . " - 9 $ ) 2 $ 1 0 ,

!" !# !$ !% !& !' !( !"#$%&'($)*+

(B).

)#% 5 )*+!,$%$-'./$(-'0 (

! #

- 1 %-'.& + < (#$ ; : 9 -

8 (#% 7 !" $%$&'( + # 6 3 - %#$ , + * %#% !"#$ !"#% !&#$ !&#% !$#$ !$#% !'#$ *+,-./0)1%($"-2345

Figure 5.8 SB290157 is a competitive and surmountable antagonist against non-peptide C3a agonist (80). (A) showing concentration - dependent (zero (l), 200 nM (n), 1 µM (), 2 µM (), 8 µM (®)) inhibition of iCa2+ release in human monocyte derived macrophage cells (HMDM) induced by varying concentrations (10 nM to 100 µM) of 80. Data is expressed as % of maximal change in fluorescence induced by 80 (100 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for SB290157 against 80 (slope = 1). Calculated pA2 value for SB290157 against 80 was 6.8.

211 CHAPTER 5 Antagonism of C3aR Agonists 5.3.2 Competition of Peptide Antagonist 25 (FLTChaAR) with C3aR Agonists

5.3.2.1 Compound 25 is a Non-competitive and Insurmountable Antagonist of hC3a

Compound 25 was first reported by our group as a peptide antagonist (Scully et al., 2010). It binds to C3aR with IC50 = 238 nM (pIC50 = 6.62 ± 1.59, n=2) and has no agonist activity at the concentration up to 100 µM in dU937 cells. It was an equipotent antagonist of C3a-induced intracellular calcium release with IC50 = 1.5 µM vs 100 nM of hC3a in dU937) compared to 125 SB290157 (IC50 = 1.3 µM) (Chapter 2). It was selective for C3aR over C5aR in competitive [ I]- C5a binding assays. The antagonist effect of 25 was then determined in HMDMs by using human C3a to stimulate intracellular calcium release from cells. Concentration response curves of C3a were performed after preincubation of macrophage cells with increasing concentrations of 25. Interestingly, the results showed that 25 inhibited C3a in an insurmountable manner because increasing concentrations of 25 reduced the maximal response achievable in a dose-dependent manner of C3a. The maximal hC3a (10 µM) mediated Ca2+ release was inhibited to 56.1 ± 9.5 % and 15.2 ± 3.3 % by peptide C3aR antagonist (25) concentration at 4 and 8 µM, respectively (Figure 5.9A).

The EC50 of hC3a in the absence of peptide C3a antagonist (25) was 48 nM (pEC50 = 7.3 ± 0.09, n

= 6) and EC50 was higher in the presence of increasing concentration of 25. EC50 was 131 nM (6.88 ± 0.11), 137 nM (6.87 ± 0.11), 153 nM (6.82 ± 0.09) and 232 nM (6.63 ± 0.28) and 42 nM (7.38 ± 0.58) in the presence of 500 nM, 1, 2, 4 and 8 µM of compound 25. A Schild plot (Figure 5.9B) suggests that compound 25 is a non-competitive and insurmountable antagonist of human C3a as shown by a linear slope of 0.37 (much less than 1.0).

212 CHAPTER 5 Antagonism of C3aR Agonists (A).

)

0 ) 2 + 0.5µM FLTChaAR 5 M µ M + 1µM FLTChaAR E 0 ( 1 100 + 2µM FLTChaAR ( e

c a + 4µM FLTChaAR 3 n e C + 8µM FLTChaAR

c y s hC3a b

e 50

r d o e u l c F u

n 0 i % -10 -9 -8 -7 -6 -5 Log [C3a] (M)

(B).

$#) 2 )*+!,%$%-'./% %-'0&

: ( ! - #

+ 1 $%-'(. 9 $#% 8 7 6 -

5 !" $%&'( $#( # 4 + 3 0 - $#' , + * $#$ !"#$ !%#& !%#$ !&#& !&#$ *+,-.'&/-012

Figure 5.9 Mechanism of C3aR antagonism. (A) Compound 25 is a non-competitive and insurmountable antagonist against hC3a, showing concentration - dependent (zero (r), 500 nM (l), 1 µM (n), 2 µM () 4 µM () and 8 µM (®)) inhibition of iCa2+ release in human monocyte derived macrophage cells (HMDM) induced by varying concentrations (0.1 nM to 3 µM) of hC3a. Data is expressed as % of maximal change in fluorescence induced by C3a (10 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for C3a antagonist 25 against hC3a (slope <1).

213 CHAPTER 5 Antagonism of C3aR Agonists

5.3.2.2 Compound 25 is a Non-competitive and Insurmountable Antagonist of Agonists 17 and 20

Compound 25 is a non-competitive and insurmountable antagonist against compound 17

The peptide antagonist 25 also inhibited 17 in an insurmountable fashion as increasing concentrations of 25 reduced the maximal response achievable in a dose-dependent manner of 17 similar to the results for hC3a. The maximal response of compound 17 mediated Ca2+ release (100 µM) was inhibited to 94.0 ± 6.6 %, 93.1 ± 5.2 %, 86.6 ± 6.1 %, 73.1 ± 4.2 %, 66.9 ± 4.4 % and 39.9 ± 2.7 %, by peptide C3aR antagonist 25 at concentrations of 200 nM, 2, 4, 6, 8 and 10 µM, respectively (Figure 5.10 A). Compound 25 had a Kb value of ~100 nM on HMDM cells.

The EC50 of 17 in the absence of peptide C3a antagonist (25) was 48 nM (pEC50 = 7.3 ± 0.2, n = 4) and EC50 was higher in the present of increasing concentration of 25. EC50 was 138 nM (6.9 ± 0.3), 419 nM (6.4 ± 0.1), 640 nM (6.2 ± 0.1) and 549 nM (6.3 ± 0.1), 1.2 uM (5.9 ± 0.1) and 2.2 µM (5.7 ± 0.1) in the presence of 200 nM, 2, 4, 6, 8 and 10 µM of compound 25. A Schild plot (Figure 5.10 B) supports the compound 25 being a non-competitive and insurmountable antagonist of 17, since there is a linear slope of 0.7 ± 0.13 (slope <1.0).

(A).

"## - " + + = 7 )

+ $# 9 0 : + 7 + 9

# %# 8 # 7 " 6 + ) ) &# < 5 + 4 7 3 + 8

: '# + ) 2 ; 8

7 # 6

!"# !$ !% !& !' ()*+,"-.+/01

214 CHAPTER 5 Antagonism of C3aR Agonists

(B). )#% 2

( *+,!-%$%.'/%(.'01 ! - 2#%$%.')( + 9 (#$ 8 7 6 -

5 !" $%&'() (#% # 4 + 3 0 - %#$ , + * %#% !"#$ !"#% !&#$ !&#% !$#$ !$#% !'#$ *+,-.)$/-012

Figure 5.10 Compound 25 is a non-competitive and insurmountable antagonist against 17, (A) shows concentration - dependent (zero (l), 200 nM (n), 2 µM () 4 µM (®), 6 µM (), 8 µM (£) and 10 µM (r) inhibition of iCa2+ release in HMDM induced by varying concentrations (3 nM to 100 µM) of 17. Data is expressed as % of maximal change in fluorescence induced by 17 (100 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for C3a antagonist 25 against 17 (slope <1) (n≥3).

Compound 25 is a non-competitive and insurmountable antagonist against compound 20

Similar results were found for 25 against agonist 20 as against 17. Compound 25 inhibited 20 in an insurmountable fashion. The increasing concentration of peptide antagonist 25 reduced the 2+ agonist response without a shift in the EC50 value. The maximal Ca release (100 µM) mediated by agonist 20 was inhibited to 95.9 ± 7.6 %, 82.6 ± 4.8 %, 85.2 ± 9.6 %, 75.3 ± 28.7 % and 28.8 ± 8.9 % by peptide C3aR antagonist (25) at concentrations increasing from 500 nM to 1, 2, 4 and 8 µM, respectively (Figure 5.11 A). The Schild plot (Figure 5.11 B) from these results is in contrast to the results obtained for 17. A Schild plot for antagonism by 25 against peptide agonist 20 showed a linear slope of 1.9 ± 0.3 (slope >1) which may indicate positive cooperativity in the binding of 25, depletion of a potent antagonist from the medium by receptor binding or non-specific binding or lack of antagonist equilibrium (Kenakin, 1997).

215 CHAPTER 5 Antagonism of C3aR Agonists (A).

(B).

( !"#$%&'&()*&(&+), 2 . ) - '&+)*, ! - $/ &'&0)1

+ . 9 ) 8 7 6 -

5 $ 4 + 3 0 - !) , + * !( !"#$ !%#& !%#$ !&#& !&#$ !'#& *+,-.(&/-012

Figure 5.11 Compound 25 is a non-competitive and insurmountable antagonist against 20, (A) shows concentration - dependent (zero (n), 500 nM (l), 1 µM (n), 2 µM () 4 µM () and 8 µM (®)) inhibition of iCa2+ release in human dU937 cells induced by varying concentrations (10 nM to 300 µM) of 20. Data is expressed as % of maximal change in fluorescence induced by 20 (1 mM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for C3aR peptide antagonist 25 against 20 (slope >1).

216 CHAPTER 5 Antagonism of C3aR Agonists 5.3.2.3 Compound 25 is a Non-competitive and Insurmountable Antagonist of Agonists 63 and 80

Compound 25 is a non-competitive and insurmountable antagonist of compound 63

Compound 25 inhibited 63 in an insurmountable fashion as increasing concentrations of 25 reduced the maximal response achievable in a dose-dependent manner of 63. At the higher concentrations of compound 25, 63 was not able to achieve full receptor activation because it was not able to displace compound 25 bound to C3aR. The maximal response of compound 63 mediated Ca2+ release (100 µM) was inhibited to 83.6 ± 3.2 %, 73.1 ± 0.06 %, 54.3 ± 3.7 % and 24.2 ± 2.2 % by peptide C3aR antagonist (25) concentration at 500 nM, 2, 4 and 8 µM, respectively (Figure 5.12

A). The EC50 of 63 in the presence and absence of 25 is not significantly changed (EC50 =10 -18 nM). Kb value = 398 nM. A Schild plot (Figure 5.12 B) reinforces the conclusion that compound 25 is a non-competitive and insurmountable antagonist of C3aR agonist 63.

(A). 3 $ / # & (

- "$$ @ 2 + - = 1 - 2 )( 9 + - ; $ < " - 9 ($ + ; ? - : 9 9 8

: #( + < 7 + 6 > 5

- $ : 9 8 4 !"# !"$ !% !& !' *+,-.&/0-123

217 CHAPTER 5 Antagonism of C3aR Agonists

(B).

$#) 2 !"#$%&'&()**&(&()+ :

! +

- , &'&()-. + 9 $#% 8 7 6 - $/ '&-)*

5 + $#' 4 + 3 0 - $#( , + * $#$ !"#$ !%#& !%#$ !&#& !&#$ !'#& *+,-.(&/-012

Figure 5.12 Compound 25 is a non-competitive and insurmountable antagonist against non-peptide agonist (63). (A) showing concentration-dependent (zero (l), 500 nM (n), 2 µM (), 4 µM (), 8 µM (®)) inhibition of iCa2+ release in human monocyte derived macrophage cells (HMDM) induced by varying concentrations (0.1 nM to 10 µM) of 63. Data is expressed as % of maximal change in fluorescence induced by 63 (10 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for C3a peptide antagonist 25 against 63 (slope <1).

Compound 25 is a non-competitive and insurmountable antagonist of agonist 80

Similar results were found for compound 80 which inhibited stimulation of calcium release by 25 in an insurmountable manner, with the increasing concentration of antagonist reducing the maximal response in a dose dependent manner. The maximal Ca2+ release mediated by 80 (100 µM) was inhibited to 84.4 ± 3.3 %, 66.1 ± 2.3 %, 71.1 ± 3.1 % and 40.8 ± 3.1 % by peptide C3aR antagonist (25) at concentrations 1, 2, 4 and 8 µM, respectively (Figure 5.13 A). A Schild plot (Figure 5.13 B) suggests that compound 25 is a non-competitive and insurmountable antagonist of C3aR agonist 80.

218 CHAPTER 5 Antagonism of C3aR Agonists

(A). $ + ' ' 8 & *(% $ 7 < 6 " $

5 *)) ) * $ + 1 $ 3

' #% 4 ' 1 ; $

3 %) " 2 : $ 1 1 0 (% 2 " 4 / .

" ) - $ 9 !" !# !$ !% !& !' 2 $

1 !(% 0 , !"#$%&'($)*+

(B).

Figure 5.13 Compound 25 is a non-competitive and insurmountable antagonist against non-peptide agonist (80). (A) shows concentration - dependent antagonism by 25 (zero (l), 1 µM (), 4 µM (®), 8 µM () of iCa2+ release in human monocyte-derived macrophage cells (HMDM) induced by varying concentrations (10 nM to 100 µM) of 80 (n ≥3). Data is expressed as % of maximal change in fluorescence induced by 80 (100 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for C3a peptide antagonist 25 against agonist 80 (slope <1).

219 CHAPTER 5 Antagonism of C3aR Agonists 5.3.3 Competition of Non-peptide Antagonist 146 with C3aR Agonists

5.3.3.1 Compound 146 is a Non-competitive and Insurmountable Antagonist of hC3a

From the studies of antagonist mechanisms, we found that human C3a mediated calcium release was non-competitive and insurmountably inhibited by C3aR antagonist, SB290157 analogue (146). This was indicated by a vertical reduction in the agonist (hC3a) induced response without any shift of EC50 even at the higher concentrations of agonist. This means that the human C3a was not able to achieve full receptor activation due to it being unable to compete with 146 binding to C3aR. These results could be because 146 binds to C3aR in an irreversible or pseudo-irreversible binding pattern or it binds at an allosteric binding site on the receptor. The maximal response of hC3a (1 µM) mediated Ca2+ release was inhibited to 55.1 ± 6.4 %, 51.8 ± 8.6 %, 59.1 ± 62.8 %, 23.0 ± 4.7 % and 18 ± 3.5 % by C3aR antagonist (146) concentration at 100 nM, 300 nM, 1, 3 and 10 µM, respectively (Figure 5.14 A).

The EC50 of hC3a in the absence of C3a antagonist (146) was 28.2 nM (pEC50 = 7.5 ± 0.07, n = 4) and EC50 was higher in the present of increasing concentrations of antagonist 146. EC50 was 124 nM (6.9 ± 0.14), 129 nM (6.9 ± 0.17), 440 nM (6.36 ± 0.44), 235 nM (6.6 ± 0.24) and 313 nM (6.5 ± 0.1) in the presence of 100 and 300 nM, 1, 3 and 10 µM of compound 146.

(A).

220 CHAPTER 5 Antagonism of C3aR Agonists

A schild plot (Figure 5.14 B) supports the compound 146 being a non-competitive and insurmountable antagonist of human C3a, since there is a linear slope of 0.25 ± 0.03 (slope < 1.0). (B).

*)& 3 *

! )*+!,$%$-'#.(-'-/ . ,

: #$

9 0 %$-'(1

8 *)( 7 . 6 5 !" $%$&'( , #

4 ()& 1 . - , + ()( !" !# !$ !% !& !' +,-./*'%0.123

Figure 5.14 Compound 146 is a non-competitive and insurmountable antagonist against hC3a on human monocyte derived macrophages. (A) shows concentration -dependence of 146 (zero (l), 0.1 (¾), 0.3 (p), 1 (), 3(®), 10 () µM) measured by the inhibition of iCa2+ release in HMDM which was induced by varying concentrations of human C3a (0.01 nM to 1 µM) (n ≥3). Data is expressed as % of maximal change in fluorescence induced by hC3a (1 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for 146 against hC3a (slope < 1).

