Open Collections

UBC Theses and Dissertations

UBC Theses Logo

UBC Theses and Dissertations

Novel anticoagulant neutralizing agents for the management of bleeding : studies on the design, mechanism… Kalathottukaren, Manu Thomas 2017

Your browser doesn't seem to have a PDF viewer, please download the PDF to view this item.

Notice for Google Chrome users:
If you are having trouble viewing or searching the PDF with Google Chrome, please download it here instead.

Item Metadata

Download

Media
24-ubc_2017_may_kalathottukaren_manuthomas.pdf [ 9.33MB ]
Metadata
JSON: 24-1.0343330.json
JSON-LD: 24-1.0343330-ld.json
RDF/XML (Pretty): 24-1.0343330-rdf.xml
RDF/JSON: 24-1.0343330-rdf.json
Turtle: 24-1.0343330-turtle.txt
N-Triples: 24-1.0343330-rdf-ntriples.txt
Original Record: 24-1.0343330-source.json
Full Text
24-1.0343330-fulltext.txt
Citation
24-1.0343330.ris

Full Text

  NOVEL ANTICOAGULANT NEUTRALIZING AGENTS FOR THE MANAGEMENT OF BLEEDING: STUDIES ON THE DESIGN, MECHANISM OF ACTION AND THEIR INFLUENCE ON BLOOD COAGULATION                                                                    by  Manu Thomas Kalathottukaren  B.Pharm; The Tamil Nadu Dr. M.G.R. Medical University, 2004 M.Pharm; The Tamil Nadu Dr. M.G.R. Medical University, 2008  A THESIS SUBMITTED IN PARTIAL FULFILLMENT OF THE REQUIREMENTS FOR THE DEGREE OF  DOCTOR OF PHILOSOPHY  in THE FACULTY OF GRADUATE AND POSTDOCTORAL STUDIES (Pathology and Laboratory Medicine)  THE UNIVERSITY OF BRITISH COLUMBIA (Vancouver)  March 2017 © Manu Thomas Kalathottukaren, 2017  ii  Abstract Heparins exert anticoagulation by potentiating anti-factor (F)Xa and anti-thrombin activity of antithrombin (AT), whereas oral anticoagulants (DOACs) directly target FXa or thrombin. FXa and thrombin are the key proteases required for blood clotting. Anticoagulants are therefore used for the prophylaxis and treatment of thrombosis and during surgeries. However, anticoagulation associated haemorrhage is a concern. The only approved antidote for unfractionated heparin (UFH), protamine do have limitations including cardiovascular complications. No approved antidotes are available for low molecular weight heparins, fondaparinux and direct FXa inhibitors. Therefore, there is a need for antidotes that are nontoxic and efficient. In this thesis, we reveal the mechanism of action, hemocompatibility and efficiency of three antidote molecules that are under development.  UHRA: UHRA is a synthetic antidote developed by the Kizhakkedathu laboratory at the UBC. Thermodynamics based on isothermal titration calorimetry (ITC) and fluorescence studies revealed the molecular design of UHRA, the importance of steric shield produced by PEG brush, the selectivity of UHRA against heparins and its mechanism of action. Clotting studies confirm the antidote activity of UHRA. Studies also show that UHRA even in the absence of heparins, do not interact with fibrinogen, alter fibrin polymerization or abrogate blood clotting. Unlike protamine, UHRA does not incorporate into blood clots, form clots with normal morphology, and lysis profile. Studies in mice reveal that UHRA reverses UFH anticoagulant activity without the lung injury as seen with protamine. Studies confirm the superiority of UHRA compared to protamine. iii  Andexanet Alfa (AnXa) and PER977: AnXa is a truncated FXa recombinant protein developed by Portola Pharmaceuticals and PER977 is a small cationic molecule developed by Perosphere Pharmaceuticals. ITC confirms high-affinity binding of AnXa to heparin/AT complex and to DOACs studied. PER977 shows weak binding to heparins and no binding to DOACs tested. Both antidotes do not influence fibrin polymerization, fibrin and blood clot architecture even in the absence of anticoagulants. Electron micrographs of blood clots containing edoxaban treated with AnXa or PER977 reveal restoration of impaired fibrin formation. However, in clotting assays, PER977 failed to show antidote activity, whereas AnXa neutralized the anticoagulation activity of all tested anticoagulants.            iv  Preface Ethics approval was received from the University of British Columbia (UBC) for studies conducted at the Centre for Blood Research (UBC Ethics approval no: H07-02198). All ethics approval for animal experiments was obtained from the UBC and complied with the Canadian Council of Animal Care guidelines and the animal care protocol (A13-0195). This research project was conducted under the supervision of Dr. Jayachandran N. Kizhakkedathu (Centre for Blood Research, UBC, Vancouver) and Dr. Charles A.Haynes (Department of Chemical and Biological Engineering, UBC, Vancouver)   Some sections of the Chapter 1 have been submitted for a book chapter entitled “Mechanisms of blood coagulation in response to biomaterials: extrinsic factors”. I was responsible for conducting the literature review, conceptualizing and writing the book chapter. The book chapter is under review. Introduction to the macromolecular antidote, Universal Heparin Reversal Agent (UHRA) is based on the study published Sci. Transl. Med. 6, 260ra150, 2014. I was responsible for thromboelastography, anti-factor Xa assays, isothermal titration calorimetry and fluorescence spectroscopy experiments.  Chapter 2 is based on the study conducted in collaboration with Dr. Charles A. Haynes laboratory at the department of Chemical and Biological Engineering, UBC, Vancouver. This study provides information regarding the mechanism of anticoagulant neutralization activity of UHRA. I was responsible for synthesizing antidote molecules, designing experiments, conducting isothermal titration calorimetry experiments, anticoagulant neutralization assays, fluorescence spectroscopy, ELISA for antithrombin and size measurements of antidote-heparin complexes using dynamic light scattering. Srinivas Abbina synthesized the precursor polymer of one antidote molecule. Kai yu v  performed atomic force microscopy to determine the size of antidote-heparin complexes. A manuscript has been drafted in which I am the first author.  Chapter 3 is based on the study conducted in collaboration with Dr. Charles A. Haynes laboratory at department of Chemical and Biological Engineering, UBC and Portola Pharmaceuticals Inc. California, USA. This study examines the mechanism of action of antidote molecules Andexanet Alfa, PER977 and UHRA developed for neutralizing anticoagulants by identifying its binding partners, and by studying its influence on blood coagulation and clot structure. I was responsible for conducting all in vitro anticoagulant neutralization and scanning electron microscopy of fibrin and whole blood clots. A manuscript has been drafted in which I am the first author.  Chapter 4 is based on the study published in the journal Blood (DOI: 10.1182/blood-2016-10-747915). This work examines the influence of the UHRA molecule on fibrinogen, clotting and clot structure. I was responsible for conducting spectroscopy, clotting assays, isothermal titration calorimetry, scanning electron microscopy and preparing samples for confocal microscopy. Libin Abraham performed confocal microscopy. Benjamin FL Lai performed the thrombin generation assay. Dr. Libin Abraham and Dr. Piyushkumar R. Kapopara, provided assistance for animal studies. Conflict of interest statement The data presented in chapter 3 of the thesis on Andexanet Alfa and PER977 in part was funded by Portola Pharmaceuticals Inc. They also provided some reagents for the study.  vi  Publications 1) Kalathottukaren MT, Abraham L, Kapopara PR, Lai BFL, Shenoi RA, Rosell FI, Conway EM, Pryzdial ELG, Morrissey JH, Haynes CA and Kizhakkedathu  JN*. Alteration of blood clotting and lung damage by protamine are avoided using the heparin and polyphosphate inhibitor UHRA. Blood. (129), 1368-1379, 2017. 2) Kalathottukaren MT, Creagh LA, Abbina S, Lu G, Karbarz MJ, Pandey A, Conley PB, Kizhakkedathu JN* and Haynes CA.* Comparing mechanisms of action of anticoagulant antidotes through analysis of binding partners, clot formation and clot structure (manuscript in preparation). 3) Kalathottukaren MT, Abbina S, Yu K, Shenoi RA, Creagh LA, Haynes CA* and Kizhakkedathu JN*. Molecular Design and Mechanism of Action of the Nontoxic Universal Heparin Reversal Agent UHRA (manuscript in preparation). 4) Shenoi RA, Kalathottukaren MT, Travers RJ , Lai BFL, Creagh LA, Lange D, Yu K, Weinhart M, Chew B,  Du C, Brooks DE, Carter CJ, Morrissey JH, Haynes CA and Kizhakkedathu JN.  Affinity-based design of a synthetic universal reversal agent for heparin anticoagulants. Sci. Transl. Med. 6, 260ra150, 2014. 5) Travers RJ, Shenoi RA, Kalathottukaren MT, Kizhakkedathu JN and Morrissey JH. Nontoxic polyphosphate inhibitors reduce thrombosis while sparing hemostasis. Blood. 22(124), 2014.    vii  Table of contents      Abstract ...................................................................................................................... ii Preface ...................................................................................................................... iv Table of contents .................................................................................................... vii List of tables ........................................................................................................... xiii List of figures ......................................................................................................... xiv List of symbols ....................................................................................................... xix List of abbreviations ............................................................................................... xx Acknowledgements .............................................................................................. xxii Dedication ............................................................................................................. xxiv Chapter 1: Introduction ............................................................................................ 1 1.1 The blood coagulation cascade..................................................................... 1 1.1.1 Initiation phase .......................................................................................... 3 1.1.2 Amplification phase ................................................................................... 5 1.1.3 Propagation phase .................................................................................... 5 1.2  Endogenous inhibitors of blood coagulation .............................................. 6 1.2.1 Antithrombin .............................................................................................. 6 1.2.2 Activated protein C .................................................................................... 7 1.2.3 Tissue factor pathway inhibitor .................................................................. 9 viii  1.3 The fibrinolytic system ................................................................................... 9 1.3.1 Inhibitors of fibrinolysis ............................................................................ 11 1.3.2 Alterations in fibrin clot structure and fibrinolysis ..................................... 11 1.4 Thrombosis: a global health concern ......................................................... 16 1.4.1 Role of FXII in the pathogenesis of thrombosis ....................................... 17 1.4.2 Newly identified prothrombotic polyanions ............................................... 19 1.4.2.1 Polyphosphate .................................................................................. 19 1.4.2.2 Cell-free nucleic acids and neutrophil extracellular traps .................. 20 1.5 Heparin-based anticoagulants ..................................................................... 23 1.6 Direct oral anticoagulants (DOACs): Factor Xa inhibitors ........................ 28 1.7 Reversing anticoagulation activity: Current options and recent .............. 32 developments ...................................................................................................... 32 1.7.1 Aripazine or ciraparantag (PER977) ........................................................ 35 1.7.2 Andexanet Alfa ........................................................................................ 37 1.7.3 Universal Heparin Reversal Agent (UHRA) ............................................. 38 1.8 Rationale and specific aims ......................................................................... 43 1.8.1 Role of PEG chains in the mechanism of action of UHRA ....................... 43 1.8.2 Validating the mechanism of action of UHRA, PER977 and Andexanet Alfa ................................................................................................................... 45 1.8.3 UHRA is a superior heparin antidote compared to protamine .................. 46 ix  Chapter 2: Molecular design, origin of selectivity and functional mechanism of the nontoxic universal heparin reversal agent UHRA .......................................... 47 2.1 Synopsis ........................................................................................................ 47 2.2 Background ................................................................................................... 48 2.3 Methods ......................................................................................................... 58 2.4 Results and discussion ................................................................................ 68 2.4.1 UHRA has a large core due to multi-valent ligands randomly displayed on the core ........................................................................................................ 68 2.4.2 Binding of PS to UFH shows classic electrostatic-association behavior .. 73 2.4.3 At physiologic pH, the strong Fel from the high core of UHRA is balanced by the entropic repulsion of the brush ............................................... 78 2.4.4 Binding of UHRA to UFH diminishes non-linearly with 𝐈 and pH .............. 80 2.4.5 Influence of antidote architecture on blood clotting, neutralization of UFH anticoagulation activity and mechanism of action ............................................. 83 2.4.6 Influence of antidote architecture on the size of antidote-heparin complexes ........................................................................................................ 87 Chapter 3: Comparing mechanisms of anticoagulant antidotes ........................ 93 3.1 Synopsis ........................................................................................................ 93 3.2 Background ................................................................................................... 94 3.3 Methods ......................................................................................................... 96 3.4 Results ......................................................................................................... 101 x  3.4.1 ITC reveals binding partners common and unique to the three antidotes ....................................................................................................................... 101 3.4.2 Binding data supports different mechanisms of action........................... 110 3.4.3 Fibrin polymerization and clot imaging permits identification of antidote effects on clot morphology .............................................................................. 112 3.4.4 SEM studies reveal the efficacy of antidotes against direct FXa inhibitors in restoring impaired fibrin fiber development in anticoagulated blood ........... 117 3.4.5 Clotting assays define anticoagulant reversal activity of antidotes ........ 119 3.5 Discussion ................................................................................................... 122 Chapter 4: Alteration of blood clotting and lung damage by protamine are avoided using the heparin and polyphosphate inhibitor, UHRA ...................... 128 4.1 Synopsis ...................................................................................................... 128 4.2 Background ................................................................................................. 129 4.3 Methods ....................................................................................................... 131 4.4 Results ......................................................................................................... 145 4.4.1 Design and synthesis of a UHRA molecule that binds pro- and anticoagulant polyanions ................................................................................ 145 4.4.2 UHRA does not interact with fibrinogen or alter thrombin-mediated fibrin polymerization ................................................................................................ 148 4.4.3 UHRA has no effect on tissue factor /recalcification-initiated plasma coagulation and clot lysis ................................................................................ 154 xi  4.4.4 UHRA has negligible impact on purified fibrin clot and whole-blood clot structure.......................................................................................................... 158 4.4.5 UHRA does not bind or incorporate into purified fibrin or whole blood clots ....................................................................................................................... 164 4.4.6 UHRA reverses anticoagulant activity of UFH without lung injury and alteration in clot morphology ........................................................................... 166 4.5 Discussion ................................................................................................... 173 Chapter 5: Conclusions and future studies ........................................................ 178 5.1 Thesis summary.......................................................................................... 178 5.2  Future studies ............................................................................................ 182 5.2.1 Disruption of PF4/UFH complexes with UHRA ...................................... 183 5.2.2 Inhibitors for prothrombotic nucleic acids ............................................... 186 Bibliography .......................................................................................................... 188 Appendices ............................................................................................................ 215 Appendix A ........................................................................................................ 215 A.1 NMR spectra of N-UHRA .............................................................................. 215 A.2 NMR spectra of mPEG750-epoxide ............................................................... 216 A.3 NMR spectra of HPG-mPEG750-36 kDa ........................................................ 217 A.4 NMR spectra of mPEG750-UHRA .................................................................. 218 A.5 A Cartoon of UHRA and mPEG750-UHRA molecule ..................................... 219 xii  A.6 GPC profile and NMR spectrum of UHRA .................................................... 220 A.7 Dynamic light scattering profiles of antidote-UFH complexes ....................... 221 A.8 Count rate and settings used for DLS........................................................... 222 Appendix B ........................................................................................................ 223 B.1 NMR and mass spectrum of PER977 ........................................................... 223 Appendix C ........................................................................................................ 224 C.1 GPC profile and NMR spectrum of UHRA .................................................... 224 C.2 Characterization of Alexa-Fluor-488 UHRA and protamine .......................... 225               xiii  List of tables Table 1.1: Pharmacological features, clinical uses and dosing of heparin-based parenteral anticoagulants……………………………………………………………………..24 Table 1.2: Pharmacological features, clinical uses and dosing of direct FXa oral anticoagulants …………………………………………………………………………………31 Table 1.3: Characteristics of UHRA molecules synthesized for the preliminary heparin binding and in vitro anticoagulant reversal activity…………………………………………40 Table 2.1: Characteristics of UHRA and the two UHRA analogs synthesized………….63 Table 2.2: Thermodynamic parameters for binding of antidote molecules to UFH at different ionic strengths determined by ITC…………………………………………………75 Table 3.1: Equilibrium dissociation constants (Kds) determined by ITC for each reversal UHRA, AnXa or PER977 titrated into various potential binding partners………………102 Table 4.1: Characteristics of the UHRA molecule………………………………………..146 Table 4.2: Thermodynamic parameters for the interaction of UHRA with UFH and polyP75 determined by ITC…………………………………………………………………..148 Table A.8: Count rate, derived count rate and attenuator settings for antidote/UFH complexes measured using DLS …………………………………………………………..222       xiv  List of figures Figure 1.1: The cascade model of coagulation depicting extrinsic and intrinsic coagulation pathways…………………………………………………………………………..2 Figure 1.2: Phases of a cell-based model of coagulation…………………………………..4 Figure 1.3: Physiologic inhibitors of blood clotting……………………………………….....8 Figure 1.4: Vaso-occulsive clots are digested by the fibrinolytic system………………..10 Figure 1.5: Schematic representation of fibrin polymerization and fibrinolysis…………12 Figure 1.6: The contact clotting pathway is initiated when artificial surfaces are exposed to blood………………………………………………………………………………………….18 Figure 1.7: Novel prothrombotic biomolecules……………………………………………..22 Figure 1.8: The structure of major repeating disaccharide unit and pentasaccharide sequence in UFH and LMWHs……………………………………………………………….23 Figure 1.9: The chemical structure of fondaparinux……………………………………….27 Figure 1.10: Structure of oral direct FXa inhibitors currently in clinics; rivaroxaban, apixaban and edoxaban………………………………………………………………………29 Figure 1.11: Aminoacid residues in protamine……………………………………………..33 Figure 1.12: The chemical structure of PER977 molecule………………………………..36 Figure 1.13: The chemical structure of UHRA molecule and heparin binding group…..39 Figure 1.14: UHRA-7 reverses anticoagulation activity of fondaparinux (1.2 IU/mL)…..41 Figure 1.15: UHRA-7 prevents bleeding in mice, induced by heparin anticoagulants....42 Figure 2.1: Schematics of macro-ion interaction energetics………………………………54 xv  Figure 2.2: The structure of methylated tris(2-aminoethyl)amine ligand on the UHRA molecule………………………………………………………………………………………..71 Figure 2.3: The potentiometric titration curve for pure water (calculated; grey trace) and for UHRA in water (experimental; black trace) at RT………………………………………72 Figure 2.4: ITC thermograms for antidotes binding to UFH at two different salt concentrations…………………………………………………………………………………74 Figure 2.5: Dependence of Ka of the UHRA/UFH interaction on salt concentration…..77 Figure 2.6: Effect of temperature on the thermodynamics of UHRA/UFH interaction…78 Figure 2.7: Effect of pH on the binding affinity of UHRA/UFH system…………………..82 Figure 2.8: Effect of antidote molecules on clotting, and the ability of each molecule to neutralize the anticoagulation activity of UFH……………………………………………...84 Figure 2.9: Turbidimetric analysis of antidote-UFH complexes…………………………..88 Figure 2.10: Characterization of sizes of antidote-UFH complexes at pH 7.4 (150 mM NaCl)…………………………………………………………………………………………....89 Figure 2.11: A cartoon representing the presence of antidote molecules in heparinized blood…………………………………………………………………………………………….92 Figure 3.1: Isothermal titration calorimetry data for UHRA binding to indirect FXa inhibitors………………………………………………………………………………………104 Figure 3.2: Isothermal titration calorimetry data for andexanet alfa binding to various anticoagulants or to the AT-enoxaparin complex………………………………………...106 Figure 3.3: ITC data for AnXa or PER977 binding to edoxaban………………………..107 Figure 3.4: Isothermal titration calorimetry data for PER977 binding to various anticoagulants……………………………………………………………………………….109 xvi  Figure 3.5: AnXa and PER977 alone at therapeutic doses do not affect fibrinogen polymerization……………………………………………………………………………….113 Figure 3.6: Fibrin architecture and fiber size remain unaltered at therapeutic doses of AnXa and PER977 alone…………………………………………………………………...115 Figure 3.7: AnXa and PER977 alone do not influence blood clot morphology……….116 Figure 3.8: AnXa and PER977 normalize impaired fibrin formation and fiber diameter of edoxaban treated clots………………………………………………………………………118 Figure 3.9: Reversal of UFH and enoxaparin anticoagulation activity by AnXa and PER977……………………………………………………………………………………….120 Figure 3.10: Reversal of the anticoagulation activity of edoxaban by AnXa and PER977……………………………………………………………………………………….122 Figure 4.1: Structure of the UHRA molecule and cationic binding group (CBGs) represented as R in the structure…………………………………………………………..146 Figure 4.2: UHRA exhibits high-affinity binding to both UFH and polyP75……………..147 Figure 4.3: UHRA does not perturb the intrinsic tryptophan fluorescence of human fibrinogen……………………………………………………………………………………..149 Figure 4.4: UHRA does not change the secondary structure of human fibrinogen…..150 Figure 4.5: Influence of UHRA and PEI on secondary structure of fibrinogen measured by CD spectroscopy………………………………………………………………………….151 Figure 4.6: A binding isotherm obtained using ITC by titrating UHRA with fibrinogen.152 Figure 4.7: UHRA does not influence fibrin polymerization……………………………..153 Figure 4.8: Effect of protamine, UHRA and PEI on fibrinogen solution….....................153 Figure 4.9: Plasma clot formation and clot turbidity is unaffected by UHRA………….155 xvii  Figure 4.10: UHRA does not promote lysis of plasma clots……………………………..157 Figure 4.11: UHRA does not alter purified fibrin clot morphology or fiber diameter…..159 Figure 4.12: Clot characteristics formed in whole blood remain unchanged in the presence of UHRA…………………………………………………………………………...161 Figure 4.13: UHRA does not influence thrombin activity and thrombin generation in human platelet-rich plasma…………………………………………………………………162 Figure 4.14: UHRA does not show intrinsic anticoagulant activity in whole blood even in the absence of heparin……………………………………………………………………...163 Figure 4.15: UHRA does not bind or incorporate into purified fibrin, blood clots or other clot components……………………………………………………………………………...165 Figure 4.16: UHRA reverses anticlotting activity of UFH without impairing clotting even with excess levels of UHRA…………………………………………………………………167 Figure 4.17: UHRA reverses anticlotting activity of tinzaparin………………………….168 Figure 4.18: UHRA neutralizes heparin anticoagulation activity without altering clot morphology……………………………………………………………………………………169 Figure 4.19: Protamine binds or incorporates into blood clot fiber or other clot components even in the presence of UFH………………………………………………...170 Figure 4.20: UHRA reverses anticlotting activity of UFH with no adverse effect on lung ultrastructure or clot morphology…………………………………………………………...173 Figure 5.1: A cartoon depicting the pathogenesis of HIT and the possible use of UHRA to disrupt PF4/UFH complex………………………………………………………………..184 Figure 5.2: PF4 is released from PF4/UFH complexes treated with UHRA…………...185 Figure 5.3: Novel multivalent ligand with more tertiary amine groups…………………186 xviii  Figure A.1: NMR spectra of N-UHRA…………………………………………………......215 Figure A.2: NMR spectra of mPEG750-epoxide……………………………………..........216 Figure A.3: NMR spectra of HPG-mPEG750-36 kDa……………………………………..217 Figure A.4: NMR spectra of mPEG750-UHRA……………………………………….........218 Figure A.5: A cartoon depicting the UHRA and mPEG750-UHRA molecules………....219 Figure A.6: Characterization of UHRA……………………………………………………220 Figure A.7: Intensity size distributions, volume size distributions and autocorrelation functions obtained for antidote-UFH complexes………………………………………....221  Figure B.1: Characterization of PER977 Structure……………………………………...223 Figure C.1: 1H-Nuclear magnetic resonance spectrum of UHRA and Gel-permeation chromatography elution profile of HPG-PEG polymer of molecular weight 25kDa in 0.1M sodium nitrate………………………………………………………………………………..224 Figure C.2: Purification of Alexa Fluor 488 conjugated protamine by gel filtration chromatography and heparin neutralizing activity of Alexa Fluor 488 conjugated UHRA and protamine……………………………………………………………………………….225       xix  List of symbols °C            Degree Celsius  g              Gram  h              Hour K             Kelvin kg            Kilogram  mg           Milligram  mL           Milliliter  mM          Millimolar  g            Microgram  L            Microliter nm           Nanometer nM           Nanomolar pM           Picomolar xx  List of abbreviations AFM             Atomic force microscopy ANOVA        Analysis of variance aPTT            Activated partial thromboplastin time AnXa           Andexanet Alfa AT                Antithrombin CHES           2-(cyclohexylamino) ethanesulfonic acid DLS              Dynamic light scattering DMSO          Dimethyl sulfoxide  EDTA           Ethylenediamine tetraacetic Acid ELISA          Enzyme linked immunosorbent assay FDA             Food and Drug Administration  HPG            Hyperbranched polyglycerol HEPES        N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid HIT              Heparin-induced thrombocytopenia ITC              Isothermal titration calorimetry  i.v.               Intravenous  IU                International unit kDa             Kilo Dalton LMWHs      Low molecular weight heparins MOPS        4-morpholinepropanesulfonic acid mPEG        Methoxypolyethylene glycol MW             Molecular weight  xxi  NIH            National institutes of health OD              Optical density  PF4            Platelet factor 4 PBS            Phosphate-buffered saline PolyP         Polyphosphate PPP            Platelet-poor Plasma  PRP           Platelet-rich Plasma  PS              Protamine sulfate (protamine) RBC           Red blood cell  RES           Reticuloendothelial System  Rh             Hydrodynamic radius  RT             Room temperature  SD            Standard deviation  s.c.            Subcutaneous  TEG          Thromboelastography tPA            Tissue plasminogen activator UHRA       Universal Heparin Reversal Agent UFH          Unfractionated heparin      xxii  Acknowledgements I thank God for showering tons of blessings upon me and leading me in a righteous path. I trust in you, God.  This research thesis would not have been possible without the support from my colleagues at the Center for Blood Research (CBR), loving friends in Brooks and Kizhakkedathu laboratories and teachers. I would like to thank my supervisor Dr. Jayachandran N. Kizhakkedathu and co-supervisor Dr. Charles A.Haynes, for giving me an opportunity to learn and do research in their laboratory. Thank you for having confidence in me, which gave me the energy to do hard work and achieve progress. A million thanks for your patience in our weekly meetings, and for constantly providing guidance and motivation, whenever needed.    I would like to thank Dr. Edward L.G. Pryzdial (Thesis chair), Dr. Cedric J. Carter (Thesis committee member), Dr. Donald E. Brooks and Dr. Edward M. Conway, for their valuable suggestions and thoughts on this project. Dr. Pryzdial, thank you very much for helping to overcome my “stage phobia”.     Special thanks to Dr. Rajesh A. Shenoi for being my mentor in the lab and sharing your knowledge. You are an inspiration for me and other students in the Kizhakkedathu laboratory. Thank you for being with me during stressful situations.   I would like to thank Dr. Libin Abraham, Dr. Frederico Rosell, Dr. Piyushkumar.R Kapopara, Dr. Srinivas Abbina, Dr. Louise A.Creagh, Benjamin Lai, Iren Constantinescu and Derrick Horne, for providing me excellent training on various techniques used for this project.  xxiii   My gratitude to Narges Hadjesfandiari, Yan Mei, Dr. Anilkumar Parambath, Dr. Mahsa Alizadeh, Dr. Johan Janzen, Dr. David Yang, Vincent Leung, Dr. Kai Yu, Dr. Nima Khademmohtaram, Dr. Imran Ul-Haq, Erika Siren, Usama Abbasi, Chanel La, Sreeparna Vappala, Na Li (April), Erika Das, Dr. Rafi Chapnian, Dr. Madhab Bagpai, Prashant Kumar, Irina Chafeeva and all past lab members of Brooks and Kizhakkedathu laboratories, for creating a fun-filled work environment.          I am grateful to Rolinda Carter, Scott Meixner, Brauna Culibrk, Ahmad Arbaeen, Christa Klein-Bosgoed, Deborah Chen, Dr. Katherine Serrano and Elena Levine, for sharing knowledge and skills.          I would like to thank the CBR for the graduate student award 2014 and providing research facilities. Also, I would like to thank my former supervisors, Prof. Francis M. Saleshiar and Dr. Manoj P.Kumar, for teachings and providing career guidance.  Special thanks to Prashant chettan, Ann Mary Chacko chechi and Anitha chechi, for support and encouragement. Thank you to all blood donors.         I owe everything to Ammachi, Appachan, Karolin, Rose mol and Santheep, for their unconditional love and prayers        xxiv        Dedication    1) To Ammachi, Appachan, Karolin, Rose mol, Santheep and Rebecca vava 2) To Cedric, Chip, Don, Ed and Jay       1Some sections of the chapter 1 have been submitted for a book chapter entitled” Mechanisms of blood     coagulation in response to biomaterial: extrinsic factors”. The book chapter is under review.               1 Chapter 1: Introduction 1.1 The blood coagulation cascade Hemostasis is a physiological mechanism that seals damaged blood vessels, maintains blood fluidity and degrades blood clots after restoration of vascular integrity [1]. In the hemostatic system, the blood coagulation process arrests bleeding at the injured sites [1]. Commonly the blood coagulation process is outlined as a Y-shaped schematic diagram, with intrinsic and extrinsic pathways. [2,3]. The pathways converge at the level of formation of the prothrombinase complex (Factor (F) Xa: FVa) [Figure 1.1]. Thrombin (FIIa) is the final protease generated in the cascade, which polymerizes fibrinogen into insoluble fibrin fibers; a major component of the hemostatic blood clot [1]. The coagulation proteases of the intrinsic pathway (FXII, FXI, FIX and FVIII) are all present in the blood; whereas TF is expressed on the subendothelial matrix, which is exposed to FVII- in blood upon vascular injury [4,5]. The intrinsic and extrinsic pathways are evaluated clinically by performing activated partial thromboplastin time (aPTT) and prothrombin time (PT), respectively [6]. However, these diagnostic coagulation tests are not reliable predictors of coagulation disorders. For example, deficiency of FXII in plasma prolongs the aPTT; however, individuals with isolated FXII deficiency never exhibit hemorrhage [7]. This suggests that there is cross-talk between these pathways. Notably, the TF:FVIIa complex can also activate FIX, included in the intrinsic pathway [5]. Hence, under in vivo conditions these pathways could not operate as independent systems, as depicted by the cascade or waterfall model.  2   Figure 1.1: The cascade model of coagulation depicting extrinsic and intrinsic coagulation pathways.  Extrinsic pathway is initiated upon vascular injury. TF present on the subendothelium is exposed to blood. TF exposure enables the activation of FVII and in complex with the activated form, FVIIa (TF:FVIIa), provides initiating extrinsic tenase activity. The extrinsic tenase complex activates FX into FXa. On the other hand, FXIIa activates FXI to FXIa, which in turn activates FIX. 3  FIXa-FVIIa complex (intrinsic tenase complex) will activate FX into FXa. FXa generated by both pathways then binds to cofactor FVa to form the prothrombinase complex. The prothrombinase complex converts prothrombin into thrombin. Thrombin converts soluble monomeric fibrinogen to fibrin, which polymerizes to form a stable hemostatic blood clot. In addition, thrombin activates cofactor V and VIII (not shown in the figure 1.1), and FXI. HK indicates high molecular weight kininogen; PK, prekallikrein. Recently, a cell-based model of coagulation is widely used to describe mechanisms of in vivo blood coagulation.  According to this model, coagulation can be divided into three phases: initiation, amplification and propagation phases [8, 9]. 1.1.1 Initiation phase  Platelets are small anuclear blood cells that play a significant role in blood clot formation at the site of injury [10]. Upon damage to the vascular endothelium, circulating platelets adhere to the subendothelial matrix through interactions between von Willebrand factor (vWF) and collagen present on the subendothelial matrix and GPIb-IX-V and GPVI receptors expressed on the platelet surface, respectively [10]. This leads to the formation of partially activated platelet plug on the ruptured endothelium. In addition, TF expressed on cell membranes of subendothelial matrix is exposed to FVIIa in blood [Figure 1.2 A]. The TF binds FVIIa and forms the TF: FVIIa complex that proteolytically cleaves FIX and FX to FIXa and FXa, respectively [8]. FXa formed on the TF-bearing cell binds its cofactor protein, FVa, to form prothrombinase (FVa: FXa), an enzymatic complex which is 300,000-fold more active than FXa in catalyzing prothrombin (FII) activation into thrombin (FIIa) [11].    4   Figure 1.2: Phases of a cell-based model of coagulation. (A) Platelets adhere to the subendothelial matrix creating a platelet plug. FXa is generated on TF-bearing cells by the TF: FVIIa complex and leads to the formation of prothrombinase, FVa:FXa. Small amounts of thrombin generated initiates coagulation.  (B) Thrombin on the surface of TF-bearing cell activates nearby platelets. In addition, thrombin activates cofactor V and VIII, and FXI. (C) Propagation phase occurs on the surface of activated platelets. Localization of FXIa and FIXa on the platelet surface, amplifies FX activation, and subsequent prothrombinase formation, leading to a burst of thrombin required for effective blood clot formation. 