"Medicine, Faculty of"@en . "Medicine, Department of"@en . "Experimental Medicine, Division of"@en . "DSpace"@en . "UBCV"@en . "Khosrovi-Eghbal, Arash"@en . "2012-08-15T22:33:30Z"@en . "2012"@en . "Master of Science - MSc"@en . "University of British Columbia"@en . "Platelet-monocyte aggregates circulating in the blood are found to play an important role in cardiovascular disease, the most common cause of death in Canada, and are now an established early marker of acute events. Upon stimulation, monocytes transmigrate across the endothelial layer into the intima where they take up oxidized low density lipoproteins (LDL) and differentiate into macrophages, and are then incorporated into atherosclerotic plaques.\tUpon plaque rupture, which accounts for the majority of fatal cardiovascular incidents, platelets are exposed to a variety of agonists such as collagen and mildly oxidized LDL. We used dimethyl labeling quantitative proteomics approach to determine the relative abundance of platelet releasate (Rel) proteins from platelets activated with thrombin, collagen or Lysophosphatidic acid (LPA; the most potent platelet activator found in mildly oxidized LDL). Using the different agonists led to releasates with unique protein compositions. In addition, we analyzed the relative abundance of protein in releasate free of microparticles (Rel-MP) when using the different agonists. Flow cytometry and proteomics studies showed that thrombin, collagen or LPA activated platelets not only produce different number of platelet microparticles (MP), but that these MP have different proteome profiles. We studied the effects of combining agonists by activating platelets with thrombin plus collagen or a subthreshold concentration of collagen plus LPA. Through biologic functional studies we saw that the Rel, Rel-MP or MP from the platelets activated with the different agonists lead to different degree of THP-1 cell migration. \nWe added the contents released from thrombin activated platelets to a human monocytic cell line THP-1, in order to find the proteins which are responsible for THP-1 cell stimulation. Based on the increased expression of proteins such as integrin \u00CE\u00B21, we found that, adding platelet releasate induces a pro-inflammatory state in THP-1 cells and prime them for transmigration. Therefore, we have taken considerable strides towards uncovering the effect of platelet activation on monocyte protein expression. The findings may aid in discovery of drug targets for prevention of inappropriate platelet activation and platelet-monocyte aggregate formation, in order to dampen the effects of these events in contributing to cardiovascular disease."@en . "https://circle.library.ubc.ca/rest/handle/2429/42942?expand=metadata"@en . "EXPLORING THE INTERACTION ENVIRONMENT OF BLOOD CELLS: PROTEOMIC ANALYSIS OF PLATELET RELEASATE AND PLATELET- MONOCYTE INTERACTION by Arash Khosrovi-Eghbal B.Sc., The University of British Columbia, 2008 A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF MASTER OF SCIENCE in The Faculty of Graduate Studies (Experimental Medicine) THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver) August 2012 \u00C2\u00A9 Arash Khosrovi-Eghbal, 2012 ii Abstract Platelet-monocyte aggregates circulating in the blood are found to play an important role in cardiovascular disease, the most common cause of death in Canada, and are now an established early marker of acute events. Upon stimulation, monocytes transmigrate across the endothelial layer into the intima where they take up oxidized low density lipoproteins (LDL) and differentiate into macrophages, and are then incorporated into atherosclerotic plaques. Upon plaque rupture, which accounts for the majority of fatal cardiovascular incidents, platelets are exposed to a variety of agonists such as collagen and mildly oxidized LDL. We used dimethyl labeling quantitative proteomics approach to determine the relative abundance of platelet releasate (Rel) proteins from platelets activated with thrombin, collagen or Lysophosphatidic acid (LPA; the most potent platelet activator found in mildly oxidized LDL). Using the different agonists led to releasates with unique protein compositions. In addition, we analyzed the relative abundance of protein in releasate free of microparticles (Rel-MP) when using the different agonists. Flow cytometry and proteomics studies showed that thrombin, collagen or LPA activated platelets not only produce different number of platelet microparticles (MP), but that these MP have different proteome profiles. We studied the effects of combining agonists by activating platelets with thrombin plus collagen or a subthreshold concentration of collagen plus LPA. Through biologic functional studies we saw that the Rel, Rel-MP or MP from the platelets activated with the different agonists lead to different degree of THP-1 cell migration. We added the contents released from thrombin activated platelets to a human monocytic cell line THP-1, in order to find the proteins which are responsible for THP-1 cell stimulation. Based on the increased expression of proteins such as integrin \u00CE\u00B21, we found that, adding platelet releasate induces a pro-inflammatory state in THP-1 cells and prime them for transmigration. Therefore, we have taken considerable strides towards uncovering the effect of platelet activation on monocyte protein iii expression. The findings may aid in discovery of drug targets for prevention of inappropriate platelet activation and platelet-monocyte aggregate formation, in order to dampen the effects of these events in contributing to cardiovascular disease. iv Preface Ethical approval for blood draw and the use of human platelets was granted by the University of British Columbia Research Ethics Board (certificate number H07-01943) and written informed consent was granted by the donors. v Table of Contents Abstract ......................................................................................................................................................... ii Preface ......................................................................................................................................................... iv Table of Contents .......................................................................................................................................... v List of Tables .............................................................................................................................................. viii List of Figures ............................................................................................................................................... ix List of Abbreviations ..................................................................................................................................... x Acknowledgements .....................................................................................................................................xiii 1 Introduction: ......................................................................................................................................... 1 1.1 Platelets ........................................................................................................................................ 3 1.1.1 Major platelet organelles ...................................................................................................... 4 1.1.2 Platelet activation by collagen .............................................................................................. 7 1.1.3 Platelet activation by thrombin ............................................................................................ 9 1.1.4 Oxidized low density lipoprotein ........................................................................................ 13 1.1.5 Lysophosphatidic acid ......................................................................................................... 14 1.1.6 Platelet secretion ................................................................................................................ 16 1.1.7 Platelet releasate ................................................................................................................ 18 1.1.8 Platelet microparticles ........................................................................................................ 19 1.2 Platelet-monocyte aggregates .................................................................................................... 21 1.3 Proteomics .................................................................................................................................. 25 1.3.1 Mass spectrometry ............................................................................................................. 25 1.3.2 Quatitative proteomics ....................................................................................................... 27 1.4 Proteomics and platelet-monocyte aggregates .......................................................................... 29 1.5 Aims and hypothesis ................................................................................................................... 31 2 Methods .............................................................................................................................................. 33 2.1 THP-1 cell culture ........................................................................................................................ 33 2.2 Ethics statement, blood donations and platelet preparation ..................................................... 33 2.3 Platelet activation by mildly oxidized low density lipoprotein (mox-LDL) .................................. 34 2.4 Preparation of platelet releasates for quantitative proteomics, aggregometry and flow cytometry ................................................................................................................................................ 35 2.5 Dimethyl labeling of platelet releasates ..................................................................................... 36 vi 2.6 Liquid chromatography tandem mass spectrometry ................................................................. 37 2.7 Dimethyl labeled protein quantitation and bioinformatics ........................................................ 39 2.8 THP-1 Cell transmigration assay ................................................................................................. 39 2.9 Membrane enrichment ............................................................................................................... 40 2.9.1 Protocol A ............................................................................................................................ 40 2.9.2 Protocol B ............................................................................................................................ 41 2.10 Western immuno blotting to test protocol A and B ................................................................... 41 2.11 Reference THP-1 cell global and membrane proteome dataset ................................................ 42 2.12 Preparation of platelet releasate for THP-1 cell stimulation ...................................................... 43 2.13 THP-1 cell stimulation using platelet releasate .......................................................................... 43 2.14 SILAC labeled protein quantitation and bioinformatics .............................................................. 46 3 Results ................................................................................................................................................. 47 3.1 Platelet releasate analysis ........................................................................................................... 47 3.2 Quantitative proteomic analysis on platelet releasate when using thrombin, collagen and thrombin plus collagen as agonists ......................................................................................................... 51 3.2.1 Analysis of Rel proteins with high relative abundance ....................................................... 53 3.2.2 Analysis on Rel-MP proteins with high relative abundance ............................................... 59 3.2.3 Proteins potentially abundant in MP .................................................................................. 59 3.2.4 GO analysis on all Rel proteins ............................................................................................ 61 3.3 Quantitative platelet releasate analysis using LPA, collagen and subthreshold collagen plus LPA as agonists ............................................................................................................................................... 69 3.3.1 Analysis on Rel proteins with high relative abundance ...................................................... 72 3.3.2 Analysis on Rel-MP proteins with high relative abundance ............................................... 78 3.3.3 Proteins potentially abundant in MP .................................................................................. 78 3.3.4 GO analysis on all Rel proteins ............................................................................................ 79 3.4 Flow cytometry analysis .............................................................................................................. 88 3.5 THP-1 Migration towards Rel ...................................................................................................... 92 3.6 Membrane protein enrichment .................................................................................................. 94 3.7 Generating reference THP-1 cell dataset .................................................................................... 96 3.8 THP-1 cell stimulation by platelet releasate ............................................................................. 100 3.8.1 Flow cytometry analysis .................................................................................................... 100 3.8.2 Global protein changes ..................................................................................................... 103 vii 3.8.3 Membrane protein changes .............................................................................................. 113 4 Discussion .......................................................................................................................................... 121 4.1 Protein content of platelet releasate ........................................................................................ 121 4.2 Quantitative proteomic analysis of releasate proteins from thrombin, collagen and thrombin plus collagen activated platelets ........................................................................................................... 123 4.2.1 Thrombin versus collagen activated platelets .................................................................. 123 4.2.2 Rel T/C versus Rel-MP T/C................................................................................................. 127 4.2.3 Thrombin versus thrombin plus collagen activated platelets ........................................... 129 4.2.4 Rel T/TC versus Rel-MP T/TC ............................................................................................. 132 4.3 Quantitative proteomic analysis of releasate proteins from LPA, collagen and subthreshold collagen plus LPA activated platelets .................................................................................................... 133 4.3.1 LPA versus collagen activated platelets ............................................................................ 134 4.3.2 Rel L/C versus Rel-MP L/C ................................................................................................. 138 4.3.3 Subthreshold collagen plus LPA versus collagen activated platelets ................................ 139 4.4 Functional biological studies using platelet releasate .............................................................. 140 4.5 Reference membrane and global proteome dataset for THP-1 cells ....................................... 142 4.6 THP-1 cell stimulation by platelet releasate ............................................................................. 143 4.6.1 THP-1 cell global proteome changes................................................................................. 144 4.6.2 THP-1 cell membrane proteome changes......................................................................... 145 5 Conclusions ....................................................................................................................................... 148 References ................................................................................................................................................ 152 Appendices ................................................................................................................................................ 169 Appendix A: Proteins primarily found in microparticles from thrombin activated platelets ............... 169 Appendix B: Proteins primarily found in microparticles from collagen activated platelets ................. 170 Appendix C: Proteins primarily found in microparticles from thrombin plus collagen activated platelets .............................................................................................................................................................. 171 Appendix D: Proteins unique to total releasate (Rel) from thrombin, collagen or thrombin plus collagen activated platelets (from Figure 3.20) .................................................................................... 173 Appendix E: Proteins unique to total releasate free of microparticles (Rel-MP) from thrombin, collagen or thrombin plus collagen activated platelets (from Figure 3.20) ........................................................ 174 Appendix F: Proteins primarily found in microparticles from LPA activated platelets ......................... 175 Appendix G: Proteins unique to total releasate (Rel) from collagen, LPA or subthreshold collagen plus LPA activated platelets (from Figure 3.25) ........................................................................................... 176 viii List of Tables Table 1.1: Granule secretion ......................................................................................................................... 6 Table 3.1: Estimated average amount of protein found in each compartment of platelet releasate upon activation with various agonists ................................................................................................................. 48 Table 3.2: Change in protein abundance when using different agonists ................................................... 55 Table 3.3: Effect of activating platelets with thrombin plus collagen ........................................................ 56 Table 3.4: Releasate proteins and their ratios grouped according to biological process gene ontology (GO) terms................................................................................................................................................... 62 Table 3.5: Releasate proteins and their ratios grouped according to cellular compartment gene ontology (GO) terms................................................................................................................................................... 66 Table 3.6: Change in protein abundance when using different agonists ................................................... 73 Table 3.7: Effect of adding subthreshold collagen to LPA .......................................................................... 74 Table 3.8: Releasate proteins and their ratios grouped according to biological process gene ontology (GO) terms................................................................................................................................................... 80 Table 3.9: Releasate proteins grouped according to cellular compartment gene ontology (GO) terms ... 83 ix List of Figures Figure 1.1: Schematic of events which take place during atherosclerosis and plaque formation ............... 3 Figure 1.2: Platelet activation by collagen .................................................................................................... 8 Figure 1.3: Platelet activation by thrombin ................................................................................................ 12 Figure 1.4: Platelet-monocyte molecular interactions ............................................................................... 22 Figure 1.5: Dimethyl labeling mechanism ................................................................................................... 29 Figure 2.1: Workflow for dimethyl labeling of releasate proteins .............................................................. 38 Figure 2.2: Work flow for SILAC quantitative experiments ........................................................................ 45 Figure 3.1: Releasate proteins from mox-LDL activated platelets .............................................................. 50 Figure 3.2: Overlap of quantified platelet releasate proteins from thrombin, collagen and thrombin plus collagen activated platelets ........................................................................................................................ 52 Figure 3.3: Quantified releasate proteins from thrombin versus collagen samples .................................. 57 Figure 3.4: Quantified releasate proteins from thrombin versus thrombin plus collagen samples ........... 58 Figure 3.5: Quantified Rel-MP proteins from thrombin versus collagen samples ..................................... 60 Figure 3.6: Aggregometry studies on activated platelets ........................................................................... 70 Figure 3.7: Overlap of quantified platelet releasate proteins from LPA, collagen and subthreshold collagen plus LPA activated platelets .......................................................................................................... 71 Figure 3.8: Quantified releasate proteins from LPA versus collagen samples ........................................... 76 Figure 3.9: Releasate proteins from subthreshold collagen plus LPA versus collagen samples ................. 77 Figure 3.10: Flow cytometer studies on activated platelets ....................................................................... 91 Figure 3.11: THP-1 cell migration towards platelet releasate .................................................................... 93 Figure 3.12: THP-1 membrane protein enrichment ................................................................................... 95 Figure 3.13: Repeat of THP-1 membrane protein enrichment using protocol A ........................................ 96 Figure 3.14: Reference global and membrane THP-1 cell datasets ............................................................ 99 Figure 3.15: MAC-1 analysis on THP-1 cells stimulated with platelet releasate....................................... 101 Figure 3.16: P-selectin binding to THP-1 cells ........................................................................................... 102 Figure 3.17: Global protein ratio distribution at 6 and 24 hours of THP-1 cell stimulation .................... 104 Figure 3.18: Venn diagrams for protein overlaps by 6 hours and 24 hours global datasets .................... 105 Figure 3.19: Significant decreases in global proteins at 6 hours of stimulation ....................................... 107 Figure 3.20: Significant increase in global proteins at 6 hours stimulation .............................................. 108 Figure 3.21: Ingenuity Pathway Analysis (IPA) looking at protein changes by 6 hours ............................ 109 Figure 3.22: Significant decreases in global proteins at 24 hours of stimulation ..................................... 111 Figure 3.23: Significant increase in global proteins at 24 hours of stimulation........................................ 112 Figure 3.24: Ingenuity Pathway Analysis (IPA) looking at protein changes by 24 hours .......................... 113 Figure 3.25: Membrane protein ratio distribution at 6 and 24 hours of THP-1 cell stimulation .............. 114 Figure 3.26: Venn diagrams for protein overlaps for membrane datasets .............................................. 115 Figure 3.27: Significant changes in membrane proteome datasets at 6 hours of stimulation ................. 117 Figure 3.28: Significant changes in membrane proteome datasets at 24 hours of stimulation ............... 119 Figure 3.29: Ingenuity Pathway Analysis of proteins found from membrane enriched datasets ............ 120 x List of Abbreviations AC Adenylyl cyclase ACTN4 Alpha actin 4 ADP Adenosine 5\u00E2\u0080\u0099-diphosphate ANXA Annexin ATP Adenosine 5\u00E2\u0080\u0099-triphosphate CALR Calretiuclin cAMP Cyclic adenosine monophosphate CCL3L1 C-C motif chemokine 3-like 1 CCL5 C-C motif ligand 5 cGMP Cyclic guanosine monophosphate CL/C Proteins from subthreshold collagen plus LPA activated platelets to collagen activated platelets ratio CSTF3 Cleavage stimulation factor subunit 3 CTSG Cathepsin G CXCL10 C-X-C motif chemokine ligand 10 DAG Diacylglycerol DCXR L-xylulose reductase EZR Ezrin FcR \u00CE\u00B3 Fc receptor \u00CE\u00B3-chain FDFT1 Squalene synthase FITC Fluorescein isothiocyanate GAPDH Glyceraldehyde 3-phosphate dehydrogenase GDP Guanosine 5\u00E2\u0080\u0099-diphosphate GNAI2 Guanine nucleotide binding protein (G protein) alpha 2 GO Gene ontology GP Glycoprotein GPCR G protein-coupled receptor GTP Guanosine 5\u00E2\u0080\u0099-triphosphate hox-LDL High oxidized low density lipoprotein IgG Immunoglobulin G IL-1\u00CE\u00B2 Interleukin-1\u00CE\u00B2 IL-8 Interleukin-8 IP3 Inositol 1,4,5-triphosphate ITGA5 Integrin alpha 5. ITGB Integrin \u00CE\u00B2 iTRAQ Isobaric tag for relative and absolute quantitation KTN1 Kinectin L/C Protein from LPA activated platelets to collagen activated platelets ratio LC-MS/MS Liquid chromatography tandem mass spectrometry LDL Low density lipoprotein LGALS1 Galectin-1 xi LPA Lysophosphatidic acid LPC Lysophosphatidylcholine MALDI- TOF Matrix-assisted laser desorption/ionisation-time of flight MAPK Mitogen-activated protein kinase MCP-1 Monocyte chemotactic protein-1 MCP- 1/CCL2 Monocyte chemotactic protein 1 MLCK Myosin light chain kinase Mlcp Phosphatidylinositol 4-phosphate 3-kinase mm-LDL Minimally modified low density lipoprotein mox-LDL Mildly oxidized low density lipoprotein MP Platelet microparticles MS Mass spectrometry MS/MS Tandem mass spectrometry MSN Moesin mTOR Rapamycin MYL12A Myosin regulatory light chain 12A NF-\u00CE\u00BA\u00CE\u00B2 Nuclear factor kappa-light-chain-enhancer of activated B cells NSF N-ethlymaleimide sensitive factor OAT Ornithine aminotransferase. oxLDL oxidized low density lipoproteins P2Y Purinoreceptor PAR Protease activated receptor PC phosphatidylcholine PE Phycoerythrin PECAM1 Platelet endothelial cell adhesion molecule PE-CyTM7 PE-cyanine 7 PF4 Platelet factor 4 PI3 Elafin preproprotein PI3K Phosphatidylinositol 3-kinase PIP2 Phosphatidylinositol-4,5-biphosphate PKC Protein kinase C PLC Phospholipase C Plt Platelets PPIF Peptidylprolyl cis\u00E2\u0080\u0094trans isomerse F PPP Platelet poor plasma PPP1CC Serine/threonine-protein phosphatase PP1-gamma PRTN3 Myeloblastin PSGL-1 P-selectin glycoprotein ligand RANGAP1 RAN GTPase-activating protein 1 Rel Total platelet releasate Rel-MP Platelet releasate free of MP (the soluble fraction of releasate) xii RIAM Rap-1-GTP-interacting molecule SAR1A GTP-binding protein SAR1A SDS-PAGE Sodium dodecyl sulfate polyacrylamide gel electrophoresis SERPINB1 Leukocyte elastase inhibitor SFTPD Pulmonary surfactant-associated protein D SILAC Stable isotope labeling by amino acids in culture SLPI Antileukoproteinase SPN Leukosialin STAT Signal transducer and activator of transcription SWAP70 Switch-associated protein 70 T/C Protein from thrombin activated platelets to collagen activated platelets ratio T/TC Protein from thrombin activated platelets to thrombin plus collagen activated platelets ratio TF Tissue factor TGF\u00CE\u00B21 Transforming growth factor \u00CE\u00B21 THP-1 The human acute monocytic leukemia cell line Timp Metalloproteinase inhibitor TNF Tumor necrosis factor TNF-\u00CE\u00B1 Tumour necrosis factor-\u00CE\u00B1 TPR Nucleoprotein TPR TRAP Thrombin receptor activating peptide TXA2 Thromboxane A2 VWF von Willebrand factor xiii Acknowledgements I would like to start by thanking Canadian Institutes of Health Research for funding this research project. Although it\u00E2\u0080\u0099s my name on the front of this thesis, there are many people who have helped me along the way, so I would like to show my gratitude now. I must give enormous thanks to my supervisor Dr. Juergen Kast for his guidance and encouragement which started during my undergraduate years, when I knew very little about scientific research and what it takes to generate meaningful results. Dr. Kast opened many doors for me and showed me the next step and helped me see the big picture when I was overwhelmed by small details. Over the years I have learnt a great deal from Dr. Kast as my instructor, supervisor and mentor. Dr. Geraldine Walsh took me under her wing from day one and got me excited about platelets and scientific research. Although she left our lab during the course of my thesis, she never stopped helping me and guiding me to the finish line. Her advice has gotten me to where I am now and I will forever be grateful. I\u00E2\u0080\u0099d next like to thank Dr. Cordula Klockenbusch for always being there and listening to my crazy experimental plans and her expertise has helped me design and complete many of my experiments. She was always willing to give me a helping hand when I needed it. Although a new addition to our lab, Dr. Jane O\u00E2\u0080\u0099Hara always took time from her own busy schedule to proof read my thesis and helped me be aware of the logistics of writing a thesis. I would like to also thank all the members of Dr. Kast group who I have worked with over the years. I would also like to thank the Biomedical Research Centre (BRC) and all the students and staff for providing a unique environment to work. Thanks to all the people who have attended my BRC talks and gave me input. Their probing questions helped improve my project and experiments. xiv I would like to thank Shujun Lin for helping me use the strong cation exchange and having it ready every time I needed it. A special thanks to Dr. Nikolay Stoynov and Jason Rogalski for running my samples on the mass spectrometers and giving me guidance and enormous knowledge about mass spectrometry and proteomics. I would like to also thank ubcFLOW, especially Andy Johnson, who over the years taught me so much about flow cytometry and what \u00E2\u0080\u009Cconvincing\u00E2\u0080\u009D data looks like. I have great gratitude for all the blood donors without whom my research would not be possible. Thanks to Brana Culibrk in Centre for Blood Research for her phlebotomy services. I\u00E2\u0080\u0099d like to also extend my appreciation to Centre for Blood Research for the informative lectures and allowing me to use their equipment for my experiments, in particular the aggregometry machine. Thank you to my committee members, Dr. Ed Conway and Dr. Kelly McNagny, for their support, advice and input over the years. Their knowledge and expertise have helped me design better experiments and overcome and the various challenges and obstacles I faced during the course of my project. Finally I would like to thank my parents Homeira and Alireza as well as my brother Ashkan. They always gave me new perspective in life and helped me get to where I am today. I would like to also thank my fianc\u00C3\u00A9e Margit Juhasz for putting up with me and always being there for me when I needed her. 1 1 Introduction: According to the World Health Organisation (WHO), cardiovascular disease is the single most common cause of death in the developed world, accounting for almost 20 % of deaths in Canada (http://www.who.int/cardiovascular_diseases/en/index.html). The sudden formation of thrombi and subsequent arterial blockage leads to a lack of downstream tissue oxygen and nutrients that can result in harsh tissue damage and death. The occlusion of myocardial arteries and those that supply the brain can result in myocardial infarction (heart attack) and cerebral infarctions (stroke) (Davi and Patrono 2007). To prevent arterial occlusions we must attain a better understanding of the root causes and mechanisms that lead to eventual circulatory blockage. Though it has been believed that progressing atherosclerosis and increasing stenosis lead to a greater chance of arterial occlusion, recent clinical studies have proven otherwise. Fourteen percent of recorded fatal cases from atherosclerosis disease are triggered by events associated with severely blocked arteries. Alternatively, the rupture of large plaques that had not yet been identified as major restrictors of blood flow were implicated in the majority of fatal cardiovascular incidents. Researchers believe that the lipid content released from these \u00E2\u0080\u009Cnon-occluding\u00E2\u0080\u009D plaques triggers platelet activation, which then attracts leukocytes, especially monocytes, to the site of plaque rupture. Furthermore, the activated platelets, clot (form thrombi), as well as form aggregates with the nearby monocytes. The thrombi and platelet-monocyte aggregates can travel through the blood stream to other sites where they adhere, forming further occlusions (Ross 1999; Libby and Theroux 2005; Davi and Patrono 2007). Platelet-monocyte aggregation has been associated with early stages of atherosclerosis, the thickening of the arterial wall due to the build-up of fatty deposits (Ross 1999). The increased permeability of the endothelium to lipoproteins results in the deposition of cholesterol-containing low density lipoproteins (LDLs) into the tunica intima. Agents originating from endothelial cells later oxidise 2 these deposits (oxLDLs) causing attraction of platelets, T-lymphocytes, and monocytes to the sites of atherosclerotic lesion caused by the oxidation. Monocytes then differentiate into macrophages, ingest the oxLDLs, and form foam cells that become fatty streaks in the arteries (Figure 1.1). The endothelial cells that are part of the lesion also secrete cytokines and chemokines increasing recruitment of platelets and leukocytes, thereby amplifying the immune response and allowing the formation of plaques that sequester the oxLDLs from the bloodstream with a fibrous cap. The repeated rupture and repair of the arteries causes the arterial wall to thicken away from the lumen, however this does not result in arterial occlusion. The rupture of plaques and the secondary thrombosis caused by the plaque rupture are thought to be the real causes of arterial occlusion which can lead to myocardial or cerebral infarctions (Libby and Theroux 2005). With platelet activation comes stronger platelet-monocyte interactions as well as the induction of a pro-inflammatory state in the monocytes. Circulating platelet-monocyte aggregates have been observed in stroke (McCabe, Harrison et al. 2004), acute as well as stable coronary syndromes (Furman, Mark et al. 1998), hypertension (Nomura, Kanazawa et al. 2002), and diabetes (Elalamy, Chakroun et al. 2008) and are now thought to be better early markers of acute myocardial infarction than proven molecular markers. It is also known that diets high in omega-3 fatty acids reduce the levels of activated platelets and therefore platelet-monocyte aggregates (Din, Harding et al. 2008). These findings all suggest that there is a link between plaque rupture, platelet activation, platelet-monocyte aggregation, and the occurrence of cardiovascular disease through the exact correlations and mechanisms have not been studied to a large extent. Platelets and the means by which they are activated will be discussed in the following sections, since this plays an important role in initiating platelet-monocyte aggregate formation. Later the tools and the techniques used in this thesis to study the content of proteins released from activated platelets, as well as, the monocyte proteins which change upon stimulation with platelet releasate are discussed. 3 1.1 Platelets Platelets are one of the essential components of the blood. Platelet are nonnucleated, disk shaped, membrane-encapsulated cell fragments. Mammalian megakaryocyte cells form and release platelets into the bloodstream, where platelets on average circulate for 10 days. In humans, the average platelet Resting platelet Activated platelet Proteins released from Platelets upon Monocyt e Endothelial cells Transmigration Thrombus PMA circulate in blood stream Damaged blood vessel LDL Resident Macrophages Foam Cell oxLDL Figure 1.1: Schematic of events which take place during atherosclerosis and plaque formation (A) Low density lipoproteins (LDL) can become oxidized by agents released from endothelial cells and migrate across the endothelial layer initiating plaque formation. These oxidized-LDL (ox-LDL) molecules can be ingested by macrophages turning them into foam cells, thereby increasing plaque size and making it more prone to subsequent ruptures. (B) Platelets in circulation can become activated as part of normal hemostasis or upon plaque rupture. These platelets can bind to monocytes forming platelet-monocyte aggregates (PMA). The content released from platelets can also bind to monocytes. PMA can shed the platelets and transmigrate into the plaque. 4 count in the blood ranges from 200-400*109 platelets/L (Mason, Carpinelli et al. 2007). Platelets are the smallest cell type circulating the blood with an average of 2.0 to 5.0 \u00C2\u00B5m in diameter, 0.5 \u00C2\u00B5m in thickness, and having a mean cell volume of only 6 to 10 femtoliters (Bessis and Weed 1973). Platelets have three major organelles which release their content upon activation: \u00CE\u00B1-granules, dense bodies (\u00CE\u00B4 granules) and lysosomes. Other organelles are also present in platelets including small numbers of mitochondria for energy metabolism, glycosomes, electron-dense chains and clusters and tubular inclusions. 1.1.1 Major platelet organelles 1.1.1.1 \u00CE\u00B1-granules \u00CE\u00B1-Granules, only found in platelets, are the most numerous organelles in platelets ranging from 40 to 80 granules per platelet with some bigger platelets containing over 100 (King and Reed 2002; Reed 2004). In resting platelets, \u00CE\u00B1-granules remain separate from each other, indicating an organized substructure in the cytoplasm. \u00CE\u00B1-granules contain procoagulant molecules, fibrinolytic regulators, growth factors, chemokines, immunologic modulators, adhesion molecules and other proteins (Table 1). \u00CE\u00B1-granules contain molecules which are only found in platelets and are synthesized only in megakaryocytes such as platelet factor 4, integrin \u00CE\u00B1IIb and \u00CE\u00B2-thromboglobulin. They also contain molecules which are primarily synthesized or endocytosed by megakaryocytes and are less abundant in other cell types such as von Willebrand factor (VWF) and fibrinogen (Niewiarowski 1977; Handagama, Rappolee et al. 1990; Browder, Folkman et al. 2000; Gear and Camerini 2003). Platelet exocytosis upon stimulation leads to the secretion of the molecules contained within \u00CE\u00B1-granules. 1.1.1.2 Dense bodies Dense bodies, also exclusively found in platelets, are smaller than \u00CE\u00B1-granules and are about 10 fold less abundant (WHITE 1969). The membrane enclosing dense bodies has some receptors identical to \u00CE\u00B1-granules and the platelet plasma membrane making it difficult to pinpoint the origins of dense 5 granules. Dense bodies are rich in calcium and calcium containing complexes as shown using analytical electron microscopy (Table 1). They are also rich in adenine nucleotides including adenosine triphosphate (ATP) and adenosine diphosphate (ADP) (Holmsen and Weiss 1979; Israels, Gerrard et al. 1992; Israels, McMillan et al. 1996). Compared to \u00CE\u00B1-granules, dense granules, contain fewer proteins and are slightly acidic. 1.1.1.3 Lysosomes Lysosomes are common to many cell types, and platelets also contain few of them, usually an average of zero to one lysosome per platelet with a maximum number of no more than three lysosomes (Fukami and Salganicoff 1977; Rendu and Brohard-Bohn 2001). Like lysosomes in other cells platelet lysosomes contain acid hydrolases, cathepsins and lysosomal membrane proteins such as CD63 (Table 1). The function of lysosomes and their contents in relation to hemostasis is not known. Lysosomes release their content when platelets are exposed to maximum stimulation, but may not secrete during vascular injury in vivo (Dangelmaier and Holmsen 1980; Israels, McMillan et al. 1996). 6 Table 1.1: Granule secretion List of proteins and molecules stored in different platelet granules. These molecules are secreted upon platelet stimulation by different agonists. \u00CE\u00B1-Granules Adhesion Molecules P-Selectin, PECAM-1, GPIIb/IIIa, VWF, thrombospondin-1, vitronectin, fibronectin Chemokines Platelet basic proteins (PF4 and its variant (CXCL4) and \u00CE\u00B2-thromboglobulin, CCL3, CCL5 (RANTES), CCL7 (MCP-3), CCL17, CXCL1, CXCL5, CXCL8 (IL-8) Coagulation factors Fibrinogen, plasminogen, protein S, kininogens, multimerin, factor V, VII, VIII, XI, XIII Mitogenic factors Platelet-growth factor, vascular endothelial growth factor, transforming growth factor-\u00CE\u00B2, basic fibroblast growth factor, epidermal growth factor, hepatocyte growth factor, insulin-like growth factor 1 Protease inhibitors C1 inhibitor, plasminogen activator inhibitor-1, tissue factor pathway inhibitor, \u00CE\u00B11- antitrypsin Immunologic molecules \u00CE\u00B21H Globulin, factor D, c1 inhibitor, IgG Others Albumin, Gas6, histidine-rich glycoprotein, high molecular weight kiniogen, amyloid beta-protein precursor, actin Dense granules Nucleotides ATP, ADP, GTP, GDP Amines Serotonin, histamine Bivalent cations Ca, Mg, P, pyrophoshate Membrane Proteins CD63 (granulophysin) LAMP 2 Lysosomes Glycosidsases, proteases, cationic proteins 7 1.1.2 Platelet activation by collagen Collagen can be found throughout the body in connective tissues, and is one of the key initiators of platelet activation (Figure 1.2). Platelets get exposed to collagen at the site of vessel injury or plaque rupture. There are two types of receptors on platelets which bind directly to collagen, Glycoprotein VI (GPVI) and integrin \u00CE\u00B12\u00CE\u00B21 (Brass, Stalker et al. 2007). Integrin \u00CE\u00B12\u00CE\u00B21 is important for adhesion to collagen, while GPVI is responsible for platelet activation by collagen. Under static conditions, particularly in in vitro experiments, collagen can capture and activate platelets, but under arterial flow conditions, collagen needs cofactors, in particular VWF. During vascular injury or after platelet secretion, platelets are also exposed to VWF. VWF binds collagen and platelet receptors GPIb\u00CE\u00B1 and integrin \u00CE\u00B1IIb\u00CE\u00B23. GPVI is crucial in initiating platelet activation and blocking GPVI using antibodies impairs platelet response to collagen (Poole, Gibbins et al. 1997; Nieswandt, Brakebusch et al. 2001). Signaling through GPVI relies on GPVI association with Fc receptor \u00CE\u00B3-chain (FcR\u00CE\u00B3). Collagen binding to GPVI causes FcR\u00CE\u00B3 phosphorylation by tyrosine kinases in the Src family. Syk is then activated by forming a complex with GPVI/\u00CE\u00B3-chain. Syk activates PLC\u00CE\u00B32, which leads to formation of inositol 1,4,5-triphosphate (IP3) and diacylglycerol (Poole, Gibbins et al. 1997). IP3 causes an increase in cytosolic Ca 2+ concentration by opening the Ca2+ channels in the platelet dense tubular system. Increase in Ca2+ concentration leads to shape change and aggregation. Diacylglycerol activates protein kinase C (PKC) isoforms which are expressed in platelets. PKC initiates signaling which lead to platelet activation and secretion (Brass, Stalker et al. 2007). Taken together, collagen generates a strong response from platelets both in vivo and in vitro which leads to platelet activation, aggregation and secretion. 