5.3.3.2 Compound 146 is a Competitive and Surmountable Antagonist of Agonist 17

Compound 146 or C3aR non-peptidic antagonist inhibited the dose response of 17 in the calcium mobilisation assay by shifting the dose response curves to the right (horizontal shift). However, the maximal agonist response of these compounds could be restored by increasing the concentration of agonist compounds. These are characteristics of a competitive and surmountable antagonist. The EC50 of 17 in the absence of 146 was 87 nM (pEC50 = 7.06 ± 0.42, n=4) and EC50 was higher in the presence of increasing concentrations of 146. EC50 was 301.8 nM (6.52 ± 0.12), 1.6 µM (5.79 ± 0.12), 12.2 µM (4.92 ± 0.15), 23.4 µM (4.63 ± 0.11) and 18.7 µM (4.73 ± 0.08) in the presence of 300 nM, 1, 4, 8, 10 µM of 146.

221 CHAPTER 5 Antagonism of C3aR Agonists

The results (Figure 5.15 A) were converted to a Schild plot (Figure 5.15 B) to calculate the

Schild slope factor and pA2 value of 146 against 17. The Schild slope is an indication of the competitive relationship between agonist and antagonist. pA2 value of antagonist represents the negative logarithm of the concentration of antagonist compound 146 required to cause a two-fold rightward shift of agonist (17). (A). $ + ' 9 8 & '%& $ 7 < " 6 $

$ '&& 5 ) * $ + 1 "& $ 3 9 4 9 #& 1 & $ 3 " 2

; $& $ 1 0 1 " 2 %& / 4 . " -

$ & :

2 !" !# !$ !% $

1 !%& 0 , !"#$%&'($)*+

(B). * / 012+3.-.(#*.(.$#( .

- )

$ 4 .-.$#/5 " , +

* ) ) $ +, -.%#/ ( ) ' "

& ( % $ # " ! $ !"#$ !%#& !%#$ !&#& !&#$ !'#& !"#$0.123$%4/

Figure 5.15 Compound 146 is a competitive antagonist against 17 on human monocyte derived macrophages. (A) showing shifting of EC50 values due to competition at various concentrations of 146 (zero (l), 0.3 (¾), 1 (p), 4 (), 8 (®), 10 () µM) as measured by the inhibition of iCa2+ release in HMDM which was induced by varying concentrations of C3a peptide agonist 17 (3 nM to 100 µM)(n ≥3). Data is expressed as % of maximal change in fluorescence induced by. 17 (100 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for 146 against 17 (slope ≥ 1). Calculated pA2 value for 146 against 17 was 6.9 (~126 nM). 222 CHAPTER 5 Antagonism of C3aR Agonists The Schild plot suggests that 146 is a surmountable antagonist of C3aR against 17. The slope factor of approximately 1.0 (1.3 ± 0.1), suggesting a competitive relationship between these two compounds is equal to 1:1. The calculated pA2 value indicates that the concentration of

SB290157 required to shift the 17 curve by two fold is ~126 nM (pA2 = 6.9, Figure 5.15 B).

5.3.3.3 Compound 146 is a Competitive and Surmountable Antagonist of Non- Peptide Agonist 63

Compound 146 is also a competitive antagonist (reversible and surmountable) against non- peptidic agonist 63. Increasing concentrations of 146 resulted in a horizontal shift in the agonist concentration-response curves (Figure 5.16 A). 146 was also a surmountable antagonist of 63, since it could be fully displaced by 63 at high concentrations.

The EC50 of 63 in the absence of 146 was 8.9 nM (pEC50 = 8.05 ± 0.24, n=3) and EC50 was higher in the present of increasing concentrations of 146. EC50 was 38 nM (7.42 ± 0.30), 74.5 nM (7.13 ± 0.15), 418.7 nM (6.38 ± 0.16), 985 nM (6.01 ± 0.1) and 2.52 µM (5.6 ± 0.19) in the presence of 100 nM, 300 nM, 1, 3, 10 µM of compound 146. The Schild plot suggested that 146 was a surmountable antagonist of C3aR against 63. The slope factor of approximately 1.0 (1.01 ± 0.08), suggesting a competitive relationship between these two compounds is equal to 1:1. The calculated pA2 value indicates that the concentration of 146 required to shift the 63 curve by two fold is ~ 32 nM (pA2 = 7.5, Figure 5.16 B).

(A). $ + ' 8 7 & "'# $ 6 < " * $ $ "## 5 ) $ * 1 + $

3 $# 8 4 8 1 ; $

3 %# " 2 : $ 1 0 1 &# " 2 / 4 . "

- '# $ 9 2 $

1 # 0 ,

!"# !$ !% !& !"#$%&'($)*+

223 CHAPTER 5 Antagonism of C3aR Agonists

(B).

* /

. +,-./010(2'(0(0'2'" - $ )

" 3 010'24" , +

* ) ) $ (

' .5)010#2% "

& ( % $ # " ! ' !" !# !$ !% !& !"#$0.123$%4/

Figure 5.16 Compound 146 is a competitive antagonist against 63 on human monocyte derived macrophages, (A) showing shifting of EC50 values due to competition at various concentrations of compound 146 (zero (n), 0.1 (), 0.3 (), 1 (®), 3 (), 10 (£) µM) as measured by the inhibition of iCa2+ release in HMDM induced by varying concentrations of C3a peptide agonist 63 (0.3 nM to 10 µM)(n ≥3). Data is expressed as % of maximal change in fluorescence induced by 63 (100 µM) mediated iCa2+ mobilisation ± S.E.M. (B) Schild plot for 146 against 63 (slope =1). Calculated pA2 value for 146 against 63 was 7.5 (~ 32 nM).

5.4 DISCUSSION

C3a is proposed to interact with C3aR in a similar manner to how C5a is thought to interact with C5aR. This may involve two major sites of C3a interacting with the receptor in a two-site binding model, involving a high affinity recognition binding site and a low affinity activation site on the receptor (Chao et al., 1999; Ember et al., 1997). Human C3aR was cloned in 1996 (Ames et al., 1996; Crass et al., 1996; Roglic et al., 1996) revealing an extraordinarily large second extracellular loop between transmembrane domain 4 and 5. This 2nd extracellular loop is thought to be involved in the recognition and binding site of C3a (Chao et al., 1999). The activation site of C3aR responsible for triggering cell function is thought to be bound by C-terminal residues of C3a

224 CHAPTER 5 Antagonism of C3aR Agonists in a site on C3aR that may be in the seven transmembrane helices in the membrane of the cell (Figure 5.1). The C-terminal octapeptide region of C3a, implicated as the effector region required for activation of C3aR, has been used as a template for analogue development of C3aR agonists and antagonists. However, until this thesis there was no truly potent or selective small molecule agonist or antagonist reported for C3aR and so there have been limited tools available for probing the pharmacological roles of human C3a in physiology or diseases. The main C3aR antagonist known is SB290157 but, while it has been used to study in vitro 2+) and vivo the roles of C3a and its receptor, it is not a particularly potent antagonist (IC50 2 µM, Ca and has also shown agonist properties in some assays (Mathieu et al., 2005). In our hands, SB290157 had no agonist activity in the calcium mobilisation assay on HMDM and other cells. We have examined the nature of the antagonism for SB290157 and new, more potent, C3aR antagonists recently developed in our laboratories in the functional assay or calcium mobilisation assay. SB290157 displayed non-competitive and insurmountable antagonism against human C3a. In contrast, it was a competitive and surmountable antagonist against both potent peptide agonists of C3aR (17 and 20) and potent non-peptidic agonists (63 and 80). Insurmountable antagonism is an inhibition that has showed a significant reduction in the maximum response attained by the agonist in the presence of increasing antagonist concentrations. This may involve different processes such as physiological antagonism (two drugs with opposing actions in the body), allosteric binding of the receptor, irreversible or pseudo-irreversible binding to the receptor, receptor internalisation (decreased number of receptors available on the cell surface) and the activation states of the receptor (Kenakin, 1997). Pseudo-irreversible or irreversible antagonism is a competitive antagonism that occurs when the antagonist dissociates very slowly, or not at all, from the receptors, with the result that no change in the antagonist occupancy takes place when the agonist is applied. The mechanism of insurmountable antagonism of SB290157 against hC3a from this study is likely to be due to it binding pseudo-irreversibly to C3aR or it may have caused receptor internalisation that reduced the number of receptors at the cell surface. However, it is unlikely that SB290157 binds at an allosteric site on the receptor as it showed that it could compete with iodine labelled C3a in competitive binding assays. In contrast, the antagonism of SB290157 against peptide agonists and non-peptide agonists it is a competitive and surmountable antagonism of these agonists. The results could explain that SB290157 binds at the same site as peptide agonists and non-peptide agonists. A peptide antagonist (25) of C3aR, derived from the C-terminal region of C3a, was reported in Chapter 2 and is now published (Scully et al., 2010). From figure 5.17 A showed the overlayed

225 CHAPTER 5 Antagonism of C3aR Agonists of 25 on the oxazole compound (88) which showed that the last four residues of 25 form a beta-turn comformation which preferred recognised of ligands for GPCRs. This would explain the high affinity of this compound for C3aR (IC50 = 238 nM). This compound inhibited the C3a (100 nM) induced response in dU937 with IC50 =1.5 µM which were equipotent to SB290157 (IC50 = 1.3 µM). The hexapeptide antagonist 25 has showed non-competitive and insurmountable antagonism against (1) hC3a, (2) peptide agonists (17 and 20) and (3) a non-peptide agonist (63) (Figure 5.17 B), since depression of agonist response by varying concentrations of any of these agonists did not change the IC50 of compound 25. This insurmountable antagonism has several possible alternative explanations. The antagonist may interact at the allosteric site on the receptor (i.e. not the orthosteric site where the agonist binds) or it may bind pseudo-irreversibly. However, the mechanism of insurmountable antagonism of 25 against all agonists (hC3a, 17, 20, 63 and 80) from this study is unlikely to be due to an allosteric interaction on the receptor as in the previous study it showed that it could compete with [125I]-C3a in competitive binding assay. It is more likely that 25 binds tightly and pseudo-irreversibly to C3aR, in other words with a slow off rate leading to a long residence time on the receptor. Compound 146, an analogue of SB290157, is a new antagonist that has been discovered in our studies (Figure 5.17 C). Figure 5.17 C demonstrated the superimposed of the furan compound (144) with SB290157 showed that the ether oxygen atom of SB290157 align with the oxygen, nitrogen or sulfur atom of furan. These atoms (oxygen, nitrogen or sulfur) are hydrogen bond acceptors which contribute to the receptor binding potency of the compounds. It had affinity for 125 C3aR (IC50 = 91 nM against 80 pM [ I]-C3a) in HMDMs, but it had no agonist activity up to 100 µM in calcium mobilisation on HMDM. It inhibited the native human C3a (100 nM) response with

IC50 = 534 nM. Interestingly, the results of antagonism of 146 against hC3a and peptide and non- peptide agonists are similar to those of SB290157, consistent with their structural similarity. Compound 146 is a non-competitive and insurmountable antagonist of hC3a (Schild plot slope < 1.0), but it is a competitive and surmountable antagonist of peptide (17) and non-peptide (63) agonists. It could be that compound 146 binds to the receptor at the same site as SB290157, while 25 binds to a different site from these two antagonists. Thus the component of the antagonists that appears to govern the finer differences in where all these compounds are binding on C3aR appears not to be the common arginine, or similar linker (ether oxygen or heterocylic hydrogen-bonding acceptor), but rather the different hydrophobic component at the N-terminus of our tripeptide-like surrogates (Figure 5.17). In conclusion, the antagonists (SB290157 and SB analogue 146) are suggested to bind to C3aR at the same site as the peptide agonists (17 and 20) and non-peptide agonists (63 and 80) which are all shown to be competitive ligands for C3aR. In contrast, peptide antagonist (25) binds

226 CHAPTER 5 Antagonism of C3aR Agonists pseudo-irreversibly (rather than allosterically) to C3aR. In competition studies, SB290157 has pA2 values (pA2 value of antagonist represents the negative logarithm of the concentration of antagonist required to cause a two-fold rightward shift of agonist) against all agonists is in following order: 20 (63 nM) < 17 (100 nM) < 63 (126 nM) < 80 (158 nM). This reveals that the non-peptide agonists (63 and 80) bind to C3aR more tightly than peptide agonists, thus they require more concentration of SB290157 to shift the EC50 value of agonist two-fold and concentration-dependent agonist curves to the right. In contrast, the value of pA2 in the competitive studies with 146 found that the concentrations required to cause a two-fold rightward shift are lower with compound 63 (pA2= 32 nM) compared to peptide agonist (17), which has pA2 = 126 nM. These results need further study by testing more compounds, conducting molecular modelling studies, and chimeric and deletion mutagenesis of C3aR.

227 CHAPTER 5 Antagonism of C3aR Agonists

Figure 5.17 A. Superimposition of the oxazole (green) (Boc-Leu-oxazole-Arg, 88) with the last four C-terminal residues of FLTChaAR (25) (cyan, NMR structure). B. SB290157 (grey) aligned with energy minimised compound 63 (green) C. Alignment of furan (cyan) and SB290157 (green) Arg residue not shown (courtesy of Dr. Reid).

228 CHAPTER 5 Antagonism of C3aR Agonists 5.5 MATERIALS AND METHODS

5.5.1 Calcium mobilization assay on HMDM cells

Harvested HMDM cells were washed with 0.9% NaCl solution by centrifugation at 2500 rpm for 5 min, followed by resuspension of the cell pellet with complete media. Cells were seeded at 5 x 104 cells/well in a 96-well black-wall, clear bottom plate (DKSH, Zurich), and left overnight to adhere at 37°C. On the day of the experiment, supernatant was removed and cells were incubated in dye loading buffer. Dye-loading buffer was prepared by using 12 mL of wash buffer (Hank’s Balanced Salt Solution (1x HBSS, 20 mM HEPES, 2.5 mM probenecid, pH 7.4) with 25 µL Fluo-3 (final concentration 4 µM), 25 µL 20% pluronic acid, 1% fetal bovine serum (FBS) and incubated in a cover plate for one hour at 37°C. The supernatant was removed and cells were then washed once with HBSS. Then 100 µL of test compound or buffer or control/well was plated out on sterile black-wall, clear-bottom 96-well plates (Corning Incorporated, NY). Plate was then transferred to a Fluorostar spectrofluorimeter (BMG LabTechnologies, Offenburg, Germany), where fluorescence was measured over 1 min with excitation at 485 nm and emission at 520 nM of Fluo-3 bound Ca2+ complex, at 28 oC. Agonist responses were expressed as a percentage of native human C3a at 100 nM (or calcimycin) as the maximum change in fluorescence through emission from the Fluo-3 bound calcium complex.

5.5.1.1 Antagonist Surmountability/Insurmountability:

Cells were prepared as for the calcium mobilisation assay after incubation with dye loading buffer for 1 h, cells were incubated with different concentrations of antagonist for 15 mins before starting the experiment. The plate was then transferred to a Fluorostar spectrofluorimeter (BMG LabTechnologies, Offenburg, Germany), and examined for concentration-dependent effects on the activity of agonists in the absence and presence of different concentrations of antagonist.

5.5.1.2 Data Analysis

In Functional Assays, the concentration curves were analysed by non-linear regression

(GraphPad Prism 5, USA) and EC50 and pEC50 determined. A pEC50 for each compound was calculated from separate experiments and expressed as an arithmetic mean ± SEM. Functional activity is expressed as a percentage of the maximal response induced by hC3a (100 nM) or 100 µM of agonist compounds for calcium mobilisation of human derived macrophage cells. Non-linear regression was performed. SB290157 (C3aR antagonist) has been shown to insurmountably antagonise the C3aR which might be attributed to pseudo-irreversible antagonism.