5  1.1.2 Amplification phase Even though the coagulation reaction is effectively triggered, the small amount of thrombin generated on the TF-bearing cells is not sufficient to produce a stable hemostatic blood clot [8]. However, thrombin generated on the TF-bearing cells will further activate platelets, leading to aggregation and release of soluble platelet activation agonists and coagulation factors from their secretory granules [12]. In addition, thrombin amplifies the coagulation reaction by converting platelet-derived FV to FVa and by generating FVIIIa, through dissociation of vWF:FVIII complex [1,8]. FXI in plasma is also activated by thrombin, bypassing the need for FXIIa [Figure 1.2 B] [13,14].  1.1.3 Propagation phase In the physiological setting, the propagation phase occurs on the surface of activated platelets, as procoagulant complexes of coagulation such as prothrombinase require negatively charged surfaces for optimal catalytic activity [15]. By exposing phosphatidylserine, activated platelets, facilitate procoagulant complex assembly on its surface. Cofactors FVa and FVIIIa localize on to the surface of activated platelets [16]. In addition, FIXa in the fluid phase, generated by the TF:FVIIa complex (mentioned in the initiation phase), and thrombin activated FXIa binds to the surface of activated platelets [17,18]. In contrast to FXa, the FIXa in the fluid phase is not inhibited by tissue factor pathway inhibitor (TFPI) [19]. Moreover, FIXa displays enhanced resistance to inhibition by antithrombin (AT) in comparison to FXa [20].   FXIa bound to the platelet surface activates more FIX, thereby generating FVIIIa: FIXa tenase complex on the platelet surface, which is ~50 times more efficient in catalyzing FX activation than the TF: FVIIa complex [21]. The crucial role of the intrinsic 6  tenase complex (FVIIIa:FIXa) in the coagulation reaction is evident from the fact that, people with hemophilia A (deficiency in FVIII) and hemophilia B ((deficiency in FIX) exhibit severe haemorrhagic abnormalities [21].    FXa combines with its cofactor FVa on the activated platelet membrane surface to form prothrombinase complexes, leading to a burst of thrombin generation [Figure 1.2 C]. Thrombin cleaves fibrinogen, into soluble fibrin monomers that subsequently polymerize to stable fibrin clot [22]. Also, thrombin activates FXIII to FXIIIa, which cross-links fibrin fibers of blood clots, and augments clot strength and stability [23]. This is indispensable for blood clots to stop bleeding from injury sites.  1.2   Endogenous inhibitors of blood coagulation Hemostasis is a tightly controlled mechanism that comprises procoagulant and anticoagulant players. Any perturbation to this balance might lead either to thrombosis or haemorrhage. Upon injury to a blood vessel, clotting is triggered to produce fibrin clots that seal the injury site, and halt bleeding from the damaged blood vessel. However, uncontrolled clotting could generate clots that could obstruct the normal blood flow (thrombosis). Natural anticoagulant proteins act as brakes on clotting reactions and restrict clot formation to the specific site of injury. There are three major anticoagulant proteins that regulate blood coagulation.  1.2.1 Antithrombin Antithrombin (AT), a serine protease inhibitor (serpin), forms 1:1 stoichiometric complexes with serine coagulation proteases such as FIIa, FIXa and FXa to inhibit their activity [24] [Figure 1.3]. AT is the most important physiological inhibitor of blood coagulation; inherited or acquired deficiency of AT increases the risk of thrombosis [25]. To achieve maximal protease inhibition, AT requires a cofactor. Proteoglycans such as 7  heparan sulphate expressed on the surface of endothelium and in the subendothelial matrix, binds to AT and induces a conformational change which dramatically increases the coagulation protease inhibitory activity of AT [26]. The long polymeric heparin moieties may also provide a scaffolding effect to increase the rate of inhibition of certain clotting factors. These are also the mechanisms of action for heparin-based anticoagulants as these drugs bind to AT for anticoagulation activity [27].   1.2.2 Activated protein C Protein C activation is an important step in the regulation of coagulation [Figure 1.3]. Like most clotting proteins, protein C is synthesized in the liver. It is present in plasma at a concentration of 70 nM [28]. Thrombin is the key protease that activates protein C. Thrombin binds thrombomodulin (TM), an integral membrane protein present on the endothelium to form a thrombin-thrombomodulin complex (T-TM complex) [29]. TM occupies the crucial exosite I of thrombin, and as a result thrombin loses its procoagulant activity. The T-TM complex converts protein C into activated protein C (APC). In addition, the binding of protein C to endothelial protein C receptor (EPCR) assists generation of APC [30]. APC along with its cofactor, protein S, proteolytically cleaves FVa and FVIIIa, respectively. The clinical significance of APC is evident from the fact that a dysfunctional protein C pathway contributes to the thrombotic conditions such as in sepsis [31]. 8    Figure 1.3: Physiologic inhibitors of blood clotting. To maintain normal blood flow in the vasculature there are inhibitors of clotting cascade. The most important anticoagulant protein is a serine protease inhibitor, antithrombin (AT). AT inhibits mainly FIXa, FXa and thrombin. However, AT requires cofactor for complete inhibitory activity. Proteoglycans containing heparan sulphate expressed on endothelial surface, bind AT and induce a conformational change on AT and the target protease, thus acting as an endogenous cofactor of AT. Activated protein C (APC) along with its cofactor protein S (Prot S) inhibits the activity of FVIII/FVIIIa and FV/Vaa, respectively. Thrombin is the key protease involved in the protein C activation pathway. Thrombin loses its procoagulant activity when it binds thrombomodulin (TM). The thrombin–thrombomodulin complex (T-TM) activates protein C (Prot C) bound to endothelial protein C receptor (EPCR) to APC. TFPI binds TF-FVIIa-FXa complexes and inhibit their procoagulant activity.    9  1.2.3 Tissue factor pathway inhibitor Another important molecule that is synthesised primarily by endothelial cells and megakaryocytes is the tissue factor pathway inhibitor (TFPI). TFPI exists in two major isoforms: TFPIα and TFPIβ [19] [Figure 1.3]. TFPIα is the predominant isoform expressed on platelets and TFPIβ is mainly expressed on endothelial cells. Both isoforms of TFPI bind TF-FVIIa-FXa ternary complexes and inhibit TF-FVIIa and FXa through distinct mechanisms. Very recently, it has been shown that TFPIα can also inhibit activity of prothrombinase assembled on platelet surfaces [32].   1.3 The fibrinolytic system Upon healing of the injured blood vessel, the clot must be removed to avoid vaso-occulsion and to maintain blood fluidity. Clots are lysed by the fibrinolytic protease, plasmin, generated from the zymogen plasminogen [Figure 1.4] [33]. Plasminogen is converted into plasmin by tissue plasminogen activator (tPA) and urokinase plasminogen activator (uPA) [34]. tPA, synthesized and released primarily by endothelial cells, is the principal activator of plasminogen for vascular fibrinolysis, whereas uPA produced by monocytes, macrophage and urinary epithelium, is predominantly involved in tissue remodeling localized through its cofactor, the uPA receptor (uPAR) [34,35]. tPA is a weak activator of plasminogen on its own; however, upon colocalization with plasminogen on the fibrin surface the catalytic efficiency of tPA increases at least two to three orders of magnitude [36]. Unlike tPA, uPA has a low binding affinity towards fibrin and does not require fibrin colocalization for plasminogen activation [36, 37].  10   Figure 1.4: Vaso-occulsive clots are digested by the fibrinolytic system. Tissue plasminogen activator (tPA) is the major serine protease that converts plasminogen (PLG) into plasmin (PN). tPA is synthesized primarily by endothelial cells and released into circulation. tPA and plasminogen (PLG) bind fibrin through C-terminal lysine residues. This trimolecular assembly enhances plasmin (PN) generation leading to removal of fibrin clots.   The plasmin generated on the fibrin surface, cleaves fibrin and exposes carboxy-terminal lysine residues. Both tPA and plasminogen contain lysine-binding sites and bind to C-terminal lysine residues on fibrin clots. This trimolecular assembly enhances plasmin generation and leads to complete digestion of fibrin [34,37]. Hence, fibrin acts 11  as a catalytic surface and augments its own degradation by exposing C-terminal lysine residues.   Studies suggest that plasmin possesses additional physiological functions. Plasmin triggers contact clotting system by activating FXII, leading to production of PK [38]. In addition, plasmin mediates complement activation in animal models of thrombosis [39]. 1.3.1 Inhibitors of fibrinolysis Fibrinolysis is highly regulated in order to control unwanted plasminogen activation and to confine the activity of plasmin to blood clots that may cause vascular occlusion. Like anticoagulants, fibrinolysis inhibitors can be grouped into serine protease inhibitors (serpins) and non-serine protease inhibitors [34]. Serpins such as plasminogen activator inhibitor type-1 (PAI-1) and plasminogen activator inhibitor type-2 (PAI-2) regulate fibrinolysis by rapidly inhibiting plasminogen in the circulation, whereas alpha-2 plasmin inhibitor (α2PI) is a serpin and binds to plasmin with 1:1 stoichiometry, inhibiting its activity [34, 40]. Fibrin modulates the activity of these inhibitors. For instance, plasmin bound to fibrin is inhibited slowly by α2PI compared to free plasmin in circulation [40].  Thrombin-activatable fibrinolysis inhibitor (TAFIa) or carboxypeptidase U is an enzymatic inhibitor [41]. It cleaves C-terminal lysine residues in fibrin and thus reduces the number of plasminogen and tPA binding sites on fibrin. This enzyme circulates as a zymogen and is activated by the thrombin-thrombomodulin complex [41].  1.3.2 Alterations in fibrin clot structure and fibrinolysis Fibrinogen is a large glycoprotein with a molecular weight of 340 kDa, and is the third most abundant plasma protein (6-12 µM) [42]. Fibrinogen consists of two identical sets of three polypeptide chains (Aα:Bβ:γ)2, connected by 29-disulphide bonds. The N-12  termini of the six polypeptide chains converge at the center of the molecule, called the E-domain [Figure 1.5].   Figure 1.5: Schematic representation of fibrin polymerization and fibrinolysis. Conversion of fibrinogen into fibrin clot is mediated by thrombin. Cleavage of fibrinopeptide A from fibrinogen 13  results in the formation of protofibrils. Subsequent cleavage of fibrinopeptide B and intermolecular αC-domain interaction facilitates lateral aggregation of protofibrils into insoluble fibrin clot. Plasmin degrades fibrin into fragments of different molecular weights, collectively referred as fibrin degradation products (FDP).  The C-termini of the Bβ and γ chains, terminates in a globular D-domain. The central E-domain is connected to two D-domains by 2 coiled–coils composed of 3 chains. The C-termini of Aα chains protrude from the D-domain, and are located near the central E-domain of the molecule and interact intra-molecularly [43-45].   Polymerization of fibrinogen into fibrin is mediated by thrombin in two steps. First, thrombin binds to the central E-domain of the fibrinogen, and catalyzes the release of fibrinopeptide A (FpA); a 16 residue peptide from the N-termini of Aα chains. This exposes a new binding site (Gly-Pro-Arg) on the E-domain, called as the ‘A’ knob. The knob ‘A’ non-covalently interacts with the complementary binding pocket (‘a’ hole) exposed in the γ-chain of the D-domain of another fibrin molecule. This A:a interaction results in the formation of half-staggered, double-stranded protofibrils [43-45].  In the second step, thrombin cleaves fibrinopeptide B (FpB; 14 residue peptide) from the N-termini of Bβ chains. This exposes a new binding site (Gly-His-Arg-Pro), called the knob ‘B’. Subsequently, knob ‘B’ interacts with the complementary binding pocket, ‘b’ hole exposed in the Bβ chains of the D-domain. This B:b interaction, together with αC-domain intermolecular interaction between fibrin molecules, facilitates the lateral association of protofibrils into insoluble fibrin fibers [43,44]. However, the precise mechanism of lateral aggregation of fibrin protofibrils is not fully understood. Interestingly, the kinetics of fibrinopeptide release is dictated by the conformational state of fibrinogen [43,46]. In fluid phase (fibrinogen in blood), the rate of release of FpA from fibrinogen is much faster in comparison to surface-bound (fibrinogen bound to vascular 14  grafts or bound to activated platelets and subendothelial proteins), where FpB is cleaved rapidly by thrombin [43,47]. The insoluble fibrin clot formed is degraded by plasmin into soluble fragments of different molecular weights, collectively referred as fibrin degradation products (FDP) [48]. First, plasmin cleaves αC domains and then multiple cleavages occur between D and E regions generating fragments of αC domains and E3 and complexes such as D-D: E1 complexes and D-D dimers [49].  The rate of clot lysis is correlated to the fibrin clot structure, which in turn is dependent on numerous factors such as concentration of thrombin, fibrinogen, ionic conditions, etc. [50]. The concentration of thrombin has a significant influence on the fibrin clot structure. High thrombin levels produce clots with thin, densely packed fibers with reduced pore size, whilst low thrombin levels produce clots with thick, loosely packed fibers with elevated pore size [44,50]. Studies suggest that clots with a tight network of thin fibers and reduced pore size are more resistant to lysis, due to reduced permeability of fibrinolytic proteins, as well as binding of tPA and plasminogen to fibers [51]. Although, the size of pores between fibrin fibers (160-380 nm) is large enough to allow diffusion of proteins, the presence of blood cells such as platelets and red blood cells (RBCs) entrapped between fibers, reduces molecular diffusion coefficients of fibrinolytic proteins [52]. For instance, activated platelets bind strongly to fibrin through αIIbβ3 integrins, and promote formation of a dense fibrin network (platelet-mediated clot contraction), that are resistant to lysis [53]. Similarly, RBCs also participate in clot contraction by binding to fibrin fibers through integrin receptors expressed on the surface of RBCs [54]. This incorporation of RBCs into platelet-rich fibrin fibers results in the compression of RBCs to polyhedral structures which reduce clot permeability and 15  diffusion of fibrinolytic proteins, respectively [55]. However, retention of RBCs in clot requires FXIIIa, a transglutaminase enzyme activated by thrombin, which cross-links fibrin fibers [56]. Recent study shows that thrombi obtained from FXIII-/- mice have fewer RBCs incorporated than thrombi from wild-type mice [56]. FXIIIa catalyzes the formation of γ-glutamyl-ε-lysyl covalent bonds between γ 398/399 Gln and γ 406Lys residues, and between C-terminal portions of αC chains to generate γ-γ dimers and αC polymers, respectively [57]. Cross-linking of fibrin fibers provides fibrin clots with mechanical stability and augmented resistance to fibrinolysis. In addition, FXIIIa stabilizes fibrin clots by cross-linking antifibrinolytic proteins such as α2PI, TAFI and PAI onto fibrin fibers [57,58].  The clinical importance of fibrin clot structure and stability is evident from the fact that, thicker clot fibers are observed in clots formed from plasma of patients with hemophilia. These clots show decreased elastic modulus, enhanced permeability and are more prone to lysis, explaining abnormal thrombin generation and subsequent bleeding complications associated with hemophilia [59]. Recombinant activated FVIIa (rFVIIa) that ameliorate bleeding in hemophilia patients, is shown to normalize thrombin generation and provide stability to fibrin clot [60]. Studies have shown that plasma clots obtained from patients with coronary artery disease, peripheral artery disorders and thromboembolism are composed of densely packed fibers and therefore, more resistant to lysis [61-63].  Therapeutic agents used to treat cardiovascular disorders are also shown to alter clot structure. For instance, clots formed in the presence of drugs like aspirin, statins, 16  and anticoagulants, are more porous and permeable, rendering them susceptible to lysis [64-66]. 1.4 Thrombosis: a global health concern Thrombosis is the pathogenic formation of blood clots. These result in cardiovascular disorders (CVD) such as ischemic heart disease, ischemic stroke and venous thromboembolism (VTE), which are the leading causes of death worldwide [67]. The Global Burden of Diseases, Injuries and Risk Factor Study (GBD study 2010) documented 7.0 million deaths and 2.8 million deaths, due to ischemic heart disease and ischemic stroke, respectively [68]. This accounts for 1 in 4 deaths worldwide [67,69].  Notably, the incidence of ischemic heart disease and ischemic stroke, increased by 35 % and 25 %, respectively, compared to the early 90’s [68]. Another report from the American heart Association revealed that in the year 2013, CVD accounted for 800,937 deaths in the United States [70]. These reports imply that, despite our significant progress in unravelling novel molecular mechanism(s) of thrombosis and developing antithrombotic therapies, the mortality rates associated with thromboembolic complications have not subsided. Moreover, the situation will become more challenging in the future, with ageing population and increase in incidence of metabolic syndromes such as obesity, diabetes etc. Although, the GBD study did not provide data on VTE-related deaths, a study by Cohen et al, estimated 370,012 VTE-related deaths per annum in six European Union countries [71]. In United States, approximately 300,000 VTE-related deaths occur each year [72].    Thrombosis can occur in the arterial (arterial thrombosis) and venous blood circulation system (venous thrombosis) [73]. The rupture of unstable atherosclerotic lesions, particularly in the coronary circulation, exposes thrombogenic molecules such 17  as TF, and platelet activators such fibrillar collagens and lysophosphatidic acid into blood stream. The composition of arterial thrombi varies considerably and is dependent on the blood flow rate. Thrombi developed in areas of high shear and disturbed flow primarily comprises of platelets and therefore is described as ‘white thrombi’. However, due to blood flow rate alteration caused by thrombi, the regions of thrombi with low shear, comprises of fibrin and trapped red blood cells and therefore is described as ‘red thrombi’. Thus, in addition to antiplatelet therapy, anticoagulant drugs to reduce thrombin generation and profibrinolytic drugs would be beneficial for the management of arterial thrombosis [74-76]. VTE is an important cardiovascular complication affecting millions worldwide. The clinical manifestations include deep-vein thrombosis (DVT) and pulmonary embolism (PE) [77]. The pathogenetic mechanism of VTE includes acquired and genetic factors that modulate the flow of blood (stasis), coagulability of blood (hypercoagulability) and endothelial function (loss of nonthrombogenic property of endothelium). In contrast to arterial thrombi, venous thrombi are mainly composed of fibrin and red blood cells. Drugs that reduce thrombin generation i.e. anticoagulants are used for both the prevention and treatment of VTE [76,77]. 1.4.1 Role of FXII in the pathogenesis of thrombosis When blood comes into contact with negatively charged molecules or surfaces, Factor XII undergoes conformational change and subsequent activation. Once activated, FXII (FXIIa) will then sequentially activate FXI and FIX. In addition, FXIIa proteolytically cleaves plasma kallikrein (PK) into active kallikrein which then produces potent proinflammatory mediator bradykinin (BK), from high molecular weight kininogen (HK) [4,7] [Figure 1.6]. Activated FIX (FIXa) complexed with cofactor FVIIIa (FIXa:FVIIIa), converts FX to FXa, leading to thrombin production [4]. 18   Figure 1.6: The contact clotting pathway is initiated when artificial surfaces are exposed to blood. Trimolecular assembly of FXII, high molecular weight kininogen (HK) and plasma kallikrein (PK) bound to the surface of the foreign material generates FXIIa, and cleaves plasma kallikrein (PK) to generate active kallikrein. Active kallikrein produces bradykinin (BK) from HK and also activates FXII. Binding of BK to kinin B2 receptor (B2R) activates proinflammatory signaling cascade. FXIIa activates FXI to FXIa, which in turn activates FIX. FIXa-FVIIa complex (intrinsic tenase complex) will generate prothrombinase complex leading to thrombin generation and fibrin clot.  Interestingly, patients with FXII deficiency do not display bleeding diathesis, suggesting that FXII might not be necessarily involved in the primary hemostatic process [7]. Recent studies reveal emerging role/s for clotting enzymes of the intrinsic coagulation pathway, especially FXI and FXII [78]. Studies performed in mouse model 19  suggest that FXI and FXII appear to be equally important drivers of thrombosis [79]. However, currently available epidemiological evidence and studies in non-human primates suggests, FXI as a dominant player involved in the pathogenesis of thrombosis [80]. So, strategies that target and avert activation of FXI or FXII could be a beneficial antithrombotic therapy with reduced risk of bleeding.    Novel antithrombotics that target FXI or FXII and which are currently under investigation include (i) antisense oligonucleotides (ii) antibodies (iii) small molecule inhibitors, and (iv) aptamers [81-83]. However, inhibiting FXII activation or activity might be safer considering the fact that, patients with severe FXI deficiency experience mild bleeding complications with surgery or trauma [83].  1.4.2 Newly identified prothrombotic polyanions Studies provide conclusive evidence that contact activation of blood clotting is not essential for hemostasis; however, may contribute to the pathogenesis of thrombosis. The contact clotting system is triggered when FXII comes into contact with polyanionic biomolecules or surfaces. Upon binding to these charged surfaces, FXII undergoes conformational change into FXIIa, followed by activation of coagulation zymogens such as FIX and FXI leading to fibrin clot formation. Studies have identified that endogenous polyanionic molecules such as polyphosphate (polyP) and extracellular RNA and DNA possess such prothrombotic activity [84].  1.4.2.1 Polyphosphate Polyphosphate (polyP) is a highly anionic inorganic polymer composed of phosphate monomers, connected by high energy phosphoanhydride bonds [85]. PolyP of varying lengths is ubiquitously found in every prokaryotic and eukaryotic cell. Microorganisms contain long-chain polyP, composed of 100 to 1000 phosphate units in their storage 20  granules (acidocalcisomes). Conversely, polyP found in dense granules of human platelets is smaller (60 to 100 phosphate units) and less heterogeneous compared to microbial polyP [86]. Studies show that the potential of polyP to trigger contact clotting by activating FXII depends on its chain length. Long chain microbial polyP is a potent FXII activator (template mechanism) compared to shorter polyP released from platelets [87].    In addition to FXII activation, polyP has multiple acting points in the coagulation cascade that can augment thrombin generation. PolyP accelerates generation of cofactor FVa and FXIa and inhibits anticoagulant activity of tissue factor pathway inhibitor (TFPI) [88]. Moreover, polyP incorporates into fibrin clots and produce thicker fibrin fibers that are more resistant to fibrinolysis [89, 90] [Figure 1.7]. PolyP also display proinflammatory effects through numerous mechanisms involving, FXII mediated bradykinin release, binding and potentiating proinflammatory actions of histones etc. [91]. Conversely, physiological relevance of platelet polyP is evident from the fact that, patients with Hermansky-Pudlak syndrome (lack of dense granules in platelets) exhibit mild bleeding diathesis due to defective thrombin generation [91]. Hence, detailed studies are needed to uncover precise role of polyP in hemostasis and thrombosis. 1.4.2.2 Cell-free nucleic acids and neutrophil extracellular traps Endogenous macromolecules that have been implicated in the pathogenesis of thrombosis are cell-free nucleic acids (cfNAs) such as DNA [92]. cfNAs are released by cells undergoing necrotic and /or apoptotic process as well as certain strains of bacteria [92]. In addition, activation of blood-cells such as neutrophils, generate web like structures known as neutrophil extra cellular traps (NETs), through a unique process 21  called NETosis [93]. NETs entrap and kill infectious microorganisms and thus assist our immune system in its fight against infections [94]. However, recent studies show that NETs initiate pathologic thrombosis through multiple mechanisms. NET fibers are decorated with scores of prothrombotic molecules such as DNA-histone complexes (DNA acts as a scaffold for FXII activation; histones activate platelets), TF and neutrophil elastase (proteolytically cleaves and abrogates anticoagulant activity of TFPI) [95] [Figure 1.7]. Studies performed in mouse models of thrombosis show that initiation and propagation of thrombosis by NETs can be reduced by administering DNAase or RNAase; enzymes that disintegrate nucleic acid structures [96].  The pathophysiological relevance of NETs is evident from the fact that elevated plasma levels of cfNAs are observed in patients with disease conditions such as sepsis, malignancy, stroke and autoimmune disorders [94].    22   Figure 1.7: Novel prothrombotic biomolecules. Microbial infections result in inflammation and subsequent activation of neutrophils. Activated neutrophils release web like structures known as neutrophil extracellular traps (NETs) through a process called NETosis. NETs are mainly composed of DNA–histone complexes. NETs can trigger clotting via the contact clotting pathway. Polyanionic DNA on NETs can act as a scaffold for contact protein assembly, leading to thrombin generation. In addition, NETs possess highly procoagulant TF molecules which can activate the extrinsic clotting pathway. Microorganisms release long chain polyphosphate (polyP) which can act as a template for FXII activation. Activated human platelets release short chain polyP which potentiates clotting by enhancing thrombin generation and fibrin polymerization.      23  1.5 Heparin-based anticoagulants Heparin-based anticoagulants are widely used for the prophylaxis and treatment of thromboembolic disorders, and also in patients undergoing surgical interventions [97]. Heparin-based anticoagulants, currently used in clinics include unfractionated heparin (UFH), low molecular weight heparins (LMWHs), and the synthetic pentasaccharide (fondaparinux) [Table 1.1] [97]. Heparins are a heterogeneous mixture of sulfated polysaccharides, comprised of repeating units of uronic acid -(1→4)- glucosamine residues [Figure 1.8] [98]  Figure 1.8: (A) The structure of major repeating disaccharide unit in UFH and LMWHs. (B) The structure of unique pentasaccharide sequence in UFH and LMWHs, that binds to AT with high-affinity.  24   Drug  Source Molecular weight (Da)  Pharmacokinetics Clinical indications and dosing  Adverse events       UFH       Biological     3000-30,000 (~45 saccharide units)     Bioavailability: 30 % Plasma half-life: 45-60 min# Clearance: RES and renal   1) Treatment of VTE (80 U/kg IV bolus and then 18 U/kg/h infusion) 2) Treatment of unstable angina and NSTEMI (60-70 U/kg IV bolus and then 12-15 U/kg/h infusion) 3) Prophylaxis of VTE in medical and surgical patients (5000 U SC every 8-12 h) 4) Cardiac surgery with CPB  (300 IU/kg or higher to maintain ACT>400 seconds)  1) Major bleeding (rates/year)* a) VTE prophylaxis (3.5%) b) VTE treatment (2.0%) c) ACS (4.5%) 2) HIT < 5% 3) Osteoporosis        LMWHs       Semi- Synthetic      2000-9000 (~15 saccharide units)    Bioavailability: 90-100 % Plasma half-life: 3-4 h Clearance: Renal          Dosing of enoxaparin 1) Treatment of VTE (1.5 mg/kg SC every 24 h. If CrCl< 30 mL/min ,then 1 mg/kg every 24 h) 2) Treatment of unstable angina and NSTEMI (1 mg/kg SC every 12 h) 3) Prophylaxis of VTE in medical and surgical patients (40 mg SC every 24h)  1) Major bleeding with enoxaparin (rates/year)* a) VTE prophylaxis (1.7%) b) VTE treatment (2.1%) c) ACS (4.7%) 2) HIT < 1% 3) Risk of osteoporosis is  lower than UFH       Fonda- parinux     Synthetic   1728 (5 saccharide units)   Bioavailability: 100 % Plasma half-life: 17 h Clearance: Renal  1) Treatment of VTE (7.5 mg SC daily for 50-100 kg patients) 2) Treatment of NSTEMI (2.5 mg SC daily)  1) Major bleeding (rates/year)*  a) VTE prophylaxis (2.7%) b) VTE treatment (1.2%) c) ACS (2.2%) 2) HIT (case reports) 3) Risk of osteoporosis (unknown)  25  # Due to unfavorable pharmacokinetics of UFH, the plasma half–life is dependent on the dose administered. For instance, the plasma half-life of UFH after a high dose of 400 U/kg is 150 min.  * Major bleeding indicates, bleeding from critical sites such as intracranial, intraocular and gastrointestinal (GI); bleeding associated with decrease in hemoglobin 2g/dL;  bleeding that require transfusion of 2 or more units of red cells; or fatal bleeding. Bleeding rates shown are obtained from meta-analyses or clinical trials; summarized in the reference [99] UFH indicates unfractionated heparin; CPB, cardiopulmonary bypass; ACT, activated clotting time; VTE, venous thromboembolism; ACS, acute coronary syndrome; LMWHs, low molecular weight heparins; MI, myocardial infarction; non-ST- elevation myocardial infarction; HIT, heparin-induced thrombocytopenia; RES, reticuloendothelial system; CrCl, Creatinine clearance; IV, intravenous; SC, subcutaneous; kg, kilograms; h, hour. Table 1.1: Pharmacological features, clinical uses and dosing of heparin-based parenteral anticoagulants [97,99-102].  Heparin exerts anticoagulation action, predominantly by catalyzing, the anti-FXa and anti-FIIa (thrombin) activity of AT. Heparin activates AT by inducing an allosteric alteration to the structure of AT and also by acting as a template, facilitating AT-protease interaction [103]. Allosteric activation of AT occurs, when the unique high-affinity pentasaccharide sequence present in a heparin molecule [Figure 1.8 B], binds to positively charged binding pocket in AT. This induces a conformational change in AT, and leads to expulsion of the reactive center loop (RCL) of AT. The interaction and subsequent incorporation of RCL into the protease substrate, for example, thrombin and FXa, causes its structural deformation and inactivation [104,105]. In addition, the binding of heparin with AT, exposes a specific binding site on AT to which FIXa and FXa can directly interact. This is the basis for the specificity of heparin-activated AT towards FXa and FIXa [106,107]. Studies have revealed that accelerated inhibition of thrombin by AT occurs through the formation of a ternary heparin-AT-thrombin complex [108]. Thrombin via its exosite II interacts with heparin-AT complex, and this facilitates the AT-thrombin interaction on the heparin surface, leading to inhibition of thrombin activity 26  [109]. Interestingly, only heparins with minimum 18 saccharide units can catalyze AT mediated thrombin inhibition by the template mechanism [97,102]. This explains why UFH (~45 saccharide units) has an anti-FXa-to-anti-FIIa ratio of 1:1, compared to LMWHs (~15 saccharide units) with anti-FXa-to-anti-FIIa ratios between 2:1 and 4:1 [97,102]. Furthermore, depending on the heparin chain length and therapeutic concentration, heparin can exert anticoagulation activity by activating serpins such as heparin-cofactor II (HCII), protein C inhibitor, protein Z-dependent protease inhibitor etc. [110].  HCII present in plasma is a specific thrombin inhibitor [111]. Unlike AT, allosteric activation of HCII does not require binding to the pentasaccharide sequence in heparin. In fact, the N-terminal tail of HCII interacts with the hirudin-binding thrombin exosite I, leading to thrombin inhibition. However, similar to AT, HCII undergoes conformational change upon binding to heparins, comprising a minimum of 24-saccharide units [111]. At low therapeutic concentration of heparin, thrombin is preferentially inhibited by AT; however, at high heparin concentration, thrombin inhibition by HCII is possible [97,110]. Studies have shown that antithrombotic activity of HCII, through its interaction with dermatan sulfate could offer protection against thrombotic occlusion [112,113]. The exact physiological role of HCII is still unknown. Understanding the fact that anticoagulation of activity of heparin is dependent on the specific AT-binding pentasaccharide sequence, led to the development of LMWHs and fondaparinux [110]. LMWHs are derived from UFH by enzymatic or chemical depolymerization [110]. Despite the development of LMWHs and fondaparinux, UFH remains as the anticoagulant of choice for major surgeries such as cardiac procedures 27  involving cardiopulmonary bypass [97,102]. UFH is the second most frequently used natural drug and is included in the list of essential drugs by the WHO [110].  UFH is a highly negatively charged biomolecule [98]. Only one third of its structure possesses AT binding sequence [97]. Due to the high anionic charge density, UFH binds to numerous blood proteins, endothelial cells and macrophages. This contributes to the unpredictable anticoagulant response, unfavorable pharmacokinetic properties and immunological adverse reactions (heparin-induced thrombocytopenia, HIT) associated with UFH therapy [Table 1.1] [97,102]. On the other hand, both LMWHs and fondaparinux have reduced binding to blood proteins and thus, exhibit predictable dose response and superior pharmacokinetic profile compared to UFH [Table 1.1] [97,102].   Unlike UFH and LMWHs, fondaparinux exhibits only anti-FXa activity. The unique pentasaccharide sequence (DEFGH) of fondaparinux, possesses 10 negatively charged groups. Groups specified in red color are important for the interaction with arginine and lysine residues of antithrombin [Figure 1.9] [114].  Figure 1.9: The chemical structure of fondaparinux. Groups indicated in red are critical for the interaction with AT, whereas groups indicated in blue are essential to enhance the biological activity.  28  For instance, 3-O-sulfate group in unit F of the pentasaccharide sequence is decisive for strong and specific interaction with antithrombin [114].  Even though, heparin-based anticoagulants are economical and effective for the prevention and treatment of thrombotic disorders, adverse events such as bleeding and HIT, and need for frequent coagulation monitoring are major limitations [97]. In addition, parenteral administration can cause inconvenience, mainly for outpatients on extended VTE prophylaxis after major orthopedic or cancer surgery. This may lead to poor patient compliance [115]. These shortcomings of heparin anticoagulants provided the impetus to develop novel oral anticoagulants that are less burdensome for patients and with extended clinical safety profiles. 1.6 Direct oral anticoagulants (DOACs): Factor Xa inhibitors For decades, vitamin K antagonists (VKA) such as warfarin were the only oral anticoagulant available for the prophylaxis of thromboembolic complications [116]. Although very effective, VKAs have numerous clinical limitations.  For instance, slow onset and offset of action, unfavorable pharmacokinetics and pharmacodynamics, inter-individual variability in dose response, and high incidence of intracranial bleeding etc. [116,117]. VKAs produce anticoagulation activity by modulating the γ-carboxylation of glutamate residues (Gla) of vitamin-K dependent coagulation proteases FII, FVII, FIX and FX [116]. Knowledge about the structure of FXa, led to the design and discovery of orally active small organic molecule anticoagulants. These molecules bind with high-affinity and specificity to the active site of the FXa [Figure 1.10] [118]. Dosing and clinical uses of direct FXa inhibitors, rivaroxaban, apixaban and edoxaban, is summarized in the [Table 1.2] [118] 29      Figure 1.10: Structure of oral direct FXa inhibitors currently in clinics; rivaroxaban, apixaban and edoxaban [118].  30      Drug Coagulation monitoring$   Pharmacokinetics Clinical indications and dosing* (US, EU & Canada) Adverse events and disadvantages          Rivaroxaban         PT or  Chromogenic anti-FXa assays       Bioavailability: 66 % Plasma half-life: 9-13 h Renal clearance: 33%  1) Prevention of stroke and systemic embolism in AF  (20 mg OD)  2) Treatment of acute VTE  (15 mg BID for 3 weeks, followed by 20 mg OD)  3) Secondary VTE prevention  (20 mg OD)  4) VTE prophylaxis in major orthopaedic surgery  (10 mg OD) 1) Major bleeding (rates/year)# a) Prevention of stroke and embolism in AF (ROCKET AF study)  (R 3.6% vs W 3.4%) (GI bleeding R 3.2% vs W 2.2%)  b) Treatment of acute VTE (EINSTEIN DVT & PE study)  (R 1.0% vs Enox/VKA 1.7%)  c) Thromboprophylaxis in orthopaedic surgery (ORTHO-TEP study) (R 2.9% vs F 4.9%)  2) Lack of specific reversal agents 3) Expensive than VKA      Edoxaban  (not approved in Canada)     PT or Chromogenic anti-FXa assays     Bioavailability: 62 % Plasma half-life: 9-14 h Renal clearance: 50%   1) Prevention of stroke and systemic embolism in AF  (60 mg OD)  2) Treatment of acute VTE   (60 mg OD after LMWH)   1) Major bleeding (rates/year)# a) Prevention of stroke and embolism in AF (ENGAGE-AF study)  (E 2.75% vs W 3.43%) (GI bleeding E 1.51% vs W 1.23%)  b) Treatment of acute VTE (HOKUSAI-VTE study)  (A 1.4% vs W 1.6%)  2) Lack of specific reversal agents 3) Expensive than VKA 31   $An ideal test with accuracy and precision for monitoring DOACs is not available. * Dose adjustments (not shown) must be required for renal impaired patients, based on the CrCl. # Major bleeding indicates, bleeding from critical sites such as intracranial, intraocular and gastrointestinal (GI); bleeding associated with decrease in hemoglobin 2g/dL;  bleeding that require transfusion of 2 or more units of red cells; or fatal bleeding. Bleeding rates shown are obtained from the following studies [119-126] AF indicates atrial fibrillation; A, apixaban; BID, twice daily; DVT, deep-vein thrombosis; E, edoxaban; Enox, enoxaparin; EU, European Union; F, fondaparinux; LMWH, low molecular weight heparin; PE, pulmonary embolism; PT, prothrombin time; R, rivaroxaban; VTE, venous thromboembolism; ACS, acute coronary syndrome; OD, once daily; CrCl, Creatinine clearance; VKA, vitamin K antagonists; W, warfarin; US, United States. Table 1.2: Pharmacological features, clinical uses and dosing of direct FXa oral anticoagulants [117,118].    