8 Figure 1.2: Platelet activation by collagen Collagen binds to GPIb-IX-V through VWF and directly to GPVI and integrin \u00CE\u00B12\u00CE\u00B21. Collagen binding results in the clustering of GPVI receptors that causes phosphorylation of tyrosine residues in the FcR\u00CE\u00B3 chain. Tyrosine kinase, Syk, binds to the \u00CE\u00B3-chain and becomes activated. As a result PLC\u00CE\u00B3 is activated leading to granule secretion and integrin \u00CE\u00B1IIb\u00CE\u00B23 activation. Abbreviations: VWF, von Willebrand factor; DAG, diacylglycerol; GP, glycoprotein; PI3K, phosphatidylinositol 3-kinase; PIP2, phosphatidylinositol-4,5- biphosphate; PKC, protein kinase C; PLC\u00CE\u00B3, phospholipase C\u00CE\u00B3; TXA2, thromboxane A2; MAPK, mitogen- activated protein kinase; RIAM, rap-1-GTP-interacting molecule. 9 1.1.3 Platelet activation by thrombin Thrombin is a multifunctional serine protease and is routinely used as a potent platelet activator in vivo and in vitro (Davey and L\u00C3\u00BCscher 1967; Hanson and Harker 1988; Eidt, Allison et al. 1989; Leung and Gibbs 1997). Thrombin plays a crucial role in normal hemostasis and thrombosis. The liver produces the zymogen prothrombin and the protein enters circulation with a concentration of 100 \u00C2\u00B5g/mL (Blanchard, Furie et al. 1981). The Prothrombinase complex made up of prothrombin, coagulation factor Xa and active cofactor Va produces active thrombin (\u00CE\u00B1-thrombin) on the surface of cellular phospholipid membrane with the aid of calcium ions (Mann, Butenas et al. 2003). Thrombin has specific cleavage sites for large proteins and the main thrombin substrates include coagulation factor VIII, factor V, factor XIII and fibrinogen to form stable fibrin clots (Blomb\u00C3\u00A4ck, Blomb\u00C3\u00A4ck et al. 1967; Pittman and Kaufman 1988). Active thrombin is rapidly cleared mainly by antithormbin III (Damus, Hicks et al. 1973), but there is evidence that clot-bound thrombin may remain active for up to 2 weeks, and possibly cause cellular activation (Bar-Shavit, Eldor et al. 1989). Thrombin signals through protease activated receptor (PAR) family of G protein-coupled receptors (GPCRs). There are four PARs identified to date, with PAR 1 (Vu, Hung et al. 1991), PAR3 (Ishihara, Connolly et al. 1997) and PAR4 (Xu, Andersen et al. 1998) being substrates for thrombin. PAR3 mRNA is minimally expressed in human platelets with no more than 150 to 200 PAR3 receptors per platelet, but is the primary thrombin receptor in mice (Schmidt, Nierman et al. 1998). PAR expression has also been shown to increase in coronary artery disease, with PAR1 expression upregulated in advance human atherosclerotic vessels and in animal models after experimental injury (Nelken, Soifer et al. 1992; Baykal, Schmedtje et al. 1995). PAR1 is by far the most abundant of the PAR receptors on platelets with about 1500 to 2000 PAR1 receptors per platelet. Two thirds of PAR1 receptors are located on the plasma membrane and the 10 rest in the intracellular surface, exposed only after activation (Brass, Vassallo Jr et al. 1992; Norton, Scarborough et al. 1993). Platelet activation by agonists other than thrombin, for example collagen, exposes the internal PAR1 receptors thereby increasing their numbers accessible to thrombin for cleavage on the platelet surface. Thrombin has a high affinity for PAR1 and after cleavage the receptors are internalized or shed into platelet microparticles (Molino, Bainton et al. 1997). Downstream PAR activation involves phophoinostide hydrolysis, protein phosphorylation, increase in free calcium in the cytosol and suppression of cyclic adenosine monophosphate (cAMP) synthesis (Figure 1.3). These signals converge and lead to platelet shape change, aggregation, secretion (Hart, Jiang et al. 1998; Kozasa, Jiang et al. 1998; Garrington and Johnson 1999; Klages, Brandt et al. 1999). PARs are associated with heterotrimeric \u00CE\u00B1\u00CE\u00B2\u00CE\u00B3 G proteins, with the \u00CE\u00B1-subunit bound to GDP in its inactive form. Upon PAR activation through thrombin, GDP is released and converted to GTP and the \u00CE\u00B1-subunit dissociates from the \u00CE\u00B2\u00CE\u00B3-subunit leaving both active. Upon cleavage by thrombin, PARs couple with G protein families found in platelets, namely G\u00CE\u00B112/13, G\u00CE\u00B1q, G\u00CE\u00B1i and G\u00CE\u00B1s (Brass, Manning et al. 1997). These G\u00CE\u00B1 proteins initiate different signaling pathways. G\u00CE\u00B1i enhances platelet activation by lowering cAMP levels via inhibiting adenylyl cyclase (AC) which converts ATP to cAMP. However, whether thrombin causes G\u00CE\u00B1i activation or subsequent release of ADP activates G\u00CE\u00B1i protein is still unclear. Platelets have at least two isoforms of phospholipase C (PLC), PLC\u00CE\u00B2 and PLC\u00CE\u00B3, but PG\u00CE\u00B1q primarily activates PLC\u00CE\u00B2 which is responsible for phosphoinositide hydrolysis which generates diacylglycerol and IP3 (Banno, Yada et al. 1988; Berridge and Taylor 1988). This leads to an increase in intracellular [Ca2+] which is responsible for tyrosine phosphorylation of other downstream effector proteins, and activation of signaling cascades that are linked to MAP kinase pathways and integrin \u00CE\u00B1IIb\u00CE\u00B23 activation, among others (Berridge and Taylor 1988; Brass, Manning et al. 1997). To date, the molecular mechanisms that take place upon thrombin activation which lead to actin polymerization and ultimately platelet shape change remain poorly elucidated. Mouse model 11 studies show that G\u00CE\u00B112/13 could be involved in platelet shape change with a selective impairment of thrombin induced activation (Offermanns, Laugwitz et al. 1994; Offermanns, Mancino et al. 1997; Van Aelst and D\u00E2\u0080\u0099Souza-Schorey 1997) (211-213;9). Again, the molecules linking these membrane proteins to cytoskeleton shape remain largely unknown, but evidence suggests the involvement of guanine nucleotide exchangers as a starting point (Van Aelst and D\u00E2\u0080\u0099Souza-Schorey 1997). The signaling pathways lead to remodeling of actin filaments which transforms the normally quiescent discoid platelet to a contractile sphere. Although, evidence suggests the involvement of Rho GTPases Rac1 and Cdc42 for actin filament uncapping and actin polymerization (Azim, Barkalow et al. 2000), there are probably other pathways for thrombin stimulated actin reorganization and platelet shape change. Although less abundant than PAR1 in platelets, PAR4 mRNA has been readily detected in platelets (Xu, Andersen et al. 1998). Having both PAR1 and PAR4 receptors present on the platelet surface leads to the presence of a dual-receptor signaling system. Studies suggest a primary role for PAR1 and a secondary role for PAR4, and to completely block thrombin response, strategies targeting both PAR1 and PAR4 are needed (Kahn, Zheng et al. 1998). In addition, evidence suggests the involvement of other receptors which respond to thrombin. One of these candidate receptors is GPIb (Mazzucato, De Marco et al. 1998), which is a heterodimer comprised of \u00CE\u00B1 and \u00CE\u00B2 subunits and exists in a complex with GPIX and GPV (De Cristofaro, De Candia et al. 2000). Blocking or deleting this receptor reduces platelet response to thrombin and impairs PAR1 cleavage, especially at low doses (Harmon and Jamieson 1988). In summary, thrombin irreversibly and fully activates platelets both in vivo and in vitro through activating the signaling pathways that start at the membrane proteins and end with platelet shape change, aggregation and secretion. In addition to activating platelets, thrombin interacts with and activates other cells including endothelials and leukocytes. Endothelial cells produce prostaglandin I2 (Weksler, Ley et al. 1978), 12 platelet activating factor (Prescott, Zimmerman et al. 1984), and platelet-derived growth factor (Daniel, Gibbs et al. 1986) in response to thrombin activation, as well as releasing VWF (Hattori, Hamilton et al. 1989) and expression of P-selectin at the membrane (Zimmerman, McINTYRE et al. 1986). When endothelial cells are activated by thrombin they change shape becoming more permeable. In contrast, they also produce nitric oxide, which is a platelet inhibitor and a vasodilator (Tesfamariam, Allen et al. 1993). In addition, thrombin directly binds to monocytes and neutrophils and increases their adhesion and chemotaxis (Naldini, Sower et al. 1998). These effects on blood cells link thrombin to inflammatory esponses and cardiovascular disease. Figure 1.3: Platelet activation by thrombin Thrombin acts through its receptors PAR1 and PAR4. These PARs are associated with heterotrimeric \u00CE\u00B1\u00CE\u00B2\u00CE\u00B3 G proteins. Once activated, GTP binds to G\u00CE\u00B1 subunit, dissociating it from the G\u00CE\u00B2\u00CE\u00B3 subunits. PARs couple to members of Gq (A), G12/13 (B) and there are conflicting reports about PARs binding to Gi (C) G protein families, resulting in activation of various different intracellular signaling pathways. Gi inhibits AC resulting in lowered cAMP concentrations which enhances platelet response, but whether thrombin causes this or the released adenosine diphosphate (ADP) is not clear. The G\u00CE\u00B1q subunit activates PLC\u00CE\u00B2 which results in IP3 and DAG production. IP3 causes release of Ca 2+ from a dense tubular system thus raising intracellular [Ca2+], which leads to granule secretion, integrin \u00CE\u00B1IIb\u00CE\u00B23 activation and platelet shape change. DAG activates PKC which also through its effectors causes granule secretion and integrin \u00CE\u00B1IIb\u00CE\u00B23 activation. The G12/13 subunits are coupled to guanine nucleotide exchange factors (GEFs) which leads to platelet shape change upon activation. Abbreviations: DAG, diacylglycerol; GDP, guanosine 5\u00E2\u0080\u0099-diphosphate; GTP, guanosine 5\u00E2\u0080\u0099-triphosphate; MLCK, myosin light chain kinase; PI3k, phosphatidylinositol 3-kinase; PAR, protease-activated receptor; PIP2, phosphatidylinositol-4,5- bisphosphate; PKC, protein kinase C; PLA2, phospholipase A2; PLC\u00CE\u00B2, phospholipase C\u00CE\u00B2; TXA2, thromboxane2; ATP, adenosine 5\u00E2\u0080\u0099-triphosphate; cAMP, cyclic adenosine monophosphate; MAPK, mitogen-activated protein kinase; RIAM, rap-1-GTP-interacting molecule. 13 1.1.4 Oxidized low density lipoprotein Oxidation of low density lipoprotein generates agents that play an important role in cardiovascular and inflammatory diseases through their pro-thrombotic and pro-inflammatory properties. Depending on the extent of oxidation, different oxidized LDLs form that differ in their composition of bioactive lipids. These oxidized LDLs include, mildly oxidized LDL (mox-LDL), minimally modified LDL (mm-LDL), and high oxidized LDL (hox-LDL). Macrophages do not uptake mox-LDL and mm-LDL, so they accumulate in the plaques of atherosclerotic patients. Unlike native LDL oxidized LDL can initiate platelet activation (Aviram and Brook 1983; Weidtmann, Scheithe et al. 1995; Naseem, Goodall et al. 1997). When plaques rupture platelets come in contact with the oxidized LDL and their products and become activated (Bocan, Schifani et al. 1986). Platelet activation by oxidized LDL could be important in forming thrombus and platelet-monocyte aggregates as well as coronary events such as myocardial and cerebral infarctions. In additions, patients with high risk of cardiovascular disease have elevated levels of circulating oxidized LDL which could also interact with platelets and possibly explain the higher platelet aggregability seen in these patients (Tanaga, Bujo et al. 2002; Kovanen and Pentik\u00C3\u00A4inen 2003; Sanchez- Quesada, Ben\u00C3\u00ADtez et al. 2004). Oxidized LDL is capable of causing platelet shape change, aggregation and secretion (Ardlie, Selley et al. 1989; Meraji, Moore et al. 1992; Weidtmann, Scheithe et al. 1995), but this response seems to be dependent on the degree of LDL oxidation. Experiments show that mox-LDL generates molecules which solicit stronger platelet activation as compared to hox-LDL (Meraji, Moore et al. 1992; Weidtmann, Scheithe et al. 1995). LDL oxidation changes both proteins and lipid properties, but in mox- LDL proteins are not modified (Steinberg 1997). Therefore, protein modification is unlikely to play a role in activation, but altered or newly produced lipid products are responsible for platelet activation by oxidized LDL. Furthermore, platelets interaction with mm-LDL and mox-LDL are of most biological 14 relevance since they accumulate in the plaques (Berliner, Navab et al. 1995), circulate in the blood stream (Sanchez-Quesada, Ben\u00C3\u00ADtez et al. 2004) and enhance platelet activation induced by other platelet stimuli as seen by an increase in platelet fibrinogen binding after TRAP activation (Korporaal, Gorter et al. 2005). Indeed, mm-LDL and mox-LDL produce bioactive lipids that are capable of stimulating endothelial cells, as well as blood cells such as monocytes and platelets. These bioactive lipids include, lysophosphatidylcholine (LPC) (Quinn, Parthasarathy et al. 1988), oxidized phosphatidylcholine molecules (PC) such as 1-palmitoyl-2-(5-oxovaleroyl)-sn-glycero-3-PC (Subbanagounder, Watson et al. 2000; Navab, Ananthramaiah et al. 2004), F2-isoprostanes (Lynch, Morrow et al. 1994) and lysophosphatidic acid (Siess, Zangl et al. 1999). 1.1.5 Lysophosphatidic acid Lysophosphatidic acid (LPA) is the biologically active component of mox-LDL for eliciting platelet activation (Siess, Zangl et al. 1999). Upon oxidation, LPA increases around 8-fold in mox-LDL and mm- LDL as compared to native LDL (Siess, Zangl et al. 1999). In addition, upon activation platelets release LPA, but LPA\u00E2\u0080\u0099s role in further feedback for generating thrombi is not clear (Sano, Baker et al. 2002). There are different types of LPA present in plaques, and they differ in their type of fatty acid and the linkage of fatty acid to the glycerol backbone. Different LPA molecules have different potency for platelet activation, with alkyl-LPA having 20 times more potency as compared to acyl-LPA with the same fatty acids (Simon, Chap et al. 1982; Rother, Brandl et al. 2003). During oxidation the amount of alkyl- LPA increases almost 6-fold whereas acyl-LPA remains almost the same as compared to native LPA (Zhang, Baker et al. 2004). Although the total amount of LPA is less than those of other lipid in mox-LDL, LPA is by far the most potent platelet activator, as shown by lipid fractionation studies. In fact, apart from LPC shows some platelet stimulatory activity, no other lipids found in mox-LDL, such as PCs, show significant platelet stimulatory activity (Siess, Zangl et al. 1999). 15 The signaling mechanisms by which LPA activates platelets and the receptors that LPA binds are unclear. To date there are seven LPA receptors identified, LPA1-7. They can be classified according to their structure with LPA1, LPA2 and LPA3 belonging to the endothelial differentiation gene (EDG-) subfamily of GPCRs (Chun, Goetzl et al. 2002) and second subfamily, LPA4, LPA5, LPA6 and LPA7 belonging to purinoreceptor (P2Y) cluster of GPCRs (Tabata, Baba et al. 2007). Platelets, at least at the mRNA level express all seven LPA receptors, but whether all seven receptors are expressed at the protein level is still unknown (Amisten, Braun et al. 2008; Pamuklar, Lee et al. 2008). LPA1 and LPA3 antagonists inhibit platelet response caused by LPA. However, platelet response to LPA is not consistent with only LPA1 and LPA3 activation, so there must be other receptors that play a role in platelet activation by LPA (Bandoh, Aoki et al. 2000; Rother, Brandl et al. 2003; Fujiwara, Sardar et al. 2005). Furthermore, as mentioned earlier, different LPA molecules have different potency for platelet activation. LPA1-4 receptors do not differentiate between different structures of LPA and respond the same to all LPA molecules, so there must be other receptors in play to explain this structure activity relationship (Smyth, Cheng et al. 2008). The role of identified LPA receptors in platelet activation is currently not well understood, and there are possibilities for yet unknown GPCRs to be involved in LPA mediated platelet activation. LPA activates GTPase Rho, which activates Rho kinase leading to activation of myosin light chain kinase and myosin phosphorylation to achieve actin reorganization and platelet shape change (Retzer and Essler 2000). GTPase Rho activation could occur through LPA receptors coupled to G12/13\u00CE\u00B1 as seen in other platelet agonists. To further complicate matters, platelet response to LPA seems to be donor dependent with platelets from about 20 % of donors not undergoing aggregation due to LPA (Khandoga, Fujiwara et al. 2008; Pamuklar, Lee et al. 2008). On the other hand, platelet shape change induced by LPA seems to be donor independent (Haser\u00C3\u00BCck, Erl et al. 2004; Khandoga, Fujiwara et al. 2008). There is much debate regarding the cause of this non-responsiveness, with some hypothesizing that LPA4 receptor and its 16 coupling to cAMP is responsible (Pamuklar, Lee et al. 2008), while others not observing any change in cAMP concentrations upon LPA activation (Khandoga, Fujiwara et al. 2008). Unfortunately, knockout mice cannot be used to study the receptors and the mechanism through which LPA activates platelets since rodent platelets are not activated in response to LPA. This could either be due to lack of key receptors on rodent platelets or the inhibitory pathways initiated by LPA dominating the stimulatory pathways. Indeed, there is a need for further studies to understand the receptors different LPA molecules interact with, the signaling mechanisms they initiate and the extent to which they cause platelet shape change, aggregation and secretion. 1.1.6 Platelet secretion Platelet secretion occurs in a similar manner to other secretory cells such as neurons. During exocytosis the actin cytoskeleton reorganizes and granules fuse with the plasma membrane, releasing their content to the extracellular matrix. Platelet secretion occurs after strong irreversible platelet activation through different receptors for agonists such as thrombin and collagen (Brass 2003). Activation leads to \u00CE\u00B1- granules, dense granules and lysosomes releasing their content. The process of releasing the content of granules occurs through steps including actin reorganization, movement of granules closer to the plasma membrane, granule-plasma membrane fusion, and finally release of molecules (White 1974; Flaumenhaft, Dilks et al. 2005). The central proteins responsible for platelet secretion are SNARE proteins. SNARE proteins have over 30 members, some of which are found in platelets, including VAMP, syntaxins and SNAP-25 family proteins (S\u00C3\u00B6llner, Whiteheart et al. 1993). These proteins enable membrane fusion (S\u00C3\u00B6llner, Whiteheart et al. 1993). SNARES depend on factors including N-ethlymaleimide sensitive factor (NSF) and Mg2+ to fuse granule and plasma membrane for exocytosis (Weber, Zemelman et al. 1998). Different organelles secretion depends on different SNARE proteins. For example, dense granule secretion depends on signaling 17 through syntaxin-2, SNAP-23 and VAMP3 (Polg\u00C3\u00A1r, Chung et al. 2002; Polg\u00C3\u00A1r, Lane et al. 2003). On the other hand \u00CE\u00B1-granule secretions occurs through syntaxin-4, syntaxin-2, VAMP 3 and VAMP 8. Like many other proteins in platelets, SNARE protein interaction and activity is tightly regulated. Proteins such as Sec1/Munc18 are responsible for regulating SNARE protein activity (Reed, Houng et al. 1999). The exact mechanism of this control however is still under much investigation. In addition to SNARE proteins, Rab proteins, part of small GTPases family, are also believed to play a part in platelet secretion (Zerial and McBride 2001). Membrane trafficking is thought to be governed at least in part by Rab cycle. For instance, thrombin activation results in Rab3b, 6 and 8 phosphorylation, so this may suggest they play a role in thrombin activated secretion (Fitzgerald and Reed 1999). There are a number of Rab proteins in platelets including Rabs 1, 11, 27 and 31, but their role in platelet secretion due to activation is still unknown (King and Reed 2002). Intracellular calcium ion concentrations play a crucial role in platelet secretion, and the increase, due to for example IP3 generation, is alone sufficient to initiate secretion (Authi, Evenden et al. 1986; Watson, Ruggiero et al. 1986). To date, the precise mechanism by which increased [Ca2+] leads to secretion is not well understood not only in platelet but other secretory cells as well. There are also other proteins which are phosphorylated immediately after platelet activation which could mean they are somehow involved in signaling leading to platelet secretion. For instance, blocking myristoylated alanine-rich C kinase substrate phosphorylation prevents dense granule secretion, suggesting they play a role somewhere between membrane receptor signaling and secretion (Elzagallaai, Ros\u00C3\u00A9 et al. 2000). PKC activation seems to play a role in platelet secretion in synergy with increased [Ca2+] (Bensing, Rubens et al. 2001). Activated PKC could phosphorylate proteins which effect secretion through SNARE mechanism. Munc18c becomes phosphorylated by a PKC dependant pathway in thrombin activated platelets (Reed, Houng et al. 1999; Schraw, Lemons et al. 2003). Once Munc18c is 18 phosphorylated it dissociates from syntaxin which increases platelet secretion (Houng, Polg\u00C3\u00A1r et al. 2003). Moreover, PKC mediates the phosphorylation of syntaxin-4 and inhibits its binding to SNAP-23 (Chung, Polg\u00C3\u00A1r et al. 2000). SNAP-23 is also phosphorylated in a timely manner which corresponds to dense granule secretion and precedes \u00CE\u00B1-granule secretion, in thrombin activated platelets through PKC activity (Polg\u00C3\u00A1r, Lane et al. 2003). These studies which investigate the connection between PKC and SNARES attempt to elucidate the mechanism that initiate SNARE dependant platelet secretion as well as signaling pathways for platelet secretion in general. Cytoskeletal reorganization plays a crucial role in granule secretion. The type and concentration of agonists could lead to platelet shape change, through platelet cytoskeleton rearrangement, without subsequent secretion (Otterdal, Pedersen et al. 2001). However, platelet secretion precedes platelet shape change which suggests cytoskeletal rearrangement could play a role in signaling pathways that lead to granule secretion (Sasakawa, OHARA\u00E2\u0080\u0090IMAIZUMI et al. 2002). Although platelet secretion triggered by different agonists seem to depend on cytoskeletal rearrangement and increase [Ca2+], there still remain missing pieces to the signaling map that results in protein secretion after platelet activation. Hence, different agonists alone or in combination depending on the receptors they signal through could potentially result in a different secreted protein profile. 1.1.7 Platelet releasate The content that platelets release into the matrix upon activation, referred here to as platelet releasate (Rel), contains both soluble and insoluble matter. The platelet releasate includes the secreted molecules from the granules (Table 1) and the proteins shed from platelets. Platelets shed proteins such as GPIb\u00CE\u00B1 (Bergmeier, Piffath et al. 2004), GPV (Rabie, Strehl et al. 2005), GPVI (Bergmeier, Rabie et al. 2004) and P-selectin (Berger, Hartwell et al. 1998). The advantage of protein shedding is not clearly understood, but it could help activate nearby cells such as platelets to elicit a stronger platelet response to injury. 19 In addition, activated platelets release membrane bound vesicles, namely platelet microparticles (MP) and exosomes. Exosomes are released from \u00CE\u00B1-granules and range from 40 nm to 100 nm (Heijnen, Schiel et al. 1999). Through ultracentrifugation one can separate the soluble and the insoluble component, including MP. The remaining supernatant free of MP and other insoluble component is referred to here as Rel-MP. 1.1.8 Platelet microparticles There has been a lot of research focused around MP and their biological function in the last few years. Originally referred to as \u00E2\u0080\u0098platelet dust\u00E2\u0080\u0099 (Wolf 1967), MP vary in size from 0.1 \u00C2\u00B5M to as big as almost platelet size at 1.0 \u00C2\u00B5M. Since MP are budded from platelet plasma membrane, activated platelets and MP have very similar membrane phospholipid composition (Bevers, Comfurius et al. 1983), and also have all the membrane receptors used for detecting activated platelets. These receptors include GP receptors, platelet endothelial adhesion molecule (PECAM-1) and integrin \u00CE\u00B1IIb\u00CE\u00B23, and in some MP P- selectin (Heijnen, Schiel et al. 1999). Depending on the agonist used the resultant MP may have different functions and surface proteins, so while MP from thrombin or collagen activated platelets possess integrin \u00CE\u00B1IIb\u00CE\u00B23 which binds fibrinogen, MP from complement C5b-9 activated platelets have integrin \u00CE\u00B1IIb\u00CE\u00B23 that doesn\u00E2\u0080\u0099t bind fibrinogen (Sims, Wiedmer et al. 1989). MP can originate from activated platelets, aging platelets during storage (Bode, Orton et al. 1991) and even from megakaryocytes (Cramer, Norol et al. 1997), but our understanding of the exact mechanism by which MP are formed remains incomplete. A number of agonists including thrombin, collagen, high shear, complement and low temperatures can induce platelet activation and MP formation (Bode and Knupp 1994; Holme, \u00C8\u0096rvim et al. 1997). In activated platelets, increase in intracellular [Ca2+] is a prerequisite to MP formation (Ratajczak, Wysoczynski et al. 2006). Also, direct activation of PKC using phorbol esters causes MP formation, so agonists such as thrombin and collagen 20 cause MP formation partially through PKC activation (Cocucci, Racchetti et al. 2009). Other proteins activated by Ca2+ such as calpains, which degrade structural proteins such as actin, facilitate MP formation, but calpain inhibition doesn\u00E2\u0080\u0099t prevent MP formation (Shcherbina and Remold-O\u00E2\u0080\u0099Donnell 1999). In addition, other proteins such as integrin \u00CE\u00B1IIb\u00CE\u00B23 could play a role in MP formation. Platelet which lack of functional integrin \u00CE\u00B1IIb\u00CE\u00B23, as seen in platelets from patients with Glanzmann thrombasthenia, release fewer platelets than normal controls after exposure to collagen or thrombin (Holme, Solum et al. 1995). Overall, more evidence is needed to understand MP formation mechanism and directly link activated proteins of different kinds to MP formation. MP have been implicated to a number of different biological functions, but their extent of contribution remains unclear. MP have been long implicated in playing a role in supporting coagulation. MP have many binding sites for activated factor V, factor VIIIa, factor IXa and can provide membrane surface to form thrombin (Sims, Wiedmer et al. 1989). Experiments show that platelet poor plasma (PPP) clots after plasma is recalcified, whereas when MP are removed from PPP by ultracentrifugation no clotting occurs after recalcification (Chargaff and West 1946; O'Brien 1955). MP also play a role in adhesion in that MP can bind to subendothelial matrix through integrin \u00CE\u00B1IIb\u00CE\u00B23 receptors and promote platelet and leukocyte adhesion under flow (Merten, Pakala et al. 1999; Forlow, McEver et al. 2000). MP also contain bioactive lipids such as arachidonic acid and platelet activating factor which can activate endothelial cells and platelets (Iwamoto, Kawasaki et al. 1996; Barry, Kazanietz et al. 1999). MP can directly affect vasculature through different pathways, since evidence suggests that in myocardial infarction patients MP interfered with vasodilation through nitric oxide (Boulanger, Scoazec et al. 2001). Overall, MP formation and their numbers in circulation are vary greatly based on the manner of platelet activation and can vary in different individuals. 21 1.2 Platelet-monocyte aggregates Platelet-monocyte aggregate formation depends on both platelet activation and monocyte activation (Li, Hu et al. 2000). Though the number of activated platelets is low under normal physiological conditions, activation can be caused by numerous events including turbulent blood flow, rolling on activated endothelium, signaling by inflammatory cytokines, or signaling by platelet factors in unstable thrombi. When the body undergoes stresses such as systemic inflammation or myocardial infarction, the number of activated platelets in the circulation increases dramatically (Cosemans, Munnix et al. 2006). Platelets express P-selectin (CD62P) on their surface that allows their binding to the leukocyte P- selectin glycoprotein ligand (PSGL-1). Activated platelets have about 10,000 P-selectin proteins on their surface which gives a density of about 350 P-selectins/\u00C2\u00B5M2 (McEver and Martin 1984; Yeo, Sheppard et al. 1994). This density is much higher than for example endothelial cells. Monocytes are especially rich in PSGL-1 expression, which may explain their tight adherence to platelets underflow even without addition receptor-ligand binding between other proteins (Rinder, Bonan et al. 1991; Ahn, Jun et al. 2005). It has been previously shown that platelet-monocyte adhesion can be interrupted with the use of P-selectin antibodies (Fernandes, Conde et al. 2003) thereby confirming a role of P-selectin and PSGL-1 in platelet-monocyte interactions. Other leukocyte surface proteins, such as CD15, are also known to interact with P-selectin, and probably also play a role in platelet adhesion. In addition, activated platelets release chemokines which play a role in monocyte signaling and activation (Figure 4). As P- selectin is shed from the platelets after several hours in vivo, these platelet-leukocyte adhesions are not sustained for a long period (Berger, Hartwell et al. 1998). On top of P-selectin/PSGL-1 binding there are other proteins which interact with each other at the interface of activated platelets and monocytes possibly creating a firmer adhesion between the two (Figure 4). Monocytes use \u00CE\u00B22 integrin Mac-1 (\u00CE\u00B1M\u00CE\u00B22,CD11b/CD18), to bind to platelets. Evidence show 22 Figure 1.4: Platelet-monocyte molecular interactions P-selectin binding to PSGL-1 initiates platelet-monocyte aggregate formations. Further binding occurs between molecules such as CD40L and CD40. This results in integrin activation and further binding between platelets and monocytes. In addition, soluble molecules such as PF4 and RANTES bind to monocytes causing secretion of cytokines such as TNF-\u00CE\u00B1 which reinforces inflammatory response. Abbreviations: GP, glycoprotein; \u00CE\u00B1M\u00CE\u00B23, integrin \u00CE\u00B1M\u00CE\u00B23; PF4, platelet factor 4; TNF-\u00CE\u00B1, tumor necrosis factor-\u00CE\u00B1; IL, interleukin; TF, tissue factor; NF\u00CE\u00BAB, nuclear factor kappa-light-chain-enhancer of activated B cells; MCP-1, monocyte chemotactic protein-1. 23 that fibrinogen can bind to both Mac-1 on monocytes and integrin \u00CE\u00B1IIb\u00CE\u00B23 on platelets at the same time, but the lack of both receptors doesn\u00E2\u0080\u0099t prevent platelet-monocyte aggregate formation (Weber and Springer 1997; Ostrovsky, King et al. 1998). Moreover, GPIb\u00CE\u00B1 on platelets can bind directly to Mac-1 (Simon, Chen et al. 2000). LFA-1 on monocytes also binds ICAM-2 on platelets, but they have limited contribution to stabilizing the platelet-monocyte aggregate (Diacovo 1994). Initially, platelet-monocyte aggregate formation triggers several events in the monocyte that increase the expression and activity of \u00CE\u00B21 and \u00CE\u00B22 integrins (da Costa Martins, van Gils et al. 2006). For instance, the deposition of platelet chemokine C-C motif ligand 5 (CCL5) and C-X-C motif chemokine ligand 10 (CXCL10) onto monocytes (Schober, Manka et al. 2002), CD40/CD40L interactions (Li, Sanders et al. 2008), as well as the Trem-1/Trem1 ligand interactions (Bouchon, Dietrich et al. 2000) are key factors in up-regulating monocyte integrin expression on at the membrane. In addition, the aforementioned factors cause the shedding of monocyte L-selectin (da Costa Martins, van Gils et al. 2006), and the monocyte secretions of various cytokines and chemokines such as interleukin-1\u00CE\u00B2 (IL-1\u00CE\u00B2), interleukin-8 (IL-8), monocyte chemotactic protein 1 (MCP-1/CCL2) and tumour necrosis factor-\u00CE\u00B1 (TNF- \u00CE\u00B1). Furthermore, tissue factor (TF) is secreted by monocytes in a NF-\u00CE\u00BA\u00CE\u00B2 (nuclear factor kappa-light-chain- enhancer of activated B cells) mediated fashion when induced by thrombin-activated platelets (Gawaz, Neumann et al. 1998). The ligation of PSGL-1 triggers monocyte signaling and activation through GTPase Ras and several kinases, as well as, activating PKC isoforms (Atarashi, Hirata et al. 2005). Secondary activation of \u00CE\u00B22 integrins leads to the activation of Akt/PKB and mammalian target of rapamycin (mTOR) both of which modulate monocyte adhesion and migration. It is known that platelet-monocyte aggregates show a greater tendency for transmigration across endothelial monolayers than monocytes do alone (da Costa Martins, van Gils et al. 2006; van Gils, da Costa Martins et al. 2008). Therefore, monocyte stimulation by activated platelets which precedes transmigration, as well as the platelet deposition on the endothelium, would suggest that there is a regulatory effect of platelets on 24 monocytes during aggregate formation that enhances monocytes\u00E2\u0080\u0099 abilities to transmigrate across the endothelium (Huo, Schober et al. 2003). Chemokines CCL5 and PF4, secreted by activated platelets, promote monocyte survival and further differentiation into macrophages (Scheuerer, Ernst et al. 2000). PF4 facilitates the development of foam cells by promoting the esterification and uptake of oxLDL and CCL5 supports fibrous plaque formation (von Hundelshausen, Weber et al. 2001). Thus, vascular injury by trauma and microlesions resulting from plaque formation in atherosclerosis show similar responses. These responses also resemble those leading to platelet-monocyte aggregate formation and subsequent endothelial interaction in ischemic events. Clinical trials have shown that platelet-monocyte aggregates may be targeted without devastating effects on hemostasis (Diener, Cunha et al. 1996). In a study done by the Second European Stroke Prevention Study (ESPS-2), combined therapy of aspirin and extended-release dipyridamole has been demonstrated to decrease the risk of a recurrent stroke as compared to a placebo or a to aspirin alone, without any signs of increased bleeding (Diener, Cunha et al. 1996). These results may be due to both drugs having anti-inflammatory properties and acting on platelets by inhibiting cyclooxygenase-1 production, reducing TXA2 production, inhibiting platelet aggregation, and increasing intra-platelet cAMP and cyclic guanosine monophosphate (cGMP) levels (Weyrich, Denis et al. 2005). However, platelet-monocyte aggregate formation stimulated by LPA is insensitive to inhibition by aspirin (Haser\u00C3\u00BCck, Erl et al. 2004), so there is still a need to identify drug targets to prevent platelet- monocyte aggregate formation. To do so, we first need to understand the mechanisms by which LPA activates platelets and identify the releasate component from LPA activated platelets. 25 1.3 Proteomics Using large-scale, high-throughput techniques we are now able to provide information on the global state of many biological systems (Weston and Hood 2004). Systems biology (Ideker, Galitski et al. 2001) is comprised of \u00E2\u0080\u009C\u00E2\u0080\u009Domics\u00E2\u0080\u009D techniques as well as bioinformatics. \u00E2\u0080\u009COmics\u00E2\u0080\u009D techniques allow us to comprehensively analyse DNA, mRNA, proteins and metabolites. Bioinformatics focuses on reducing, interpreting, and translating the information gathered through \u00E2\u0080\u009Comics\u00E2\u0080\u009D approaches into a manageable format. In the past few years, proteomics has made significant advances in novel techniques as well as technical capabilities (Cox and Mann 2007). New instrumentation is allowing faster, more sensitive sequencing, efficient sample enrichment techniques, and sophisticated data processing tools. For instance, the number of unique human proteins identified in cells has grown from a few hundred to more than 5,000 in less than five years (Graumann, Hubner et al. 2008). 1.3.1 Mass spectrometry Mass spectrometry (MS) is one of the key technologies used in proteomic research. Analyte molecules are converted into ions, by adding or removing charged particles, which are then accelerated, deflected and focused to determine a mass-to-charge ratio (m/z) of each analyte ion. Ions with different m/z ratios are separated and detected so that the mass spectrum delineates a statistical representation of the type and abundance of each analyte species in a complex mixture. Using tandem mass spectrometry (MS/MS), detailed information on specific analytes becomes accessible. Upon analyte detection, the analyte is isolated in situ, collided with gas atoms to gain sufficient energy to break covalent bonds and then a resulting mass spectrum of the fragments generated is obtained. With addition of separation devices such as a chromatograph (for liquid chromatography tandem mass spectrometry (LC-MS/MS), the acquisition of a series of mass spectra over time allows for the analysis of complex mixtures. Enzymatic digestion produces peptides which are easier to separate, 26 ionize, and detect in the conditions normally used for mass spectrometry, thereby increasing sensitivity and allowing coverage of a wider range of proteins. Even with the ability to detect a wider range of proteins, the increased complexity of analyte mixtures has consequences. Since signal intensity is associated to physiochemical properties and protein concentration, abundant proteins are identified by the presence of multiple peptides, but other proteins in low concentration are missed. In addition, identification relies on comparing the mass spectrum obtained for each peptide fragment to an in silico prediction for the entire proteome, then reporting the most probable protein origin of each fragmented peptide. Therefore, a fewer number of sequenced peptides means less confident protein identification in an analyte mixture. This confidence is reflected in the protein\u00E2\u0080\u0099s score given by bioinformatics programs based on the identified peptides from a given spectra. By designing experiments that target certain peptides, we can better identify low abundant peptides. Using methods, such as fractionation or enrichment/depletion, sample complexity may be reduced prior to LC-MS/MS. For example, with ion exchange chromatography, peptides can be separated into multiple fractions of different ionic strength. To reduce the complexity of an analyte mixture on a protein level techniques such as sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE), can be used for protein separation. Even though fractionation provides a reliable method of reducing mixture complexity, enrichment and/or depletion protocols are the method of choice. The use of ultracentrifugation, and other fractionation techniques, to isolate specific cellular structures (including membrane proteins, which are sorely underrepresented in early proteomics) allows the detection of proteins that are usually not found during global analyses. Though mass spectrometry-based proteomics is vastly useful, the processing steps results in only a loose correlation between the peptide signal intensity, as seen on the mass spectrum, and 27 relative protein abundance, thereby eliminating any reliable quantitative analysis of the proteome data. Throughout the years, crude methods of approximation to determine relative protein abundance by mass spectrometry have been developed (Carvalho, Hewel et al. 2008). 1.3.2 Quatitative proteomics Quantitative proteomics was developed with the introduction of isotope dilution, a common technique in metabolite analysis. Today there are many techniques and products for both metobalic based and chemical based quantiative pretomics including stable isotope labeling by amino acids in culture, isobaric tag for relative and absolute quantitation (iTRAQ) and dimethyl labeling among others. 1.3.2.1 Stable isotope labeling by amino acids in culture (SILAC) Recently SILAC, an approach based on metabolic incorporation, was pioneered (Ong, Blagoev et al. 2002). By depleting the cell culture medium of specific amino acids, leucine, lysine and/or arginine, these amino acids would be reintegrated with different isotope-labelled versions. SILAC thus permits more accurate quantitative comparisons between different cell states by allowing cells or lysates to be mixed, maximizing reproducibility at later stages of sample processing. Because SILAC depends on cell cultures, the one limitation is that primary tissues are only available in their natural, light isotopic form. SILAC is analyzed and quantified using MS scans through comparing the peak area for differentially labeled peptides. There are numerous bioinformatics software programs, both through companies and freely available, which interpret and quantify the data available from different mass spectrometers. 1.3.2.2 Dimethyl labeling for quantitative proteomics Stable isotope dimethyl labeling, is a more recent technique which represents a relatively simple, fast and inexpensive alternative for multiplex quantitative proteomics method based on chemical labeling and analysing the MS spectra (Boersema, Raijmakers et al. 2009). Currently up to three different samples can be analyzed in one experiment (Figure 1.5). Protein samples are first digested using a 28 protease of choice, such as trypsin, and the resulting peptides of the different samples are labeled with the different dimethyl labels. The labeled samples are then mixed together and usually fractionated and different fractions are analyzed by LC-MS/MS. The peak area in the MS scan for each differently dimethyl labeled peptide is used to determine relative abundance of each peptide and ultimately the proteins for each sample. The dimethyl labels react with all primary amines (the N terminus and the side chain of lysine residues) and convert them to dimethylamines, except for N-terminal proline in which a monomethylamine is formed. The dimethyl labels differ from each other by 4 Da, a minimum mass difference for accurate interpreting and quantification of the MS data, through using combination of several isotopomers of formaldehyde and cyanoborohydride. Labeling of intact proteins using dimethyl labels is also possible, but the choices of proteases will be limited as trypsin and Lys-C no longer cleave modified lysine residues (Boersema, Raijmakers et al. 2009). Indeed, dimethyl labeling represents a quick and inexpensive way for relative quantitation of protein from different sample. A 29 Figure 1.5: Dimethyl labeling mechanism (A) Formaldehyde reacts with primary amine in solution, including lysine residues, protein N-terminus and peptide N-terminus. The second step of the reaction involves reaction with sodium cyanoborohydride which results in the addition of two methyl groups to the nitrogen atom. (B) By using different combination of isotopic labeled formaldehyde and cynoborohydride, peptides with different mass to charge ratios can be generated. Peptides labeled with different tags can be mixed and analyzed on the mass spectrometer. The ratios of peptide peak areas can be used to calculate the relative abundance of each protein from three different samples. 