229 CHAPTER 5 Antagonism of C3aR Agonists The Kb was calculated from the equation Kb=[antagonist]/slope-1 (Kenakin, 1997) This was obtained from the linear regression of 1/[A] versus 1/[A’] where A is the agonist concentration and A’ is the equi-active concentration of agonist in the presence of antagonist. The more accurate estimates of Kb are derived from experiments in which the maximal response to the agonist is depressed by the antagonist to less than 50% of the maximal response, so this calculation was performed using a concentration of antagonist which gave at least 50% inhibition.

230 CHAPTER 5 Antagonism of C3aR Agonists

5.6 REFERENCES

1. AMES, RS, LI, Y, SARAU, HM, NUTHULAGANTI, P, FOLEY, JJ, ELLIS, C, ZENG, Z, SU, K, JUREWICZ, AJ, HERTZBERG, RP, BERGSMA, DJ, KUMAR, C (1996) Molecular cloning and characterization of the human anaphylatoxin C3a receptor. Journal of Biological Chemistry 271(34): 20231-20234.

2. ANDERSSON, M, MICHEL, L, LLULL, JB, PIPKORN, U (1994) Complement activation on the nasal mucosal surface- a feature of the immediate allergic reaction in the nose. Allergy 49(4): 242-245.

3. BAO, LH, OSAWE, I, HAAS, M, QUIGG, RJ (2005) Signaling through up-regulated C3a receptor is key to the development of experimental lupus nephritis. J. Immunol. 175(3): 1947-1955.

4. CAPORALE, LH, TIPPETT, PS, ERICKSON, BW, HUGLI, TE (1980) The active-site of C3a anaphylatoxin Journal of Biological Chemistry 255(22): 758-763.

5. CHAO, TH, EMBER, JA, EMBER, JA, WANG, MY, BAYON, Y, HUGLI, TE, YE, RD (1999) Role of the second extracellular loop of human C3a receptor in agonist binding and receptor function. Journal of Biological Chemistry 274(14): 9721-9728.

6. CHAZIN, WJ, HUGLI, TE, WRIGHT, PE (1988) H-1-NMR studies of human C3a anaphylatoxin in solution-sequential resonance assignments, secondary structure, and global fold. Biochemistry 27(26): 9139-9148.

7. CHENOWETH, DE, HUGLI, TE (1980) Human C5a and C5a analogs as probes of the neutrophil C5a receptor. Mol. Immunol. 17(2): 151-161.

8. CRASS, T, AMES, RS, SARAU, HM, TORNETTA, MA, FOLEY, JJ, KOHL, J, KLOS, A, BAUTSCH, W (1999) Chimeric receptors of the human C3a receptor and C5a receptor (CD88). Journal of Biological Chemistry 274(13): 8367-8370.

9. CRASS, T, RAFFETSEDER, U, MARTIN, U, GROVE, M, KLOS, A, KOHL, J, BAUTSCH, W (1996) Expression cloning of the human C3a anaphylatoxin receptor (C3aR) from differentiated U- 937 cells. European Journal of Immunology 26(8): 1944-1950.

10. DAFFERN, PJ, PFEIFER, PH, EMBER, JA, HUGLI, TE (1995) C3a is a chemotaxin for human eosinophils but not for neutrophils. I. C3a stimulation of neutrophils is secondary to eosinophil activation. Journal of Experimental Medicine 181(6): 2119-2127.

11. DROUIN, SM (2001) Expression of the Complement Anaphylatoxin C3a and C5a receptors on Bronchial Epithelial and Smooth muscle cells in Models of Sepsis and Asthma. J. Immunol. 166: 2025-2032.

12. DROUIN, SM, CORRY, DB, HOLLMAN, TJ, KILDSGAARD, J, WETSEL, RA (2002) Absence of the complement anaphylatoxin C3a receptor suppresses Th2 effector functions in a murine model of pulmonary allergy. J. Immunol. 169(10): 5926-5933.

231 CHAPTER 5 Antagonism of C3aR Agonists

13. DROUIN, SM, CORRY, DB, KILDSGAARD, J, WETSEL, RA (2001) Cutting edge: The absence of C3 demonstrates a role for complement in Th2 effector functions in a murine model of pulmonary allergy. J. Immunol. 167(8): 4141-4145.

14. EMBER, JA, HUGLI, TE (1997) Complement factors and their receptors. Immunopharmacology 38(1-2): 3-15.

15. FISCHER, WH, JAGELS, MA, HUGLI, TE (1999) Regulation of IL-6 synthesis in human peripheral blood mononuclear cells by C3a and C3a(desArg). J. Immunol. 162(1): 453-459.

16. FUKUOKA, Y, EMBER, JA, HUGLI, TE (1998) Molecular cloning of two isoforms of the guinea pig C3a anaphylatoxin receptor: Alternative splicing in the large extracellular loop. J. Immunol. 161(6): 2977-2984.

17. HARTMANN, K, HENZ, BM, KRU?GER-KRASAGAKES, S, KO?HL, J, BURGER, R, GURTL, S, HAASE, I, LIPPERT, U, ZUBERBIER, T (1997) C3a and C5a stimulate chemotaxis of human mast cells. Blood 89(8): 2863-2870.

18. HOEPRICH, PD, HUGLI, TE (1986) Helical conformation at the carboxy-terminal portion of human C3a is required for full activity. Biochemistry 25(8): 1945-1950.

19. HUGLI, TE (1975) Human anaphylatoxin (C3a) from 3rd component of complement- primary structure. Journal of Biological Chemistry 250(21): 8293-8301.

20. HUMBLES, AA, LU, B, NILSSON, CA, LILLY, C, ISRAEL, E, FUJIWARA, Y, GERARD, NP, GERARD, C (2000) A role for the C3a anaphylatoxin receptor in the effector phase of asthma. Nature 406(6799): 998-1001.

21. HUTAMEKALIN, P, TAKEDA, K, TANI, M, TSUGA, Y, OGAWA, N, MIZUTANI, N, YOSHINO, S (2010) Effect of the C3a-Receptor Antagonist SB 290157 on Anti-OVA Polyclonal Antibody-Induced Arthritis. J. Pharmacol. Sci. 112(1): 56-63.

22. JUN, SW, KIM, TH, LEE, HM, LEE, SH, KIM, WJ, PARK, SJ, KIM, YS (2008) Overexpression of the anaphylatoxin receptors, complement anaphylatoxin 3a receptor and complement anaphylatoxin 5a receptor, in the nasal mucosa of patients with mild and severe persistent allergic rhinitis. J. Allergy Clin. Immunol. 122(1): 119-125.

23. KAPP, A, WOKALEK, H, SCHOPF, E (1985) Involvement of complement in psoriasis and atopic-dermatitis measurement of C3a and C5a,C-3, C-4 and C1 inactivator. Arch. Dermatol. Res. 277(5): 359-361.

24. KAWAI, M, QUINCY, DA, LANE, B, MOLLISON, KW, LULY, JR, CARTER, GW (1991) Identification and synthesis of a receptor-binding site of human anaphylatoxin C5a. Journal of Medicinal Chemistry 34(7): 2068-2071.

25. KENAKIN, TP (1997) Pharmacologic analysis of drug-receptor interaction. 3rd edn. Lippincott-Raven Publishers: Philadelphia

26. KILDSGAARD, J, HOLLMANN, TJ, MATTHEWS, KW, BIAN, K, MURAD, F, WETSEL, RA (2000) Cutting edge: Targeted disruption of the C3a receptor gene demonstrates a novel protective anti-inflammatory role for C3a in endotoxin-shock. J. Immunol. 165(10): 5406-5409.

232 CHAPTER 5 Antagonism of C3aR Agonists

27. KOHL, J, CASARETTO, M, GIER, M, KARWATH, G, GIETZ, C, BAUTSCH, W, SAUNDERS, D, BITTERSUERMANN, D (1990) Reevaluation of the C3a active-site using short synthetic C3a analogs. European Journal of Immunology 20(7): 1463-1468.

28. KRETZSCHMAR, T, JEROMIN, A, GIETZ, C, BAUTSCH, W, KLOS, A, KOHL, J, RECHKEMMER, G, BITTERSUERMANN, D (1993) Chronic myelogenous leukemia-derived basophilic granulocytes express a functional active receptor for anaphylatoxin C3a. European Journal of Immunology 23(2): 558-561.

29. MAMANE, Y, CHAN, CC, LAVALLEE, G, MORIN, N, XU, LJ, HUANG, JQ, GORDON, R, THOMAS, W, LAMB, J, SCHADT, EE, KENNEDY, BP, MANCINI, JA (2009) The C3a Anaphylatoxin Receptor Is a Key Mediator of Insulin Resistance and Functions by Modulating Adipose Tissue Macrophage Infiltration and Activation. Diabetes 58(9): 2006- 2017.

30. MARTIN, U, BOCK, D, ARSENIEV, L, TORNETTA, MA, AMES, RS, BAUTSCH, W, KOHL, J, GANSER, A, KLOS, A (1997) The human C3a receptor is expressed on neutrophils and monocytes, but not on B or T lymphocytes. Journal of Experimental Medicine 186(2): 199- 207.

31. MATHIEU, MC, SAWYER, N, GREIG, GM, HAMEL, M, KARGMAN, S, DUCHARME, Y, LAU, CK, FRIESEN, RW, O'NEILL, GP, GERVAIS, FG, THERIEN, AG (2005) The C3a receptor antagonist SB 290157 has agonist activity. Immunology Letters 100(2): 139-145.

32. MOCCO, J, MACK, WJ, DUCRUET, AF, SOSUNOV, SA, SUGHRUE, ME, HASSID, BG, NAIR, MN, LAUFER, I, KOMOTAR, RJ, HOLLAND, MCH, PINSKY, DJ, CONNOLLY, ES (2006) Complement component C3 mediates inflammatory injury following focal cerebral ischemia. Circ.Res. 99(2): 209-217.

33. PROCTOR, LM, ARUMUGAM, TV, SHIELS, I, REID, RC, FAIRLIE, DP, TAYLOR, SM (2004) Comparative anti-inflammatory activities of antagonists to C3a and C5a receptors in a rat model of intestinal ischaemia/reperfusion injury. British Journal of Pharmacology 142(4): 756-764.

34. ROGLIC, A, PROSSNITZ, ER, CAVANAGH, SL, PAN, ZX, ZOU, AH, YE, RD (1996) cDNA cloning of a novel G protein-coupled receptor with a large extracellular loop structure. Biochim. Biophys. Acta-Gene Struct. Expression 1305(1-2): 39-43.

35. SCULLY, CCG, BLAKENEY, JS, SINGH, R, HOANG, HN, ABBENANTE, G, REID, RC, FAIRLIE, DP (2010) Selective Hexapeptide Agonists and Antagonists for Human Complement C3a Receptor. Journal of Medicinal Chemistry 53(13): 4938-4948.

36. SICILIANO, SJ, ROLLINS, TE, DEMARTINO, J, KONTEATIS, Z, MALKOWITZ, L, VANRIPER, G, BONDY, S, ROSEN, H, SPRINGER, MS (1994) 2-Site binding of C5a by its receptor- an alternative binding paradigm for G-protein coupled receptors. Proc. Natl. Acad. Sci. U. S. A. 91(4): 1214-1218.

37. TAKABAYASHI, T, VANNIER, E, BURKE, JF, TOMPKINS, RG, GELFAND, JA, CLARK, BD (1998) Both C3a and C3a(desArg) regulate interleukin-6 synthesis in human peripheral blood mononuclear cells. J. Infect. Dis. 177(6): 1622-1628.

233 CHAPTER 5 Antagonism of C3aR Agonists

38. TAKABAYASHI, T, VANNIER, E, CLARK, BD, MARGOLIS, NH, DINARELLO, CA, BURKE, JF, GELFAND, JA (1996) A new biologic role for C3a and C3a desArg - Regulation of TNF- alpha and IL-1 beta synthesis. J. Immunol. 156(9): 3455-3460.

39. TANG, ZY, LU, B, HATCH, E, SACKS, SH, SHEERIN, NS (2009) C3a Mediates Epithelial-to- Mesenchymal Transition in Proteinuric Nephropathy. J. Am. Soc. Nephrol. 20(3): 593-603.

40. VALLOTA, EH, MULLEREB.HJ (1973) Formation of C3a and C5a anaphylatoxins in whole human-serum after inhibition of anaphylatoxin inactivator. Journal of Experimental Medicine 137(5): 1109-1123.

41. WENDERFER, SE, WANG, HY, KE, BZ, WETSEL, RA, BRAUN, MC (2009) C3a receptor deficiency accelerates the onset of renal injury in the MRL/Ipr mouse. Mol. Immunol. 46(7): 1397-1404.

42. ZHANG, XL, BOYAR, W, TOTH, MJ, WENNOGLE, L, GONNELLA, NC (1997) Structural definition of the C5a C terminus by two-dimensional nuclear magnetic resonance spectroscopy. Proteins 28(2): 261-267.

43. ZWIRNER, J, GOTZE, O, MOSER, A, SIEBER, A, BEGEMANN, G, KAPP, A, ELSNER, J, WERFEL, T (1997) Blood- and skin-derived monocytes/macrophages respond to C3a but not to C3a(desArg) with a transient release of calcium via a pertussis toxin-sensitive signal transduction pathway. European Journal of Immunology 27(9): 2317-2322.

234 CHAPTER 5 Antagonism of C3aR Agonists

235 CHAPTER 6 Summary

CHAPTER 6

SUMMARY

C3aR, which is a G protein–coupled receptor predominantly coupled to Gi. C3aR is found in both myeloid and non-myeloid cells. C3a and its receptor have been implicated in the pathogenesis of various autoimmune diseases and inflammatory conditions (e.g. asthma, sepsis, allergies, lupus erythematosus (SLE), diabetes, psoriasis, arthritis, nephropathy, ischemia-reperfusion injury and others). Potent and selective agonists or antagonists that selectively act on C3aR have enormous potential as physiological probes, anti-inflammatory and perhaps other classes of drugs. The agonist activity of C3a is due to the octapeptide fragment at the C-terminus, which triggers C3a receptor activation. The truncated and modified 15 residue C-terminal end of C3a, with addition of two tryptophans to it’s N-terminus, was reported to be 15-fold more potent than C3a in a guinea pig platelet aggregation assay. This superagonist activity has been disputed in the literature (Ames et al., 1997; Chao et al., 1999; Ischenko et al., 1998; Sun et al., 1999) and by our group in a previous study (Blakeney, 2007) since the compound showed less potency and no selectivity for C3aR. In any case, there is no potent agonist of C3aR composed of less than eight amino acids. The first chapter of our study focused on the development of C3a peptide derived ligands by studying both selectivity and potency of compounds. Firstly, we studied the affinity and selectivity of compounds using a competitive radioligand binding assay with [125I]-labelled C3a and [125I]- labelled C5a on peripheral blood mononuclear cells (PBMCs), which were isolated from human plasma. The general selectivity was also determined by a C3a-induced receptor desensitisation experiment which shows effects other than those mediated through C3aR. Finally, we determined the functional potency of the compounds by using an intracellular calcium release or mobilisation assay, which detected intracellular calcium trafficking by binding calcium to a cell permeable fluorophore.