Drug Coagulation monitoring$ Pharmacokinetics Clinical indications and dosing* (US, EU & Canada) Adverse events and disadvantages       Apixaban      PT or Chromogenic anti-FXa assay     Bioavailability: 50 % Plasma half-life: 8-15 h Renal clearance: 25%  1) Prevention of stroke and systemic embolism in AF  (5 mg BID)  2) Treatment of acute VTE  (10 mg BID for 1 week, followed by 5 mg BID)  3) Secondary VTE prevention  (2.5 mg BID)  4) VTE prophylaxis in major orthopaedic surgery  (2.5 mg BID) 1) Major bleeding (rates/year)# a) Prevention of stroke and embolism in AF (ARISTOTLE study)  (A 2.13% vs W 3.09%)  b) Treatment of acute VTE (AMPLIFY study)  (A 0.6% vs Enox+W 1.8 %)  c) Thromboprophylaxis in orthopaedic surgery (ADVANCE 1-3 study) (A 0.8% vs Enox 0.7%)  2) Lack of reversal agents 3) Expensive than VKA 32  FXa, an important component of the prothrombinase complex, plays a crucial role in converting prothrombin into thrombin [15]. Direct FXa inhibitors such as rivaroxaban bind to the S1 and S4 active sites of FXa, thereby abrogating the procoagulant activity of FXa [118]. Unlike AT-dependent anticoagulants, direct FXa inhibitors are capable of inhibiting prothrombinase bound and clot-associated FXa, respectively. Unlike heparins, this provides FXa inhibition directly on a thrombus, impeding its further development [117,118]. Data from extensive clinical studies suggest that, licensed FXa inhibitors [shown in Table 1.2] are as effective as VKAs. Moreover, in comparison to VKAs, DOACs have favorable pharmacokinetics, fixed dose regimen and, importantly, fewer incidences of intracranial bleeding [118]. However, the risk of gastrointestinal (GI) bleeding is higher with DOACs, possibly due to the accumulation of active drug in the GI tract [118].  1.7 Reversing anticoagulation activity: Current options and recent            developments Adverse drug reaction (ADR) is a major cause for morbidity and mortality during medical care [127]. A study by Shepherd et al, revealed that in US hospitals, ADR-related deaths were higher in patients receiving drugs such as anticoagulants, opioids and immunosuppressants [128]. Bleeding is the major ADR associated with anticoagulation therapy. The propensity to bleed during anticoagulation therapy is influenced by numerous factors such as patient characteristics, dosing and intensity of anticoagulation etc. Generally, the risk of bleeding is higher in elderly patients and in patients with impaired renal function (CrCl of < 30 mL/min) [120]. Studies show that bleeding complications have a significant impact on anticoagulant treatment outcomes [129,130]. 33   For instance, OASIS-5 trial (enoxaparin vs fondaparinux in patients with non-ST-elevation ACS) showed high incidence of mortality in patients who experienced major bleeding complications with enoxaparin [129]. Hence, reducing the risk of bleeding by administering anticoagulant antidote is a clinical necessity.   Worldwide, millions of patients undergo cardiac surgeries, which involve the use of extracorporeal circuits [131]. In this clinical scenario, blood clotting is avoided, by administering high dose of UFH, intravenously [131]. To prevent post-surgical bleeding complication, excess UFH in the blood stream must be removed. To date, protamine is the only clinically approved antidote for reversing UFH anticoagulation activity [97]. Protamine is a 5 kDa cationic peptide (60% arginine aminoacid residues) derived from fish sperm [Figure 1.11] [97]. Protamine electrostatically interacts with negatively charged heparins and form stable complexes, thus neutralizing heparin’s anticoagulation activity [132].  Figure 1.11: Aminoacid residues in protamine. The guanidine groups of arginine aminoacid are protonated at physiological pH [133].  However, protamine administration may lead to adverse events such as systemic hypotension, intrinsic anticoagulation activity, complement activation by UFH-protamine complexes, protamine-induced thrombocytopenia etc. [134-137]. Most importantly, protamine is only partially effective in reversing LMWHs anticoagulation activity, and is not capable of reversing the anticoagulation activity of fondaparinux [132].  34  To mitigate adverse events associated with protamine use, researchers have developed variants of protamine. For instance, Stanley et al, developed synthetic protamine analogues [+18 BE] and [+18 RGD]. A low molecular weight protamine was prepared by Yang et al, by digesting protamine using thermolysin. In another approach, to reduce protamine toxicity, polyethylene glycol (PEG) chains were attached to protamine. Even though, these protamine derivatives showed UFH and LMWHs anticoagulation reversal activity and nontoxicity, none of them advanced to clinical trials [138-140].  Over the past years, numerous alternatives to protamine were developed and tested. Recombinant platelet factor 4 (rPF4), a cationic peptide, produced using an Escherichia coli expression system, was tested in humans. No life threatening adverse events were observed, and rPF4 was found to be as effective as protamine. However, rPF4 is structurally related to human PF4, which forms ultra-large complexes with UFH, implicated in the pathogenesis of HIT. Therefore, further human trials with rPF4 as a UFH reversal agent were suspended [141,142]. Heparinase I (Neutralase), an enzyme produced by the bacterium Flavobacterium heparinum was evaluated for UFH reversal activity. Heparinase I cleave 1-4 glycosidic linkages within heparin, and degrades heparin to smaller oligosaccharide fragments thereby reducing the anticoagulation activity. However, efficacy and safety evaluation in humans revealed that heparinase I has poor safety profile compared to protamine. Administration of heparinase I to patients undergoing Coronary Artery Bypass Grafting (CABG) resulted in post-operative bleeding complications, and subsequent transfusions [143,144].  35  Recently, Bainchini et al. prepared recombinant AT (AT-N135Q-Pro394) by introducing two mutations in the native human AT. The AT-N135Q-Pro394 antidote was found to be efficient in neutralizing all clinically available heparin-based anticoagulants [145]. Studies have explored the potential of using cationic peptides, cationically modified natural polysaccharides (N-(2-hydroxypropyl)-3-trimethylammonium chitosan chloride, Dex40-GTMAC3, etc.) and cationic synthetic polymers (pegylated poly(3-(methacryloylamino)propyl trimethylammonium chloride block copolymers, poly(ethylene glycol)–poly(2-(dimethylamino)ethyl methacrylate) as heparin reversal agents [146-150].  Currently, treatment options available for emergency reversal of VKAs anticoagulation activity include transfusion of fresh frozen plasma or prothrombin complex concentrate (PCC) such as three-factor PCC (FII,FIX and FX) or four-factor PCC (FII,FVII,FIX and FX). Both PCC also contain protein C and protein S. Vitamin K has the limitation of a slower onset of reversal. Studies are ongoing to explore the possibility of using four-factor PCC for reversing anticoagulation activity of DOACs [151,152].  This research thesis is based on studies that involve three novel anticoagulant antidotes, which are currently in different phases of development. Details are provided in the following sections.  1.7.1 Aripazine or ciraparantag (PER977) PER977 is a low molecular weight (512 Da) synthetic organic molecule developed by Perosphere Pharmaceuticals (Danbury, Connecticut). The chemical structure of PER977 is shown [Figure 1.12] [153] 36   Figure 1.12: The chemical structure of PER977 molecule [153]. PER977 is reported to bind, heparin anticoagulants (UFH, LMWHs and fondaparinux) and DOACs (dabigatran, rivaroxaban, apixaban, and edoxaban) through noncovalent hydrogen bonding and charge-charge interactions, and reverse their anticoagulation activity [154]. PER977 does not bind to any coagulation factors and does not exhibit prothrombotic activity [155]. Toxicity studies in dogs and rats showed no signs of toxicity and accumulation of PER977 and its primary metabolite 1,4-bis(3-aminopropyl)piperazine. Maximum tolerated dose of PER977 in rat and dog was found to 40 and 35 mg/kg, respectively [155].  PER977 has been shown to halt bleeding from a rat tail transection, induced by dabigatran, rivaroxaban and apixaban. PER977 also reversed anticoagulation activity of edoxaban as evaluated by rat tail, liver transection and whole blood clotting time assay. Chromogenic FXa assay demonstrated the capability of PER977 to restore impaired FXa activity in human plasma spiked with rivaroxaban and apixaban, respectively [154,156,157]. In human subjects treated with enoxaparin (1.5 mg/kg), a single dose of PER977 (100 mg or 200 mg) completely normalized the elevated whole blood clotting time, within 20 minutes in subjects treated with 100 mg dose and within 5 minutes with the 200 mg dose [158]. In another clinical study, human subjects treated with edoxaban (60 mg), a single dose of PER977 (100 and 300 mg) completely normalized the elevated whole blood clotting time within 10 minutes. Moreover, scanning electron 37  micrographs of blood clots revealed PER977’s ability to restore fibrin fiber formation in edoxaban containing blood clots [153]. In 2015, PER977 received fast-tract designation from the Food and Drug Administration (FDA) [159].  1.7.2 Andexanet Alfa Andexanet is a recombinant truncated form of catalytically inactive human FXa, developed by Portola Pharmaceuticals (South San Francisco, California). Modifications were made in three regions of native human FXa to produce andexanet: A 34-residue fragment was removed from the light chain of native human FXa that contains 11 Gla residues (eliminates the ability of native FXa to incorporate into prothrombinase assembly); change in the linker sequence between the light chain and heavy chain; and substitution of the serine residue with an alanine residue in the catalytic domain (eliminates ability of native FXa to cleave prothrombin into thrombin) [160]. However, the modified human FXa (andexanet) retains its ability to bind direct FXa inhibitors and LMWH [160]. In preclinical studies, andexanet was found to be effective in reducing blood loss caused by direct FXa inhibitors and AT-dependent heparins in animal models of bleeding such as rat tail transection and rabbit liver laceration [160]. In the phase–II study performed in nonbleeding volunteers treated with apixaban, rivaroxaban, edoxaban or enoxaparin, administration of andexanet rapidly normalized plasma anti-FXa levels, indicating anticoagulation reversal activity [161,162]. In the phase-III, double blind and placebo-controlled study, nonbleeding healthy volunteers received 5 mg apixaban (ANNEXA-A) or 20 mg rivaroxaban (ANNEXA-R), followed by bolus and continuous infusion of andexanet for 2-hours. In the ANNEXA-A study, the apixaban-treated group received andexanet 400 mg IV bolus followed by a continuous infusion of 38  4 mg/minute for 2-hours. The total dose administered reached 480 mg.  In the ANNEXA-R study, the rivaroxaban-treated group received andexanet 800 mg IV bolus followed by a continuous infusion of 8 mg/minute for 2-hours. The total dose administered reached 960 mg. Andexanet restored the impaired thrombin generation and normalized FXa levels in both apixaban and rivaroxaban treated groups [163].  Currently, the ANNEXA-4 trial is evaluating the efficacy and safety of andexanet in elderly patients experiencing bleeding following treatment with FXa inhibitor anticoagulants. Preliminary results indicate that andexanet administration restored hemostasis in 37 out of 47 patients [164]. Portola pharmaceuticals received break-through therapy designation from FDA for andexanet [159]. 1.7.3 Universal Heparin Reversal Agent (UHRA)  UHRA is a synthetic macromolecular antidote for heparin-based anticoagulants, developed in the Kizhakkedathu laboratory at the University of British Columbia, Canada. The chemical structure of the UHRA molecule is shown [Figure 1.13] [165]. UHRA is a dendritic macromolecule comprising a hyperbranched polyglycerol core (HPG core), methoxy polyethylene glycol chains emanating from the core and heparin binding groups (HBGs) incorporated on to the core [165].  39   Figure 1.13: The chemical structure of UHRA molecule and heparin binding group. HBG is indicated as ‘R’ in the UHRA structure. Adapted from Sci Transl Med 2014; 6(260):260ra150. Reprinted with permission from AAAS.   The heparin binding groups in UHRA contain tertiary amine groups that acquire cationic charges at physiological pH. By modulating the molecular weight and the number of HBG, Kizhakkedathu laboratory developed a library of UHRA molecules [Table 1.3]. Isothermal titration calorimetry studies were used to identify the lead antidote molecule and the data showed that UHRA-6, UHRA-7 and UHRA-8 exhibits micromolar binding affinity to all clinically available heparin anticoagulants [165].     40  UHRA Mn Mw/Mn Rh (nm) No. of HBGs (R) Zeta potential (mV)  UHRA-1  116,700  1.2  10  33  8.9 ± 1.6  UHRA-2  48,000  1.45  4  18  14.8 ± 0.9  UHRA-3  23,000  1.52  3  4  10.5 ± 1.9  UHRA-4  23,000  1.52  3  5  12.7 ± 1.1  UHRA-5  23,000  1.52  3  11  15.5 ± 0.4  UHRA-6  23,000  1.52  3  16  17.1 ± 3  UHRA-7  23,000  1.52  3  20  19.8 ± 1.6  UHRA-8  23,000  1.52  3  24  22.7 ± 3  Table 1.3: Characteristics of UHRA molecules synthesized for the preliminary heparin binding and in vitro anticoagulant reversal activity. Adapted from Sci Transl Med 2014; 6(260):260ra150. Reprinted with permission from AAAS.   By analysis of plasma clotting using aPTT, the efficiency of UHRA-6, UHRA-7 and UHRA-8, respectively, to neutralize UFH and tinzaparin anticoagulation activity was first evaluated. All three UHRA molecules were found to be effective in reversing the anticoagulation activity of UFH and tinzaparin (2 IU/mL). However, from the observation that UHRA-8 (23kDa, 24 R groups) showed slight intrinsic anticoagulation activity at a concentration above 0.1 mg/mL, it was learned that the number of HBGs on a 23kDa UHRA molecule is critical to achieve maximum anticoagulation neutralization efficiency without adversely affecting blood coagulation. By performing thromboelastography in whole human blood, UHRA-7 at concentrations 0.1 mg/mL and 0.2 mg/mL, respectively, 41  is capable of reversing anticoagulation activity of fondaparinux (1.2 IU/mL) [Figure 1.14 A] and this data was corroborated by performing chromogenic FXa assay in human plasma [Figure 1.14 B]. In addition, UHRA-7 did not activate complement system and platelets, confirming its hemocompatibility.   Figure 1.14: UHRA-7 reverses anticoagulation activity of fondaparinux (1.2 IU/mL). (A) A representative whole blood TEG clotting profile demonstrates that the enhanced clotting time induced by fondaparinux is normalized by UHRA-7 at a dose of 0.1 mg/mL and 0.2 mg/mL, respectively. (B) A chromogenic FXa assay was performed in fondaparinized human plasma spiked with UHRA-7 or protamine. In comparison to protamine, UHRA-7 achieved >90% neutralization of fondaparinux activity. Adapted from Sci Transl Med 2014; 6(260):260ra150. Reprinted with permission from AAAS. 42   The effect of UHRA-7 on bleeding time using a mouse tail bleeding model was evaluated, and unlike protamine (20 mg/kg), UHRA-7 (20 mg/kg and 50 mg/kg) did not induce any bleeding side-effects and arrested bleeding induced by all clinically used heparin anticoagulants, corroborating the universal antidote action of UHRA-7. Dose tolerance studies in mice revealed that UHRA-7 is non-toxic. No change in body weights of mice treated with UHRA-7 were observed for 29 days. Moreover, histopathological analysis of vital organs isolated from rodents treated with UHRA-7 did not show any abnormalities [165].  Figure 1.15: (A) UHRA-7 does not cause bleeding even in the absence of heparin anticoagulants. (B) UHRA-7 normalizes the enhanced bleeding induced by all clinically available heparin anticoagulants.  *P < 0.05, ***P < 0.0005, n.s. represents P > 0.05. Adapted from Sci Transl Med 2014; 6(260):260ra150. Reprinted with permission from AAAS.    43  1.8 Rationale and specific aims Blood clotting is an intricate and highly regulated physiological process, which involves a fine balance between clotting and anticlotting pathways [Figure 1.1, 1.2 and 1.3]. For instance, the presence of abnormally high levels of polyphosphate or neutrophil extracellular traps in blood, or the hereditary deficiency of physiological anticlotting proteins such as AT, would favour clotting, resulting in thrombosis. Conversely, therapeutic inhibition of activity of clotting factors such as FX and thrombin with anticoagulant drugs or inherited deficiency of clotting proteins such as FVIII or FIX would impair the clotting reaction leading to hemorrhagic complications. Therefore, therapeutic modulation of blood clotting (anticoagulation therapy) intended to mitigate thrombotic disorders should be performed with utmost care. Despite significant advances in the field of anticoagulant drug development, there is no anticoagulant currently available without bleeding side-effects [Table 1.1 and 1.2]. Therefore, an anticoagulant antidote that is nontoxic and effective, preferably with universal anticoagulant reversal action is required. The main objective of this research thesis is to understand the design and mechanism of action of the anticoagulant antidote UHRA, and investigate its anticoagulant reversal activity, influence on clotting and clot structure. In addition, compare the activity and mechanism of action of UHRA with that of PER977 and Andexanet Alfa.  1.8.1 Role of PEG chains in the mechanism of action of UHRA Currently, protamine is widely used for reversing UFH anticoagulation activity. The adverse side-effects and limitations associated with protamine are well-documented. It is known that polycationic protamine binds highly anionic UFH via electrostatic 44  interaction. However, protamine exhibits nonspecific binding to plasma proteins and this contributes towards its toxicity [97,134].   Extensive in vitro and in vivo studies performed in our laboratory, revealed that UHRA-7 molecule is capable of reversing anticoagulation activity of all clinically available heparin-based anticoagulants and is nontoxic. UHRA-7 exhibits high-affinity multivalent binding to all heparin derivatives, due to the arrangement of HBGs on a HPG scaffold. It is important to state that although both UHRA-7 and protamine exhibit electrostatic interactions with heparins, UHRA-7 demonstrates a superior toxicity profile compared to protamine, which could be due to the presence of PEG chains on UHRA-7 [165].                                     Studies reported in Chapter 2, unearth the molecular design of UHRA and important role of a PEG shell in the UHRA molecule. Our hypothesis is that the heparin binding affinity and anticoagulant neutralization potential of a UHRA molecule could be modulated by the nature of the PEG shell. The PEG chains prevent nonspecific interactions of UHRA molecule with plasma proteins and provide necessary specificity towards highly anionic heparins. To test this hypothesis, the specific aims are: 1. To synthesize, UHRA molecule (similar to UHRA-7), a UHRA molecule with no PEG chains (N-UHRA) and a UHRA molecule with longer PEG chains than the conventional UHRA molecule (mPEG750-UHRA). 2. To measure the binding affinity of protamine, UHRA, N-UHRA and mPEG750-UHRA with UFH at different ionic strengths, temperature and pH using isothermal titration calorimetry (ITC).  45  3. To evaluate and compare the UFH anticoagulation reversal activity of N-UHRA, UHRA and mPEG750-UHRA to protamine, and study the complex formation between different UHRAs, protamine and UFH. 1.8.2 Validating the mechanism of action of UHRA, PER977 and Andexanet Alfa    UHRA is being currently developed in the Kizhakkedathu laboratory. PER977 and Andexanet Alfa are anticoagulant antidotes currently in the advanced stages of clinical trials. It is reported that PER977 binds directly to anticoagulants via noncovalent hydrogen bonding and electrostatic interactions. This information is based on dynamic light scattering and in silico modelling data [154]. However, no data is available in the literature, regarding the binding affinity of PER977 to anticoagulants. A study conducted by Portola Pharmaceuticals, demonstrated that PER977 halts bleeding induced by rivaroxaban in a rabbit liver laceration model; however, PER977 failed to normalize PT, anti-FXa activity and thrombin generation [166]. Hence, the mechanism of action of PER977 is not fully understood. The binding affinity of Andexanet Alfa to their intended anticoagulant targets is also not available.   Studies reported in Chapter 3, provide insights into the mechanism of action of PER977, Andexanet Alfa and UHRA, anticoagulant neutralization potential and the influence these antidotes have on clot structure in the absence and presence of anticoagulant. The specific aims are: 1. To identify anticoagulant binding partners of UHRA, PER977 and Andexanet Alfa, and to measure binding affinity of this interaction. 2. To understand the influence of PER977 and Andexanet Alfa on fibrin polymerization and clot structure in the absence and presence of anticoagulant.  46  1.8.3 UHRA is a superior heparin antidote compared to protamine Studies suggest that thrombosis could be avoided by counteracting prothrombotic activity of biomolecules such as polyP and extracellular nucleic acids using cationic inhibitors [86,167]. Recently, Travers et al. has demonstrated that selected UHRA molecules reduce thrombotic occlusions in mouse model of thrombosis, without causing bleeding in comparison to heparin [168].   However, it is well-known that, conventional polycations such as polyamidoamine dendrimers (PAMAM), protamine etc. adversely affect blood clotting. For instance, PAMAM dendrimers trigger blood clotting, whereas protamine shows intrinsic anticoagulation activity [135,169]. Therefore, it is important to evaluate the impact of UHRA molecule on blood clotting and clot structure, in the absence and presence of polyanions such as heparin to demonstrate its nontoxicity.   Studies reported in the Chapter 4, corroborate the superior heparin reversal efficacy and nontoxic nature of the UHRA molecule. Our hypothesis is that the UHRA molecule with an optimal number of heparin binding groups, PEG chain length and density will have minimal influence on blood proteins, clotting and clot structure, and will be superior to protamine in terms of efficacy and toxicity. To test this hypothesis, the specific aims are:  1. To synthesize UHRA molecule (similar to UHRA-7) and confirm it’s binding to heparin using ITC measurements. 2. To evaluate the influence of UHRA on fibrinogen, fibrin polymerization, clotting and clot structure in comparison to protamine. 3. To demonstrate that UHRA reverses UFH anticoagulation without lung injury in mouse in comparison to protamine.2The chapter 2 has been prepared for submission to Journal of American Chemical Society                 47  Chapter 2: Molecular design, origin of selectivity and functional mechanism of the nontoxic universal heparin reversal agent UHRA 2.1 Synopsis To prevent blood clotting in patients undergoing cardiopulmonary bypass, or other surgical procedures, unfractionated heparin (UFH), a highly sulphated anticoagulant is administered at a high dose. However, due to the risk of bleeding, excess circulatory UFH must be neutralized with protamine sulphate (PS), which is the only approved UFH antidote. At physiological pH, the arginine clusters of PS are protonated generating cationic charges, which facilitate PS binding to any oppositely charged molecules through electrostatic interactions. This forms the mechanistic basis for PS reversal of UFH activity. The unshielded cationic charges on PS are accountable for its nonspecific interaction with blood proteins leading to adverse cardiovascular effects, and could be also linked to the failure of PS to neutralize other heparin variants, such as low molecular weight heparins and fondaparinux. Understanding this clinical need, we developed Universal Heparin Reversal Agent (UHRA), a synthetic and nontoxic antidote capable of reversing all clinically available heparin-based anticoagulants. Studies presented in this chapter reveal the unique design of UHRA and roles of its structural components that makes UHRA, a superior heparin antidote compared to PS. UHRA consists of a hyperbranched polyglycerol core with incorporated methylated tris(2-aminoethyl)amine ligands that acquire cationic charges at physiological pH, and methoxy polyethylene glycol (mPEG350) chains emanating from the core. To understand the contribution of mPEG chains towards heparin neutralization activity and the nontoxic nature of UHRA, we synthesized two UHRA analogs: naked UHRA (N-UHRA) with no48   mPEG chains and mPEG750-UHRA with longer mPEG750 chains. From the thermodynamics of binding of antidotes with UFH at different ionic strengths, we found that mPEG chains in the UHRA molecule impose an entropic penalty to binding. Clotting assays reveal that UHRA molecules with mPEG chains did not adversely affect clotting, and neutralized UFH over a wide range of concentrations. Conversely, N-UHRA and PS, possessing unshielded cationic charges displayed intrinsic anticoagulant activity and showed a narrow concentration window for UFH neutralization. This supports the notion that mPEG chains avert nonspecific interactions of UHRA with blood proteins, and provide selectivity towards heparins through a combined steric repulsion and Donnan shielding effect. In addition, we found that mPEG chains regulate the size of antidote/UFH complexes. UHRA molecules with mPEG chains formed smaller and compact complexes with UFH, compared to N-UHRA and PS complexes with UFH, respectively. Finally, fluorescence and ELISA experiments reveal that UHRA neutralizes UFH anticoagulation activity by disrupting antithrombin/UFH complexes. Thus studies presented in this chapter reveal the unique design of UHRA and the role of its structural components towards achieving heparin reversal activity. 2.2 Background Unfractionated heparin (UFH) is a linear, highly sulfated polysaccharide mostly comprised of repeating units of uronic acid–(1→4)–glucosamine disaccharide residues [170]. UFH and other heparin variants, such as low molecular weight heparins (LMWHs) and the synthetic penta-saccharide fondaparinux, bind antithrombin (AT), an endogenous inhibitor of blood clotting, and potentiate its activity to inhibit blood coagulation proteases [27]. Heparin-based therapeutics are therefore widely employed 49   as anticoagulants, including prior to invasive surgeries to prevent blood clots, as well as to prevent and treat pulmonary embolism, deep vein thrombosis, and arterial thromboembolism [99]. However, bleeding risk associated with heparin therapy is a major concern for clinicians. The ability to neutralize heparin activity is therefore a clinical requirement [132]. Protamine sulphate (PS), a cationic peptide, is currently the only FDA-approved antidote to UFH. It is thought to neutralize heparin’s anticoagulation activity via a direct binding interaction that is driven by electrostatic attraction and ion pairing between the two oppositely charged macromolecules. PS is not approved as a neutralizing agent to other heparin variants, as it only partially reverses the anti-FXa activity of LMWHs and is ineffective in reversing the anticoagulant activity of fondaparinux [132]. Moreover, PS administration can affect cardiovascular health adversely, increasing the likelihood of severe hypotension, cardiovascular collapse, anaphylaxis, pulmonary edema, and complement activation by PS-UFH complexes [171,172]. The clinical need for a more effective and nontoxic alternative to PS that can fully and more safely neutralize all heparin-based anticoagulants is therefore widely recognized.   To address this unmet clinical need, we recently developed a universal antidote for heparin anticoagulants, including UFH, LMWHs and fondaparinux. The antidote, named Universal Heparin Reversal Agent (UHRA), is a cationic dendritic hyperbranched polymer comprised of three principal components: 1) a hyperbranched polyglycerol (HPG) core that presents 2) a set of methylated tris(2-aminoethyl)amine ligands beneath 3) a brush of methoxy-terminated polyethylene glycol (mPEG350, where 350 denotes the number-average molecular weight in Da) chains emanating from the ligand-50   bearing core [165]. In later chapters of this thesis, in vivo and ex vivo studies will be presented to show that UHRA provides anticoagulation reversal activity against all clinically relevant heparin variants, does not interact with blood clotting proteins or clot structural components, and is nontoxic [165,173]. Here, the mechanistic basis for the unique activity and hemocompatibility of UHRA is described.   Polymer-based nanomedicines have been explored for drug delivery, as well as diagnostic, imaging and therapeutic applications [174]. Studies on those molecules have shown that certain macromolecular characteristics, including hydrodynamic size, chemistry and charge (density), can significantly impact the efficacy, pharmacokinetics and immunogenicity/toxicity of the polymeric drug [175]. This is likewise evident with the arginine-rich peptide therapeutic PS, which has an isoelectric point of ca. 13.8 and is therefore strongly basic [176]. Tight binding to the highly sulfated polysaccharide UFH is achieved via two arginine-rich clusters within PS [177]. The fact that PS is not able to fully neutralize the anti-FXa activity of LMWHs suggests that PS binding and the extent of anticoagulant activity reversal depend on heparin molecular weight (MW), the degree of heparin sulphonation (charge density), and possibly other physico-chemical metrics such as the persistence length of the heparin variant. As the density s of sulfonate groups per disaccharide correlates with heparin MW (s for UFH > LMWHs > fondaparinux) [132], it is difficult to parse the relative contribution of each to the anticoagulant reversal activity of PS.  The dense cationic charge within PS is displayed in a naked (i.e., sterically unshielded) fashion. Regrettably, this enables PS to interact non-specifically with endogenous anionic macromolecules in blood, including fibrinogen, and various 51   components of blood clot such as platelets and fibrin fibers [173,178,179]. Those unwanted interactions are known to contribute to the adverse complications associated with PS administration [172,173,134,180]. Together, these findings suggest that the efficacy of PS in neutralizing UFH resides in a direct binding interaction dominated by electrostatic attraction, while complications associated with PS therapy appear to be related, at least in part, to the fact that its cationic charges are sterically unshielded and therefore available to freely interact and pair with any anionic surfaces and ligands within the vasculature.    Here, these insights are combined with basic theories describing the molecular physics of charged macromolecules to establish the principles that guided the design of UHRA. The basic theoretical underpinnings of the molecular architecture of UHRA that enables it to selectively bind and neutralize heparin-based anticoagulants are described. The chemistry, density and protonation states of the cationic ligand employed, as well as the size of HPG core presenting those ligands, are characterized to understand how UHRA binds heparin. The design and properties of the mPEG brush emanating from the ligand-functionalized core are then studied in more detail.  PEGylation (i.e. the incorporation of polyethylene glycol (PEG) into a drug scaffold) is a commonly adopted strategy to inhibit adverse effects arising from nonspecific interactions between a polymeric drug molecule and components of biological fluids [181,182]. To characterize the contribution of PEGylation to the mechanism of action of UHRA, two dendritic molecules in addition to UHRA have been synthesized and characterized. The first, N-UHRA, is a truncated analog of UHRA with no mPEG chains. The second, mPEG750-UHRA (where 750 denotes the number-average MW in Da), is a 52   UHRA-like molecule in which the mPEG350 chains have been replaced with longer mPEG750 chains. Isothermal titration calorimetry (ITC) is used to characterize the binding of each of these macromolecules, as well as PS, to UFH as a function of solution conditions, including ionic strength (I). The thermodynamic data obtained show that binding of either PS or N-UHRA depends on I in a manner predicted by classic polyelectrolyte theory. In contrast, the mPEG brush of UHRA is found to modulate binding through a combined steric repulsion and Donnan shielding effect that can be changed by altering the PEG chain length and/or segment density. As a result, the strength of binding falls off more rapidly with increasing I. Antidote-UFH complexes are further characterized using atomic force microscopy (AFM) and dynamic light scattering (DLS) to reveal the crucial role of the PEG chains in controlling the stability and size of complexes formed. Finally, the ability of the unique components of UHRA to collectively achieve anticoagulant reversal activity against UFH is demonstrated by performing anticoagulant neutralization assays.  Fundamentals of UHRA molecular design Electrostatic interactions between charged macromolecules in aqueous solutions containing a strong electrolyte (background salt) have been studied extensively [183-186]. The molecular physics of such interactions are known to be qualitatively captured by Derjaguin, Landau, Verwey and Overbeek’s (DLVO) classic theory of colloid stability that invokes the Debye-Hückel approximation to linearize the Poisson-Boltzmann equation [187,188].  DLVO theory models the potential energy of interaction of a pair of uniformly charged spherical particles within an aqueous electrolyte solution by superimposing the electrostatic double-layer interaction and the van der Waals 53   attraction. Though the theory is approximate, direct force measurements have shown its electrostatic term to be accurate to a surface-separation distance of a few nanometers [189].  In DLVO theory, the electrostatic interaction force Fel(r*) between two spherical colloids (i and j) having radii ai and aj and surface charge densities per unit particle surface area  and j, respectively (see Figure 2.1A for a schematic of this system), is given by 𝐹𝑒𝑙(𝑟∗) = 4𝜋 (𝑎𝑖𝑎𝑗𝑎𝑖 + 𝑎𝑗)𝜎𝑖𝜎𝑗𝜀𝜀𝑜𝜅𝑒−𝜅𝑟∗ (2.1) where is the inverse of the Debye length, , and is given by  𝜅−1 = 𝜆 = √𝜀𝜀𝑜𝑘𝐵𝑇2𝑒2𝑁𝐴𝐼 (2.2) Here, e is the charge of an electron, and o are the solvent dielectric constant and the permittivity of free space, respectively, kB is Boltzmann’s constant, T is the absolute temperature, and I is the ionic strength of the added background salt.  In equation 2.1, r* is the separation distance between the surfaces of the two particles whose centres of mass are separated by a distance r = r* + ai + aj.  Thus, the particles touch when r* = 0, and as they are modeled as hard spheres, an infinite force is required to move their centres of mass any closer [Figure 2.1A].     54    Figure 2.1: Schematics of macro-ion interaction energetics. (A) Interaction between unshielded oppositely charged macromolecules – for example, positively charged PS and negatively charged UFH; (B) Interaction between a shielded (fuzzy) macromolecule and an unshielded oppositely charged macromolecule – for example, positively charged UHRA and UFH.  Though qualitative, equation 2.1 nevertheless provides insights into the nature of the interaction between unshielded oppositely charged macromolecules (i.e. when i and j are opposite in sign) that help explain certain aspects of the interaction of PS with UFH, and that also served as a foundation for the design of UHRA. As Fel is found proportional to ij exp (– r*), a nonlinear and increasing force of attraction is predicted as the distance between the two oppositely charged macromolecules decreases toward r* = 0 [Figure 2.1A]. As a result, at sufficiently low I and/or high  values, coacervation (i.e, hetero-aggregation) is predicted for oppositely charged macro-ions where the 55   charge on each is displayed in a naked fashion. This is approximately the case for PS and UFH, and coacervation is indeed observed experimentally at low or even physiological I. In particular, both Maurer et al. [190] and Valimaki et al. [149] found that the interaction of PS with UFH at physiologic conditions generates hetero-aggregates of average diameter close to 200 nm.  