1.4 Proteomics and platelet-monocyte aggregates Proteomics is an especially attractive approach to use for the study of platelets. Since platelets lack nuclei, all genetic experimentation, including methods such as transfection to study protein interactions/functions, is precluded. Knocking-in exogenous proteins into platelet progenitor cells, megakaryocytes, is not practical as culturing these cells is hard. Knocking-down mRNA transcripts is possible, but the level of de novo synthesis in platelets is rather low. Since platelet mRNA levels are even lower than that of nucleate cells, transcriptome analysis is often difficult. As platelets are small, 2 to 4 B 30 \u00C2\u00B5M, microscopic studies are impeded as top-of-the-line light-microscopes are needed in order to distinguish unique platelet features. As a consequence, most knowledge to date on platelet structure, function and morphology has been obtained using electron microscopy with gold particles to visualize individual proteins (Denis, Tolley et al. 2005; Schwertz, Tolley et al. 2006). The lack of DNA and the low levels of mRNA require that platelet function be regulated on a post-translational level (with very little post-transcriptional support): protein modification, protein- protein interactions, and protein localisation (Denis, Tolley et al. 2005; Schwertz, Tolley et al. 2006). Thus, platelet-centered proteomics have been used repeatedly as method of choice to complete platelet studies. Early studies used 2D gel electrophoresis and matrix-assisted laser desorption/ionisation-time of flight (MALDI-TOF) techniques to investigate the activation, and other aspects of function, of platelets (Maguire, Wynne et al. 2002; Garc\u00C3\u00ADa, Prabhakar et al. 2004; Garc\u00C3\u00ADa 2006; Gnatenko, Perrotta et al. 2006). These studies were mostly comprised of \u00E2\u0080\u009Cinventory proteomics\u00E2\u0080\u009D, cataloguing the various proteins expressed in the system. Membrane proteins identified in these analyses were separated, using strategies such as two-phase partitioning or chemical enrichment of glycoproteins (Lewandrowski, Moebius et al. 2006; Lewandrowski, Zahedi et al. 2007; Lewandrowski, Wortelkamp et al. 2009), and became the focus of subsequent experiments (Garc\u00C3\u00ADa, Prabhakar et al. 2004; Garcia, Zitzmann et al. 2004). Platelet features such as shed microparticles (Garcia, Smalley et al. 2005; Smalley, Root et al. 2007) and proteins found in platelet \u00E2\u0080\u009Creleasate\u00E2\u0080\u009D have also been studied (Maguire 2003; Coppinger, Cagney et al. 2004; Maguire, Foy et al. 2005; Coppinger, Fitzgerald et al. 2007; Maynard, Heijnen et al. 2007; Piersma, Broxterman et al. 2009). However, none of the studies to date have involved using proteomics to investigate functional interactions between platelets and monocyte. Using global and targeted proteomic tools will identify proteins involved in platelet activation and in the releasate, as well as the proteins involved in the formation of platelet-monocyte aggregates and their coordination, all of which will help design treatments to reduce the risks of associated with cardiovascular disease. The 31 results from studying the platelet releasate and monocyte stimulation by platelet releasate could potentially identify new drug targets both to prevent ill-timed platelet activation and prevent platelet- monocyte aggregate formation. 1.5 Aims and hypothesis We hypothesize that, since platelet agonists act through different platelet receptors, the resulting platelet releasate proteome will differ depending on the agonist used, and the releasate alone is sufficient to stimulate monocytes and prime them for transmigration to different degrees. We aimed to establish a model system based on quantitative proteomics to identify potential biomarkers that could play an important role in platelet-monocyte aggregate formation. We started by studying the content of platelet releasate after activation via different agonists. The agonists studied in this thesis were, thrombin, collagen (type 1), mox-LDL and 1-hexadecyl-LPA (alkyl-LPA 16:0, referred to from here on as simply LPA). Thrombin represents a soluble agonist which signals through GPCRs and is a potent platelet activator both in vivo and in vitro. Collagen activates platelets through adhesion receptor-mediated signaling and is exposed to platelets upon vessel injury and plaque rupture. Upon plaque rupture, as related to cardiovascular disease, platelets are exposed to mox-LDL which is shown to activate platelets. LPA is the most potent platelet activator found in mox- LDL, so its effects on platelet activation will be studied in isolation. It is worth mentioning that in vivo, platelets are rarely exposed to just one agonist and usually platelets are exposed to a combination of different agonists. To this end, we studied the effect of activating platelets and comparing their resultant releasate when using a combination of thrombin plus collagen or sub-threshold levels of collagen plus LPA. We decided to use subthreshold levels of collagen, so we prime platelets for LPA activation, instead of having collagen and LPA both having their full effect on platelets. To carry out a quantitative proteomic study of the proteins released as a result of using 32 different agonists we utilized dimethyl labeling. We also separated the insoluble fraction of platelet releasate from the soluble fraction and compared the resulting proteins in the releasate free of MP (Rel- MP) to the total releasate (Rel) when activating platelet with different agonists. As well, we used flow cytometry to study the effects of using different agonists on platelets and the resulting MPs. Once, we learnt the content of platelet releasate and identified potential candidates that could interact with monocytes, we shifted our focus to study the effect of platelet releasate on the target cell, namely the monocytes. We used the human acute monocytic leukemia cell line (THP-1) as model system for monocytes. We used migration assay studies, to look at the extent of THP-1 cell migration induced by different compartments of platelet releasate when activated with the various agonists. Next, we focused on pathways and proteins which could potentially lead to platelet-monocyte aggregate formation. We used quantitative proteomics to compare the protein changes which take place in the THP-1 cells upon adding platelet releasate from thrombin stimulated platelets. We compared the proteome of unstimulated and stimulated THP-1 cells after addition of platelet releasate. We utilized SILAC technology to carry out this comparative quantitative proteomic study. We analyzed both the global protein changes as well as protein changes that take place at the membrane of THP-1 cells. To study the proteins at the membrane responsible for signaling into and out of THP-1 cells, we optimized a membrane enrichment protocol to get as much membrane proteome coverage as possible. These studies attempt to shed light on the content of platelet releasate when using different agonists, and their effect on monocyte stimulation in the context of platelet monocyte aggregates formation. We established a model system to identify candidate biomarkers, in the platelet releasate as well as in THP-1 cells, that can potentially play an important role in monocyte stimulation. These biomarkers lend themselves to farther biological investigations to elucidate their role and importance in the progression of cardiovascular disease. 33 2 Methods 2.1 THP-1 cell culture The human monocytic cell line THP-1 (American Type Culture Collection, Rockville, MD) was maintained in Roswell Park Memorial Institute medium (RPMI) 1640 media (Invitrogen, Frederick, MD) supplemented with 10 % fetal bovine serum (FBS; Invitrogen Corporation, Carlsbad, CA) and 100 units/ml of penicillin/streptomycin. For SILAC labeling THP-1 cells were maintained in lysine, arginine and glutamine depleted RPMI 1640 (Caisson Labs, North Logan, UT) supplemented with 10 % dialyzed FBS (Invitrogen Corporation), 100 units/ml of penicillin/streptomycin, 100 units/ml of L-glutamine and either normal, 1H1-L-lysine and 12C6-L-arginine (light SILAC media) or 2H1 (D)-L-lysine and 13C6-L-arginine (heavy SILAC media) (Cambridge Isotope Labs, Andover, MA). Cells were grown for at least 6 doublings to allow full incorporation of labeled amino acids and maintained between 2 x 105 and 9 x 105 cells/ml. 2.2 Ethics statement, blood donations and platelet preparation Ethical approval for this study was granted by the University of British Columbia Research Ethics Board (certificate number H07-01943) and written informed consent was granted by the donors. Whole blood was drawn from the antecubital vein of healthy human volunteers into acid-citrate-dextrose (ACD) (11.5 mM citrate acid monohydrate; 88.5 mM trisodium citrate dehydrate; 111 mM dextrose; pH 6.0) anticoagulant at a final volume of 15 % ACD. Platelet rich plasma (PRP) was isolated by centrifugation at 150 x g for 15 minutes at room temperature. The PRP volume was supplemented with half the volume ACD and spun at 720 x g for 10 minutes. The platelet poor plasma (PPP) supernatant was removed and discarded. The platelets were washed twice in physiological buffer (10 mM trisodium citrate, 30 mM dextrose, 10 units/ml apyrase) by centrifugation at 720 x g for 10 minutes. The resulting platelet pellet was used for subsequent platelet activation. 34 2.3 Platelet activation by mildly oxidized low density lipoprotein (mox-LDL) Washed platelets from section 2.2 were resuspended in HEPES buffer saline (70 mM NaCl, 0.75 mM Na2HPO4\u00C2\u00B72H2O, 25 mM HEPES in dH2O) with 1.8 mM CaCl2, to 300 x 10 6 platelets/ml. Platelets were then activated with mox-LDL (Avanti Polar Lipids, Inc., Alabaster, AL) at a final concentration of 1 mg/ml for 10 minutes at 37 \u00C2\u00B0C. The platelets were then separated from the releasate by centrifugation at 700 x g at 4\u00C2\u00B0C. The supernatant was centrifuged again and the resulting supernatant represents the Rel. Portion of the Rel was centrifuged at 150,000 x g for 90 minutes at 4 \u00C2\u00B0C in a TLA 100.3 rotor (Beckman, Palo Alto, CA). The resulting supernatant represents the releasate minus the microparticles (Rel-MP). The remaining pellet, the microparticles (MP) was resuspended in 50 \u00C2\u00B5l of HEPES buffer saline. The samples were assayed for protein content, using the Bichinchonic Assay (BCA; Pierce Biotechnology, Thermo Fisher Scientific, Rockford, IL). 70 \u00C2\u00B5g of protein from Rel, Rel-MP and MP were separated by sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on a 10 % gel. Gels were stained with Coomassie Brilliant Blue, imaged and each lane was cut into 11 bands. Standard reduction, alkylation and tryptic in-gel digestions were performed overnight at 37 \u00C2\u00B0C (Shevchenko, Henrik Tomas et al. 2007). Following digestion, peptides were extracted, desalted and analyzed on the mass spectrometer according to section 2.6. The raw files for Rel, Rel-MP, and MP samples, were searched against Swiss-Prot database with locally hosted Mascot server (version 2.3.0). The searches were taxonomically restricted to \u00E2\u0080\u009CHomo sapiens\u00E2\u0080\u009D with carbamidomethyl as fixed modification and oxidized methionine as variable modification. Maximum allowed mass deviation was set to 20 ppm for monoisotopic precursor ions and 0.6 Da for MS/MS peaks, with a maximum of three missed cleavages. The cutoff used on the peptide level ensures that the worst identified peptide has a probability of 0.05 of being false. The peptides were grouped in to proteins for Rel, Rel-MP and MP samples. 35 2.4 Preparation of platelet releasates for quantitative proteomics, aggregometry and flow cytometry Washed platelets from section 2.2 were resuspended in HEPES buffer saline (70 mM NaCl, 0.75 mM Na2HPO4\u00C2\u00B72H2O, 25 mM HEPES in dH2O) with 1.8 mM CaCl2, to 300 x 10 6 platelets/ml. Platelets were then activated with 1 U/ml of thrombin (Sigma, St. Louis, MO), 0.19 mg/ml collagen (soluble calf skin, type I, 101562, Bio/Data, Horsham, PA), a combination of 1U/ml thrombin and 0.19 mg/ml collagen or 20 mM 1-X-hexadecyl-2-hydroxy-sn-glycero-3-phosphate (LPA, Avanti Polar Lipids, Alabaster, AL) for 10 minutes at 37\u00C2\u00B0C. Platelets were also incubated for one minute with subthreshold concentration (3 \u00C2\u00B5g/ml) of collagen before addition of 20 \u00C2\u00B5M LPA. The platelets were then separated from the releasate by centrifugation at 700 x g at 4\u00C2\u00B0C. The supernatant was centrifuged again and the resulting supernatant represents the Rel. Portion of the Rel was centrifuged at 150,000 x g for 90 minutes at 4 \u00C2\u00B0C. The resulting supernatant represents the releasate minus the microparticles (Rel-MP). The remaining pellet, the microparticles (MP) was resuspended in 50 \u00C2\u00B5l of HEPES buffer saline. In addition, platelets (1 ml) were incubated with thrombin (1 U/ml), collagen (0.19 mg/ml), thrombin plus collagen or LPA (20 mM) and stirred in a dual chamber aggregometer (Chrono-Log model 560CA, Chrono-Log Corp, Havertown, PA) and traces were recorded for 5 minutes at 37 \u00C2\u00B0C. Before platelet activation with LPA for migration or proteomic studies, 500 \u00C2\u00B5l of platelets from each donor were tested for response to LPA on the aggregometer. Platelets were considered aggregated by LPA if the trace changed by more than 1 % as compared to the baseline. 100 \u00C2\u00B5l of the platelets activated with the different agonists were subjected to staining with 5 \u00C2\u00B5l of PE conjugated mouse anti-human CD62P (12-0628, eBioscience), 5 \u00C2\u00B5l of PE-cyanine 7 (PE-CyTM7) mouse anti-human CD41a (561424, BD Biosciences) and 20 \u00C2\u00B5l of FITC mouse anti-human PAC-1 (340507, BD Biosciences). The platelets were incubated for 20 minutes at room temperature in the dark. 900 \u00C2\u00B5l of FACS buffer was added to each sample and they were analyzed on LSR II flow cytometer (BD 36 Biosciences) and the platelets and MP were identified based on CD41a positive events. These events were compared to inactivated platelets treated under the same experimental conditions. Unstained platelets were used as control. All the platelet activation occurred within 3 hours blood collection. 2.5 Dimethyl labeling of platelet releasates Platelets from each donor provided Rel and Rel-MP from thrombin, collagen or thrombin plus collagen activated platelets as per section 2.3. In addition, platelets from each donor were activated with collagen, LPA or subthreshold collagen plus LPA to obtain Rel and Rel-MP as per section 2.3. 3 ml of Rel and 3 ml of Rel-MP from each donor was concentrated using 3 K centrifugal filters (EMD Millipore) to 60 \u00C2\u00B5l, and BCA assay was performed on 10 \u00C2\u00B5l from all the samples to determine the protein concentrations (Figure 2.1). All the Rel and Rel-MP samples were denatured using 1 \u00C2\u00B5l of 2 % SDS and reduced for 1 hour at 60 \u00C2\u00B0C with 2 \u00C2\u00B5l of 50 mM Tris-(2-carboxyethyl)phosphine (TCEP) in HEPES buffer. The cysteine residues were alkylated with 1 \u00C2\u00B5l of 200 mM methyl methanethiosulfonate (MMTS) in isopropanol at room temperature for 10 minutes. The samples were trypsin digested overnight at 37 \u00C2\u00B0C. Different isotopic combination of formaldehyde (either 4 % (vol/vol) of CH2O (light), CD2O (intermediate) or 13CD2O (heavy)) was added to each Rel and Rel-MP peptide solution. Immediately after, sodium cyanoborohydride was added to the samples (1M NaBH3CN was added to the light and intermediate samples and 1M NaBD3CN was added to the heavy labeled samples). The samples were incubated for 2 hours at room temperature with vortexing every 15 minutes. 16 \u00C2\u00B5l of 1 % (vol/vol) ammonia solution was added to each sample followed by the addition of three times the volume of the sample in 5 % formic acid. Rel from thrombin (light labeled) collagen (intermediate labeled) and thrombin plus collagen (heavy labeled) activated platelets were mixed. The labeled Rel-MP samples from these activated platelets were also mixed. In addition, Rel from collagen (light labeled) LPA (intermediate 37 labeled) and subthreshold collagen plus LPA (heavy labeled) were mixed together. The labeled Rel-MP samples from these activated platelets were also mixed. Each sample was loaded on to Ettan MDLC (GE Healthcare Biosciences AB, Uppsala, Sweden) with a strong cation exchange column. The column used was a 50mm x 0.32mm BioBasic strong cation exchange column with particle size 5\u00CE\u00BCm, with sulfonic acid based cation exchange ligand (Thermo Fisher Scientific Inc., Waltham, MA) with water:acetonitrile:ammonium chloride as the mobile phase with gradient elution. Each sample was separated into 25 fractions and analyzed by mass spectrometry according to section 2.6 (Figure 2.1). 2.6 Liquid chromatography tandem mass spectrometry Separation and identification of peptides was performed by nano-HPLC MS/MS on an Agilent 1100 (Agilent, Santa Clara, CA) coupled to an FT-ICR mass spectrometer (Bruker Daltonics, Billerica, MA; PLT 2- 4) using a 15 cm long, 75 \u00CE\u00BCm I.D. fused silica column packed with 3 \u00CE\u00BCm particle size reverse phase (C18) beads (Dr Maisch GmbH, Germany) with water:acetonitrile:formic acid as the mobile phase with gradient elution. 38 Figure 2.1: Workflow for dimethyl labeling of releasate proteins Donor\u00E2\u0080\u0099s blood provided releasate (Rel) and releasate free of microparticles (Rel-MP) from thrombin, collagen or thrombin plus collagen activated platelets. The proteins were digested and peptides were labeled with different dimethyl tags, and combined for Rel and Rel-MP analysis as described in the methods. In addition, different donor platelets were also taken for activation with collagen, LPA or subthreshold collagen plus LPA and underwent the same workflow as depicted here. In this scenario, before starting, platelets from each donor were tested for response to LPA using aggregometry as described in the methods. In total 12 datasets were generated for Rel and Rel-MP samples. 39 2.7 Dimethyl labeled protein quantitation and bioinformatics The raw files from each Rel and Rel-MP sample from section 2.5 were submitted to Mascot Distiller (version 2.4.2.0, Matrix Science, London, UK) and searched against Swiss-Prot database with locally hosted Mascot server (version 2.3.0). The searches were taxonomically restricted to \u00E2\u0080\u009CHomo sapiens\u00E2\u0080\u009D with MMTS as fixed modification and oxidized methionine as well as the light, medium and heavy dimethyl tags as variable modifications. Maximum allowed mass deviation was set to 20 ppm for monoisotopic precursor ions and 0.6 Da for MS/MS peaks, with a maximum of three missed cleavages. The cutoff used on the peptide level ensures that the worst identified peptide has a probability of 0.05 of being false. For protein quantitation, a minimum of 1 unique peptide in addition to having two or more quantitated peptides were required. The reported protein ratios are the weighted average of the peptide ratios with no normalization. For analysis only proteins with both intermediate/light and heavy/light ratios were chosen. The quantitated proteins were submitted to DAVID Bioinformatics Resources 6.7 (http://david.abcc.ncifcrf.gov/, National Institute of Allergy and Infectious Diseases, NIH) (Da Wei Huang and Lempicki 2008). Using DAVID software the proteins were grouped based on their biological function and cellular compartment gene ontology (GO) terms. 2.8 THP-1 Cell transmigration assay Transmigration assay was performed with 8 \u00C2\u00B5m pore size cell culture inserts and 24 well tissue culture plates (BD bioscience). 225 \u00C2\u00B5l of THP-1 cells in RPMI 1640 media at 1 million cell/ml were placed on the top chamber. 350 \u00C2\u00B5l of RPMI 1640 was added to the bottom chamber along with 50 \u00C2\u00B5l of Rel, Rel-MP or MP from thrombin, collage, thrombin plus collagen, LPA or subthreshold collagen plus LPA activated platelets prepared in the same manner described in section 2.4. 1 ml of Rel or Rel-MP was concentrated and used for the migration assay and was repeated three times. Each repeat used platelets from 40 different individual donors. In addition, 25 \u00C2\u00B5g of platelet factor 4 (PF4) active protein (ab80477, Abcam) was added to 375 \u00C2\u00B5l of media in the bottom chamber. 2.5 \u00C2\u00B5g, 5 \u00C2\u00B5g and 20 \u00C2\u00B5g of PF4 antibody (ab9561, Abcam) was added to 50 \u00C2\u00B5l of Rel (concentrated from 1 ml of Rel) from thrombin activated platelets and added to the bottom chamber along with media. For control, 5 \u00C2\u00B5g of PF4 antibody plus media, media alone, and media supplemented with 10 % FBS was added to bottom chamber. The assay was incubated for 6 hours at 37 \u00C2\u00B0C. 100 \u00C2\u00B5l of the migrated cells were injected into MACSQuant Analyzer flow cytometer (Miltenyi Biotec, Auburn, CA) for counting the number of cells in the solution. 2.9 Membrane enrichment 2.9.1 Protocol A 15 x 106 THP-1 cells were harvested and washed twice in ice cold Tris buffered saline (TBS: 10 mM Trizma base, 0.15 NaCl in dH2O) pH 7.4. The cell pellet was resuspended in 1-2 ml of membrane lysis buffer containing 10 mM TBS pH 7.4, 150 mM NaCl, 1 mM MgCl2 and 1mM protease inhibitors (Complete tablet; Roche, Mannheim, Germany) and incubated on ice for 1 hour. Cells were homogenized by 25 strokes in a 2 ml Dounce homogenizer and passed through a 26-gauge syringe needle 6 times. The homogenate was gently layered on top of a 5-60 % (wt/wt) discontinuous sucrose gradient in gradient buffer (20 mM TBS, 150 mM NaCl, 1 mM EDTA, pH 7.4), and centrifuged at 100,000 x g at 4 \u00C2\u00B0C in a SW41Ti rotor (Beckman, Palo Alto, CA). The membrane proteins fraction was harvested from the 5 %/60 % interface, and diluted 1:1 or 1:3 with gradient buffer. The solution was centrifuged at 100,000 x g at 4 \u00C2\u00B0C for 10 min in a TLA 100.3 rotor (Beckman, Palo Alto, CA). The membrane protein pellet was resuspended in gradient buffer containing 10 % glycerol. 41 2.9.2 Protocol B 15 x 106 THP-1 cells were washed twice with cold TBS and resuspended in 9 % (wt/wt) sucrose-10 mM PIPES [piperazine-N,N\u00E2\u0080\u0099-bis(2-ethanesulfonic acid)], pH 7.2. The cells were disrupted by passage through a 26-guage needle 6 times. The disrupted cells were centrifuged at 600 x g for 3 min. The supernatant was loaded on top of the sucrose gradient. The sucrose gradient was prepared by layering 55 % (0.2 ml), 47 % (0.3 ml), 40 % (1.2 ml), 34 % (0.8 ml), 25 % (0.8 ml) and 20 % (0.6 ml) sucrose solutions (all given as wt/wt in 10 mM PIPES, pH 7.2). The sucrose layer was allowed to diffuse into linearity for 1 hour at 37 \u00C2\u00B0C. The sample was centrifuged at 250,000 x g 4 \u00C2\u00B0C for 1 hour in a SW41 rotor. The top cytosolic fraction and then two membrane bands of increasing equilibrium density (band 1, 1.12 g/ml and band 2, 1.31 g/ml, from top to bottom) were collected. The two membrane bands were combined and centrifuged at 200,000 x g at 4 \u00C2\u00B0C for 10 min in a TLA 100.3 rotor. The membrane protein pellet was resuspended in gradient buffer containing 10 % glycerol. 2.10 Western immuno blotting to test protocol A and B The membrane protein enriched samples from section 2.9 were spun at 20,000 x g at 4 \u00C2\u00B0C for 5 minutes and the pellet was washed with 1 ml of TBS. The pellet was lysed in lysis buffer containing 1 mM EDTA, 2 % Triton X-100 and 1 mM protease inhibitors in TBS, for 1 hour on ice with vortexing every 10 minutes. In addition, proteins from whole cell lysate were also collected by incubating THP-1 cells in lysis buffer for 1 hour on ice with vortexing every 10 minutes. All the lysates were centrifuged for 20 min at 20,000 x g, at 4 \u00C2\u00B0C and supernatants were removed and assayed for protein content, using the BCA. 50 \u00C2\u00B5g of each protein sample was electrophoresed by SDS-PAGE on a 10 % gel in a running buffer solution containing 0.192 M glycine, 25 mM Trizma base 0.1 % SDS in dH2O at pH 8.3. The gel was equilibrated in transfer buffer solution containing 39 mM glycine, 48 mM Trizma base, 0.037 % SDS, 2 % methanol in dH2O at pH 8.3, for 15 min. Proteins were transferred to a methanol activated PVDF membrane buffered by transfer 42 buffer solution. Proteins were incubated in blocking solution (5 % milk powder dissolved in PBST (8 g NaCl, 0.2g KCl, 144 g Na2HPO4, 0.24g KH2PO4, 2 ml of Tween-20 dissolved in 800 ml of dH2O)) for 1 hour at room temperature. The membrane was incubated overnight at 4\u00C2\u00B0C with a primary antibody (MAB1965, EMD Millipore, Billerica, MA) against integrin \u00CE\u00B21, at a concentration of 1:5000 in blocking solution. Following three washes in PBST the membrane was incubated with anti-mouse HRP- conjugated secondary antibody at a concentration of 1:10000 in blocking solution. Following three washes in PBST, protein expression was detected using enhanced chemiluminescent reagent (ECL; Pierce Biotechnology), and exposed to X-ray film. To check for cytosolic protein depletion, the membrane was then exposed to a primary antibody (sc-32233, Santa Cruz Biotechnology, Santa Cruz, CA) against GAPDH, at a concentration of 1:5000 in blocking solution and an anti-mouse HRP-conjugated secondary (1:10,000 in blocking solution). To confirm reproducibility of membrane protein enrichment using protocol A, the above experiment was repeated along with also using a primary antibody (ab14112, Abcam, Cambridge, MA) agaist 14.3.3 protein at a concentration of 1:5000 in blocking solution and an anti- mouse HRP- conjugated secondary (1:10,000 in blocking solution). 2.11 Reference THP-1 cell global and membrane proteome dataset Membrane proteins from THP-1 cells grown in light SILAC media were enriched using protocol A (section 2.8). Proteins from whole cell lysate (global proteins) and from membrane enriched fraction (membrane proteins) were collected and protein concentration was determined by BCA assay as before. 70 \u00C2\u00B5g of global and membrane proteins were each separated by SDS-PAGE on a 10 % gel. Gels were stained with Coomassie Brilliant Blue, imaged and each lane was cut into 24 bands. Standard reduction, alkylation and tryptic in-gel digestions were performed overnight at 37 \u00C2\u00B0C. Following digestion, peptides were extracted, desalted and analyzed on the mass spectrometer. These experiments were repeated two 43 more times to obtain three biological repeats for global and membrane proteomes. The proteins were identified in Mascot in the same manner as section 2.3. The resulting shared proteins from the global and membrane proteome were grouped based on their cellular compartment gene ontology (GO) terms. Proteins were submitted to DAVID Bioinformatics Resources 6.7 (http://david.abcc.ncifcrf.gov/, National Institute of Allergy and Infectious Diseases, NIH) for GO analysis. 2.12 Preparation of platelet releasate for THP-1 cell stimulation Washed platelets from section 2.2 were resuspended in RPMI 1640 media (supplemented with CaCl2 to a final concentration of 1.8 mM) to 10 x 108 platelets/ml before being activated with 0.1 U/ml human \u00CE\u00B1- thrombin (Sigma, St. Louis, MO) for 90 minutes to induce release of platelet components into the media. The samples were then placed on ice and hirudin (0.4 U/ml, Sigma) was added to neutralize thrombin. The platelets were then removed by centrifugation at 700 x g for 10 minutes at 4\u00C2\u00B0C. The supernatant was centrifuged again and the resulting supernatant represents the total platelet releasate which was added to THP-1 cells for stimulation. 2.13 THP-1 cell stimulation using platelet releasate Platelet releasate (from section 2.12) from 4 different donors were combined and added to heavy SILAC labeled THP-1 cells at 1 ml of Rel per 1 x 106 THP-1 cells and incubated for 6 and 24 hours at 37 \u00C2\u00B0C. Equal volume of RPMI 1640 was added to light SILAC labeled THP-1 cells for control and incubated for 6 and 24 hours at 37 \u00C2\u00B0C. Global proteins and membrane proteins from light and heavy labeled THP-1 cells were collected as per section 2.10, and protein concentration was determined by BCA assay as before. Equal amounts of light and heavy SILAC labeled global proteins, and equal amounts of light and heavy SILAC labeled membrane proteins from 6 and 24 hours of stimulation were mixed and separated by SDS-PAGE 44 on a 10 % gel. Gels were stained with Coomassie Brilliant Blue, imaged and each lane was cut into 24 bands. Standard reduction, alkylation and tryptic in-gel digestions were performed overnight at 37 \u00C2\u00B0C. Following digestion, peptides were extracted, desalted and analyzed by tandem mass spectrometry according to section 2.5 (Figure 2.2). These experiments were repeated two more times to obtain three biological repeats for global and membrane proteome after 6 and 24 hours of stimulation. In addition, THP-1 cells stimulated for 4, 6 or 10 hours were analyzed for MAC-1 expression. 2.5 x 105 cells at 4, 6 or 10 hours were washed twice in cold TBS, and pelleted. The cells were stained with 5 \u00C2\u00B5l of fluorescein isothiocyanate (FITC) conjugated anti-human CD11b (MAC-1) (11-0113-73, eBiosciences, San Diego, CA) for 20 minutes in the dark on ice. The cells were then washed twice in FACS buffer (2 % FBS, 0.002 M EDTA, 3.7 % formaldehyde in PBS), resuspended in 300 \u00C2\u00B5l of FACS buffer and analyzed on LSR II flow cytometer (BD Biosciences, San Jose, CA). THP-1 cells stimulated for 6 and 24 hours were stained with 20 \u00C2\u00B5l of phycoerythrin (PE) anti-human CD62P (348107, BD Biosciences) and also washed and analyzed on LSR II flow cytometer. Unstimulated THP-1 cells were stained in the same manner, and as well as unstained THP-1 cells were used as control. 45 Figure 2.2: Work flow for SILAC quantitative experiments THP-1 cells grown in RPMI 1640 media supplemented with 1H1-L-lysine and 12C6-L-arginine (\u00E2\u0080\u009Clight\u00E2\u0080\u009D media) were treated as control. THP-1 cells grown in RPMI 1640 media supplemented with 2H1 (D)-L- lysine and 13C6-L-arginine (\u00E2\u0080\u009Cheavy\u00E2\u0080\u009D media) were treated with releasate (Rel) from thrombin activated platelets. Equal amounts of global proteins (whole cell lysate) from \u00E2\u0080\u009Clight\u00E2\u0080\u009D and \u00E2\u0080\u009Cheavy\u00E2\u0080\u009D labeled THP-1 cells were combined and identified as described in the methods. Protein ratios were calculated after 6 and 24 hours of THP-1 cell stimulation. In addition, membrane proteins were enriched from THP-1 cells and analyzed in the same manner. These experiments were repeated 3 times to generate 12 datasets in total. THP-1 cells grown in \u00E2\u0080\u009CHeavy\u00E2\u0080\u009C (+4 Lys. +6 Arg) media THP-1 cells grown in \u00E2\u0080\u009Clight\u00E2\u0080\u009D media Control \u00E2\u0080\u0093 treated with media Rel-treated (6, 24 hours) I m/z Harvest cells by centrifugation 1) Global analysis: lyse THP-1 cells. 2) Membrane analysis: homogenize cells in hypotonic buffer, place on a 60 %/5 % sucrose gradient and centrifuge at 100,000 x g; enriched cell membranes are collected from the interface 6 h 24 h Membrane Combine 1:1 ratio for global and membrane proteome and separate by SDS-PAGE Excise, digest and identify proteins by MS/MS. Quantify proteins from MS spectra using MaxQuant 6 h 24 h Global 46 2.14 SILAC labeled protein quantitation and bioinformatics Raw MS spectra from each experiment from section 2.13 were analyzed using MaxQuant (version 1.0.13.13, http://maxquant.org/, Max Planc Institute, Germany). The data were searched against the human International Protein Index protein sequence database (IPI version 3.69 (The International Protein Index: an integrated database for proteomics experiments) using a locally hosted Mascot server (version 2.3.0, Matrix Science, London, UK). Enzyme specificity was set to trypsin, and carbamidomethyl cysteine was set as a fixed modification, and oxidized methionine was set as a variable modification. Spectra determined to result from heavy labeled peptides by presearch MaxQuant analysis were searched with the additional fixed modifications Arg6 and Lys4, whereas spectra with a SILAC state not determinable a priori were searched with Arg6 and Lys4 as additional variable modifications. Maximum allowed mass deviation was set to 20 ppm for monoisotopic precursor ions and 0.6 Da for MS/MS peaks. A maximum of three missed cleavages and three labeled amino acids (arginine and lysine) were allowed. The cut-off used on the peptide level ensures that the worst identified peptide has a probability of 0.05 of being false. If the identified peptide sequence set of one protein was equal to or contained the peptide set of another protein, these two proteins were grouped together by MaxQuant and not counted as independent protein hits. For SILAC quantitation, only proteins with two or more quantitated peptides were considered. MaxQuant generated the list of high confidence proteins with their normalized ratios based on the median for the 12 SILAC experiments from section 2.12. The global and membrane proteins along with their average normalized heavy to light ratios were submitted to Ingenuity Pathway Analysis (IPA) software (http://www.ingenuity.com/) for core analysis. The result of the core analysis showed the pathways and the biological functions through which the identified proteins connect. 47 3 Results 3.1 Platelet releasate analysis Through proteomic studies we analyzed the protein content of platelet releasate when activated with different agonists. These studies helped us identify the proteins in the releasate which can potentially stimulate monocytes, and the biological role the releasate proteins might have in the body. Washed platelets were activated with thrombin (1U/ml), collagen (0.19 mg/ml), a combination of thrombin plus collagen, mildly oxidized low density lipoprotein (mox-LDL; 1 mg/ml), LPA (20 mM) or subthreshold collagen plus LPA (3 \u00C2\u00B5g/ml collagen and after 1 minute 20 mM LPA) for 10 minutes at 37 \u00C2\u00B0C. These agonists activate platelets for hemostasis as well as playing a role in platelet activation during plaque rupture in patients with cardiovascular disorder. Thrombin represents a strong platelet activator which has been extensively studied. Collagen activates platelets in order to form platelet plugs during vessel injury and is also exposed during plaque rupture in patients with cardiovascular disorder. Mox- LDL is another agonist which is exposed to platelets during plaque rupture, and LPA is the main component of mox-LDL responsible for platelet activation. Using these agonists, we can analyze the similarities and differences in platelet activation for the purposes of hemostasis as compared to platelet activation during plaque rupture. The total content released from the platelets due to activation (Rel) was collected and the amount of proteins in each sample measured using BCA assay analysis (Table 3.1). There is variation in these values from sample to sample and only represent trends seen in the amount of proteins released after using each stimuli. The results in the table are based on at least three repeats. Although mox-LDL seems to cause the release of the most amounts of proteins from platelets, in fact, the majority of the proteins measured by BCA are from the mox-LDL added to platelets for activation. Activation of platelets 48 by LPA seems to actually cause the most amounts of proteins released at an estimated amount of 159 \u00C2\u00B5g per 3 mL of washed platelets at 300 x 106 platelets/mL (physiological platelet concentration). Table 3.1: Estimated average amount of protein found in each compartment of platelet releasate upon activation with various agonists These estimated values were obtained by BCA (bicinchoninic acid) assay analysis of the platelet releasate from washed platelets resuspended in HEPES buffer with 1.8 mM added CaCl2. In general thrombin produces the least amount of microparticles (MP) based on these studies. The amount of proteins released from mildly oxidized low density lipoproteins (mox-LDL) activated platelets is over estimated due to the presence of proteins from oxLDL which were added to the platelets. Each average is based on at least 3 independent measurements. Rel-MP represents the fraction of releasate after removing MP via ultracentrifugation. Abbreviations: Thr+Coll, thrombin plus collagen; Coll+LPA, subthreshold collagen plus LPA. Compartments Average estimated amount of proteins from 3mL of platelet releasate at [phys.] (\u00C2\u00B5g) Thrombin Collagen Thr+Coll Mox-LDL LPA Coll+LPA Releasate 80 132 92 498 159 184 Rel-MP 68 98 80 474 150 134 MP 3 30 25 73 28 30 The releasate obtained from the activated platelets were also subject to ultracentrifugation to isolate the platelet microparticles (MP) and the supernatant free of MP (Rel-MP). These compartments of platelet releasate were also analyzed using BCA assay to find the estimated amount of proteins they contain. Thrombin activated platelets seem to release the fewest amount of MP as evident by the amount of proteins measured using BCA assay (Table 3.1). From these results alone we can suspect that the releasate proteome profile will look very different when the various agonists are used. 49 We identified the proteins released from platelets when mox-LDL is used as an agonist. Rel, Rel- MP and MP proteins from mox-LDL were separated using SDS-PAGE (Figure 3.1A). There is a strong band at around 130 KDa in all three lanes, with very few distinguishable bands in the other parts of the lanes. Proteins in the Rel, Rel-MP and MP samples from mox-LDL activated platelets were identified by cutting the bands from the corresponding gel lanes and analyzing them using mass spectrometer. There are 491 proteins from the Rel, 314 proteins from the Rel-MP and 455 proteins from the MP when mox-LDL is used as an agonist (Figure 3.1B). Not surprisingly, the most abundant protein (highest Mascot score) in the samples is apolipoproteins, and made identification of other proteins difficult. Due to lipoprotein contamination in the samples from mox-LDL activated platelets, LPA was used as an agonist for quantitative proteomics experiments. Proteins in Rel and Rel-MP from thrombin, collagen or thrombin plus collagen activated platelets were quantitatively compared to analyze the effect of using agonists which are exposed to platelets during hemostasis. Similarly, protein in Rel and Rel-MP from LPA, collagen or subthreshold collagen plus LPA were quantitatively compared to analyze the effect of using agonists which are exposed to platelets during plaque rupture in patients with cardiovascular disorder. By comparing Rel to Rel-MP we can get a better idea of whether a given protein is primarily in the soluble fraction of platelet releasate or packaged into MP. Quantitative proteomics studies helped shed light on the relative abundance of proteins released from platelets activated by these agonists. 50 Figure 3.1: Releasate proteins from mox-LDL activated platelets Washed platelets resuspended at physiological concentrations were activated with mildly oxidized low density lipoprotein (mox-LDL). The releasate (Rel) was isolated from the activated platelets and part of the releasate was subject to ultracentrifugation to isolate the platelet microparticles (MP). The proteins from the Rel, MP and the remaining supernatant (Rel-MP) were also separated using SDS-PAGE gel and analyzed using LC-MS/MS for identification (A). (B) Venn diagram showing the overlap of identified proteins from the different compartments of platelet releasate. Rel (491 proteins) Rel-MP (314 proteins) 91 269 118 224 48 57 83 MP (455 proteins) B. A. mox-LDL Rel-MP 170 130 100 70 55 40 35 25 MWM (kDa) MP Rel 51 3.2 Quantitative proteomic analysis on platelet releasate when using thrombin, collagen and thrombin plus collagen as agonists Equal numbers of washed platelets resuspended at physiological concentrations were activated with thrombin, collagen or thrombin plus collagen. The proteins from the Rel and Rel-MP were digested in- solution and the peptides were labeled with light, medium or heavy dimethyl tags, for quantitative proteomic analysis. The Rel peptides from the differentially activated platelets were combined and analyzed together using tandem mass spectrometry and Mascot Distiller, and the Rel-MP peptides were also combined and analyzed together (Figure 2.1). We found 97 quantified proteins in common between the three biological repeat experiments when analyzing the Rel samples (Figure 3.2A) and 97 quantified proteins in common from three Rel-MP samples (Figure 3.2B). There are also 86 common proteins in common between the Rel and Rel-MP datasets (Figure 3.2C). The proteins in common from the Rel and Rel-MP samples were analyzed based on their thrombin/collagen (T/C) and thrombin/thrombin plus collagen (T/TC) ratios. In other words, these ratios indicate the relative abundance of a given protein in the sample when the platelets are activated with the different agonists. (i.e. a thrombin/collagen ratio of above 1 indicates that a given protein is more abundant in the releasate from thrombin activated platelets relative to the releasate from collagen activated platelets). The ratios from the three samples were averaged and a change of more than 20 % is considered a significant difference between the samples. We performed GO analysis on the proteins identified and grouped them according to their biological function and location within a cell. We looked at the Rel and Rel-MP proteins and based on their ratios we determined each protein\u00E2\u0080\u0099s relative abundance and where they can potentially originate from. 52 Figure 3.2: Overlap of quantified platelet releasate proteins from thrombin, collagen and thrombin plus collagen activated platelets (A) Platelet releasate proteins from thrombin, collagen or thrombin plus collagen activated washed platelets resuspended to physiological concentrations were trypsin digested in solution. Peptides from different platelet activated samples were dimethyl labeled and combined for proteins identification and quantitation using LC-MS/MS. The Venn diagrams represent the common quantified proteins between three biological repeat of the experiment. (B) The releasate free of microparticles (Rel-MP) proteins were also analyzed in the same manner. (C) Venn diagram showing the number of proteins which overlap between the releasate and the Rel-MP datasets. Dataset 1 190 Proteins Dataset 2 155 Proteins Dataset 3 159 Proteins 97 58 22 25 17 19 18 Dataset 1 150 Proteins Dataset 2 159 Proteins Dataset 3 202 Proteins 97 23 34 65 9 19 21 C. Releasate versus Rel-MP Releasate all 3 datasets (97 proteins) 7 2 Rel-MP all 3 datasets (97 proteins) 11 86 11 A. Releasate proteins B. Rel-MP proteins 53 3.2.1 Analysis of Rel proteins with high relative abundance By studying the proteins which are relatively more abundant when a particular agonist is used, we can get an idea of the biological consequences of platelet activation. There were 25 proteins with average T/C ratio of greater than 1.2 as compared to 49 proteins with an average T/C ratio of less than 0.8 (Table 3.2A). The 25 Rel proteins with average T/C greater than 1.2 were submitted to DAVID Bioinformatics Resources 6.7 biological processes GO analysis (Figure 3.3A). These 25 proteins were relatively more abundant in Rel from thrombin activated platelets, as compared to collagen activated platelets, and by grouping them based on biological processes GO term we can get a better idea of their function in the body. The top 20 GO terms with the number of proteins that belong to each term was graphed. These GO terms include response to wounding (10 proteins), immune response (9 proteins), coagulation (5 proteins) and inflammatory response (6 proteins) among others. Similarly, the 49 proteins that were relatively more abundant in Rel when collagen was used as an agonist were also submitted to DAVID for GO analysis (Figure 3.3B). The top biological processes from this list include cell motion (14 proteins), actin filament-based process (10 proteins) and actin cytoskeleton organization (11 proteins) among others. Based on these data we can see that abundant proteins from thrombin activated platelets play a different functional role in the body as compared to abundant proteins from collagen activated platelets. Next, we looked at T/TC ratios and found that on average, there were 46 proteins more abundant in the Rel sample when thrombin was used as compared to 22 abundant proteins when thrombin plus collagen was used (Table 3.2A). Once again, we performed GP analysis, to find out the biological functions that these 46 proteins perform, and saw that a lot of GO terms are the same as the ones in Figure 3.3A, but the numbers of proteins for each GO term were different (Figure 3.4A). These 54 terms include coagulation (13 proteins) and response to wound healing (20 proteins). There were also some new GO categories such as cell motion (10 proteins), which was the top GO term for proteins more abundant in the releasate from collagen activated platelets when compared to thrombin activated platelets. Figure 3.4B shows the GO terms which were generated based on the 22 proteins with a T/TC ratio less than 0.8. These proteins include collagen which was added to the solution, actin cytoplasmic 1 and platelet glycoprotein Ib. These proteins and others were grouped into GO terms such as platelet activation (5 proteins), protein polymerization (5 proteins) and blood coagulation (5 proteins). Overall we see that abundant protein from thrombin activated platelets play a different role to abundant protein from thrombin plus collagen activated platelets. There were in fact more proteins with a T/TC ratio of greater than 1.2 as compared to those with a T/TC ratio less than 0.8, suggesting that overall combining the two agonists lead to less relative secretion when compared to using each agonist alone. In addition, Table 3.3 shows the effect of activating platelets with thrombin plus collagen. There were only 6 proteins whose average T/C ratio was significantly greater than their average T/TC ratio. On the other hand, there were 67 proteins whose average T/C ratio was significantly less than their average T/TC ratio. These 67 proteins were grouped based on their T/C ratios. The majority of proteins affected by the addition of thrombin to collagen for platelet activation are cytoskeletal proteins (Table 3.3B). These data show the effect of combining agonists as opposed to activating platelets with each agonist alone. To understand the contribution of MP to the total releasate we performed the same analysis on the Rel-MP samples and looked at how the T/C and T/TC ratios changed. 