236 CHAPTER 6 Summary We found that previously reported compounds, including the C3a “superagonist” (WWGKKYRASKLGLAR), were not selective for C3aR in a receptor desensitisation experiment. Four decapeptides, originally reported as C5aR agonists, also bound tightly to C3aR in PBMC cells. Two reported oryzatensin pentapeptide derivatives, WPLPR and YPLPR, also showed no selectivity for C3aR over C5aR in the competitive radioligand binding assay. A new series of hexapeptides based on the C-terminus of human C3a was designed and studied using structure- activity relationships (SAR). Two small peptide agonists were identify to be highly potent agonists and selective for C3aR. FLTLAR and FWTLAR bound to C3aR tightly with IC50 ~ 40 and 80 nM, respectively on PBMC cells and in the calcium mobilisation assay had EC50 = 0.32 and 0.37 µM, respectively. Structure activity relationships and competitive binding results have led to a new

C3aR peptide antagonist, F1LTChaAR6. Changing leucine at position 4 to a bulkier amino acid such as cyclohexylalanine (Cha) showed no agonist activity (>100 µM) in a calcium mobilisation assay in dU937 cells, but did bind tightly to C3aR and displayed antagonist activity (IC50 1.5 µM). However, C3a-derived peptides have significant disadvantages for uses in biology and medicine because of the poor bioavailability of peptides in general, as well as poor stability to protease enzymes, high clearance rates, and poor membrane permeability. Therefore, it is highly desirable to develop nonpeptidic C3a ligands that are more suitable for use as pharmacological/physiological probes for interrogating the roles of hC3a in vivo. As a step towards this goal, we introduced a heterocycle to restrict the conformational freedom of short peptides, in an attempt to mimic the receptor-binding conformation of the C-terminal activating domain of human C3a. A molecular modelling study (Figure 3.2) revealed structural similarity between the C3a carboxyl terminal residues (GLAR), modelled in a beta turn conformation, and the model of compound 56. This compound contains an oxazole heterocycle that acts as a beta turn-inducing constraint and mimics the conformation found for the C-terminus of C3a in its crystal structure. This result supports a large set of observations that GPCRs have a general tendency to recognise turn structural motifs in their protein and peptide ligands (Fairlie et al., 2005; Madala et al., 2010; Tyndall et al., 2005). In this study, multiple potent agonist compounds were discovered for C3aR. They competed strongly with 125I-labelled C3a in binding to, and activating intracellular calcium release in, human monocyte-derived macrophages (HMDM). Indeed, some of the compounds had equal or even greater affinity than C3a itself. Building on the discovery in Chapter 2 that hexapeptide agonists for C3aR agonist (e.g. FLTLAR) adopted a beta turn conformation at their C- terminus, it was decided to introduce an oxazole heterocycle into the LAR region to impart conformational rigidity to the agonist peptides (Figure 3.3). This modification produced new compounds with improved affinities for human C3aR. The oxazole has electronegative oxygen and 237 CHAPTER 6 Summary nitrogen atoms that can function as hydrogen bond acceptors, while introducing a rigid turn-like conformation to the molecule. The most potent and selective compounds in this study were 56, 63 and 75 (EC50 = 19, 7 and 17 nM, respectively) in HMDM cells. There was a linear correlation between the binding affinity of the compounds and their agonist potency. A potent C3aR agonist (56) was chosen for further study in Chapter 4. The incorporation of a phenyl group at the 5-position of the oxazole ring of compound 56 gave an improvement in binding affinity and eventually led to the novel and potent nonpeptidic C3aR antagonist compound (105). Further study by modification of the Boc group at the N-terminus of compound 105 also generated another potent C3aR antagonist compound (109), which suggested that the fifth position of the oxazole is the critical position for transforming an agonist to an antagonist. In addition, substitution at the fifth position of the oxazole ring and at the N-terminus with a bulkier group gave improvement in binding affinity and antagonist potency of the compound. A second approach to developing C3aR antagonists in Chapter 4 was to use the known, albeit weak, antagonist compound SB290157 as a lead compound for further modifications (Figure 4.1 B). SB290157 was the first reported “potent” C3a receptor antagonist in several in vitro assays and two in vivo models. In 2005, SB290157 was reported to also have “potent” agonist activities in a variety of in vitro and in vivo assays. We proposed to develop new non-peptidic antagonists using SB290157 as a lead compound, because it binds tightly to C3aR and, in our hands, it showed no agonist activity in the Ca2+ assay on human monocyte derived macrophages. Moreover, it also showed no cross reactivity at, or binding to, C5aR. The structure of SB290157 was modified in five different ways: (i) modifications to the linker region, while leaving unchanged the diphenylmethylene rings and arginine moieties, (ii) modifications to the aromatic and linker regions, where the ether bond and the linker were replaced by an amide functional group (R- glycine-arginine scaffold), (iii) replacement of the aromatic region by cinnamic acids and derivatives, while the ether bond and the linker region were replaced with an amide group (cinnamic acid-glycine-arginine), (iv) introduction to the linker region of a tryptophan residue and analogues with different substituents at the aromatic region, and (v) replacement of the linker by furan moieties, including functionalised analogues. We discovered two potent C3aR antagonists, which bound tightly to C3aR in HMDM cells. These compounds feature derivations of SB290157 with linker region replaced by a furan ring. The heterocycle was modified with varying hydrogen bond acceptors. The most potent and selective compounds in this study were 145 and 146 (IC50 460 and 534 nM, respectively) in the calcium mobilisation assays in HMDM cells. Binding affinity correlated with antagonist potency, and these compounds were by far the most potent C3aR antagonists reported to date.

238 CHAPTER 6 Summary To better understanding the binding site, the new potent and selective peptide antagonist 25 and non-peptide antagonist 146 have been compared in this chapter with SB290157 for their capacity to inhibit intracellular calcium release in HMDM stimulated by up to 100 µM of three different classes of C3aR agonists: human C3a, peptide agonists (17 and 20), and non-peptide agonists (63 and 80). The antagonists SB290157, 25 and 146 all displayed insurmountable and non-competitive antagonism against C3a for C3aR. This is attributed to either a possible allosteric binding site for the compounds on C3aR or possibly irreversible or pseudo-irreversible binding to the C3a receptor. By contrast, SB290157 was a surmountable and competitive antagonist against peptide agonists (17 and 20) and non-peptide agonists (63 and 80). Compound 25 was an insurmountable antagonist for all C3a agonists (hC3a, peptide and non-peptide agonists). Compound 146 behaved similarly to SB290157 in exhibiting surmountable and competitive antagonism of peptide and non-peptide C3aR agonists. These results point to a mechanism of insurmountable antagonism for compound 25 against all agonists (hC3a, 17, 20, 63 and 80) involving binding to C3aR that is unlikely to be at an allosteric site, since it competed with [125I]-C3a in the competitive binding assay. It is more likely that compound 25 binds pseudo-irreversibly to C3aR. The competitive antagonism study of C3aR antagonist C3aRA (SB290157) against all peptide and non-peptide agonists showed a concentration of C3aRA (SB290157) that caused a two-fold rightward shift of agonists in the following order: 20 (63 nM) < 17 (100 nM) < 63 (126 nM) < 80 (158 nM). This indicated that the non-peptide agonists (63 and 80) bound to C3aR more tightly than peptide agonists (17 and 20), thus requiring higher concentrations of C3aRA (SB290157) to shift the EC50 value of agonists two-fold to the right. The work presented in this thesis has extensively studied the discovery and development of C3aR-selective, small molecule agonists/antagonists with high affinity and potency for human C3aR. These could serve as chemical probes for revealing the roles of C3a and C3aR in immune and inflammatory physiology and pathophysiology. They might also provide some clues to drive the development of specific drugs to target C3aR and are produce benefits for the treatment of immune-deficiency or immunosuppression (C3aR agonist) or autoimmune, inflammatory, metabolic diseases (C3aR antagonist).

239 CHAPTER 6 Summary

240 APPENDIX

Selective Hexapeptide Agonists and Antagonists for Human Complement C3a Receptor. J. Med.Chem. 2010, 53, 4938-4948.

241 4938 J. Med. Chem. 2010, 53, 4938–4948 DOI: 10.1021/jm1003705

Selective Hexapeptide Agonists and Antagonists for Human Complement C3a Receptor

Conor C. G. Scully, Jade S. Blakeney, Ranee Singh, Huy N. Hoang, Giovanni Abbenante, Robert C. Reid, and David P. Fairlie* Division of Chemistry and Structural Biology, Institute for Molecular Bioscience, The University of Queensland, Brisbane, Queensland 4072, Australia

Received March 24, 2010

Human anaphylatoxin C3a, formed through cleavage of complement protein C3, is a potent effector of innate immunity via activation of its G protein coupled receptor, human C3aR. Previously reported short peptide ligands for this receptor either have low potency or lack receptor selectivity. Here we report the first small peptide agonists that are both potent and selective for human C3aR, derived from structure-activity relationships of peptides based on the C-terminus of C3a. Affinity for C3aR was examined by competitive binding with 125I-labeled C3a to human macrophages, agonist versus antagonist activity measured using fluorescence detection of intracellular calcium, and general selectivity monitored by C3a-induced receptor desensitization. An NMR structure for an agonist in DMSO showed a β-turn motif that may be important for C3aR binding and activation. Derivatization produced a noncompetitive and insurmountable antagonist of C3aR. Small molecule C3a agonists and antagonists may be valuable probes of immunity and inflammatory diseases.

Introduction opsonization, lysis, and removal of pathogens and infected or 1 3 damaged cells and cellular debris. Among complement acti- Human complement is a complex network - of plasma proteins that initiates inflammatory and cellular immune vation products are the anaphylatoxins C3a and C5a, which responses to stimuli like infectious organisms (bacteria/ prime and amplify the immune response through chemoat- viruses/parasites), chemical/physical injury, radiation, and traction of immune cells to inflammatory sites and by promoting secretion of proinflammatory agents, lysosomal enzymes, cancer. Complement activation is an important proteolytic 1-3 signaling cascade in the immune response to infection and and reactive oxygen species. Unlike C5a, which is almost tissue injury, its proteins cooperatively effecting recognition, undetectable in healthy individuals, there is considerable C3a (>100 nM) present due to continuous degradation of C3. 4a,b *To whom correspondence should be addressed. Phone: 61 7 While the pharmacology of C5a and its receptor has been 33462989. Fax: 61 7 33462990. E-mail: [email protected].þ strongly implicated in the pathogenesis of inflammatory dis- aAbbreviations:þ 3D53, C5aR antagonist cyclo-(2,6)-AcF[OP- eases through effects of gene deletion, genetic deficiency, anti- (dCha)WR]; Ahx, 6-aminohexanoic acid; Aib, 2-aminoisobutyric acid; bodies, and a synthetic antagonist, C3a or its receptor C3aRa Bt2-cAMP, 30,50- dibutyryladenosine monophosphate; ATP, adenosine triphosphate; C3aR, C3a receptor; C5aR, C5a receptor; Cha, cyclohex- has only more recently been implicated in the pathogenesis ylalanine; dU937, human monocytic cells from histiocytic lymphoma and progression of inflammatory diseases such as asthma, differentiated with Bt2-cAMP; DIPEA, N,N-diisopropylethylamine; allergies, sepsis, lupus erythematosus, diabetes, arthritis, psor- DMF, N,N-dimethylformamide; EC50, molar concentration that pro- 5-7 duces 50% of the maximum response of an agonist; FBS, fetal bovine iasis, nephropathy, ischemia-reperfusion injury, and others. serum; Fmoc, fluorenylmethoxycarbonyl; G-protein, guanosine mono- Antimicrobial and antifungal properties have also been re- phosphate protein; GPCR, G-protein-coupled receptor; HATU, 2-(1H- ported for C3a derived peptides.8 The 21-residue C-terminus 7-azabenzotriazol-1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate; HBSS, Hank’s balanced salt solution; HBTU, O-(1H-benzotriazol- of C3a and the 20-residue C-terminus of C3adesArg are also 1-yl)-1,1,3,3-tetramethyluronium hexafluorophosphate; HEK-293, hu- antibacterial against E. faecalis and P. aeruginosa,9 Escherichia man embryonic kidney 293 cells; HEPES, 4-(2-hydroxyethyl)-1-piper- coli, and Staphylococcus aureus10 and antifungal against Candi- azineethanesulfonic acid); HT-29, human colon adenocarcinoma grade 11 II cell line; hF, homophenylalanine; 125I, iodine isotope 125; IC , molar da albicans. Two nonapeptide derivatives of the C3a C-termi- 50 2 concentration of an unlabeled agonist/antagonist that inhibits 50% nus interfere with intracellular Ca þ release and ERK 1/2 radioligand binding or molar concentration of antagonist that inhibits phosphorylation by binding to and promoting phosphorylation 50% of a known concentration of agonist activity; Ig, immunoglobulin; 12 IMDM, Iscove’s modified Dulbucco’s medium; MBHA, methylbenzhy- of type 1 Fc ε receptor in mast cells. As more C3a-C3aR drylamine; Nal, 1-naphthylalanine; Nle, norleucine; NEAA, nonessen- pharmacology has become established, interest has been re- tial amino acids; NOE, nuclear Overhauser effect; NOESY, nuclear newed in C3a and there is a realization of the need for synthetic Overhauser effect spectroscopy; Orn, ornithine; Pbf, 2,2,4,6,7-penta- C3a agonists and antagonists for interrogating and potentially methyldihydrobenzofuran-5-sulfonyl; PBMC, peripheral blood mono- 5-7 nuclear cells; rmsd, root-mean-square deviation; ROE, rotating-frame treating human C3a-mediated conditions. Overhauser enhancement; ROESY, rotating-frame NOE spectroscopy; Human complement component C3a is a 76 residue protein rpHPLC, reversed phase high performance liquid chromatography; that binds to a single known seven-transmembrane domain TFA, trifluoroacetic acid; TIPS, triisopropylsilane; TOCSY, total correlation spectroscopy; U937, human leukemic monocyte lymphoma receptor C3aR, a rhodopsin-like GPCR that is expressed cell line. ubiquitously on the surface of many cells (particularly pubs.acs.org/jmc Published on Web 06/08/2010 r 2010 American Chemical Society Article Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 4939