Moreover, at sufficiently low I, equation 2.1 predicts binding of UFH to any poly-cationic macromolecule i having a sufficiently large i. It is therefore not surprising that other (largely) unshielded cationic macro-ions, including N-(2-hydroxypropyl)-3-trimethylammonium chitosan chloride [150], dextran decorated with glycidyltrimethylammonium chloride groups [148], and polybrene [191], have also proven effective in neutralizing the anticoagulant activity of UFH. For most of these proposed polymeric neutralizing agents, the strength and stoichiometry of the interaction with UFH has not been carefully studied. Equation 2.1 likewise predicts electrostatic attraction between PS and any endogenous polyanionic macromolecule or surface within the vasculature (e.g., circulating plasma proteins, free DNA, etc.). Because UFH is generally administered at very high dosages, mass action will still favor PS binding to it, despite the potential for PS to participate in these competing interactions. However, mg/kg dosages of LMWHs are typically lower when compared to those of UFH [97], raising the possibility that the reduced activity of PS in neutralizing LMWHs is due in part to competing reactions. At the very least, these observations make clear the intrinsic challenge in achieving binding specificity when the interaction force between the drug and target is dominated by electrostatic attraction up to the point of molecular contact. Moreover, they are 56   consistent with the fact that off-target binding of drug molecules often leads to functional failure and adverse side effects [192]. But it is known that the nature of interactions between a so-called fuzzy (or soft) charged particle and an oppositely charged unshielded macro-ion is quite different from that between two unshielded macro-ions [193-196]. Consider, for example, the fuzzy macromolecule shown in Figure 2.1B where, as with UHRA, a dense core structure displaying a large positive charge has been coated with a neutral end-grafted polymer brush. That neutral polymer layer alters the force of interaction in several important ways. As the polymer layer is permeable to both solvent and the ions of the background salt, at least some of the diffuse layer will be contained within the brush void volume. As a result, the electrokinetic potential brush at the outer edge (r = ai + ; see Figure 2.1B) of the brush layer will be attenuated relative to the electrostatic potential core at the surface of the ligand-bearing core. Indeed, in the extreme case where the height of the polymer brush is made to be larger than the Brinkman screening length (given by the square root of the permeability of the polymer layer), Donnan equilibrium will be achieved within the brush volume, making brush negligible. The neutral polymer brush can further attenuate the force of attraction through steric repulsion arising from compression of the brush as the interacting macromolecule partitions into it. Scaling theories are available for estimating the separation-distance-dependent force Fsteric required to partition a spherical particle into a neutral polymer brush [197,198].  One such relation can be derived from the work of Steels et al. [199,200]  57   𝐹𝑠𝑡𝑒𝑟𝑖𝑐(𝑟∗) ∝𝑘𝐵𝑇𝑎𝑗232√2𝜙𝑏𝑟𝑢𝑠ℎ2 (𝑟∗)𝑁𝑚2 𝑎𝑚3 (𝜆𝑏𝑟𝑢𝑠ℎ𝑟∗)5/4 (2.3) where r* is the separation distance defined in Figure 2.1B, aj is the radius of the macromolecule partitioning into the brush layer of fuzzy particle i, brush is the volume fraction of the brush at position r*, Nm and am are the number and radius of monomers in each end-grafted polymer chain, respectively, and brush is a scaling length.  Equation 3 predicts a complex non-linear dependence of Fsteric on r*, with Fsteric tending to 0 as r* → , where  is the height of the brush, and to a maximum repulsion as r* → 0 [Figure 2.1B].   The cumulative effect of the neutral brush on both the pair-potential energy uij(r*) and the equilibrium association constant Ka for the interaction of a fuzzy macro-ion and an unshielded macro-ion of opposite charge is illustrated in Figure 2.1B. At physiologic conditions and lower I, unshielded macro-ions carrying values that are opposite in sign can exhibit a highly attractive uij(r*) at or near molecule contact [Figure 2.1A].  For that interaction, one finds from equation 2.1 that ln Ka should to a first approximation scale linearly (with negative slope) with √𝐼 [Figure 2.1A].  A significant increase in ionic strength is therefore required to dissociate the complex by making  of order ai + aj.  In contrast, at low I the interaction of a fuzzy macro-ion with an unshielded macro-ion of opposite charge sign is predicted to be somewhat weaker, with ln Ka decreasing non-linearly and more sharply with increasing √𝐼 [Figure 2.1B].  As a result, binding of the two macro-ions can be made to be favored only under certain conditions, most notably when 1) core is sufficiently large that an electrostatic attraction to the oppositely charged macromolecule is maintained in the presence of the neutral polymer layer, and 58   2) the end-grafted polymer adopts a soft brush configuration that provides a steric resistance to non-specific binding while not being sufficiently punitive to negate tight association when condition 1 is met.  Below we show how the flexibility of the UHRA synthesis platform allows one to achieve this desired specific-association outcome by tuning the core size, the ligand density displayed on it, and/or the grafting density and length of the mPEG brush.  2.3 Methods General information  All reactions with air- and/or water sensitive reagents were performed in a Schlenk flask under dry argon atmosphere. All chemicals were purchased from Aldrich (ON, Canada) unless otherwise mentioned. Human AT and sheep anti-human AT conjugated with peroxidase were purchased from Haematologic Technologies Inc. Heparin-binding 96-well micro-titre plates were purchased from BD Biosciences. 1H and 13C nuclear magnetic resonance (NMR) spectra were acquired in deuterium oxide (D2O) or deuterated chloroform (CDCl3) on a Bruker Avance AV-400 NMR spectrometer. Absolute molecular weights of precursor HPG or HPG-mPEG350/750 polymers were determined by gel permeation chromatography (GPC) with a Waters 2695 separation module fitted with Ultrahydrogel™ columns (guard, linear and 120), as well as a DAWN HELEOS II multiangle laser light scattering (MALLS) detector and an Optilab T-rEX refractive index detector, all from Wyatt Technology. In all cases, the mobile phase was 0.1 M NaNO3 (10 mM phosphate buffer).  59   Synthesis of polymeric heparin antidotes  Synthesis of UHRA  Synthesis of precursor polymer, HPG-mPEG350 (25 kDa), and its conversion into the UHRA molecule is detailed in the Methods section of chapter 4 and in our publications [7,8]. Synthesis of N-UHRA a) Synthesis of the 6-kDa HPG core (precursor of N-UHRA) Glycidol was purified by vacuum distillation before use and stored over dried molecular sieves at 4 °C. A flame-dried three-neck round bottomed flask was cooled under vacuum and filled with argon. To this, 1,1,1-tris(hydroxymethyl)propane (0.480 g, 3.5 mmol) and potassium methylate (25 wt % solution in methanol, 0.440 mL, 1.491 mmol) were added and stirred for 30 minutes at room temperature (RT) (22 °C). Methanol was removed under vacuum for 4 hr. The flask was heated to 95 °C and glycidol (8.0 mL, 0.120 mol) was added over a period of 15 hr and stirred for an additional 4 hr at the same temperature. The polymerization reaction was quenched with methanol and the polymer dissolved in methanol. The polymer was precipitated twice from diethyl ether. The resulting polymer was dissolved in water and dialyzed against water using a 1500 Da molecular weight cut-off (MWCO) membrane for 12 hours with multiple exchanges in water. The polymer was characterized by NMR spectroscopy and the absolute molecular weight of the polymer was determined by gel permeation chromatography (GPC) combined with light scattering.  60   b) Conversion of 6-kDa HPG core into N-UHRA The HPG precursor polymer (0.8 g, 0.13 mmol) was dissolved in 3 mL of anhydrous pyridine. To this, p-toluenesulfonyl chloride (0.497 g, 2.6 mmol) was added slowly and stirred at RT for 48 hr. Pyridine was removed by rotary evaporation; the polymer was dissolved in 0.1 N HCl and dialyzed overnight. The lyophilized HPG-tosylate (0.5 g) and tris(2-aminoethylamine) (2 mL) were dissolved in dimethyl sulfoxide (DMSO, 5 mL) and refluxed for 24 hr. DMSO was removed under vacuum and the polymer was dissolved in a minimum amount of methanol and precipitated twice from diethyl ether. The polymer was then dissolved in water and dialyzed against water using a 1500 Da MWCO membrane for 2 days. The resulting polymer solution was added to a mixture of formaldehyde (2 mL) and formic acid (2 mL) at 0 °C and the reaction mixture was refluxed overnight. After cooling the reaction flask to RT, the pH of the solution was adjusted to 10 using 1M NaOH; the polymer was extracted with dichloromethane (DCM) and the DCM then removed on a rotary evaporator. The polymer was dissolved in distilled water and dialyzed against water using a 1500 Da MWCO membrane with frequent changes in water for 2 days. The number of amine groups on the polymer was determined by conductometric titration. NMR spectra of N-UHRA confirming its expected chemical composition are shown in the Appendix A.1. 1H NMR (400 MHz, CDCl3): δ (ppm) 2.31 (s, 12H, -CH3), 2.50-2.66 (m, 12H, -CH2CH2), 3.49-4.30 (m, 5H, -CH2O). 13C NMR (100 MHz, CDCl3): δ (ppm) 42.39, 44.11, 50.68, 54.32, 55.14, 59.61, 60.95, 62.70, 66.02, 67.55, 68.98, 69.25, 70.64, 70.91, 71.28, 74.12, 74.44, 74.63, 78.05, 79.56, 94.52.   61   Synthesis of mPEG750-UHRA a) Synthesis of mPEG750 epoxide  To a solution of methoxy(polyethylene glycol) 750 (m-PEG-OH-750; 30 g, 0.04 mol) in tetrahydrofuran (350 mL) was added sodium hydride (4.8 g, 0.2 mol) in small portions at 50 °C under argon. The cloudy solution was stirred for 2 hr and epichlorohydrin (16 mL, 0.2 mol) was added slowly to that solution then stirred for another 2 hr. The reaction mixture was filtered and the unreacted NaH was quenched with ethanol very slowly at 0 °C. The filtrate was concentrated and dissolved in DCM. The epoxide was precipitated twice in cold hexane (1:10) and centrifuged to isolate the pure epoxide. The polymer was dried in vacuum at 50 °C for one day (yield: 25 G, 77%). NMR spectra of mPEG750 epoxide are shown in Appendix A.2. 1H NMR (400 MHz, CDCl3): δ (ppm) 2.56 (dd, J = 2.74, 2.80 Hz, 1H, CH2OCH), 2.82 (t, J = 4.11 Hz, 1H, CH2OCH), 3.19 (m, 1H, CH2OCH), 3.31 (s, 1H, -O CH3), 3.39 (dd, J = 5.75, 5.89 Hz, 1H, CHCH2O), 3.82-3.49 (m, 64H, CH2O). 13C NMR (100 MHz, CDCl3): δ (ppm) 44.18 (CH2OCH), 50.82 (CH2OCH), 58.91 (-CCH2O and -OCH3), 70.40, 70.55, 71.55 (CH2O-) b) Synthesis of the 36-kDa HPG-mPEG750 molecule (precursor of mPEG750-UHRA) A flame dried three neck round-bottom flask was loaded with TMP (122 mg, 0.910 mmol) and potassium methoxide (113 µL, 25 % in methanol, 0.383 mmol) under argon, stirred for 1 hr at RT, and that reaction mixture was dried at 50 °C overnight. An overhead stirrer was connected to the reaction flask and the reaction temperature was raised to 90 °C. Under the constant stirring (100 rpm), glycidol (3.5 mL, 51.5 mmol) was added for 17 hr (205 µL/hr) and stirred for another 4 hr. To this sticky colorless mixture, 62   a pinch of potassium hydride (KH in oil) was added and stirred for 10 minutes. A solution of anhydrous m-PEG750 epoxide (16.4 g, 20.5 mmol) in dry dimethyl formamide (16 mL) was added at a flow rate of 2 mL/hr and the resultant pale brown solution stirred for another 4 hr. The reaction mixture was quenched with water and methanol (2/6 mL). After concentrating the reaction mixture, the polymer was dissolved in methanol and precipitated from ether. The solid obtained after centrifugation was subjected to fractional precipitation (methanol and ether combination) to obtain the desired polymer. Incorporation of mPEG into the polymer was determined by NMR. Molecular weight was determined by GPC. NMR spectrum of the 36-kDa HPG-mPEG750 is shown in Appendix A.3. 1H NMR (400 MHz, CDCl3): δ (ppm) 3.40 (s, 3H, -OCH3), 3.50-4.10 (m, -OCH2) 13C NMR (100 MHz, CDCl3): δ (ppm) 58. 41 (-OCH3), 60.81, 62.70, 68.95, 69.24, 69.54, 69.70, 70.06, 70.50, 70.79, 71.01, 71.98, 72.30, 78.09, 79.54 c) Conversion of HPG-mPEG750 into mPEG750-UHRA The precursor polymer HPG-mPEG (36 kDa, 0.2 g, 5.5 µmol) was dissolved in 5 mL of pyridine. To this, p-toluenesulfonyl chloride (0.079 g, 0.410 mmol) was added and stirred at RT for 48 hr. Pyridine was removed by rotary evaporation; the polymer was dissolved in 0.1 N HCl and dialyzed overnight. The HPG-mPEG750-tosylate was dried by freeze drying. The dried HPG-mPEG750-tosylate (0.150 g) and tris(2-aminoethylamine) (2 mL) were dissolved in 1,4-dioxane (5 mL) and refluxed for 24 hours. Dioxane was removed under vacuum and the polymer was dissolved in a minimum amount of methanol and precipitated twice from diethyl ether. The polymer was dissolved in water, and dialyzed against water using a 1000 Da MWCO membrane 63   for 2 days. The resulting polymer solution was added to a mixture of formaldehyde (3 mL) and formic acid (3 mL) at 0 °C. The reaction mixture was refluxed overnight. After cooling to RT, the pH of the solution was adjusted to 10 using NaOH and the polymer was extracted with DCM. DCM was removed under vacuum; the polymer dissolved in distilled water and dialyzed against water using a 1000 Da MWCO membrane with frequent changes in water for 2 days. The amine groups on the polymer were determined using conductometric titration. NMR spectra of mPEG750-UHRA are shown in Appendix A.4. 1H NMR (400 MHz, CDCl3): δ (ppm) 2.50-3.03 (m, 20H, -CH3, CH2CH2), 3.20 (br, 4H, CH2CH2), 3.35 (s, 3H, OMe), 3.51-3.99 (m, 102H, -CH2O). 13C NMR (100 MHz, CDCl3): δ (ppm) 43.16, 47.90, 54.44, 58.16, 68.92, 69.16, 69.68, 70.05, 71.08, 72.35, 77.89. Schematic representations of UHRA and the mPEG750-UHRA molecule are presented in Appendix A.5. The characteristics of these synthesized antidote molecules are described in Table 2.1        Da= Daltons; PDI= Polydispersity Index; Ligands = methylated Tris(2-aminoethyl)amine Antidote molecule Molecular weight (Da)a Polydispersity Index (PDI)a mPEG content  (mol %) b Number of  ligandsc  UHRA  25,000  1.4  28  25  N-UHRA      6000  1.29  0  7                       mPEG750-UHRA  36,000  1.5  23  32 64   a Determined by gel permeation chromatography in 0.1 M sodium nitrate, b Nuclear magnetic resonance spectroscopy using a Bruker Avance AV-400 NMR spectrophotometer, and  c Conductometric titrations.    Table 2.1: Characteristics of UHRA and the two UHRA analogs synthesized   General procedure for ITC ITC experiments were performed on a VP-ITC microcalorimeter from Microcal, Inc. (Northampton, MA). Antidotes and UFH solution were prepared in buffer as indicated. The average molecular weight of UFH (Leo pharma) was considered to be 15 kDa. UFH from a single vial was used for this study to minimize variation between experiments. The average molecular weight of protamine sulphate (Aldrich) was considered to be 5.1 kDa. All samples were filtered using 0.2 μm filters (Millex, Merck Millipore Ltd) and degassed prior to titrations. Injections of antidote solution were added from a computer controlled microsyringe at an interval of 5 minutes into the UFH sample solution in the cell at 298 K or at the indicated temperature (cell volume = 1.4 mL). Heat of dilution was measured by injecting antidote solution into the UFH-free buffer, and subtracted from the experimental curves prior to data analysis. A single identical-sites Langmuir-type binding model was fit to the resulting data using the MicroCal ORIGIN software supplied with the instrument to obtain the binding stoichiometry n, binding enthalpy ∆H, and equilibrium association constant Ka. The binding entropy ∆S was then estimated using the standard thermodynamic relation:  𝛥𝐺 =  𝛥𝐻 – 𝑇𝛥𝑆 = –  𝑅𝑇 ln 𝐾𝑎 (2.4)                            65   Neutralization of UFH anticoagulation activity by antidotes Ethics statement   The protocol for blood collection was approved by the Clinical Research Ethics Board (CREB; certificate number H10-01896) of the University of British Columbia, and written consent was obtained from donors in accordance with the Declaration of Helsinki.  Blood collection and plasma preparation Blood specimens from healthy consenting donors were collected by venipuncture into BD vacutainer glass tubes containing 0.105 M trisodium citrate. Blood was centrifuged at 150 x g for 10 minutes to separate platelet-rich plasma (PRP), and then spun at 1000 x g for 15 minutes for platelet-poor plasma (PPP) at 22°C. Activated partial thromboplastin time (aPTT) assay  Polymeric antidote solutions were prepared in 20 mM HEPES buffer (pH 7.4 and 150 mM NaCl). Heparinized human PPP was prepared by mixing UFH solution and citrated human PPP. The final concentration of UFH in PPP was 4 IU/mL. The anticoagulant neutralization activity was examined by mixing 20 μL of polymeric antidote solution with 180 µL of heparinized plasma (1:10 v/v). The final concentration of antidotes in plasma ranged from 12.5 to 250 µg/mL. Two hundred microliters of aPTT reagent (Dade® Actin® FS Activated PTT, Siemens/Dade-Behring) was then added to the sample and 100 µL of this resulting mixture was transferred to cuvette-strips at 37 0C. The clotting time was measured on a STart®4 coagulometer (Diagnostica Stago, France). HEPES buffer, 20 mM (pH 7.4 and 150 mM NaCl) added to heparinized plasma was used as a control for the experiments. All experiments were performed in triplicates and the average values (mean ± standard error of the mean) are reported. 66   Disruption of AT-UFH complexes: mechanism of action of UHRA  a) Fluorescence measurement of AT-UFH complexes   Fluorescence measurements were performed on a Cary Eclipse fluorescence spectrophotometer (Agilent). UHRA, AT and UFH solutions were prepared in 20 mM sodium phosphate buffer (150 mM NaCl, pH 7.4). The wavelength of excitation was 280 nm and the emission spectra were recorded from 300 nm to 400 nm using a 0.1 cm quartz fluorescence cuvette. AT (180 µL) (100 nM, final) was mixed with 25 µL of UFH (1000 nM, final) at 25 °C to obtain AT-UFH complexes. This mixture was then incubated with UHRA (25 and 50 µg/mL) for 5 minutes and the fluorescence was measured. Fluorescence measurements were corrected for dilution of AT or AT-UFH complex solution following the addition of UHRA. AT-UFH complex solution treated with buffer was considered to contain intact AT-UFH complex (no free AT) for calculating the percentage of AT bound to UFH in UHRA treated samples. b) Enzyme linked immunosorbent assay (ELISA) for AT UFH solution (4 IU/mL, final) was prepared in PBS (10 mM phosphate, 150 mM NaCl and pH 7.4) and coated onto wells of a heparin binding micro-titre plate by overnight incubation at room temperature. After 3X washing with PBS containing 0.05 %v/v Tween-20 (PBS-Tween), the plate was blocked with 0.2 % gelatin for 1 h at 37 °C. After 3X washing with PBS-Tween, the plate was incubated with AT (0.1 mg/mL, final) for 1 h at 37 °C. The plate was then incubated with UHRA or PBS (control) for 15 minutes at 37 °C. Following a 3X wash with PBS-Tween, sheep anti-human AT antibody conjugated with peroxidase (1:5000, final) was added, and the plate was incubated for 1 h at 37 °C. Following a 3X wash with PBS-tween, peroxidase substrate (ortho-phenylenediamine 6 67   mg/mL and hydrogen peroxide 0.012 %v/v) was added and incubated for 15 minutes at RT. The reaction was stopped by adding 1 M sulfuric acid. The plate was read at 492 nm using a plate reader. Controls consisted of a series of wells with PBS buffer instead of UHRA. For calculating the relative amount of AT bound (%), absorbance from the control sample was considered to be 100%.  Size of antidote-UFH complexes a) Atomic force microscopy (AFM) AFM was performed as described previously [165]. UFH (4 IU/mL, final) and antidotes in PBS (150 mM NaCl, pH=7.4) was mixed and incubated for 10 min at 37 °C to obtain antidote-UFH complexes. Three microliters of this mixture was then deposited on a thin mica disc and allowed to adsorb overnight. The mica disc was then washed with ethanol and air-dried. Images of complexes were acquired in tapping mode in air using AFM with a multimode Nanoscope IIIa controller (Digital Instruments, Santa Barbara, CA). The AFM is equipped with an atomic head offering a 130×130 µm2 scan range using a rectangular-shaped silicon probe (spring constant of 42 N/m and frequency of 320 kHz) (Bruker, TESP). b) Dynamic Light Scattering (DLS) The size of polymeric antidote-UFH complexes was measured by DLS using a Zetasizer Nano ZS (Malvern, Worcestshire, United Kingdom) The Nano ZS contains a 4 mW helium-neon laser operating at a wavelength of 633 nm. Antidote and UFH solution was prepared in PBS (150 mM NaCl, pH=7.4). All solutions were filtered through a Millipore filter with a 0.22 µm pore size. UFH and antidotes in PBS was mixed and incubated for 5 minutes at 25 °C to obtain antidote - UFH complexes. Solution was then transferred 68   into a polystyrene cuvette (SARSTEDT) for size measurements using the Non-Invasive Back-Scatter (NIBS) technique, which measures the light scattered by the sample at a backscatter angle of 173º. The program was set to conduct two sequential runs per measurement. Data was processed using the associated Zetasizer software (version 6.34; Malvern). Z-average hydrodynamic size (diameter in nm) is reported as the average of four values obtained from two independent experiments. Potentiometric titrations of UHRA UHRA (14.55 mg) was dissolved in 40 mL of water. The solution was acidified to pH ~2.0 by adding 200 µL of 1 N HCl, and then titrated with 0.01N NaOH. Equivalence point (pKa) was calculated from a plot of ΔpH vs ΔVolume of 0.01N NaOH. General procedure for conductometric titrations of UHRA Conductometric titrations were performed using a YSI model 35 conductance meter and 3403 cell with platinum electrode at room temperature. A syringe pump (Harvard Instruments) was used to inject NaOH at constant flow rate of 0.02 mL/min. For a typical titration, UHRA (8 mg) was dissolved in distilled water and acidified with 1 mL of 0.05 N HCl followed by titration with 0.05 N NaOH. Conductance of the solution was measured at every 15 seconds. From the volume of NaOH consumed at the equivalence point, the number of amine groups per polymer molecule was calculated. 2.4 Results and discussion 2.4.1 UHRA has a large core due to multi-valent ligands randomly displayed on the core UFH on average carries 2.7 sulphonate groups per disaccharide unit [201], which equates to 73 to 75 mono-anionic charges per 15 kDa heparin chain, or ca. – 5/kDa, at 69   physiologic pH. PS in turn carries ca. 21 to 25 mono-cationic charges (protonated arginine and lysine residues) per peptide chain [202], giving it an absolute charge density (~ + 4.2 to + 5 per kDa) comparable to that of UFH. Crowther et al. [132] have provided good evidence that this favorable matching of ligands of opposite charge sign is integral to the ability of PS to fully neutralize the anti-FXa activity of UFH.  In particular, they and others [203-205] have shown that the methods used to prepare LMWHs serve to decrease the average number of sulphonate groups per disaccharide unit (for example, enoxaparin, which is prepared by benzylation of UFH followed by alkaline depolymerisation, carries on average 2.6 sulphonate groups per disaccharide [203]). More significantly, those processing methods also increase the variability in the charge density of the polysaccharide chains, with Crowther et al. [132] reporting a low MW component of LMWH having a charge density ca. 25% lower than that of UFH (i.e. ~ – 3.8/kDa). This finding clearly supports their conclusion that the reduced anti-FXa neutralization activity of PS when applied to LMWH is not only due to a reduction in heparin MW, but also to a lower average sulphonate density and a higher variability in that charge density per chain. As a result, the distribution and spacing of negative charge differ more among the chains comprising LMWH. This analysis, though valid, ignores the potential impact of differences in the structure and stiffness of heparin variants on PS binding and activity.  Heparin exists in aqueous solution as an extended helix with a persistence length, which provides a measure of the stiffness (or structural order) of a polymer, of 2.11 nm, or approximately 5 disaccharide units [206].  This represents ~ 15% of the average length of UFH, but ~ 45% – 50% of that of LMWH. Thus, on a per molecule scale, the chains comprising 70   LMWH are stiffer, which in turn may affect their capacity to form a sufficient number of favorable ion pairs during complexation with PS, at least in the absence of an Fel that is attractive enough to disrupt the otherwise preferred chain order in aqueous solution.    The core of UHRA was therefore designed to achieve two objectives: 1) creation of a large-enough core to make Fel sufficiently attractive when the fuzzy macro-ion is interacting with its intended target, and 2) random positioning/patterning of the required number of cationic ligands on the core to enable high-affinity binding of heparin molecules (including chains within a given heparin preparation) that vary in the average number of sulphonate groups per disaccharide unit and in the manner in which those negative charges are distributed on each chain. Meeting the first objective is facilitated by our use of the methylated tris(2-aminoethyl)amine multi-cationic ligand [Figure 2.2A], which at sufficiently low pH carries a maximum charge of +4.  This creates the potential to achieve cationic charge densities within the HPG core that exceed that of PS since each cationic ligand in PS is monovalent.  As the average number of ligands displayed on the HPG core can be controlled, it also permits one to tune core to enable selective binding of heparins.   71    Figure 2.2: (A) The structure of methylated tris(2-aminoethyl)amine ligand on the UHRA molecule. At pH 7.4, the ligand is tri-protonated. (B) 2-dimensional cartoon of the structure (slice) of the UHRA molecule. The ligand is denoted as “R” in the structure of the UHRA molecule.   In UHRA, 25 (± 2) methylated tris(2-aminoethyl)amine ligands are displayed on a HPG core [Table 2.1, Figure 2.2B] that is 6 kDa in size based on GPC data [Appendix A.6] and comprised of 28 mol% mPEG350, as determined from 1H NMR spectral data [Appendix A.6].  Potentiometric titration data for UHRA are reported in Figure 2.3 and show that it carries an average molecular charge of ~ +100 at pH 2, with each ligand deprotonating one of its four charged amino groups at a pKa of 6.1, and the remaining cationic amino groups deprotonating near or above pH 8. Thus, at physiologic pH the HPG core carries a charge density of ~ +12/kDa, which is more than 2-fold higher than that of PS.  Compared to PS, both i and i (core in the case of UHRA) are therefore considerably larger. 72    Figure 2.3: The potentiometric titration curve for pure water (calculated; grey trace) and for UHRA in water (experimental; black trace) at RT.  The starting pH of the solution was adjusted to pH 2 using 0. 005M HCl. Synthesis of the HPG core of UHRA follows the method described by Shenoi et al. [165] and pioneered by the Frey laboratories [207], who have used small-angle neutron scattering to show that hyperbranched polyglycerols so formed are spherical in shape [208]. The hydroxyls within the HPG core are then tosylated and a nucleophilic substitution reaction used to link tris(2-aminoethyl)amine ligands to them. That 6 kDa HPG core carries about 150 hydroxyls. The 25 ligands can therefore be attached in up to W(25,150) = 150!/(25!(150 – 25)!) = 1.95 x 1028 different possible arrangements. As is generally the case for synthetic polymers, UHRA therefore exists as an ensemble of very similar molecules that differ in the pattern in which the multi-valent ligands are presented, enabling effective binding to the full ensemble of heparins, which as discussed are likewise comprised of molecules that vary in terms of charge distribution. 73   2.4.2 Binding of PS to UFH shows classic electrostatic-association behavior ITC-derived thermograms [Figure 2.4] show that binding of PS to UFH at pH 7.4 is exothermic, as expected for an electrostatically driven association reaction. Binding enthalpy ∆H values regressed from those data [Table 2.2] are expressed per mole of UFH, and therefore indicate that, on average, the pairing of each available sulphonate ion to a cationic amine on PS contributes ~ 0.5kT to at most 1kT to the binding enthalpy (or ~ 0.25kT to the binding energy ∆G). This is weak, less the thermal energy of the system, which emphasizes the fact ion pairs are generally not strong interactions in aqueous solutions because the relatively large dielectric constant of water weakens the interaction, and because the donors and acceptors of the water molecules themselves compete for binding to each ion in the pair.               74    Figure 2.4: ITC thermograms for antidotes binding to UFH at two different salt concentrations. In each thermogram, the top panel shows the raw heat signal and the bottom panel shows the differential binding curve showing both experimental data and the nonlinear least square fit. Titrations were performed in 50 mM sodium phosphate buffer with pH 7.4 at 298 K. (A) ITC thermograms obtained for antidotes binding to UFH at 0.01M NaCl concentration. (B) ITC thermograms obtained for antidotes binding to UFH at 0.15 M NaCl concentration.        75    NaCl (M) Antidote molecule n Ka ( M-1) ΔH (kcal/ mol) ΔG (kcal/ mol) TΔS (kcal/ mol)     0.010 PS 4.03  (±0.01)  1.07 (±0.08)×107 -19.8  (±0.014) -9.54 (±0.046) -10.3 (±0.06) UHRA 0.88  (±0.005)  2.0  (±0.04)×106 -124.5  (±0.71) -8.55 (±0.014) -115.9 (±0.06) N-UHRA 3.27  (±0.021)  4.44 (±0.176)×106 -22.6  (±0.11) -9.03 (±0.023) -13.57 (±0.08) mPEG750-UHRA 0.84  (±0.003)  1.21 (±0.084)×106 -101.3  (±3.08) -8.26 (±0.041) -93.05 (±3.12)     0.025 PS 3.67  (±0.06)  6.57 (±0.46)×106 -19.01 (±0.11) -9.26 (±0.041) -9.7 (±0.15) UHRA 0.88  (±0.007)  1.72 (±0.014)×106 -117.5  (±0.71) -8.46  (±0.01) -109.03 (±0.7) N-UHRA 2.8  (±0.035)  3.74 (±0.197)×106 -22.8  (±0.06) -8.92 (±0.031) -13.94 (±0.03) mPEG750-UHRA 0.74  (±0.164)  1.25  (±0.014)×106 -84.73  (±0.34) -8.28 (±0.006) -76.45 (±0.33)     0.050 PS 3.28  (±0.04)  4.87 (±0.52)×106 -19.36  (±0.19) -9.08  (±0.06) -10.3 (±0.25) UHRA 0.9  (±0.01)  1.34 (±0.01)×106 -107.5  (±0.7) -8.32 (±0.005) -99.2 (±0.7) N-UHRA 2.49  (±0.042)  2.92 (±0.176)×106 -21.15  (±0.48) -8.78 (±0.035) -12.36 (±0.44) mPEG750-UHRA 0.92  (±0.002)  8.51 (±0.084)×105 -78.54  (±0.27) -8.05 (±0.005) -70.48 (±0.28)     0.1 PS 3.48  (±0.08)  2.96 (±0.4)×106 -17.06  (±0.19) -8.79 (±0.08) -8.25 (±0.3) UHRA 0.93  (±0.02)  8.9 (±0.01)×105 -91.2  (±1.26) -8.08 (±0.001) -83.2 (±1.3) N-UHRA 2.52  1.47 (±0.134)×106 -17.07  (±0.42) -8.37 (±0.053) -8.69 (±0.37) mPEG750-UHRA 1.07  5.37 (±0.134)×105 -58.02 (±0.169) -7.78 (±0.014) -50.23 (±0.15)     0.150 PS 4.75  (±0.12)  4.61 (±0.33)×106 -16.3  (±0.26) -9.05 (±0.042) -7.21 (±0.3) UHRA 1.05  (±0.014)  4.46 (±0.13)×105 -84.14  (±0.11) -7.67 (±0.014) -76.5 (±0.098) N-UHRA 3.79  (±0.09)  5.84 (±0.79)×105 -15.57  (±2.08) -7.83  (±0.08) -7.74 (±2.16) mPEG750-UHRA 1.0  (±0.012)  2.44 (±0.056)×105 -70.83  (±0.59) -7.31 (±0.014) -63.51 (±0.607) 76   All titrations were performed in 50 mM sodium phosphate buffer (pH=7.4) at 298 K. Data is the mean ± standard deviation (indicated in parentheses) of results from two independent experiments. Ionic strengths were varied by adding sodium chloride (NaCl) into the buffer solution used to prepare the titrant and titrand.  Table 2.2: Thermodynamic parameters for binding of antidote molecules to UFH at different ionic strengths determined by ITC. The formation of each ion pair is therefore transient, with the tight and stable complex between UF and PS arising through an avidity effect associated with the presence of a relatively constant number of transient hetero-ion pairs between the two chains at equilibrium. In accordance with equation 2.1, log Ka for binding of PS to UFH depends linearly on √𝐼 [Figure 2.5A], showing a maximum intrinsic value as I → 0 (linear extrapolation) of ~7.2x107 M-1. This dependence is indeed observed for many macro-ion/ polyelectrolyte interactions [209,210].  But to further validate that this ionic strength dependence applies to the interaction of an unshielded macro-ion and an oppositely charged polyelectrolyte, we synthesized a HPG core bearing a sufficient number of methylated tris(2-aminoethyl)amine ligands to give it a charge density near that of PS. The molecule produced, which we call naked UHRA or N-UHRA, carries a charge density near + 4/kDa, which is slightly less than that of PS. Binding of N-UHRA to UFH is therefore characterized by a correspondingly weaker Ka [Table 2.2]; but more importantly, a linear dependence of log Ka on √𝐼 is again observed within experimental error [Figure 2.5B], despite the significant differences in the structures of PS and N-UHRA.    77     Figure 2.5: Dependence of Ka on salt concentration. (A) PS/UFH interaction, (B) N-UHRA/UFH interaction, (C) UHRA/UFH interaction and (D) mPEG750-UHRA/UFH interaction. The error bars on plots are the standard deviation for two independent determinations of each Ka.     78   2.4.3 At physiologic pH, the strong Fel from the high core of UHRA is balanced by the entropic repulsion of the brush The high charge density (~ +12/kDa) of the HPG core of UHRA enables it to bind UFH with a ∆H that, compared to the PS/UFH interaction at the same solution conditions, is 6-times more exothermic [Table 2.2]. Thus, in terms of binding enthalpy the UHRA/UFH interaction is quite favorable; at 37 °C; indeed, it is similar in magnitude to the formation enthalpy of a carbon-hydrogen bond (~ –100 to –110 kcal/mol).  Moreover, ∆H becomes more favorable with increasing temperature [Figure 2.6A].    Figure 2.6: Effect of temperature on the thermodynamics of UHRA/UFH interaction. (A) The plot of variation of enthalpy of binding with temperature. A linear relationship is observed. The heat capacity value was determined from the slope of the fitted line (ΔCp = -2.29 ± 0.2 kcal mol-1 K-1). (B) Enthalpy-entropy compensation plot for the UHRA/UFH interaction.  This is expected for a binding enthalpy dominated by electrostatic attraction in water, as it is predicted by equation 2.1 through the fact that  decreases with increasing 79   temperature, resulting in reduced solvent-mediated dielectric screening of the electrostatic force of attraction between the oppositely charged macro-ions.  From slope of the data reported in Figure 2.6A, it is also clear that the binding reaction is characterized by negative and temperature-invariant heat-capacity change ∆Cp of –2.3 ± 0.2 kcal mol-1 K-1 (R2 = 0.97). While changes in the hydration states of the two macro-ions resulting from their complexation may contribute to the observed ∆Cp [211], that contribution is likely to be small given the hydrophilic nature of interacting species. The large and negative ∆Cp therefore more likely simply reflects a decrease in system fluctuations about equilibrium [212], suggesting that the state of the solvent, and possibly ion-pairs formed between UHRA and UFH, becomes less transient with increasing temperature.   The ITC data also show that the UHRA/UFH interaction is characterized by an unfavorable change in entropy ∆S. Linear enthalpy-entropy compensation is observed for this complexation reaction [Figure 2.6B], as is the case for many if not most macromolecular binding events in aqueous solution [213]. What is therefore more important is the relatively large absolute value of ∆S compared to that recorded for the PS/UFH system. To understand this, we used pulsed-field gradient NMR experiments [214] to measure the diffusivity D at 25 °C of the HPG core of UHRA (D = 9.47 x 10-11 m2/s), which was then combined with the Stokes-Einstein equation to estimate the hydrodynamic radius rh of the core (rh = 2.1 nm). This enabled estimation of the average distance dg between the grafted mPEG350 chains (dg = √4𝜋𝑟ℎ2/(# of grafted mPEG chains) = 1.17 nm) [215].  As the Flory diameter dF for an end-grafted mPEG350 chain in 80   a good solvent is 2.8 nm (dF > dg), we find that the grafted polymer layer on the surface of UHRA adopts a soft brush conformation [197,216]. We then applied the self-consistent field theory of Steels et al [200], from which equation 2.3 was derived, to estimate the contribution to ∆S made by compression of a PEG350 brush at this grafting density. Given the uncertainty in the size and conformation of the interacting UFH molecule, this estimate is arguably more qualitative than quantitative in nature, but it suggests that at pH 7.4 ~ 30% to 40% of the absolute value of ∆S can be attributed to the entropy loss associated with UFH-mediated compression of the brush on the surface of UHRA.  As a result, the Ka for binding of UHRA to UFH is commensurate with that for the PS/UFH complex [Table 2.1], despite the considerably larger charge density of the UHRA core. Importantly, however, a 1:1 binding stoichiometry (n = 0.9) is observed for the UHRA/UFH complex [Figure 2.4], and that n is invariant with ionic strength. In contrast, asymmetric binding of multiple PS molecules to UFH is observed in the PS/UFH system, with n (molecules of antidote bound per UFH molecule) decreasing from 4 to 3 with increasing I. As noted above, the PS/UFH system is known to form hetero-aggregates [149,190], and the observed change in n is consistent with that observation. 2.4.4 Binding of UHRA to UFH diminishes non-linearly with √𝑰 and pH In a co-authored publication not included in this thesis, additional ITC data [165] are reported and show that the Ka for binding of UHRA to enoxaparin (LWMH) and fondaparinux is similar to that reported here [Table 2.2] for binding to UFH. The Chapter 4 of the thesis also presents data showing that UHRA interacts either quite weakly or not at all with fibrinogen, thrombin, AT or FXa at physiologic conditions [173]. UHRA 81   therefore shows selectivity towards heparin variants. Some insights into the physico-chemical origin of that selectivity are provided in Figure 2.5C, which shows that the affinity (log Ka) of the UHRA/UFH interaction diminishes non-linearly with √𝐼, as predicted for fuzzy-particle/macro-ion interactions [Figure 2.1]. Compared to that observed for the PS/UFH interaction [Figure 2.5A], this more complex dependence on ionic strength arises because Fel, which diminishes with I due to an increase in diffuse layer screening, is now compensated by the steric repulsion force Fsteric arising from brush compression. For a given brush configuration, stable complex formation therefore requires a threshold Fel and associated core.   