55 Table 3.2: Change in protein abundance when using different agonists (A) Proteins with average thrombin/collagen (T/C) ratios greater than 1.2 represent proteins which are more abundant in releasate (Rel) from thrombin activated platelets as compared to collagen. Proteins with T/C ratios lower than 0.8 are the proteins more abundant in the Rel from collagen activated platelets. The same logic applies when comparing thrombin/thrombin plus collagen (T/TC) ratios. (B) Here the ratios of Rel-MP (the soluble fraction of platelet releasate free of microparticles) are analyzed in the same manner. (C) This table represent the number of proteins which can potentially be abundant in the MP. If Rel T/C or T/TC is significantly greater than the Rel-MP T/C or T/TC then the proteins could come from MP generated through thrombin platelet activation. If Rel T/C ratio is significantly less than the Rel-MP T/C ratio then the proteins could originate from MP generated through collagen platelet activation. A. Releasate # of proteins with avg. T/C greater than 1.2 25 # of proteins with avg. T/C less than 0.8 49 # of proteins with avg. T/TC greater than 1.2 46 # of proteins with avg. T/TC less than 0.8 22 B. Rel-MP # of proteins with avg. T/C greater than 1.2 34 # of proteins with avg. T/C less than 0.8 34 # of proteins with avg. T/TC greater than 1.2 65 # of proteins with avg. T/TC less than 0.8 9 C. Common proteins btw Rel and Rel-MP # of proteins with avg. Rel T/C more than 20 % greater than avg. Rel-MP T/C 11 # of proteins with avg. Rel T/C more than 20 % less than avg. Rel-MP T/C 21 # of proteins with avg. Rel T/TC more than 20 % less than avg. Rel-MP T/TC 41 56 Table 3.3: Effect of activating platelets with thrombin plus collagen (A) There are 6 proteins which have a thrombin/collagen (T/C) ratio significantly greater than thrombin/thrombin plus collagen (T/TC) ratio, compared to 67 proteins which have a T/C ratio less than T/TC in Rel sample. (B) Some of the 67 proteins have a T/C ratio significantly over 1 and their T/TC ratio increases. There are 16 proteins which are equally abundant when comparing releasate from thrombin activated platelets to collagen activated platelets, but their T/TC ratios are over 1.2. There are 36 proteins which have a T/C ratio of less than 0.8 but their T/TC ratio is around 1. The top cellular compartment gene ontology terms for these proteins are included in the table. A. Effect of adding thrombin plus collagen Protein category Number of proteins AVG T/C ratio over 20 % greater than T/TC 6 AVG T/C ratio over 20 % less than T/TC 67 B. Break down of the 67 proteins Protein category Number of proteins Top cellular compartment T/C ratio of greater than 1.2 15 Extracellular region (13 proteins) T/C ratio around 1 16 \u00CE\u00B1-Granule (5 proteins) T/C ratio less than 0.8 36 Cytoskeleton (20 proteins) 57 Figure 3.3: Quantified releasate proteins from thrombin versus collagen samples The platelet releasate quantified proteins in common from thrombin, collagen and thrombin plus collagen activated platelets were subject to gene ontology (GO) analysis on DAVID software. (A) The top 20 biological process associated with proteins which are more abundant in the releasate from thrombin activated platelets as compared to collagen. (B) The top 20 biological process associated with proteins which are more abundant in the releasate from collagen activated platelets as compared to thrombin. 0 2 4 6 8 10 12 # o f p ro te in s Biological process GO terms A. Thrombin overexpressed proteins 0 2 4 6 8 10 12 14 16 # o f p ro te in s Biological process GO terms B. Collagen over expressed proteins 58 Figure 3.4: Quantified releasate proteins from thrombin versus thrombin plus collagen samples Releasate proteins were subject to gene ontology (GO) analysis using DAVID software. (A) The top 20 biological process associated with proteins which are more abundant in the releasate from thrombin activated platelets as compared to releasate from thrombin plus collagen activated platelets. (B) The top 20 GO terms associated with proteins which are more abundant in the releasate from thrombin plus collagen activated platelets. Overall, we see that when platelets are activated with thrombin they have a very different platelet releasate proteome profile as compared to using thrombin plus collagen. 0 5 10 15 20 25 # o f p ro te in s Biological process GO terms A. Thrombin over expressed proteins 0 1 2 3 4 5 6 7 # o f p ro te in s Biological process GO terms B. Thrombin plus collagen over expressed proteins 59 3.2.2 Analysis on Rel-MP proteins with high relative abundance In the Rel-MP samples, there were 34 proteins with an average T/C greater than 1.2 or less than 0.8 (Table 3.2B). On the other hand, there were 65 proteins with an average T/TC ratio greater than 1.2 as compared to only 9 proteins with an average T/TC ratio of less than 0.8 in the Rel-MP (Table 3.2B). When we look at the 34 protein in Rel-MP sample with T/C ratio of less than 0.8, we see that the top biological function GO terms are very different than those for the corresponding Rel graph (Figure 3.5). The GO terms include platelet activation (7 proteins) and coagulation (8 proteins). Terms such as the ones associated with cytoskeletal proteins moved down the list and had fewer proteins associated with them when compared to Figure 3.4B, when we looked at abundant protein in Rel from collagen activated platelets as compared to thrombin activated platelet. Cell motion which was the top term in the Rel sample, moved down to the sixteenth term in the Rel-MP sample. Next, by comparing the Rel dataset to the Rel-MP dataset we can get an idea of proteins potentially abundant in the MP when using the different agonists. 3.2.3 Proteins potentially abundant in MP From the 86 proteins in common between the Rel and Rel-MP datasets, there were 11 proteins which have significantly greater average T/C ratios in Rel sample as compared to Rel-MP sample (Table 3.2C). These 11 proteins can potentially be most abundant in the MP from thrombin activated platelets and include, transforming growth factor \u00CE\u00B21 (TGF\u00CE\u00B21) and nidogen-1 (Appendix A). On the other hand, there were 21 proteins which had an average Rel T/C ratio significantly less than average Rel-MP T/TC ratio. These 21 proteins could potentially be most abundant in MP when platelets are activated with collagen. Examples of these proteins include 14-3-3 protein, coagulation factor V, integrin \u00CE\u00B1IIb (Appendix B). Moreover, there were 41 proteins which had an average Rel T/TC ratio significantly less than average Rel-MP T/TC ratio. These proteins can potentially be most abundant in MP when platelets are activated 60 with thrombin plus collagen and examples include talin-1, Platelet factor 4 and integrin \u00CE\u00B1IIb (Appendix C). Proteins which were unique to the Rel sample can also be potentially most abundant in the MP. There were11 proteins which were unique to Rel samples such as nidogen-2 (Appendix D), and 11 proteins unique to Rel-MP samples such as platelet glycoprotein VI (Appendix E). The proteins unique to Rel could potentially be most abundant in the MP and proteins only found in the Rel-MP can potentially be primarily in the soluble fraction of the releasate. To understand the overall functions of Rel proteins and analyze the different proteome profile of releasate proteins when using the three agonists, we grouped all the Rel proteins using GO analysis on DAVID software. Figure 3.5: Quantified Rel-MP proteins from thrombin versus collagen samples The platelet releasate free of microparticles (Rel-MP) quantified proteins in common between three Rel- MP datasets were subject to gene ontology (GO) analysis on DAVID software. The top 20 biological process associated with proteins which are significantly more abundant in the Rel-MP from collagen activated platelets as compared to Rel-MP from thrombin activated platelets. 0 2 4 6 8 10 Biological process GO Terms Collagen overexpressed proteins 61 3.2.4 GO analysis on all Rel proteins To understand the biological role each Rel protein has, all the proteins in common between the three Rel datasets (97 proteins), were submitted to DAVID GO analysis. Table 3.4 shows the top biological process and table 3.5 shows the top cellular compartment GO terms. The protein grouping based on biological process shows the functional role the Rel proteins play in the body, while the grouping based on cellular compartment gives us an idea of where in the platelets these proteins can originate from. The calculated T/C and T/TC ratios for each dataset along with the average ratios are depicted in the table. The red color represents T/C or T/TC ratios which are greater than 1.2 (i.e. proteins which are more abundant in the Rel from thrombin activated platelets). The green color represents T/C or T/TC which are less than 0.8 (i.e. proteins which are more abundant in the Rel from collagen or thrombin plus collagen activated platelets). All three datasets are shown so that platelet heterogeneity between the three datasets can be highlighted. The top GO terms for biological process are response to wounding, cell adhesion, immune response and inflammatory response (Table 3.4). These GO terms are shown based on the fact that platelets are known to play important roles in each of these categories. Proteins belonging to these categories are important in hemostasis as well as in platelet-monocyte aggregate formation and cardiovascular disease. There were 29 proteins associated with response to wounding. There was variability between the three datasets especially with T/TC ratios. There were more proteins with average T/TC ratios greater than 1.2 as compared to T/C ratios greater than 1.2. Proteins such as platelet factor 4 (PF4) or transforming growth factor \u00CE\u00B21 (TGF\u00CE\u00B21) have average T/C and T/TC ratios of greater than 1.2. Moreover, some proteins such as PF4 have one dataset where the T/C or T/TC ratio is less than 0.8, but the other two datasets have T/C and T/TC ratio of more than 1.2. Some proteins such as coagulation factor XIII have T/C ratios of less than 0.8, but their T/TC ratios are around 1. 62 Looking at immune and inflammatory response, there are more proteins with average T/C ratios highlighted in red. Dataset 1 seems to have the highest T/C and T/TC ratios for proteins such as PF4, C-C motif chemokine 5 and complement C3. Table 3.4: Releasate proteins and their ratios grouped according to biological process gene ontology (GO) terms All the releasate proteins in common between the three datasets were searched against biological process GO terms using DAVID bioinformatics software. The proteins are grouped according to their involvement in the different GO terms including wound healing, cell motion, immune response and inflammatory response. The red color indicates a ratio of more than 1.2. The green color indicates a ratio of less than 0.8. The three T/C and T/TC ratios for all three datasets are shown along with the average ratios. Abbreviations: T/C, thrombin/collagen; T/TC, thrombin/thrombin plus collagen. Response to wounding/wound healing/coagulation/blood coagulation/hemostasis/regulation of body fluid levels T/C 1 T/C 2 T/C 3 AVG T/C Protein name T/TC 1 T/TC 2 T/TC 3 AVG T/TC 16.17 16.58 47.33 26.69 Prothrombin 1.78 1.22 1.57 1.52 3.94 0.51 1.88 2.11 Platelet factor 4 2.37 0.69 2.78 1.95 2.12 1.43 1.62 1.72 Coagulation factor V 1.71 1.69 1.91 1.77 2.41 1 1.46 1.63 Complement factor H 2.17 1.35 4.6 2.71 1.41 2.32 0.64 1.46 Transforming growth factor beta-1 1.47 2.92 0.94 1.78 2.6 0.51 1.14 1.42 C-C motif chemokine 5 3.69 0.7 1.28 1.89 1.83 0.9 1.29 1.34 Multimerin-1 1.84 1.3 2.02 1.72 1.81 1.13 1.04 1.33 Serotransferrin 2.77 1.58 0.38 1.58 1.69 1.09 1.06 1.28 Complement C3 1.87 1.43 1.48 1.59 1.65 1.05 0.92 1.21 Platelet glycoprotein V 1.3 1.36 1.25 1.3 1.83 0.77 1.04 1.21 Plasminogen activator inhibitor 1 2.31 0.99 1.69 1.67 1.58 1.02 0.95 1.19 Thrombospondin-1 1.44 1.3 1.29 1.34 1.61 0.58 1.19 1.13 Alpha-1-antitrypsin 2.08 0.99 2.37 1.81 1.78 0.63 0.8 1.07 Fibronectin 7.61 0.89 1.11 3.2 1.48 0.87 0.84 1.06 Plasminogen 1.53 1.33 1.37 1.41 1.38 0.56 1 0.98 von Willebrand factor 1.55 1.21 2.39 1.71 1.32 0.76 0.79 0.96 Kininogen-1 1.83 2.03 0.7 1.52 63 Response to wounding continued\u00E2\u0080\u00A6 T/C 1 T/C 2 T/C 3 AVG T/C Protein name T/TC 1 T/TC 2 T/TC 3 AVG T/TC 0.97 0.72 0.99 0.89 Gelsolin 0.89 0.67 1.4 0.99 1.09 0.64 0.92 0.88 Vitamin K-dependent protein S 1.25 0.91 1.63 1.26 0.6 1.08 0.8 0.83 Pleckstrin 0.66 2.03 1.72 1.47 0.91 0.72 0.82 0.82 Platelet glycoprotein Ib alpha chain 0.46 0.69 0.65 0.6 0.93 0.38 0.92 0.74 Clusterin 1.14 0.64 1.32 1.03 0.78 0.6 0.39 0.59 Trem-like transcript 1 protein 0.51 0.56 0.35 0.47 0.59 0.57 0.43 0.53 14-3-3 protein zeta/delta 0.61 0.96 0.55 0.71 0.48 0.48 0.46 0.47 Coagulation factor XIII A chain 0.54 1.2 1.73 1.15 0.2 0 0.01 0.07 Fibrinogen gamma chain 6.53 0.02 0.26 2.27 0.1 0.02 0.03 0.05 Fibrinogen alpha chain 1.71 0.07 0.47 0.75 0.07 0.01 0.02 0.03 Fibrinogen beta chain 2.1 0.04 0.24 0.8 4.E-04 4.E-04 1.E-03 7.E-04 Collagen alpha-1(III) chain 3.E-04 7.E-04 2.E-03 9.E-04 Cell adhesion T/C 1 T/C 2 T/C 3 AVG T/C Protein name T/TC 1 T/TC 2 T/TC 3 AVG T/TC 1.76 2.24 6.06 3.36 Nidogen-1 1.59 3.54 11.29 5.47 2.78 4.06 1.26 2.7 Nidogen-2 1.95 5.38 1.35 2.89 2.12 1.43 1.62 1.72 Coagulation factor V 1.71 1.69 1.91 1.77 1.73 1.69 1.08 1.5 Amyloid beta A4 protein 3.41 2.13 1.45 2.33 2.6 0.51 1.14 1.42 C-C motif chemokine 5 3.69 0.7 1.28 1.89 1.83 0.9 1.29 1.34 Multimerin-1 1.84 1.3 2.02 1.72 1.65 1.05 0.92 1.21 Platelet glycoprotein V 1.3 1.36 1.25 1.3 1.58 1.02 0.95 1.19 Thrombospondin-1 1.44 1.3 1.29 1.34 1.78 0.63 0.8 1.07 Fibronectin 7.61 0.89 1.11 3.2 1.38 0.56 1 0.98 von Willebrand factor 1.55 1.21 2.39 1.71 0.6 1.08 0.8 0.83 Pleckstrin 0.66 2.03 1.72 1.47 0.91 0.72 0.82 0.82 Platelet glycoprotein Ib alpha chain 0.46 0.69 0.65 0.6 0.28 0.16 1.83 0.76 Integrin alpha-IIb 0.08 0.34 1.23 0.55 0.72 0.64 0.76 0.71 Alpha-actinin-1 0.88 1.06 1.18 1.04 0.45 0.42 1.19 0.69 Myosin-9 0.58 0.79 2.23 1.2 0.61 0.56 0.84 0.67 Vinculin 0.57 0.9 1.49 0.98 0.35 0.53 0.76 0.55 Zyxin 0.32 0.87 1.4 0.86 0.09 0.15 0.57 0.27 Integrin-linked protein kinase 0.17 0.8 1.37 0.78 4.E-04 4.E-04 1.E-03 7.E-04 Collagen alpha-1(III) chain 3.E-04 7.E-04 2.E-03 9.E-04 64 Immune response T/C 1 T/C 2 T/C 3 AVG T/C Protein name T/TC 1 T/TC 2 T/TC 3 AVG T/TC 3.94 0.51 1.88 2.11 Platelet factor 4 2.37 0.69 2.78 1.95 2.41 1 1.46 1.63 Complement factor H 2.17 1.35 4.6 2.71 1.41 2.32 0.64 1.46 Transforming growth factor beta-1 1.47 2.92 0.94 1.78 1.91 1.35 1.03 1.43 Ig gamma-1 chain C region 1.91 1.94 1.34 1.73 2.6 0.51 1.14 1.42 C-C motif chemokine 5 3.69 0.7 1.28 1.89 1.79 1.48 0.76 1.34 Ig kappa chain C region 1.76 1.66 0.84 1.42 1.77 1.02 1.07 1.29 Ig lambda chain C regions 1.69 1.4 1.24 1.44 1.69 1.09 1.06 1.28 Complement C3 1.87 1.43 1.48 1.59 1.71 0.82 1.11 1.21 Platelet basic protein 1.09 0.83 1.64 1.19 1.58 1.02 0.95 1.19 Thrombospondin-1 1.44 1.3 1.29 1.34 0.93 0.38 0.92 0.74 Clusterin 1.14 0.64 1.32 1.03 0.78 0.6 0.39 0.59 Trem-like transcript 1 protein 0.51 0.56 0.35 0.47 0.59 0.57 0.43 0.53 14-3-3 protein zeta/delta 0.61 0.96 0.55 0.71 0.38 0.31 0.52 0.4 Tubulin beta chain 0.42 0.61 0.97 0.67 Inflammatory response T/C 1 T/C 2 T/C 3 AVG T/C Protein name T/TC 1 T/TC 2 T/TC 3 AVG T/TC 16.17 16.58 47.33 26.69 Prothrombin 1.78 1.22 1.57 1.52 2.41 1 1.46 1.63 Complement factor H 2.17 1.35 4.6 2.71 1.41 2.32 0.64 1.46 Transforming growth factor beta-1 1.47 2.92 0.94 1.78 2.6 0.51 1.14 1.42 C-C motif chemokine 5 3.69 0.7 1.28 1.89 1.81 1.13 1.04 1.33 Serotransferrin 2.77 1.58 0.38 1.58 1.69 1.09 1.06 1.28 Complement C3 1.87 1.43 1.48 1.59 1.58 1.02 0.95 1.19 Thrombospondin-1 1.44 1.3 1.29 1.34 1.61 0.58 1.19 1.13 Alpha-1-antitrypsin 2.08 0.99 2.37 1.81 1.78 0.63 0.8 1.07 Fibronectin 7.61 0.89 1.11 3.2 1.32 0.76 0.79 0.96 Kininogen-1 1.83 2.03 0.7 1.52 0.93 0.38 0.92 0.74 Clusterin 1.14 0.64 1.32 1.03 0.59 0.57 0.43 0.53 14-3-3 protein zeta/delta 0.61 0.96 0.55 0.71 65 The proteins in common between all three Rel datasets were also grouped according to the cellular compartment GO terms (Table 3.5). These GO terms include platelet \u00CE\u00B1-granule, cytoskeleton and plasma membrane. Proteins grouped based on platelet \u00CE\u00B1-granule is shown since \u00CE\u00B1-granules are important secretory organelles which release their content upon platelet activation. There were 25 proteins belonging to platelet \u00CE\u00B1-granules, 7 of which were more abundant when platelets are activated with thrombin and 6 of which when they were activated by collagen. When looking at the T/TC ratios for these proteins there were more proteins with T/TC ratio of greater than 1.2. Again there is variability between the datasets. Cytoskeleton GO term is included for the fact that most of the proteins in this GO term are more abundant in collagen activated platelets. Moreover, these proteins have average T/TC ratios which are higher than their average T/TC values. There are 37 proteins (including collagen) associated with plasma membrane. Since platelets shed membrane proteins during activation, we wanted to see which of the identified proteins could potentially come from shedding. These proteins include glycoproteins and integrins. There were 8 proteins with average T/C ratio greater than 1.2, as compared to 21 proteins with average T/C ratio of less than 0.8. Again adding thrombin and collagen together causes greater variability between datasets with T/TC ratios generally being larger than T/C ratios. Overall, we see the differences in platelet releasate proteome which arises from activating platelets with the different agonists. We also see how two agonists working in combination can alter the proteome profile as compared to using each agonist separately. We can now compare these data with the data generated when LPA is used as an agonist, and use collagen as a bridge to compare all the releasates from platelets activated with the various agonists. 66 Table 3.5: Releasate proteins and their ratios grouped according to cellular compartment gene ontology (GO) terms The proteins in common between the three releasate datasets are grouped according to their cellular compartment such as belonging to \u00C9\u0091-granules or plasma membrane. The red color indicates a ratio of more than 1.2, The green color indicates a ratio of less than 0.8. The ratios for all three datasets along with their average ratios are shown. Abbreviations: T/C, thrombin/collagen; T/TC, thrombin/thrombin plus collagen. Platelet \u00CE\u00B1-granule T/C 1 T/C 2 T/C 3 AVG T/C Protein name T/TC 1 T/TC 2 T/TC 3 AVG T/TC 3.94 0.51 1.88 2.11 Platelet factor 4 2.37 0.69 2.78 1.95 2.12 1.43 1.62 1.72 Coagulation factor V 1.71 1.69 1.91 1.77 1.73 1.69 1.08 1.5 Amyloid beta A4 protein 3.41 2.13 1.45 2.33 1.41 2.32 0.64 1.46 Transforming growth factor beta-1 1.47 2.92 0.94 1.78 1.83 0.9 1.29 1.34 Multimerin-1 1.84 1.3 2.02 1.72 1.41 1.13 1.16 1.23 Metalloproteinase inhibitor 1 1.34 1.15 2.62 1.7 1.71 0.82 1.11 1.21 Platelet basic protein 1.09 0.83 1.64 1.19 1.65 1.05 0.92 1.21 Platelet glycoprotein V 1.3 1.36 1.25 1.3 1.58 1.02 0.95 1.19 Thrombospondin-1 1.44 1.3 1.29 1.34 1.61 0.58 1.19 1.13 Alpha-1-antitrypsin 2.08 0.99 2.37 1.81 1.67 0.52 1.11 1.1 Serum albumin 1.73 0.76 1.27 1.26 1.45 0.85 0.98 1.09 SPARC 1.28 0.82 1.5 1.2 1.78 0.63 0.8 1.07 Fibronectin 7.61 0.89 1.11 3.2 1.38 0.56 1 0.98 von Willebrand factor 1.55 1.21 2.39 1.71 1.09 0.64 0.92 0.88 Vitamin K-dependent protein S 1.25 0.91 1.63 1.26 1.34 0.59 0.71 0.88 Serglycin 1.25 1.17 1.15 1.19 0.91 0.72 0.82 0.82 Platelet glycoprotein Ib alpha chain 0.46 0.69 0.65 0.6 0.28 0.16 1.83 0.76 Integrin alpha-IIb 0.08 0.34 1.23 0.55 0.93 0.38 0.92 0.74 Clusterin 1.14 0.64 1.32 1.03 0.72 0.64 0.76 0.71 Alpha-actinin-1 0.88 1.06 1.18 1.04 0.78 0.6 0.39 0.59 Trem-like transcript 1 protein 0.51 0.56 0.35 0.47 0.48 0.48 0.46 0.47 Coagulation factor XIII A chain 0.54 1.2 1.73 1.15 0.2 0 0.01 0.07 Fibrinogen gamma chain 6.53 0.02 0.26 2.27 0.1 0.02 0.03 0.05 Fibrinogen alpha chain 1.71 0.07 0.47 0.75 0.07 0.01 0.02 0.03 Fibrinogen beta chain 2.1 0.04 0.24 0.8 67 Cytoskeleton T/C 1 T/C 2 T/C 3 AVG T/C Protein name T/TC 1 T/TC 2 T/TC 3 AVG T/TC 11.3 0.84 0.91 4.35 WD repeat-containing protein 1 15.36 1.34 1.13 5.94 1.73 1.69 1.08 1.5 Amyloid beta A4 protein 3.41 2.13 1.45 2.33 1.49 0.81 0.93 1.08 Cofilin-1 1.55 1.29 1.39 1.41 0.97 0.72 0.99 0.89 Gelsolin 0.89 0.67 1.4 0.99 0.52 1.19 0.94 0.88 Calmodulin 0.62 63.6 4.04 22.75 1.06 0.75 0.76 0.85 Actin-related protein 2 0.53 1.21 1.09 0.94 0.62 0.97 0.7 0.76 Profilin-1 0.63 1.37 1.03 1.01 0.95 0.43 0.81 0.73 Myosin light polypeptide 6 5.57 0.65 1.66 2.62 0.72 0.64 0.76 0.71 Alpha-actinin-1 0.88 1.06 1.18 1.04 0.45 0.42 1.19 0.69 Myosin-9 0.58 0.79 2.23 1.2 0.61 0.56 0.84 0.67 Vinculin 0.57 0.9 1.49 0.98 0.61 0.68 0.67 0.65 Fructose-bisphosphate aldolase A 0.79 1.25 0.92 0.98 0.54 0.61 0.66 0.6 Filamin-A 0.59 0.98 1.02 0.86 0.5 0.62 0.67 0.6 Tropomyosin alpha-4 chain 0.81 3.15 1.11 1.69 0.61 0.63 0.53 0.59 Septin-6 0.75 1.15 2.32 1.41 0.49 0.53 0.74 0.59 Thymosin beta-4-like protein 1 0.61 0.71 1.07 0.79 0.39 0.54 0.78 0.57 PDZ and LIM domain protein 1 0.49 0.92 1.38 0.93 0.35 0.53 0.76 0.55 Zyxin 0.32 0.87 1.4 0.86 0.41 0.38 0.81 0.54 Talin-1 0.48 0.74 1.8 1.01 0.59 0.57 0.43 0.53 14-3-3 protein zeta/delta 0.61 0.96 0.55 0.71 0.4 0.59 0.57 0.52 Caldesmon 0.54 0.71 1.73 0.99 0.43 0.37 0.49 0.43 Tubulin alpha-1A chain 0.53 0.7 1.06 0.76 0.38 0.31 0.52 0.4 Tubulin beta chain 0.42 0.61 0.97 0.67 0.35 0.54 0.25 0.38 Actin, cytoplasmic 1 0.35 0.9 0.94 0.73 0.3 0.21 0.48 0.33 Tubulin beta-1 chain 0.38 0.46 0.89 0.58 0.41 0.23 0.3 0.31 Adenylyl cyclase-associated protein 1 0.43 0.49 0.85 0.59 0.09 0.15 0.57 0.27 Integrin-linked protein kinase 0.17 0.8 1.37 0.78 Plasma membrane T/C 1 T/C 2 T/C 3 AVG T/C Protein name T/TC 1 T/TC 2 T/TC 3 AVG T/TC 16.17 16.58 47.33 26.69 Prothrombin 1.78 1.22 1.57 1.52 1.11 6.03 7.69 4.94 Apolipoprotein A-I 1.42 1 15.49 5.97 2.12 1.43 1.62 1.72 Coagulation factor V 1.71 1.69 1.91 1.77 1.73 1.69 1.08 1.5 Amyloid beta A4 protein 3.41 2.13 1.45 2.33 1.91 1.35 1.03 1.43 Ig gamma-1 chain C region 1.91 1.94 1.34 1.73 1.81 1.13 1.04 1.33 Serotransferrin 2.77 1.58 0.38 1.58 1.65 1.05 0.92 1.21 Platelet glycoprotein V 1.3 1.36 1.25 1.3 1.83 0.77 1.04 1.21 Plasminogen activator inhibitor 1 2.31 0.99 1.69 1.67 1.58 1.02 0.95 1.19 Thrombospondin-1 1.44 1.3 1.29 1.34 1.78 0.63 0.8 1.07 Fibronectin 7.61 0.89 1.11 3.2 68 Plasma membrane continued\u00E2\u0080\u00A6 T/C 1 T/C 2 T/C 3 AVG T/C Protein name T/TC 1 T/TC 2 T/TC 3 AVG T/TC 0.52 1.19 0.94 0.88 Calmodulin 0.62 63.6 4.04 22.75 0.77 0.39 1.41 0.86 Chloride intracellular channel protein 1 0.95 5.12 3.34 3.14 0.6 1.08 0.8 0.83 Pleckstrin 0.66 2.03 1.72 1.47 0.91 0.72 0.82 0.82 Platelet glycoprotein Ib alpha chain 0.46 0.69 0.65 0.6 0.3 0.45 1.71 0.82 Serum deprivation-response protein 0.37 0.6 6.68 2.55 0.28 0.16 1.83 0.76 Integrin alpha-IIb 0.08 0.34 1.23 0.55 0.75 0.55 0.96 0.75 Alpha-enolase 0.8 0.88 1.94 1.21 0.72 0.64 0.76 0.71 Alpha-actinin-1 0.88 1.06 1.18 1.04 0.45 0.42 1.19 0.69 Myosin-9 0.58 0.79 2.23 1.2 0.48 0.63 0.97 0.69 Transgelin-2 0.59 0.97 1.54 1.03 0.54 0.61 0.66 0.6 Filamin-A 0.59 0.98 1.02 0.86 0.59 0.7 0.47 0.59 Apolipoprotein B-100 1.27 1.12 1.35 1.25 0.78 0.6 0.39 0.59 Trem-like transcript 1 protein 0.51 0.56 0.35 0.47 0.35 0.53 0.76 0.55 Zyxin 0.32 0.87 1.4 0.86 0.41 0.38 0.81 0.54 Talin-1 0.48 0.74 1.8 1.01 0.4 0.59 0.57 0.52 Caldesmon 0.54 0.71 1.73 0.99 0.38 0.31 0.52 0.4 Tubulin beta chain 0.42 0.61 0.97 0.67 0.32 0.32 0.32 0.32 Ras-related protein Rap-1b 0.23 0.56 0.54 0.44 0.41 0.23 0.3 0.31 Adenylyl cyclase-associated protein 1 0.43 0.49 0.85 0.59 0.09 0.15 0.57 0.27 Integrin-linked protein kinase 0.17 0.8 1.37 0.78 0.2 0 0.01 0.07 Fibrinogen gamma chain 6.53 0.02 0.26 2.27 0.1 0.02 0.03 0.05 Fibrinogen alpha chain 1.71 0.07 0.47 0.75 0.07 0.01 0.02 0.03 Fibrinogen beta chain 2.1 0.04 0.24 0.8 2.E-04 5.E-05 3.E-04 2.E-04 Collagen alpha-1(I) chain 2.E-04 6.E-05 4.E-04 2.E-04 1.E-04 3.E-03 3.E-03 2.E-03 Collagen alpha-2(I) chain 0.01 0.01 0.01 0.01 69 3.3 Quantitative platelet releasate analysis using LPA, collagen and subthreshold collagen plus LPA as agonists Before starting proteomics studies on LPA activated platelets, we used aggregometry to make sure that the platelets are responsive and are activated by LPA (Figure 3.6). Equal amount of washed platelet were activated with the different agonists for 5 minutes with stirring on the aggregometer. LPA response was compared to thrombin, collagen and thrombin plus collagen activation. Collagen seems to cause the most washed platelet aggregation, while the trace generated by adding thrombin plus collagen does not level out after 5 minutes of activation. LPA was added at a final concentration of 1, 5, 10 or 20 \u00C2\u00B5M to test platelet response, and find the optimum concentration of LPA to use for platelet activation. The 10 and 20 \u00C2\u00B5M final LPA concentration generated about the same platelet response with platelet aggregating slightly more with 20 \u00C2\u00B5M of LPA. Based on these studies, 20 \u00C2\u00B5M concentration of LPA was used for all subsequent proteomic studies. LPA, collagen or subthreshold concentration of collagen plus LPA was used to activate equal numbers of washed platelets at physiological concentrations. Subthreshold concentration of collagen was added a minute prior to LPA addition to prime the platelet for activation. The proteins from the Rel and Rel-MP were quantitatively analyzed as before (Figure 2.1). There are 180 proteins in common between the three datasets from Rel samples (Figure 3.7A) and 148 proteins in Rel-MP samples (Figure 3.7B). There are 129 proteins in common between the Rel and Rel-MP datasets (Figure 3.7C). The proteins in common from the Rel and Rel-MP samples were analyzed based on their LPA/collagen (L/C) ratio and subthreshold collagen plus LPA/collagen (CL/C) ratios. Overall, most of the proteins found are more abundant when LPA is used as an agonist (Table 3.6). Again we looked at the proteins which are more abundant when LPA or collagen is used and grouped them based on their biological function to understand the biological consequences that could occur when platelets are activated by these agonists. 70 Figure 3.6: Aggregometry studies on activated platelets Aggregation studies showing washed platelet response to thrombin, collagen, thrombin plus collagen and LPA. Washed platelets were incubated at 37 \u00C2\u00B0C for 5 minutes before addition of agonists. The platelet activation reaction was carried out for 5 minutes for each recorded trace. Different final concentrations of LPA were used to find an optimal concentration for platelet activation. For thrombin, collagen and thrombin plus collagen activated platelets two repeats are shown, in blue and in red. 1 min. 1 min. 1 min. 1 min. 1 min. 10 \u00C2\u00B5M LPA 20 \u00C2\u00B5M LPA 0.19 mg/ml collagen + 1 U/ml thrombin 1 \u00C2\u00B5M LPA 5 \u00C2\u00B5M LPA 1 U/ml thrombin 0.19 mg/ml collagen % c h an ge in li gh t tr an sm is si o n % c h an ge in li gh t tr an sm is si o n % c h an ge in li gh t tr an sm is si o n % c h an ge in li gh t tr an sm is si o n % c h an ge in li gh t tr an sm is si o n 71 Figure 3.7: Overlap of quantified platelet releasate proteins from LPA, collagen and subthreshold collagen plus LPA activated platelets (A) Protein releasate from LPA, collagen or subthreshold collagen plus LPA activated washed platelets were trypsin digested. Peptides from different platelet activated samples were dimethyl labeled and their ratios quantified. The Venn diagrams represent the common proteins quantified according to their LPA/collagen and collagen plus LPA/collagen ratios between three biological repeat of the experiment. C. Releasate versus Rel-MP Releasate all 3 datasets (180 proteins) 7 2 Rel-MP all 3 datasets (148 proteins) 51 129 19 Dataset 1 314 Proteins Dataset 2 287 Proteins Dataset 3 215 Proteins 148 68 57 23 68 14 30 A. Releasate B. Rel-MP Dataset 1 353 Proteins Dataset 2 313 Proteins Dataset 3 268 Proteins 180 97 66 41 48 19 28 72 (B) The releasate free of microparticles (Rel-MP) were also analyzed in the same manner. (C) Venn diagram showing the number of proteins which overlap between the releasate and the Rel-MP datasets. 3.3.1 Analysis on Rel proteins with high relative abundance There were 135 Rel proteins with an average L/C ratio of greater than 1.2 (Table 3.6A) and these were submitted to DAVID software for biological process GO analysis. Figure 3.8A shows the top 20 GO terms generated and the number of proteins from the list that belongs to each term. Some of these terms include cell motion (31 proteins) and cytoskeletal organization (25 proteins). Similarly, the 27 proteins with an average L/C ratio of less than 0.8 (Table 3.6A) were submitted to DAVID for the same analysis (Figure 3.8B). Some of the terms include response to wounding (17 proteins) wound healing (12 proteins) and coagulation (10 proteins). The same analysis was done with the 133 proteins with an average CL/C ratio greater than 1.2 (Figure 3.9A), and the top GO terms were for the most part the same as Figure 3.8A, which shows the GO terms for proteins with L/C ratio greater than 1.2. Figure 3.9B shows the top 20 GO terms associated with the 40 proteins with an average CL/C ratio less than 0.8. The top GO terms such as response to wounding (23 proteins) and coagulation (13 proteins) are similar to that of Figure 3.8B. To study the effect of activating platelets with subthreshold collagen plus LPA as compared to collagen and LPA alone we compared how the L/C and CL/C for each protein compares. Table 3.7 shows the effect of adding subthreshold collagen to platelets before LPA activation. There were 32 proteins which were secreted more by using subthreshold collagen plus LPA as compared to LPA alone. On the other hand, there were 87 proteins which were secreted less due to addition of subthreshold collagen. Again, we looked at the Rel-MP dataset to find out the extent of MP contribution to the total releasate. MP play a crucial role in health and disease state and knowing their composition could help design drug targets to prevent a wide array of diseases. 73 Table 3.6: Change in protein abundance when using different agonists (A) Proteins with average LPA/collagen (L/C) ratios greater than 1.2 represent proteins which are significantly more abundant in releasate (Rel) from LPA activated platelets as compared to collagen. Proteins with L/C ratios lower than 0.8 are the proteins which are significantly more abundant in the releasate from collagen activated platelets. The same logic applies when comparing collagen plus LPA/ collagen (CL/C) ratios. (B) Here the ratios of Rel-MP (the soluble fraction of platelet releasate free of microparticles) are analyzed in the same manner. (C) This table represent the number of proteins which can potentially be abundant in MP. If Rel L/C or CL/C ratio is significantly (20 %) greater than the Rel-MP L/C or CL/C ratio then the proteins could come from MP generated through LPA platelet activation. If Rel L/C ratio is significantly (20 %) less than the Rel-MP L/C ratio then the proteins could originate from MP generated through collagen platelet activation. A. Releasate # of proteins with avg. L/C greater than 1.2 135 # of proteins with avg. L/C less than 0.8 27 # of proteins with avg. CL/C greater than 1.2 133 # of proteins with avg. CL/C less than 0.8 40 B. Rel-MP # of proteins with avg. L/C greater than 1.2 109 # of proteins with avg. L/C less than 0.8 25 # of proteins with avg. CL/C greater than 1.2 108 # of proteins with avg. CL/C less than 0.8 33 C. Common proteins btw Rel and Rel-MP # of proteins with avg. Rel L/C 20 % or more greater than avg. Rel-MP L/C 9 # of proteins with avg. Rel L/C 20 % or more less than avg. Rel-MP L/C 89 # of proteins with avg. Rel CL/C 20 % or more greater than avg. Rel-MP CL/C 9 # of proteins with avg. Rel CL/C 20 % or more less than avg. Rel-MP CL/C 81 74 Table 3.7: Effect of adding subthreshold collagen to LPA There are 32 proteins which have an average LPA/collagen (L/C) ratio significantly less than subthreshold collagen plus LPA/collagen (CL/C) ratio, compared to 87 proteins which have an average L/C ratio greater than CL/C in Rel sample. Overall, the amount of proteins released decreases by priming platelets by subthreshold collagen before LPA activation, although the amount of LPA needed for secretion might decrease. A. Effect of adding subthreshold collagen plus LPA Protein category Number of proteins AVG L/C ratio over 20 % less than CL/C 32 AVG L/C ratio over 20 % greater than CL/C 87 75 0 5 10 15 20 25 30 35 # o f p ro te in s Biological process GO terms A. LPA overexpressed proteins 76 Figure 3.8: Quantified releasate proteins from LPA versus collagen samples The platelet releasate quantified proteins in common between three datasets from LPA and collagen activated platelets were subject to gene ontology (GO) analysis on DAVIS program. (A) The top 20 biological process associated with proteins which are more abundant in the releasate from LPA activated platelets as compared to collagen. (B) The top 20 biological process associated with proteins which are more abundant in the releasate from collagen activated platelets as compared to LPA. 0 2 4 6 8 10 12 14 16 18 # o f p ro te in s Biological process GO terms B. Collagen overexpressed proteins 77 Figure 3.9: Releasate proteins from subthreshold collagen plus LPA versus collagen samples (A) The top 20 biological process, as found by DAVID software, associated with proteins which are more abundant in the releasate from collagen plus LPA activated platelets as compared to collagen. (B) The top 20 biological process associated with proteins which are more abundant in the releasate from collagen activated platelets as compared to collagen plus LPA. 0 5 10 15 20 25 30 35 # o f p ro te in s Biological process GO terms A. Subthreshold collagen+LPA overexpressed proteins 0 5 10 15 20 25 # o f p ro te in s Biological process GO terms B. Collagen overexpressed proteins 78 3.3.2 Analysis on Rel-MP proteins with high relative abundance There were a lot more proteins with average L/C ratios and CL/C ratio of greater than 1.2, 109 and 108 respectively, in the Rel-MP samples (Table 3.6B). In comparison, there were 25 proteins with an average L/C ratio less than 0.8, and 33 proteins with an average CL/C ratio less than 0.8 in the Rel-MP sample. The same biological function GO analysis was done on these proteins as the ones from Rel sample, and the GO terms and the number of proteins for each term looked very similar to their corresponding Rel dataset graphs (data not shown). By comparing the Rel and Rel-MP datasets we can learn which proteins can potentially come from MP when LPA is used as an agonist. 3.3.3 Proteins potentially abundant in MP From the 129 proteins in common between the Rel and Rel-MP datasets, there were 9 proteins which had significantly greater average L/C ratio in Rel samples as compared to Rel-MP (Table 3.6C). These proteins include P-selectin, complement factor H and PF4 (Appendix F). Similarly there were 9 proteins which had significantly greater average CL/C ratio in Rel samples as compared to Rel-MP. On the other hand there were 89 proteins for which the L/C ratio decreases significantly going from Rel to Rel-MP samples, and 81 proteins for which CL/C ratio decreases (Table 3.6C). Most of these proteins had ratios greater than 1.2 in the Rel sample and they remained greater than 1.2 in the Rel-MP sample. There were 51 proteins unique to Rel dataset (Appendix G) in comparison to only 19 unique proteins in Rel-MP dataset. Most of the proteins which were unique to Rel dataset were more abundant in the releasate from LPA activated platelets. However, proteins such as integrin \u00CE\u00B16, \u00CE\u00B23 and \u00CE\u00B1IIb and C-C motif chemokine 5 were more abundant in Rel from collagen activated platelets. Again we analyzed all the proteins released so we can have a better understanding of the proteome profile when LPA is used an agonist. 79 3.3.4 GO analysis on all Rel proteins Similar to section 3.2.4, all the proteins in common between the Rel datasets were submitted to DAVID for biological process and cellular compartment GO analysis. The top biological process for the Rel samples includes response to wounding, cell adhesion, immune and inflammatory response (Table 3.8). Here we analyze the same GO terms as before, to highlight the differences when LPA is used as an agonist in comparison to thrombin. For response to wounding, there were 16 proteins with an average L/C ratio of greater than 1.2 and 17 proteins (including collagen) with L/C ratio of less than 0.8. All the integrin proteins, nidogen-1 and C-C motif chemokine 5 have L/C ratio of less than 0.8 as can be seen in the cell adhesion category. Some of the proteins important for immune and inflammatory response such as TGF\u00CE\u00B21 had L/C ratio of less than 0.8 while others like PF4 have an average L/C ratio of greater than 1.2. Next we looked where from the platelets, these proteins might originate from. The top cellular compartment GO terms for the Rel dataset include platelet \u00CE\u00B1-granule, cytoskeleton and plasma membrane (Table 3.9). There were 28 proteins which had an average L/C ratio of greater than 1.2, 10 proteins with L/C ratio of around 1 and 14 proteins with L/C ratio of less than 0.8 for platelet \u00CE\u00B1-granules. Of the 57 proteins associated with the cytoskeleton, 55 of them had an average L/C ratio greater than 1.2. There were 41 proteins associated with the plasma membrane with most having L/C ratios of greater than 1.2. Some key proteins that help in coagulation and platelet function, such as integrins and fibrinogen, had an average L/C and CL/C ratio of less than 0.8. Next, to confirm some of the results found from these proteomics studies, we used flow cytometry to see the differences in platelet activation that the various agonists cause. 