Table 1. Affinity and Potency of Known Truncated Anaphylatoxin Agonists on dU937 Cells C3a receptor affinitye C5a receptor affinitye apparent agonist activityf a b c compd peptide n -log IC50 ( SE IC50 (nM) n pIC50 ( SE IC50 (nM) n -log EC50 ( SE EC50 (μM) 1 C3a 13 9.9 ( 0.07 0.12 3 j 10 7.3 ( 0.06 0.052 2 C5a 1 j 4 8.9 ( 0.24 1.4 3 8.2 ( 0.09 0.007 3 SB290157 7 6.9 ( 0.16 140 3 j 3 φd 4 3D53 3 j 3 7.8 ( 0.17 15 3 φ 5 WWGKKYRASKLGLAR 3 8.7 ( 0.20 2.0 3 j 5 5.9 ( 0.16 1.3 6 C3a63-77 3 6.6 ( 0.23 250 3 6.3 ( 0.33 560 4 5.8 ( 0.11 1.8 7 C5a60-74 3 5.8 ( 0.21 1600 3 5.8 ( 0.17 1700 4 4.6 ( 0.11 24 8 YSFKPMPL(Me-a)R 3 7.9 ( 0.13 13 3 5.7 ( 0.14 2000 5 6.9 ( 0.06 0.13 9 YSFK(Me-D)MPLaR 3 6.1 ( 0.19 740 3 4.5 ( 0.45 29000 5 5.9 ( 0.05 1.4 10 YSHKPMPLaR 3 j 3 j 3 5.2 ( 0.03 5.7 11 YSFKPMPLaR 3 6.9 ( 0.10 140 3 5.3 ( 0.21 4700 2 6.7 ( 0.07 0.22 12 HLGLAR 3 5.2 ( 0.21 5900 3 j 3 5.4 ( 0.31 3.8 13 HLALAR 3 5.2 ( 0.29 6300 3 j 3 φ 14 YPLPR 3 j 3 j 2 φ 15 WPLPR 3 4.5 ( 0.50 27000 3 j 2 φ a Concentrations causing 50% of maximum binding of 125I-C3a to intact human monocytes. b Concentration causing 50% inhibition of maximum 125 c binding of I-C5a to intact human monocytes. Concentration causing 50% maximum calcium mobilization from Bt2-cAMP differentiated U937 cells. d e f No agonist activity at 1 mM, antagonist with pIC50 = 5.9 ( 0.1 (IC50 = 1.3 μM). j: no binding detected at 1 mM. φ: no intracellular calcium release detected at 10 μM. immune cells such as macrophages, neutrophils, T cells) and Comparisons between reported activities of these syn- 13a signals through GRi and GR16 proteins. Both crystal and thetic C3a agonists are difficult to make because of the NMR-derived solution13b structures have been reported variety of assay systems used to evaluate them, ranging for human C3a, as well as for complement C3 incor- from tissue organ preparations for measurement of smooth porating C3a.13c The crystal structure of C3a (actually muscle contraction16a,b,17b,c,h,k to mast cell histamine-release18 C3adesArg) was of low resolution (3.2 A˚ ) but showed the assays and ATP-release1-7,19 assays from guinea pig platelets C-terminus of C3a in a turn conformation, while the solution and affinity measurements on isolated membranes or whole structure showed disorder in the last seven residues. For the cells. The activities of synthetic ligands are further compli- C3 structure, the electron density was too diffuse to identify cated by the use of different species, especially tissues and cells the C-terminal seven residues of the sequence corresponding from guinea pigs that have since been found to have two to C3a. A similar turn structure was, however, also observed functional C3a receptors20 compared with only one receptor for the C-terminus of C5a.14 Like C5a, it is the C-terminal in humans. Moreover, GPCRs are notorious for having octapeptide region of C3a that is implicated as the effector widely variable species-dependent responses to ligands due domain required for triggering receptor activation,15 the to different amino acid polymorphisms in the receptor, C5a remaining 69 amino acids at the N-terminus of C3a being a being a case in point.21 This may in part explain why many of high affinity binding domain. Numerous peptides based on these peptides reevaluated by us against intact human cells are the C-terminus of C3a have been reported to compete with either not potent or not selective in their binding to C3aR over 125I-C3a for the surfaces of cells. Removal of the C-terminal C5aR or also bind and activate other receptors. Our experi- arginine of C3a abolishes the biological activity of C3a, so all ences have also highlighted some potency and affinity incon- synthetic C3a ligands to date have a C-terminal arginine. sistencies with literature data, warranting reevaluation in Despite three decades of research, endeavors to produce more relevant human cells. It is our opinion that there have highly potent peptidic C3a ligands of less than eight residues, to date been no truly selective and potent agonists or antago- with agonist activity against human cells at submicromolar nists reported for probing the pharmacology of human C3a concentrations, have been unsuccessful. In fact, there have in vivo. been relatively few reports of new synthetic C3a ligands since There have been very few reports of small non-peptide the early 1990s. ligands for C3aR. The most studied is a hydrophobic capped 2 Among the most potent reported C3aR binding peptide arginine N -[(2,2-diphenylethoxy)acetyl]-L-arginine (SB290157, ligands is a synthetic peptide comprising the C-terminal 21 3) reported to be a C3a receptor antagonist22a but also shown residues of C3a, which was equipotent with C3a in lung to be an agonist in transfected rat basophilic leukemia cells, strip and ileal contraction assays but 44% as effective as Chinese hamster ovary cells, and human U937 cells,22b,c as C3a in suppressing Ig secretion by peripheral blood well as acting on other unknown receptors.22d Antagonists for lymphocytes.16a Incorporation of helix-inducing Ala and C3aR have been reported based on diiminoisoindoline23 and Aib residues at judicious positions resulted in a 250% biphenylimidazole24 core structures, while more potent partial enhancement in activity of the peptide, correlating with agonists have been based on arginine derivatives with a 2,5- 125 25a increased helicity measured by circular dichroism furyl component (pIC50 of 7.1 vs 20 pM I-C3a) and 16b 25b spectra. Truncation and modification to 15 residues with aminopiperidines without arginine (pIC50 = 7.5). One path incorporation of two tryptophans at the N-terminus re- to rational development of nonpeptidic ligands for C3aR is to sulted in a “superagonist” WWGKKYRASKLGLAR that first identify potent and selective peptides, which proffer was 15-fold more potent than C3a in a guinea pig platelet information about side chain fitting to the receptor. This aggregation assay.16c Numerous other attempts have been paper identifies some important properties for high affinity made to shorten peptide sequences while retaining either ligand binding to human C3aR that can be crucial for future competitive binding with 125I-C3a or C3a-like function.17 design of druggable compounds. 4940 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 Scully et al.

Figure 1. Receptor affinity and agonist potency for selected compounds. (A) Affinities for C3a receptor measured by displacement of 125I-C3a from human PBMCs: 1 (b), 3 (9), 5 ([), 8 (1), and 9 (2). (B) Affinities for C5a receptor measured by displacement of 125I-C5a on human 2 PBMCs: 2 (b), 4 (9), 5 (0), 8 (1), 9 (2), and 10 (O). (C) Activity of 1 (9), 2 (b), 5 ([), and 8 (1) in intracellular Ca þ mobilization assay on human dU937 cells. (D) Intracellular calcium release by 5 in C3aR-desensitized human dU937 cells.

Results potency measured by the capacity to induce intracellular Activity of Truncated Anaphylatoxins. N-Terminal trunca- calcium release from dU937 cells (Table 1) (C3a63-77 6, tion of C3a has previously been used to determine what EC50 = 1.8 μM; C3a 1, EC50 =52 nM; 35-fold), a reduction constitutes the active portion of C3a, but past efforts to that is 100 times even more pronounced for C5a (C5a60-74 7, access short, potent peptides to mimic C3a function have met EC50 = 24 μM; C5a 2, EC50 = 7 nM; 3500-fold). These 15 with limited success. It has been proposed that, as for C5a- residue peptides based on the C-terminal native sequences of C5aR, C3a-C3aR interactions can be described by a two-site C3a and C5a were, however, not selective for their receptors model, with the N-terminal helix bundle of C3a providing (Table 1), C5a60-74 7 having some affinity for C3aR (IC50 = affinity while the C-terminus carries the receptor-activating 1.6 μM) and C3a63-77 6 having even greater affinity for motif.15 We chose to begin addressing binding affinity and C5aR (560 nM) as well as 2000-fold lower affinity than C3a 2 agonist induced Ca þ release from the promonocytic cell line for C3aR. As a result of the nonselectivity of C5a60-74 7, we U937 differentiated with Bt -cAMP (dU937). When both decided to also investigate the binding of reported C5a 2 26 binding and functional assays were used, progressive trunca- decapeptide agonists (8-11), which we find have much tion of C3a dramatically reduced both affinity and activity of higher affinity for hC3aR than hC5aR (Table 1). The even 17b truncated peptides (Table 1). shorter hexapeptide analogues (12, 13) of the C3a C-ter- C3a (1) and C5a (2) are selective for their respective receptors minus had marginal affinity for human C3a receptor at on human PBMCs, with no crosstalk between them, as well as micromolar concentrations (Table 1) and were the minimum 2 potent inducers of intracellular Ca þ release in dU937 cells sequences able to induce any calcium mobilization in human (Table 1). Similarly, the C3a antagonist 322a and the C5a anta- dU937 cells. They were not very active even at 100 μM, the gonist 44a,d-f also bind specifically to C3aR and C5aR, respec- highest concentration tested, while pentapeptides 14 and tively, on PBMCs, with no cross-reactivity, and neither com- 1517j had even weaker or no affinity for C3aR and did not 2 2 pound induces Ca þ release in dU937 cells at 1 mM (Table 1). induce Ca þ release. For 2 decades the 15-residue peptide (5), reported as a supera- These results highlight both the difficulty in designing gonist,16c has been used as a benchmark for measuring com- potent and selective small molecules for these receptors, parative agonist activity against C3aR. Although it has high which have high sequence homology in their transmembrane affinity for C3aR on PBMCs (Table 1), it had only comparable regions, and the need to examine affinity for both C3aR and 2 weak Ca þ-inducing activity to that of C3a63-77 (6) in dU937 C5aR when looking for truly potent and selective ligands. On cells. Also, although it did not bind to C5a receptors on human the basis of this and our other unpublished work, we do not 2 PBMCs, 5 was not selective for C3aR because it induced Ca þ consider that there currently exists any short peptide (8 release even after C3aR in dU937 cells was first densensitized by residues or less) that is both potent (at submicromolar con- human C3a (Figure 1D). Thus, it would appear that future centrations) and selective (g100-fold) for human C3aR over functional studies with 5 could be misleading because of lack of C5aR or other receptors. selectivity for human C3aR. Structure-Activity Relationships for Agonist Derivatives Truncation of C3a to its 15-residue C-terminal native pep- of FKPLAR. Our goal was to create short peptides that are tide sequence leads to a substantial reduction in functional both selective for human C3aR over human C5a receptors Article Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 4941

Figure 2. Agonist activity for selected peptides in intracellular calcium release in human dU937 cells: (A) 1 ([), 2 (b), 17 (4), 26 (2), 31 (1); (B) 1 (b), 42 (1), 54 ([), 55 (9).

Table 2. Agonist Activity of FKPLAR (17) Analogues on Human Table 3. Activity of FWPLAR (31) and FLPLAR (26) Analogues on dU937 Cells Measured by Calcium Release Assay Human dU937 Cells Measured by Calcium Release Assay apparent agonist activityb apparent agonist activityb a a compd peptide n pEC50 ( SE EC50 (μM) compd peptide n pEC50 ( SE EC50 (μM)

16 FKP(dCha)(Cha)r-NH2 2 5.7 ( 0.25 2.2 42 WWTLAR 3 6.04 ( 0.06 0.91 17 FKPLAR 4 4.5 ( 0.20 33 43 hFLTLAR 2 4.96 ( 0.08 11 18 HKPLAR 2 j 44 Ac-FLTLAR 2 6.28 ( 0.15 0.52 19 ChaKPLAR 3 j 45 YLTLAR 2 5.61 ( 0.15 2.5 20 WKPLAR 3 j 46 RLTLAR 4 j 21 FOPLAR 3 j 47 FOTLAR 1 j 22 FRPLAR 4 j 48 FRTLAR 1 5.17 ( 0.36 6.8 23 FEPLAR 4 j 49 FYTLAR 2 5.32 ( 0.35 4.8 24 FIPLAR 7 6.0 ( 0.08 0.95 50 FNleTLAR 3 6.51 ( 0.30 0.31 25 FNlePLAR 4 5.7 ( 0.14 1.8 51 FWGLAR 2 6.33 ( 0.09 0.47 26 FLPLAR 3 6.4 ( 0.09 0.42 52 FWALAR 3 5.93 ( 0.12 1.2 27 FChaPLAR 3 6.1 ( 0.05 0.78 53 FWSLAR 5 6.31 ( 0.09 0.49 28 FQPLAR 4 5.3 ( 0.03 5.3 54 FWTLAR 6 6.43 ( 0.09 0.37 29 FFPLAR 2 5.3 ( 0.09 5.4 55 FLTLAR 5 6.50 ( 0.12 0.32 30 FhFPLAR 3 6.0 ( 0.06 1.1 56 FLGLAR 3 6.29 ( 0.33 0.52 31 FWPLAR 8 6.2 ( 0.14 0.58 57 FLALAR 3 5.96 ( 0.23 1.1 32 FNalPLAR 3 6.1 ( 0.06 0.72 58 FLSLAR 4 6.45 ( 0.19 0.36 33 FKTLAR 3 5.4 ( 0.19 3.8 59 FLTlAR 2 j 34 FKFLAR 2 j 60 FLTNleAR 1 5.58 ( 0.26 2.7 35 FKPNleAR 4 4.4 ( 0.20 38 61 FLTChaAR 4 j 36 FKPChaAR 2 j 62 FLT(dCha)AR 2 j 37 FKPFAR 3 j 63 FLTSAR 3 j 38 FKPWAR 2 j 64 FLTLLR 4 j 39 FKPLChaR 3 j 65 FLTLPR 1 4.54 ( 0.19 29 40 FKPLFR 3 j 66 FLTLAr 2 4.24 ( 0.12 57 41 FKPLWR 3 4.6 ( 0.34 25 67 FLTLAR-NH2 1 4.44 ( 0.19 37 a Concentration causing 50% maximum calcium mobilization from a Concentration causing 50% maximum calcium mobilization from b b Bt2-cAMP differentiated human U937 cells. j: no calcium release Bt2-cAMP differentiated human U937 cells. j: no calcium release detected at 1 mM. detected at 1 mM. and with submicromolar potency as agonists in the intracel- but none of the resulting peptides (18-20) were functional 2 lular Ca þ release assay. Among the most potent C5aR short agonists at up to 100 μM, as there was no detectable calcium peptide agonists is the hexapeptide FKP(dCha)(Cha)r-NH2 release (Table 2). (16).27 Given the high sequence homology between C3a and Lys5 in 17 was substituted by nonpolar (Ile, Leu, Cha, C5a, we decided to start with this hexapeptide to derive Nle), aromatic (Trp, Phe, homoPhe, naphthylalanine), acidic structure-activity relationships for C3aR binding peptides (Glu), neutral and basic (Gln, Orn and Arg) amino acids using a functional agonist assay in dU937 cells. Since the (Table 2). Changing Lys to an alternative basic residue Orn Leu-Ala-Arg motif is present at the C-terminus of C3a and (21) or Arg (22) or to an acidic residue Glu (23) abolished has been shown to be vital for activity at C3aR,17b,f these activity. However, changing to a hydrophobic aliphatic residues were incorporated as replacements for (dCha)- (24-27) or aromatic (29-32) residue or to a Gln residue 2 (Cha)r to produce hexapeptide F6K5P4L3A2R1 (17), (28) generated compounds that stimulated Ca þ release with EC50 =33 μM in the intracellular calcium mobilization within the IC50 range 0.5-6.0 μM. The two most potent assay for dU937 cells (Figure 2). Table 2 shows effects of agonists were 26 (Leu) and 31 (Trp) with EC50 of 0.4 and substitution in FKPLAR on agonist activity, as measured 0.6 μM, respectively (Figure 2), and they were subjected∼ to a 2 by intracellular Ca þ release from dU937 cells. When stan- further round of sequence optimization as shown in Table 3. dard numbering28 was used, Phe6 in 17 was successively Table 3 reports some derivatives of FWPLAR (31). replaced with His, Trp, and the unnatural amino acid Cha Replacing Pro4 with residues Gly, Ala, Ser, and Thr 4942 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 Scully et al.

Figure 3. C3aR selectivity for peptides 24 (A), 31 (B), 54 (C), and 55 (D). Intracellular calcium release was measured in human dU937 cells desensitized to C3a agonists by C3a. Compounds 24, 54, and 55 were selective, while 31 induced calcium release even after the cells were desensitized by C3a, suggesting some activation through a different receptor to C3aR.