High-affinity binding to heparin also requires UHRA to be in a certain charge state. The potentiometric titration data for UHRA [Figure 2.3] show that at physiologic pH (7.4), most of the methylated tris(2-aminoethyl)amine ligands carry a charge of +3, with no more than one ligand carrying a +4 charge due to the first pKa at ~ 6, and no more than two ligands carrying a charge of +2 due to the fact that the second pKa on some ligands lies above but relatively near 8. Perturbations away from physiological pH are therefore expected to change I; and they do. The binding affinity of UHRA for UFH plateaus around physiologic pH, and is then seen to rise sharply at lower pH due to protonation of the ligand, and to fall off sharply with increasing pH to progressive deprotonation of the ligand [Figure 2.7].   82    Figure 2.7: Effect of pH on the binding affinity of UHRA/UFH system. Near physiologic pH, the binding affinity of the UHRA/UFH system becomes relatively insensitive to pH. The binding affinity diminishes strongly with an increase in pH above 8. No measurable binding interaction is observed at pH 9.2. The connected line is for guidance to eyes.  Within the pH range defined by this plateau, Fel and Fsteric are properly balanced.  At lower pH, the increased ligand charge strengthens Fel such that the opposing force arising from brush compression becomes comparably negligible.  While at pH values above 8, the loss of ligand charge due to deprotonation renders the force of electrostatic attraction too weak to achieve brush compression and binding. Thus, because of the added steric repulsion force provided by the brush layer, binding equilibrium between UHRA and a polyanion is sensitive to UHRA’s charge state relative to that of the interacting polyanion. Based on this argument, we would expect binding to also be sensitive to subtle changes in the properties of the brush; and it is. Replacing the mPEG350 chains on UHRA with mPEG750 chains creates a slightly denser brush (dg and dF are now 1.27 nm and 2.1 nm, respectively) that still lies in the soft brush regime.  As 83   a result, the binding affinity is reduced [Table 2.2], and ln Ka shows a somewhat sharper decline with increasing √𝑰 [Figure 2.5D]. For UHRA, binding selectivity is therefore achieved both through proper matching of charge states by tuning the core size and ligand density, and through optimization of the brush properties, within UHRA. 2.4.5 Influence of antidote architecture on blood clotting, neutralization of UFH anticoagulation activity and mechanism of action Next, the impact of structural/chemical variation of PS, UHRA and the two UHRA analogues on the blood clotting cascade were investigated, including evaluation of their UFH anticoagulation neutralization activity using an activated partial thromboplastin time (aPTT) assay in human plasma. Studies have shown that polycationic molecules such as PS, polyethylenimine, polyamidoamine (PAMAM) dendrimers, and chitosan adversely affect blood clotting by inhibiting generation of thrombin, the blood protease that polymerizes fibrinogen into fibrin clot, and by binding to anionic blood components [173,135]. Figure 2.8A, shows clotting of plasma in the presence of antidotes at increasing concentrations. The data show that at concentrations above 50 µg/mL, both protamine and N-UHRA exhibit intrinsic anticoagulation activity, as demonstrated by an increase in plasma clotting time that reaches the maximum measurable level. This intrinsic anticoagulation activity may be due to impairment of thrombin generation by unshielded (naked) polycationic moieties [135]. Conversely, UHRA and mPEG750-UHRA both show minimal influence on the clotting even at 250 µg/mL. Next, the ability of antidotes to neutralize UFH (4 IU/mL, final- a high dose equivalent to cardio bypass surgery dose) anticoagulation activity was evaluated. Heparins exert anticoagulation activity by binding to AT with high affinity (K d ~ 50 nM) 84   [103]. This binding induces conformational changes in AT, resulting in a 300-fold increase in its inhibitory activity against the clotting protein factor Xa, thus reducing thrombin generation [Figure 2.8B] [217].  Figure 2.8: Effect of antidote molecules on clotting, and the ability of each molecule to neutralize the anticoagulation activity of UFH. (A) Plasma clotting time measured by aPTT in the presence of increasing concentrations of antidote. (B) A schematic illustrating the mechanism of action of 85   heparin-based anticoagulants. (C) Efficacy of antidotes to neutralize UFH (4 IU/mL, final) activity, evaluated by aPTT. (D) Percentage of AT present as AT-UFH complex following treatment with UHRA, analyzed by fluorescence spectroscopy. (E) Percentage of AT present as AT-UFH complex following treatment with UHRA, measured by ELISA for AT. (F) A schematic illustrating the mechanism of action of the UHRA antidote molecule. Structures presented in this figure are merely for illustrative purposes.   As shown in Figure 2.8C, PS at 50 or 75 µg/mL, respectively, completely neutralizes UFH anticoagulation activity. Similarly, 25 to 75 µg/mL N-UHRA neutralizes UFH anticoagulation activity. However, at concentrations above 75 µg/mL, both PS and N-UHRA show intrinsic anticoagulation activity, which demonstrates the narrow window for heparin neutralization by these unshielded polycationic antidote molecules.  In contrast, UHRA and mPEG750-UHRA completely neutralize UFH anticoagulation activity and normalize the elevated clotting over a wide range of antidote concentrations (50 to 250 µg/mL), without displaying any intrinsic anticoagulation activity. An essential feature of the molecular design, the soft mPEG brush on the surface of UHRA and mPEG750 -UHRA therefore increases the specificity of the antidote and its neutralization activity. In addition, the increased concentration range over which heparin neutralization is observed could increase the therapeutic window for UHRA molecules.  Further understanding of the mechanism by which UHRA neutralizes heparin anticoagulation activity was achieved by measuring the fluorescence spectra of the AT-UFH complex in the presence of UHRA. Binding of heparin to AT to form an AT-heparin complex enhances the intrinsic fluorescence of AT [218,219]. The increase in fluorescence intensity of AT in AT-UFH complexes (~40%) is due in part to 86   perturbations in the environment of tryptophan residues within AT following conformational changes upon heparin binding [219]. The intrinsic fluorescence intensity profile of AT within AT-UFH complexes and within AT-UFH complexes treated with UHRA was therefore measured and compared. The addition of UHRA is found to reduce the fluorescent intensity of AT. This reveals that the UHRA molecule disrupts the AT-UFH interaction to re-establish the native conformation of AT. From the fluorescent intensity of native AT and AT within the AT-UFH complex, the percentage of AT present in AT-UFH-UHRA complexes were calculated. Those data are presented in Figure 2.8D, which shows that addition of 50 µg/mL of UHRA into a solution containing AT-UFH complex results in the release of ~ 80% of AT from the complex.  To further confirm this mechanism, ELISA assays for AT binding were performed to quantify the percentage of AT present as AT-UFH complex in the presence of UHRA. Results of ELISA assay show that at either 50 or 75 µg/mL UHRA, the residual amount of AT bound to UFH is ~20 % [Figure 2.8E]. Thus, UHRA disrupts ~80% of the AT-UFH complex, thereby neutralizing the anticoagulation activity of UFH [Figure 2.8C].  Collectively, these results support a mechanism of action of UHRA where [Figure 2.8F], upon its addition into heparinized blood, UHRA specifically binds heparins, including those in AT-heparin complexes, leading to disruption of that active complex in blood to abrogate heparin’s anticoagulation activity.  It is important that neutralization activity is provided in a biocompatible manner.  Plasma clotting assays reveal that, unlike PS and N-UHRA, both UHRA and mPEG750-UHRA have minimal impact on clotting [Figure 2.8A]. The charge shielding and entropic penalty imposed by the mPEG chains provides that biocompatibility as unshielded 87   cationic molecules such as PS are known to bind anionic blood components [173,179] in ways that reduce hemocompatibility (shown by intrinsic anticoagulation activity) and antidote efficacy [Figure 2.8A,C]. 2.4.6 Influence of antidote architecture on the size of antidote-heparin complexes It has been suggested that toxicity associated with PS administration, including protamine-induced thrombocytopenia, complement activation and subsequent organ failure, is due to formation of large protamine-UFH complexes [220]. The size of such complexes will also define immune recognition, making regulation of complex size crucial to avoid immuno-toxic effects [221]. Turbidimetry analysis was therefore combined with dynamic light scattering and atomic force microscopy imaging techniques to analyze the sizes of various antidote- UFH complexes at physiologic pH.   In clinical practice, a dose of 1.0-1.3 mg of protamine is used to neutralize 100 IU of UFH [97]. Thus, for the complete neutralization of 4 IU/mL UFH, the dose of protamine should be at least 50 µg/mL. Figure 2.8C shows that, at a concentration of 75 µg/mL, each antidote completely neutralizes the anticoagulation activity of UFH (4 IU/mL). However, the oligomeric states of neutralizing agents differ when present at that concentration. As shown in Figure 2.9, a pronounced increase in the turbidity of protamine-UFH mixtures is observed when the concentration of protamine reaches or exceeds 50 µg/mL, which suggests formation of PS-UFH aggregates. Likewise, an increase in turbidity is recorded for the N-UHRA-UFH mixture when the concentration of N-UHRA is greater than 25 µg/mL. 88    Figure 2.9: Turbidimetric analysis of antidote-UFH complexes. No change in turbidity was observed in mixtures of UHRA and UFH (4 IU/mL) or of mPEG750-UHRA and UFH. However, a pronounced increase in turbidity is observed in mixtures of PS and UFH or N-UHRA and UFH at antidote concentrations above ~ 50 µg/mL, indicating formation of aggregates. However, no change in turbidity is observed with either the UHRA-UFH or the mPEG750-UHRA-UFH mixture up to antidote concentrations of 200 µg/mL [Figure 2.9].  To confirm these observations, the size of each antidote-UFH complex was measured using atomic force microscopy (AFM) and dynamic light scattering (DLS). AFM analyses were conducted by incubating each antidote (75 µg/mL) with UFH (4 IU/mL, final) for 10 minutes at 37 °C, and then placing the reaction product on mica substrates for imaging under dry conditions.  89   Figure 2.10: Characterization of sizes of antidote-UFH complexes at pH 7.4 (150 mM NaCl). (A) AFM images of antidote-UFH complexes. (B) Average size of dried antidote-UFH complexes calculated from AFM images. (C) z-average hydrodynamic diameter of antidote-UFH complexes measured by DLS. AFM images [Figure 2.10A] show that mixing of protamine or N-UHRA with UFH at pH 7.4 produces aggregates of large size, while mixtures of UHRA or mPEG750-UHRA with UFH do not. The average size of protamine-UFH and N-UHRA-UFH complexes visible in the AFM images is ~130 nm and ~160 nm, respectively. On the other hand, the average size of imaged UHRA-UFH and mPEG750-UHRA-UFH complexes is ~ 60 nm [Figure 2.10B]. Because the drying process required for AFM imaging can induce aggregation, the hydrodynamic diameter of antidote-UFH complexes in solution was measured using DLS. Complexes were prepared by 90   incubating antidotes with UFH in PBS buffer (150 mM NaCl, pH 7.4) for 5 minutes at RT. Figure 2.10C shows the resulting z-average hydrodynamic diameters of antidote-UFH complexes. Representative intensity size distributions, volume size distributions and correlation functions for these systems are shown in the Appendix A.7. As expected, mixing of protamine or N-UHRA with UFH, results in the formation of very large aggregates having hydrodynamic diameters in solution between 500 nm and 700 nm. In contrast, UHRA or mPEG750-UHRA mixed with UFH produces well-defined complexes with hydrodynamic diameters less than 20 nm [Figure 2.10C]. The high values of derived count rates (kcps, kilo counts per second) for mixtures of protamine or N-UHRA with UFH are also consistent with the formation of large aggregates [Appendix A.8]. Thus, though the average sizes determined from AFM images are likely influenced by the drying and washing procedure used, both studies show that protamine and N-UHRA form complex aggregates of a high oligomeric state with UFH, while the DLS studies show that the size of the UHRA-UFH complex and the mPEG750-UHRA complex formed in aqueous solution (pH 7.4) is consistent with the 1:1 stoichiometry recorded in the ITC experiments. Flexible, unshielded polyelectrolytes of opposite charge are known to aggregate [222], particularly if the complex formed is electroneutral or carries a greatly reduced charge density. If the polymers are amphiphilic, these complexes can show properties of complex-coacervate core micelles, where the inner insoluble core of each aggregate is shielded by more soluble chains that prevent further aggregation and macroscopic phase separation [223].  This suggests that the neutral polymer brush on the surfaces of UHRA and mPEG750-UHRA might serve to shield against aggregation because the 1:1 91   complex formed carries both a low charge density and a relatively high density of neutral, yet quite hydrophilic PEG chains [149]. Collectively, the results presented in this chapter therefore support the fundamental underpinnings used to design the UHRA molecule, and the crucial role of each of its structural components in achieving heparin neutralization activity and selectivity. The mPEG chains on UHRA impose an entropic penalty to binding. As a result, only those polyanionic molecules offering a complementary charge density, such as heparins, can make contacts with the cationic methylated tris(2-aminoethyl)amine ligands presented on the HPG core that are sufficient in both number and strength to form a stable complex. The neutral brush on UHRA thereby minimizes nonspecific interactions with blood proteins and other anionic species in blood [Figure 2.11].   As a result, while PS and N-UHRA display intrinsic anticoagulant activity and neutralize UFH within only a narrow range of concentrations, UHRA and mPEG750-UHRA show no adverse effect on clotting, and completely neutralize UFH activity over a wide range of concentrations. Finally, fluorescence spectroscopy and ELISA experiments reveal the mechanism of action of UHRA by showing that it disrupts AT-UFH complexes to reverse the anticoagulation activity of UFH.   92          Figure 2.11: A cartoon representing the presence of antidote molecules in heparinized blood. (A) UHRA molecule binds to UFH to form 1:1 stoichiometric complexes. The mPEG chains prevent nonspecific interactions of UHRA to blood proteins. (B) N-UHRA molecule binds UFH, forming hetero-aggregates containing 3 to 4 N-UHRA molecules per UFH molecule. Blood proteins bind to exposed cationic charges, which could likewise promote formation of large hetero-aggregates with N-UHRA. N-UHRA molecule could also bind to anionic glycosaminoglycans on the endothelium. (C) mPEG750-UHRA displays properties similar to the UHRA molecule. Structures presented in this figure are merely for illustrative purposes.      93   Chapter 3: Comparing mechanisms of anticoagulant antidotes  3.1 Synopsis Anticoagulants used to prevent and treat thrombosis include unfractionated heparin (UFH), low molecular weight heparins (LMWHs), fondaparinux and direct oral anticoagulants (DOACs). Bleeding risks associated with these anticoagulants demand careful monitoring and neutralization, and current antidotes, including protamine, have many known limitations. Improved antidotes are therefore in development and include the following: UHRA, a universal antidote for natural and synthetic heparins; andexanet alfa (AnXa), a biologic antidote for certain DOACs, as well as LMWH; and, ciraparantag (PER977), a synthetic small molecule reported to reverse UFH, LMWHs and certain DOACs. The binding affinities of each of these antidotes for their presumed targets have not been fully defined or compared, leaving questions about mechanisms of action unanswered. Here, isothermal titration calorimetry (ITC) was used to determine the affinity of each antidote for its putative targets and various coagulation-pathway components. Clotting assays were used to characterize neutralization activity, and scanning electron microscopy used to visualize the influence of each antidote on clot morphology in the absence or presence of anticoagulants. ITC confirms binding of UHRA to all heparins, AnXa to all of the DOACs studied and to AT-complexed with UFH or enoxaparin, and PER977 to UFH and enoxaparin, but not to the DOACs studied or to fondaparinux. A µM or stronger Kd was found to correlate with neutralization activity, except for the PER977/UFH and PER977/enoxaparin systems, where no reversal activity was observed. Standard metrics of clot structure were found to correlate only weakly with neutralization activity. This is the first study comparing three anticoagulant 94   antidotes in development, each of which appears to exert its reversal activity, when observed, through a unique mechanism of action. 3.2 Background Anticoagulants, including those based on unfractionated heparin (UFH), low–molecular weight heparins (LMWHs), and the synthetic pentasaccharide fondaparinux, as well as direct factor Xa (FXa) (e.g. apixaban, edoxaban and rivaroxaban) or thrombin (FIIa) inhibitors, also known as direct oral anticoagulants (DOACs), are widely administered to treat and prevent thromboembolism [224,225]. UFH and LMWHs remain the primary anticoagulants used to prevent and treat acute thrombotic events, including those arising in procedures requiring extracorporeal circulation such as hemodialysis and cardiopulmonary bypass surgery [226], while DOACs are increasingly used to prevent strokes due to atrial fibrillation, treat pulmonary embolism and deep-vein thrombosis, and prevent venous thrombosis following surgery [227]. Dosages must be carefully monitored and controlled to achieve the required anticoagulant activity while avoiding bleeding complications [228]. Regrettably, major bleeding events associated with the administration of either indirect or direct FXa inhibitors, as well as direct FIIa inhibitors such as dabigatran, remain a far too common and serious problem [227-229]. Protamine is the only FDA-approved anticoagulant antidote [230] and its use is restricted to neutralization of UFH [99,132]. Protamine only partially reverses the activity of LMWHs, shows no neutralization activity against fondaparinux, and is known to exhibit an unpredictable dose response and severe side effects [231,172]. Reliable clinical neutralization of the oral anticoagulant warfarin by vitamin K is plagued by similar challenges due to the narrow therapeutic index of vitamin K antagonists [116]. 95   Four-factor prothrombin complex concentrates (PCCs) have therefore recently been developed to reverse life-threatening bleeding due to warfarin anticoagulation [232].   In contrast, effective neutralization of the activities of UFH, LMWHs and fondaparinux, as well as DOACs, remains a medical deficiency that is motivating development of new antidotes. Recently, idarucizumab, a specific antidote to the FIIa inhibitor dabigatran, has been approved [233]. Other antidotes either nearing FDA approval or currently in development include UHRA [165] – a synthetic universal heparin reversal agent; andexanet alfa (AnXa; Portola Pharmaceuticals, Inc.) [160,163] – a recombinant variant of FXa designed to reverse the activity of both direct and indirect FXa inhibitors; and ciraparantag (PER977; Perosphere Inc.), an arginine-based small molecule reported to reverse direct FXa inhibitors, UFH and LMWHs, as well as some thrombin inhibitors [156,153]. To date, the binding affinities of each of these antidotes in development for their presumed targets have not been fully defined or compared. Isothermal titration calorimetry (ITC) was therefore used to identify unique and common binding partners among representative DOACs and heparins, as well as binding to relevant components of the blood coagulation cascade. The measured equilibrium dissociation constant Kd is reported for each binding partner detected by ITC. This collective dataset is combined with 1) clot formation kinetics and strengths before and after neutralization measured by clotting assays and thromboelastography (TEG), and 2) structures of fibrin and blood clots before and after neutralization determined by scanning electron microscopy (SEM) to further delineate molecular mechanism(s) of action for each anticoagulant antidote. 96   Each of the three antidotes was found to exert its reversal activity, when observed, through a unique mechanism of action. 3.3 Methods Ethics statement  Blood collection and the protocols used were approved by the Clinical Research Ethics Board (CREB; certificate number H10-01896) of the University of British Columbia, and written consent from donors was obtained in accordance with the Declaration of Helsinki.  Reagents Unfractionated heparin (LEO Pharma Inc. Canada), enoxaparin (Lovenox®, Sanofi-Aventis Canada), fondaparinux (Arixtra®, GlaxoSmithKline Canada); edoxaban and rivaroxaban were kindly provided by Portola Pharmaceuticals Inc. Human fibrinogen (Fibrinogen) and N-2-hydroxyethyl piperazine-N’-2-ethanesulfonic acid (HEPES) were obtained from Sigma-Aldrich. Human proteins such as human FXa, thrombin, FIX, FIXa and antithrombin were obtained from Haematologic Technologies Inc.  Ciraparantag (PER977) and Andexanet alfa were provided by Portola Pharmaceuticals Inc., with their protocol for synthesizing PER977 detailed below. The UHRA molecule (UHRA-7) used for this study was synthesized as described in Shenoi et al [165]. Polystyrene 96-well microplates (Costar) used for all clotting assays were purchased from Corning. Synthesis of PER977 Portola Pharmaceuticals Inc, synthesized PER977, (2S,2’S)-N,N’-(piperazine-1,4-diylbis(propane-3,1-diyl)bis(2-amino-5-guanidinopentanamide) (L-enantiomer), as described in the patent US2013/0137702.  PER977, (2S,2’S)-N,N’-(piperazine-1,4-97   diylbis(propane-3,1-diyl)-bis-(2-amino-5-guanidinopentanamide) was synthesized in two steps via coupling of 3,3’-(piperazine-1,4-diyl)-bis-(propan-1-amine)  and (S)-5-(1,3-Bis ((benzyloxy)carb onyl)guanidino)-2-((tert-butoxy-carbonyl)amino) pentanoic acid (S-Boc-DiBoc-Arg-OH) followed by removal of the Boc protecting group. The product was characterized by proton NMR and mass spec analysis to confirm the structure identity [Appendix B.1], and the purity was determined by HPLC analysis. Isothermal titration calorimetry All ITC experiments were performed in phosphate buffered saline (PBS) pH 7.40 (± 0.1) (Gibco). To enhance solubility, rivaroxaban and edoxaban titrations were performed in 1% Dimethyl sulfoxide (DMSO) solutions prepared from 10 mM anticoagulant stock solutions in DMSO. UFH, enoxaparin or fondaparinux samples were prepared by dilution into PBS of concentrated stock solutions in isotonic NaCl.  Buffer exchange into PBS of AnXa (received in 10 mM Tris, 45 mM arginine HCl, 2% sucrose, 5% mannitol, 0.01% Tween 80, pH 7.8), PER977 (received in saline), or fibrinogen (Sigma F4883, received lyophilized and prepared in PBS) was done by gel filtration using a GE PD MidiTrap G25 column. Factors Xa, IX and IXa (Haematologic Technologies Inc.) were prepared from stock solutions in 50% (v/v) glycerol/water by dialysis into PBS using a Pierce Slide-a-lyzer 10kD cassette. All solutions were degassed with stirring at room temperature before loading in the ITC. ITC experiments were performed using a MicroCal iTC200 (Malvern Instruments). UHRA experiments were performed by injecting consecutive 2 µL aliquots of ligand solution (200 – 220 µM UFH, enoxaparin or fondaparinux) into the ITC cell (volume = 200 µL) containing UHRA (8 – 10 µM).  AnXa titrations were performed by 98   injecting consecutive 2 µL aliquots of ligand solution (40 µM rivaroxaban, 100 µM edoxaban, or 400 µM UFH, enoxaparin or fondaparinux) into AnXa (5–36 µM). PER977 titrations were performed by injecting consecutive 2 µL aliquots of PER977 solution (200 µM - 1.82 mM) into rivaroxaban, edoxaban, fibrinogen, FXa, FIX or FIXa (20 – 182 µM). PER977 titrations were also conducted by injecting consecutive 2 µL aliquots of ligand solution (400 µM UFH, enoxaparin or fondaparinux; various concentrations of rivaroxaban or edoxaban up to 1.5 mM) into PER977 (5 – 200 µM). The ITC data were corrected for the heat of dilution of the titrant by subtracting mixing enthalpies for 2 µL injections of ligand into buffer. At least 2 independent titration experiments were performed for each system at 25C. The binding stoichiometry, N, and Kd for each measured interaction were determined by fitting the corrected data to a standard bimolecular interaction model [234,235]. Fibrin polymerization assay PER977 or AnXa (15 µL at various concentrations) was added to 135 µL of purified fibrinogen (3 mg/mL, initial concentration) in 20 mM HEPES (pH 7.4, 150 mM NaCl) buffer. Clotting at 37C was initiated in 100 µL of this solution by adding 5 µL of CaCl2 (3 mM, final) and 5 µL of thrombin (2.5 NIHU/mL, final). Fibrin formation was monitored by recording changes in turbidity (A405) every 30 seconds on a Spectramax microplate reader (Molecular Devices) for 1 hour at 37°C. Final turbidity of fibrin clots was also measured.   99   Scanning electron microscopy (SEM) of fibrin and whole-blood clots The morphology of clots formed in the presence of AnXa or PER977 was determined by SEM. All samples were randomly coded and blinded to the individual performing imaging analyses to avoid bias. Fibrin clots were prepared in sterile, 5 mL round-bottom polypropylene tubes (BD Falcon) by mixing 200 µL of human fibrinogen (3 mg/mL, initial concentration) in 20 mM HEPES (pH 7.4 and 150 mM NaCl) buffer with 2.5 NIHU/mL thrombin (final), 3 mM CaCl2 (final) and AnXa or PER977 (in HEPES buffer). Control clots were prepared in the absence of anticoagulant and/or antidote. After incubation for 1 hour at 37C, clots were fixed using Karnovsky fixative (2.5% glutaraldehyde and 4% formaldehyde), repeatedly washed with 0.1 M sodium cacodylate buffer (pH=7.4), then subjected to post-fixation with 1% v/v osmium tetroxide. The fixed samples were washed three times with distilled water and dehydrated with a gradient series of ethanol (20-95% v/v). Clots were then dried with CO2 in a Tousimis Autosamdri 815B critical-point dryer, mounted onto stubs, and gold sputter-coated for SEM using a Hitachi S-4700 field emission microscope at different magnifications. Multiple images from different areas of each clot were captured. Fiber diameters of clots were measured with ImageJ [26]. For fibrin fiber diameter calculations, images from two independent experiments were analyzed. Fibrin fiber diameters (n = 60) from 8 separate areas of each clot were used to calculate the mean fiber diameter.  Blood clots were also prepared in 5 mL polypropylene tubes by incubating 180 µL of non-citrated blood with or without edoxaban (200 ng/mL, final) and with or without 20 µL of AnXa or PER977 in 20 mM HEPES buffer at 37°C. Blood clots were then 100   processed for SEM. Fiber diameters of clots were measured with ImageJ. Fibrin fiber diameter calculations were performed as described previously [237]. Clotting and anticoagulant neutralization assays Activated partial thromboplastin time (aPTT) assay AnXa, UHRA or PER977 solutions were prepared in HEPES-buffered saline. Non-anticoagulant whole blood was mixed with UFH or enoxaparin (2 IU/mL, final concentration) and then centrifuged at 3000 rpm for 5 minutes to prepare heparinized plasma. The anticoagulant neutralization activity was examined by mixing 20 μL of AnXa, UHRA or PER977 solution with 180 µL of heparinized plasma (1:10 v/v). The final concentration of antidotes in plasma ranged from 25 to 200 µg/mL. aPTT reagent (200 µL) was then added to the sample and 100 µL of this resulting mixture was transferred to cuvette-strips at 37C. The clotting time was measured on a STart®4 coagulometer (Diagnostica Stago, France). HEPES-buffered saline added to heparinized plasma was used as a control for the experiments. All experiments were performed in duplicate and the average values (mean ± standard error of the mean) are reported. Turbidimetric plasma clotting assay Turbidimetric plasma clotting assays were performed with platelet-poor plasma (PPP) in 96-well microplates (costar). PER977, UHRA and AnXa solutions were prepared in 20 mM HEPES (pH 7.4, 150 mM NaCl) buffer. Non-anticoagulant whole blood was mixed with UFH (2 IU/mL, final concentration) and then centrifuged at 3000 rpm for 5 minutes to prepare heparinized plasma. Clotting was initiated in 180 µL of 50% diluted heparinized PPP (2 IU/mL UFH, final concentration), spiked with 20 µL of UHRA, AnXa 101   or PER977 (dilution 1:10). Clotting was evaluated by monitoring changes in turbidity (A405) every 30 seconds on a Spectramax microplate reader for 2 hours at 37C. Clotting time was taken as the time point when an exponential increase in absorbance was observed.  Thromboelastography Non-AC whole blood was mixed with edox (200 ng/mL, final). The neutralization of edoxaban by PER977 or AnXa was then analyzed using a Thromboelastograph Hemostasis System 5000 (TEG) from Haemoscope Corporation at 37C. Anticoagulant whole blood (360 µL) was mixed with each antidote (40 µL) to give a final concentration of 0.1 mg/mL. 360 µL of this blood mixture was transferred into a TEG cup within 5 minutes of blood draw and the coagulation then monitored. HEPES-buffered saline and edoxaban anticoagulated blood were used as controls. Data analysis Data are presented as mean ± standard error (SE) values from n (≥ 3) independent experiments unless otherwise specified. ITC data analysis and plots were done using Origin software from Microcal, Inc. (Northampton, MA).  All other results were plotted and analyzed using GraphPad Prism 6.0 (La Jolla, CA). Statistical significance was determined using a Student’s t-test or by one-way ANOVA followed by a Dunnett post hoc test. P values < 0.05 were considered statistically significant. 3.4 Results 3.4.1 ITC reveals binding partners common and unique to the three antidotes The interaction of UHRA, AnXa or PER977 with their putative binding partners, as well as with relevant components of the blood coagulation cascade was studied by ITC. The 102   procedure used follows that of Kormos et al [238]., with the specific conditions used described in the methods section. Each of the antidotes studied here is thought to directly or indirectly target specific anticoagulants to provide a reversal activity that restores normal hemostatic mechanisms. To better delineate those targets, ITC was used to determine binding affinities to each putative target as well as to relevant components of the coagulation pathway.  Table 3.1 reports the mean Kd and standard deviation (n = 3) measured by ITC for each binding partner for which a measureable (mM or tighter) binding interaction was observed. Putative binding partner Kd (M)         UHRA Andexanet alfa Ciraparantag  UFH Enoxaparin Enoxaparin/AT complex Fondaparinux  8.9 (± 0.3) x 10-7 6.0 (± 0.4) x 10-6 - 2.4 (± 0.6) x 10-6  3.4 (± 0.4) x 10-5 4.0 (± 0.3) x 10-5 7.1 (± 0.5) x 10-7 1.1 (± 2.0) x 10-4  2.8 (± 0.3) x 10-5 1.7 (± 0.3) x 10-5 6.9 (± 2.8) x 10-4 9.6 (± 3.9) x 10-3  Edoxaban Rivaroxaban   N.B.* N.B.  3.4 (± 0.6) x 10-9 0.9 (± 0.2) x 10-9  N.B. N.B.  h-Factor Xa h-Factor IX h-Factor IXa h-Thrombin h-Antithrombin h-Fibrinogen  N.B. - - N.B. N.B. N.B.  N.B. - - - - N.B.  N.B. N.B. 3.5 (± 0.4) x 10-5 N.B. N.B. N.B. 103   * N.B. – No binding detected Table 3.1: Equilibrium dissociation constants (Kds) determined by isothermal titration calorimetry for each reversal agent titrated into various potential binding partners.  UHRA has sub-µM binding affinity for each heparin-based anticoagulants studied [Figure 3.1], including fondaparinux, while showing no measureable interaction with FXa, antithrombin III (ATIII) or fibrinogen. UHRA was not designed to neutralize DOACs, so they were not included in the panel of putative UHRA-binding partners studied.     0.0 0.5 1.0 1.5 2.0 2.5 3.0-80-60-40-200-7.5-50Time (min)-10 10 30 50 70 90 110 130-2.5µcal/seckcal/mole of injectantUFH/UHRA molar Ratio0.0 0.5 1.0 1.5-120-100-80-60-40-20-10-500 20 40 60 80 100 120 140Time (min)µcal/secEnoxaparin/UHRA molar Ratiokcal/moleofinjectantA B 104                 Figure 3.1: Isothermal Titration Calorimetry (ITC) data for UHRA binding to indirect FXa inhibitors. Raw (top panel) and cumulative (bottom panel) heat data for titrations at 25°C of:  A) 200-µM UFH into 10-µM UHRA; B) 220-µM enoxaparin into 10-µM UHRA; and C) 220-µM fondaparinux into 10-µM UHRA.  All differential heat peaks for 2 µL injections; buffer: PBS with 1% DMSO. AnXa bound the largest range of anticoagulants. Direct binding to UFH and enoxaparin was recorded [Figure 3.2] at affinities roughly an order of magnitude weaker than that observed for UHRA. However, with the addition of human AT, which binds to a pentasaccharide-sulfation sequence within heparin to stimulate FIIa and FXa inhibition [226], a Kd commensurate with that reported for the UHRA:UFH complex is recorded, suggesting that AnXa preferably binds to heparin-activated AT. In addition, AnXa binds 0.0 0.5 1.0-60-50-40-30-20-10010-50-10 10 30 50 70 90 110 130Time (min)µcal/secFondaparinux/UHRA molar ratiokcal/moleofinjectant0.25 0.75-2.5C 105   each DOAC studied with nM affinity [e.g., Figure 3.3]; binding to rivaroxaban [Figure 3.2] is exceptionally high affinity (Kd = 0.9 nM).   0.0 0.5 1.0 1.5 2.0 2.5-25-20-15-10-50-1.0-0.50.00.50 10 20 30 40 50 60 70Time (min)µcal/secEnoxaparin/Andexanet Molar Ratiokcal/mole of injectantUFH/Andexa et alfa	Molar	Ratio						0.0 0.5 1.0 1.5 2.0 2.5-18-16-14-12-10-8-6-4-20-0.8-0.7-0.6-0.5-0.4-0.3-0.2-0.10.00.10 10 20 30 40 50 60 70Time (min)µcal/secEnoxaparin/Andexanet Molar Ratiokcal/mole of injectantA B 106    Figure 3.2: Isothermal Titration Calorimetry (ITC) data for andexanet alfa (AnXa) binding to various anticoagulants or to the AT-enoxaparin complex. Raw (top panel) and cumulative (bottom panel) heat data for titrations at 25°C of:  A) 400-µM UFH into buffer (top titration of upper panel) or 40-µM UFH (bottom titration of upper panel); B) 400-µM enoxaparin into buffer (top upper) or 36-µM AnXa (bottom upper); C) 44.5-µM AnXa into buffer (top upper) or 8-µM each of enoxaparin and ATIII (bottom upper); D) 40-µM rivaroxaban into buffer (top upper) or 5-µM AnXa (bottom upper). All differential heat peaks for 2 µL injections; buffer: PBS with 1% DMSO. 0.0 0.5 1.0-40-35-30-25-20-15-10-50-0.2-0.10.00 10 20 30 40 50 60 70Time (min)µcal/secMolar Ratiokcal/mole of injectant0.0 0.5 1.0 1.5-22-20-18-16-14-12-10-8-6-4-202-0.2-0.10.00.10 10 20 30 40 50 60 70Time (min)µcal/secRivaroxaban/Andexanet Molar Ratiokcal/mole of injectantC D 107    Figure 3.3: ITC data for AnXa or PER977 binding to edoxaban. Raw (top panel) and cumulative (bottom panel) heat data for titrations at 25°C of:  A) 100-µM edoxaban into buffer (top upper panel) or 13-µM Andexanet alfa (bottom upper panel); B) 200-µM edoxaban into buffer (top upper panel) or 20-µM Per977 (bottom upper panel).  All differential heat peaks for 2 µL injections; buffer:  PBS with 1% DMSO. PER977 binds fondaparinux  weakly, and binds UFH or enoxaparin [Figure 3.4] with an affinity similar to that observed for direct binding of AnXa to the same targets (Kd ~ 25 µM). Unlike with AnXa, binding of PER977 is weakened when enox is complexed with AT. Finally, using ITC, we did not observe binding of PER977 to either edoxaban or rivaroxaban [Figure 3.3B and Figure 3.4], but did observe direct binding to FIXa [Figure 3.4]. Direct binding to other components of the blood coagulation cascade was not observed for any of the three neutralizing agents.                                                      