80 Table 3.8: Releasate proteins and their ratios grouped according to biological process gene ontology (GO) terms All quantified releasate proteins from LPA, collagen and subthreshold collagen plus LPA activated platelets in common between the three different datasets were searched against biological process GO terms using DAVID bioinformatics software. The proteins were grouped according to their involvement in the different GO terms such as wound healing, and immune response. The red color indicates a ratio of more than 1.2, i.e. in L/C ratios this means proteins in the releasate from LPA activated platelets are more abundant than collagen activated platelets. The green color indicates a ratio of less than 0.8. Abbreviations: L/C, LPA/collagen; CL/C, subthreshold collagen plus LPA/collagen. Response to wounding/hemostasis/wound healing/blood coagulation/coagulation L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 2.40 2.81 38.59 14.60 Ras-related C3 botulinum toxin substrate 1 1.83 2.21 31.42 11.82 12.25 7.10 11.42 10.26 Peroxiredoxin-5, mitochondrial 10.22 6.00 9.35 8.52 5.62 6.55 16.46 9.54 Tyrosine-protein phosphatase non-receptor type 6 6.66 4.83 12.14 7.87 3.11 3.03 12.93 6.36 Glucose-6-phosphate isomerase 1.89 2.53 5.45 3.29 4.88 4.40 5.60 4.96 Myotrophin 5.78 3.96 4.73 4.82 4.20 4.93 5.38 4.84 Pleckstrin 2.84 13.98 4.67 7.17 3.75 5.14 3.89 4.26 14-3-3 protein zeta/delta 2.60 2.17 3.59 2.78 3.49 3.89 3.80 3.73 Glutathione peroxidase 1 9.81 8.09 4.46 7.46 3.81 3.51 3.65 3.66 Coagulation factor XIII A chain 1.95 4.03 6.33 4.10 1.04 8.51 0.73 3.43 Inter-alpha-trypsin inhibitor heavy chain H4 0.72 80.52 0.30 27.18 0.80 0.52 4.65 1.99 P-selectin 0.54 0.18 1.44 0.72 2.16 1.73 2.04 1.98 Gelsolin 1.88 2.03 1.62 1.84 1.84 1.78 1.59 1.74 Platelet glycoprotein Ib alpha chain 1.14 2.49 1.21 1.61 0.97 2.64 1.52 1.71 Complement factor H 0.50 0.84 0.95 0.76 2.11 1.16 1.47 1.58 Platelet factor 4 1.48 1.22 1.27 1.32 1.74 0.97 1.17 1.29 Thrombospondin-1 1.48 1.42 0.88 1.26 0.88 1.01 0.87 0.92 Plasminogen 0.35 0.84 0.74 0.64 0.74 0.70 1.19 0.88 Trem-like transcript 1 protein 0.91 0.60 1.26 0.92 0.87 0.89 0.86 0.87 Platelet glycoprotein V 0.52 0.62 0.74 0.62 1.11 0.71 0.75 0.86 Serotransferrin 0.49 0.39 0.65 0.51 0.86 0.72 0.94 0.84 Vitamin K-dependent protein S 0.42 0.35 0.68 0.48 0.90 0.69 0.70 0.76 Complement C3 0.42 0.58 0.55 0.52 81 Response to wounding continued\u00E2\u0080\u00A6 L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 1.03 0.67 0.58 0.76 Alpha-1-antitrypsin 0.33 0.37 0.56 0.42 0.74 0.71 0.83 0.76 Plasminogen activator inhibitor 1 0.32 0.52 0.85 0.56 0.87 0.69 0.61 0.72 von Willebrand factor 0.48 0.80 0.82 0.70 0.54 0.49 0.83 0.62 Clusterin 0.32 0.29 0.76 0.46 0.54 0.58 0.48 0.53 Alpha-1-antichymotrypsin 0.33 0.38 1.04 0.59 0.58 0.40 0.60 0.53 Coagulation factor V 0.37 0.20 0.54 0.37 0.59 0.48 0.48 0.52 Transforming growth factor beta-1 0.44 0.32 0.59 0.45 0.50 0.73 0.22 0.48 Integrin beta-3 0.19 0.91 0.39 0.50 0.64 0.25 0.54 0.48 Multimerin-1 0.44 0.25 0.74 0.48 0.41 0.40 0.18 0.33 Fibrinogen gamma chain 0.23 0.18 0.32 0.24 0.33 0.39 0.21 0.31 Fibrinogen beta chain 0.37 0.23 0.38 0.33 0.32 0.32 0.20 0.28 Fibronectin 0.35 0.35 0.59 0.43 0.27 0.30 0.21 0.26 Fibrinogen alpha chain 0.33 0.20 0.43 0.32 0.40 0.16 0.10 0.22 C-C motif chemokine 5 0.23 0.16 0.00 0.13 0.15 0.12 0.17 0.15 Erythrocyte band 7 integral membrane protein 0.13 0.50 0.30 0.31 5.E-04 7.E-05 1.E-04 2.E-04 Collagen alpha-1(III) chain 0.02 0.01 0.01 0.01 Cell adhesion L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 4.15 61.52 5.26 23.64 Nucleoside diphosphate kinase A 4.43 316.90 3.68 108.33 2.40 2.81 38.59 14.60 Ras-related C3 botulinum toxin substrate 1 1.83 2.21 31.42 11.82 10.74 5.30 4.91 6.98 Beta-parvin 1.95 7.35 4.99 4.76 6.00 6.25 7.26 6.50 Zyxin 5.46 6.33 6.99 6.26 5.67 5.98 5.71 5.79 Coronin-1A 3.88 5.34 5.00 4.74 4.20 4.93 5.38 4.84 Pleckstrin 2.84 13.98 4.67 7.17 4.23 3.66 4.04 3.98 Alpha-actinin-1 3.34 3.75 4.16 3.75 3.81 3.48 4.43 3.90 Vinculin 3.92 3.34 3.88 3.71 2.96 3.79 4.88 3.88 Integrin-linked protein kinase 1.54 4.61 5.25 3.80 2.47 2.57 4.97 3.34 Proto-oncogene tyrosine-protein kinase Src 1.06 4.21 4.12 3.13 2.78 3.29 3.57 3.21 Fermitin family homolog 3 2.12 5.13 3.90 3.72 3.09 0.95 3.87 2.64 Moesin 3.13 1.27 3.08 2.49 2.94 1.78 2.27 2.33 Myosin-9 1.93 2.43 2.27 2.21 0.80 0.52 4.65 1.99 P-selectin 0.54 0.18 1.44 0.72 1.84 1.78 1.59 1.74 Platelet glycoprotein Ib alpha chain 1.14 2.49 1.21 1.61 1.58 1.12 1.52 1.41 Amyloid beta A4 protein 1.24 0.93 1.35 1.17 1.74 0.97 1.17 1.29 Thrombospondin-1 1.48 1.42 0.88 1.26 0.87 0.89 0.86 0.87 Platelet glycoprotein V 0.52 0.62 0.74 0.62 0.87 0.69 0.61 0.72 von Willebrand factor 0.48 0.80 0.82 0.70 0.70 0.50 0.50 0.57 Nidogen-1 0.35 0.34 0.88 0.52 82 Cell adhesion continued\u00E2\u0080\u00A6 L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 0.58 0.40 0.60 0.53 Coagulation factor V 0.37 0.20 0.54 0.37 0.53 0.42 0.59 0.51 Integrin alpha-6 0.35 0.66 0.52 0.51 0.50 0.73 0.22 0.48 Integrin beta-3 0.19 0.91 0.39 0.50 0.64 0.25 0.54 0.48 Multimerin-1 0.44 0.25 0.74 0.48 0.31 0.36 0.76 0.48 Integrin alpha-IIb 0.34 0.72 0.52 0.53 0.32 0.32 0.20 0.28 Fibronectin 0.35 0.35 0.59 0.43 0.40 0.16 0.10 0.22 C-C motif chemokine 5 0.23 0.16 0.00 0.13 5.E-04 7.E-05 1.E-04 2.E-04 Collagen alpha-1(III) chain 0.02 0.01 0.01 0.01 Immune response L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 11.13 1.26 27.33 13.24 Alpha-synuclein 8.09 0.03 20.55 9.56 4.29 6.69 22.41 11.13 Drebrin-like protein 4.53 9.23 18.35 10.70 3.11 3.03 12.93 6.36 Glucose-6-phosphate isomerase 1.89 2.53 5.45 3.29 5.67 5.98 5.71 5.79 Coronin-1A 3.88 5.34 5.00 4.74 3.57 5.37 6.25 5.06 Rho GDP-dissociation inhibitor 2 4.11 4.38 4.83 4.44 3.75 5.14 3.89 4.26 14-3-3 protein zeta/delta 2.60 2.17 3.59 2.78 3.44 2.51 6.75 4.23 Peroxiredoxin-1 2.77 2.59 4.66 3.34 0.53 2.96 8.31 3.93 Purine nucleoside phosphorylase 0.14 2.34 7.58 3.35 1.24 1.88 2.87 2.00 Tubulin beta-2C chain 0.95 1.64 3.01 1.87 1.31 1.94 2.70 1.98 Tubulin beta chain 1.13 1.70 3.04 1.96 0.97 2.64 1.52 1.71 Complement factor H 0.50 0.84 0.95 0.76 2.11 1.16 1.47 1.58 Platelet factor 4 1.48 1.22 1.27 1.32 1.74 0.97 1.17 1.29 Thrombospondin-1 1.48 1.42 0.88 1.26 0.96 0.82 1.15 0.98 Ig kappa chain C region 0.72 0.73 0.98 0.81 0.92 0.90 1.00 0.94 Ig lambda chain C regions 0.61 0.79 0.77 0.73 0.93 0.82 0.97 0.91 Beta-2-microglobulin 0.83 0.64 0.75 0.74 1.01 0.78 0.84 0.88 Ig gamma-1 chain C region 0.56 0.73 0.74 0.68 0.74 0.70 1.19 0.88 Trem-like transcript 1 protein 0.91 0.60 1.26 0.92 0.92 0.77 0.88 0.86 Platelet basic protein 0.40 0.78 0.85 0.68 0.95 0.59 1.01 0.85 Ig gamma-2 chain C region 0.48 0.49 0.82 0.59 0.90 0.69 0.70 0.76 Complement C3 0.42 0.58 0.55 0.52 0.54 0.49 0.83 0.62 Clusterin 0.32 0.29 0.76 0.46 0.59 0.48 0.48 0.52 Transforming growth factor beta-1 0.44 0.32 0.59 0.45 0.40 0.16 0.10 0.22 C-C motif chemokine 5 0.23 0.16 0.00 0.13 83 Inflammatory response L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 2.40 2.81 38.59 14.60 Ras-related C3 botulinum toxin substrate 1 1.83 2.21 31.42 11.82 12.25 7.10 11.42 10.26 Peroxiredoxin-5, mitochondrial 10.22 6.00 9.35 8.52 3.75 5.14 3.89 4.26 14-3-3 protein zeta/delta 2.60 2.17 3.59 2.78 1.04 8.51 0.73 3.43 Inter-alpha-trypsin inhibitor heavy chain H4 0.72 80.52 0.30 27.18 0.80 0.52 4.65 1.99 P-selectin 0.54 0.18 1.44 0.72 0.97 2.64 1.52 1.71 Complement factor H 0.50 0.84 0.95 0.76 1.74 0.97 1.17 1.29 Thrombospondin-1 1.48 1.42 0.88 1.26 1.11 0.71 0.75 0.86 Serotransferrin 0.49 0.39 0.65 0.51 0.90 0.69 0.70 0.76 Complement C3 0.42 0.58 0.55 0.52 1.03 0.67 0.58 0.76 Alpha-1-antitrypsin 0.33 0.37 0.56 0.42 0.54 0.49 0.83 0.62 Clusterin 0.32 0.29 0.76 0.46 0.54 0.58 0.48 0.53 Alpha-1-antichymotrypsin 0.33 0.38 1.04 0.59 0.59 0.48 0.48 0.52 Transforming growth factor beta-1 0.44 0.32 0.59 0.45 0.32 0.32 0.20 0.28 Fibronectin 0.35 0.35 0.59 0.43 0.40 0.16 0.10 0.22 C-C motif chemokine 5 0.23 0.16 0.00 0.13 Table 3.9: Releasate proteins grouped according to cellular compartment gene ontology (GO) terms The proteins quantified in releasate datasets were grouped according to their cellular compartment such as belonging to \u00C9\u0091-granules or plasma membrane. The red color indicates a ratio of more than 1.2. The green color indicates a ratio of less than 0.8. Abbreviations: L/C, LPA/collagen; CL/C, subthreshold collagen plus LPA/collagen. Platelet alpha granule L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 2.40 2.81 38.59 14.60 Ras-related C3 botulinum toxin substrate 1 1.83 2.21 31.42 11.82 4.07 7.04 16.01 9.04 EH domain-containing protein 3 3.21 11.02 14.23 9.49 6.19 6.94 9.94 7.69 Peroxiredoxin-6 5.13 7.74 10.76 7.87 6.46 5.54 9.70 7.23 Protein disulfide-isomerase A6 6.66 5.79 3.81 5.42 5.96 5.83 7.99 6.59 Protein disulfide-isomerase 16.99 8.92 6.06 10.66 5.67 5.98 5.71 5.79 Coronin-1A 3.88 5.34 5.00 4.74 8.94 4.37 3.75 5.69 Peptidyl-prolyl cis-trans isomerase B 2.39 5.52 2.77 3.56 4.01 4.62 8.31 5.65 Ras-related protein Rab-7a 3.04 4.87 6.91 4.94 5.29 4.85 5.23 5.12 Heat shock cognate 71 kDa protein 3.91 7.37 4.66 5.32 4.34 4.08 6.95 5.12 Protein disulfide-isomerase A3 4.06 4.27 5.71 4.68 84 Platelet alpha granule continued\u00E2\u0080\u00A6 L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 3.57 5.37 6.25 5.06 Rho GDP-dissociation inhibitor 2 4.11 4.38 4.83 4.44 5.08 5.37 4.62 5.02 Alpha-actinin-4 2.75 3.86 5.12 3.91 4.90 4.40 5.48 4.93 78 kDa glucose-regulated protein 3.72 5.57 4.08 4.46 5.86 4.83 2.93 4.54 14-3-3 protein epsilon 2.22 1.77 2.54 2.18 3.75 5.14 3.89 4.26 14-3-3 protein zeta/delta 2.60 2.17 3.59 2.78 3.44 2.51 6.75 4.23 Peroxiredoxin-1 2.77 2.59 4.66 3.34 5.23 3.60 3.64 4.16 Endoplasmin 3.46 5.28 3.44 4.06 4.23 3.66 4.04 3.98 Alpha-actinin-1 3.34 3.75 4.16 3.75 3.71 1.E-03 7.60 3.77 Chloride intracellular channel protein 4 3.08 0.00 7.22 3.43 3.81 3.51 3.65 3.66 Coagulation factor XIII A chain 1.95 4.03 6.33 4.10 3.37 3.20 4.10 3.56 Fructose-bisphosphate aldolase A 16.53 3.27 3.66 7.82 1.43 3.85 3.61 2.96 EH domain-containing protein 1 1.89 5.32 4.73 3.98 2.59 0.61 3.40 2.20 Heat shock protein HSP 90-alpha 1.78 0.72 4.86 2.45 0.80 0.52 4.65 1.99 P-selectin 0.54 0.18 1.44 0.72 1.84 1.78 1.59 1.74 Platelet glycoprotein Ib alpha chain 1.14 2.49 1.21 1.61 2.11 1.16 1.47 1.58 Platelet factor 4 1.48 1.22 1.27 1.32 1.58 1.12 1.52 1.41 Amyloid beta A4 protein 1.24 0.93 1.35 1.17 1.74 0.97 1.17 1.29 Thrombospondin-1 1.48 1.42 0.88 1.26 1.12 1.19 1.10 1.14 Serglycin 1.04 1.08 0.96 1.03 1.17 0.97 1.08 1.07 Serum albumin 0.63 0.94 0.87 0.81 1.47 0.60 1.01 1.03 SPARC 0.69 0.59 0.88 0.72 0.94 0.82 1.27 1.01 Metalloproteinase inhibitor 1 0.33 0.64 19.43 6.80 1.03 0.79 1.14 0.99 Calumenin 0.61 0.54 1.49 0.88 0.74 0.70 1.19 0.88 Trem-like transcript 1 protein 0.91 0.60 1.26 0.92 0.87 0.89 0.86 0.87 Platelet glycoprotein V 0.52 0.62 0.74 0.62 1.11 0.71 0.75 0.86 Serotransferrin 0.49 0.39 0.65 0.51 0.92 0.77 0.88 0.86 Platelet basic protein 0.40 0.78 0.85 0.68 0.86 0.72 0.94 0.84 Vitamin K-dependent protein S 0.42 0.35 0.68 0.48 1.03 0.67 0.58 0.76 Alpha-1-antitrypsin 0.33 0.37 0.56 0.42 0.87 0.69 0.61 0.72 von Willebrand factor 0.48 0.80 0.82 0.70 0.54 0.49 0.83 0.62 Clusterin 0.32 0.29 0.76 0.46 0.58 0.40 0.60 0.53 Coagulation factor V 0.37 0.20 0.54 0.37 0.59 0.48 0.48 0.52 Transforming growth factor beta-1 0.44 0.32 0.59 0.45 0.50 0.73 0.22 0.48 Integrin beta-3 0.19 0.91 0.39 0.50 0.64 0.25 0.54 0.48 Multimerin-1 0.44 0.25 0.74 0.48 0.31 0.36 0.76 0.48 Integrin alpha-IIb 0.34 0.72 0.52 0.53 0.06 1.01 0.04 0.37 WAS/WASL-interacting protein family member 1 0.05 0.88 0.03 0.32 0.41 0.40 0.18 0.33 Fibrinogen gamma chain 0.23 0.18 0.32 0.24 85 Platelet alpha granule continued\u00E2\u0080\u00A6 L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 0.33 0.39 0.21 0.31 Fibrinogen beta chain 0.37 0.23 0.38 0.33 0.32 0.32 0.20 0.28 Fibronectin 0.35 0.35 0.59 0.43 0.27 0.30 0.21 0.26 Fibrinogen alpha chain 0.33 0.20 0.43 0.32 0.15 0.12 0.17 0.15 Erythrocyte band 7 integral membrane protein 0.13 0.50 0.30 0.31 Cytoskeleton L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 24.21 12.14 101.52 45.96 PDZ and LIM domain protein 7 16.89 10.21 87.54 38.21 79.39 21.45 34.12 44.99 Nexilin 15.70 22.72 27.16 21.86 4.15 61.52 5.26 23.64 Nucleoside diphosphate kinase A 4.43 316.90 3.68 108.33 5.37 10.07 28.42 14.62 Cytoskeleton-associated protein 5 5.73 5.80 16.18 9.24 8.34 24.65 7.80 13.60 Src substrate cortactin 12.67 20.92 17.00 16.86 14.62 12.48 13.52 13.54 Calponin-2 11.33 12.09 13.41 12.28 11.13 1.26 27.33 13.24 Alpha-synuclein 8.09 0.03 20.55 9.56 4.29 6.69 22.41 11.13 Drebrin-like protein 4.53 9.23 18.35 10.70 13.36 11.91 7.41 10.89 Coronin-1C 4.67 12.56 6.11 7.78 7.48 7.28 17.30 10.69 Caldesmon 4.90 8.28 15.06 9.41 3.57 3.24 22.72 9.85 Septin-2 2.58 1.31 23.15 9.01 6.95 6.83 15.00 9.59 Vasodilator-stimulated phosphoprotein 7.13 9.46 12.12 9.57 8.40 13.12 6.48 9.33 Na(+)/H(+) exchange regulatory cofactor NHE-RF1 5.33 16.30 6.21 9.28 5.06 6.92 13.69 8.56 PDZ and LIM domain protein 1 4.82 7.49 7.74 6.68 9.78 4.85 9.30 7.98 Dihydropyrimidinase-related protein 2 3.66 5.07 4.53 4.42 5.80 4.99 12.00 7.59 LIM and SH3 domain protein 1 5.64 5.28 14.37 8.43 10.74 5.30 4.91 6.98 Beta-parvin 1.95 7.35 4.99 4.76 6.00 6.25 7.26 6.50 Zyxin 5.46 6.33 6.99 6.26 7.30 5.06 6.76 6.38 Tropomodulin-3 7.28 5.38 4.87 5.84 5.67 5.98 5.71 5.79 Coronin-1A 3.88 5.34 5.00 4.74 9.02 2.83 5.50 5.78 Myosin regulatory light polypeptide 9 5.57 4.31 5.52 5.13 5.50 5.33 5.60 5.48 Tropomyosin beta chain 18.98 3.42 2.34 8.25 3.57 5.37 6.25 5.06 Rho GDP-dissociation inhibitor 2 4.11 4.38 4.83 4.44 5.08 5.37 4.62 5.02 Alpha-actinin-4 2.75 3.86 5.12 3.91 4.19 5.77 4.83 4.93 Tropomyosin alpha-3 chain 24.67 4.15 4.35 11.06 6.48 2.78 5.04 4.77 Myosin light polypeptide 6 17.45 5.59 4.17 9.07 4.39 4.04 5.65 4.69 Coactosin-like protein 4.15 4.67 4.68 4.50 4.22 4.14 5.71 4.69 Cofilin-1 1.35 4.03 4.47 3.29 3.85 5.54 4.55 4.65 Adenylyl cyclase-associated protein 1 2.02 7.88 3.62 4.51 0.51 6.27 6.90 4.56 F-actin-capping protein subunit beta 2.81 7.30 5.38 5.16 5.25 3.35 5.07 4.56 Tropomyosin alpha-4 chain 8.52 3.21 4.81 5.51 86 Cytoskeleton continued\u00E2\u0080\u00A6 L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 3.71 4.18 5.69 4.53 Talin-1 3.65 5.23 5.76 4.88 3.95 3.82 5.17 4.31 Filamin-A 3.36 4.63 4.77 4.26 3.75 5.14 3.89 4.26 14-3-3 protein zeta/delta 2.60 2.17 3.59 2.78 3.41 4.28 5.03 4.24 Profilin-1 3.63 6.48 4.20 4.77 5.20 3.12 3.64 3.99 WD repeat-containing protein 1 2.48 3.94 4.20 3.54 4.23 3.66 4.04 3.98 Alpha-actinin-1 3.34 3.75 4.16 3.75 3.35 3.23 5.35 3.98 F-actin-capping protein subunit alpha-1 1.88 4.48 4.46 3.61 0.53 2.96 8.31 3.93 Purine nucleoside phosphorylase 0.14 2.34 7.58 3.35 3.81 3.48 4.43 3.90 Vinculin 3.92 3.34 3.88 3.71 2.96 3.79 4.88 3.88 Integrin-linked protein kinase 1.54 4.61 5.25 3.80 3.71 0.00 7.60 3.77 Chloride intracellular channel protein 4 3.08 0.00 7.22 3.43 3.37 3.20 4.10 3.56 Fructose-bisphosphate aldolase A 16.53 3.27 3.66 7.82 2.78 3.29 3.57 3.21 Fermitin family homolog 3 2.12 5.13 3.90 3.72 1.77 2.76 3.39 2.64 Actin, cytoplasmic 1 1.85 2.76 2.92 2.51 3.09 0.95 3.87 2.64 Moesin 3.13 1.27 3.08 2.49 2.94 1.78 2.27 2.33 Myosin-9 1.93 2.43 2.27 2.21 2.33 1.22 2.85 2.13 Actin-related protein 2 1.10 1.41 2.01 1.51 1.77 1.91 2.60 2.09 Tubulin alpha-4A chain 1.44 1.75 2.93 2.04 1.24 1.88 2.87 2.00 Tubulin beta-2C chain 0.95 1.64 3.01 1.87 1.31 1.94 2.70 1.98 Tubulin beta chain 1.13 1.70 3.04 1.96 2.16 1.73 2.04 1.98 Gelsolin 1.88 2.03 1.62 1.84 0.66 1.56 2.48 1.57 Tubulin beta-1 chain 0.74 1.44 3.06 1.75 1.33 1.20 1.99 1.51 Actin-related protein 3 1.13 2.96 1.94 2.01 1.58 1.12 1.52 1.41 Amyloid beta A4 protein 1.24 0.93 1.35 1.17 0.06 1.01 0.04 0.37 WAS/WASL-interacting protein family member 1 0.05 0.88 0.03 0.32 0.15 0.12 0.17 0.15 Erythrocyte band 7 integral membrane protein 0.13 0.50 0.30 0.31 Plasma membrane L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 79.39 21.45 34.12 44.99 Nexilin 15.70 22.72 27.16 21.86 2.40 2.81 38.59 14.60 Ras-related C3 botulinum toxin substrate 1 1.83 2.21 31.42 11.82 14.62 12.48 13.52 13.54 Calponin-2 11.33 12.09 13.41 12.28 6.95 6.83 15.00 9.59 Vasodilator-stimulated phosphoprotein 7.13 9.46 12.12 9.57 8.40 13.12 6.48 9.33 Na(+)/H(+) exchange regulatory cofactor NHE-RF1 5.33 16.30 6.21 9.28 4.07 7.04 16.01 9.04 EH domain-containing protein 3 3.21 11.02 14.23 9.49 5.80 4.99 12.00 7.59 LIM and SH3 domain protein 1 5.64 5.28 14.37 8.43 10.74 5.30 4.91 6.98 Beta-parvin 1.95 7.35 4.99 4.76 6.00 6.25 7.26 6.50 Zyxin 5.46 6.33 6.99 6.26 87 Plasma membrane continued\u00E2\u0080\u00A6 L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 4.32 7.88 5.63 5.94 Calreticulin 3.61 8.19 4.30 5.36 5.67 5.98 5.71 5.79 Coronin-1A 3.88 5.34 5.00 4.74 4.20 4.93 5.38 4.84 Pleckstrin 2.84 13.98 4.67 7.17 3.71 4.18 5.69 4.53 Talin-1 3.65 5.23 5.76 4.88 3.60 3.68 5.56 4.28 Carbonic anhydrase 2 2.97 2.92 4.34 3.41 4.23 3.66 4.04 3.98 Alpha-actinin-1 3.34 3.75 4.16 3.75 4.01 3.41 4.36 3.93 Serum deprivation-response protein 4.02 3.54 2.33 3.30 3.81 3.48 4.43 3.90 Vinculin 3.92 3.34 3.88 3.71 2.96 3.79 4.88 3.88 Integrin-linked protein kinase 1.54 4.61 5.25 3.80 2.47 2.57 4.97 3.34 Proto-oncogene tyrosine-protein kinase Src 1.06 4.21 4.12 3.13 2.78 3.29 3.57 3.21 Fermitin family homolog 3 2.12 5.13 3.90 3.72 1.43 3.85 3.61 2.96 EH domain-containing protein 1 1.89 5.32 4.73 3.98 3.09 0.95 3.87 2.64 Moesin 3.13 1.27 3.08 2.49 2.94 1.78 2.27 2.33 Myosin-9 1.93 2.43 2.27 2.21 0.80 0.52 4.65 1.99 P-selectin 0.54 0.18 1.44 0.72 1.31 1.94 2.70 1.98 Tubulin beta chain 1.13 1.70 3.04 1.96 1.84 1.78 1.59 1.74 Platelet glycoprotein Ib alpha chain 1.14 2.49 1.21 1.61 1.33 1.20 1.99 1.51 Actin-related protein 3 1.13 2.96 1.94 2.01 1.58 1.12 1.52 1.41 Amyloid beta A4 protein 1.24 0.93 1.35 1.17 1.74 0.97 1.17 1.29 Thrombospondin-1 1.48 1.42 0.88 1.26 0.93 0.82 0.97 0.91 Beta-2-microglobulin 0.83 0.64 0.75 0.74 0.87 0.89 0.86 0.87 Platelet glycoprotein V 0.52 0.62 0.74 0.62 1.11 0.71 0.75 0.86 Serotransferrin 0.49 0.39 0.65 0.51 0.87 0.69 0.61 0.72 von Willebrand factor 0.48 0.80 0.82 0.70 0.53 0.42 0.59 0.51 Integrin alpha-6 0.35 0.66 0.52 0.51 0.50 0.73 0.22 0.48 Integrin beta-3 0.19 0.91 0.39 0.50 0.31 0.36 0.76 0.48 Integrin alpha-IIb 0.34 0.72 0.52 0.53 0.41 0.40 0.18 0.33 Fibrinogen gamma chain 0.23 0.18 0.32 0.24 0.33 0.39 0.21 0.31 Fibrinogen beta chain 0.37 0.23 0.38 0.33 0.32 0.32 0.20 0.28 Fibronectin 0.35 0.35 0.59 0.43 0.27 0.30 0.21 0.26 Fibrinogen alpha chain 0.33 0.20 0.43 0.32 0.15 0.12 0.17 0.15 Erythrocyte band 7 integral membrane protein 0.13 0.50 0.30 0.31 88 3.4 Flow cytometry analysis In addition to quantitative proteomics studies, washed platelets activated with thrombin, collagen, thrombin plus collagen, LPA or subthreshold collagen plus LPA were subject to flow cytometry analysis. These studies highlighted how the size and shape of platelets change by activation and the MP population that is generated. The platelets were stained with CD41a (also known as integrin \u00CE\u00B1IIb, a platelet marker) antibody, PAC-1 (an antibody which recognizes active integrin \u00CE\u00B1IIb\u00CE\u00B23), and P-selectin antibody (to detect the level of platelet activation). Figure 3.10 (A-F) shows the side and forward light scatter for CD41a positive events. As well the figure shows the level of PAC-1 binding in platelets for each agonist used (Figure 3.10G). Based on the side and forward scatter plot for unstimulated platelets (Figure 3.10A) as well as the expected size of platelets, gates for platelets and MP were created. Collagen had a bigger MP population as compared to thrombin but a combination of thrombin plus collagen generated a large unique MP population. When LPA was used as an agonist, distinguishing the platelets and MP population became harder. The platelet population seems to have gotten larger in size as event from the population have moved up and to the right in the side versus forward light scatter plot. The MP population is also larger in size. The majority of thrombin activated platelets bond to PAC-1 (around 90% as calculated with quadrant plots). On the other hand, platelets activated with the other agonist, hardly had any platelets that binds to PAC-1. Even when thrombin plus collagen was used only about 11 % of the gated platelets are PAC-1 positive. These experiments were repeated three times and each time the results looked very similar to the one seen in figure 3.10. 89 A. Unstimulated platelets B. Thrombin activated platelets D. Thrombin + Collagen activated C. Collagen activated platelets F. Subthreshold Collagen + LPA activated platelets E. LPA activated platelets 90 I. Microparticle CD62P surface expression H. Platelet CD62P surface expression G. Platelet active integrin \u00CE\u00B1IIb\u00CE\u00B23 (conformational change) expression 91 Figure 3.10: Flow cytometer studies on activated platelets Washed platelets resuspended at physiological concentrations in HEPES/1.8 mM CaCl2 were activated with thrombin (B), collagen (C), thrombin plus collagen (D), LPA (E) or subthreshold collagen plus LPA (F). As a reference uncultivated platelets treated under same experimental conditions are also shown (A). Platelets were labeled with CD41a antibody (a platelet marker), PAC-1 (an antibody which recognized integrin \u00CE\u00B1IIb\u00CE\u00B23 on activated platelets) and P-selectin (CD62P) antibody. The side light scatter (SSC-A) versus forward light scatter (FSC-A) plots show all the CD41a positive events. These events are gated for platelets and microparticles (MP) based on their size as they appear on the SSC-A versus FSC-A plot. (G) Level of PAC-1 expression on platelets activated with different agonists. (H) Level of P-selectin expression of platelets activated with different agonists. (I) Level of P-selectin expression on MP. The same populations of platelets and MP from Figure 3.10 A-G were analyzed for P-selectin (CD62P) expression (Figure 3.10H and I). While the platelets activated by the different agonist had P- selectin expression, platelets activated by collagen seem to have the least P-selectin expression, followed by thrombin plus collagen activated platelets. LPA and thrombin activated platelets had the most P-selectin expression. On the other hand, MP from LPA activated platelets had the highest P- selectin expression, as compared to thrombin or collagen. These results, confirm some of the results found from the proteomic studies which include the fact that thrombin activated platelets produce the least amount of MP. Moreover, to study the functional consequences caused by the different platelet releasates, and whether the different releasate proteome profiles have biological impacts we performed migration assays using THP-1 cells. 92 3.5 THP-1 Migration towards Rel To study the biological function of Rel, Rel-MP and MP, THP-1 cell migration assays were performed. The Rel, Rel-MP and MP from 1 mL of washed platelets activated with the different agonists were tested for their ability to cause THP-1 cell migration across an 8 \u00C2\u00B5M membrane after incubation at 37 \u00C2\u00B0C for 6 hours (Figure 3.11A). The Rel from thrombin, collagen, thrombin plus collagen, LPA and subthreshold collagen plus LPA activated platelets caused around the same number of THP-1 migration based on three repeats. The Rel-MP also had the same migration effects between the different agonists, but the number of THP-1 cell migration was less than when Rel was used. On the other hand, the MP from collagen or thrombin plus collagen activated platelets caused the most THP-1 migration. When PF4 inhibitory antibody was added to the Rel from thrombin activated platelets THP-1 migration decreased in a dose dependent manner, with 20 \u00C2\u00B5g essentially blocking all migration (Figure 3.11B). Moreover, 25 \u00C2\u00B5g of purified PF4 protein did not cause nearly as much THP-1 migration as Rel sample did. These experiments identified the content of platelet releasate and confirmed their intraction with THP-1 cells through the migration assay experiments. To study the protein changes which take place in THP-1 cells upon stimulation with platelet releasate, we incubated THP-1 cells with platelet releasate from thrombin activated platelets for 6 and 24 hours. We used Rel from thrombin activated platelets since traditionally it is the strong agonist known, and there have been lot of studies done on thrombin activated platelets both in vivo and in vitro. We used THP-1 cells as a model system for monocytes, to perform SILAC quantitative proteomics. We used the results from our experiments and bioinformatics to identify the proteins which significantly change due to Rel addition, in an effort to elucidate the pathways responsible for THP-1 stimulation and platelet-monocyte aggregate formation. However, before starting our quantitative studies, we generated reference datasets to identify the proteins which can be identified in THP-1 cells using proteomics. To get the greatest protein coverage 93 we identified the proteins from whole lysate as well as proteins from the membrane fraction. Membrane proteins are normally harder to isolate and identify due to their hydrophobic regions. To this end, we compared and optimized two protocols for enriching and identify membrane proteins. Figure 3.11: THP-1 cell migration towards platelet releasate (A) Washed platelets were resuspended at physiological concentrations and stimulated with thrombin (Thr), collagen, Thrombin plus collagen (Thr+Coll), LPA or subthreshold collagen plus LPA (Coll+LPA). One 0 50000 100000 150000 200000 250000 300000 Th r C o lla ge n Th r+ C o ll LP A C o ll+ LP A Th r C o lla ge n Th r+ C o ll LP A C o ll+ LP A Th r C o lla ge n Th r+ C o ll LP A C o ll+ LP A M e d ia FB S Rel Rel-MP MP Control # ce lls m ig ra ti n g/ m L A. THP-1 cell migration analysis 0 50000 100000 150000 200000 250000 300000 Thr Thrombin + 2.5 \u00C2\u00B5g ab Thrombin + 5 \u00C2\u00B5g ab Thrombin + 20 \u00C2\u00B5g ab 25 \u00C2\u00B5g PF4 5 \u00C2\u00B5g antibody alone FBS Media # o f ce lls m ig ra ti n g/ m L B. Effect of platelet factor 4 antibody on THP-1 migration 94 mL of total releasate (Rel) and 1 mL of releasate minus microparticles (Rel-MP) were concentrated to 50 \u00C2\u00B5L and added to bottom chamber for THP-1 migration. The microparticles (MP) were resuspended in 50 \u00C2\u00B5L of HEPES buffer after ultracentrifugation and used for migration assay analysis. Here we incubated for 6 hours with three different donors for N=3, using transmigration well with 8 \u00C2\u00B5M pores. (B) Varying amount of platelet factor 4 (PF4) antibody (ab) was added to releasate from thrombin activated platelets. In addition, the migration due to 25 \u00C2\u00B5g of PF4 active protein was recorded as well as migration due to PF4 antibody. The numbers of THP-1 cells migrating towards the different releasates were counted using flow cytometry. RPMI media was used as to see the number of random migration and RPMI media supplemented with fetal bovine serum (FBS) was sued as a positive control. 3.6 Membrane protein enrichment Two protocols, A and B, were implemented for membrane enrichment and cytosolic protein depletion for THP-1 cells. Both protocols used ultracentrifugation methods to isolate membrane proteins. Protocol A used a sucrose bilayer, whereas protocol B used a sucrose gradient. Antibodies against integrin \u00CE\u00B21 (a THP-1 cell membrane protein) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) (a highly abundant THP-1 cell cytosolic protein) were used for immuno blot to compare the amount of protein from each sample based on the protein intensity. Equal amounts of proteins from membrane enriched and whole lysate samples were loaded on the gel for performing immuno blot. Using protocol A for membrane protein enrichment resulted in an enrichment of integrin \u00CE\u00B21 as compared to whole cell lysate (Figure 3.12A). In addition, protocol A resulted in a greater depletion of GAPDH as compared to protocol B (Figure 3.12B) as evident from the intensity of GAPDH band at around 36 KDa. To confirm the reproducibility of the results when using protocol A, another immuno blot with antibodies against integrin \u00CE\u00B21, GAPDH and 14.3.3 (a highly abundant cytosolic protein) was carried out. Based on the intensities on the immuno blot there indeed is an enrichment of membrane proteins while 95 depleting cytosolic proteins, when compared to equal amount of total protein from whole cell lysate sample (Figure 3.13). Protocol A was used for all the subsequent THP-1 cell membrane protein enrichments to generate reference datasets and for quantitative proteomics studies. Figure 3.12: THP-1 membrane protein enrichment Two protocol A, and protocol B were tested for their ability to enrich THP-1 cell membrane proteins. Protocol A and two sucrose gradients (60% and 5%) and membrane proteins were collected from the boundary of these two layers after ultracentrifugation, while protocol B used a linear sucrose gradient for isolation of membrane proteins after ultracentrifugation. Antibodies against integrin \u00CE\u00B21 (A), a membrane protein, and GAPDH (B), a highly abundant protein found in the cytosol of THP-1 cells, were used to show the level of enrichment in membrane proteins and the depletion in cytosol proteins for each protocol. The isolated membrane proteins from each protocol were compared to the whole cell lysate and as can be seen from the immuno blot using protocol A to isolate membrane proteins resulted in a higher amount of membrane proteins and a lower level of cytosol proteins in the sample as compared to protocol B. Protocol B 130kDa Integrin \u00CE\u00B21 Membrane Whole Membrane Whole Protocol A A. Protocol A GAPDH Protocol B 36kDa Membrane Whole Membrane Whole B. 96 Figure 3.13: Repeat of THP-1 membrane protein enrichment using protocol A THP-1 membrane enrichment was repeated according to protocol A, in order to confirm previous results. The level of enrichment was tested again using an antibody against integrin \u00CE\u00B2, a membrane protein, GAPDH, an abundant cytosolic protein, and 14.3.3, another highly abundant cytosol protein. As evident from the immuno blot, membrane proteins are indeed present in a greater quantity after implementing protocol A as compared to whole cell lysates. On the other hand, cytosolic proteins are depleted in the membrane fraction as compared to the whole cell lysate. 3.7 Generating reference THP-1 cell dataset To identify THP-1 proteins in the whole cell lysate (global dataset) and the membrane protein enriched fraction (membrane dataset) by mass spectrometry, 70 \u00C2\u00B5g of protein from each sample was separated by SDS-PAGE (Figure 3.14A). From each lane, 24 bands were cut, reduced, alkylated and trypsin digested. The peptides from each lane were loaded on to FT-ICR for tandem mass spectrometry and protein identification. These experiments were repeated two more times for global and membrane proteins to generate a reference dataset. In all, there were 1747 proteins in common between the three datasets from global THP-1 cell samples based on search using CHiBi Mascot server (Figure 3.14B). There were 1375 proteins in common between the membrane datasets (Figure 3.14C). There were 827 proteins which are in common between the global and membrane datasets, with 920 unique proteins for global and 549 unique proteins for membrane dataset (Figure 3.14D). Membrane Membrane Membrane Whole Whole Whole Integrin \u00CE\u00B21 GAPDH 14.3.3 97 The proteins in common between the 3 global datasets and the 3 membrane protein datasets were submitted to DAVID Bioinformatics Resources 6.7 for gene ontology (GO) analysis. Through the GO analysis we grouped the proteins based on their cellular compartment, to find out whether the proteins identified are membrane proteins or cytosolic proteins. The top GO term for global dataset was cytosol with 405 proteins (Figure 3.14E), while cytosol was the seventeenth GO term with only 231 proteins (Figure 3.14F). Overall, the global fraction consists of many cytosolic and nuclear proteins, while membrane fraction has many membrane proteins from the plasma as well as organelles such as the mitochondria. Before starting with quantitative proteomics studies, we first used flow cytometry to confirm whether key proteins interactions between the releasate proteins and THP-1 cells surface proteins occur. 250 100 70 55 35 27 15 10 KDa Global Ladder Membrane A. Membrane and global proteins separated on SDS-PAGE B. THP-1 Global Dataset 1 4068 Proteins Dataset 2 3061 Proteins Dataset 3 3270 Proteins 1747 1221 571 762 541 202 559 98 0 50 100 150 200 250 300 350 400 450 500 N u m b e r o f P ro te in s Gene Ontology Term E. Global Cellular Compartment Increasing P-Value D. THP-1 Global versus membrane fraction THP-1 membrane fraction common proteins 1747 THP-1 global common proteins 1747 827 920 548 C. THP-1 Membrane Fraction Dataset 1 3403 Proteins Dataset 2 3051 Proteins Dataset 3 2499 Proteins 1375 1046 699 655 745 232 237 99 Figure 3.14: Reference global and membrane THP-1 cell datasets (A) 100\u00C2\u00B5g of proteins from THP-1 membrane fraction and whole lysate were loaded on to a 10% SDS- PAGE and separated. The gel was cut into 24 bands for each protein lane. The proteins were reduced, alkylated and after trypsin digestion were submitted to FT-ICR for MS/MS analysis. This experiment was repeated three times to generate three global and membrane datasets. There are 1747 proteins in common between the 3 global datasets (B) and 1375 proteins in common between the 3 membrane protein datasets (C) with 827 proteins shared between the global and membrane datasets (D). Top 30 gene ontology terms for cellular compartment as grouped by DAVID Bioinformatics Resources 6.7 (http://david.abcc.ncifcrf.gov/home.jsp), are shown for the 1747 proteins from global dataset (E) and the 1375 proteins from the membrane dataset (F). The top 30 GO terms were determined according to their p-value as assigned by DAVID software based on the likelihood that the grouped proteins are from a given GO term. 0 50 100 150 200 250 300 350 400 N u m b e r o f P ro te in s Gene Ontology Term F. THP-1 Membrane Fraction: Cellular Compartment Increasing P-Value 100 3.8 THP-1 cell stimulation by platelet releasate 3.8.1 Flow cytometry analysis Washed platelets at a concentration of 10 x 108 platelet/ml (concentrations found around plaque regions during rupture in patients with cardiovascular disorder) were activated with 0.1 U/ml of thrombin for 90 minutes. The Rel was collected by removing the platelet through centrifugation at 700 xg. Rel from four donors were combined and added to THP-1 cells during their exponential growth phase at a concentration of 1 ml of platelet releasate per 1 million THP-1 cells. Stimulated THP-1 cells were labeled with MAC-1 (integrin \u00CE\u00B1M) anti-human antibody after 4, 6 and 10 hours of stimulation and analyzed on the FACSCalibur flow cytometer. MAC-1 antibody binding to THP-1 cell surface decreases by 4 hours and stays the same after 6 and 10 hours (Figure 3.15). Next THP-1 cells were treated with Rel for 6 and 24 hours as explained in the methods, section 2.13. These THP-1 cells were stained with phycoerythrin (PE) conjugated clone of anti-human CD62P antibody which recognizes P-selectin (CD62P) bound to its ligand. After 6 hours of stimulation (Figure 3.16A) there were P-selectin proteins bound to THP-1 cells surface. After 24 hours (Figure 3.16B) there was still P-selectin bound to most THP-1 cells, but the peak shifted left indicating decreased amount of P-selectin on THP-1 cell surface. Now that we know these key protein interactions occur, we can analyze how these interactions lead to THP-1 cell stimulation by studying the global and membrane THP-1 proteome changes due to stimulation with Rel. We used a model system to find potential biomarkers which could play an important role in platelet-monocyte aggregate formation. Although our model system does not take into account the in vivo conditions, such as shear flow or the impact of cells such as red blood cells, under which activated platelets or proteins from the releasate interact with monocytes, the results from our studies could give us a starting point and potential proteins to further investigate for their biological role in platelet monocyte aggregate formation. 101 Figure 3.15: MAC-1 analysis on THP-1 cells stimulated with platelet releasate Releasate from thrombin activated platelets were added to THP-1 cells at a concentration of 1 mL of releasate per 1 million THP-1 cells for temporal stimulation according to section 2.12 and 2.13. THP-1 cells were stimulated for 4, 6 and 10 hours. The THP-1 cells were stained with fluorescein isothiocyanate (FITC) conjugated MAC-1 anti-human antibody. The THP-1 cells were then subject to flow cytometery analysis for effect of platelet releasate of THP-1 cells. MAC-1 also know an integrin \u00CE\u00B1M binds proteins such as glycoprotein ib\u00CE\u00B1. Antibody binding to MAC-1 on THP-1 cell surface decreases due to addition of releasate to THP-1 cells at 4, 6 and 24 hours. 102 Figure 3.16: P-selectin binding to THP-1 cells THP-1 cells were stimulated by releasate from thrombin activated platelets for 6 hours (A) and 24hours (B) as described in methods section. The THP-1 cells were stained with phycoerythrin (PE) conjugated clone of anti-human CD62P antibody which recognizes CD62P (P-selectin) bound to its ligand. The P- selectin from the releasate binds to THP-1 cells at 6 hours and to a lesser extent at 24 hours. A. Stimulation after 6 hours B. Stimulation after 24 hours 103 3.8.2 Global protein changes THP-1 cells were grown in light and heavy SILLAC media to compare the protein changes which take place upon THP-1 cell stimulation with releasate. THP-1 cells grown in heavy SILAC media were treated with Rel for 6 and 24 hours while THP-1 cells grown in light SILAC media were treated with vehicle control. Equal amounts of protein from light and heavy labeled THP-1 cell global proteins were mixed and analyzed using quantitative proteomics (Figure 2.2). Proteins whose normalized heavy to light ratios (H/L) deviated from 1 (or zero when using log2 scale) were either overexpressed or underexpressed in the THP-1 stimulated cells as compared to control (Figure 3.17). There were three repeats of these experiments to look for protein whose ratios significantly change after 6 (Figure 3.17A-C) and 24 hours of stimulation (Figure 3.17D-F). Most of the proteins found had H/L ratio of around 1 meaning they do not significantly change upon THP-1 stimulation with releasate. This resulted in a normal distribution around zero in the histograms depicted in Figures 3.17. There were also proteins with extreme ratios on either direction, but they were only found in one of the three datasets, and so were disregarded for the downstream bioinformatics analysis. There were 923 quantified proteins in common between the three datasets after 6 hours of stimulation (Figure 3.18A). There were 901 quantified proteins in common between the three datasets after 24 hours of stimulation (Figure 3.18B). In addition, there were 728 proteins in common between the \u00E2\u0080\u009C6 hours\u00E2\u0080\u009D dataset and \u00E2\u0080\u009C24 hours\u00E2\u0080\u009D dataset (Figure 3.18C). These 728 proteins represent all the proteins in common between the six global protein datasets and are used for downstream temporal analysis of protein changes due to stimulation with platelet releasate. 104 Figure 3.17: Global protein ratio distribution at 6 and 24 hours of THP-1 cell stimulation Heavy SILAC labeled THP-1 cells (H) were stimulated with releasate from thrombin activated platelets. The light SILAC THP-1 cells (L) were used as control and a 1:1 protein mix of light labeled and heavy labeled proteins were analyzed on the mass spectrometer and their normalized ratios calculated using MaxQuant (Max Plank institute of Biochemistry, maxquant.org). Ratios above zero indicate proteins that a given protein is more abundant in stimulated THP-1 cells. The figures show the biological repeats after 6 hours of stimulation (A-C) and 24 hours of stimulation (D-F). 105 Figure 3.18: Venn diagrams for protein overlaps by 6 hours and 24 hours global datasets There are 923 quantified proteins in common between the three datasets from 6 hours of THP-1 stimulation with platelet releasate (A). There are 901 quantified proteins in common between the three datasets from THP-1 cell stimulation for 24 hours (B). (C) The proteins in common between all six datasets. These 728 proteins were used for farther analysis to study effects of platelet releasate on THP- 1 cells. 24 h THP-1 global all 3 datasets (901 proteins) 6 h THP-1 global all 3 datasets (923 proteins) 728 195 173 6 h THP-1 global dataset 1 (1906 proteins quantified) 6 h THP-1 global dataset 2 (1833 proteins quantified) 923 375 132 681 565 213 43 6 h THP-1 global dataset 3 (1860 proteins quantified) 24 h THP-1 global dataset 1 (2416 proteins quantified) 24 h THP-1 global dataset 2 (1594 proteins quantified) 901 918 74 587 548 71 49 24 h THP-1 global dataset 3 (1608 proteins quantified) A. B. C. 106 Statistical analysis was performed on the 923 proteins in common between the \u00E2\u0080\u009C6 hours\u00E2\u0080\u009D datasets. The mean heavy to light ratio for these proteins was 0.96\u00C2\u00B10.18. Any protein that had a ratio of 1.5 x \u00CF\u0083 (0.18) less than the mean was considered significantly underexpressed (i.e. any proteins with a ratio of less than 0.69). By 6 hours, there were 52 proteins that were less abundant in THP-1 cells stimulated with platelet releasate as compared to control THP-1 cells. On the other hand, any protein whose ratio was 1.5 x \u00CF\u0083 greater than the mean was considered significantly overexpressed (i.e. any proteins with a ratio of greater than 1.23). There were 42 such proteins. To follow the changes which take place by 6 and 24 hours of stimulation the 728 proteins in common between all datasets were used for analysis. Figure 3.19 shows the proteins which were significantly underexpressed at 6 hours of stimulation. Some of the proteins which were significantly underexpressed at 6 hours remain underexpressed at 24 hours (Figure 3.19A), while others increase in abundance by 24 hours of stimulation (Figure 3.19B). Figure 3.20 follows the fate of proteins which were significantly overexpressed at 6 hours. Some of these proteins remain overexpressed at 24 hours (Figure 3.20A), while others drop in expression level (Figure 3.20B). We used Ingenuity Pathway Analysis (IPA) software to find out how the 923 proteins identified in the \u00E2\u0080\u009C6 hour\u00E2\u0080\u009D dataset connect together and their biological function. We focused on the pathways which are important in cardiovascular disorder and platelet-monocyte aggregate formation. One of the top findings for biological role of the proteins was inflammatory response (Figure 3.21). This pathway highlights how platelets are capable of initiating an inflammatory response in THP-1 cells, in addition to their role in hemostasis. All the proteins found in this pathway, are overexpressed in the stimulated samples as compared to the control sample as indicated by the red color. These proteins include cathepsin G, proteinase 3, serpin B8 and kinectin. Next we looked at how protein expression changes in 107 0 0.5 1 1.5 2 2.5 0 h 6 h 24 h H /L r at io n ( n o rm al iz e d ) Time points (hours) B. Even or over expressed by 24 h 60S ribosomal protein L18 60S ribosomal protein L7a 60S ribosomal protein L23 Sulfotransferase 1A1 Ribosomal protein L10 Squamous cell carcinoma antigen recognized by T-cells 3 60S ribosomal protein L8 40S ribosomal protein S3a 40S ribosomal protein S11 Transcriptional activator protein Pur-alpha 60S ribosomal protein L15 60S ribosomal protein L3 OCIA domain-containing protein 1 60S ribosomal protein L19 La-related protein 1 EF-hand domain-containing protein D2 Histone H1x after a longer time, after 24 hours, where the cells had more time for proteins synthesis and degradation of stimulation. Figure 3.19: Significant decreases in global proteins at 6 hours of stimulation Proteins which are significantly decreased in abundance when comparing heavy labeled THP-1 proteins (H) treated with platelet releasate to control light labeled THP-1 proteins (L) are graphed here. Graph A shows the group of proteins which either stay at the same ratio or increase slightly after 24 hours of THP-1 stimulation. Graph B shows the group of proteins which increase their expression at 24 hours to a ratio of about 1:1 or above. 0 0.2 0.4 0.6 0.8 1 1.2 0 h 6 h 24 h H /L r at io n ( n o rm al iz e d ) Time points (hours) A. Underexpressed by 24 h 39S ribosomal protein L4 40S ribosomal protein S14 cDNA FLJ56357 Putative uncharacterized protein MRPL23 39S ribosomal protein L9 Microtubule-associated protein 1B Phosphoribosyl pyrophosphate synthetase-associated protein 1 Reticulon-4 40S ribosomal protein S2 Cytosolic non-specific dipeptidase 39S ribosomal protein L40 Splicing factor 1 Heterogeneous nuclear ribonucleoprotein H2 cDNA FLJ58953 Poly(U)-binding-splicing factor PUF60 Ras GTPase-activating protein-binding protein 2 Neuroblast differentiation-associated protein AHNAK Syntaxin-7 Glutathione peroxidase 1 Coronin-7 108 0 0.5 1 1.5 2 2.5 0 h 6 h 24 h H /L r at io n ( n o rm al iz e d ) Time Points (hours) A. Similar or increased expression levels by 24 Calreticulin Lamin-A/C Ornithine aminotransferase Peptidyl-prolyl cis-trans isomerase H Proteasome subunit beta type-6 Guanine nucleotide-binding protein G(I)/G(S)/G(T) Transaldolase Lactoylglutathione lyase Switch-associated protein 70 Serpin B8 UDP-glucose 6-dehydrogenase RNA-binding protein 39 0 0.5 1 1.5 2 2.5 0 h 6 h 24 h H /L r at io n ( n o rm al iz e d ) Time points (hours) B. Drop in expression levels by 24 h Myeloblastin Lamin-B2 Lamin-B1 Isopentenyl-diphosphate Delta-isomerase 1 Nuclear pore membrane glycoprotein 210 Cathepsin G Caprin-1 Kinectin Serine/threonine-protein phosphatase Ran GTPase-activating protein 1 Nucleoprotein TPR THO complex subunit 2 L-xylulose reductase UPF0556 protein C19orf10 Figure 3.20: Significant increase in global proteins at 6 hours stimulation Proteins which are significantly increased in abundance when comparing heavy labeled THP-1 proteins (H) treated with platelet releasate to light labeled THP-1 proteins (L) treated with vehicle control are graphed here. Based on the common proteins between the three 6 hour datasets any proteins which are more than 1.5 times the standard deviation above the average ratio are considered significantly over expressed. Graph A shows the group of proteins which either stay at the same ratio or increase after 24 hours of THP-1 stimulation. Graph B shows the group of proteins which decrease their expression by 24 hours. 109 Figure 3.21: Ingenuity Pathway Analysis (IPA) looking at protein changes by 6 hours The 923 proteins found in common between all dataset from 6 hour stimulation of THP-1 cell with platelet release were submitted to IPA to find pathways connecting the proteins together. This figure shows that the proteins found strongly correlate together through inflammatory response. Proteins in red are the proteins which are found in the \u00E2\u0080\u009C6 hours\u00E2\u0080\u009D datasets and the shade of red indicates the ratio. The darker the red the more abundant a given protein in THP-1 cells stimulated with platelet releasate as compared to the protein in the control sample. Abbreviations: SFTPD, pulmonary surfactant- associated protein D; SERPINB1, leukocyte elastase inhibitor; PI3, elafin preproprotein; PPIF, peptidyl=prolyl cis\u00E2\u0080\u0094trans isomerse F; IgG, immunoglobulin G; SWAP70, switch-associated protein 70; Timp, metalloproteinase inhibitor; PRTN3, myeloblastin; CCL3L1, C-C motif chemokine 3-like 1; SLPI, antileukoproteinase; FDFT1, squalene synthase; CTSG, cathepsin G; TNF, tumor necrosis factor; CALR, calretiuclin; TPR, nucleoprotein TPR; RANGAP1, RAN GTPase-activating protein 1; PPP1CC, serine/threonine-protein phosphatase PP1-gamma; KTN1, kinectin; SAR1A, GTP-binding protein SAR1A; MYL12A, myosin regulatory light chain 12A; DCXR, L-xylulose reductase; Mlcp, phosphatidylinositol 4- phosphate 3-kinase; CSTF3, cleavage stimulation factor subunit 3; OAT, ornithine aminotransferase. 