(51-54) enhanced agonist potency except for Ala. The Table 4. Affinity of Selected Hexapeptides for C3aR by Competition with Radioligand Binding, Assayed in Human PBMCs threonine-containing 54 (EC50 = 0.37 μM) was the most a b c potent agonist, but changes from Thr4 to Ser or Gly did not compd peptide n -log IC50 ( SE IC50 (nM) IC50 (nM) alter agonist potency much. A second family of compounds 1 hC3a 13 9.75 ( 0.12 0.18 j was based on FLTLAR (55) which was slightly more potent 24 FIPLAR 3 6.64 ( 0.13 230 j than 54. Changing Phe6 to homophenylalanine or tyrosine 31 FWPLAR 4 6.56 ( 0.30 274 j or capping the C-terminus with acetyl (43-45) reduced agonist 42 WWTLAR 3 6.61 ( 0.12 244 j potency, while an arginine at this position (46) abolished 49 FYTLAR 3 6.78 ( 0.17 166 j activity. Replacing Leu5 with ornithine, arginine, or tyrosine 54 FWTLAR 3 7.09 ( 0.33 82 j (47-49) significantly reduced activity, but a change to nor- 55 FLTLAR 3 7.38 ( 0.26 42 j leucine (50) had little affect. Changing Thr4 to glycine, 56 FLGLAR 3 6.81 ( 0.12 157 j 61 FLTChaAR 2 6.62 ( 1.59 238 alanine, or serine (56-58) also had little effect. Replacing j 63 FLTLAr 3 5.10 ( 0.50 8200 j any of Leu3, Ala2, or Arg1 (59-67) substantially reduced or a Number of experiments. b Competitive binding with 80 pM 125I- abolished agonist function of these hexapeptides. c 125 C3a. Competitive binding with 80 pM I-C5a. j: no binding at 1 mM. Further investigations were carried out on the tolerance of each position to different residues in order to confirm the tions, but these reduced agonist potency, suggesting that tests carried out in the earlier part of the study. As Table 3 L3A2R1 at the C-terminus may be optimal among the shows, all changes to FLTLAR resulted in reduced activity, compounds examined here for agonist activity. including addition of an acetyl N-terminal cap (44) and Affinity of C3a Ligands for Human C3aR. The affinity for changing of the C-terminal acid to an amide (67), suggesting C3aR and C5aR of some of the more potent agonists (and that both the free amino group and the carboxylic acid are one antagonist 61) is reported in Table 4 and Figure 4. We important for potency. Figure 2 shows agonist responses for used competitive radioligand binding experiments using potent compounds from Table 2 and 3. whole PBMCs isolated from human plasma to determine Figure 3 shows receptor desensitization experiments to IC50 for each ligand (1, 24, 31, 42, 49, 53, 54, 55, 61 and 63) by examine possible selectivity for two of the most active com- displacement of 125I labeled human C3a or C5a measured pounds in this series, 24 and 31. The cells were first exposed with a scintillation proximity assay. The ligands were tested to C3a, which triggers phosphorylation of the receptor by in triplicate on three different occasions to obtain a full dose tyrosine kinases and desensitization to further addition of C3a response curve for each compound. There was a pseudo- or C3aR-selective agonists. Compound 24 does not stimulate linear correlation between C3a-competitive ligand binding 2 calcium release following C3a receptor desensitization, and we affinity and Ca þ-inducing agonist potency (Figure 7), and conclude that it is selective for C3aR. However, 31 does induce none of the compounds showed any affinity at submillimolar 2 some Ca þ release following desensitization with C3a, indicat- concentrations for human C5a receptors. 2 ing that at least some Ca þ release is not mediated through Solution Structure of Agonist 55. Agonist peptide 55 was activation of C3aR; thus, 31 is not selective. Compounds examined by 1D and 2D proton NMR spectroscopy in an 35-41 and 59-67 represent changes to the L3 and A2 posi- attempt to identify any propensity for adopting well-defined Article Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 4943

Figure 4. Competitive ligand binding affinity for human PBMCs. (A) Affinity of selected peptides for C3aR in competitive radioligand binding vs 125I-C3a (80 pM): 1 (b), 3 (9), 4 ([), 54 (1), 55 (2). (B) Affinity of selected peptides for C5aR as determined in a radioligand binding assay vs 125I-C5a (80 pM): 1 (b), 2 (0), 3 (9), 4 ([), 54 (1), 55 (2).

Table 5. 1H NMR Chemical Shift (δ, ppm) Assignments for 55 in The peptide turn motif in Figure 5 is similar to what we 4f DMSO-d6 at 25 °C have reported for the C5aR specific antagonist FKP- residue NH HR Hβ others (dCha)Wr in DMSO and appears to be encouraged by a F (6) na 4.04 2.88, 3.09 Hδ 7.24; Hε 7.30 hydrogen bond formed between the hydroxyl side chain of L (5) 8.65 4.50 1.48 Hγ 1.65; Hδ 0.88 Thr4 and the backbone amide proton of Leu3. Variable T (4) 8.05 3.97 3.97 Hγ1 4.92; Hγ2 1.04 temperature NMR data reveal a low temperature coefficient L (3) 7.82 4.30 1.48 Hδ 0.82 (3.2 ppb/K) for this proton, indicating its possible involve- A (2) 7.96 4.26 1.19 ment in a hydrogen bond. A serine at this position, which R (1) 8.06 4.15 1.73, 1.62 Hγ 1.57; Hδ 3.09; Hε 7.46 can also form the same hydrogen bond, resulted in similar agonist potency. Proline- and glycine-containing peptides structure in solution. There was no well ordered structure in (31 and 56) have similar agonist activity and likely also favor water, as expected for short peptides, because water com- a turn motif but without the need for this hydrogen bond. petes strongly for hydrogen bonds and only a random The amide proton of Leu3 in peptide F6LALAR1 57 did not assembly of rapidly interconverting structures is normally exhibit a low temperature coefficient, presumably because it 4f observed. However, in the aprotic solvent DMSO-d6, stable is missing its hydroxyl oxygen bonding partner concomitant secondary structures can sometimes be observed even for with its relatively low activity (1 μM) in the calcium release short peptides.4e,f Indeed, at 298 K distinct doublet reso- assay. nances were observed for each of the amide peaks of 55 C3aR Antagonist 61. Figure 7 shows an almost linear (Table 5). The phi and psi dihedral angles of the residues were relationship between C3aR affinity (PBMCs) and activity obtained from the Karplus equation using coupling con- (dU937 cells) for C3a hexapeptides from Table 4, consistent stants displayed for the amides. High (>8Hz) or low (<6Hz) with the expected affinity-activity correlation despite using coupling constants were indicative of secondary structure in two different cell types. It was clear that the LAR tripeptide solution (Leu5, 8.2 Hz; Thr4, 8.6 Hz; Leu3, 8.3 Hz), whereas motif comprising the C-terminus of C3a agonists was not a disordered structure has uniform amide coupling constants amenable to much change without loss of substantial agonist activity in a calcium release assay. However, this was not an between 6 and 8 Hz. absolute requirement for tight binding to the C3a receptor. Secondary structure in peptides is often stabilized by main For example, FLTChaAR (61) was unique among the pep- chain to main chain or main chain to side chain hydrogen tides examined that showed no agonist activity at 100 μM in bonds. The low temperature coefficient for the L3 amide the calcium release assay, but it did bind strongly to the proton (3.2 ppb/K) (Supporting Information) suggests that receptor (IC = 240 nM, Table 4). it may be involved in hydrogen bonding. When T4 was 50 This indicated that 61 was likely an antagonist, with no replaced by alanine, the L3 amide proton no longer had a agonist activity below 100 μM. This was subsequently de- low temperature coefficient, indicating that the T4 side chain 2 monstrated through noncompetitive antagonism of Ca þ hydroxyl is the hydrogen bond acceptor. ROESY experi- release in human PBMCs induced by either C3a (Figure ments for 55 showed 19 distinctive short and medium range 8A) or agonist 54 (Figure 8B). It was not a competitive ligand ROEs, indicating a well-defined structure for 55. Two weak with either of these two agonists, since depression of the ROEs d and d are indicative of a β turn R,N(T4, A2) R,N(L5, A2) maximal agonist response by varying concentrations of 61 structure between residues L5 and A2 (Figure 6). did not change the apparent EC values of C3a or agonist Solution structures were determined using 19 ROE distance 50 54. Thus, either compound 61 binds to a different site on the constraints and 3 dihedral angle constraints, resulting in a C3a receptor or it is an insurmountable antagonist, much family of 15 lowest energy structures which are displayed in like certain hexapeptide antagonists of human C5aR.7d-f We Figure 5 (see Supporting Information for the Ramachandran favor the latter explanation in view of the similarity of the plot, dihedral angles, CR(i,i 3) distances and H-bond dis- peptide sequences. tances). The ROEs (Figureþ 6), distances separating CR protons of Leu5 and Ala2 (<7 A˚ ) are indicative of a β-turn, Discussion the dihedral angles of the two middle residues (T4 and L3) closely mimicking a β-turn. The rmsd of the peptide backbone Human C3a or its receptor C3aR is implicated in the within the family of 15 structures is 1.7 A˚ , which is a reasonable pathology of multiple inflammatory and immune conditions, convergence for a short unconstrained peptide in solution. for which either an agonist or an antagonist might have 4944 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 Scully et al.

Figure 5. NMR-derived solution structure for compound 55 in in DMSO-d6 (298 K). (A) 15 of 20 lowest energy minimized NMR structures showing the backbone of 55. Residues are labeled with three-letter amino acid codes. (B) Representative NMR structure of 55 showing the H-bond formed between the side chain hydroxyl group of Thr4 and the backbone amide proton of Leu3. A distance of 2.8 A˚ separates the heavy atoms involved in the bond. bloodstream into tissue.1 Furthermore, C3a-desArg, in which the C-terminal arginine is cleaved from C3a by N-carboxypeptidase, is known as acylation stimulating protein (ASP) and is implicated in triglyceride synthesis and adipocyte function in metabolic syndrome.29 Clearly, small molecules that can reproduce the function of C3a and its metabolites can at the very least be useful biological probes of C3a function in vivo, where it is often too expensive and impractical to experiment with the native protein or specific antibodies to it. Small molecule agonists may also have important uses by reproducing beneficial immunological properties of C3a. The known GPCR activating functions of C3a are localized largely within its C-terminal octapeptide region, and there have been numerous attempts to mimic the functions of C3a by creating short peptides based on its C-terminus. However, most reported small molecule ligands based on C3a have been tested in a diverse array of assays, often in guinea pigs that have more than one C3a receptor, so we needed to reevaluate Figure 6. NMR summary diagram for 55. Shown are sequential some of the leading agonists against human C3aR first, before and medium-range ROEs, the bar thickness being proportional to devising new ligands. We found (Table 1, Figure 1) that some strong (upper distance constraint 2.7 A˚ ), medium (3.5 A˚ ), weak (5.0 ˚ ˚ 3 reported C5aR ligands actually bound much more tightly to A), and very weak (6.0 A) ROE intensities: JNHCH g 8 Hz (V); amide NH chemical shifts 4δ e 4 ppb/K (b). The medium range ROEs, C3aR, while other reported peptide ligands for C3aR were not 125 especially dRN(i,i 2), dRN(i,i 3) and dNN(i,i 1), are evidence for a β turn selective and often bound competitively with I-C5a to cells þ þ þ 2 conformation. or induced Ca þ release from cells in which C3aR was first desensitized by C3a. Another compound reported to be a superagonist of C3aR (e.g., 5) was found to bind tightly to human macrophage C3aR but had only weak agonist activity 2 in the Ca þ release assay and was also not receptor-selective. Consequently, we set out to obtain potent agonists that unambiguously bound selectively and tightly to human C3aR. From two series of new hexapeptides, we discovered FLPLAR 26 and FIPLAR 24 (EC50 of 0.4-0.95 μM, Table 2) and FWTLAR 54 and FLTLAR 55 (EC50 0.3 μM, Table 3) as the most potent and selective agonists≈ for hC3aR. There was a pseudolinear correlation between agonist 2 potency (measured as release of Ca þ into human dU937 cells) and C3aR binding affinity (measured by competition with 125I-labeled human C3a on PBMCs). An NMR-derived solu- Figure 7. Correlation between binding affinity (IC50) and agonist activity (EC50) for peptides in Table 4. 61 had no agonist activity at tion structure for agonist 55 showed significant “turn” or 100 μM; thus, it is given pEC50 g 4. “bend” propensity in the aprotic solvent DMSO, a feature of the C-terminus of C3a itself, hinting at a potential requirement beneficial effects. However, the ubiquitous expression of C3a for this structural feature for binding within the transmem- and C3aR on many cell types in addition to immune cells may brane region of human C3aR. This result is consistent with the point to additional functions of C3a that await to be discov- findings that over 100 peptide-activated G protein coupled ered. C3a itself induces chemotaxis, secretion of proinflam- receptors have shown a tendency to recognize protein and matory cytokines that define immunostimulatory properties, peptide ligands with turn structure30a,b and that cyclic pep- contraction of smooth muscle, activation of mast cells and tides that mimic turn conformations tend to be potent ligands eosinophils, and an increase in vascular permeability leading for GPCRs and other protein targets.30a-c Future incorpora- to increased migration of monocytes and neutrophils from the tion of turn-inducing peptidomimetic constraints into such Article Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 4945

Figure 8. Noncompetitive antagonism for 61 on human dU937 cells. (A) Against C3a and (B) against agonist 54, showing diminished fluorescence of activation by each agonist due to competition from 61 at 0.5 μM(b), 1 μM(9), 2 μM(2), 4 μM(1), and 8 μM([) as measured by intracellular calcium release assay.

peptides, their cyclic or nonpeptidic analogues, could prove Laufelfingen, Switzerland), TFA, DIPEA, piperidine and DMF fruitful for obtaining even more potent and selective ligands (peptide synthesis grade; Auspep, Parkville, Victoria, Australia), for human C3aR, en route to discovery of candidates suitable HBTU (IRIS Biotech, Germany) and HATU (Sigma Aldrich) for clinical trials. were obtained from commercial sources. Preparative scale reversed- The reported C3aR antagonist 3, which does not bind to phase HPLC (rpHPLC) separations were performed on Phenom- enex C18 columns (100 A˚ , 250 mm 20 mm). Analytical rpHPLC human C5aR (Table 1), is currently the main antagonist were measured on Phenomenex LunaÂ 5 μm C18 columns [Rt(1)], reported in the literature for probing the pharmacology of using gradient mixtures of water/0.1% TFA (solvent system A) and C3aR in vivo and in vitro. It has been used in animal models water 10%/acetonitrile 90%/TFA 0.1% (solvent system B). Purity for treating asthma, allergy, arthritis, intracerebral hemorrha- was checked under different gradients to confirm single products. ging, ischemia/reperfusion injury, among others.5-7,22 How- The molecular weight was determined by electrospray mass spec- ever, it has also been shown to be an agonist in some assays trometry (LCT MICRO-MASS). All compounds were analyzed and to induce non-C3a related activities in other assays.22 Our for purity and molecular weight by rpHPLC and mass spectro- new antagonist FLTChaAR (61) has been found to have metry, respectively, with all compounds having >95% purity. High comparable binding affinity to 3 for human C3aR on PBMCs resolution mass spectrometry was performed on a Bruker micro- (IC 238 nM (61) vs 140 nM (3)). Like 3, it did not bind to TOF by direct infusion in MeCN at 3μL/min using sodium formate 50 clusters as an internal calibrant. hC5aR, and in our study 61 was a more potent antagonist of 2 Peptide Synthesis. Short peptides were assembled by manual C3a-induced intracellular Ca þ release than 3 (IC50 of 238 nM stepwise solid phase peptide synthesis using HBTU activation vs 1.3 μM). New compound 61 has cyclohexylalanine in place and DIPEA in situ neutralization using Fmoc chemistry. Pep- of leucine at position 3 and is the most potent selective tides 16 and 67 were synthesized on Rink amide MBHA resin antagonist for human C3aR yet reported. Interestingly, this (Novabiochem, 0.64 mmol/g) to obtain amide functionality at compound was an insurmountable antagonist and may pro- the C-terminus. All other peptides were synthesized on Fmoc- vide a valuable clue to future design of even more potent and Arg(Pbf)-Wang resin, 0.54 mmol/g substitution (Novabiochem), selective small molecule C3aR antagonists. to obtain peptides with the free C-terminal carboxylic acid group. Some conclusions from this study that can be drawn about The peptides were fully deprotected and cleaved from resin by the requirements of human C3aR for potent and selective treatment with 9.5 mL of TFA/0.25 mL of TIPS/0.25 mL of water at room temperature for 2 h. After evaporation preparative hexapeptide ligands are the following: (i) the C-terminal Leu3- HPLC was used for peptide purification (gradient 0-100% B over Ala2-Arg1 motif is important for agonist activity, with changes 30 min). to this region leading to significant reductions in agonist NMR Spectra Acquisition. 1H NMR spectra were recorded potency; (ii) threonine at the fourth position gave the most on a Bruker Avance 600 spectrometer on samples containing potent compounds, although other amino acids with small 1-4 mM peptide in 90% water 10% D2O or DMSO-d6. side chains (serine, alanine, proline, glycine) were not that Proton assignments were made usingþ TOCSY (80 ms mixing much less potent; (iii) the most potent agonists had trypto- time), NOESY, and ROESY (350 ms mixing time) spectra 31a phan, norleucine, or leucine at the fifth position; (iv) only according to the sequential assignment method. Water suppression in 2D experiments was performed using a 3-9-19 ligands with phenylalanine at the sixth or N-terminal position 1 were potent agonists; (v) replacing leucine at the third position Watergate pulse sequence. Variable-temperature 1D H and TOCSY spectra were typically collected at 5 K increments from of short peptide agonist 55 with a bulkier group leads to 298 to 303 K. For identification of slowly exchanging amides, a compound 61, which is a functional antagonist of both C3a series of 1D 1H and TOCSY spectra were run immediately after and the short peptide agonist 54. These results for ligands with dissolving the peptide (5 mg) in 90% D2O/H2O (550 μL). All both high affinity and high selectivity for human C3aR finally spectra were analyzed in TopSpin 1.3 (Bruker, Germany). provide some long awaited clues that have been needed to Structure Calculation. Cross-peaks in ROESY spectra were drive the development of drug leads specific for this important integrated and calibrated in TopSpin 1.3, and distance con- receptor in human immune defense and inflammatory and straints from ROE intensities were assigned as strong (2.7 A˚ autoimmune diseases. upper limit), medium (3.5 A˚ upper limit), weak (5 A˚ upper limit), or very weak (6 A˚ upper limit). Corrections for pseudo-atoms were added to distance constraints where needed. Backbone Experimental Methods 3 dihedral angle restraints were inferred from JRH-NH coupling General. Protected amino acids and resins (Auspep, Parkville, constants in 1D spectra at 298 K. φ was restrained to -120 ( 30° 3 Victoria, Australia; Advanced ChemTech, U.S.; Novabiochem, for JRH-NH g 8 Hz. Peptide bond ω angles were all set to trans, 4946 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 Scully et al.