0.0-0.2-0.40 10 20 30 40 50 60-0.4-0.20.00 10 20 30 40 50 60 70BTime (min)µcal/secA  Time (min)µcal/sec108       0.0 0.5 1.0 1.5-12-10-8-6-4-20-2-100 10 20 30 40 50 60 70Time (min)µcal/secEnoxaparin/Per977 Molar Ratiokcal/mole of injectant0.0 0.5 1.0 1.5 2.0 2.5-25-20-15-10-50-0.4-0.20.00.20.40 10 20 30 40 50 60 70Time (min)µcal/secFondaparinux/Per977 Molar Ratiokcal/mole of injectant	0.0 0.5 1.0 1.5 2.0 2.5 3.0 3.5-25-20-15-10-50-0.4-0.20.00.20.40 10 20 30 40 50 60 70Time (min)µcal/secPer977/Rivaroxaban Molar Ratiokcal/mole of injectant	A B C 109       0 1 2 3 4-25-20-15-10-50-0.4-0.20.00.20.40 10 20 30 40 50 60 70Time (min)µcal/secPer977/Fibrinogen Molar Ratiokcal/mole of injectant0.0 0.5 1.0 1.5 2.0 2.5-25-20-15-10-50-0.4-0.20.00.20.40 10 20 30 40 50 60 70Time (min)µcal/secPer977/F-IX Molar Ratiokcal/mole of injectant	D E F 110   Figure 3.4: Isothermal Titration Calorimetry (ITC) data for PER977 binding to various anticoagulants. Raw (top panel) and cumulative (bottom panel) heat data for titrations at 25°C of:  A) 1.5-mM enoxaparin into buffer (top upper panel) or 200-µM PER977 (bottom upper panel); B) 400-µM fondaparinux into buffer (top upper panel) or 40-µM PER977 (bottom upper panel); C) 200-µM PER977 into buffer (top upper panel) or 20-µM rivaroxaban (bottom upper panel); D) 200-µM PER977 into buffer (top upper panel) or 10-µM human fibrinogen (bottom upper panel); E) 200-µM PER977 into buffer (top upper panel) or 20-µM human Factor IX (bottom upper panel); F) 10-µM PER977 into buffer (top upper panel) or 200-µM human Factor IXa (bottom upper panel). All differential heat peaks for 2 µL injections; buffer: PBS with 1% DMSO.  3.4.2 Binding data supports different mechanisms of action UHRA: Using intrinsic fluorescence (IF) studies, Shenoi et al. [165] found that the UFH: AT complex is disrupted by the addition of UHRA, resulting in loss of anticoagulant activity. This finding aligns with our ITC results, which show that UHRA directly binds UFH, LMWH (enoxaparin), or fondaparinux with sub-µM affinity [Table 3.1]. Taken together, the IF and ITC results therefore support direct binding of UHRA to heparin-based anticoagulants (at a strength that enables displacement of AT in the anticoagulant: AT complex) as the primary mechanism by which UHRA neutralizes the activity of UFH or LMWH. Andexanet Alfa: AnXa is a catalytically inactive form of human FXa that contains an S419A mutation at the active site of its protease domain and lacks the γ-carboxyglutamic acid domain required for prothrombinase complex formation [160]. Siegel et al. have recently reported on two clinical trials designed to test the safety and efficacy of AnXa in reversing the anticoagulant activity of rivaroxaban or apixaban in healthy older volunteers [163]. For either DOAC, their results show that AnXa neutralizes anticoagulant activity as evidenced by the rapid reduction (>90%) in anti-111   FXa activity and free fraction of the anticoagulant, as well as restoration of thrombin generation [239]. Our ITC data support those findings by showing that AnXa directly binds rivaroxaban with nM affinity (and 1:1 stoichiometry), consistent with its proposed mechanism of action in reversing DOAC-mediated FXa inhibition by reducing unbound levels of anticoagulant in plasma [163,240]. Reversal of the activity of edoxaban by AnXa has also been reported [163], and our ITC results suggest that neutralization is likewise achieved via direct and tight binding of the DOAC to AnXa [Table 3.1]. Direct binding of AnXa to fondaparinux, enoxaparin and UFH is also observed. However, drug discovery programs are usually built around the concept that candidate drugs should display µM or tighter affinities to their molecular target [241]. The Kd for direct binding of AnXa to any of these heparins does not meet that threshold; but the binding affinity of AnXa to either the enoxaparin:AT or UFH:AT complex does [Table 3.1]. These results, taken together with the fact that direct binding of AnXa to AT was not observed, suggest that the major mechanism of AnXa’s activity in neutralizing AT-dependent inhibitors is via binding to the AT: anticoagulant complex, and not through direct sequestration of the anticoagulant [242,243]. PER977: By ITC, we saw no conclusive evidence of a binding interaction between PER977 and either edoxaban or rivaroxaban [e.g. Figure 3.3B] in PBS (with no citrate or other chelator present). This result is at odds with a previous study [244] reporting non-covalent binding of PER977 to DOACs based on indirect dynamic light scattering (DLS) data. Based in part on those DLS data [244], direct binding of DOACs has been proposed as the putative mechanism of action of PER977, but our findings suggest that PER977 might derive its purported DOAC-neutralization activity in a different manner.   112   Some insight into that alternate mechanism of action is provided by our finding that PER977 binds FIXa. Since FIXa is the key enzyme in the intrinsic Xase that activates FX to FXa leading to thrombin generation, it has long been recognized as a potential target for modulating coagulation [1,245]. Indeed, FIXa inhibitors are an emerging class of anticoagulants that includes synthetic oral inhibitors, monoclonal antibodies, and aptamers [245]. The observed binding of PER977 to FIXa raises the question as to whether it might serve to alter the activity of the FIXa/FVIIIa complex in recruiting FX to the surface of platelets for activation, similar to polylysine [246,247]. In concordance with previously reported DLS data [244] we do observe direct binding of PER977 to enoxaparin [Table 3.1]. However, the Kd for binding of PER977 to enoxaparin is weaker than that for either UHRA binding to enoxaparin or for AnXa binding to the enoxaparin: AT complex. Interestingly, that binding affinity is attenuated when enoxaparin is complexed with AT.  Finally, activity against fondaparinux through a direct binding mechanism is not supported by our ITC data, as weak affinity (Kd ~ mM) is observed for that complex. 3.4.3 Fibrin polymerization and clot imaging permits identification of antidote effects on clot morphology Cessation of bleeding at the site of injury is mediated by polymerization of fibrin from thrombin-cleaved fibrinogen to form a fibrin clot [44]. A number of factors, including thrombin, calcium and fibrinogen concentration, as well as pH, modulate the formation kinetics, structure and stability of a fibrin clot [44,50]. In general, high thrombin levels produce thinner clot fibers that are more resistant to lysis when compared to the generally thicker fibers formed at lower thrombin concentrations [44,50].  Both AnXa 113   and PER977 are reported to halt bleeding caused by DOACs and heparin anticoagulants to generate a hemostatic clot [248]. To better define their mechanisms of action, we therefore asked whether these anticoagulant reversal agents have any impact on fibrin clot formation and structure. Our study focused on AnXa and PER977, as we have recently shown UHRA does not alter fibrin polymerization or clot structure [173] (see also the chapter 4).  Figure 3.5: Antidotes alone at therapeutic doses do not affect fibrinogen polymerization. Clotting was initiated by adding 2.5 NIHU/mL of thrombin and 3 mM CaCl2 to 3 mg/mL fibrinogen solution, incubated with Andexanet Alfa or PER977 for 10 minutes at 37OC. Fibrin formation was monitored 114   by measuring turbidity at 405 nm (A405nm) for 1 hour. (A, C) Absorbance changes following fibrin fiber formation in the presence of Andexanet Alfa or PER977. For clarity, error bars are avoided. (B, D) The final turbidity of mature fibrin clots produced from fibrinogen in the presence of Andexanet Alfa or PER977. Even though, PER977 (500 µg/mL) showed a subtle increase in fibrin turbidity the value did not reach statistical significance compared to buffer control. However, fibrin turbidity was significantly reduced with Andexanet Alfa (500 µg/mL), compared to the buffer control. All experiments were performed in triplicate. Results are expressed as the mean ± SE of six measurements from two independent experiments. Unpaired 2-tailed t tests were performed to determine significance, with p < 0.05 indicating a statistically significant change. Fibrin polymerization in the presence of AnXa (25-200 µg/mL) or PER977 (25-500 µg/mL) did not significantly change the maximum optical density, which correlates with the average fibre-cross-sectional area [249], in comparison to the buffer control [Figure 3.5 A-D]. However, AnXa at a supra-therapeutic concentration (500 µg/mL, ~12 µM) produced clots with reduced optical density, suggesting impaired clot formation and/or the formation of thinner fibrin fibers.  To gain further insights, we visualized the ultra-structures of fibrin and blood clots formed in the presence of either antidote using SEM to determine whether these molecules have any effect on fibrin fiber diameter. No anticoagulant was present in these studies. Fibrin clots formed in the presence of PER977 at concentrations of 25 to 500 µg/mL show no change in clot structure or fibrin fiber diameter relative to the antidote-free control.  The same is true for AnXa at 25 and 200 µg/mL [Figure 3.6A-C]. However, consistent with the fibrin-polymerization data, AnXa at 500 µg/mL gave thinner fiber structures [Figure 3.6A, B].   115    Figure 3.6: Fibrin architecture and fiber size remain unaltered at therapeutic doses of antidotes alone. Clots were made by incubating 3 mg/mL human fibrinogen in 3.0 mM CaCl2 plus Andexanet Alfa or PER977, then initiating clotting with 2.5 NIHU/mL thrombin. Clots were then allowed to mature for 1 hour and processed for SEM imaging. Images were acquired using Hitachi S-4700 field emission scanning electron microscope (A) Scanning electron micrographs of fibrin clots formed in the presence of antidotes at different concentrations (25, 200 and 500 µg/mL) are depicted. Clot architectures formed in the presence of antidotes are comparable to buffer control. Clot images were taken at four magnifications 2500X, 5000X, 10,000X and 25,000X, respectively. Only images from the 25,000X magnification are depicted. (B, C) Fibrin fiber diameters of clots 116   formed in the presence of Andexanet Alfa or PER977. Fiber diameter is measured from scanning electron micrographs using ImageJ software. A total of 60 fibers were analyzed. Fibers for size analysis were selected from four images by probing four different spots in each image. Data are mean ± SE (n=60) from two independent experiments. Statistical significance for fiber diameter was determined by comparing the antidote treated group to the control using a one-way ANOVA followed by a Dunnett post hoc test.  Fibrin fibers formed in the presence of 500 µg/mL Andexanet Alfa are significantly thinner than those in the control clot (***p <0.001). Likewise, anticoagulant-free blood clots formed in presence of AnXa or PER977 do not show any significant morphological changes, having normal clot components and normal fibrin fibers when compared to the buffer control [Figure 3.7A, B]. At a supra-therapeutic concentration (500 µg/mL), AnXa yields thinner fibers [Figure 3.7A, B].    117   Figure 3.7: Antidotes alone do not influence blood clot morphology. Blood clots were generated by incubating antidotes in non-citrated whole blood at 370C. (A) Clots formed in the presence of both Andexanet Alfa and PER977 did not show any major morphological changes. Clot images were taken at two magnifications 2500X and 5000X, respectively. Only images from the 5000X magnification are depicted. (B, C) Blood clot fiber diameters formed in the presence of Andexanet Alfa or PER977. Fiber diameter is measured from scanning electron micrographs using ImageJ software. A total of 60 fibers were analyzed from two independent experiments. Fibers for size analysis were selected from four images by probing four different spots in each image. Data are mean ± SE (n=60). Statistical significance for fiber diameter was determined by comparing the antidote treated group to the control using a one-way ANOVA followed by a Dunnett post hoc test.  Fibers formed in the presence of 500 µg/mL Andexanet Alfa are significantly thinner than those in the control clot (**p <0.01).  3.4.4 SEM studies reveal the efficacy of antidotes against direct FXa inhibitors in restoring impaired fibrin fiber development in anticoagulated blood Ansell et al. reported the ability of PER977 to normalize the mean fibrin fiber diameter in edoxaban-anticoagulated blood based on whole-blood clotting time data [153].  Imaging of the clots and their morphologies was not included in that study. To advance a mechanistic understanding of PER977 activity, we therefore explored the restoration of impaired clot formation in edoxaban-anticoagulated blood using SEM analysis. AnXa was included in the study as well. Neutralization of edoxaban (200 ng/mL, final) with AnXa at either 100 or 200 µg/mL normalized the impaired fibrin fiber formation [Figure 3.8A,B]. PER977 also restored fibrin formation and fiber diameter upon addition to edoxaban-anticoagulated blood (200 ng/mL, final) [Figure 3.8A,B]. This result is consistent with the whole-blood clotting data of Ansell et al [153]. However, neutralization of edoxaban-anticoagulated blood using 200 µg/mL of PER977 results in fibrin fibers that are significantly thicker than those observed in the buffer control; this was not observed with AnXa.  118     Figure 3.8: Antidotes normalize impaired fibrin formation and fiber diameter of edoxaban treated clots. Blood clots were generated by incubating antidotes in non-citrated whole blood containing 200 ng/mL of edoxaban at 370C.  (A) Clots formed in the presence of edoxaban (200 ng/mL, final) have fewer fibers compared to buffer control. However, normal clot signatures can be observed in 119   edoxaban clots, treated with 100 and 200 µg/mL of Andexanet Alfa or PER977, respectively. Clot images were taken at two magnifications 2500X and 5000X, respectively. Only images from the 5000X magnification are depicted. (B) Blood clot fiber diameters in edoxaban-treated clots formed in the absence or presence of an antidote. Fiber diameters are measured from scanning electron micrographs using ImageJ software. A total of 60 fibers were analyzed from two independent experiments. Fibers for size analysis were selected from four images by probing four different spots in each image. Data are mean ± SE (n=60). Statistical significance for fiber diameter was determined by comparing the antidote treated group to the buffer control using a one-way ANOVA followed by a Dunnett post hoc test. A significant reduction in fiber diameter is observed in edoxaban containing clots compared to the buffer control.  The diameter of fibrin fibers formed in edoxaban clots containing 100 and 200 µg/mL of Andexanet Alfa is comparable to that of the buffer control. Similar effect can be observed with 100 µg/mL of PER977. However, fiber diameters formed in the presence of 200 µg/mL of PER977 are significantly larger compared to the buffer control clot (****p <0.0001).  3.4.5 Clotting assays define anticoagulant reversal activity of antidotes Our ITC data show that UHRA, AnXa and PER977 have two common binding partners: UFH and enoxaparin. Indeed, although AnXa and PER977 are principally designed to neutralize DOACs, activity against indirect FXa inhibitors has been indicated for both [160,250]. To determine if binding of each of these antidotes to UFH and enoxaparin correlates with anticoagulant neutralization, we performed plasma and whole blood clotting assays. UHRA, which is specifically designed to neutralize heparin-based anticoagulants, was included in this study as a positive control.   The aPTT assay results reported in Figures 3.9A and 3.9B indicate that AnXa completely neutralizes the activity of UFH or enoxaparin in human plasma (without citrate or EDTA) over a wide range of antidote concentrations (25-200 µg/mL); as expected, UHRA does as well. However, neutralization of these heparins by PER977 (25-200 µg/mL) was not observed by aPTT assay. A dynamic turbidimetric plasma 120   clotting assay was therefore performed as an orthogonal test of the heparin neutralization activity of each antidote at 200 µg/mL. Results [Figure 3.9C] were consistent with those of the aPTT assay; no heparin neutralization activity was observed with PER977 (i.e., no change in optical density indicative of clot development was recorded).  121   Figure 3.9: Reversal of UFH and enoxaparin anticoagulation activity. (A, B) aPTT assay was performed in heparinized, non-citrated human PPP. Andexanet Alfa and UHRA neutralized UFH and enoxaparin anticoagulation activity. However, PER977 at all tested concentrations showed no anticoagulation reversal activity. (C) Microplate turbidimetric clotting assay was performed in 50% diluted, heparinized and non-citrated human PPP. Neutralization of UFH activity was verified by measuring changes in absorbance at 405nm. Neutralization of UFH and subsequent clot generation increases absorbance. Clot formation was observed in heparinized plasma containing 200 µg/mL of UHRA or Andexanet Alfa, but not with PER977. UHRA and Andexanet Alfa can reverse the anticoagulation activity of UFH. Results are expressed as the mean ± SE of six measurements from two independent experiments (n=2). Given these results, we performed TEG in non-citrated whole blood to evaluate the efficacy of AnXa or PER977 to reverse the anticoagulant activity of a direct FXa inhibitor, edoxaban. The resulting TEG data show that 100 µg/mL of AnXa normalizes the prolonged clotting time induced by edox [Figure 3.10A,B]. In contrast, we observed no edoxaban reversal activity with PER977 (100 µg/mL). 122    Figure 3.10: Reversal of the anticoagulation activity of edoxaban. TEG assay was performed in non-citrated human blood. Andexanet Alfa neutralized the anticoagulation activity of edoxaban. However, PER977 did not neutralize the anticoagulation activity of edoxaban. (A) A representative TEG profile showing neutralization of edoxaban anticoagulation activity by antidotes. (B). 100 µg/mL of Andexanet Alfa, normalized the augmented clotting time induced by edoxaban.  However, 100 µg/mL of PER977 did not show edoxaban reversal activity. No significant change was observed in the clot strength (maximum amplitude, mm) between drug treated group and the buffer control. Data are mean ± SE (n=5). Unpaired 2-tailed t tests were performed to determine significance, with p < 0.05 indicating a statistically significant change. (**p <0.01).  3.5 Discussion Worldwide, millions of patients receive anticoagulant therapy for the treatment and prevention of thromboembolic disorders [251]. Despite the introduction of new oral 123   anticoagulants (direct FXa and FIIa inhibitors) with excellent dose response and fewer drug-drug interactions, anticoagulant-related bleeding remains a serious concern [230]. To address this limitation and make anticoagulant therapy safer, improved reversal strategies with effective antidotes that specifically target anticoagulants and/or anticoagulant activity are needed. Currently, clinicians rely on protamine sulfate to neutralize the activity of UFH, and the complications associated with its administration are well documented [227]. Moreover, no effective antidotes to the LMWHs (e.g. enoxaparin) or fondaparinux are approved for clinical use, and no antidotes to direct FXa inhibitors are commercially available at the present time. For each of the more general antidotes under development (AnXa, PER977, and UHRA), binding affinities to their putative targets have not been fully defined or compared, creating knowledge gaps that limit understanding of common and/or unique mechanisms of action. The aim of this study was therefore to provide further insight into the mechanisms of the anticoagulant reversal activities of UHRA, AnXa and PER977 by probing the binding affinities of these antidotes to their presumed targets. Neutralization activities and effects on clot structure in the absence or presence of anticoagulants were also investigated using a panel of orthogonal assays to connect the binding thermodynamics data to function.  UHRA is specifically designed to neutralize the activity of heparin-based anticoagulants [165], and the ITC data reported here show that UHRA directly binds heparin-based anticoagulants with sub-µM affinity. Moreover, they show that UHRA does not bind relevant coagulation factors and proteins central to clotting, including FXa, AT, thrombin or fibrinogen. As intended, the methoxy-capped polyethylene glycol 124   exterior of the UHRA molecule therefore effectively prevents non-specific interactions with clotting proteins. These results collectively suggest that the previously demonstrated ability of UHRA to reverse the activity of all clinically available heparin-based anticoagulants is established through its selective affinity for heparin, which enables UHRA to disrupt heparin-activated AT to form a UHRA: heparin complex [165]. Our ITC data further show that AnXa has exceptional affinity (nM) for the direct FXa inhibitors rivaroxaban and edoxaban, binding each with 1:1 stoichiometry. That tight binding correlates with reversal of edoxaban anticoagulant activity measured both by TEG and plate-based coagulation assays. Reversal of anticoagulant activity was also indicated by restoration of impaired clot fiber formation and clot morphology as measured by SEM analysis. Components of clot structure obtained after neutralization of edoxaban with AnXa were comparable to buffer control clots and suggested complete recovery of normal hemostatic mechanisms. Collectively, the data therefore support high-affinity, direct stoichiometric binding as the mechanism by which AnXa neutralizes each of these DOACs.  The ITC data also record a Kd in the order of 10-7 M for binding of AnXa to heparin-activated AT. This affinity is ~50-fold tighter than the corresponding AnXa: UFH complex, suggesting that AnXa reverses the activity of indirect FXa inhibitors primarily by binding heparin-activated AT and thereby inhibiting the ability of that complex to interact with FXa [252]; direct sequestration of heparin possibly contributes to some extent as well. Orthogonal clotting (aPTT) assays and clot morphology analyses by SEM confirm that neutralization activity.  125   AnXa shows no binding to relevant coagulation factors [Table 3.1], except to tissue factor pathway inhibitor (TFPI) [252]. Moreover, AnXa does not adversely affect fibrin polymerization or alter fibrin and whole-blood clot structures except at supra-therapeutic concentrations. At the one supra-therapeutic concentration studied (500 µg/mL), AnXa shows only subtle effects on fibrin polymerization (reduced optical density for fibrin clots) and overall clot morphology.  Though previous literature cites it as a putative mechanism of action [156], direct binding of PER977 to rivaroxaban or edoxaban was not observed by ITC, consistent with previous observations that PER977 does not reverse the anti-FXa activity of rivaroxaban or edoxaban in plasma or buffer systems [247]. However, we observed that clot structure and fibrin fiber diameter were normalized to that of the control when edox-treated blood was treated with PER977, a result consistent with previously reported whole-blood clotting time data [153]. The exception was neutralization with 200 µg/mL PER977 (400 µM or ~10X therapeutic dose), which resulted in significantly thicker fibrin fibers than those within the control clots (in analyses performed with utmost care to avoid measuring the diameter of intertwined fibers). When applied to whole blood containing no anticoagulants, PER977 did not affect fibrin polymerization or alter clot structure at any concentrations studied. Thus, if one takes restoration of impaired clot fiber diameter as a metric of activity, PER977 is found to reverse anticoagulant activity induced by edoxaban.  Given that direct binding of edoxaban by PER977 was not seen by ITC, the mechanism underpinning this observed restoration of impaired clot fiber diameter is unclear, but may be related to the measured affinity of PER977 for FIXa 126   [Table 3.1], as we did not observe an interaction between PER977 and other relevant coagulation proteins.  PER977 showed near-µM affinity for UFH or enoxaparin, with binding affinity to enoxaparin weakened when it was pre-complexed with AT. The observed binding to enoxaparin aligns with data indicating capture of PER977 on heparin affinity chromatography columns [250]. Once again, however, the ability of PER977 to directly bind these anticoagulants did not correlate with standard measures of reversal activity, as neutralization of UFH or enoxaparin by PER977 was not observed in either TEG (in non-citrated blood) or clotting assays. Thus, unlike for UHRA and AnXa, a correlation between binding partners and reversal function could not be established for PER977, raising questions with regard to its mode of action depicted in a recent report indicating neutralization of enoxaparin by PER977 based on whole blood clotting time data [250].  Our data, particularly for PER977, provide several insights. First and foremost is the finding that a single metric, such as whole-blood clotting time or fibrin fiber diameter measurements [153], though informative, may not on its own provide a definitive indication of anticoagulant neutralization activity. Though PER977 displays evidence of neutralization based on whole-blood clotting time and formation of clot-like structures, reversal of edoxaban-mediated anticoagulant activity is not detected in either TEG or plate-based clotting assays. A panel of orthogonal tests of anticoagulant reversal activity, including biologic function assessments provided by aPTT, prothrombin time/international normalized ratio, thrombin time, and/or anti-FXa activity [163,239], may therefore be required (or at least advised) to prove therapeutic efficacy by mitigating uncertainties in results obtained from any given assay. For example, while 127   electron micrographs of clots can provide in-depth understanding about the impact of an anticoagulant or antidote on clot components and structure, they provide no direct information on clot formation kinetics or strengths. Conversely, the microplate-based clotting assays and TEG data most often used to characterize neutralization activity do not ensure that drug use results in desired clot morphologies.  Moreover, a degree of uncertainty is associated with each of these assays.  For example, clotting endpoints in whole-blood clotting time, aPTT, TEG and microplate clotting assays can show inter-individual variation depending on several factors, including abundances of coagulation factors within the assayed plasma. Second, for drugs designed as anticoagulant neutralizing agents, verification of specific and selective binding to the intended target is critical, and likely should be included in New Drug Application (NDA) submissions to the FDA. The FDA should therefore work to include in their NDA guidance documents recommendations for a sufficient set of orthogonal data/metrics (and associated sample sizes) for any anticoagulant neutralizing agent (synthetic or biologic) to show that the drug is safe and effective in its proposed use(s), and that the benefits of the drug outweigh the risks. Evidence supporting this position is clearly provided in this study, which reveals that each of the three antidotes studied exhibits evidence of neutralization activity against different types of anticoagulants based on different mechanisms of action. However, and more importantly, for PER977 we observe evidence of anticoagulant neutralization activity in some standard metrics but not others, indicating that no single metric on its own is certain to provide an infallible measure of drug efficacy and safety. 3 A version of chapter 4 has been published. Kalathottukaren MT, Abraham L, Kapopara PR, et.al. Alteration of blood clotting and lung damage by protamine are avoided using the heparin and polyphosphate inhibitor, UHRA (Blood, 2017). (DOI: 10.1182/blood-2016-10-747915)                      128 Chapter 4: Alteration of blood clotting and lung damage by protamine are avoided using the heparin and polyphosphate inhibitor, UHRA 4.1 Synopsis Anticoagulant therapy-associated bleeding and pathological thrombosis pose serious risks to hospitalized patients. Both complications could be mitigated by developing new therapeutics that safely neutralize anticoagulant activity and inhibit activators of the intrinsic blood clotting pathway, such as polyphosphate (polyP) and extracellular nucleic acids. The latter strategy could reduce the use of traditional anticoagulants and potentially decrease bleeding events. However, previously described cationic inhibitors of polyP and extracellular nucleic acids, exhibits both non-specific binding and adverse effects on blood clotting that has limited their use. Indeed, the polycation, protamine used to counteract heparin-associated bleeding in surgical settings, exhibit adverse side-effects. To address these clinical shortcomings, we developed a synthetic polycation, Universal Heparin Reversal Agent (UHRA), which is nontoxic and can neutralize the anticoagulant activity of heparins and the prothrombotic activity of polyP. Sharply contrasting protamine, we show that UHRA does not interact with fibrinogen, affect fibrin polymerization during clot formation or abrogate plasma clotting. Using scanning electron microscopy, confocal microscopy and clot lysis assays, we confirm that UHRA does not incorporate into clots, and clots are stable with normal fibrin morphology. Conversely, protamine binds to the fibrin clot, which could explain how protamine instigates clot lysis and increases bleeding after surgery. Finally, studies in mice reveal that UHRA reverses heparin anticoagulant activity without the lung injury 129  seen with protamine. The data presented here illustrate that UHRA could be safely used as an antidote during adverse therapeutic modulation of hemostasis.            4.2 Background Heparins are highly sulphated anionic polymers that are commonly used as prophylactic anticoagulants, and to treat acute thromboembolism [251]. However, the risk of bleeding associated with heparin therapy remains a serious concern, and no approved antidotes are available for low molecular weight natural or synthetic (fondaparinux) formulations of heparin [99]. Hence, effective strategies to manage bleeding associated with heparin therapy remain a clinical need.  In the past decade, investigations have revealed that other polyanionic biomolecules can participate directly or indirectly in the development of thrombosis. In particular, extracellular nucleic acids, polyphosphate (polyP) and neutrophil extracellular trap-associated DNA are potent activators of blood clotting. Neutralizing these prothrombotic polyanions could therefore be used to treat thrombosis [253-255,92]. Polycations are used widely to counteract polyanions. For example, protamine sulphate is the only clinically approved antidote for unfractionated heparin (UFH) [99]. Adverse outcomes to protamine therapy are well-documented and include intrinsic anticoagulant effects, precipitation, nonenzymatic polymerization of fibrinogen, and complement activation [135,172,178,179]. Likewise, cationic polyethyleneimine (PEI) and polyamidoamine dendrimers have been tested as polyP and extracellular nucleic acids inhibitors [167,255]. Both molecules exhibit cellular toxicity and are not hemocompatible [169,256].   130  The development of new hemocompatible polycations that effectively regulate pro- or anticoagulant effects of polyanions would therefore be of significant clinical value. We have therefore developed a new class of molecules collectively named Universal Heparin Reversal Agents (UHRAs) [165]. UHRAs have a unique design comprised of a dendritic core decorated with multivalent cationic heparin binding groups (HBGs) shielded within a neutral brush layer of methoxy polyethylene glycol (mPEG) emanating from the core.  The general nature of the synthesis platform allows facile creation of libraries of UHRA molecules that differ in their molecular weight and number of HBGs on the scaffold. Through extensive calorimetric screening of binding affinities and stoichiometries among library members, we identified UHRA molecules that bind and inhibit heparins and polyP, respectively, in animal models of bleeding and thrombosis [165,168].  Here, we extend our earlier discovery by reporting the impact on plasma coagulation, clot stability, morphology and organ toxicity of a specific UHRA molecule having high binding affinity to heparin and polyP. Previously tested polycations against heparin and polyP, exhibit nonspecific binding to blood proteins, and are known to cause alterations in clot morphology and augmented clot dissolution [257,258]. Our ex vivo and in vivo studies reported here therefore address the following questions: (1) What is the influence of UHRA on fibrinogen, fibrin polymerization and thrombin generation? (2) Are fibrin or blood clots that have been formed in the presence of UHRA stable and do they exhibit normal clot morphology? (3) Does UHRA bind to and/or incorporate into clots? (4) And finally, does UHRA reverse the anticoagulation activity of UFH without lung injury?   131  The results demonstrate that, in contrast to protamine, UHRA has negligible influence on fibrinogen, blood clotting and clot morphology. This is supported by our finding that UHRA, unlike protamine, is excluded from fibrin clots and does not bind to blood clot components. Also, UHRA is capable of completely reversing the anticoagulant activity of UFH in mice without causing lung injury. Our results therefore indicate that UHRA is a suitable candidate for further clinical development as an inhibitor of polyanionic pro- and anticoagulant molecules. 4.3 Methods Reagents and proteins  Human fibrinogen (Fibrinogen), polyethyleneimine (PEI), thrombin, protamine sulphate and N-2-hydroxyethyl piperazine-N’-2-ethanesulfonic acid (HEPES) were from Sigma-Aldrich. Recombinant tissue factor (TF, Innovin) was from Dade-Behring. Citrated, pooled normal human plasma was from Affinity Biologicals. Recombinant human tissue plasminogen activator (t-PA) was from Aniara. Polystyrene 96-well microplates (Costar) used for all clotting assays were from Corning. Ethics statement  The protocol for blood collection was approved by the Clinical Research Ethics Board (CREB; certificate number H10-01896) of the University of British Columbia, and written consent was obtained from donors in accordance with the Declaration of Helsinki.  Blood collection and plasma preparation Blood specimens from healthy consenting donors were collected by venipuncture into BD vacutainer glass tubes containing 0.105 M trisodium citrate. Blood was centrifuged at 150 x g for 10 minutes to separate platelet-rich plasma (PRP), and then spun at 1000 x g for 15 minutes for platelet-poor plasma (PPP) at 22°C. 132  Synthesis of hyperbranched polyglycerol-methoxy polyethylene glycol-25kDa (HPG-mPEG-25kDa) precursor polymer  Chemicals and reagents were from Sigma-Aldrich (ON) and used without further purification unless specified. Glycidol was purified by distillation under reduced pressure before use and stored over molecular sieves at 4oC.  In the first step, a three-necked round bottomed flask was cooled under vacuum and filled with argon. To this, 1,1,1-tris(hydroxymethyl)propane (TMP, 0.240 g) and potassium methylate (25 wt % solution in methanol, 0.220 mL) were added and stirred for 30 minutes. Methanol was removed under vacuum for 4 hours. The flask was heated to 95°C and glycidol (6.0mL) was added over a period of 15 hours. After the complete addition of glycidol, the reaction mixture was stirred for additional 3 hours. mPEG-epoxide 350 (12.5 mL) was added over a period of 12 hours. The reaction mixture was stirred for additional 4 hours. The polymer was dissolved in methanol, and precipitated twice from diethyl ether. The resulting polymer was dissolved in water and dialyzed against water using MWCO-1000 membrane for 3 days with periodic changes in water.   Synthesis of UHRA The HPG-mPEG-25kDa precursor polymer (1.0 g) was then dissolved in 5 mL of pyridine. To this, p-toluenesulfonyl chloride (0.230 g) was added and stirred at room temperature for 24 hours. Pyridine was removed by rotary evaporation; the polymer was dissolved in 0.1 N HCl and dialyzed overnight. The HPG-mPEG-tosylate was isolated by freeze drying. The HPG-mPEG-tosylate (1.2 g) and tris (2-aminoethylamine) (4 mL) were dissolved in 1,4-dioxane (10 mL) and refluxed for 24 hours. Dioxane was removed under vacuum and the polymer was dissolved in a minimum amount of methanol and precipitated twice from diethyl ether. The polymer was then dissolved in water and 133  dialyzed against water using MWCO-1000 membrane for 2 days. The resulting polymer solution was added to a mixture of formaldehyde (3 mL) and formic acid (3 mL) at 0°C. The reaction mixture was refluxed overnight. After cooling to room temperature, the pH of the solution was adjusted to 10 using NaOH and the polymer was extracted with dichloromethane. Dichloromethane was removed under vacuum; the polymer dissolved in distilled water and dialyzed against water using MWCO-1000 membrane with frequent changes in water for 2 days.  Characterization of UHRA 1H Nuclear magnetic resonance spectra were acquired in deuterium oxide on a Bruker Avance AV-300 spectrometer. Absolute molecular weights and polydispersity of the polymer were determined by Gel Permeation Chromatography (GPC) with a Waters 2695 separation module fitted with different Ultrahydrogel columns (guard, linear and 120), a DAWN HELEOS II multiangle laser light scattering (MALLS) detector and an Optilab T-rEX refractive index detector, all from Wyatt Technology.  In all cases the mobile phase was 0.1 M NaNO3 (10 mM phosphate buffer).  1H NMR (CDCl3, 300 MHz): δ ppm 3.38 (-O-CH3 from mPEG), 3.4-3.95 (main chain protons from HPG and mPEG), 2.4-2.7 (-N-CH2) and 2.27 (-N-CH3) [Appendix C.1]. The number of cationic heparin binding groups (HBGs) in UHRA was calculated to be 25 based on the conductometric titration analysis. Synthesis of Alexa Fluor 488 labelled UHRA and protamine sulphate  UHRA (57.5 mg, 2.3 µmol) was dissolved in anhydrous DMF. To this solution, sodium hydride (~1 mg, 42 µmol) was added and stirred at room temperature for 4 hours. After cooling the reaction mixture to room temperature, (R)-N-glycidylpthalimide (5.1 mg, 0.25 134  µmol) was added and the mixture was stirred at 65°C for 24 h.  The DMF was removed under reduced pressure. The polymer was then dissolved in methanol and precipitated from diethyl ether. The polymer was dissolved in methanol and refluxed at 60°C for 48 h with hydrazine hydrate solution (1.1 mmol) to generate primary amine groups on the UHRA. The methanol was removed by rotary evaporation and the resultant solution was dialyzed against deionized water using 1000 Da MWCO dialysis membrane for 12 h with frequent changes in water. The polymer was recovered by lyophilization of the dialyzate. The polymer was then dissolved in anhydrous DMF and Alexa Fluor 488 succnimidyl (NHS) ester (1.5 µmol) was added and stirred for 6 h. The reaction mixture was then dialyzed against deionized water using 3.5kDa MWCO dialysis membrane for 8 h to remove free Alexa Flour 488.  