110 The same statistical analysis was performed using the 901 proteins in common between the three \u00E2\u0080\u009C24 hours\u00E2\u0080\u009D datasets. The mean H/L ratio for these proteins was 1.04\u00C2\u00B10.27. By 24 hours of stimulation, there were only 10 proteins which were significantly decreased in abundance (H/L ratio of less than 0.64) as compared to control. In contrast, any protein with a ratio of greater than 1.44 is considered significantly overexpressed and there were 43 such proteins. Figure 3.22 shows the proteins which were significantly underexpressed by 24 hours of stimulation. As can be seen all of these proteins were also underexpressed at 6 hours and most decreased in ratio by 24 hours. There were more proteins overexpressed by 24 hours of stimulation than underexpressed. Figure 3.23 shows the proteins which were significantly overexpressed at 24 hours. Some of these proteins such as annexin (ANXA) and Signal Transducers and Activators of Transcription 1 (STAT1) are evenly expressed at 6 hours and increase in expression at 24 hours (Figure 3.23A). Other proteins were already overexpressed by 6 hours and increase in abundance in stimulated THP-1 cells by 24 hours (Figure 3.23B). Once again, the 901 proteins found in common between the \u00E2\u0080\u009C24 hours\u00E2\u0080\u009D datasets were submitted to IPA to find the different biological functions which these proteins are involved in. Figure 3.24 shows some the proteins such as STAT1 and STAT3 which are overexpressed by 24 hours and the biological function they can potentially participate in. These functions include, immune response, binding of blood platelets and activation of leukocytes. These pathways again highlight how platelet releasate is capable of initiating immune response in THP-1 cells. Rel protein interaction with THP-1 cells leads to feedback mechanisms in the THP-1 cells which can lead to further cell to cell interactions, and forming platelet-monocyte aggregates. To further study the interactions between the platelet releasate and THP-1 cells, and to look specifically at the proteins in the surface of THP-1 cells we focused on enriched membrane proteins. We looked at how membrane proteins change expression at 6 and 24 111 hours of stimulation to have an idea of short and long term changes which take place. These changes could result from both initial interactions with proteins in the releasate and feedback mechanisms that reinforce platelet-monocyte aggregate formation. Figure 3.22: Significant decreases in global proteins at 24 hours of stimulation This graph shows the proteins which are significantly decreased in abundance when comparing heavy labeled THP-1 proteins (H) treated with platelet releasate to light labeled THP-1 proteins (L) treated with vehicle control. Based on the common proteins between the three 24 hours datasets any proteins which are 1.5 times the standard deviation under the average ratio are considered significantly over expressed. All the proteins found to be significantly underexpressed at 24 hours are also underexpressed at 6 hours. 0 0.2 0.4 0.6 0.8 1 1.2 0 h 6 h 24 h H /L r at io n ( n o rm al iz e d ) Time points (hours) Neutral amino acid transporter B(0) KH domain-containing ARHGAP4 protein Nuclease-sensitive element-binding protein 1 39S ribosomal protein L4 Reticulon-4 112 0 0.5 1 1.5 2 2.5 3 3.5 0 h 6 h 24 h H /L r at io n ( n o rm al iz e d ) Time points (hours) A. Under or evenly expressed by 6 h Histone H1x EF-hand domain-containing protein D2 Interferon-induced Histone H1.5 Histone H1.0 2'-5'-oligoadenylate synthetase 2 Signal transducer and activator of transcription 1 Heterogeneous nuclear ribonucleoprotein A0 Heterochromatin protein 1-binding protein 3 Cathepsin D Protein DEK Putative neutrophil cytosol factor 1B High mobility group protein B3 Coiled-coil domain-containing protein 124 Annexin A2 Annexin A5 Figure 3.23: Significant increase in global proteins at 24 hours of stimulation Proteins which are significantly increased in abundance when comparing heavy labeled THP-1 proteins (H) treated with platelet releasate to light labeled THP-1 proteins (L) treated with vehicle control by 24 hours of stimulation are graphed. Based on the common proteins between the three 24 hour datasets any proteins which are 1.5 times the standard deviation above the average ratio are considered significantly over expressed. Graph A shows the group of proteins which are either underexpressed or evenly expressed at 6 hours. Graph B shows the group of proteins which are overexpressed at 6 hours. 0 0.5 1 1.5 2 2.5 3 0 h 6 h 24 h H /L r at io n ( n o rm al iz e d ) Time points (hours) B. Over expressed by 6 h Heterogeneous nuclear ribonucleoproteins A2/B1 Heterogeneous nuclear ribonucleoprotein H3 Filamin B Alpha-actinin-4 Splicing factor Formin-like protein 1 Eukaryotic translation initiation factor 4H SH3 domain-binding protein 1 RNA-binding protein 39 Serpin B8 UDP-glucose 6-dehydrogenase Lamin-A/C 113 Figure 3.24: Ingenuity Pathway Analysis (IPA) looking at protein changes by 24 hours List of proteins found in common between all dataset was submitted to IPA to find pathways connecting the proteins together. This figure shows how the proteins connect together through different functions such as immune response and binding of blood platelets and their ratio changes by 24 hours. The shades of red indicate the ratios. The darker the red the more abundant a given protein in THP-1 cells stimulated with platelet releasate as compared to the protein in the control sample. Abbreviations: PECAM1, platelet endothelial cell adhesion molecule; STAT, signal transducer and activator of transcription; LGALS1, galectin-1; ANXA, annexin; ITGA5, integrin alpha 5. 3.8.3 Membrane protein changes THP-1 cells were grown in SILAC media and treated with platelet releasate, in the same manner as for global protein analysis, section 3.3.2. Membrane proteins from stimulated heavy labeled and control light labeled THP-1 cells were enriched according to protocol A and analyzed using quantitative proteomics. There were three repeats of these experiments to look for proteins whose H/L ratios significantly change after 6 hours and 24 hours of stimulation. Figure 3.25 shows the histograms for the normalized Log2 H/L ratios for each dataset. Once again, there was a normal distribution around zero in the histograms depicted. There were 285 quantified proteins in common between the three datasets from 6 hours of stimulation (Figure 3.26A). Looking at the \u00E2\u0080\u009C24 hours\u00E2\u0080\u009D membrane protein datasets, there were 413 114 proteins in common between the three datasets (Figure 3.26B). There were 208 proteins in common between the \u00E2\u0080\u009C6 hours\u00E2\u0080\u009D and \u00E2\u0080\u009C24 hours\u00E2\u0080\u009D membrane enriched datasets (Figure 3.26C). These 208 proteins represent all the proteins in common between the six membrane protein datasets and are used for downstream temporal analysis of protein changes due to stimulation with platelet releasate. Figure 3.25: Membrane protein ratio distribution at 6 and 24 hours of THP-1 cell stimulation Heavy SILAC labeled THP-1 cells (H) were stimulated with releasate from thrombin activated platelets. The light SILAC THP-1 cells (L) were used as control. Membrane proteins from heavy and light labeled THP-1 cells were enriched according to protocol A and 1:1 mix of the resulting proteins were analyzed on the mass spectrometer and their normalized ratios calculated using MaxQuant (Max Plank institute of Biochemistry, maxquant.org). Ratios above zero indicate proteins that a given protein is more abundant in stimulated THP-1 cells. The figures show the biological repeats after 6 hours of stimulation (A-C) and 24 hours of stimulation (D-F). 115 Figure 3.26: Venn diagrams for protein overlaps for membrane datasets Venn diagrams representing the protein overlap from the three membrane protein enriched datasets after 6 hours (A) and 24 hours (B) of stimulation with platelet releasate. Overall there are 285 quantified proteins in common between the three dataset after 6 hours of THP-1 stimulation and 413 quantified proteins in common after 24 hours of stimulation. Venn diagram C shows the proteins in common between the \u00E2\u0080\u009C6 hours\u00E2\u0080\u009D dataset and the \u00E2\u0080\u009C24 hours\u00E2\u0080\u009D datasets. The 208 proteins in common between all six datasets are used for all downstream analysis to study platelet releasate effect on THP-1 cells. 24h THP-1 membrane dataset 1 (1042 proteins quantified) 24h THP-1 membrane dataset 2 (863 proteins quantified) 413 339 216 30 188 46 102 24h THP-1 membrane dataset 3 (591 proteins quantified) 6h THP-1 Membrane all 3 datasets (285 proteins) 208 77 205 6h THP-1 Membrane all 3 datasets (413 proteins) A. B. C. 6h THP-1 membrane dataset 1 (797 proteins quantified) 6h THP-1 membrane dataset 2 (633 proteins quantified) 285 320 80 239 149 119 43 6h THP-1 membrane dataset 3 (776 proteins quantified) 116 Statistical analysis was performed on the proteins 285 proteins in common between the \u00E2\u0080\u009C6 hours\u00E2\u0080\u009D membrane protein datasets in the same manner as section 3.3.2. The mean H/L ratio for these proteins was 1.02\u00C2\u00B10.19. By 6 hours of stimulation, there were 18 proteins which were significantly underexpressed (H/L ratio of less than 0.73) in THP-1 cells treated with releasate as compared to control. On the other hand, there were 20 proteins significantly overexpressed (H/L ratio of greater than 1.30). To follow the changes which take place by 6 and 24 hours of stimulation the 208 proteins in common between all membrane datasets were used for analysis. Figure 3.27 shows the proteins which significantly changed in expression level by 6 hours. Some of the proteins which were significantly underexpressed by 6 hours remained underexpressed by 24 hours, while others such as platelet endothelial cell adhesion molecule (PECAM-1) increase in abundance by 24 hours (Figure 3.27A). Figure 3.27B shows the proteins which were significantly overexpressed by 6 hours and how their ratio changes by 24 hours. These protein changes are important in signaling which eventually can reinforce platelet- monocyte aggregate formation. To analyze the long term changes we once again looked at the \u00E2\u0080\u009C24 hour\u00E2\u0080\u009D dataset. Again the same statistical analysis was performed using the 413 proteins in common between the three \u00E2\u0080\u009C24 hours\u00E2\u0080\u009D datasets to highlight the proteins which significantly change upon THP-1 cell stimulation. The mean H/L ratio for these proteins was 1.03\u00C2\u00B10.29. There were only 7 proteins which were significantly decreased in abundance upon THP-1 stimulation (proteins with a ratio less than 0.60). Any protein with a ratio greater than 1.46 is considered significantly overexpressed, and there were 19 such proteins. 117 0 0.5 1 1.5 2 2.5 3 0h 6h 24h H /L r at io ( n o rm al iz e d ) Time points (hours) A. Significant decrease by 6 h Membrane-associated progesterone receptor Polyadenylate-binding protein 1 Fermitin family homolog 3 60S ribosomal protein L6 Gelsolin Thioredoxin-related transmembrane protein 1 Adenylyl cyclase-associated protein 1 60S ribosomal protein L4 40S ribosomal protein S8 Pyruvate kinase isozymes M1/M2 Neuroblast differentiation-associated protein AHNAK Band 4.1-like protein 2 Coronin-1A Platelet endothelial cell adhesion molecule Figure 3.27: Significant changes in membrane proteome datasets at 6 hours of stimulation The graphs show proteins which are significantly changed in abundance when comparing heavy labeled THP-1 proteins (H) treated with platelet releasate to light labeled THP-1 proteins (L) treated with vehicle control at 6 hours. Graph A shows the group of proteins which are significantly underexpressed at 6 hours. Some proteins remain at the same ratio at 24 hours while some increase their expression at 24 hours. Graph B shows the group of proteins which are significantly over expressed at 6 hours. Some of these proteins remain over expressed at 24 hours and some such as myelobastin decrease in expression at 24 hours of stimulation. 0 0.5 1 1.5 2 2.5 0h 6h 24h H /L r at io ( n o rm al iz e d ) Time points (hours) B. Significant increase by 6 h Cytochrome b-c1 complex subunit 2 Aspartate aminotransferase Mannosyl-oligosaccharide glucosidase Putative uncharacterized protein DKFZp686M24262 Isocitrate dehydrogenase [NADP] Solute carrier family 2 member 1 variant Chitinase-3-like protein 1 Cathepsin G Citrate synthase Lysosome-associated membrane glycoprotein 1 Calreticulin Spectrin beta chain Ornithine aminotransferase Ras-related protein Rab-27A Succinyl-CoA ligase [ADP-forming] subunit beta Myeloblastin 118 Figure 3.28 follows the fate of proteins which significantly change by 24 hours of stimulation. Figure 3.28A shows the two proteins, transferrin receptor protein 1) and myeloblastin which were significantly underexpressed at 24 hours. Figure 3.28B shows the proteins which were significantly overexpressed by 24 hours. Most of these proteins were evenly expressed at 6 hours and their expression increases by 24 hours. Examples of these proteins include PECAM1, integrin beta-1 (ITB1) and annexin A2 (ANXA2). In addition, the 413 proteins in common between the \u00E2\u0080\u009C24 hours\u00E2\u0080\u009D membrane dataset were submitted to IPA to find the relationship between the proteins and the pathways they are involved in. We focused on important surface proteins such as integrins and leukosialin which are capable of interacting with platelets and other blood cells creating aggregates. These proteins are closely related to each other through leukocyte extravasation signaling. Figure 3.29 shows the molecules that were involved in leukocyte extravasation signaling such as platelet endothelial cell adhesion molecule, integrin beta-1 and leukosialin (SPN) all of which were overexpressed by 24 hours of stimulation. This pathway is very important in monocytes stimulation and their transmigration across the endothelial cell and forming plaques in cardiovascular disease. Overall, these results show that platelet releasate are sufficient to stimulate THP-1 cells and initiate signaling pathways that lead to platelet-monocyte aggregate formation and transmigration across to the endothelial cell, all of which are important events that aid in the progression of cardiovascular disease. 119 0 0.5 1 1.5 2 2.5 3 3.5 0h 6h 24h H /L r at io ( n o rm al iz e d ) Time Points (hours) B. Significant increase by 24 h Ras GTPase-activating-like proteinIQGAP1 cDNA FLJ55458 4F2 cell-surface antigen heavy chain Ezrin Integrin beta-1 Moesin Integrin alpha-5 Platelet endothelial cell adhesion molecule Cathepsin D Annexin A2 Figure 3.28: Significant changes in membrane proteome datasets at 24 hours of stimulation The graphs show the proteins which are significantly changed in abundance when comparing heavy labeled THP-1 proteins (H) treated with platelet releasate to light labeled THP-1 proteins (L) treated with vehicle control after 24 hours. Graph A shows the group of proteins which are significantly underexpressed at 24 hours. Graph B shows the group of proteins which are significantly over expressed at 24 hours. 0 0.2 0.4 0.6 0.8 1 1.2 1.4 1.6 1.8 2 0h 6h 24h H /L r at io ( n o rm liz e d ) Time points (hours) A. Significant decrease by 24 h Transferrin receptor protein 1 Myeloblastin 120 Figure 3.29: Ingenuity Pathway Analysis of proteins found from membrane enriched datasets The 413 proteins in common between the three \u00E2\u0080\u009C24 hours\u00E2\u0080\u009D membrane dataset was submitted to IPA for pathway analysis. This figure depicts the proteins as they relate to leukocyte extravasation signaling. The red color indicates over expression of proteins of heavy labeled THP-1 cells stimulated with platelet releasate as compared to the light labeled control THP-1 cells at 24 hours of stimulation. The green color indicates underexpression of heavy labeled proteins at 24 hours of stimulation. Some of the key proteins found to be overexpressed include integrin \u00CE\u00B21 (ITGB1) and leukosialin (SPN). Abbreviations: ACTN4, alpha actin 4; MSN, moesin; EZR, ezrin; GNAI2, guanine nucleotide binding protein (G protein) alpha 2; ITGB, integrin \u00CE\u00B2; SPN, leukosialin; PECAM1, platelet endothelial adhesion molecule. 121 4 Discussion Through our work, we managed to establish a successful working model system for comparing the platelet releasate proteome profile when using different agonists. Based on our analysis, we managed to identify candidate proteins in platelet releasate that could potentially play a unique role in normal or disease sates. In addition, we developed a model system for the study of proteins involved in platelet- monocyte aggregate formation, using THP-1 cells to study mechanisms of stimulation by platelet releasate. The proteins and their changes identified in our model system were validated by observations from previous studies of platelet-monocyte interactions. In addition, we identified candidate biomarkers that could potentially play an important role in platelet-monocyte aggregate formation. Further biological studies may reveal and confirm the mechanisms by which these proteins effect platelet- monocyte aggregate formation and the progression of cardiovascular disease. 4.1 Protein content of platelet releasate Before stimulating THP-1 cells with platelet releasate to study the protein changes that take place and the pathways that lead to THP-1 cell stimulation and platelet-monocyte aggregate formation, we analyzed the content of platelet releasate when using different agonists, and found out which proteins in the releasate can potentially interact with THP-1 cells. Solely based on the amount of protein in the total releasate (Rel) when platelets are activated with different agonists, we can tell that the protein releasate content depends on the agonist used (Table 3.1). Furthermore, based on the amount of proteins in the releasate minus the microparticles (Rel-MP) and MP fractions of platelet releasate we can tell which agonist causes the greater MP generation, and from these results it seems thrombin activated platelets secrete the least amount of MP. We further explored the content of Rel and MP and their biological significance through quantitative proteomics and other techniques such as flow cytometry. 122 We chose to identify the releasate proteins from platelets activated with mildly oxidized low density lipoprotein (mox-LDL), since it plays a key role as an agonist upon plaque rupture in patients with cardiovascular disorder. The amount of protein released from platelets activated with mox-LDL activation is not reliable, since mox-LDL has a lot of apolipoproteins that get added to the sample. This is evident when the proteins are separated on SDS-PAGE, where releasate from mox-LDL activated platelets displayed a strong band at around 130 KDa (Figure 3.1A). We identified relatively few Rel, Rel- MP and platelet MP proteins using LC-MS/MS after protein digestion. The majority of proteins in Rel and Rel-MP were different isoforms of apolipoproteins (data not shown). Despite separation using SDS- PAGE, apolipoprotein presence made the identification of other proteins difficult. To gain a more in-depth analysis of releasate proteins, and compare differences between the Rel and Rel-MP, a quantitative proteomic approach was used with dimethyl labeling. To overcome the problems mentioned for mox-LDL (mainly the abundance of apoliproteins), LPA was chosen as an agonist. As mentioned in section 1.5, LPA is the most potent platelet agonist in mox-LDL, so we used LPA to represent the major agonist which activates platelets upon plaque rupture in patients with cardiovascular disorder in our quantitative proteomics studies. Rel and Rel-MP proteins from thrombin, collagen or thrombin plus collagen were quantitatively analyzed together using the three dimethyl labels to represent the more traditional agonists responsible for hemostasis. Rel and Rel-MP proteins from collagen, LPA or subthreshold collagen plus LPA were analyzed together to represent the agonist exposed to platelets during plaque rupture in patients with cardiovascular disease. Furthermore, collagen can be used as a reference point to compare LPA to thrombin. 123 4.2 Quantitative proteomic analysis of releasate proteins from thrombin, collagen and thrombin plus collagen activated platelets Quantitated proteins shared between the three Rel datasets and quantitated proteins shared between the three Rel-MP datasets were used for analyzing the different proteome profiles when using thrombin, collagen or thrombin plus collagen as agonists (Figure 3.2). The resulting quantitative data produced thrombin/collagen (T/C) ratios and thrombin/thrombin plus collagen (T/TC) ratios. In other words, we calculated the relative abundance of Rel and Rel-MP proteins from thrombin activated platelets compared to collagen activated platelets, as well as thrombin plus collagen activated platelets. There are no proteins in the Rel or the Rel-MP sample that we expect to have a 1:1 ratio, so we did not normalize the data. Based on similar chemical quantitative proteomic methods and the number of protein found and their ratios, any deviation greater than 20 % from 1 is considered significantly different. For example, a T/C ratio greater than 1.2 shows that a protein is significantly more abundant in the sample from thrombin activated platelets and a T/C ratio of less than 0.8 shows that a protein is significantly more abundant in the sample from collagen activated platelets. To see how the proteome profile of releasates from thrombin or collagen activated platelets differ we looked at the T/C ratios of each protein. 4.2.1 Thrombin versus collagen activated platelets First, to get an overview of the functions of proteins which are more abundant in thrombin or collagen activated platelets, we submitted the abundant proteins to DAVID software to determine the biological functions of these proteins and how they interact with each other. Overall, proteins which are significantly more abundant in thrombin activated platelets are involved in wound healing and immune response (Figure 3.3A). On the other hand, collagen activation leads to the release of more cytoskeletal proteins which can be categorized into gene ontology (GO) terms such as cell motion and actin filament 124 based process (Figure 3.3B). From these data we can see that there is great potential for Rel from platelets activated with different agonists to play roles in different cellular processes. Next we looked at all the proteins quantitated and grouped them according to their biological function and based on the location in platelets they could originate from. The 97 proteins in common between the Rel datasets were grouped according to their biological process and cellular compartment GO terms (Tables 3.4 and 3.5) using DAVID software. Not surprisingly, response to wounding, coagulation and hemostasis were the top biological GO terms. The protein with the highest T/C ratio was of course prothrombin, due to the addition of thrombin to platelets, and the lowest T/C ratio was collagen, again due to its addition to the sample. A lot of proteins in this group had an average T/C ratio of around 1, such as fibronectin or plasminogen, which shows that both agonists are capable of activating platelets to perform their important role in coagulation. Next we studied which Rel proteins can have the potential to interact with other blood cells which could have great implications in monocyte stimulation and platelet-monocyte aggregate formation. Looking at proteins grouped into cell adhesion GO terms, we see a lot of key proteins involved in cell adhesion and stimulation. Nidogen-1 and nidogen-2 are cell adhesion glycoproteins which bind to collagen and are more abundant when thrombin is used as an agonist. This could make biological sense since if platelets are already bound to collagen, they do not need to secrete more proteins which also bind collagen, whereas, thrombin activated platelets do. Another important protein which was more abundant in thrombin activated platelets is coagulation factor V, which makes part of the complex that converts prothrombin to its active thrombin form. This could lead into a positive feedback mechanism and activation of more platelets through thrombin in the body. In addition, we wanted to see which of the proteins released through activation could play a role in immune and inflammatory response which could aid in monocyte stimulation and the progression of cardiovascular disorder. 125 In general, there were proteins in different groups which had varying T/C ratios between datasets, but this inconsistency is especially evident for proteins grouped into immune and inflammatory response. For this reason, ratios for all three datasets as well as their averages are displayed in the tables. Platelet factor 4 (PF4), TGF\u00CE\u00B21 and C-C motif chemokine 5 (cleaved into RANTES which plays an important role in platelet monocyte aggregate formation) all have varying T/C ratios between the three datasets. However, on average all the three examples mentioned have a T/C ratio of more than 1.2. These differences in ratios between datasets could result from the experimental design and instrumentation. On the other hand, this variation could very well be biological and due to heterogeneity of platelets from different donors. Dataset 1 seems to have the most proteins with T/C ratio over 1.2 compared to the other two datasets, possibly because the platelets are more responsive to thrombin and/or they have different content stored in the platelets. Breland et al. found that THP-1 cells sometimes did not respond to stimulation by platelet releasate from thrombin activated platelets from some donors (Breland, Michelsen et al. 2010). These results could explain the different responses of THP-1 cells. There are also other proteins in these categories which could play an important role in monocyte stimulation. There are different types of immunoglobulin proteins released upon activation and on average they are more abundant in Rel from thrombin activated platelets. While immunoglobulin proteins have been found in platelet releasate by other groups (Piersma, Broxterman et al. 2009) their exact locations in platelets is not yet confirmed. These proteins which on average are more abundant in Rel from thrombin activated platelets could play an important role in immune response and interaction with monocytes. One important protein missing from this list is P-selectin. P-selectin was only found in dataset 1 and 2 with T/C ratio of 0.30 and 0.49 respectively. Upon platelet activation with collagen, more P- 126 selectin is secreted into the matrix as compared to when platelets are activated by thrombin. This can be explained by flow cytometry studies which show that thrombin activated platelets have higher P- selectin expression on their surface as compared to collagen activated platelets, which could mean that collagen activated platelets shed more of their P-selectin after 10 min of activation (Figure 3.10H). This is very important to platelet-monocyte aggregates, as more P-selectin on platelet surface leads to a stronger adhesion between platelets and monocytes. In addition, to get a better idea of where in the platelets the releasate proteins originate from we grouped all the proteins based on cellular compartment GO term. Not surprisingly, platelet \u00CE\u00B1-granule was the top cellular compartment GO term. Abundant \u00CE\u00B1- granule proteins such as thrombospondin-1 (the most abundant protein in releasate based on its Mascot score) and VWF had T/C ratios of around 1. However, we do see variability between the three datasets. This suggests that either the content of \u00CE\u00B1-granule in platelets from dataset 1 was different or that there was a greater extent of secretion due to thrombin activation. Moreover, there were some proteins found in the \u00CE\u00B1-granules that were more abundant in Rel from thrombin activated platelets and some which were more abundant from collagen activated platelets. One reason for this could be that these proteins have other sources, for example they might be shed from the membrane like integrin \u00CE\u00B1IIb. However, based on the range of T/C ratios we see for each protein from \u00CE\u00B1-granules, we believe that thrombin and collagen platelet activation lead to differential secretion of \u00CE\u00B1-granule content. The mechanisms by which this could occur are yet unknown, but our findings show the need to perform further research to look specifically at mechanism responsible for granule secretion. As mentioned, in addition to secretion, platelets shed membrane proteins during activation. An important adhesion protein found on the membrane which is shed during platelet activation is integrin \u00CE\u00B1IIb. On average this protein was more abundant in Rel from collagen activated platelets. It is curious 127 that its T/C ratio for dataset 3 was 1.83, but these experiments are not sufficient to explain this observation. On other hand, proteins which belong to cytoskeletal GO term are consistently more abundant when platelets are activated with collagen. All of the cytoskeletal proteins, except for two, were more abundant in Rel from collagen activated platelets. These proteins include actin, talin-1 and filamin among others. Some of these proteins were also associated with the membrane. For example, filamin and 14.3.3 are associated with GP Ib/IX/V complex (Section 1.1.2) so there is a good chance they can also be shed during platelet activation by collagen. However, by comparing the T/C ratios in Rel samples to those from Rel-MP samples we can get a better idea of where these cytoskeletal proteins come from. From these data, we can conclude that using thrombin or collagen leads to a releasate proteome profile that looks very different based on the relative abundance of each protein. These differences could have important biological implications both in hemostasis and in the context of cardiovascular disorder. Moreover, to get a better understanding of the nature of the releasate and to study the content of MP we also compared the thrombin to collagen ratios in Rel versus the Rel-MP. 4.2.2 Rel T/C versus Rel-MP T/C The MP were removed from the Rel via ultracentrifugation and the Rel-MP was quantitatively analyzed in the same manner as for the Rel proteins. There were 34 proteins with an average T/C ratio of less than 0.8 (as compared to 49 in the Rel) (Table 3.2). This shows that a lot of proteins in the Rel from collagen activated platelets were packaged into MP. Upon performing biological process GO analysis on these 34 proteins we found that cell motion and actin filament-based process GO terms were no longer the top GO terms (as was the case in Rel sample). The top GO terms in the Rel-MP sample were platelet activation and coagulation (Figure 3.5). However, there were still cytoskeletal proteins in the Rel-MP sample with T/C ratio of less than 0.8 such as actin and talin, as evident from the graph where actin 128 cytoskeleton organization has 7 proteins (as compared to 9 in Rel sample, Figure 3.3B). Based on these data we may conclude that a lot of the cytoskeletal proteins could have been in the MP when platelets are activated with collagen. We studied which specific proteins could be abundant in MP from thrombin or collagen activated platelets by looking at the proteins unique to Rel samples and the Rel-MP samples as well as comparing the T/C ratios between the Rel and Rel-MP samples. By looking at the proteins unique to Rel samples we can make comments about their abundance in MP. If a protein is present in the Rel dataset and absent from the Rel-MP, there is a good chance that the protein primarily comes from the MP. Kininogen-1 and integrin-linked proteins kinase are only found in Rel sample and are important in contributing to platelet function (Appendix D). Kininogen-1 is an inhibitor of thiol proteases and plays an important role in blood coagulation. Integrin-linked proteins kinase regulates integrin-mediated signaling and may act as a mediator for inside-out integrin signaling. Since integrin-linked protein kinase has a very low T/C ratio, this protein came mostly from MP generated form collagen activated platelets as opposed to MP from thrombin activated platelets. Theoretically, there should be no proteins unique to Rel-MP sample, but the instrument could miss peptides for protein identification in the Rel samples. An example of a protein unique to Rel-MP is platelet GP VI which was much more abundant in the Rel-MP of collagen activated platelets (Appendix E). There were other GP identified in the Rel sample, but a peptide unique to GP VI was not sequenced in the Rel sample. By looking at proteins which had significantly different average Rel T/C ratio compared to average Rel-MP T/C ratio, we can see which proteins primarily come from the MP. In other words, if the Rel T/C ratio is significantly greater than the Rel-MP T/C ratio, then that protein came from the MP generated by thrombin activated platelets. Appendix A shows the 11 proteins which were abundant in MP from thrombin activated platelets. One such protein was nidogen-1, which had previously been 129 identified in MP from thrombin activated platelets (Piersma, Broxterman et al. 2009). Interestingly nidogen-2, also identified previously (M\u00C3\u00A1jek, Reicheltov\u00C3\u00A1 et al. 2010) is exclusive to the Rel fraction and had a T/C ratio significantly over 1. We already know, based on the amount of proteins in MP (Table 3.1) and flow cytometry (Figure 3.10B and C) that collagen activated platelets produce a lot more MP than thrombin activated platelets. Appendix B shows the 21 proteins which were potentially abundant in MP from collagen activated platelets. These proteins include 14-3-3, clusterin, PF 4 and VWF. Overall, proteins from MP of collagen activated platelets makeup a significant proportion of total proteins released. Whether, these proteins are on the membrane or within the MP and whether MP release their contents at a later time, at a different location, is not yet clear. Potentially these MP can play an important role in health and disease state in the body. Next, we looked at the effects of activating platelets with a combination of thrombin plus collagen and compared the results to activating platelets with each agonist alone. We expected to see differences based on the fact that different signaling pathways can be initiated by the two agonists which could cross talk and produce a unique releasate proteome profile. 4.2.3 Thrombin versus thrombin plus collagen activated platelets The T/TC ratios for the 97 proteins identified in the Rel sample looks very different than the T/C ratios. There were only 6 proteins whose average T/C ratio was significantly greater than their average T/TC ratio (Table 3.3). These six proteins include prothrombin, PG Ib\u00CE\u00B1 and integrin \u00CE\u00B1IIb. We added equal amount of thrombin to the thrombin and thrombin plus collagen activated platelets so we expect the T/TC to be 1. Since it is 1.52, more prothrombin is generated from thrombin activated platelets as compared to thrombin plus collagen activated platelets. On the other hand, there could be differential binding of added thrombin to activated platelets in the different samples, so there is still a need for further biological investigation to find the exact cause of this observation. 130 In contrast, there were 67 proteins with an average T/C ratio significantly less than their average T/TC ratio (Table 3.3). The majority of these proteins are cytoskeletal proteins such as vinculin, filamin-A, tubulin and actin. In other words, it is possible that activating platelets with thrombin plus collagen prevents the increased release of many of these cytoskeletal proteins as compared to using collagen alone. A lot of the proteins mentioned here, such as talin-1 and \u00CE\u00B1-actin had average T/TC ratios of around 1. The top 20 biological process for the 22 proteins with T/TC ratio of less than 0.8, no longer had cell motion and actin filament-based process as was the case comparing thrombin and collagen activated platelets, again showing the reduced secretion of these cytoskeletal proteins by addition of thrombin to collagen for platelet activation (Figure 3.4B). Looking at Tables 3.4 and 3.5, we can see how the T/TC ratios differ from T/C ratios for each individual protein. Overall, there were lot more T/TC ratios which have red backgrounds signifying proteins which were more abundant in thrombin activated platelets as compared to thrombin plus collagen activated platelets. Proteins such as factor V, had about the same T/C ratio as T/TC, showing that thrombin activation alone causes as much factor V release as possible. On the other hand, proteins such as coagulation factor XIII which had T/C ratio of less than 0.8, now have T/TC ratios of around 1 or more. There were some key cell adhesion molecules which had T/TC ratios over 1.2. The two nidogen proteins were more abundant in the releasate when thrombin was used alone, showing that once collagen is bound to platelets, nidogen secretion decreases. Some of the signaling pathways initiated by collagen or thrombin alone which were responsible for secretion were not affected by addition of both agonists at the same time, while other signaling pathways were inhibited or had reduced intensity. Interestingly, the pattern of variability in T/C ratios between the three datasets looks very similar to the pattern of variability in the T/TC ratio, for key immune and inflammatory proteins such as PF4, C-C motif chemokine 5 and TGF\u00CE\u00B21. This makes a stronger case for heterogeneity in platelets from 131 different donors. Overall, thrombin alone causes a greater release of immune and inflammatory proteins than thrombin plus collagen, which could have great biological impact on immune response and monocyte stimulation. Moreover, by looking at the cellular compartment GO terms we can see how secretion is effected by addition of both agonists from platelet compartments such as \u00CE\u00B1-granules. Proteins which were almost exclusively found in the \u00CE\u00B1-granule, such as VWF or thrombospondin-1, that had average T/C ratios of around 1, had average T/TC ratio of over 1.2 (Table 3.5). It is more likely that collagen and thrombin on their own cause maximum granule secretion, whereas, adding thrombin and collagen together does not cause \u00CE\u00B1-granule secretion to the same extent. Perhaps, when platelets are exposed to more agonists, it signals that the platelet plug formation is nearly complete, so platelets do not secrete as many proteins to sustain coagulation, but perhaps the greatest change is seen in proteins belong to the cytoskeleton GO term. The average T/C ratio compared to average T/TC ratio for proteins grouped in cytoskeleton GO term looks very different. The ratio for most of the cytoskeletal proteins were now around 1 or had increased significantly as compared to their T/C ratio. Proteins such as talin-1, vinculin and filamin-A have ratios around 1. The addition of thrombin plus collagen prevents the release of cytoskeletal proteins as compared to activating platelets with collagen alone. Similarly, there were not as many membrane proteins secreted when platelets were activated with thrombin plus collagen as compared to activation with each agonist alone. Overall, many of the signaling mechanisms initiated by each agonist alone were greatly affected when the two were added together. In addition, to study the protein content of the MP generated through using thrombin plus collagen we again looked at the Rel T/TC ratios versus Rel-MP T/TC ratios. 132 4.2.4 Rel T/TC versus Rel-MP T/TC There was a greater abundance of proteins in the soluble fraction of Rel when platelets were activated with thrombin as compared to thrombin plus collagen. There were only 9 proteins with average Rel-MP T/TC less than 0.8 including GP Ib\u00CE\u00B1 and VI as well as three different isoforms of collagen. Appendix C shows the 41 proteins which were largely found in the MP when platelets were activated by thrombin plus collagen. These proteins were identified based on the fact that their average Rel T/TC ratio was significantly greater than their average Rel-MP T/TC ratio. Based on flow cytometry, we can see that there was a large concentrated population of MP generated when platelets were activated with thrombin plus collagen, which looks unique compared to the MP from collagen or thrombin activated platelets (Figure 3.10D). As part of these 41 proteins we found cytoskeletal proteins such as the actin proteins, talin-1, vinculin and tubulin. When platelets were activated with collagen alone, not only were these cytoskeletal proteins in the MP, but they were also abundant in the Rel-MP, so overall there were a lot more cytoskeletal proteins released as compared to thrombin or thrombin plus collagen activated platelets. There were about the same amount of cytoskeletal proteins released from thrombin or thrombin plus collagen activated platelets. The difference is that these cytoskeletal proteins are packaged into MP when platelets are activated by thrombin plus collagen, but when platelets are activated by thrombin alone these cytoskeletal proteins are in the soluble fraction of Rel. How this occurs is not clear from these experiments. The cytoskeletal proteins found in the thrombin plus collagen activated platelet Rel are mostly packed into MP along with other proteins important for coagulation and immune response. These data have given us a better understanding of the difference between the relative abundance of proteins when thrombin, collagen or thrombin plus collagen are used as agonists. These differences could have significant biological implications in platelet function both in hemostasis and in 133 cardiovascular disorder. To get a better understanding of the releasate proteins when platelets are activated during plaque rupture in patients with cardiovascular disorder we used quantitative proteomics to compare the relative abundance of releasate proteins when LPA or collagen were used as agonists. We used the same experimental setup as before with the use of dimethyl labels to perform these quantitative proteomics studies. 4.3 Quantitative proteomic analysis of releasate proteins from LPA, collagen and subthreshold collagen plus LPA activated platelets As mentioned there are individuals whose platelets do not get activated by LPA. We have also found one donor where washed platelets did not respond to LPA based on aggregometry trace and the very low amount of proteins secreted (data not shown). We performed aggregometry studies on the platelets from every donor before carrying out quantitative proteomics studies. To come up with an ideal concentration of LPA to use, we performed platelet aggregometry with different LPA dosages (Figure 3.6). A final concentration of 20 \u00C2\u00B5M of LPA was used for activation of platelets for quantitative proteomics study. As compared to collagen or thrombin, LPA activated platelets did not aggregate as much. As before, the proteins in the Rel and the Rel-MP from LPA, collagen or subthreshold level of collagen plus LPA activated platelets were quantitatively compared in three biological repeats (Figure 3.7). There were 180 proteins in common between the three Rel dataset when LPA or collagen was used as compared to 97 proteins when thrombin or collagen was used. When LPA was used a lot more different types of proteins were released. Since there were more proteins, the mass spectrometer has a better chance of finding high confidence peptides for protein quantification. However, almost all the 180 proteins quantified are also found in the releasate from thrombin activated platelets, when the Rel proteins from thrombin activated platelets are separated on SDS-PAGE and analyzed for identification 134 only. Due to in-solution digestion, the separation technique and instrumentation, only the abundant proteins were quantified. To compare the releasate proteins from LPA activated platelets versus collagen activated platelets, we analyzed the resulting LPA/collagen (L/C) ratio for each protein. 