and structures were calculated without explicit hydrogen bond as the maximum change in fluorescence through emission from restraints. Initial structures were generated using XPLOR-NIH the Fluo-3 bound calcium complex. 2.19.31d Starting structures with randomized φ and ψ angles and Statistical Analysis. For both receptor binding and agonist extended side chains were generated using an ab initio simulated assays, nonlinear regression was performed using Prism 5 annealing protocol.31b The calculations were performed using (GraphPad Software, San Diego, CA). the standard force field parameter set (PARALLHDG.PRO) and topology file (TOPALLHDG.PRO). Refinement of structures was achieved using the conjugate gradient Powell algo- Acknowledgment. We thank the Australian Research rithm with 1000 cycles of energy minimization and a refined Council (ARC) for a Federation Fellowship to D.P.F. and force field based on the program CHARMm.31c Structures were both ARC and the National Health and Medical Research visualized with MacPymol (The PyMOL Molecular Graphics Council of Australia for research support. System, Schrodinger,€ LLC). Cell Isolation and Culture. Human peripheral blood mono- Supporting Information Available: NMR characterization of nuclear cells (PBMC) were isolated from buffy coat (obtained agonist 55 including variable temperature NMR, a Ramachan- from Australian Red Cross Blood Service, Kelvin Grove) using dran plot, dihedral angles, and ROE restraints; HPLC and Ficoll-Paque density centrifugation (GE Healthcare Bio- HRMS data for all compounds. This material is available free Science, Uppsala, Sweden) and cultured in complete media, con- of charge via the Internet at http://pubs.acs.org. sisting of IMDM with 10% FBS, 10 U/mL penicillin, 10 U/mL streptomycin, and 2 mM L-glutamine (Invitrogen at 37 °C, with References 5% CO2). Promonocytic U937 cells (cultured in RPMI media with 10% FBS, L-Glu, PenStrep, and NEAA) were pretreated (1) (a) Haas, P. J.; van Strijp, J. Anaphylatoxins: their role in bacterial 72 h prior to assay with membrane-permeable cAMP analogue infection and inflammation. Immunol. Res. 2007, 37, 161–175. Bt -cAMP (0.5 mM), which induces cell differentiation to a (b) Gerard, N. P.; Gerard, C. Complement in allergy and asthma. Curr. 2 Opin. Immunol. 2002, 14, 705–708. monocyte/macrophage- or granulocyte-like phenotype. Cellu- (2) (a) Hugli, T. E. Biochemistry and biology of anaphylatoxins. lar differentiation induces increased C3aR and C5aR expression Complement 1986, 3 (3), 111–127. (b) Hugli, T. E. Structure and 32 on the cell membrane. function of the anaphylatoxins. Springer Semin. Immunopathol. 1984, Receptor Binding. Receptor binding was performed using 7, 193–219. (c) Hugli, T. E. The structural basis for anaphylatoxin and 125I-C3a, 80 pM (2200 Ci/mmol, Perkin-Elmer, Torrance, chemotactic functions of C3a, C4a, and C5a. Crit. Rev. Immunol. CA), PBMC cells (15 106 cells/mL) and in the absence or 1981, 1, 321–366. (d) Hugli, T. E.; Muller-Eberhard, H. J. Anaphyla- Â toxins: C3a and C5a. Adv. Immunol. 1978, 26,1–53. presence of various concentrations of unlabeled C3a or C3a (3) (a) Zipfel, P. F.; Skerka, C. Complement regulators and inhibitory hexapeptide agonist for 60 min at room temperature with proteins. Nat. Rev. Immunol. 2009, 9, 729–740. (b) Masters, S. L.; shaking in 50 mM Tris, 3 mM MgCl2, 0.1 mM CaCl2, 0.5% Simon, A.; Aksentijevich, I. Horror autoinflammaticus: the molecular (w/v) bovine serum albumin, pH 7.4. Unbound radioactivity pathophysiology of autoinflammatory disease. Annu. Rev. Immunol. was removed by filtration through glass microfiber filters GF/B 2009, 27, 621–668. (Whatman Iner. Ltd., England) which had been soaked in 0.6% (4) (a) Monk, P. N.; Scola, A.-M.; Madala, P. K.; Fairlie, D. P. Function, structure and therapeutic potential of complement C5a polyethylenimine to reduce nonspecific binding. The filter was receptors. Br. J. Pharmacol. 2007, 152, 429–448. (b) Allegretti, M.; washed 3 times with cold buffer (50 mM Tris-HCl), pH 7.4. Moriconi, A.; Beccari, A. R.; Di Bitondo, R.; Bizzarri, C.; Bertini, R.; Bound 125I-C3a was then assessed by scintillation counting on a Colotta, F. Targeting C5a: recent advances in drug discovery. Curr. β-counter. Specific 125I-C3a binding is defined as a difference Med. Chem. 2005, 12, 217–236. (c) Kohl, J. Drug evaluation: the C5a between total binding and nonspecific binding as determined in receptor antagonist PMX-53. Curr. Opin. Mol. Ther. 2006, 8, 529– 538. (d) March, D. R.; Proctor, L. M.; Stoermer, M. J.; Sbaglia, R.; the presence of 1 μM unlabeled C3a. The IC50 value is the Abbenante, G.; Reid, R. C.; Wadi, K.; Paczkowski, N.; Tyndall, J. D. A.; concentration of antagonist to inhibit the binding of labeled Taylor, S. M.; Fairlie, D. P. potent cyclic antagonists of the complement ligand by 50%. Nonlinear regression analysis (GraphPad Prism C5a receptor on human polymorphonuclear leukocytes. relationships 5) was performed on concentration-response curves to deter- between structures and activity. Mol. Pharmacol. 2004, 65, 868–879. (e) Finch, A. M.; Wong, A. K.; Wadi, S. K.; Paczkowski, N. J.; Fairlie, mine IC50 and -log IC50. The -log IC50 for each compound was calculated for separate experiments and expressed as an arith- D. P.; Taylor, S. M. Low molecular weight peptidic and cyclic antagonists of the receptor for the complement factor C5a. J. Med. metic mean standard error (SE). IC50 values were expressed as a Chem. 1999, 42, 1965–1974. (f) Wong, A. K.; Finch, A. M.; Pierens, geometric mean. G. K.; Craik, D. J.; Taylor, S. M.; Fairlie, D. P. Small molecular probes Intracellular Calcium Release Assay. Pretreated U937 cells and for G-protein-coupled C5a receptors: conformationally constrained culture medium were centrifuged (25 rpm, 5 min). The super- antagonists derived from the C terminus of the human plasma protein natant was removed, and the cells were resuspended in 2 mL of C5a. J. Med. Chem. 1998, 41, 3417–3425. wash buffer (1 HBSS, 20 mM HEPES, 2.5 mM probenecid, pH (5) (a) Kirschfink, M. Targeting complement in therapy. Immunol. Â Rev. 2001, 180, 177–189. (b) Drouin, S. M.; Corry, D. B.; Kildsgaard, 7.4). Cells were counted on a hemocytometer. One dye-loading J.; Wetsel, R. A. Cutting edge: the absence of C3 demonstrates a role for buffer (12 mL of wash buffer, 1% FBS, 25 μL of Fluo-3 (final complement in Th2 effector functions in a murine model of pulmonary concentration 4 μM), 25 μL 20% pluronic acid) was added per allergy. J. Immunol. 2001, 167, 4141–4145. (c) Drouin, S. M.; Corry, 5-7 million cells. Cells were suspended in dye loading buffer, D. B.; Hollman, T. J.; Kildsgaard, J.; Wetsel, R. A. Absence of the incubated in a covered culture flask for 1 h at 37 C, then complement anaphylatoxin C3a receptor suppresses Th2 effector func- ° tions in a murine model of pulmonary allergy. J. Immunol. 2002, 169, centrifuged (25 rpm, 5 min). The supernatant was removed, and 5926–5933. (d) Boos, L.; Campbell, I. L.; Ames, R.; Wetsel, R. A.; cells were resuspended in 3 mL of wash buffer and centrifuged Barnum, S. R. Deletion of the complement anaphylatoxin C3a receptor (25 rpm, 5 min). The process was repeated for cell density of 2 attenuates, whereas ectopic expression of C3a in the brain exacerbates, 106 cells/mL (or 100 000 cells per well). Then 50 μL of testÂ experimental autoimmune encephalomyelitis. J. Immunol. 2004, 173, compound or buffer or control/well was plated out on sterile 4708–4714. (e) Boos, L.; Szalai, A. J.; Barnum, S. R. C3a expressed in the central nervous system protects against LPS-induced shock. Neu- black-wall, clear-bottom 96-well plates (Corning Incorporated, rosci. Lett. 2005, 387, 68–71. (f) Mueller-Ortiz, S. L.; Hollmann, T. J.; NY). Plate was then loaded into the FLUOstar instrument Haviland, D. L.; Wetsel, R. A. Ablation of the complement C3a (BMG LabTechnologies, Offenburg, Germany), where fluores- anaphylatoxin receptor causes enhanced killing of Pseudomonas aeru- cence was measured over 1 min with excitation at 485 nm and ginosa in a mouse model of pneumonia. Am. J. Physiol.: Lung Cell. 2 Mol. Physiol. 2006, 291, L157–L165. (g) Rahpeymai, Y.; Hietala, emission at 520 nm of Fluo-3 bound Ca þ complex, at 28 °C. M. A.; Wilhelmsson, U.; Fotheringham, A.; Davies, I.; Nilsson, A. K.; Differentiated U937 cells were administered in situ 10 s into the Zwirner, J.; Wetsel, R. A.; Gerard, C.; Pekny, M.; Pekna, M. Comple- 1 min reading. Agonist responses were expressed as a percentage ment: a novel factor in basal and ischemia-induced neurogenesis. of calcimycin (A23187) or control compound activity, measured EMBO J. 2006, 25, 1364–1374. Article Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 4947