The concentration of the final Alexa Fluor 488 labelled UHRA solution was determined by thermogravimetric analysis.  Solution was stored at 4°C until further use. Formation of UHRA-Alexa conjugate was confirmed by fluorescence spectroscopy [Appendix C.2A] after passing the mixture through Sephadex G25 column.   Protamine with an average molecular weight 5kDa (Sigma-Aldrich) was labelled using the procedure described in the Alexa Fluor 488 protein labelling kit (Molecular probes). Sodium bicarbonate (0.1 M) buffer with 1 M NaCl at pH 8.3 was used as reaction medium.  Alexa Fluor-488 conjugated protamine was purified by passing the reaction mixture through Sephadex G25 column [Appendix C.2B-C]. The final concentration of F-Protamine was determined by Pierce TM BCA protein assay kit. Solution was stored at 4°C until further use.  135  The heparin neutralizing activity of Alexa Fluor 488 conjugated UHRA and protamine was determined by activated partial thromboplastin time (aPTT) assay [Appendix C.2D].  Interaction of UHRA with fibrinogen Fluorometric experiments were performed on a Cary Eclipse Fluorescence Spectrophotometer (Agilent). Human fibrinogen (Fibrinogen), UHRA and polyethyleneimine (PEI) solutions were prepared in 10 mM PBS buffer (137 mM NaCl, pH 7.4). Fibrinogen solution (0.25 µM) was incubated with increasing concentrations of UHRA or PEI (25kDa) at 37C for 5 minutes. The intrinsic fluorescence of fibrinogen (at λex = 280 nm, λem from 300 to 400 nm) was recorded in a 0.1 cm quartz cuvette with the absorbance of all solutions at λ280 nm kept below 0.1. The excitation and emission slit width (each 5 nm) and system temperature (37C) was fixed for all experiments. To characterize secondary structures of fibrinogen in the presence of UHRA or PEI, circular dichroism (CD) analyses were performed using a J-810 CD spectrophotometer and a 1 mm path length cuvette. Before scans from 190 to 260 nm were conducted, fibrinogen in 20 mM sodium phosphate (pH 7.4) was incubated with varying concentrations of UHRA or PEI for 5 minutes at 37°C. Data was collected every 1 nm at 100 nm/min and then averaged over three scans. The final concentration of fibrinogen used was determined by the PierceTM BCA protein assay and adjusted to 0.4 µM for all experiments. All concentrations reported are final, and changes in fibrinogen secondary structure were analyzed using CDNN deconvolution software.  Finally, the interaction of UHRA with fibrinogen, UFH and polyP was analyzed by isothermal titration calorimetry (ITC).  136  Isothermal titration calorimetry (ITC) The interaction of UHRA with UFH and PolyP (PolyP75) was carried out at 25.0°C using a MicroCal iTC200 calorimeter (MicroCal, Northampton, USA). PolyP, UFH and UHRA solutions used for titrations were prepared in 10 mM phosphate buffered saline (0.137 M NaCl, pH 7.4). Samples were filtered using 0.2 μm filters (Millex, Merck Millipore Ltd) and degassed prior to addition. Titration of UHRA (300 μM) to polyP (5 μM, estimated polyP concentration based on average polyP size of 75 phosphates), consisted of 22 consecutive injections of 1 μL volume and 5 sec duration each, with a 3 min interval between injections. Titration of UHRA (400 μM) to UFH (25 μM) consisted of 25 consecutive injections of 1 μl volume and 5 sec duration each, with a 3 min interval between injections. Heat of dilution of UHRA was measured by injecting UHRA solution into PBS buffer alone and was subtracted from the experimental curves prior to data analysis. The resulting data were fit into a single set of identical sites model using the MicroCal ORIGIN software supplied with the instrument.  The interaction of UHRA with human fibrinogen was performed using a VP-ITC microcalorimeter from Microcal, Inc. (Northampton, MA) with a cell volume of 1.4 mL at 298K. Human fibrinogen and UHRA used for titrations were prepared in 50 mM sodium phosphate buffer (0.150 M NaCl, pH 7.4). Samples were filtered using 0.2 μm filters (Millex, Merck Millipore Ltd) and degassed prior to addition. Injections of 10 μL of UHRA (20 µM) solution were performed from a computer controlled micro syringe at an interval of 5 minutes into 2.5 µM fibrinogen solution in the cell. The heats of dilution from titrations of UHRA solution into buffer only (without fibrinogen) were subtracted from the heats obtained from titrations of UHRA solution into fibrinogen solution to obtain net 137  binding heats. All the experiments were carried out in duplicate. Raw ITC data of UHRA binding to fibrinogen was analyzed with Origin software from Microcal, Inc. (Northampton, MA).  The resulting data were fit into a single set of identical sites model using the MicroCal ORIGIN software supplied with the instrument.  Fibrin polymerization assay  Fibrinogen (150 µL, 3 mg/mL, initial concentration) in 20 mM N-2-hydroxyethylpiperazine-N’-2-ethanesulfonic acid (HEPES) buffer (pH 7.4, 150 mM NaCl) supplemented with 20 µL of CaCl2 (3 mM, final) was incubated with 10 µL of UHRA or protamine (dilution 1:20, final) for 10 minutes. Clotting was then initiated by adding 20 µL of thrombin (2.5 NIHU/mL, final). Fibrin formation was monitored by recording changes in turbidity (A405nm) every 30 seconds on a Spectramax microplate reader (MOLECULAR DEVICES) for 1 hour at 37°C. The final turbidity of fibrin clots was also recorded.  Plasma clot formation and lysis by turbidity analysis Microplate turbidimetric clotting assays were performed with platelet-poor plasma (PPP) obtained from three donors. Protamine and UHRA solutions were prepared in 20 mM HEPES (pH 7.4, 150 mM NaCl) buffer. Clotting was initiated in 90 µL of 30% diluted PPP spiked with protamine or UHRA (dilution 1:10) by adding 5 µL of recombinant tissue factor (TF; Innovin (1:10,000; 0.73 pM)) and 5 µL of CaCl2 (20 mM). Clotting was evaluated by monitoring changes in turbidity (A405nm) every 30 seconds with the Spectramax microplate reader for 2 hours at 37°C. Clotting of PPP by recalcification was performed as described above in the absence of TF. Clotting parameters such as lag time and maximum absorbance (MaxAbs) were calculated. Lag time is considered 138  as the time point when an exponential increase in absorbance was first observed. MaxAbs is the absorbance at which at least 5 readings were identical (plateau phase, corrected for the lag time).  The lysis of TF–induced plasma clots formed in the presence of UHRA or protamine exposed to exogenous recombinant human tissue plasminogen activator (t-PA) was monitored by a microplate turbidimetric assay. Control pooled plasma (30% diluted, 85 µL) spiked with UHRA or protamine (dilution 1:10) was added to microplate well containing 5 µL of TF (Innovin (1:10,000; 0.73 pM)), 5 µL of t-PA (25 ng/mL) and 5 µL of CaCl2 (20 mM). Changes in turbidity (A405nm) at 37°C were monitored every 30 seconds for 30 minutes, and then every minute thereafter up to 300 minutes. Clot lysis half-time (CLT50) is defined as the mid-point of the lysis curve excluding the plateau phase and the clear transition. The area under the clot lysis curve (AUC), a measure of clotting time, clot density and lysis potential, was calculated by applying the trapezoidal rule to transition data after baseline (lowest A405nm value) subtraction. All dilutions and concentrations mentioned are final.  Scanning electron microscopy (SEM) of fibrin and whole blood clots  The morphology of clots formed in the presence of UHRA or protamine was determined by SEM [237]. All samples were randomly coded and blinded to the individual performing imaging analyses to avoid bias. Fibrin clots were prepared in sterile, 5 mL round-bottom polypropylene tubes (BD Falcon) by mixing 200 µL of fibrinogen (3 mg/mL, initial concentration) in 20 mM HEPES (pH 7.4 and 150 mM NaCl) buffer with 2.5 NIHU/mL thrombin (final), 3 mM CaCl2 (final) and UHRA or protamine (in HEPES buffer). Control clots were prepared without UHRA or protamine. After incubation for 1 139  hour at 37°C, clots were fixed using Karnovsky fixative (2.5% glutaraldehyde and 4% formaldehyde), repeatedly washed with 0.1 M sodium cacodylate buffer (pH=7.4), then subjected to post-fixation with 1% v/v osmium tetroxide. The fixed samples were washed three times with distilled water and dehydrated with a gradient series of ethanol (20-95% v/v). Clots were then dried with CO2 in a Tousimis Autosamdri 815B critical-point dryer, mounted onto stubs, and gold sputter-coated for SEM using a Hitachi S-4700 field emission scanning electron microscope at 5,000X, 10,000X and 25,000X magnifications. Multiple images from different areas of each clot were captured. Fiber diameters of clots were measured with ImageJ (National Institutes of Health, USA) [236]. Fibrin fiber diameters (n= 30 fibers) from 8 separate areas of two separate clots (prepared in a blinded fashion) were used to calculate the mean fiber diameter.  Blood clots containing UHRA or protamine were also prepared in 5 mL polypropylene tubes (BD Falcon) by recalcifying 180 µL of citrated blood with 20 µL of CaCl2 (11.1 mM, final) in the presence of UHRA or protamine in 20 mM HEPES buffer at 37°C.    For blood clots containing heparin and antidotes, 20 µL UHRA or protamine was added to 180 µL of heparinized blood (2 IU/mL, final). Clotting was initiated in 180 µL of this mixture by adding 20 µL of CaCl2 (11.1 mM, final) at 37°C. All the clots were prepared in sterile 5mL round-bottomed polypropylene tubes (BD Falcon). After one hour incubation at 37°C, clots were carefully processed for SEM. Confocal microscopy of fibrin and blood clots  Fibrin clots for confocal microscopy were prepared as previously described [259]. Briefly, Fibrin clots were prepared at 37°C by adding 2.5 NIHU/mL thrombin (final) and 3 mM CaCl2 (final) to 200 µL of human fibrinogen (3 mg/mL, initial concentration) mixed 140  with Alexa Fluor 546-conjugated fibrinogen (Invitrogen, 2% total in 20 mM HEPES (pH 7.4 and 150 mM NaCl)) in the presence of fluorescently labelled UHRA or protamine (in HEPES buffer). Whole blood clots were prepared at 37°C by recalcifying 180 µL of whole blood with 20 µL of CaCl2 (11.1 mM, final) in the presence of fluorescently labelled UHRA or protamine in 20 mM HEPES buffer (pH 7.4 and 150 mM NaCl). Control clots were prepared in the absence of fluorescently labelled UHRA or protamine. All the clots were prepared in sterile 5mL round-bottomed polypropylene tubes (BD Falcon). After one hour incubation at 37°C, clots were then carefully transferred into 14 mm microwell dishes (MatTek Corp.) and washed three times with 20 mM HEPES buffer (pH 7.4 and 150 mM NaCl).              For confocal microscopy of blood clots containing heparin and antidotes, 20 µL of fluorescently labelled UHRA or protamine was added to 180 µL of heparinized blood (2 IU/mL, final). Clotting was initiated in 180 µL of this mixture by adding 20 µL of CaCl2 (11.1 mM, final) at 37°C. All the clots were prepared in sterile 5mL round-bottomed polypropylene tubes (BD Falcon). After one hour incubation at 37°C, clots were carefully transferred into 14 mm microwell dishes (MatTek Corp.) and washed three times with 20 mM HEPES buffer (pH 7.4 and 150 mM NaCl). Clots were then treated with CellMask™ deep red plasma membrane stain (C10046, ThermoFisher) at 1:1000 dilution (final) for 30 minutes. Clots were washed three times with 20 mM HEPES buffer (pH 7.4 and 150 mM NaCl) before imaging.           Confocal micrographs were acquired with a spinning disk confocal system (3i – Intelligent Imaging Innovations) based on an inverted Zeiss Axiovert 200M microscope equipped with 100 NA 1.45 Oil Plan-Fluor objective and a QuantEM 512SC 141  Photometrics camera. All images were acquired using identical acquisition settings. For quantitative image analysis, the raw integrated density (sum of all pixel intensities) in green channel was determined using FIJI. The values were normalized to mean values obtained from polycation free buffer, in order to obtain normalized raw integrated density. Thrombin generation and amidolytic activity assay  Thrombin generation assay was carried out in human platelet rich plasma (PRP) at 37°C by measuring the fluorescence intensity upon cleavage of a fluorogenic substrate, Z-Gly-Gly-Arg-AMC by the generated thrombin. For every experiment, fresh PRP collected from two different donors of same blood type were pooled together. Stock solutions of UHRA were prepared at concentrations 1 and 2 mg/mL. For the study with UHRA alone, 40 µL of UHRA (final concentration of 0.1 and 0.2 mg/mL, respectively) was incubated with 360 µL of the pooled PRP at 37°C for 20 min. Eighty microliters of the UHRA-PRP mixture in triplicates were added to a 96 well plate. Twenty microliters of recombinant tissue factor (TF, 0.5 pM final concentration) from Innovin was then added to initiate clotting. PRP incubated with thrombin-α2-macroglobulin was used as a thrombin calibrator. The thrombin generation assay was initiated by the addition of 20 µL of fluorogenic substrate (final concentration 0.416 mM) and CaCl2 solution (final concentration 17 mM) to each well. The fluorescence intensity was recorded at 37°C every 30 sec over a period of 1.5 h on a POLARstar Optima spectrofluorometer at excitation and emission wavelengths of 390 nm and 460 nm respectively. Two important parameters such as endogenous thrombin potential (ETP) and peak amount of thrombin generated were measured.  142  Thrombin amidolytic activity assays were performed in 96-well plates by mixing 10 µL of UHRA in 10 mM PBS buffer (pH 7.4) with 90 µL of 0.5 NIHU/mL thrombin. The plate was then incubated at 37 °C for 30 min, and the reaction was initiated by adding 50 µL of diluted Chromozym TH (50 µM), which is a colorimetric pseudo-substrate that releases p-nitroanilide (pNA) upon cleavage. Changes in absorbance were monitored at 405 nm for 5 min, sampling at 8 seconds intervals to quantitate thrombin activity based on the initial slopes of these kinetic profiles.  Thromboelastography The whole blood clotting time in the presence of UHRA or protamine was analyzed using a Thromboelastograph Hemostasis System 5000 (TEG) from Haemoscope Corporation at 37C. Citrated human whole blood (360 µL) was mixed with UHRA or protamine (40 µL) to get the desired concentration. 360 µL of this blood mixture was transferred into a TEG cup containing 11.1 mM CaCl2 (final) and the clotting was monitored. HEPES-buffered saline was used as the control.  Activated partial thromboplastin time assay (aPTT) in human PPP  UHRA or protamine solutions were prepared in 20 mM HEPES buffer (pH 7.4 and 150 mM NaCl). Heparinized human PPP was prepared by mixing UFH solution and citrated human PPP. The final concentration of UFH in PPP was 4 IU/mL. The anticoagulant neutralization activity was examined by mixing 20 μL of UHRA or protamine solution with 180 µL of heparinized plasma (1:10 v/v). The final concentration of antidotes in plasma ranged from 12.5 to 250 µg/mL. 200 µL of aPTT reagent (Dade® Actin® FS Activated PTT, Siemens/Dade-Behring) was then added to the sample and 100 µL of this resulting mixture was transferred to cuvette-strips at 37C. The clotting time was 143  measured on a STart®4 coagulometer (Diagnostica Stago, France).  20 mM HEPES buffer (pH 7.4 and 150 mM NaCl) added to heparinized plasma was used as a control for the experiments. All experiments were performed in triplicates and the average values (mean ± standard error of the mean) are reported. Similar aPTT protocol was used for analyzing neutralization of 1 IU tinzaparin anticoagulation activity by UHRA or protamine. Neutralization of UFH activity with UHRA and evaluation of lung injury in mice  Female Balb/C mice (18-20 weeks, 25-30 g) were individually weighed and were divided into 4 treatment groups. Mice were anesthetized using isoflurane-oxygen mixture. For multiple intravenous administration (via tail-vein), mice were cannulated by inserting a tail-vein catheter (C10SS‐MTV1301, Instech Laboratories, Inc., PA USA). Buffer control group (N=3) received phosphate-buffered saline (PBS, pH =7.4) only. UFH control group (N=3) received 200 IU/kg of UFH followed by PBS. UHRA treatment group (N=7) and protamine treatment group (N=7) received 200 IU/kg UFH followed by UHRA 10 mg/kg or protamine 5 mg/kg, respectively. Prior to the administration of antidotes, five minutes was allowed for distribution of UFH in the circulation. Antidotes were injected slowly in a constant rate using a programmable syringe pump. After 45 minutes, mice were euthanized by overdose of isoflurane (5 %). Blood was collected by cardiac puncture into eppendorf tubes containing sodium citrate solution (0.105 M, final).  aPTT in mouse plasma  Sodium citrate anticoagulated blood was spun at 10,000 rpm for 10 minutes to prepare heparinized platelet-poor plasma. 150 µL of plasma is mixed with equal volume of aPTT 144  reagent (Dade® Actin® FS Activated PTT, Siemens/Dade-Behring). 100 uL of this mixture was warmed up in cuvette-strips at 37°C for 3 min. The clotting was initiated by adding 50 µL of 25 mM CaCl2 into each cuvette. Clotting time was measured on a STart®4 coagulometer (Diagnostica Stago, France). All experiments were performed in triplicates and the average values (mean ± SE) are reported. Lung sample preparation and histopathology analysis  Upon termination of the mice, the chest cavity was opened and the diaphragm incised. Lungs were then infused with 10% neutral buffered formalin through the endotracheal tube. After 15 minutes, lungs were removed and placed in 10% neutral buffered formalin solution at 40C until further tissue processing. Left lungs (N=2) from each treatment group were randomly selected for sectioning. Lungs embedded in paraffin wax were then longitudinally sectioned into 5 µm thick slices using microtome, mounted on glass slides and stained with hematoxylin and eosin (H&E). Six slices of each lung was placed on glass slides (two slices per slide). Slices were then examined using bright field light microscope (Zeiss Axioskop 2 Plus, Carl Zeiss Microimaging Inc.). Color images at 20x magnification were captured using the digital microscope camera (AxioCam ICc 1, Carl Zeiss Microimaging Inc.); 18 images per mouse were acquired. Generally, major airways and blood vessels were avoided from images to measure alveolar area. For alveolar area measurements, images were thresholded and analyzed using ImageJ as previously described [260,261]. Data analysis  Data are presented as mean ± standard error (SE) values from at least 3 independent experiments unless otherwise specified. All results were plotted and analyzed using 145  GraphPad Prism 6.0 (La Jolla, CA). Statistical significance was determined using Student’s t test or by one-way ANOVA followed by a Dunnett post hoc test. P values < 0.05 were considered statistically significant. 4.4 Results 4.4.1 Design and synthesis of a UHRA molecule that binds pro- and anticoagulant polyanions Our previous findings showed that the characteristics of a given UHRA molecule, including molecular weight (MW), number of HBGs, and density of the mPEG brush layer, determine its specificity and efficacy against a particular polyanionic target, as well as its toxicity [165,168]. This is expected, as affinity and specificity are dictated in part by a network of coulombic interactions between anions on the target and cationic ligands appropriately spaced on the UHRA to form cognate ion pairs. The mean distance of separation between ligands on a UHRA molecule is therefore a determinant of binding affinity and specificity, and the number of HBGs per MW ratio (#HBGs/MW) provides a metric of that spacing. When UFH is the target polyanion, we showed previously that tight and specific binding can be achieved, while excellent biocompatibility was maintained, with a UHRA molecule (UHRA-7) having a #HBGs/MW ratio of ca. 1.1 [165]. Through those investigations, we further showed that binding properties change with MW at a fixed #HBGs/MW due to the fact that, although their spacing remains similar, the ligands on average reside closer to the surface of the UHRA as MW is decreased. The entropic shielding provided by the brush is therefore reduced. To exploit these properties for drug selection, a library of UHRA molecules of different MW, #HBGs/MW and brush densities was created and screened to identify a UHRA molecule that exhibits good binding affinity to both heparin and polyP. The 146  synthesis of UHRA molecule evaluated in the current study is similar to that for UHRA-7 described previously [14], and details are provided in the Methods section [Table 4.1; Figure 4.1 and Appendix C.1A-B].  Figure 4.1: Structure of the UHRA molecule and cationic binding group (CBGs) represented as R in the structure. The darker shaded area represents the core of the molecule, consisting HPG and cationic binding groups (R). These groups acquire positive charges at physiological pH and facilitate binding of UHRA molecule to negatively charged polyanions such as heparins and polyphosphates. The outer shaded area represents the stealth zone created by mPEG chains, which prevents nonspecific interactions between UHRA and blood proteins. UHRA molecule possesses good structural homogeneity   Molecule Molecular weight (Da) * Polydispersity Index (PDI) * mPEG content (mol %) # Number of HBGsɸ  UHRA  25,000  1.4  28  25 * Gel permeation chromatography in 0.1 M NaNO3. # 1H Nuclear magnetic resonance spectroscopy of UHRA using Bruker Avance AV-300 spectrophotometer.  ɸ Conductometric titrations    Table 4.1: Characteristics of the UHRA molecule 147   ITC data [Figure 4.2] demonstrates stoichiometric, well-defined complexes formed between UHRA and either UFH or polyP.  In either system, the equilibrium binding constant (Ka) approaches ≈ 106 M-1 [Table 4.2]      Figure 4.2: UHRA exhibits high-affinity binding to both UFH and polyP75. (A) Binding isotherm obtained after titrating UHRA and polyP75. (B) Binding isotherm obtained after titrating UHRA and unfractionated heparin (UFH). In both (A) and (B), the upper panels demonstrate the raw data (power signal) and the lower panels show the integrated areas corresponding to each injection, normalized to the moles of UHRA injected into polyanion solution and plotted as a function of molar ratio (UHRA/polyanion). Raw heat data of UHRA titration into polyanion solution (UFH and polyP75) shows strong exothermic peaks and therefore binding is enthalpy driven. The association constants (Ka) show that UHRA binds with high-affinity to both UFH and polyP75. Thermodynamic parameters are shown in the Table 4.2.      148   #Obtained from isothermal titration calorimetry experiments  $Calculated from the equation ΔG = ΔH‐TΔS = ‐RTlnKa  All data were collected in PBS at pH 7.4 and 25 oC. Values given represent an average from two independent titrations and standard deviations are indicated in parentheses. n: number of moles of UHRA binding per mole of polyP75 and UFH; Ka: binding constant; ΔG: free energy change; ΔH: enthalpy change; TΔS: entropy change. UHRA exhibits high-affinity binding to polyP75 (a polyP preparation with median polymer lengths of approximately 75 phosphate units) and UFH. Table 4.2: Thermodynamic parameters for the interaction of UHRA with UFH and polyP75 determined by isothermal titration calorimetry. 4.4.2 UHRA does not interact with fibrinogen or alter thrombin-mediated fibrin polymerization The clotting cascade culminates with thrombin-catalyzed polymerization of fibrinogen into fibrin [44]. Fibrinogen is an anionic plasma protein, circulating at a concentration of 6-12 µM [44]. Polycations such as protamine bind fibrinogen largely through intermolecular coulombic attraction, which result in protein aggregation causing coagulopathy and pulmonary hypertension [169,172,178]. We studied the interaction of UHRA with fibrinogen by measuring changes in its intrinsic tryptophan fluorescence [257]. Binding of polycations quenches fibrinogen fluorescence as shown with polyethyleneimine [Figure 4.3A] where a concentration-dependent quenching was observed. From such data, the association constant Ka (M-1) and number of binding sites per fibrinogen molecule, n may be determined by regression of the relation [262]:                         𝑙𝑜𝑔 [(𝐹𝑜−𝐹)𝐹] = 𝑙𝑜𝑔(𝐾𝑎) + 𝑛𝑙𝑜𝑔([𝑈])  Polyanion #n #Ka M-1 $ΔG (kcal/ mol) #ΔH (kcal/ mol) $TΔS (kcal/ mol) UFH  1.13  6.86(±0.01)×105 -7.93 (±0.001) -81.36 (±0.3) -73.43 (±0.3)  polyP75 2.02 (±0.02)          8.51(±0.07)×105 -8.056 (±0.004) -76.97 (±0.44) -68.917 (±0.45) 149  where F and Fo represent fluorescence intensities of fibrinogen in the presence and absence of the polycation, respectively, and [U] is the equilibrium concentration (M) of unbound polycation. Linear least-squares regression of the double logarithm plotted data (R2 = 0.99) provides estimates of Ka and n, which for polyethyleneimine – fibrinogen binding are 4.7 × 106 M-1 and 0.75, respectively.  In contrast, the fluorescence profile of fibrinogen was unaffected by addition of increasing amounts of UHRA [Figure 4.3B], indicating that UHRA has minimal interaction, if any.   Figure 4.3: UHRA does not perturb the intrinsic tryptophan (Trp) fluorescence of human fibrinogen (fibrinogen). (A) The effect of increasing PEI concentrations on the Trp fluorescence emission of fibrinogen. PEI quenched the fluorescence signal in a concentration dependent fashion. (B) The effect of increasing UHRA concentration on the Trp fluorescence emission of fibrinogen. The intensity of fluorescence signal remains unchanged even in the presence of 1000 µg/mL UHRA, indicating minimal interaction of UHRA with fibrinogen. The depicted fluorescence spectra are the average of three independent experiments. Error values are small and not shown for data-presentation clarity. The insets show the quenching effect of UHRA or PEI on fibrinogen fluorescence. Data in the inset are mean ± SD (n=3). To support this observation, we analyzed the secondary structure of fibrinogen in the presence of UHRA or PEI using CD spectroscopy. The α-helical content of pure fibrinogen gives two pronounced negative ellipticity peaks at 208 and 222 nm  [257]. Co-150  incubation of UHRA with fibrinogen does not alter ellipticity at 208 and 220 nm. The presence of polyethyleneimine (500 µg/mL), however, increases ellipticity [Figure 4.4A-B].   Figure 4.4: UHRA does not change the secondary structure of human fibrinogen (fibrinogen). (A) CD spectrum of fibrinogen- PEI mixture showing perturbation in ellipticity values. PEI at 500 or 1000 µg/mL causes significant changes in the secondary structure of fibrinogen. (B) CD spectrum of fibrinogen-UHRA mixture. No significant change in ellipticity at 208 and 222 nm is observed. CD spectra shown are baseline-corrected using mean protein-free control scans (n=2).  Deconvolution of the spectra shows that the native fibrinogen used in this study is comprised of ca. 30(±2)% α-helix, 19(±1)% β-sheet and 35(±1)% random coil secondary structure content. In the presence of 1000 µg/mL of UHRA, fibrinogen secondary structure is unchanged (30(±3)% α-helix, 19(±2)% β-sheet, and 34(±2)% random coil), again suggestive of at most a weak intermolecular interaction. In contrast, addition of polyethyleneimine alters fibrinogen secondary structure (24(±1)% α-helix, 23(±1)% β-sheet, and 40(±1)% random coil) [Figure 4.5].  151   Figure 4.5: Influence of UHRA and PEI on secondary structure of fibrinogen measured by CD spectroscopy. Secondary structures of human fibrinogen remain unchanged following incubation with UHRA at different concentrations. Conversely, PEI at 500 µg/mL decreased the α-helical content by 19.04 %, concurrently increasing β-sheet and random coil by 23.2 % and 13.55 %, respectively. However, no statistical significance was observed between control and treatment groups. Data shown here is the mean ± SEM from two independent experiments.  The interaction between UHRA and fibrinogen was also examined directly using ITC, with representative raw and integrated heat data shown in Figure 4.6. Titration of UHRA into fibrinogen yields weak endothermic peaks indicating no significant interaction between the two components.      152    Figure 4.6: A binding isotherm obtained using ITC by titrating UHRA with fibrinogen. The upper panel depicts the raw data (power signal) and the lower panel shows the integrated areas corresponding to each injection, normalized to the moles of UHRA injected into fibrinogen the solution and plotted as a function of molar ratio (UHRA/fibrinogen). The raw heat data of UHRA titration into fibrinogen solution showed very weak endothermic peaks. This confirmed that UHRA did not bind to fibrinogen. Finally, we investigated fibrinogen polymerization by thrombin in the presence of UHRA or protamine. Results demonstrated that UHRA does not significantly change turbidity profiles or final fibrin turbidities relative to the polycation-free control [Figure 4.7]. However, protamine addition increased the turbidity significantly in a manner consistent with its interaction with fibrinogen. The final turbidity of fibrin clots correlates with fiber size, with an increase in the final turbidity indicative of thicker fibers, and thus, altered clot structure [44,263].  153   Figure 4.7: UHRA does not influence fibrin polymerization. The final turbidity of mature fibrin clots produced from purified fibrinogen in the presence of UHRA or protamine. UHRA (500 µg/mL) does not alter the final turbidity relative to the polycation-free control. In comparison, significant increases in final turbidity are observed as protamine concentration is increased to 50 µg/mL (****p < 0.0005). All experiments were performed in triplicate. Results are expressed as the mean ± SE of nine measurements from three independent experiments. Unpaired 2-tailed t tests were performed to determine significance, with p < 0.05 indicating a statistically significant change. The data therefore demonstrate that, unlike protamine or polyethyleneimine, UHRA does not alter fibrin structure, or cause fibrinogen aggregation and/or precipitation [Figure 4.8].   154  Figure 4.8: Effect of protamine, UHRA and Polyethyleneimine (PEI) on fibrinogen solution. The change in turbidity of 3 mg/mL human fibrinogen solution following incubation with protamine, UHRA and PEI at room temperature was monitored for 15 minutes. Images shown here were captured at t =15 min. (A, C) Turbidity was observed in both fibrinogen solution treated with 50 µg/mL protamine and 25 µg/mL PEI, respectively. (B) No change in turbidity was seen in fibrinogen solution incubated with UHRA even at 500 µg/mL. 4.4.3 UHRA has no effect on tissue factor /recalcification-initiated plasma coagulation and clot lysis  Polycations can perturb the blood-clotting cascade either by initiating or delaying clotting. Polyamidoamine dendrimers, for example, are procoagulant, while protamine possesses intrinsic anticoagulant properties [135,169]. To investigate the impact of UHRA on clotting, we performed turbidimetric plasma clotting assays. When coagulation was triggered by adding tissue factor (TF) to recalcified plasma containing UHRA, no significant differences in lag time or maximum absorbance (MaxAbs) were observed – even at 1000 g/mL of UHRA (p = 0.14 and 0.10, respectively) – compared to the buffer control [Figure 4.9A-C]. The clotting profile of recalcified plasma containing UHRA is likewise comparable to the buffer control [Figure 4.9D-F]. No significant alteration in lag time or Max Abs was observed even at 1000 g/mL of UHRA (p = 0.08 and p = 0.29, respectively). In contrast, protamine, even at 50 µg/mL, increased both the lag time and MaxAbs (p <0.001 and <0.005, respectively) of TF-induced coagulation of recalcified plasma. It also increased the final turbidity, while demonstrating anticoagulant activity and abnormal fiber formation in the coagulation of recalcified plasma (p = 0.01 and p = 0.009, respectively).    155    156  Figure 4.9: Plasma clot formation and clot turbidity is unaffected by UHRA. Clot formation in diluted human plasma titrated with varying amounts of UHRA or protamine was investigated using a turbidimetric assay as described in the Methods section. (A) Turbidity curves (A405nm) were obtained upon addition of TF and CaCl2 (20 mM) to plasma. (B) Lag time (sec) characterizing the time taken for initial protofibril formation during clotting were determined from (A). Significant prolongation of lag time was observed by 50 µg/mL protamine (***p <0.001). Remarkably, even at 1000 µg/mL UHRA, no significant change in lag time was observed. (C) Maximum absorbance values of plasma clots determined from (A) formed with UHRA remain unchanged, while significant (**p < 0.005) changes are recorded for clots formed in 50 µg/mL protamine. This indicated that UHRA neither inhibits nor alters fibrin polymerization in plasma. (D) Turbidity curves (A405nm) were obtained upon recalcification (20 mM) of plasma. (E) Lag times observed in the presence of UHRA or protamine. No significant change in lag time was observed with UHRA, while impaired plasma clotting caused prolongation of lag times at all protamine concentrations studied (*p < 0.015). (F) Maximum absorbance of plasma clots containing UHRA or protamine. No statistically significant differences in final absorbance values were recorded for the UHRA containing and polycation-free systems, while significant changes (** p < 0.01) where observed in the 50 µg/mL protamine system. Moreover, no clot was formed at 100 at 200 µg/mL protamine, demonstrating its potent intrinsic anticoagulation activity. (A) and (D) report the average turbidity obtained from three separate experiments. Absorbance measured at 3 minutes interval is depicted. Results are expressed as the mean ± SE of nine measurements from three independent experiments. Unpaired 2-tailed t tests were performed to determine the significance. In turbidimetric plasma clotting assays, prolongation of the lag time suggests a defective clotting reaction, while higher turbidity indicates either thicker fibrin fibers or precipitation of plasma proteins [263]. In this instance, precipitation of plasma proteins by protamine was discounted as there was no increase in the initial absorbance at higher concentrations of protamine. We then used a turbidimetric fibrinolysis assay to investigate the stability of plasma clots formed in the presence of UHRA or protamine. Lysis of plasma clots formed in the presence of UHRA was similar to that observed in the polycation-free control [Figure 4.10A].  Moreover, the clot lysis half-time (CLT50) and the area under the clot lysis curve (AUC) revealed no significant differences between plasma clots formed in the presence or absence of UHRA [Figure 4.10B-C].  157  In contrast, clots formed in the presence of protamine, lyse rapidly compared to the polycation-free control as shown by CLT50 measurements [Figure 4.10B] (p = 0.03).  Taken together, these results show that clot formation and lysis are not influenced by the UHRA molecule, but are altered significantly by protamine.   158  Figure 4.10: UHRA does not promote lysis of plasma clots. The influence of UHRA or protamine on human plasma clot lysis was investigated using turbidimetry as described in the Methods section. (A) Turbidity curves (A405nm) showing lag, clot-formation and lysis phases; error bars were omitted for clarity. TF and CaCl2 (20 mM) initiated clot formation. Clot lysis was enhanced by adding exogenous tissue plasminogen activator (t-PA) at the initiation of clot formation. Absorbance measured at 2 minutes interval is depicted for lag time. Then absorbance measured at 8 minutes interval is depicted. (B, C) CLT50 and AUC values in the presence of UHRA or protamine. Relative to the polycation-free control, clots formed in the presence of UHRA show no significant change in CLT50 and AUC values, indicating they are stable and have a normal degradation profile. CLT50 values were reduced significantly (* p < 0.035) in the presence of 50 and 100 µg/mL protamine, respectively, suggesting faster lysis of these clots compared to the polycation-free control. Results are expressed as the mean ± SE of six measurements from two independent experiments. Unpaired 2-tailed t tests were performed to determine the significance. 4.4.4 UHRA has negligible impact on purified fibrin clot and whole-blood clot structure We next investigated the impact of UHRA or protamine on fibrin clot architecture by direct visualization using SEM. Previous studies show that polycations alter clot structure through non-specific binding effects [264]. As evidence of this, we found that protamine (25 µg/mL) changes clot morphology and increases the mean fiber diameter (p < 0.001) [Figure 4.11B-C]. The thicker fibrin fibers formed in the presence of protamine correlated with the elevated final turbidities (A405nm) recorded in the corresponding fibrin polymerization assay [Figure 4.7]. However, as shown in Figure 4.11A, fibrin clots formed in the presence of UHRA are homogeneous in structure and exhibit no differences from those formed in the buffer control, even at 500 µg/mL of UHRA. In addition, UHRA does not change the mean fiber diameter [Figure 4.11C].   159   160  Figure 4.11: UHRA does not alter purified fibrin clot morphology or fiber diameter. Clots were made by incubating 3 mg/mL human fibrinogen in 3.0 mM CaCl2 plus UHRA or protamine, then initiating clotting with 2.5 NIH U/mL thrombin. Clots were then allowed to mature for 1 hour and processed for SEM imaging. (A) SEM images of fibrin clots formed in the presence of UHRA at different concentrations (50 to 500 µg/mL) were determined at both low (10000 X, scale bar= 2 µm) and high (25000 X, scale bar= 1 µm) magnifications.  Clot architectures formed in the presence of UHRA are similar to that for the polycation-free control, even up to UHRA concentrations of 500 µg/ml. (B) SEM images of fibrin clots formed in the presence of protamine, exhibit altered morphologies compared to the control clot. (C) Fibrin fiber diameters of clots formed in the presence of UHRA or protamine. Fiber diameter is measured from scanning electron micrographs using ImageJ software. A total of 30 fibers were analyzed. Fibers for size analysis were selected by probing four different spots in each image. Data are mean ± SE (n=30 fibers; measured from 4 images of two independent clots). Statistical significance for fiber diameter was determined by comparing the UHRA or protamine treated group to the control using a one-way ANOVA followed by a Dunnett post hoc test.  