4.3.1 LPA versus collagen activated platelets Again, we used DAVID software to get an overview of the biological functions of the Rel proteins which were more abundant when LPA was used as compared to using collagen. The 135 proteins which were more abundant in Rel from LPA activated platelets as compared to collagen activated platelets, participate in processes such as cell motion and actin filament based process (Figure 3.26A). On the other hand, the 27 proteins with average L/C ratio less than 0.8 participate in response to wounding and coagulation (Figure 3.8B). From these graphs we can conclude that proteins released from LPA activated platelets participate in functions very different to that of collagen activated platelets and that the proteome profile for the two are very different. The same GO analysis as per section 4.2.1 was performed on the 180 quantified proteins found in common between the three Rel datasets to identify the role of releasate proteins and identify their location in the platelets. The proteins were grouped based on their biological function and cellular compartment GO terms (Table 3.8 and 3.9). Not surprisingly, response to wounding and related GO terms such as hemostasis were the top biological functions found from the list. Based on the L/C and T/C ratio for factor V and factor XIII, we see that thrombin activated platelets release the most factor V and LPA activated platelets release the most factor XIII. Since factor V aids in thrombin generation, this would makes sense for a positive feedback, and since LPA does not rely on thrombin, factor V is not released as much. As evident from L/C and T/C ratios for proteins such as plasminogen, proteins important for clot retraction are secreted in about equal abundance when they are activated with thrombin, collagen or LPA. Overall, we see that LPA activated platelets are also capable of releasing proteins which can play a role in hemostasis and 135 some of these proteins are actually more abundant in Rel from LPA activated platelets. The next biological function we focused on due to its importance in cell to cell interactions and platelet-monocyte formation is cell adhesion GO term. There were interesting proteins which had a L/C ratio greater than 1.2 such as \u00CE\u00B2-parvin and proto-oncogene tyrosine-protein kinase Src. The mechanisms by which LPA causes greater secretion of these proteins is not yet clear, but again they also have been identified in Rel from thrombin receptor activating peptide (TRAP) activated platelets (Piersma, Broxterman et al. 2009). Furthermore, based on the L/C and T/C ratios for GP Ib and GP V it seems that even proteins in the same complex do not secrete at the same level. The data suggests that there are different cell adhesion molecules which are released in higher abundance when LPA is used as an agonist as compared to using thrombin or collagen. We also looked at the proteins which belong to inflammatory and immune response GO terms to study the impact of using LPA in the context of monocyte stimulation and cardiovascular disease. There were some surprising proteins which we found that belong to immune and inflammatory response GO terms including \u00CE\u00B1-synuclein. \u00CE\u00B1-Synuclein plays a role in immune response and was substantially more abundant in Rel from LPA activated platelets. This protein has not been identified in platelet releasate samples from TRAP activated platelets, but it is known to be in platelets based on our work looking at total platelet lysate (data not shown) and others. \u00CE\u00B1-Synuclein is not very abundant in platelets and varies in abundance from different donors (Michell, Luheshi et al. 2005) and we see this varied abundance in our datasets. \u00CE\u00B1-Synuclein plays a central role in Parkinson\u00E2\u0080\u0099s disease, so why it is found in platelets and why LPA activation causes so much secretion of this protein that we can quantify it using quantitative mass spectrometry is not very clear. In addition, we also found proteins in these GO terms which we expected to see such as complement factor proteins and cytokines such as PF4. 136 Based on the L/C ratios for complement C3 and complement factor H we can reason that the two complement proteins are governed by different mechanisms of secretion which could have a biological impact since they do play opposing role in complement cascade. Moreover, based on the observed T/C and L/C ratios we could conclude that PF4 is governed through a different mechanism of secretion as compared to TGF\u00CE\u00B21 or C-C motif chemokine 5. For the most part, the L/C ratios between the three datasets are consistent, but we see the greatest variability in proteins grouped into immune and inflammatory response. P-selectin seems to have the greatest variability between the three datasets with dataset 3 secreting the most P-selectin due to LPA activation. In general, collagen and LPA activated platelets secrete the greatest amount of P-selectin as compared to thrombin activated platelets. This can play an important role in platelet-monocyte aggregate formation since P-selectin is the key protein for monocyte stimulation. With a better idea of the biological role of these secreted proteins, we shifted our focus on to finding out which part of the platelet these proteins are secreted from. Using DAVID software, we grouped the 180 proteins identified between the three datasets into cellular compartment GO terms (Table 3.9). There were \u00CE\u00B1-granule proteins such as heat shock proteins which were not in the previous Rel dataset from section 4.2.1. The extent of secretion seems to be the same based on the ratios for abundant \u00CE\u00B1-granule proteins such as thrombospondin-1 and VWF. The difference in L/C ratios and the fact that there are proteins in this Rel list which was not seen in the previous one, suggest that perhaps not all \u00CE\u00B1-granules have the same protein content, or that they differentially release their content depending on the agonist used. Moreover, there are some important proteins responsible for coagulation that are stored in the \u00CE\u00B1-granules and seem to be more abundant when one agonist is used over another. 137 Fibrinogen, an important protein for coagulation, has an average L/C ratio of 0.32. Similarly the T/C ratio for fibrinogen is around 0.05, which means there were even less fibrinogen in Rel from thrombin activated platelets than LPA activated platelets. Fibrinogen binds activated integrin \u00CE\u00B1IIb\u00CE\u00B23, and upon cleavage by thrombin, forms a stable fibrin clot to help stop bleeding. Other groups crudely comparing thrombin activated platelets to collagen activated platelets have also seen this difference (Coppinger, O'Connor et al. 2007; M\u00C3\u00A1jek, Reicheltov\u00C3\u00A1 et al. 2010). A possible explanation suggests that the presence of thrombin in the solution cleaves soluble fibrinogen into fibrin clots, which due to their higher weight will be spun out of the solution and not undergo digestion and identification in the thrombin activated platelet samples. However, the change we see in the T/TC ratio does not account for this explanation (Table 3.5). Furthermore, collagen seems to cause much more fibrinogen release than LPA activated platelets too, where theoretically no fibrin is formed. There could be a biological purpose to collagen causing the most fibrinogen release, since upon vessel damage collagen is exposed to platelets. In addition, the level of activated integrin \u00CE\u00B1IIb, a receptor for fibrinogen, is greatly increased on the surface of thrombin activated platelets (Figure 3.10G). This could be due to the fact there isn\u00E2\u0080\u0099t much fibrinogen available to bind to integrin \u00CE\u00B1IIb, making PAC-1 binding possible. A lot of the proteins found in the Rel dataset have not yet been identified in the \u00CE\u00B1-granule. GO term \u00E2\u0080\u0098cytoskeleton\u00E2\u0080\u0099 is where Rel from LPA activated samples particularly standout. Overall, LPA platelet activation caused the release of cytoskeletal proteins in much greater abundance than collagen activation. All of these proteins have been identified in platelet releasate before, and almost all of them in TRAP induced platelet releasate (Piersma, Broxterman et al. 2009), but LPA simply causes the secretion of these proteins in greater abundance. To get a better of understanding of protein abundance in different compartment of platelet Rel, specifically MP, we also studied the Rel-MP for LPA and collagen activated platelets. 138 4.3.2 Rel L/C versus Rel-MP L/C There were 148 proteins in common between the three Rel-MP datasets, which shows that perhaps MP from LPA activation have some unique proteins which were not found in the soluble fraction of the releasate. There were 129 proteins in common between the Rel and Rel-MP datasets, with the Rel dataset having 51 unique proteins versus only 19 proteins unique to the Rel-MP dataset (Figure 3.7). The proteins unique to Rel-MP are mostly different isoforms of the proteins already identified in the Rel sample such as 14.3.3\u00CE\u00B8 or heat shock protein \u00CE\u00B2-1. The proteins unique to the Rel sample could potentially be associated with the MP. Appendix G shows the unique proteins for the Rel sample. There were a lot of proteins which could be potentially abundant in MP from LPA activated platelets including cytoskeletal proteins. Proteins such as integrin \u00CE\u00B16, \u00CE\u00B23 and \u00CE\u00B1IIb as well as C-C motif chemokine 5 are unique to the Rel sample. This does not necessarily mean that they are only in the MP, but it could mean that they are most abundant in the MP, and since their L/C ratio is small, they are mostly in the MP from collagen activated platelets. Having MP with functional adhesion and chemokine molecules could have a significant biological impact, since they can flow to other locations and interact with other blood cells and potentially activate them. By comparing how the L/C ratios change for a given protein in the Rel versus the Rel-MP, we could make comments about the abundance in the MP versus in the soluble fraction of the total releasate. Appendix F shows the proteins that could potentially be abundant in MP generated from LPA platelet activation. One interesting protein in the list is P-selectin, which was found to be very abundant in MP. Based on flow cytometry studies (Figure 3.10I), there was more P-selectin expression on the surface of MP from LPA activated samples as compared to other agonists, which corresponds very well with the proteomic data. This could have great implications for platelet-monocyte aggregate formation 139 and cardiovascular disease, since MP from LPA activated platelets can attach to monocytes and stimulate them. There was also PF4 which is abundant in these MP, and perhaps upon the rupture they can activate other blood cells leading to inflammatory and immune response. Moreover, to study whether secretion was affected by addition of a subthreshold concentration of collagen to LPA for platelet activation, we analyzed the CL/C ratios in the same manner as before. 4.3.3 Subthreshold collagen plus LPA versus collagen activated platelets In general, addition of subthreshold collagen to LPA reduces the total amount of proteins released. However, overall the change in ratios observed by adding subthreshold collagen to LPA does not make a huge impact in the overall proteome profile seen. In other words a given protein that had an average L/C ratio of greater than 1.2 also had an average CL/C ratio greater than 1.2. The change seen was not as dramatic as the change seen when thrombin and collagen were added together to activated platelets. This is reflected when we look at Figure 3.9, where the top 20 biological process GO terms for proteins with CL/C ratios greater than 1.2 (Figure 3.9A) or less than 0.8 (Figure 3.9B) are shown. Overall these graphs look very similar to those in Figure 3.8 where C/L protein ratios were analyzed. This is important because while priming platelets with subthreshold collagen could make platelet activation possible with lower concentrations of LPA the general biological significance of released proteins is the same. Overall, from the quantitative proteomics data, activating platelets with subthreshold collagen plus LPA did not generate MP which were much different in their protein composition than LPA activated platelets. By comparing the side versus forward light scatter plots in Figure 3.10E and F, the MP population seems to be at the same place. From the evidence, adding subthreshold collagen to LPA for platelet activation does not greatly affect MP production. Overall, activation of platelets by LPA seems to greatly affect the shape of platelets and MP generated based on Figure 3.10 E and F. The cause of the increased observed size is not readily apparent from the data. The MP generated from LPA 140 activated platelets are also a different size than MP generated by other agonists. From the proteomics data there were unique, mainly cytoskeletal proteins, which could potentially be present in high abundance in MP generated from LPA activated platelets (Appendix G). Overall, the releasate proteins from LPA activated platelets could potentially have great biological impact in the context of cardiovascular disease. To study the biological significant of the releasate proteins when the different agonists are used and whether the observed ratio differences have any biological impact we performed a series of migration assay studies. We counted the number of THP-1 cells which migrated towards the Rel, Rel-MP or the MP from platelets activated with the various agonists after 6 hours of incubation. 4.4 Functional biological studies using platelet releasate We studied the effects of different compartments of platelet releasate on THP-1 cell migration. Overall, for Rel, the number of THP-1 cells migrating stayed the same no matter which agonist was used (Figure 3.11A). The number of THP-1 cells migrating toward Rel from thrombin plus collagen activated platelets has the greatest standard deviation which corresponds to the proteomics data. In general the Rel-MP causes less THP-1 migration than the Rel, therefore from these data alone we can conclude that MP play a role in causing THP-1 migration. The MP from thrombin, LPA or subthreshold collagen plus LPA activated platelets caused significantly less THP-1 migration compared to the corresponding Rel samples. On the other hand, MP from collagen or thrombin plus collagen activated platelets caused almost as much THP-1 migration as the Rel. These MP can travel around the blood and stimulate other blood cells and increase inflammatory and immune response. They can also stimulate monocytes and prime them for transmigration across the endothelial layer. Even addition of subthreshold collagen seems to generate MP which caused greater THP-1 cell migration as compared to MP from platelets which had been activated with LPA alone. 141 To target specific proteins which could cause THP-1 cell migration, we used an antibody against PF4 to inhibit its actions. PF4 is a known chemokine which causes THP-1 cell migration. Figure 3.11B shows a concentration-dependent blocking of THP-1 cell migration by Rel from thrombin activated platelets. By using 20 \u00C2\u00B5g of antibody, effectively all migration is blocked. However, based on these experiments, PF4 is not the only chemokine which causes THP-1 cell migration. Using 25 \u00C2\u00B5g of purified PF4 did not result in nearly as much THP-1 cell migration as using Rel from thrombin activated platelets. Only 1 mL of Rel from 300 x 106 thrombin activated platelets was used for THP-1 cell migration, so most likely there was less than 25 \u00C2\u00B5g of PF4 in the Rel. Therefore, there are other proteins in the Rel sample which work with PF4 and contribute to THP-1 cell migration. But once enough PF4 is blocked, all migration stops despite other proteins present. Proteins which could contribute to causing THP-1 cell migration include TGF\u00CE\u00B21 and C-C motif chemokine 5. These data show that platelet releasates are capable of stimulating THP-1 cells and causing transmigration. Next, we shifted our focus to study the target of platelet releasate proteins, namely the THP-1 cells. To further study the effect of platelet releasate on THP-1 cells we incubated THP-1 cells with releasate from thrombin activated platelets for 6 and 24 hours. We confirmed the presence of P-selectin in the platelet releasate which is an important protein that binds to monocytes for initiating platelet- monocyte aggregate formation. By analyzing the proteome changes in THP-1 cells which take place upon stimulation with platelet releasate we can gain an insight into the signaling pathways which are responsible for causing THP-1 transmigration and platelet-monocyte aggregate formation. However, to have a better understanding of the model system we are working with we first generated a reference dataset for THP-1 cells. 142 4.5 Reference membrane and global proteome dataset for THP-1 cells As mentioned, to perform SILAC quantitative proteomics studies, THP-1 cells were used as model a system for monocytes. To understand the signaling mechanisms which take place upon addition of releasate from thrombin activated platelets to THP-1 cells, we need as much proteome coverage as possible. Since cells interact with their surroundings through their membrane proteins, an in-depth analysis of these proteins is essential to developing a signaling map. To this end, two protocols were tested in their abilities to enrich for membrane proteins (protocols A and B) (Figure 3.12). In proteomics, depleting unwanted proteins is just as important as enriching for proteins of interest (Figure 3.13). Therefore, we picked protocol A, which used a sucrose bilayer and ultracentrifugation as oppose to protocol B which used a continuous sucrose gradient, for quantitative membrane proteome analysis, due to its ability to enrich for membrane proteins while depleting unwanted cytosolic ones. We generated a reference global and membrane proteome dataset to identify as many proteins in THP-1 cells as possible. When compared to the global protein dataset, we found 548 proteins unique to the membrane dataset (Figure 3.14), which could potentially be membrane proteins which weren\u00E2\u0080\u0099t identified in the global dataset. To compare the type of proteins we identified from the global and membrane dataset we performed GO analysis to look for cellular compartment GO terms for each dataset (Figure 3.14E and F). From this analysis we saw a great depletion of cytosolic proteins and nuclear proteins. At the same time, we a saw lot of membrane proteins identified from the organelle and plasma membrane of THP-1 cells in the membrane dataset. Overall, by using protocol A to enrich for membrane proteins, we were able to identify more membrane proteins with a higher confidence as compared to using whole cell lysate. These datasets show the proteins which we can potentially identify through quantitative proteomics when studying the effects of platelet releasate on THP-1 cells. However, before starting our proteomic studies we used flow cytometry to confirm key proteins 143 interactions between Rel proteins and THP-1 cells. We used thrombin as an agonist since thrombin is a strong and widely used agonist for platelet activation studies both in vivo and in vitro. The local concentration of platelets at the site of plaque rupture is found to be around 10 x 108 platelets/ml (in comparison, physiological concentration of platelets is 150-400 x 106 platelets/ml) (Breland, Michelsen et al. 2010). We resuspended our washed platelets to platelet concentrations found around plaque rupture to mimic disease conditions. The platelets for these experiments were activated for 90 minutes to account for proteins which are synthesized through translation by platelets, so we get more complete set of protein releasate which is close to what happens in the body. Here we used Rel from 4 donors to overcome donor dependent responses. THP-1 cells were incubated with total platelet releasate at a concentration of 106 THP-1 cells/ml of platelet releasate, to study the proteome change in THP-1 cells. 4.6 THP-1 cell stimulation by platelet releasate We utilized flow cytometry, to study the effects of Rel from thrombin activated platelets on THP-1 cells. Upon addition of releasate from thrombin activated platelets to THP-1 cells, Mac-1 (integrin \u00CE\u00B1M) antibody binding to THP-1 cells decreased (Figure 3.15). Mac-1 antibody binds to Mac-1 on cell surfaces, so this decrease could occur due to Mac-1 shedding from the membrane or the antibody not binding to its receptor. Since this decrease did not seem to be time dependant the more likely cause of this decrease is Mac-1 antibody not binding. In other words there were one or more proteins in the platelet releasate which bound to Mac-1 preventing the antibody from binding. It has been previously shown that glycoprotein (GP) Ib\u00CE\u00B1 is a counter receptor for Mac-1 and its binding to leukocytes is important for their recruitment and inflammatory response (Simon, Chen et al. 2000). In addition other proteins such as factor X and intercellular adhesion molecule 1 have the potential to bind to Mac-1. Since we identified GP Ib\u00CE\u00B1 in the Rel from thrombin activated platelets, the most likely cause of the decrease binding is that Gb Ib\u00CE\u00B1 binds Mac-1 preventing the antibody binding. 144 Another important interaction which takes place for the initiation of platelet-monocyte aggregate formation is P-selectin binding from platelets to PSGL-1. Through flow cytometry analysis we detected P-selectin on THP-1 cell surface (Figure 3.16). P-selectin binding on THP-1 surface was decreased at 24 hours compared to 6 hours, which wasexpected since platelet monocyte aggregates eventually break apart in vivo and shed P-selectin. Through these flow cytometry studies we confirmed key interactions between the proteins from the releasate and THP-1 cells which also take place during platelet monocyte aggregate formation. After confirmation of key protein interactions, we performed quantitative global and membrane proteomic analysis to find the proteins which are responsible for THP-1 stimulation by platelet releasate at 6 and 24 hours of stimulation. 4.6.1 THP-1 cell global proteome changes THP-1 cells grown in light SILAC media were treated with vehicle control while the THP-1 cells grown in heavy SILAC media were treated with platelet releasate for 6 and 24 hours. Using MaxQuant, we identified the proteins and calculated their normalized heavy to light SILAC ratio. As expected based on the proteins found from the reference datasets, the majority of proteins did not significantly change by 6 or 24 hours of stimulation. In each dataset there is a Gaussian protein ratio distribution (Figure 3.17). By 6 hours of stimulation, there were proteins which were significantly underexpressed as well as proteins which were significantly overexpressed. Some of these proteins display a biphasic response i.e., if they were underexpressed at 6 hours they increased their expression levels by 24 hours and vice versa (Figure 3.19B and Figure 3.20B). Among these proteins were ribosomal proteins and transcriptional activators such as transcriptional activator protein pur-\u00CE\u00B1. Some of the proteins found to be overexpressed at 6 hours were involved in inflammatory response (Figure 3.21). This suggests that by 6 hours of stimulation, THP-1 cells can be recruited and stimulated through interaction with proteins found in the Rel, in accordance with our transmigration studies. 145 The 6 proteins found to be significantly underexpressed by 24 hours (Figure 3.33) do not seem to play an important role in platelet-monocyte aggregate formation. On the other hand, 28 proteins were significantly overexpressed by 24 hours. These proteins, for the most part, did not display a biphasic response, rather a group of proteins had delayed response and only increased in expression level at 24 hours (Figure 3.23A). We analyzed the proteins found using Ingenuity Pathway Analysis (IPA) software to look for their role in immune response. Some of the proteins which were overexpressed at 24 hours play important roles in binding to platelets, immune response and activation of monocytes such as annexin proteins, STAT3 and PECAM1 (Figure 3.24). The proteins that changed significantly were found to play important roles in platelet-monocyte aggregate formation and monocyte activation. To study the proteins responsible for THP-1 stimulation at the surface of the cells and the proteins responsible for extracellular communication, we performed the same SILAC quantitative studies using THP-1 protein samples that were enriched for membrane proteins using protocol A. 4.6.2 THP-1 cell membrane proteome changes Protocol A was implemented for proteins from THP-1 cells treated with vehicle control (grown in light SILAC media) or Rel (grown in heavy SILAC media) to enrich for membrane proteins. Once again as expected from the reference membrane dataset, the normalized heavy to light protein ratio had a Gaussian distribution for the \u00E2\u0080\u009C6 hours\u00E2\u0080\u009D and \u00E2\u0080\u009C24 hours\u00E2\u0080\u009D datasets (Figure 3.25). Similar to global proteins, some of the membrane proteins also displayed a biphasic response (Figure 3.15). For example, cathepsin G, a protease, was significantly up regulated by 6 hours, but at 24 hours the expression level was levelled. This could be due to either protein degradation or secretion. To explain the biphasic response seen for some of the proteins, both for the membrane and global proteome, identification and analysis of proteins secreted by THP-1 cells can be very useful. In addition, by studying the secreted proteins we can shed light on some of the feedback mechanisms which are 146 important to platelet monocyte aggregate formation. Interestingly, cathepsin G is a very potent platelet activator (LaRosa, Rohrer et al. 1994). If released, cathepsin G further initiates platelet-monocyte aggregate formation. By 24 hours of stimulation, THP-1 cells have had more time for protein synthesis through transcription and translation, so we see even a greater change in ratios for some of the proteins. By 24 hours of stimulation only two proteins were significantly underexpressed as compared to 10 proteins which were overexpressed (Figure 3.28), a theme which was evident in the global proteome also. IPA identified the proteins which play a role in cell adhesion and leukocyte extravasation signaling, a pathway important for monocyte stimulation and transmigration. Most of these proteins were overexpressed at 24 hours, such as platelet endothelial cell adhesion molecule (PECAM-1), integrin \u00CE\u00B21 and leukosialin (Figure 3.29). Indeed, with using our model system to study platelet-monocyte aggregate formation, we confirmed the importance and contribution of key proteins involved in pathways such as leukocyte extravasation signaling, which help form platelet-monocyte aggregates. Despite our model\u00E2\u0080\u0099s shortcomings when compared to platelet-monocyte aggregate formation in vivo, such as the absence of shear flow which contributes to platelet releasate dilution and impact of other cells such as red blood cells, we still managed to identify key proteins such as integrins which play an important role in platelet- monocyte aggregate formation. Overall, the results from the global and membrane proteome analysis suggest the induction of a more adhesive, pro-inflammatory state in monocytes. These findings support our hypothesis that platelet releasate is sufficient to stimulate monocytes and prime them for an immune response and migration even without direct binding to activated platelets. So far by using our bioinformatics tools, we managed to show events already known to take place in platelet-monocyte aggregate formation, thus validating our model system. Now that we know our model system can 147 confirm events known to take place during platelet-monocyte aggregate formation, we can look more in-depth at additional proteins to identify their role in platelet-monocyte aggregate formation. We can also look to investigate the biological impact of each of our proteins identified as potential biomarkers, in a more biological setting by looking at platelet-monocyte aggregates in patients with cardiovascular disease. One protein we identified that increased in expression level at 6 hours of stimulation was cathepsin G. Cathepsin G has not been significantly implicated in the context of platelet-monocyte aggregate formation. We can use in vivo experiments, or isolate platelets and monocytes in vitro and look at the secretion of cathepsin G and its contribution to feedback signaling for the production of additional platelet-monocyte aggregates. In addition, we can look at the role of leukosialin in monocyte stimulation. For example, we can block leukosialin and look at the biological consequences in transmigration and platelet-monocyte aggregate formation. The same sort of studies can be done with integrin \u00CE\u00B21 and platelet endothelial adhesion molecule 1. Now that we have identified potential biomarkers and proteins that change in expression level in THP-1 cells by stimulation with platelet releasate, we can block P-selectin binding to PSGL-1 and look at the impact of THP-1 stimulation without P-selectin binding. Using our model system, we identified a number of candidate biomarkers such as PECAM-1, integrin \u00CE\u00B21, STAT1 and annexin-5 which are responsible for transducing signals for platelet- monocyte aggregate formation and priming THP-1 cells for transmigration. By performing follow up functional experiments, we can elucidate the role of some of these candidate biomarkers in platelet- monocyte aggregate formation and progression of cardiovascular disease. 148 5 Conclusions By using quantitative proteomics we were able to address all the hypotheses put forward for this thesis. The first part of this thesis focused on the content of the platelet releasate when using different agonists. We analyzed the total releasate as well as the soluble fraction of releasate. We used a model system using washed platelets and quantitative proteomics based on dimethyl labeling to compare the proteome profile of platelet releasate when using thrombin, collagen or LPA as agonists. Despite our model\u00E2\u0080\u0099s shortcomings, such as the absence of shear flow or serum proteins, we managed to identify proteins previously seen in platelet releasates by other groups as well as finding key proteins involved in immune and inflammatory response which differed in abundance depending on the agonist used. Using our model system we identified key proteins such as P-selectin that could be involved in platelet- monocyte aggregate formation which were released to different extents depending on the agonist used. We also identified important proteins involved in coagulation and maintaining platelet plugs such as fibrinogen which were released to different extents depending on the agonist used. Overall, we saw that abundant proteins in platelet releasate when using different agonists play a unique biological role in the body. In addition, there are also donor dependant variation and response to the agonists, but the root cause of this variation is still unclear and whether the same thing occurs in vivo is still open for investigation. To this day, no one knows the exact mechanisms leading to platelet secretion. Italiano et al suggests that platelets differentially store and release their granule cargo in a controlled manner (Whiteheart 2011). There are two aspects to this hypothesis, firstly that there is differential storage in platelets and second that there is differential releasing of protein content, neither of which has been fully addressed. Kaykowski et al has shown, using high-resolution immunofluorescence microscopy, that there is a Gaussian cargo distribution in \u00CE\u00B1-granules, and essentially all \u00CE\u00B1-granules have the same 149 content. However, there is segregation of cargo within the \u00CE\u00B1-granules (Whiteheart 2011). The second aspect of the hypothesis, the differential releasing of protein content, has yet to be fully studied and has mainly focused on \u00CE\u00B1-granule and dense granule release. Most of the studies have used a single agonist at a single time point, just to see the proteins released. Here we used quantitative proteomics to compare the releasate of six different agonists and showed that indeed, there is differential release based on the agonist used. The clustering of proteins within \u00CE\u00B1-granules could be key to the differential releasate seen in the experiments in this thesis. Perhaps \u00CE\u00B1-granules do not simply fuse to the plasma membrane and release their content. Depending on the agonist used there could be MP that form within \u00CE\u00B1-granules, some of which never get released by activation using for example thrombin. We also showed that there were major differences in MP from platelets activated by the different agonists, both in quantity and composition. The MP play a key role in the content of releasate as seen from the THP-1 migration studies. These quantitative proteomic studies leave little doubt that platelets differentially release their content depending on the agonist used. It is very likely that differential secretion, has great biological impact and it is important for the normal function of platelets. Furthermore, activating platelets with thrombin plus collagen is not simply the sum of the final outcome of thrombin activated platelets plus collagen activated platelets. Using two agonists leads to the activation of unique signaling mechanisms which lead to releasate with a characteristic proteome profile. Platelets seem to be smart delivery devices which respond based on the unique environment they are in. We can take advantage of this unique response to different agonists, to come up with drug targets to prevent unwanted platelet activation and platelet releasate, while not affecting platelets\u00E2\u0080\u0099 important role in hemostasis. For example, we can find how to stop platelet response to LPA or how to prevent LPA activated platelets from releasing MP with high P-selectin expression. This can be key to when plaques rupture and platelets are activated by mox-LDL products such as LPA in patients with cardiovascular disorder. 150 Using our model system, we were able to identify candidate proteins which showed agonist dependent secretion. These proteins such as platelet factor 4, transforming growth factor \u00CE\u00B21 or P- selectin play an important role in monocyte stimulation, platelet-monocyte aggregate formation and progression of cardiovascular disease. Whether the same thing occurs in vivo and the extent at which differential secretion occurs is not yet clear. However, with our experiments we were able to generate a starting point for investigating differential secretion in vivo and provided target proteins to look for such as fibrinogen which play an important role in coagulation. We can perform the same types of experiments in blood and then look at specific proteins which we identified in our proteomics studies, by other techniques such as immuno blot or multi reaction monitoring mass spectrometry. We can use imaging techniques such as immunofluorescence microscopy or high power light microscopy to look at how MP are formed or why thrombin activated platelets produce fewer microparticles. We can follow the fate of proteins such as nidogen which were abundant in MP from thrombin activated platelets, using immunofluorescence microscopy to look at how MP are generated from platelets activated using thrombin. Although our model system is very different to the conditions under which platelets are normally activated in the body, by using our model system, we were able to identify key proteins which could potentially play an important role in monocyte stimulation and progression of cardiovascular disease. To understand the biological significance of our observations we needed to perform additional functional experiments such as using antibodies against additional chemokines, and migration assays. From these experiments we identified key proteins which play roles in monocyte stimulation and transmigration, such as P-selectin. The second part of this thesis focused on the ability of platelet releasate to stimulate monocytes. We successfully used THP-1 cells as a model system for monocytes and incubated them with releasate from thrombin activated platelets for 6 and 24 hours. We implemented the use of SILAC quantitative proteomics to discover the proteins whose expression changes upon addition of platelet 151 releasate. We confirmed that P-selectin from the releasate binds to the surface of THP-1 cells. Based on the overexpression of key adhesion and pro-inflammatory proteins such as integrin \u00CE\u00B21 and leukosialin we can conclude that indeed, the proteins in platelet releasate are sufficient to induce a more pro- inflammatory state in THP-1 cells and prime them for transmigration across the endothelial layer. However, there is still a lot more work to be done in order to elucidate the signaling mechanisms by which monocytes are stimulated and platelet-monocyte aggregates are formed. Our model system has provided key candidate proteins which could potentially play important roles in platelet-monocyte aggregate formation. Now that we have validated our model system we can look for proteins that could be involved in platelet-monocyte aggregate formation such as cathepsin G or leukosialin and perform follow up functional studies to elucidate their contribution to platelet-monocyte aggregate formation. Since we saw that different platelet agonists lead to releasate with unique proteome profile, we should add releasate from LPA activated platelets to THP-1 cells and study the effects on THP-1 cells using the same SILAC strategy. These experiments can give more insight into the events which take place upon plaque rupture and platelet activation which leads to monocyte stimulation. Indeed, we have identified biomarkers which can eventually be targeted to prevent platelet-monocyte aggregate formation and stop unwanted monocyte stimulation and transmigration. Eventually, these studies can help design drugs that can prevent the progression of cardiovascular disease and prevent artery blockage by thrombi or platelet-monocyte aggregates after plaque rupture. 152 References Ahn, K. C., A. J. Jun, et al. (2005). \"Preferential binding of platelets to monocytes over neutrophils under flow.\" Biochemical and biophysical research communications 329(1): 345-355. Amisten, S., O. \u00C3\u0096. Braun, et al. (2008). \"Gene expression profiling for the identification of G-protein coupled receptors in human platelets.\" Thrombosis research 122(1): 47-57. Ardlie, N. G., M. L. Selley, et al. (1989). \"Platelet activation by oxidatively modified low density lipoproteins.\" Atherosclerosis 76(2): 117-124. Atarashi, K., T. Hirata, et al. (2005). \"Rolling of Th1 cells via P-selectin glycoprotein ligand-1 stimulates LFA-1-mediated cell binding to ICAM-1.\" The Journal of Immunology 174(3): 1424-1432. Authi, K. S., B. J. Evenden, et al. (1986). \"Metabolic and functional consequences of introducing inositol 1, 4, 5-trisphosphate into saponin-permeabilized human platelets.\" Biochemical Journal 233(3): 707. Aviram, M. and J. G. Brook (1983). \"Platelet interaction with high and low density lipoproteins.\" Atherosclerosis 46(3): 259-268. Azim, A., K. Barkalow, et al. (2000). \"Activation of the small GTPases, rac and cdc42, after ligation of the platelet PAR-1 receptor.\" Blood 95(3): 959-964. Bandoh, K., J. Aoki, et al. (2000). \"Lysophosphatidic acid (LPA) receptors of the EDG family are differentially activated by LPA species:: Structure-activity relationship of cloned LPA receptors.\" FEBS letters 478(1-2): 159-165. Banno, Y., Y. Yada, et al. (1988). \"Purification and characterization of membrane-bound phospholipase C specific for phosphoinositides from human platelets.\" Journal of Biological Chemistry 263(23): 11459- 11465. Bar-Shavit, R., A. Eldor, et al. (1989). \"Binding of thrombin to subendothelial extracellular matrix. Protection and expression of functional properties.\" Journal of Clinical Investigation 84(4): 1096. Barry, O. P., M. G. Kazanietz, et al. (1999). \"Arachidonic acid in platelet microparticles up-regulates cyclooxygenase-2-dependent prostaglandin formation via a protein kinase C/mitogen-activated protein kinase-dependent pathway.\" Journal of Biological Chemistry 274(11): 7545-7556. 153 Baykal, D., J. F. Schmedtje, et al. (1995). \"Role of the thrombin receptor in restenosis and atherosclerosis.\" Am J Cardiol 75(6): 82B-87B. Bensing, B. A., C. E. Rubens, et al. (2001). \"Genetic loci of Streptococcus mitis that mediate binding to human platelets.\" Infection and immunity 69(3): 1373-1380. Berger, G., D. W. Hartwell, et al. (1998). \"P-selectin and platelet clearance.\" Blood 92(11): 4446-4452. Bergmeier, W., C. L. Piffath, et al. (2004). \"Tumor Necrosis Factor-\u00CE\u00B1\u00E2\u0080\u0093Converting Enzyme (ADAM17) Mediates GPIb\u00CE\u00B1 Shedding From Platelets In Vitro and In Vivo.\" Circ Res 95(7): 677-683. Bergmeier, W., T. Rabie, et al. (2004). \"GPVI down-regulation in murine platelets through metalloproteinase-dependent shedding.\" THROMBOSIS AND HAEMOSTASIS-STUTTGART- 91(5): 951- 958. Berliner, J. A., M. Navab, et al. (1995). \"Atherosclerosis: basic mechanisms: oxidation, inflammation, and genetics.\" Circulation 91(9): 2488-2496. Berridge, M. J. and C. Taylor (1988). Inositol trisphosphate and calcium signaling, Cold Spring Harbor Laboratory Press. Bessis, M. and R. I. Weed (1973). \"Living blood cells and their ultrastructure.\" Bevers, E. M., P. Comfurius, et al. (1983). \"Changes in membrane phospholipid distribution during platelet activation.\" Biochimica et biophysica acta 736(1): 57. Blanchard, R. A., B. C. Furie, et al. (1981). \"Acquired vitamin K-dependent carboxylation deficiency in liver disease.\" New England journal of medicine 305(5): 242-248. Blomb\u00C3\u00A4ck, B., M. Blomb\u00C3\u00A4ck, et al. (1967). \"Structure of N-terminal fragments of fibrinogen and specificity of thrombin.\" Nature 215: 1445-1448. Bocan, T., T. Schifani, et al. (1986). \"Ultrastructure of the human aortic fibrolipid lesion. Formation of the atherosclerotic lipid-rich core.\" Am J Pathol 123(3): 413. Bode, A. and C. Knupp (1994). \"Effect of cold storage on platelet glycoprotein Ib and vesiculation.\" Transfusion 34(8): 690-696. 154 Bode, A., S. Orton, et al. (1991). \"Vesiculation of platelets during in vitro aging.\" Blood 77(4): 887. Boersema, P. J., R. Raijmakers, et al. (2009). \"Multiplex peptide stable isotope dimethyl labeling for quantitative proteomics.\" Nature protocols 4(4): 484-494. Bouchon, A., J. Dietrich, et al. (2000). \"Cutting edge: inflammatory responses can be triggered by TREM- 1, a novel receptor expressed on neutrophils and monocytes.\" The Journal of Immunology 164(10): 4991. Boulanger, C. M., A. Scoazec, et al. (2001). \"Circulating microparticles from patients with myocardial infarction cause endothelial dysfunction.\" Circulation 104(22): 2649-2652. Brass, L., D. Manning, et al. (1997). \"Signaling through G proteins in platelets: to the integrins and beyond.\" Thromb Haemost 78(1): 581. Brass, L., R. Vassallo Jr, et al. (1992). \"Structure and function of the human platelet thrombin receptor. Studies using monoclonal antibodies directed against a defined domain within the receptor N terminus.\" Journal of Biological Chemistry 267(20): 13795-13798. Brass, L. F. (2003). \"Thrombin and Platelet Activation*.\" Chest 124(3 suppl): 18S-25S. Brass, L. F., T. J. Stalker, et al. (2007). \"Signal transduction during platelet plug formation.\" Platelets: 319- 346. Breland, U. M., A. E. Michelsen, et al. (2010). \"Raised MCP-4 levels in symptomatic carotid atherosclerosis: an inflammatory link between platelet and monocyte activation.\" Cardiovascular research 86(2): 265. Browder, T., J. Folkman, et al. (2000). \"The hemostatic system as a regulator of angiogenesis.\" Journal of Biological Chemistry 275(3): 1521. Carvalho, P. C., J. Hewel, et al. (2008). \"Identifying differences in protein expression levels by spectral counting and feature selection.\" Genetics and molecular research: GMR 7(2): 342. Chargaff, E. and R. West (1946). \"The biological significance of the thromboplastic protein of blood.\" Journal of Biological Chemistry 166(1): 189-197. Chun, J., E. J. Goetzl, et al. (2002). \"International union of pharmacology. XXXIV. Lysophospholipid receptor nomenclature.\" Pharmacological reviews 54(2): 265-269. 155 Chung, S. H., J. Polg\u00C3\u00A1r, et al. (2000). \"Protein kinase C phosphorylation of syntaxin 4 in thrombin- activated human platelets.