(6) (a) Kildsgaard, J.; Hollmann, T. J.; Matthews, K. W.; Bian, K.; complement component C5a in solution from nuclear magnetic reso- Murad, F.; Wetsel, R. A. Cutting edge: targeted disruption of the nance data. Biochemistry 1989, 28, 172–185. C3a receptor gene demonstrates a novel protective anti-inflamma- (15) (a) Hoeprich, P. D.; Hugli, T. E. Helical conformation at the tory role for C3a in endotoxin-shock. J. Immunol. 2000, 165, 5406– carboxy-terminal portion of C3a is required for full activity. 5409. (b) Kawamoto, S.; Yalcindag,A.; Laouini, D.; Brodeur, S.; Bryce, Biochemistry 1986, 25, 1945. (b) Ember, J. A.; Hugli, T. E. Comple- P.; Lu, B.; Humbles, A. A.; Oettgen, H.; Gerard, C.; Geha, R. S. The ment factors and their receptors. Immunopharmacology 1997, 38 anaphylatoxin C3a downregulates the Th2 response to epicutaneously (1-2), 3–15. introduced antigen. J. Clin. Invest. 2004, 114, 399–407. (c) Wenderfer, (16) (a) Huey, R.; Erickson, B. W.; Bloor, C. M.; Hugli, T. E. Contrac- S. E.; Wang, H.; Ke, B.; Wetsel, R. A.; Braun, M. C. C3a receptor tion of guinea pig lung by synthetic oligopeptides related to human deficiency accelerates the onset of renal injury in the MRL/lpr mouse. C3a. Immunopharmacology 1984, 8, 37–45. (b) Hoeprich, P. D., Jr.; Mol. Immunol. 2009, 46, 1397–1404. Hugli, T. E. Helical conformation at the carboxy-terminal portion of (7) (a) Humbles, A. A.; Lu, B.; Nilsson, C. A.; Lilly, C.; Israel, E.; human C3a is required for full activity. Biochemistry 1986, 25, 1945– Fujiwara, Y.; Gerard, N. P.; Gerard, C. A role for the C3a 1950. (c) Ember, J. A.; Johansen, N. L.; Hugli, T. E. Designing synthetic anaphylatoxin receptor in the effector phase of asthma. Nature superagonists of C3a anaphylatoxin. Biochemistry 1991, 3, 3603– 2000, 406, 998–1001. (b) Mocco, J.; Mack, W. J.; Ducruet, A. F.; 3612. Sosunov, S. A.; Sughrue, M. E.; Hassid, B. G.; Nair, M. N.; Laufer, I.; (17) (a) Hugli, T. E.; Erickson, B. W. Synthetic peptides with the Komotar, R. J.; Claire, M.; Holland, H.; Pinsky, D. J.; Connolly, E. S., Jr. biological activities and specificity of human C3a anaphylatoxin. Complement component C3 mediates inflammatory injury following Proc. Natl. Acad. Sci. U.S.A. 1977, 74, 1826–1830. (b) Caporale, focal cerebral ischemia. Circ. Res. 2006, 99, 209–217. (c) Jacob, A.; L. H.; Tippett, P. S.; Erickson, B. W.; Hugli, T. E. The active site of C3a Bao, L.; Brorson, J.; Quigg, R. J.; Alexander, J. J. C3aR inhibition anaphylatoxin. J. Biol. Chem. 1980, 255, 10758–10763. (c) Unson, reduces neurodegeneration in experimental lupus. Lupus 2009, 19, 73– C. G.; Erickson, B. W.; Hugli, T. E. Active site of C3a anaphylatoxin: 82. (d) Mizutani, N.; Nabe, T.; Yoshino, S. Complement C3a regulates contributions of the lipophilic and orienting residues. Biochemistry late asthmatic response and airway hyperresponsiveness in mice. 1984, 23, 585–589. (d) Gerardy-Schahn, R.; Ambrosius, D.; Casaretto, J. Immunol. 2009, 183, 4039–4046. (e) Mamane, Y.; Chung Chan, M.; Grotzinger, J.; Saunders, D.; Wollmer, A.; Brandenburg, D.; C.; Lavallee, G.; Morin, N.; Xu, L. J.; Huang, J.; Gordon, R.; Thomas, Bitter-Suermann, D. Design and biological activity of a new generation W.; Lamb, J.; Schadt, E. E.; Kennedy, B. P.; Mancini, J. A. The C3a of synthetic C3a analogues by combination of peptidic and non-peptidic anaphylatoxin receptor is a key mediator of insulin resistance and elements. Biochem. J. 1988, 255, 209–216. (e) Ambrosius, D.; Casaretto, functions by modulating adipose tissue macrophage infiltration and M.; Gerardy-Schahn, R.; Saunders, D.; Brandenburg, D.; Zahn, H. activation. Diabetes 2009, 58, 2006–2017. (f) Garrett, M. C.; Otten, Peptide analogues of the anaphylatoxin C3a; syntheses and properties. M. L.; Starke, R. M.; Komotar, R. J.; Magotti, P.; Lambris, J. D.; Biol. Chem. Hoppe-Seyler 1989, 370, 217–227. (f) Kohl, J.; Casaretto, Rynkowski, M. A.; Connolly, E. S. Synergistic neuroprotective effects M.; Gier, M.; Karwath, G.; Gietz, C.; Bautsch, W.; Saunders, D.; of C3a and C5a receptor blockade following intracerebral hemorrhage. Bitter-Suermann, D. Reevaluation of the C3a active site using short Brain Res. 2009, 1298, 171–177. (g) Tang, Z.; Lu, B.; Hatch, E.; Sacks, synthetic C3a analogues. Eur. J. Immunol. 1990, 20, 1463–1468. S. H.; Sheerin, N. S. C3a mediates epithelial-to-mesenchymal transition (g) Kretzschmar, T.; Pohl, M.; Casaretto, M.; Przewosny, M.; Bautsch, in proteinuric nephropathy. J. Am. Soc. Nephrol. 2009, 20, 593–603. W.; Klos, A.; Saunders, D.; Kohl, J. Synthetic peptides as antagonists of (h) Rynkowski, M. A.; Kim, G. H.; Garrett, M. C.; Zacharia, B. E.; the anaphylatoxin C3a. Eur. J. Biochem. 1992, 210, 185–191. (h) Pohl, Otten, M. L.; Sosunov, S. A.; Komotar, R. J.; Hassid, B. G.; Ducruet, M.; Ambrosius, D.; Grotzinger, J.; Kretzschmar, T.; Saunders, D.; A. F.; Lambris, J. D.; Connolly, E. S. C3a receptor antagonist attenuates Wollmer, A.; Brandenburg, D.; Bitter-Suermann, D.; Hocker, H. Cyclic brain injury after intracerebral hemorrhage. J. Cereb. Blood Flow disulfide analogues of the complement component C3a. Synthesis and Metab. 2009, 29, 98–107. (i) Hutamekalin, P; Takeda, K; Tani, M; conformational investigations. Int. J. Pept. Protein Res. 1993, 41, Tsuga, Y; Ogawa, N; Mizutani, N; Yoshino, S. Effect of the C3a- 362–375. (i) Takahashi, M.; Moriguchi, S.; Ikeno, M.; Kono, S.; Ohata, receptor antagonist SB 290157 on anti-OVA polyclonal antibody- K.; Usui, H.; Kurahashi, K.; Sasaki, R.; Yoshikawa, M. Studies on the induced arthritis. J. Pharmacol. Sci. 2010, 112, 56–63. ileum-contracting mechanisms and identification as a complement C3a (8) Malmsten, M.; Schtmdtchen, A. Antimicrobial C3a. Biology, receptor agonist of oryzatensin, a bioactive peptide derived from rice biophysics, and evolution. Curr. Top. Innate Immun. 2007, 598, albumin. Peptides 1996, 17,5–12. (j) Jinsmaa, Y.; Takenaka, Y.; 141–158. Yoshikawa, M. Designing of an orally active complement C3a agonist (9) Nordahl, E. A.; Rydengard, V.; Nyberg, P.; Nitsche, D. P.; peptide with anti-analgesic and anti-amnesic activity. Peptides 2001, Morgelin, M.; Malmsten, M.; Bjorck, L.; Schmidtchen, A. Activa- 22, 25–32. (k) Lu, Z. X.; Fok, K. F.; Erickson, B. W.; Hugli, T. E. tion of the complement system generates antibacterial peptides. Conformational analysis of COOH-terminal segments of human C3a. Proc. Natl. Acad. Sci. U.S.A. 2004, 101, 16879–16884. Evidence of ordered conformation in an active 21-residue peptide. (10) Pasupuleti, M.; Walse, B. r.; Svensson, B.; Malmsten, M.; J. Biol. Chem. 1984, 259, 7367–7370. Schmidtchen, A. Rational design of antimicrobial C3a analogues (18) Mousli, M.; Hugli, T. E.; Landry, Y.; Bronner, C. A mechanism of with enhanced effects against Staphylococci using an integrated action for anaphylatoxin C3a stimulation of mast cells. J. Immunol. structure and function-based approach. Biochemistry 2008, 47, 1992, 148, 2456–2461. 9057–9070. (19) Federwisch, M.;Casaretto, M.; Gerardy-Schahn, R.; Bitter-Suermann, (11) Sonesson, A.; Ringstad, L.; Nordahl, E. A.; Malmsten, M.; Morgelin, D.; Wollmer, A. Enhanced biopotency of synthetic C3a analogues M.; Schmidtchen, A. Antifungal activity of C3a and C3a-derived by membrane binding. A fluorescence anisotropy decay study. peptides against Candida. Biochim. Biophys. Acta, Biomembr. Biophys. Chem. 1992, 44, 151–161. 2007, 1768, 346–353. (20) Fukuoka, Y.; Ember, J. A.; Hugli, T. E. Molecular cloning of two (12) Peterfy, H.; Toth, G.; Pecht, I.; Erdei, A. C3a-derived peptide binds isoforms of the guinea pig C3a anaphylatoxin receptor: alternative to the type I Fc epsilon R and inhibits proximal-coupling signal splicing in the large extracellular loop. J. Immunol. 1998, 161, 2977– processes and cytokine secretion by mast cells. Int. Immunol. 2008, 2984. 20, 1239–1245. (21) Woodruff, T. M.; Strachan, A. J.; Sanderson, S.; Monk, P. N.; (13) (a) Huber, R.; Scholze, H.; Paques, E. P.; Deisenhofer, J. Crystal Wong, A. K.; Fairlie, D. P.; Taylor, S. M. Species dependence for structure analysis and molecular model of human C3a anaphyla- binding of small molecule agonists and antagonists of the C5a toxin. Hoppe-Seyler’s Z. Physiol. Chem. 1980, 361, 1389–1399. receptor on polymorphonuclear leukocytes. Inflammation 2001, 25, (b) Nettesheim, D. G.; Edalji, R. P.; Mollinson, K. W.; Greer, J.; 171–177. Zuiderweg, E. R. P. Secondary structure of complement component (22) (a) Ames, R. S.; Lee, D.; Foley, J. J.; Jurewicz, A. J.; Tornetta, C3a anaphylatoxin in solution as determined by NMR spectroscopy: M. A.; Bautsch, W.; Settmacher, B.; Klos, A.; Erhard, K. F.; differences between crystal and solution conformations. Proc. Natl. Cousins, R. D.; Sulpizio, A. C.; Hieble, J. P.; McCafferty, G.; Acad. Sci. U.S.A. 1988, 85, 5036–5040. (c) Janssen, B. J. C.; Ward, K. W.; Adams, J. L.; Bondinell, W. E.; Underwood, D. C.; Huizinga, E. G.; Raaijmakers, H. C. A.; Roos, A.; Daha, M. R.; Osborn, R. R.; Badger, A. M.; Sarau, H. M. Identification of a Nilsson-Ekdahl, K.; Nilsson, B.; Gros, P. Structures of complement selective nonpeptide antagonist of the anaphylatoxin C3a receptor component C3 provide insights into the function and evolution of that demonstrates antiinflammatory activity in animal models. immunity. Nature 2005, 437, 505–511. J. Immunol. 2001, 166, 6341–6348. (b) Therien, A. G. Agonist activity (14) (a) Zhang, X.; Boyar, W.; Toth, M. J; Wennogle, L.; Gonnella, of the small molecule C3aR ligand SB 290157. J. Immunol. 2005, 174, N. C. Structural definition of the C5a C terminus by two-dimen- 7479–7480. (c) Mathieu, M. C.; Sawyer, N.; Greig, G. M.; Hamel, M.; sional nuclear magnetic resonance spectroscopy. Proteins 1997, 28, Kargman, S.; Ducharme, Y.; Lau, C. K.; Friesen, R. W.; O'Neill, G. P.; 261–267. (b) Zuiderweg, E. R.; Fesik, S. W. Heteronuclear three- Gervais, F. G.; Therien, A. G. The C3a receptor antagonist SB 290157 dimensional NMR spectroscopy of the inflammatory protein C5a. has agonist activity. Immunol. Lett. 2005, 100, 139–145. (d) Proctor, Biochemistry 1989, 28, 2387–2391. (c) Zuiderweg, E. R.; Nettesheim, L. M.; Arumugam, T. V.; Shiels, I.; Reid, R. C.; Fairlie, D. P.; Taylor, D. G.; Mollison, K. W.; Carter, G. W. Tertiary structure of human S. M. Comparative anti-inflammatory activities of antagonists to C3a 4948 Journal of Medicinal Chemistry, 2010, Vol. 53, No. 13 Scully et al.

and C5a receptors in a rat model of intestinal ischaemia/reperfusion Marceau, F. Synthetic C5a receptor agonists. Pharmacology, metabo- injury. Br. J. Pharmacol. 2004, 142, 756–764. lism and in vivo cardiovascular and hematologic effects. Biochem. (23) Grant, E. B.; Guiadeen, D.; Singer, M.; Argentieri, D.; Hlasta, Pharmacol. 1993, 45, 1289–1299. (d) Konteatis, Z. D.; Siciliano, S. J.; D. J.; Wachter, M. Design, synthesis, and biological activity of Van Riper, G.; Molineaux, C. J.; Pandya, S.; Fischer, P.; Rosen, H.; diiminoisoindolines as complement component 3a antagonists. Mumford, R. A.; Springer, M. S. Development of C5a receptor Bioorg. Med. Chem. Lett. 2001, 11, 2817–2820. antagonists. Differential loss of functional responses. J. Immunol. (24) Claffey, M. M.; Goldstein, S.; Jung, S.; Nagel, A.; Shulze, V. 1994, 153, 4200–4205. Biphenylimidazole Compounds as C3a Receptor Antagonists. (28) Nomenclature used here for peptides of length X: the C-terminal WO 2007034282, 2007. residue is termed P1; the second is P2 up to the N-terminus which is (25) (a) Denonne, F.; Binet, S.; Burton, M.; Collart, P.; Dipesa, A.; called PX. Ganguly, T.; Giannaras, A.; Kumar, S.; Lewis, T.; Maounis, F.; (29) (a) Baldo, A.; Sniderman, A. D.; Stluce, S.; Avramoglu, R. K.; Nicolas, J. M.; Mansley, T.; Pasau, P.; Preda, D.; Stebbins, K.; Maslowska, M.; Hoang, B.; Monge, J. C.; Bell, A.; Mulay, S.; Volosov, A.; Zou, D. Discovery of new C3aR ligands. Part 1: Cianflone, K. The adipsin acylation stimulating protein system and Arginine derivatives. Bioorg. Med. Chem. Lett. 2007, 17, 3258– regulation of intracellular triglyceride synthesis. J. Clin. Invest. 3261. (b) Denonne, F.; Binet, S.; Burton, M.; Collart, P.; Defays, S.; 1993, 92, 1543–1547. (b) Maslowska, M.; Cianfione, K.; Rosenbloom, Dipesa, A.; Eckert, M.; Giannaras, A.; Kumar, S.; Levine, B.; Nicolas, M. Acylation stimulating protein (ASP): role in adipose tissue. Prog. J. M.; Pasau, P.; Pegurier, C.; Preda, D.; Van houtvin, N.; Volosov, A.; Obes. Res. 1999, 65–70. Zou, D. Discovery of new C3aR ligands. Part 2: Amino-piperidine (30) (a) Tyndall, J. D.; Pfeiffer, B.; Abbenante, G.; Fairlie, D. P. Over derivatives. Bioorg. Med. Chem. Lett. 2007, 17, 3262–3265. one hundred peptide-activated G protein-coupled receptors recog- (26) (a) Sanderson, S. D.; Kirnarsky, L.; Sherman, S. A.; Ember, J. A.; nize ligands with turn structure. Chem. Rev. 2005, 105, 793–826. Finch, A. M.; Taylor, S. M. Decapeptide agonists of human C5a: (b) Ruiz-Gomez, G.; Tyndall, J. D.; Pfeiffer, B.; Abbenante, G.; Fairlie, the relationship between conformation and spasmogenic and D. P. UPDATE 1: Over one hundred peptide-activated G protein- platelet aggregatory activities. J. Med. Chem. 1994, 37, 3171– coupled receptors recognize ligands with turn structure. Chem. Rev. 3180. (b) Kawatsu, R.; Sanderson, S. D.; Blanco, I.; Kendall, N.; Finch, DOI: 10.1021/cr900344w, in press. (c) Fairlie, D. P.; Abbenate, G.; A. M.; Taylor, S. M.; Colcher, D. Conformationally biased analogs of March, D. R. Macrocyclic peptidomimetics—forcing peptides into human C5a mediate changes in vascular permeability. J. Pharmacol. bioactive conformations. Curr. Med.Chem. 1995, 2, 655–686. Exp. Ther. 1996, 278, 432–440. (c) Vogen, S. M.; Paczkowski, N. J.; (31) (a) Wuthrich,€ K.; Billeter, M.; Braun, W. J. Mol. Biol. 1983, 169, Kirnarsky, L.; Short, A.; Whitmore, J. B.; Sherman, S. A.; Taylor, S. M.; 949–961. (b) Nilges, M.; Gronenborn, A. M.; Br€unger, A. T.; Clore, Sanderson, S. D. Differential activities of decapeptide agonists of G. M. Protein Eng. 1988, 2, 27–38. (c) Brooks, B. R.; Bruccoleri, R. E.; human C5a: the conformational effects of backbone N-methylation. Olafson, B. D.; States, D. J.; Swaminathan, S.; Karplus, M. Int. Immunopharmacol. 2001, 1, 2151–2162. (d) Taylor, S. M.; Sherman, J. Comput. Chem. 1983, 4, 187–217. (d) Schwieters, C. D.; Kuszewski, S. A.; Kirnarsky, L.; Sanderson, S. D. Development of response- J. J.; Clore, G. M. Using Xplor-NIH for NMR molecular struc- selective agonists of human C5a anaphylatoxin: conformational, ture determination. Prog. Nucl. Magn. Reson. Spectrosc. 2006, 48, biological, and therapeutic considerations. Curr. Med. Chem. 2001, 8, 47–62. 675–684. (32) (a) Burg, M.; Martin, U.; Bock, D.; Rheinheimer, C.; Kohl, J.; (27) (a) Kawai, M.; Quincy, D. A.; Lane, B.; Mollison, K. W.; Luly, Bautsch, W.; Klos, A. Differential regulation of the C3a and C5a J. R; Carter, G. W. Identification and synthesis of a receptor receptors (CD88) by IFN-gamma and PMA in U937 cells and binding site of human anaphylatoxin C5a. J. Med. Chem. 1991, related myeloblastic cell lines. J. Immunol. 1996, 157, 5574–5581. 34, 2068–2071. (b) Kawai, M.; Wiedeman, P. E.; Luly, J. R; Or, Y. S. (b) Klos, A.; Bank, S.; Gietz, C.; Bautsch, W.; Kohl, J.; Burg, M.; New Hexa- and Heptapeptides Are Anaphylatoxin Antagonists and Kretzschmar, T. C3a receptor on dibutyryl-cAMP-differentiated U937 Agonists—For Treating Inflammatory and Immunodeficiency Dis- cells and human neutrophils: the human C3a receptor characterized by eases, Cancers and Severe Infections. WO 9212168-A1, 1992 functional responses and 125I-C3a binding. Biochemistry 1992, 31, (c) Drapeau, G.; Brochu, S.; Godin, D.; Levesque, L.; Rioux, F.; 11274–11282.