\" Journal of Biological Chemistry 275(33): 25286-25291. Cocucci, E., G. Racchetti, et al. (2009). \"Shedding microvesicles: artefacts no more.\" Trends in cell biology 19(2): 43-51. Coppinger, J., D. J. Fitzgerald, et al. (2007). \"Isolation of the platelet releasate.\" METHODS IN MOLECULAR BIOLOGY-CLIFTON THEN TOTOWA- 357: 307. Coppinger, J. A., G. Cagney, et al. (2004). \"Characterization of the proteins released from activated platelets leads to localization of novel platelet proteins in human atherosclerotic lesions.\" Blood 103(6): 2096-2104. Coppinger, J. A., R. O'Connor, et al. (2007). \"Moderation of the platelet releasate response by aspirin.\" Blood 109(11): 4786-4792. Cosemans, J., I. Munnix, et al. (2006). \"Continuous signaling via PI3K isoforms beta and gamma is required for platelet ADP receptor function in dynamic thrombus stabilization.\" Blood 108(9): 3045. Cox, J. and M. Mann (2007). \"Is proteomics the new genomics?\" Cell 130(3): 395-398. Cramer, E. M., F. Norol, et al. (1997). \"Ultrastructure of platelet formation by human megakaryocytes cultured with the Mpl ligand.\" Blood 89(7): 2336-2346. da Costa Martins, P. A., J. M. van Gils, et al. (2006). \"Platelet binding to monocytes increases the adhesive properties of monocytes by up-regulating the expression and functionality of \u00C3\u009F1 and \u00C3\u009F2 integrins.\" Journal of leukocyte biology 79(3): 499-507. Da Wei Huang, B. T. S. and R. A. Lempicki (2008). \"Systematic and integrative analysis of large gene lists using DAVID bioinformatics resources.\" Nature protocols 4(1): 44-57. Damus, P. S., M. Hicks, et al. (1973). \"Anticoagulant action of heparin.\" Dangelmaier, C. A. and H. Holmsen (1980). \"Determination of acid hydrolases in human platelets.\" Analytical biochemistry 104(1): 182-191. Daniel, T., V. Gibbs, et al. (1986). \"Thrombin stimulates c-sis gene expression in microvascular endothelial cells.\" Journal of Biological Chemistry 261(21): 9579-9582. 156 Davey, M. and E. L\u00C3\u00BCscher (1967). \"Actions of thrombin and other coagulant and proteolytic enzymes on blood platelets.\" Nature 216: 857-858. Davi, G. and C. Patrono (2007). \"Platelet activation and atherothrombosis.\" N Engl J Med 357(24): 2482- 2494. De Cristofaro, R., E. De Candia, et al. (2000). \"The Asp272\u00E2\u0080\u0093Glu282 Region of Platelet Glycoprotein Ib\u00CE\u00B1 Interacts with the Heparin-binding Site of \u00CE\u00B1-Thrombin and Protects the Enzyme from the Heparin- catalyzed Inhibition by Antithrombin III.\" Journal of Biological Chemistry 275(6): 3887-3895. Denis, M. M., N. D. Tolley, et al. (2005). \"Escaping the nuclear confines: signal-dependent pre-mRNA splicing in anucleate platelets.\" Cell 122(3): 379-391. Diacovo, T. (1994). \"A functional integrin ligand on the surface of platelets: intercellular adhesion molecule-2.\" Journal of Clinical Investigation 94(3): 1243. Diener, H., L. Cunha, et al. (1996). \"European Stroke Prevention Study 2. Dipyridamole and acetylsalicylic acid in the secondary prevention of stroke.\" Journal of the neurological sciences 143(1): 1-13. Din, J. N., S. A. Harding, et al. (2008). \"Dietary intervention with oil rich fish reduces platelet-monocyte aggregation in man.\" Atherosclerosis 197(1): 290-296. Eidt, J., P. Allison, et al. (1989). \"Thrombin is an important mediator of platelet aggregation in stenosed canine coronary arteries with endothelial injury.\" Journal of Clinical Investigation 84(1): 18. Elalamy, I., T. Chakroun, et al. (2008). \"Circulating platelet\u00E2\u0080\u0093leukocyte aggregates: A marker of microvascular injury in diabetic patients.\" Thrombosis research 121(6): 843-848. Elzagallaai, A., S. D. Ros\u00C3\u00A9, et al. (2000). \"Platelet secretion induced by phorbol esters stimulation is mediated through phosphorylation of MARCKS: a MARCKS-derived peptide blocks MARCKS phosphorylation and serotonin release without affecting pleckstrin phosphorylation.\" Blood 95(3): 894- 902. Fernandes, L. S., I. D. Conde, et al. (2003). \"Platelet-monocyte complex formation: effect of blocking PSGL-1 alone, and in combination with alphaIIbbeta3 and alphaMbeta2, in coronary stenting.\" Thromb Res 111(3): 171-177. 157 Fitzgerald, M. L. and G. L. Reed (1999). \"Rab6 is phosphorylated in thrombin-activated platelets by a protein kinase C-dependent mechanism: effects on GTP/GDP binding and cellular distribution.\" Biochemical Journal 342(Pt 2): 353. Flaumenhaft, R., J. R. Dilks, et al. (2005). \"The actin cytoskeleton differentially regulates platelet \u00CE\u00B1- granule and dense-granule secretion.\" Blood 105(10): 3879-3887. Forlow, S. B., R. P. McEver, et al. (2000). \"Leukocyte-leukocyte interactions mediated by platelet microparticles under flow.\" Blood 95(4): 1317-1323. Fujiwara, Y., V. Sardar, et al. (2005). \"Identification of residues responsible for ligand recognition and regioisomeric selectivity of lysophosphatidic acid receptors expressed in mammalian cells.\" Journal of Biological Chemistry 280(41): 35038-35050. Fukami, M. and L. Salganicoff (1977). \"Human platelet storage organelles. A review.\" Thromb Haemost 38(4): 963. Furman, M., I. Mark, et al. (1998). \"Increased platelet reactivity and circulating monocyte-platelet aggregates in patients with stable coronary artery disease.\" Journal of the American College of Cardiology 31(2): 352-358. Garc\u00C3\u00ADa, \u00C3\u0081. (2006). \"Proteome analysis of signaling cascades in human platelets.\" Blood Cells, Molecules, and Diseases 36(2): 152-156. Garc\u00C3\u00ADa, A., S. Prabhakar, et al. (2004). \"Extensive analysis of the human platelet proteome by two\u00E2\u0080\u0090 dimensional gel electrophoresis and mass spectrometry.\" Proteomics 4(3): 656-668. Garc\u00C3\u00ADa, A., S. Prabhakar, et al. (2004). \"Differential proteome analysis of TRAP-activated platelets: involvement of DOK-2 and phosphorylation of RGS proteins.\" Blood 103(6): 2088-2095. Garcia, A., N. Zitzmann, et al. (2004). Analyzing the platelet proteome, New York: Stratton Intercontinental Medical Book Corporation, c1974-. Garcia, B. A., D. M. Smalley, et al. (2005). \"The platelet microparticle proteome.\" Journal of proteome research 4(5): 1516-1521. Garrington, T. P. and G. L. Johnson (1999). \"Organization and regulation of mitogen-activated protein kinase signaling pathways.\" Current opinion in cell biology 11(2): 211-218. 158 Gawaz, M., F. J. Neumann, et al. (1998). \"Activated platelets induce monocyte chemotactic protein-1 secretion and surface expression of intercellular adhesion molecule-1 on endothelial cells.\" Circulation 98(12): 1164-1171. Gear, A. R. L. and D. Camerini (2003). \"Platelet chemokines and chemokine receptors: linking hemostasis, inflammation, and host defense.\" Microcirculation 10(3\u00E2\u0080\u00904): 335-350. Gnatenko, D. V., P. L. Perrotta, et al. (2006). \"Proteomic approaches to dissect platelet function: half the story.\" Blood 108(13): 3983-3991. Graumann, J., N. C. Hubner, et al. (2008). \"Stable isotope labeling by amino acids in cell culture (SILAC) and proteome quantitation of mouse embryonic stem cells to a depth of 5,111 proteins.\" Molecular & Cellular Proteomics 7(4): 672-683. Handagama, P., D. A. Rappolee, et al. (1990). \"Platelet alpha-granule fibrinogen, albumin, and immunoglobulin G are not synthesized by rat and mouse megakaryocytes.\" Journal of Clinical Investigation 86(4): 1364. Hanson, S. R. and L. A. Harker (1988). \"Interruption of acute platelet-dependent thrombosis by the synthetic antithrombin D-phenylalanyl-L-prolyl-L-arginyl chloromethyl ketone.\" Proceedings of the National Academy of Sciences 85(9): 3184. Harmon, J. T. and G. Jamieson (1988). \"Platelet activation by thrombin in the absence of the high-affinity thrombin receptor.\" Biochemistry 27(6): 2151-2157. Hart, M. J., X. Jiang, et al. (1998). \"Direct stimulation of the guanine nucleotide exchange activity of p115 RhoGEF by G\u00CE\u00B113.\" Science 280(5372): 2112-2114. Haser\u00C3\u00BCck, N., W. Erl, et al. (2004). \"The plaque lipid lysophosphatidic acid stimulates platelet activation and platelet-monocyte aggregate formation in whole blood: involvement of P2Y1 and P2Y12 receptors.\" Blood 103(7): 2585-2592. Hattori, R., K. Hamilton, et al. (1989). \"Stimulated secretion of endothelial von Willebrand factor is accompanied by rapid redistribution to the cell surface of the intracellular granule membrane protein GMP-140.\" Journal of Biological Chemistry 264(14): 7768-7771. Heijnen, H. F. G., A. E. Schiel, et al. (1999). \"Activated Platelets Release Two Types of Membrane Vesicles: Microvesicles by Surface Shedding and Exosomes Derived From Exocytosis of Multivesicular Bodies and -Granules.\" Blood 94(11): 3791-3799. 159 Holme, P., N. Solum, et al. (1995). \"Stimulated Glanzmann's thrombasthenia platelets produced microvesicles. Microvesiculation correlates better to exposure of procoagulant surface than to activation of GPIIb-IIIa.\" Thromb Haemost 74(6): 1533. Holme, P. A., U. \u00C8\u0096rvim, et al. (1997). \"Shear-induced platelet activation and platelet microparticle formation at blood flow conditions as in arteries with a severe stenosis.\" Arterioscler Thromb Vasc Biol 17(4): 646-653. Holmsen, H. and H. J. Weiss (1979). \"Secretable storage pools in platelets.\" Annual review of medicine 30(1): 119-134. Houng, A., J. Polg\u00C3\u00A1r, et al. (2003). \"Munc18-syntaxin complexes and exocytosis in human platelets.\" Journal of Biological Chemistry 278(22): 19627-19633. Huo, Y., A. Schober, et al. (2003). \"Circulating activated platelets exacerbate atherosclerosis in mice deficient in apolipoprotein E.\" Nature medicine 9(1): 61-67. Ideker, T., T. Galitski, et al. (2001). \"A new approach to decoding life: systems biology.\" Annual review of genomics and human genetics 2(1): 343-372. Ishihara, H., A. J. Connolly, et al. (1997). \"Protease-activated receptor 3 is a second thrombin receptor in humans.\" Israels, S., J. Gerrard, et al. (1992). \"Platelet dense granule membranes contain both granulophysin and P-selectin (GMP-140).\" Blood 80(1): 143-152. Israels, S., E. McMillan, et al. (1996). \"The lysosomal granule membrane protein, LAMP-2, is also present in platelet dense granule membranes.\" Thromb Haemost 75(4): 623. Iwamoto, S., T. Kawasaki, et al. (1996). \"Platelet microparticles: a carrier of platelet-activating factor?\" Biochemical and biophysical research communications 218(3): 940-944. Kahn, M. L., Y. W. Zheng, et al. (1998). \"A dual thrombin receptor system for platelet activation.\" Nature 394(6694): 690-694. Khandoga, A., Y. Fujiwara, et al. (2008). \"Lysophosphatidic acid-induced platelet shape change revealed through LPA1-5 receptor-selective probes and albumin.\" Platelets 19(6): 415-427. King, S. M. and G. L. Reed (2002). Development of platelet secretory granules, Elsevier. 160 Klages, B., U. Brandt, et al. (1999). \"Activation of G12/G13 Results in Shape Change and Rho/Rho- Kinase\u00E2\u0080\u0093mediated Myosin Light Chain Phosphorylation in Mouse Platelets.\" The Journal of cell biology 144(4): 745-754. Korporaal, S. J. A., G. Gorter, et al. (2005). \"Effect of oxidation on the platelet-activating properties of low-density lipoprotein.\" Arterioscler Thromb Vasc Biol 25(4): 867-872. Kovanen, P. T. and M. O. Pentik\u00C3\u00A4inen (2003). \"Circulating lipoproteins as proinflammatory and anti- inflammatory particles in atherogenesis.\" Current opinion in lipidology 14(5): 411. Kozasa, T., X. Jiang, et al. (1998). \"p115 RhoGEF, a GTPase activating protein for G\u00CE\u00B112 and G\u00CE\u00B113.\" Science 280(5372): 2109-2111. LaRosa, C. A., M. J. Rohrer, et al. (1994). \"Human neutrophil cathepsin G is a potent platelet activator.\" Journal of vascular surgery: official publication, the Society for Vascular Surgery [and] International Society for Cardiovascular Surgery, North American Chapter 19(2): 306. Leung, L. and C. Gibbs (1997). \"Modulation of thrombin's procoagulant and anticoagulant properties.\" Thromb Haemost 78(1): 577. Lewandrowski, U., J. Moebius, et al. (2006). \"Elucidation of N-glycosylation sites on human platelet proteins.\" Molecular & Cellular Proteomics 5(2): 226-233. Lewandrowski, U., S. Wortelkamp, et al. (2009). \"Platelet membrane proteomics: a novel repository for functional research.\" Blood 114(1): e10-e19. Lewandrowski, U., R. P. Zahedi, et al. (2007). \"Enhanced N-glycosylation site analysis of sialoglycopeptides by strong cation exchange prefractionation applied to platelet plasma membranes.\" Molecular & Cellular Proteomics 6(11): 1933-1941. Li, G., J. M. Sanders, et al. (2008). \"CD40 ligand promotes Mac-1 expression, leukocyte recruitment, and neointima formation after vascular injury.\" Am J Pathol 172(4): 1141-1152. Li, N., H. Hu, et al. (2000). \"Platelet-leukocyte cross talk in whole blood.\" Arterioscler Thromb Vasc Biol 20(12): 2702-2708. Libby, P. and P. Theroux (2005). \"Pathophysiology of coronary artery disease.\" Circulation 111(25): 3481- 3488. 161 Lynch, S., J. Morrow, et al. (1994). \"Formation of non-cyclooxygenase-derived prostanoids (F2- isoprostanes) in plasma and low density lipoprotein exposed to oxidative stress in vitro.\" Journal of Clinical Investigation 93(3): 998. Maguire, P., M. Foy, et al. (2005). \"Using proteomics to identify potential therapeutic targets in platelets.\" Biochemical Society Transactions 33(2): 409-412. Maguire, P. B. (2003). \"Platelet proteomics: identification of potential therapeutic targets.\" Pathophysiology of Haemostasis and Thrombosis 33(5-6): 481-486. Maguire, P. B., K. J. Wynne, et al. (2002). \"Identification of the phosphotyrosine proteome from thrombin activated platelets.\" Proteomics 2(6): 642-648. M\u00C3\u00A1jek, P., Z. Reicheltov\u00C3\u00A1, et al. (2010). \"Proteome changes in platelets activated by arachidonic acid, collagen, and thrombin.\" Proteome Science 8(1): 56. Mann, K. G., S. Butenas, et al. (2003). \"The dynamics of thrombin formation.\" Arterioscler Thromb Vasc Biol 23(1): 17-25. Mason, K. D., M. R. Carpinelli, et al. (2007). \"Programmed anuclear cell death delimits platelet life span.\" Cell 128(6): 1173-1186. Maynard, D., H. Heijnen, et al. (2007). \"Proteomic analysis of platelet \u00CE\u00B1\u00E2\u0080\u0090granules using mass spectrometry.\" Journal of Thrombosis and Haemostasis 5(9): 1945-1955. Mazzucato, M., L. De Marco, et al. (1998). \"Characterization of the initial \u00CE\u00B1-thrombin interaction with glycoprotein Ib\u00CE\u00B1 in relation to platelet activation.\" Journal of Biological Chemistry 273(4): 1880-1887. McCabe, D. J. H., P. Harrison, et al. (2004). \"Platelet degranulation and monocyte\u00E2\u0080\u0093platelet complex formation are increased in the acute and convalescent phases after ischaemic stroke or transient ischaemic attack.\" Br J Haematol 125(6): 777-787. McEver, R. P. and M. N. Martin (1984). \"A monoclonal antibody to a membrane glycoprotein binds only to activated platelets.\" Journal of Biological Chemistry 259(15): 9799. Meraji, S., C. Moore, et al. (1992). \"The importance of oxidation or glycosylation of low-density lipoproteins in relation to platelet activation.\" Platelets 3(3): 155-162. 162 Merten, M., R. Pakala, et al. (1999). \"Platelet microparticles promote platelet interaction with subendothelial matrix in a glycoprotein IIb/IIIa\u00E2\u0080\u0093dependent mechanism.\" Circulation 99(19): 2577-2582. Michell, A. W., L. M. Luheshi, et al. (2005). \"Skin and platelet \u00CE\u00B1-synuclein as peripheral biomarkers of Parkinson's disease.\" Neuroscience letters 381(3): 294-298. Molino, M., D. F. Bainton, et al. (1997). \"Thrombin receptors on human platelets.\" Journal of Biological Chemistry 272(9): 6011-6017. Naldini, A., L. Sower, et al. (1998). \"Thrombin receptor expression and responsiveness of human monocytic cells to thrombin is linked to interferon-induced cellular differentiation.\" Journal of cellular physiology 177(1): 76-84. Naseem, K., A. Goodall, et al. (1997). \"Differential effects of native and oxidatively modified low-density lipoproteins on platelet function.\" Platelets 8(2-3): 163-173. Navab, M., G. Ananthramaiah, et al. (2004). \"Thematic review series: the pathogenesis of atherosclerosis the oxidation hypothesis of atherogenesis: the role of oxidized phospholipids and HDL.\" Journal of lipid research 45(6): 993-1007. Nelken, N., S. Soifer, et al. (1992). \"Thrombin receptor expression in normal and atherosclerotic human arteries.\" Journal of Clinical Investigation 90(4): 1614. Nieswandt, B., C. Brakebusch, et al. (2001). \"Glycoprotein VI but not \u00CE\u00B12\u00CE\u00B21 integrin is essential for platelet interaction with collagen.\" The EMBO journal 20(9): 2120-2130. Niewiarowski, S. (1977). \"Proteins secreted by the platelet.\" Thromb Haemost 38(4): 924. Nomura, S., S. Kanazawa, et al. (2002). \"Effects of efonidipine on platelet and monocyte activation markers in hypertensive patients with and without type 2 diabetes mellitus.\" Journal of human hypertension 16(8): 539-547. Norton, K. J., R. M. Scarborough, et al. (1993). \"Immunologic analysis of the cloned platelet thrombin receptor activation mechanism: evidence supporting receptor cleavage, release of the N-terminal peptide, and insertion of the tethered ligand into a protected environment.\" Blood 82(7): 2125-2136. O'Brien, J. (1955). \"The Platelet\u00E2\u0080\u0090like Activity of Serum.\" Br J Haematol 1(2): 223-228. 163 Offermanns, S., K. L. Laugwitz, et al. (1994). \"G proteins of the G12 family are activated via thromboxane A2 and thrombin receptors in human platelets.\" Proceedings of the National Academy of Sciences 91(2): 504. Offermanns, S., V. Mancino, et al. (1997). \"Vascular system defects and impaired cell chemokinesis as a result of G\u00CE\u00B113 deficiency.\" Science 275(5299): 533-536. Ong, S. E., B. Blagoev, et al. (2002). \"Stable isotope labeling by amino acids in cell culture, SILAC, as a simple and accurate approach to expression proteomics.\" Molecular & Cellular Proteomics 1(5): 376- 386. Ostrovsky, L., A. J. King, et al. (1998). \"A juxtacrine mechanism for neutrophil adhesion on platelets involves platelet-activating factor and a selectin-dependent activation process.\" Blood 91(8): 3028-3036. Otterdal, K., T. M. Pedersen, et al. (2001). \"Platelet shape change induced by the peptide YFLLRNP.\" Thrombosis research 103(5): 411-420. Pamuklar, Z., J. S. Lee, et al. (2008). \"Individual Heterogeneity in Platelet Response to Lysophosphatidic Acid.\" Arterioscler Thromb Vasc Biol 28(3): 555-561. Piersma, S. R., H. J. Broxterman, et al. (2009). \"Proteomics of the TRAP-induced platelet releasate.\" Journal of proteomics 72(1): 91-109. Pittman, D. D. and R. J. Kaufman (1988). \"Proteolytic requirements for thrombin activation of anti- hemophilic factor (factor VIII).\" Proceedings of the National Academy of Sciences 85(8): 2429. Polg\u00C3\u00A1r, J., S. H. Chung, et al. (2002). \"Vesicle-associated membrane protein 3 (VAMP-3) and VAMP-8 are present in human platelets and are required for granule secretion.\" Blood 100(3): 1081-1083. Polg\u00C3\u00A1r, J., W. S. Lane, et al. (2003). \"Phosphorylation of SNAP-23 in activated human platelets.\" Journal of Biological Chemistry 278(45): 44369. Poole, A., J. Gibbins, et al. (1997). \"The Fc receptor \u00CE\u00B3-chain and the tyrosine kinase Syk are essential for activation of mouse platelets by collagen.\" The EMBO journal 16(9): 2333-2341. Prescott, S. M., G. A. Zimmerman, et al. (1984). \"Human endothelial cells in culture produce platelet- activating factor (1-alkyl-2-acetyl-sn-glycero-3-phosphocholine) when stimulated with thrombin.\" Proceedings of the National Academy of Sciences 81(11): 3534. 164 Quinn, M. T., S. Parthasarathy, et al. (1988). \"Lysophosphatidylcholine: a chemotactic factor for human monocytes and its potential role in atherogenesis.\" Proceedings of the National Academy of Sciences 85(8): 2805. Rabie, T., A. Strehl, et al. (2005). \"Evidence for a role of ADAM17 (TACE) in the regulation of platelet glycoprotein V.\" Journal of Biological Chemistry 280(15): 14462-14468. Ratajczak, J., M. Wysoczynski, et al. (2006). \"Membrane-derived microvesicles: important and underappreciated mediators of cell-to-cell communication.\" Leukemia 20(9): 1487-1495. Reed, G. L. (2004). Platelet secretory mechanisms, New York: Stratton Intercontinental Medical Book Corporation, c1974-. Reed, G. L., A. K. Houng, et al. (1999). \"Human platelets contain SNARE proteins and a Sec1p homologue that interacts with syntaxin 4 and is phosphorylated after thrombin activation: implications for platelet secretion.\" Blood 93(8): 2617-2626. Rendu, F. and B. Brohard-Bohn (2001). \"The platelet release reaction: granules' constituents, secretion and functions.\" Platelets 12(5): 261-273. Retzer, M. and M. Essler (2000). \"Lysophosphatidic acid-induced platelet shape change proceeds via Rho/Rho kinase-mediated myosin light-chain and moesin phosphorylation.\" Cellular signalling 12(9-10): 645-648. Rinder, H. M., J. L. Bonan, et al. (1991). \"Dynamics of leukocyte-platelet adhesion in whole blood.\" Blood 78(7): 1730-1737. Ross, R. (1999). \"Atherosclerosis--an inflammatory disease.\" N Engl J Med 340(2): 115-126. Rother, E., R. Brandl, et al. (2003). \"Subtype-selective antagonists of lysophosphatidic acid receptors inhibit platelet activation triggered by the lipid core of atherosclerotic plaques.\" Circulation 108(6): 741- 747. Sanchez-Quesada, J. L., S. Ben\u00C3\u00ADtez, et al. (2004). \"Electronegative low-density lipoprotein.\" Current opinion in lipidology 15(3): 329. Sano, T., D. Baker, et al. (2002). \"Multiple mechanisms linked to platelet activation result in lysophosphatidic acid and sphingosine 1-phosphate generation in blood.\" Journal of Biological Chemistry 277(24): 21197-21206. 165 Sasakawa, N., M. OHARA\u00E2\u0080\u0090IMAIZUMI, et al. (2002). \"Roles of Actin Filaments and the Actin\u00E2\u0080\u0090Myosin Interaction in the Regulation of Exocytosis in Chromaffin Cells.\" Annals of the New York Academy of Sciences 971(1): 273-274. Scheuerer, B., M. Ernst, et al. (2000). \"The CXC-chemokine platelet factor 4 promotes monocyte survival and induces monocyte differentiation into macrophages.\" Blood 95(4): 1158-1166. Schmidt, V. A., W. C. Nierman, et al. (1998). \"The human proteinase-activated receptor-3 (PAR-3) gene.\" Journal of Biological Chemistry 273(24): 15061-15068. Schober, A., D. Manka, et al. (2002). \"Deposition of platelet RANTES triggering monocyte recruitment requires P-selectin and is involved in neointima formation after arterial injury.\" Circulation 106(12): 1523-1529. Schraw, T. D., P. P. Lemons, et al. (2003). \"A role for Sec1/Munc18 proteins in platelet exocytosis.\" Biochemical Journal 374(Pt 1): 207. Schwertz, H., N. D. Tolley, et al. (2006). \"Signal-dependent splicing of tissue factor pre-mRNA modulates the thrombogenecity of human platelets.\" The Journal of experimental medicine 203(11): 2433-2440. Shcherbina, A. and E. Remold-O\u00E2\u0080\u0099Donnell (1999). \"Role of caspase in a subset of human platelet activation responses.\" Blood 93(12): 4222-4231. Shevchenko, A., J. H. Henrik Tomas, et al. (2007). \"In-gel digestion for mass spectrometric characterization of proteins and proteomes.\" Nature protocols 1(6): 2856-2860. Siess, W., K. J. Zangl, et al. (1999). \"Lysophosphatidic acid mediates the rapid activation of platelets and endothelial cells by mildly oxidized low density lipoprotein and accumulates in human atherosclerotic lesions.\" Proceedings of the National Academy of Sciences 96(12): 6931. Simon, D. I., Z. Chen, et al. (2000). \"Platelet glycoprotein Ib\u00CE\u00B1 is a counterreceptor for the leukocyte integrin Mac-1 (CD11b/CD18).\" The Journal of experimental medicine 192(2): 193-204. Simon, M. F., H. Chap, et al. (1982). \"Human platelet aggregation induced by 1-alkyl-lysophosphatidic acid and its analogs: a new group of phospholipid mediators?\" Biochemical and biophysical research communications 108(4): 1743-1750. 166 Sims, P., T. Wiedmer, et al. (1989). \"Assembly of the platelet prothrombinase complex is linked to vesiculation of the platelet plasma membrane. Studies in Scott syndrome: an isolated defect in platelet procoagulant activity.\" Journal of Biological Chemistry 264(29): 17049. Smalley, D. M., K. E. Root, et al. (2007). \"Proteomic discovery of 21 proteins expressed in human plasma- derived but not platelet-derived microparticles.\" THROMBOSIS AND HAEMOSTASIS-STUTTGART- 97(1): 67. Smyth, S. S., H. Y. Cheng, et al. (2008). \"Roles of lysophosphatidic acid in cardiovascular physiology and disease.\" Biochimica et Biophysica Acta (BBA)-Molecular and Cell Biology of Lipids 1781(9): 563-570. S\u00C3\u00B6llner, T., S. W. Whiteheart, et al. (1993). \"SNAP receptors implicated in vesicle targeting and fusion.\" Nature 362(6418): 318-324. Steinberg, D. (1997). \"Low density lipoprotein oxidation and its pathobiological significance.\" Journal of Biological Chemistry 272(34): 20963-20966. Subbanagounder, G., A. D. Watson, et al. (2000). \"Bioactive products of phospholipid oxidation: isolation, identification, measurement and activities.\" Free Radical Biology and Medicine 28(12): 1751- 1761. Tabata, K., K. Baba, et al. (2007). \"The orphan GPCR GPR87 was deorphanized and shown to be a lysophosphatidic acid receptor.\" Biochemical and biophysical research communications 363(3): 861-866. Tanaga, K., H. Bujo, et al. (2002). \"Increased circulating malondialdehyde-modified LDL levels in patients with coronary artery diseases and their association with peak sizes of LDL particles.\" Arterioscler Thromb Vasc Biol 22(4): 662-666. Tesfamariam, B., G. Allen, et al. (1993). \"Involvement of the\" tethered ligand\" receptor in thrombin- induced endothelium-mediated relaxations.\" American Journal of Physiology-Heart and Circulatory Physiology 265(5): H1744-H1749. Van Aelst, L. and C. D\u00E2\u0080\u0099Souza-Schorey (1997). \"Rho GTPases and signaling networks.\" Genes & Development 11(18): 2295-2322. van Gils, J. M., P. A. da Costa Martins, et al. (2008). \"Transendothelial migration drives dissociation of plateletmonocyte complexes.\" Thromb Haemost 100(2): 271-279. von Hundelshausen, P., K. S. C. Weber, et al. (2001). \"RANTES deposition by platelets triggers monocyte arrest on inflamed and atherosclerotic endothelium.\" Circulation 103(13): 1772-1777. 167 Vu, T. K., D. T. Hung, et al. (1991). \"Molecular cloning of a functional thrombin receptor reveals a novel proteolytic mechanism of receptor activation.\" Cell 64(6): 1057. Watson, S., M. Ruggiero, et al. (1986). \"Inositol 1, 4, 5-trisphosphate induces aggregation and release of 5-hydroxytryptamine from saponin-permeabilized human platelets.\" Journal of Biological Chemistry 261(12): 5368. Weber, C. and T. A. Springer (1997). \"Neutrophil accumulation on activated, surface-adherent platelets in flow is mediated by interaction of Mac-1 with fibrinogen bound to alphaIIbbeta3 and stimulated by platelet-activating factor.\" Journal of Clinical Investigation 100(8): 2085. Weber, T., B. V. Zemelman, et al. (1998). \"SNAREpins: minimal machinery for membrane fusion.\" Cell 92(6): 759-772. Weidtmann, A., R. Scheithe, et al. (1995). \"Mildly oxidized LDL induces platelet aggregation through activation of phospholipase A2.\" Arterioscler Thromb Vasc Biol 15(8): 1131-1138. Weksler, B. B., C. W. Ley, et al. (1978). \"Stimulation of endothelial cell prostacyclin production by thrombin, trypsin, and the ionophore A 23187.\" Journal of Clinical Investigation 62(5): 923. Weston, A. D. and L. Hood (2004). \"Systems biology, proteomics, and the future of health care: toward predictive, preventative, and personalized medicine.\" Journal of proteome research 3(2): 179-196. Weyrich, A. S., M. M. Denis, et al. (2005). \"Dipyridamole selectively inhibits inflammatory gene expression in platelet\u00E2\u0080\u0093monocyte aggregates.\" Circulation 111(5): 633-642. WHITE, J. G. (1969). \"The dense bodies of human platelets: inherent electron opacity of the serotonin storage particles.\" Blood 33(4): 598-606. White, J. G. (1974). \"Electron microscopic studies of platelet secretion.\" Progress in hemostasis and thrombosis 2(0): 49. Whiteheart, S. W. (2011). \"Platelet granules: surprise packages.\" Blood 118(5): 1190-1191. Wolf, P. (1967). \"The nature and significance of platelet products in human plasma.\" Br J Haematol 13(3): 269-288. 168 Xu, W., H. Andersen, et al. (1998). \"Cloning and characterization of human protease-activated receptor 4.\" Proceedings of the National Academy of Sciences 95(12): 6642. Yeo, E. L., J. Sheppard, et al. (1994). \"Role of P-selectin and leukocyte activation in polymorphonuclear cell adhesion to surface adherent activated platelets under physiologic shear conditions (an injury vessel wall model).\" Blood 83(9): 2498-2507. Zerial, M. and H. McBride (2001). \"Rab proteins as membrane organizers.\" Nature reviews Molecular cell biology 2(2): 107-117. Zhang, C., D. L. Baker, et al. (2004). \"Lysophosphatidic acid induces neointima formation through PPAR\u00CE\u00B3 activation.\" The Journal of experimental medicine 199(6): 763-774. Zimmerman, G. A., T. M. McINTYRE, et al. (1986). \"Thrombin Stimulates Neutrophil Adherence by an Endothelial Cell\u00E2\u0080\u0090Dependent Mechanism: Characterization of the Response and Relationship to Platelet\u00E2\u0080\u0090 Activating Factor Synthesisa.\" Annals of the New York Academy of Sciences 485(1): 349-368. 169 Appendices Appendix A: Proteins primarily found in microparticles from thrombin activated platelets The 86 proteins in common between the three datasets from total platelet releasate (Rel) and releasate free of microparticles (Rel-MP) (Figure 3.20) for thrombin, collagen or thrombin plus collagen activated platelets were looked at to find proteins where the average thrombin/collagen (T/C) in the Rel significantly increases as compared to the average T/C ratio in the Rel-MP sample. These proteins can potentially come from microparticles generated through thrombin platelet activation. The red background designates proteins which are significantly more abundant in thrombin activated platelets and the green background designates proteins which are significantly more abundant in collagen activated platelets. T/C 1 Rel T/C 2 Rel T/C 3 Rel AVG T/C Rel Protein name T/C 1 Rel-MP T/C 2 Rel-MP T/C 3 Rel-MP AVG T/C Rel-MP 1.11 6.03 7.69 4.94 Apolipoprotein A-I 1.38 7.69 1.33 3.46 11.30 0.84 0.91 4.35 WD repeat-containing protein 1 0.58 0.91 2.73 1.41 1.76 2.24 6.06 3.36 Nidogen-1 1.37 6.06 1.98 3.14 2.66 0.94 6.36 3.32 L-lactate dehydrogenase B chain 0.46 6.36 1.05 2.62 3.38 0.82 1.07 1.76 Vitamin D-binding protein 1.38 1.07 1.20 1.21 2.41 1.00 1.46 1.63 Complement factor H 1.57 1.46 1.18 1.40 1.41 2.32 0.64 1.46 Transforming growth factor beta-1 0.94 0.64 2.23 1.27 1.91 1.35 1.03 1.43 Ig gamma-1 chain C region 1.21 1.03 1.26 1.17 0.59 2.94 0.74 1.42 Malate dehydrogenase, cytoplasmic 0.41 0.74 1.37 0.84 1.79 1.48 0.76 1.34 Ig kappa chain C region 1.30 0.76 1.14 1.07 1.83 0.90 1.29 1.34 Multimerin-1 1.38 1.29 0.59 1.09 170 Appendix B: Proteins primarily found in microparticles from collagen activated platelets The 86 proteins in common between the three datasets from total platelet releasate (Rel) and releasate free of microparticles (Rel-MP) (Figure 3.20) for thrombin, collagen or thrombin plus collagen activated platelets were looked at to find proteins where the average thrombin/collagen (T/C) in the Rel significantly decreases as compared to the average T/C ratio in the Rel-MP sample. These proteins can potentially be most abundant in microparticles from collagen activated platelets. The red background designates proteins which are significantly more abundant in thrombin activated platelets and the green background designates proteins which are significantly more abundant in collagen activated platelets. T/C 1 Rel T/C 2 Rel T/C 2 Rel AVG T/C Rel Protein name T/C 1 Rel-MP T/C 2 Rel-MP T/C 2 Rel-MP AVGT/C Rel-MP 3.94 0.51 1.88 2.11 Platelet factor 4 1.95 1.88 4.12 2.65 2.12 1.43 1.62 1.72 Coagulation factor V 1.96 1.62 2.29 1.96 1.04 2.00 1.28 1.44 Glyceraldehyde-3-phosphate dehydrogenase 1.55 1.28 4.15 2.33 1.41 1.13 1.16 1.23 Metalloproteinase inhibitor 1 1.44 1.16 3.26 1.95 1.61 0.58 1.19 1.13 Alpha-1-antitrypsin 1.09 1.19 2.29 1.52 1.78 0.63 0.80 1.07 Fibronectin 1.44 0.80 1.89 1.38 1.48 0.87 0.84 1.06 Plasminogen 1.34 0.84 1.70 1.29 1.38 0.56 1.00 0.98 von Willebrand factor 1.40 1.00 2.13 1.51 0.52 1.19 0.94 0.88 Calmodulin 0.51 0.94 10.85 4.10 0.77 0.39 1.41 0.86 Chloride intracellular channel protein 1 0.67 1.41 2.40 1.49 0.28 0.16 1.83 0.76 Integrin alpha-IIb 0.70 1.83 5.02 2.52 0.75 0.55 0.96 0.75 Alpha-enolase 1.06 0.96 1.18 1.07 0.93 0.38 0.92 0.74 Clusterin 1.06 0.92 1.08 1.02 0.81 0.56 0.84 0.74 Ubiquitin 0.45 0.84 3.63 1.64 0.45 0.42 1.19 0.69 Myosin-9 0.67 1.19 1.47 1.11 0.38 0.68 0.86 0.64 L-lactate dehydrogenase A chain 0.57 0.86 1.52 0.98 0.59 0.70 0.47 0.59 Apolipoprotein B-100 0.55 0.47 2.26 1.10 0.48 0.53 0.76 0.59 Glutathione S-transferase P 0.47 0.76 1.70 0.98 0.67 0.41 0.42 0.50 14-3-3 protein epsilon 0.55 0.42 1.67 0.88 0.48 0.48 0.46 0.47 Coagulation factor XIII A chain 0.16 0.46 1.36 0.66 0.41 0.23 0.30 0.31 Adenylyl cyclase-associated protein 1 0.41 0.30 0.88 0.53 171 Appendix C: Proteins primarily found in microparticles from thrombin plus collagen activated platelets The 86 proteins in common between the three datasets from total platelet releasate (Rel) and releasate free of microparticles (Rel-MP) (Figure 3.20) for thrombin, collagen or thrombin plus collagen activated platelets were looked at to find proteins where the average thrombin/thrombin plus collagen (T/TC) in the Rel significantly decreases as compared to the average T/TC ratio in the Rel-MP sample. These proteins can potentially be most abundant in microparticles from thrombin plus collagen activated platelets. The red background designates proteins which are significantly more abundant in thrombin activated platelets and the green background designates proteins which are significantly more abundant in thrombin plus collagen activated platelets. T/TC 1 Rel T/TC 2 Rel T/TC 2 Rel AVG T/TC Rel Protein name T/TC 1 Rel-MP T/TC 2 Rel-MP T/TC 2 Rel-MP AVGT/TC Rel-MP 1.42 1.00 15.49 5.97 Apolipoprotein A-I 2.04 15.49 3.10 6.88 2.17 1.35 4.60 2.71 Complement factor H 2.50 4.60 2.03 3.04 0.37 0.60 6.68 2.55 Serum deprivation-response protein 0.89 6.68 1.36 2.98 2.37 0.69 2.78 1.95 Platelet factor 4 3.57 2.78 1.85 2.73 3.69 0.70 1.28 1.89 C-C motif chemokine 5 3.11 1.28 2.32 2.24 2.08 0.99 2.37 1.81 Alpha-1-antitrypsin 1.80 2.37 2.15 2.10 1.84 1.30 2.02 1.72 Multimerin-1 2.32 2.02 2.09 2.15 1.55 1.21 2.39 1.71 von Willebrand factor 4.39 2.39 1.99 2.92 1.61 1.40 1.85 1.62 Latent-transforming growth factor beta- binding protein 1 2.44 1.85 1.67 1.99 1.53 1.66 1.11 1.43 Peptidyl-prolyl cis-trans isomerase B 1.67 1.11 5.16 2.65 1.53 1.33 1.37 1.41 Plasminogen 2.30 1.37 1.79 1.82 1.44 1.30 1.29 1.34 Thrombospondin-1 1.91 1.29 1.82 1.67 1.30 1.36 1.25 1.30 Platelet glycoprotein V 2.24 1.25 1.11 1.53 1.25 0.91 1.63 1.26 Vitamin K-dependent protein S 2.46 1.63 1.52 1.87 1.73 0.76 1.27 1.26 Serum albumin 2.23 1.27 1.44 1.65 0.80 0.88 1.94 1.21 Alpha-enolase 3.10 1.94 1.19 2.08 0.58 0.79 2.23 1.20 Myosin-9 5.60 2.23 14.54 7.46 1.25 1.17 1.15 1.19 Serglycin 2.00 1.15 1.84 1.66 1.09 0.83 1.64 1.19 Platelet basic protein 1.38 1.64 1.27 1.43 0.86 1.17 1.38 1.13 Heat shock cognate 71 kDa protein 1.19 1.38 1.44 1.34 172 T/TC 1 Rel T/TC 2 Rel T/TC 2 Rel AVG T/TC Rel Protein name T/TC 1 Rel-MP T/TC 2 Rel-MP T/TC 2 Rel-MP AVGT/TC Rel-MP 0.96 0.74 1.54 1.08 Protein disulfide-isomerase A3 1.32 1.54 1.77 1.55 0.98 1.06 1.12 1.05 78 kDa glucose-regulated protein 1.38 1.12 2.19 1.56 0.88 1.06 1.18 1.04 Alpha-actinin-1 1.50 1.18 1.67 1.45 1.14 0.64 1.32 1.03 Clusterin 1.58 1.32 1.81 1.57 0.48 0.74 1.80 1.01 Talin-1 1.13 1.80 1.53 1.49 0.90 0.98 1.11 0.99 Vitamin D-binding protein 1.60 1.11 1.62 1.44 0.89 0.67 1.40 0.99 Gelsolin 1.22 1.40 1.01 1.21 0.57 0.90 1.49 0.98 Vinculin 1.36 1.49 1.04 1.30 1.43 0.68 0.72 0.94 Ubiquitin 3.95 0.72 0.86 1.84 0.51 0.55 1.72 0.93 L-lactate dehydrogenase A chain 1.53 1.72 0.97 1.41 0.65 1.05 0.76 0.82 Phosphoglycerate kinase 1 1.71 0.76 1.20 1.22 2.10 0.04 0.24 0.80 Fibrinogen beta chain 1.71 0.24 1.15 1.03 0.62 0.67 1.07 0.79 Glutathione S-transferase P 0.93 1.07 1.03 1.01 1.71 0.07 0.47 0.75 Fibrinogen alpha chain 2.25 0.47 1.09 1.27 0.35 0.90 0.94 0.73 Actin, cytoplasmic 1 1.11 0.94 0.89 0.98 0.46 0.86 0.75 0.69 Peptidyl-prolyl cis-trans isomerase A 0.99 0.75 1.22 0.98 0.67 0.59 0.54 0.60 14-3-3 protein epsilon 1.58 0.54 1.68 1.27 0.43 0.49 0.85 0.59 Adenylyl cyclase-associated protein 1 1.04 0.85 1.53 1.14 0.38 0.46 0.89 0.58 Tubulin beta-1 chain 0.76 0.89 1.13 0.93 0.08 0.34 1.23 0.55 Integrin alpha-IIb 0.98 1.23 0.96 1.06 0.23 0.56 0.54 0.44 Ras-related protein Rap-1b 0.67 0.54 1.07 0.76 173 Appendix D: Proteins unique to total releasate (Rel) from thrombin, collagen or thrombin plus collagen activated platelets (from Figure 3.20) T/C 1 T/C 2 T/C 3 AVG T/C Protein name T/TC 1 T/TC 2 T/TC 3 AVG T/TC 2.E+03 3.E+03 3.E+03 2.E+03 Homeobox protein Hox-D1 5.E+02 8.E+02 6.E+02 6.E+02 2.78 4.06 1.26 2.70 Nidogen-2 1.95 5.38 1.35 2.89 1.49 0.81 0.93 1.08 Cofilin-1 1.55 1.29 1.39 1.41 1.32 0.76 0.79 0.96 Kininogen-1 1.83 2.03 0.70 1.52 1.06 0.75 0.76 0.85 Actin-related protein 2 0.53 1.21 1.09 0.94 0.61 0.63 0.53 0.59 Septin-6 0.75 1.15 2.32 1.41 0.39 0.54 0.78 0.57 PDZ and LIM domain protein 1 0.49 0.92 1.38 0.93 0.63 0.90 0.15 0.56 Glia-derived nexin 78.85 5.18 0.27 28.10 0.65 0.40 0.46 0.50 14-3-3 protein gamma 0.65 0.57 0.55 0.59 0.38 0.31 0.52 0.40 Tubulin beta chain 0.42 0.61 0.97 0.67 0.09 0.15 0.57 0.27 Integrin-linked protein kinase 0.17 0.80 1.37 0.78 174 Appendix E: Proteins unique to total releasate free of microparticles (Rel-MP) from thrombin, collagen or thrombin plus collagen activated platelets (from Figure 3.20) T/C 1 T/C 2 T/C 2 AVG T/C Description 2 T/TC 1 T/TC 2 T/TC 2 AVG T/TC 4.13 0.10 1.11 1.78 Angiopoietin-1 4.55 0.08 2.57 2.40 0.57 0.78 3.25 1.53 Antithrombin-III 2.47 2.40 3.56 2.81 1.73 0.55 0.80 1.03 Beta-2-glycoprotein 1 1.83 0.48 1.10 1.14 1.25 1.00 0.78 1.01 Extracellular matrix protein 1 2.31 1.35 1.13 1.60 0.15 0.21 0.14 0.17 Platelet glycoprotein VI 0.16 0.23 0.15 0.18 0.95 0.22 1.04 0.74 Haptoglobin 1.31 0.28 2.39 1.32 1.32 0.46 1.25 1.01 Ig gamma-2 chain C region 1.73 0.45 1.30 1.16 0.39 0.61 0.81 0.61 Moesin 0.78 1.34 1.37 1.16 2.E+03 9.E+02 1.E+04 5.E+03 Regulator of G-protein signaling protein-like 4.E+03 2.E+04 1.E+07 4.E+06 2.06 0.84 0.96 1.28 Proactivator polypeptide 2.26 1.18 1.36 1.60 3.47 4.14 16.46 8.02 Transthyretin 6.70 31.44 15.31 17.82 175 Appendix F: Proteins primarily found in microparticles from LPA activated platelets The 129 proteins in common between the three datasets from total platelet releasate (Rel) and releasate free of microparticles (Rel-MP) (Figure 3.25) for collagen, LPA or subthreshold collagen plus LPA activated platelets were looked at to find proteins where the average LPA/collagen (L/C) in the Rel significantly increases as compared to the average L/C ratio in the Rel-MP sample. These proteins can potentially be most abundant in microparticles from LPA activated platelets. The red background designates proteins which are significantly more abundant in LPA activated platelets and the green background designates proteins which are significantly more abundant in collagen activated platelets. L/C 1 Rel L/C 2 Rel L/C 3 Rel AVG L/C Rel Protein name L/C 1 Rel-MP L/C 2 Rel-MP L/C 3 Rel-MP AVG L/C Rel-MP 14.26 5.66 149.59 56.50 Cysteine and glycine-rich protein 1 21.99 20.08 14.89 18.99 4.15 61.52 5.26 23.64 Nucleoside diphosphate kinase A 5.00 4.98 4.51 4.83 12.25 7.10 11.42 10.26 Peroxiredoxin-5, mitochondrial 6.46 10.94 12.01 9.80 0.80 0.52 4.65 1.99 P-selectin 0.20 0.87 1.64 0.91 0.97 2.64 1.52 1.71 Complement factor H 0.13 0.53 1.24 0.64 2.11 1.16 1.47 1.58 Platelet factor 4 1.52 1.13 1.36 1.34 1.17 0.97 1.08 1.07 Serum albumin 0.83 0.62 1.14 0.86 0.88 1.01 0.87 0.92 Plasminogen 0.45 0.20 0.82 0.49 0.75 0.96 0.90 0.87 Vitamin D-binding protein 0.04 0.67 0.36 0.36 176 Appendix G: Proteins unique to total releasate (Rel) from collagen, LPA or subthreshold collagen plus LPA activated platelets (from Figure 3.25) L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 24.21 12.14 101.52 45.96 PDZ and LIM domain protein 7 16.89 10.21 87.54 38.21 3.57 2.79 99.92 35.43 Dynamin-1-like protein 2.70 3.27 97.36 34.45 39.19 11.99 11.74 20.97 Four and a half LIM domains protein 1 21.62 12.90 7.71 14.08 5.37 10.07 28.42 14.62 Cytoskeleton-associated protein 5 5.73 5.80 16.18 9.24 2.40 2.81 38.59 14.60 Ras-related C3 botulinum toxin substrate 1 1.83 2.21 31.42 11.82 8.34 24.65 7.80 13.60 Src substrate cortactin 12.67 20.92 17.00 16.86 11.13 1.26 27.33 13.24 Alpha-synuclein 8.09 0.03 20.55 9.56 11.26 12.05 10.43 11.25 Hsp90 co-chaperone Cdc37 11.11 11.08 9.13 10.44 4.23 5.44 22.57 10.75 Glycogen phosphorylase, brain form 3.09 4.80 19.93 9.27 3.57 3.24 22.72 9.85 Septin-2 2.58 1.31 23.15 9.01 2.81 8.04 17.99 9.61 Ester hydrolase C11orf54 3.76 2.16 17.11 7.68 5.62 6.55 16.46 9.54 Tyrosine-protein phosphatase non-receptor type 6 6.66 4.83 12.14 7.87 4.07 7.04 16.01 9.04 EH domain-containing protein 3 3.21 11.02 14.23 9.49 9.78 4.85 9.30 7.98 Dihydropyrimidinase-related protein 2 3.66 5.07 4.53 4.42 13.27 4.76 4.44 7.49 Mesencephalic astrocyte-derived neurotrophic factor 3.49 1.85 3.44 2.92 4.15 5.61 10.82 6.86 Protein-L-isoaspartate(D- aspartate) 3.68 5.89 6.70 5.42 7.30 5.06 6.76 6.38 Tropomodulin-3 7.28 5.38 4.87 5.84 3.11 3.03 12.93 6.36 Glucose-6-phosphate isomerase 1.89 2.53 5.45 3.29 9.02 2.83 5.50 5.78 Myosin regulatory light polypeptide 9 5.57 4.31 5.52 5.13 5.50 5.33 5.60 5.48 Tropomyosin beta chain 18.98 3.42 2.34 8.25 6.41 4.32 4.27 5.00 Hypoxanthine-guanine phosphoribosyltransferase 2.20 5.67 3.19 3.69 4.88 4.40 5.60 4.96 Myotrophin 5.78 3.96 4.73 4.82 4.19 5.77 4.83 4.93 Tropomyosin alpha-3 chain 24.67 4.15 4.35 11.06 7.42 2.93 4.41 4.92 Serpin B6 2.95 3.92 11.97 6.28 4.39 4.04 5.65 4.69 Coactosin-like protein 4.15 4.67 4.68 4.50 0.11 3.71 10.06 4.63 Catalase 0.00 3.46 7.15 3.54 3.92 3.83 6.09 4.61 SH3 domain-binding glutamic acid-rich-like protein 2 3.75 3.99 4.73 4.16 4.03 1.98 7.68 4.56 Glucosidase 2 subunit beta 3.36 1.99 7.87 4.41 3.35 3.23 5.35 3.98 F-actin-capping protein subunit alpha-1 1.88 4.48 4.46 3.61 0.53 2.96 8.31 3.93 Purine nucleoside phosphorylase 0.14 2.34 7.58 3.35 177 L/C 1 L/C 2 L/C 3 AVG L/C Protein name CL/C 1 CL/C 2 CL/C 3 AVG CL/C 3.71 0.00 7.60 3.77 Chloride intracellular channel protein 4 3.08 0.00 7.22 3.43 3.49 3.89 3.80 3.73 Glutathione peroxidase 1 9.81 8.09 4.46 7.46 3.04 3.28 4.70 3.67 Malate dehydrogenase, cytoplasmic 1.66 5.87 3.84 3.79 2.76 4.58 3.42 3.59 Proteasome activator complex subunit 1 1.72 4.80 3.23 3.25 1.04 8.51 0.73 3.43 Inter-alpha-trypsin inhibitor heavy chain H4 0.72 80.52 0.30 27.18 1.43 3.85 3.61 2.96 EH domain-containing protein 1 1.89 5.32 4.73 3.98 2.54 0.08 6.09 2.90 Isochorismatase domain- containing protein 1 1.69 0.07 4.69 2.15 2.59 0.61 3.40 2.20 Heat shock protein HSP 90-alpha 1.78 0.72 4.86 2.45 2.33 1.22 2.85 2.13 Actin-related protein 2 1.10 1.41 2.01 1.51 1.42 2.06 1.29 1.59 Endonuclease domain-containing 1 protein 0.88 4.44 1.07 2.13 1.33 1.20 1.99 1.51 Actin-related protein 3 1.13 2.96 1.94 2.01 3.36 0.00 0.02 1.13 Neurogranin 2.85 0.02 0.03 0.97 0.94 0.82 1.27 1.01 Metalloproteinase inhibitor 1 0.33 0.64 19.43 6.80 0.93 0.82 0.97 0.91 Beta-2-microglobulin 0.83 0.64 0.75 0.74 0.57 0.50 0.78 0.62 Extracellular matrix protein 1 0.38 0.26 0.61 0.42 0.53 0.42 0.59 0.51 Integrin alpha-6 0.35 0.66 0.52 0.51 0.45 0.39 0.68 0.51 Heparanase 0.36 0.42 0.59 0.46 0.50 0.73 0.22 0.48 Integrin beta-3 0.19 0.91 0.39 0.50 0.31 0.36 0.76 0.48 Integrin alpha-IIb 0.34 0.72 0.52 0.53 0.40 0.16 0.10 0.22 C-C motif chemokine 5 0.23 0.16 1.E-03 0.13 0.15 0.12 0.17 0.15 Erythrocyte band 7 integral membrane protein 0.13 0.50 0.30 0.31 "@en . "Thesis/Dissertation"@en . "2012-11"@en . "10.14288/1.0073003"@en . "eng"@en . "Experimental Medicine"@en . "Vancouver : University of British Columbia Library"@en . "University of British Columbia"@en . "Attribution-NonCommercial-NoDerivatives 4.0 International"@en . "http://creativecommons.org/licenses/by-nc-nd/4.0/"@en . "Graduate"@en . "Exploring the interaction environment of blood cells : proteomic analysis of platelet releasate and platelet-monocyte interaction"@en . "Text"@en . "http://hdl.handle.net/2429/42942"@en .