Fibrin fibers formed in the presence of 25 µg/mL protamine are significantly thicker than those in the control clot (***p < 0.001).  To further delineate the effect of protamine and UHRA on whole blood clot morphology, we prepared clots with varying amounts of either polycation, and then analyzed them by SEM. Blood clots prepared in the presence of UHRA showed normal shaped erythrocytes entrapped in a fibrin mesh, as well as fibrin strands anchored to platelet aggregates similar to buffer control clots [Figure 4.12A]. Blood clots formed in the presence of 25 and 50 µg/mL of protamine showed normal clot architecture. However, blood clots formed in the presence of 100 µg/mL protamine, showed thicker and disarrayed fibrin fibers, with no platelet–fibrin networks and platelet aggregates [Figure 4.12B].     161   Figure 4.12: Clot characteristics formed in whole blood remain unchanged in the presence of UHRA. Clotting was initiated by recalcifying human whole blood with 11.1 mM CaCl2. Clot samples were then processed for SEM imaging. (A) Clots formed in the presence of 500 µg/mL UHRA did not undergo detectable morphological changes. (B) However, clots formed in the presence of 100 µg/mL protamine showed thicker clot fibers, less platelet aggregates and complete abnormality in clot architecture. Also, at this concentration of protamine, tiny clots were obtained due to intrinsic anticoagulation effect of protamine. Clotting was inhibited at higher concentrations. Clot images were taken at two magnifications 2500X and 5000X, respectively. Only images from the 5000X magnification are depicted. Scale bar indicate 4 µm. The activity and amount of thrombin present in blood influence clot formation and morphology [44]. Studies have shown that thrombin activity is affected by protamine [135,265]. To test whether UHRA has any influence on thrombin activity, we assessed the ability of thrombin to cleave a chromogenic peptide substrate in the presence of UHRA.  We did not observe change in the initial rate of chromophore release from the substrate, suggesting that UHRA does not affect thrombin activity [Figure 4.13A] As impaired thrombin generation by protamine is responsible for its intrinsic anticoagulant effect, we also evaluated the impact of UHRA on TF-initiated thrombin generation by performing a fluorogenic thrombin generation assay. Upon clotting of platelet-rich plasma incubated with 100 or 200 µg/mL of UHRA, we did not observe any 162  significant changes in endogenous thrombin potential or the amount of thrombin generated [Figure 4.13B-D], which is consistent with the normal blood clot morphology observed [Figure 4.12A].    Figure 4.13: UHRA does not influence thrombin activity and thrombin generation in human platelet-rich plasma. (A) To investigate whether thrombin activity is directly affected by UHRA, a fixed concentration 0.5 NIHU/mL of thrombin was incubated with 100 and 200 µg/mL concentration of UHRA for 30 min at 37°C. The ability of thrombin to cleave a chromogenic substrate (50 µM) was then measured by monitoring the change in absorbance at 405 nm as a function of time. The initial rates of thrombin generation is unaffected by the presence of UHRA indicating that UHRA has no influence on thrombin amidolytic activity. (B) Coagulation was initiated in pooled human platelet-rich plasma (PRP) with 0.5 pM TF and 17 mM CaCl2, and thrombin generation was monitored using 0.416 mM fluorogenic substrate  (Z-Gly-Gly-Arg-AMC ), as described above. Thrombin generation curves of PRP containing UHRA at 100 and 200 µg/mL follows a somewhat 163  similar profile compared to the normal PRP (control). (C) ETP was calculated from the thrombin generation curves obtained following incubation of UHRA with the normal PRP. No significant change in values was observed compared to the normal PRP.  (D) The peak amount of thrombin (nM) produced during coagulation in PRP incubated with UHRA was measured. UHRA did marginally reduce the peak amount of thrombin generated at both concentrations studied. However, the values were not statistically different from the normal PRP. Results are expressed as mean ± SEM of six measurements from two independent experiments. Individual thrombin generation assay was performed in pooled human PRP obtained from two different donors. Unpaired 2-tailed t tests were performed to determine the significance. Moreover, whole blood clotting in the presence of UHRA or protamine was evaluated by thromboelastography. Data shown in the Figure 4.14 corroborates that UHRA does not possess intrinsic anticoagulant activity compared to protamine.  Figure 4.14: UHRA does not show intrinsic anticoagulant activity in whole blood even in the absence of heparin. No significant change was observed in the clotting time between the UHRA treated group and the buffer control. However, the presence of protamine (25 µg/mL) significantly elevated the clotting time, demonstrating the intrinsic anticoagulant activity. Data are mean ± SE (n=3). Unpaired 2-tailed t tests were performed to determine significance, with p < 0.05 indicating a statistically significant change. (**p =0.004).  164  4.4.5 UHRA does not bind or incorporate into purified fibrin or whole blood clots We next examined whether UHRA is incorporated into clot structures by performing confocal microscopy on fibrin clots and blood clots prepared in the presence of Alexa Fluor 488 labelled UHRA or protamine. Fibrinogen was labeled with Alexa Fluor 546 (red). Results are summarized in the Figure 4.15. We observed that the architecture of clots formed in the presence of increasing amounts of [Figure 4.15A e-g, Figure 4.15B V-VII] was quite similar to the control clot [Figure 4.15A a, Figure 4.15B I]. In contrast, clots formed in the presence of protamine showed fibrin(ogen) aggregates that appear as numerous distinct dots [Figure 4.15A b-d, Figure 4.15B II-IV; indicated by white arrow heads]. In addition, protamine was incorporated throughout the fibrin structure in a concentration dependent manner [Figure 4.15A i-k, Figure 4.15C) and exhibited co-localization in both channels [Figure 4.15A p-r]. No incorporation of UHRA was observed even at 200 µg/mL, with results comparable to the buffer control [Figure 4.15A l-n, Figure 4.15C].  Similar findings were obtained for whole blood clots, where UHRA showed minimal incorporation into clots. Clots formed in the presence of protamine fluoresce green, with a pattern suggesting protamine binding to both fibrin fibers and platelet aggregates [Figure 4.15D]. Taken together, the results indicate that UHRA does not bind to or incorporate within fibrin, blood clots or their structural components.  165  166  Figure 4.15: UHRA does not bind or incorporate into purified fibrin, blood clots or other clot components. Clots for confocal microscopy were prepared as described in the Methods section. Briefly, human fibrinogen  3 mg/mL of which 2% was Alexa Fluor 546–Fibrinogen was mixed with UHRA or protamine (both conjugated with Alexa Fluor 488) and clotting was initiated with thrombin 2.5 NIH U/mL and 3 mM CaCl2 (A) The top panel (red channel) shows confocal micrographs of fibrin clots formed from Alexa Fluor 546–labeled fibrinogen. The middle panel (green channel) shows confocal micrographs of protamine or UHRA localization within fibrin clots. Fibers of clots (i-k) formed in the presence of protamine exhibit green fluorescence, indicating binding of protamine to fibrin fibers. Green signal is absent in clots formed in the presence of UHRA (l-n), suggesting no binding or incorporation of UHRA into fibrin fibers. The lower panel (merge) shows overlay of images from the red and green channels confirming the colocalization of fibrin and protamine. Scale bar indicate 10 µm (B) Magnified images from the white box shown in the red channel. Visual inspection of clots (II-IV) formed in the presence of Alexa Fluor 488-labelled protamine reveals different fibrin architecture (presence of more compact and round fibrin (ogen) aggregates indicated by white arrow heads) compared to the control clot (I). No architectural differences are observed for clots formed in the presence and absence of UHRA (V-VII). Scale bar indicate 2 µm. (C) Normalized raw integrated density (sum of all pixel intensities normalized to polycation-free buffer) of the green channel analyzed using FIJI. Depicted are mean values ± SD with 10 or more randomly acquired images analyzed per condition. The p values were determined using the unpaired t-test. (***p < 0.001). (D) Representative confocal micrographs of whole blood clots formed in the presence of UHRA or protamine. No green fluorescence is observed in blood clots formed in the presence of UHRA. This shows that UHRA does not interact or incorporate into the clot components. In contrast, fluorescence from F-Protamine containing blood clot fibers and platelet aggregates/clumps show binding of protamine to these clot components. Scale bar indicate 10 µm. 4.4.6 UHRA reverses anticoagulant activity of UFH without lung injury and alteration in clot morphology  To avoid heparin-induced bleeding complications in postoperative patients, heparin is neutralized by administering protamine [266]. However, to restore hemostasis, the optimal protamine dose must be administered as excess heparin or excess protamine may cause bleeding [267]. To test the effect of excess UHRA levels on heparin neutralization and clotting, we simulated an antidote overdose scenario by titrating UHRA or protamine into heparinized plasma (4 IU/mL UFH, final). Effects on 167  coagulation were assessed by measuring changes in the activated thromboplastin time assay. Results show that UHRA (50-250 µg/mL) normalized the elevated clotting time induced by heparin, and with excess UHRA levels showing no adverse effect on clotting. Protamine, as is known, showed a narrow therapeutic window, with excess protamine levels impairing clotting [Figure 4.16].   Figure 4.16: UHRA reverses anticlotting activity of UFH without impairing clotting even with excess levels of UHRA. Neutralization of heparin by UHRA or protamine was studied by aPTT assay in heparinized human plasma. UHRA neutralizes UFH (4 IU/mL) over a wide range of concentrations. Conversely, excess protamine impairs clotting. Results are expressed as the mean ± SE of four measurements from two independent experiments. In addition, we showed that, UHRA can completely neutralize anticoagulation activity of low-molecular weight heparin (tinzaparin), compared to protamine, which is only partially effective [Figure 4.17].   168   Figure 4.17: UHRA reverses anticlotting activity of tinzaparin. Neutralization of anticoagulation activity of tinzaparin 1 IU/mL by UHRA or protamine was studied by aPTT assay in human plasma. Dotted line represents the clotting time for control plasma. UHRA completely neutralizes anticoagulation activity of tinzaparin. Conversely, protamine could only partially neutralize anticoagulation activity of tinzaparin.  Using turbidimetric and clot elastic modulus studies, other investigators have shown that clots formed after neutralization of heparin with excess protamine possess altered and weaker clot structure [268]. We therefore analyzed the morphology of blood clots obtained after neutralization of UFH (2 IU/mL, final) with UHRA or protamine using SEM. Visual inspection of clot micrographs revealed that at all tested concentrations of UHRA, clots exhibited normal architecture [Figure 4.18A]. The normal clot morphology observed could be considered as an indirect measure of anticoagulant neutralization because clots were not formed in samples treated with UFH only. Also, clots formed in the presence of UHRA possess fiber diameters that are comparable to the buffer control [Figure 4.18B]. Interestingly, clots formed with excess protamine (75 µg/mL) contain 169  significantly thicker fibrin fibers compared to the buffer control [Figure 4.18B] (p < 0.0001) possibly due to binding and incorporation of protamine into clot fibers.    Figure 4.18: UHRA neutralizes heparin anticoagulation activity without altering clot morphology. Morphology of clots formed after neutralization of UFH with UHRA or protamine analyzed by SEM. Clot micrographs obtained in the presence of UHRA revealed normal morphology in comparison to the buffer control clot. Clot images at 5000X magnification are depicted. Scale bar indicate 5 µm 170  (C) Thickness of clot fibers was measured from clot micrographs using ImageJ. Fibers for size analysis were selected by probing four different spots in each image. Data are mean ± SE (n=30 fibers; measured from 2 images of each clot). Fibrin fibers formed in the presence of 75 µg/mL protamine are significantly thicker than those in the buffer control clot (one-way ANOVA followed by a Dunnett post hoc test; ****p < 0.0001). Size of clot fibers obtained after neutralizing UFH with UHRA at all tested concentrations is comparable to buffer control. To show that protamine could incorporate into clots even in the presence of UFH, we performed confocal microscopy on blood clots obtained after neutralizing 2 IU /mL UFH with Alexa 488 tagged UHRA or protamine. Results show that in either heparinized or non-heparinized blood, protamine binds to clot fibers or its components [Figure 4.19].    171  Figure 4.19: Protamine binds or incorporates into blood clot fiber or other clot components even in the presence of UFH. Clots for confocal microscopy were prepared as described in the methods section. Briefly, heparinized blood (UFH 2 IU/mL) was mixed with UHRA or Protamine (both conjugated with Alexa Fluor 488). Clotting was then initiated by adding CaCl2 (11.1 mM, final) (A) The top panel  (red channel) shows confocal micrographs of blood clots in which blood cells are stained with Cell mask deep red stain. The middle panel (green channel) shows confocal micrographs of clots with localization of Alexa 488 protamine or UHRA within clots. Clots formed in the presence of protamine exhibit green fluorescence, indicating binding of protamine to fibrin fibers. Green signal is absent in clots formed in the presence of UHRA suggesting no binding or incorporation of UHRA into fibrin fibers. The lower panel (merge) shows overlay of images from the red and green channels, confirming the colocalization of fibrin and protamine. Scale bar indicate 10 µm.    Finally, we investigated the effect of heparin neutralization by UHRA or protamine on the mouse lung morphology. Lung injury is a common postoperative complication observed in patients undergoing cardiopulmonary bypass (CPB) [269]. Although the etiology of CPB-induced lung injury is multifactorial, studies suggest that administration of protamine to neutralize heparin after CPB may contribute adverse pulmonary events such as noncardiogenic pulmonary edema (NCPE), characterized by loss of alveolocapillary membrane integrity [270-272]. Therefore, we administered UFH (200 IU/kg) i.v. followed by UHRA (10 mg/kg) or protamine (5 mg/kg) i.v. into mice. Clotting assays revealed that both UHRA and protamine neutralized the anticoagulant activity of UFH [Figure 4.20A]. In the UFH only-treated group, we observed mild hemorrhage characterized by the presence of red blood cells in the alveolar space. Consistent with clinical manifestations, in the protamine-treated group, we observed disruption of alveolar membrane and subsequent enlargement of alveolar sacs [Figure 4.20B]. Quantification of the alveolar area in those protamine-treated mice showed significant alveolar space enlargement due to loss of membrane integrity (p < 0.0001) 172  [Figure 4.20C]. In contrast, normal lung ultrastructure was observed in UHRA-treated group, which is comparable to the buffer-treated group [Figure 4.20 B].      173   Figure 4.20: UHRA reverses anticlotting activity of UFH with no adverse effect on lung ultrastructure or clot morphology. (A) In vivo neutralization of UFH activity were studied in mice by injecting  UFH 200 IU/kg intravenously (via tail-vein), followed by UHRA (10 mg/kg) or protamine (5 mg/kg). aPTT confirms neutralization of UFH activity by antidotes. (B) Histopathological sections of lungs after heparin neutralization were obtained of buffer and UHRA-treated groups, which were comparable. However, in the protamine-treated group, significant damage to lung alveolar structure was observed. Lungs from the heparin- only treated group showed mild hemorrhage (presence of red blood cells in alveolar space). Scale bar indicates 50 µm. In magnified images, scale bar indicate 20 µm. A= Alveolar space, AH= Alveolar hemorrhage and AD= Alveolar disruption. (C) Relative cumulative distribution of lung alveolar area. Depicted are values measured from 144 images from the lungs of eight mice (two mice per treatment group). Measurements confirm significant enhancement of alveolar area in the protamine-treated group compared to the buffer control (one-way ANOVA followed by a Dunnett post hoc test; ****p <0.0002). No significant difference in the alveolar area was observed between the buffer control and UHRA-treated group (one-way ANOVA followed by a Dunnett post hoc test; p=0.45)  4.5 Discussion Hemocompatible polycations that can specifically bind and inhibit polyanionic modulators of hemostasis, such as heparins and polyP, with negligible adverse side-effects could provide improved treatment of disorders of hemostasis and thrombosis. However, previously proposed polycations are non-specific, binding to anionic plasma proteins, such as fibrinogen, in ways that lead to adverse effects [169,179,256]. We are therefore working toward developing cationic inhibitors with improved specificity and 174  reduced toxicity. Combined with our previous studies [165,168], the results shown here demonstrate that UHRA molecules can be safe and effective.  Although UHRA relies on HBGs for activity, these ligands are engineered within the mPEG brush that acts to limit nonspecific binding and the associated adverse effects while retaining specificity for highly polyanionic heparins and polyP [165,168]. Here we show that a UHRA molecule designed in this manner has minimal influence on blood proteins, clotting or clot morphology, even at concentrations 20-fold and 10-fold higher than required for heparin and polyP inhibitory activity, respectively [165,168].  The interaction of UHRA and fibrinogen was investigated because the latter is known to adsorb onto hydrophobic and positively charged surfaces and polycationic molecules such polyethyleneimine [257,273]. Our fluorometric and CD data show that UHRA does not interact with or induce fibrinogen conformational changes, even at high concentrations. The mPEG brush layer creates steric stabilization effects [274] capable of preventing non-specific binding to the underlying HBGs, therefore serves to effectively prevent fibrinogen binding to the polycationic UHRA molecule. Intermolecular ion-pairs formed with the relatively low density of anionic groups within fibrinogen are not sufficient to overcome the energetic penalty of brush compression [275], while our ITC data shows that high-affinity binding of heparin and polyP is retained due to more optimal formation of intermolecular ion-pairs.  The impact of the unique manner in which cationic ligands are presented within UHRA molecules is made clear through our comparative studies with the conventional unshielded polycation protamine. Clotting occurs through thrombin-mediated polymerization of fibrinogen to fibrin [50]. Binding of exogenous molecules such as 175  antibodies or synthetic polymers to fibrinogen can either enhance or inhibit fibrin polymerization, thereby affecting clot stability [276-279]. This stability depends on clot architecture, which is affected by thrombin and fibrinogen concentrations, pH and ionic strength [44,275]. A polycation that acts to perturb any of these factors can therefore influence clotting.  We found from turbidity profiles of fibrin clots that protamine addition causes increased clot turbidity, possibly due to nonenzymatic fibrinogen polymerization, incorporation of protamine into fibrinogen/fibrin, and/or polycation-induced precipitation of fibrinogen. SEM images of fibrin clots showed that fibrin structures formed in the presence of protamine were thicker, and the overall clot morphology was markedly different than the polycation-free control. In plasma clotting assays, higher concentrations of protamine delayed clotting and showed anticoagulant properties. Protamine also greatly elevated turbidity, possibly by impairing thrombin generation, which is known to result in formation of thick clot fibers [135,179] as was observed here. Clot architecture was considerably altered with increasing protamine concentrations, most notably observed above 50 µg/mL. Turbidimetric assays were used to show that clots formed in the presence of protamine exhibited enhanced fibrinolysis. This enhanced clot lysis might occur through several possible mechanisms. Our confocal images of purified fibrin and blood clots show that protamine is incorporated within the clot structure and promotes abnormal clot architecture (electron microscopy images shown in Figure 4.11 and 4.12) that could facilitate lysis. Clots containing mostly thicker fibers formed in presence of protamine are more susceptible to lysis than clots with thinner fibers [50]. Alternatively, polycationic surfaces can localize plasminogen and thereby enhance clot lysis 176  [280,281]. But, regardless of mechanism, our results confirm and help to explain why unshielded polycations such as protamine, most notably at concentrations > 30 µg/mL, adversely affect blood coagulation. However, Ainle FN, et al. have shown that protamine (in the absence of heparin) 10 µg/mL is enough to prolong the aPTT and prothrombin time. This clearly demonstrates that protamine at lower concentrations induces deleterious effects [135]. In CPB, protamine (< 50 µg/mL) is administered to neutralize UFH [135,231]. Protamine dosing correlates with bleeding during CPB, and delayed bleeding problems associated with CPB are linked to fibrinolysis instigated by protamine [258,282]. Our studies suggest that protamine-induced changes in clot architecture, possibly in combination with protamine-mediated changes in plasminogen localization, may serve to accelerate degradation of nascent clots formed upon heparin neutralization, thereby increasing the risk haemorrhagic complications. Generally, 1 mg of protamine neutralizes ~100 units of UFH [97]. To achieve hemostasis after cardiac surgeries, optimal dosing of protamine, based on plasma heparin concentration, has to be administered; otherwise excess UFH or protamine may result in bleeding [97]. Imbalances in heparin-protamine titration due to unfavorable pharmacokinetics and/or nonspecific binding to plasma proteins exhibited by UFH and protamine, may yield a relative “overdose” of heparin or protamine in surgical patients [97,283]. Our heparin-UHRA titration in plasma revealed that, unlike protamine, UHRA neutralizes heparin anticoagulant activity over a wide range of concentrations. Most importantly, UHRA overdose did not impair clotting. Based on our ex vivo results shown in this manuscript and from our previous studies [165], we can posit that to completely neutralize UFH 4 IU in blood [Figure 4.16], we would require a final concentration of 50 177  µg/mL UHRA. Thus, to neutralize UFH 5000 IU, we would require ~62.5 mg of UHRA [165].  We further show that UHRA reverses anticoagulant activity of UFH in mice without causing lung toxicity. Mice that received protamine for heparin neutralization showed significant lung damage with characteristic alveolar membrane damage. Protamine associated NCPE occurs in 0.2% of CPB patients with 30% mortality [284]. The exact mechanism for protamine induced NCPE is not known. However, complement activation by heparin-protamine complexes and subsequent activation of neutrophils in lungs release proteolytic enzymes that can disintegrate lung ultrastructure, while unshielded cationic charge in protamine can cause pulmonary endothelial cell injury [285]. Haemostatic complications and organ toxicity are avoided with UHRA technology. UHRA binds both UFH and polyP with high affinity, but shows no discernable interaction with fibrinogen or clot components. Normal clot formation, architecture, strength and fibrinolysis kinetics are therefore preserved in the presence of UHRA, and this blood compatibility aligns with the lack of toxicity of UHRA in rodents [165,168]. The findings reported here therefore further demonstrate the potential of UHRA as a next-generation antidote for UFH following CPB.  Finally, information from this study could be applied to design UHRA with extended safety profile and inhibition specificity. Additional treatment opportunities may therefore arise, for example, through screening of UHRA libraries to identify lead molecules to dismantle or neutralize detrimental histone-DNA complexes implicated in sepsis. 178  Chapter 5: Conclusions and future studies 5.1 Thesis summary Anticoagulants play a pivotal role in the prophylaxis and treatment of cardiovascular disorders such as ischemic heart disease, ischemic stroke and VTE. Heparin-based anticoagulants such UFH, LMWHs and the synthetic fondaparinux are widely used in clinics [97]. However, limitations associated with heparin therapy include HIT (commonly observed with UFH therapy) and bleeding [Table 1.1]. To date, UFH is the anticoagulant of choice for patients undergoing invasive surgical procedures such as CPB. This is partly due to the availability of protamine, the only clinically approved antidote for UFH to neutralize the excess circulatory UFH after surgeries [97]. Unfortunately, protamine has an unpredictable dose response, narrow therapeutic window and cause cardiovascular side-effects including systemic hypotension, anaphylaxis, and complement activation by protamine/UFH complexes [134-136]  Clinical limitations of heparin therapy led to the development of DOACs that either target FXa or FIIa in the coagulation cascade. In this thesis, we presented studies related to direct FXa inhibitors such as rivaroxaban, edoxaban etc. These novel oral anticoagulants have definite advantages over VKAs including improved dose response, fewer drug interactions, lower bleeding rates, however, there are no approved antidotes currently available to reverse anticoagulant activity following an undesirable event of life-threatening bleeding [Table 1.2].    From the information shown in Table 1.1 and 1.2, it is obvious that traditional and novel anticoagulants are associated with bleeding risk. The propensity to bleed is much higher among elderly population and patients with renal insufficiency. Therefore, there is an utmost and urgent clinical need for a highly efficient, nontoxic antidote molecule with 179  universal anticoagulant reversal activity. This will significantly improve the safety of anticoagulation therapy.    As described in the section 1.7, several investigators have reported development of antidote molecules. In this thesis, we presented data related to UHRA, andexanet alfa (AnXa) and PER977. UHRA is a synthetic heparin antidote developed by the Kizhakkedathu laboratory (University of British Columbia), and is currently undergoing extensive preclinical studies. AnXa is a truncated FXa recombinant protein developed by Portola Pharmaceuticals (South San Francisco, California) and currently waiting for FDA approval. PER977 is a small cationic molecule developed by Perosphere Pharmaceuticals (Danbury, Connecticut) and this molecule also in advanced stages of clinical development.   Using mouse models of bleeding, our previous publication [165], reported that UHRA can reverse anticoagulation activity of all clinically available heparin-based anticoagulants. Dose tolerance studies in mice revealed that UHRA is nontoxic. However, the mechanism of action and the role of each structural components of UHRA towards heparin antidote activity and nontoxic property were not explored. Thus in this thesis, to understand in-depth information about the molecular design of UHRA, we synthesized two UHRA analogs: N-UHRA with no mPEG chains and mPEG750-UHRA molecule with longer mPEG750 chains. Results were presented in the chapter 2. We studied the binding of these molecules along with UHRA and protamine to UFH. Thermodynamics of binding of antidotes to UFH at different ionic strength, revealed that the presence of mPEG chains impose entropic penalty to binding. The binding affinity of UHRA and mPEG750-UHRA to UFH diminished rapidly with increase in ionic strength. 180  The effect was more prominent with mPEG750-UHRA, as it possesses longer mPEG chains. Our studies suggest that the mPEG chains which exist in a brush conformation on the cationic core of UHRA provide charge shielding effect. Thus, only those anionic molecules with sufficient charge density such as heparins can compress the mPEG brush layer and achieve contacts with the cationic core of UHRA. Thus provides the much needed specificity for UHRA towards heparins and prevents its nonspecific interactions with blood proteins. In the case of N-UHRA and PS, unshielded cationic charges cause nonspecific interaction with blood proteins leading to adverse side-effects. Indeed, clotting assays reveal that both PS and N-UHRA exhibit intrinsic anticoagulant activity and possess narrow window for UFH neutralization. Conversely, both UHRA and mPEG750-UHRA exhibit UFH neutralization over a wide range of concentrations. We also observed that the size of complexes formed between protamine and N-UHRA with UFH is large compared to UHRA and mPEG750-UHRA complexes with UFH. This suggested that mPEG chains also regulate the size of complexes. Finally, fluorescence and ELISA experiments revealed that UHRA disrupt AT/UFH complexes to neutralize anticoagulation activity which provided important information regarding the mechanism of action of UHRA antidote activity.   In chapter 3, we compared the mechanism of action of UHRA, AnXa and PER977. First, we analyzed the binding affinity of antidotes to their intended target anticoagulants. As expected, UHRA bound to all forms of heparin with micromolar affinity. AnXa bound to DOACs with nanomolar affinity and with weaker affinity to heparins than UHRA. However, AnXa showed higher affinity binding to heparin/AT complex than uncomplexed heparins. This predicts that AnXa neutralize heparin 181  anticoagulation by direct binding to the AT/Heparin complex. PER977 bound to heparins with binding affinity comparable to that of AnXa. Interestingly, PER977 did not show any binding to DOACs. UHRA and AnXa did not bind to vital coagulation proteins such as thrombin, fibrinogen etc., whereas PER977 showed binding to FIXa. Fibrin polymerization assay show that AnXa and PER977 did not affect fibrin clot formation. AnXa only at supratherapeutic concentration marginally affected fibrin clot formation. We then studied the influence AnXa and PER977 on fibrin and blood clot architecture in the absence of anticoagulants. Both AnXa and PER977 did not alter fibrin and whole blood clot architecture. Interestingly, electron micrographs of edoxaban containing blood clots treated with AnXa or PER977, showed the capability of antidotes to restore the impaired fibrin generation. However, we observed that the clots obtained after neutralizing edoxaban with 200 µg/mL of PER977 possessed larger diameter fibers compared to the control clots. Finally, data obtained from clotting assays aligns with our binding studies. AnXa reversed anticoagulation activity of UFH, enoxaparin and edoxaban. UHRA reversed anticoagulation activity of UFH and enoxaparin. PER977 did not neutralize activity any of the tested anticoagulants. Overall, results presented in this chapter reveal the mechanism of action of AnXa and UHRA, respectively. However, more studies are necessary to elucidate the mechanism of action of PER977.  Results shown in the chapter 4 corroborated that mPEG chains in UHRA prevent nonspecific interaction with blood proteins and improve the hemocompatibility. First, by performing fluorescence, CD, ITC and fibrin polymerization assay, we showed that UHRA did not interact with fibrinogen and fibrin clot formation. On the other hand, PEI interacted with fibrinogen and perturbed its secondary structure. Protamine was 182  shown to increase the final turbidity of fibrin clots which suggests formation of thicker fibrin fibers. Unlike protamine, UHRA did not adversely affect TF and recalcification initiated plasma clotting. Protamine elevated lag time and maximum absorbance of plasma clots. Further, SEM images revealed that the presence of UHRA did not alter the architecture of fibrin and blood clots even in the absence of heparins. Similarly, blood clots obtained after reversing UFH with UHRA showed clot architecture comparable to the control clot. In addition, the size of fibrin fibers was comparable to the control clot. Conversely, protamine significantly increased the fibrin fiber size and altered the blood clot morphology. Confocal micrographs of fibrin and blood clots incubated with UHRA showed no binding of UHRA to fibrin and blood clot structures even in the absence of anticoagulants, whereas protamine bound to vital clot components. Implication of protamine binding to clot structures was revealed in the clot lysis assays. We observed that plasma clots formed in the presence of PS showed faster lysis than control clots. UHRA containing clots showed normal degradation profile. Finally, studies in mice showed that unlike PS, neutralization of UFH with UHRA did not cause lung damage. Based on the results presented in this chapter and our previous studies such as in vitro hemocompatibility, in vivo dose tolerance and bleeding time measurements suggests that UHRA is superior UFH reversal agent with minimal side effects in comparison to protamine [165,173]. In addition, studies also revealed a novel mechanism by which protamine affects clot structure and clot lysis.  5.2 Future studies Studies in this thesis showed superior anticoagulant activity and hemocompatibility of the UHRA molecule due to its molecular design and high binding affinity to heparins.  183  Due to these excellent properties, a number of new applications or studies can be envisioned. Some of them are proposed here as future studies.  5.2.1 Disruption of PF4/UFH complexes with UHRA  HIT is an immune mediated complication observed in a subset of patients receiving parenteral anticoagulants, primarily UFH [286]. Clinical features include mild to moderate thrombocytopenia with or without life-threatening thrombosis. Incidence of HIT is dependent on several factors including the type of anticoagulant administered, duration of exposure, type of surgery, and patient population [287]. HIT occur in 0.2 to 5% of patients receiving heparin based anticoagulants. However, there is a greater risk in patients receiving UFH during cardiopulmonary bypass graft and orthopaedic surgeries [288]. Use of LMWHs and fondaparinux as alternatives to UFH in certain clinical settings has reduced the incidence of clinical HIT to some extent [289]. However, bleeding complications associated with these anticoagulants due to dearth of proven antidotes complicates the scenario [99].  Thrombotic complications develop in 20 to 50% of patients diagnosed with HIT [290] Occlusions could arise in any vascular bed (predominantly venous), and contributes significantly to morbidity, mortality and amputations associated with HIT [290] Unfortunately, direct thrombin inhibitors approved for HIT treatment are expensive, lack proven antidotes, and ineffective in circumventing need for amputations [291]. The pathogenesis of HIT initiates when highly negatively charged UFH interact with cationic peptide PF4, resulting in ultra large PF4/heparin antigenic complexes [292]. Immune recognition of epitopes on complexes primarily by immunoglobulin G (IgGs) and subsequent binding of PF4/heparin-IgG immune complexes to platelets via FcgammaIIa receptor (FcγRIIa) activate platelets leading to thrombocytopenia [292].  184   UHRA achieves high specificity towards UFH. As indicated in the chapter 2, UHRA disrupts the AT/UFH complex to inhibit UFH anticoagulation activity. It has been shown previously that disruption of PF4/UFH complex could be a strategy to reduce complications in HIT [293]. Therefore, we hypothesize that UHRA could disrupt PF4/UFH complexes and reduce HIT antibody mediated platelet activation, aggregation and thrombotic complications [Figure 5.1].  Figure 5.1: (A) Cationic PF4 binds UFH and forms ultra large PF4/UFH complexes. These complexes are recognized by HIT antibodies. The PF4/UFH/IgG immune complexes bind to platelets via FcγRIIa, leading to platelet activation and aggregation. (B) Addition of UHRA disrupts PF4/UFH complex. As a proof-of-concept, we performed a bicinchoninic acid assay (BCA assay) to estimate the amount of PF4 released from PF4/UFH complex following treatment with UHRA.  A B 185   Figure 5.2: PF4 is released from PF4/UFH complexes treated with UHRA. Briefly, PF4/UFH complexes were prepared by adding 500 µL of UFH (4 IU, final) to 500 µL of PF4 (200 µg/mL, final) at room temperature. Photon correlation spectroscopy measurements were performed to confirm the presence of PF4/UFH complexes. In an eppendorf tube containing 10 µL of UHRA/buffer, 40 µL of PF4/UFH complex solution was added and incubated at 37 °C for 60, 180, 360 and 720 minutes (overnight) . Following incubation, samples were gently vortexed and then centrifuged @ 10,000 rpm for 5 minutes. 25 µL of the supernatant from each sample was transferred into wells of a 96-well microplate (costar) and 200 µL of BCA reagent was added into each well. Samples were then incubated for 30 min at 37 °C. Absorbance was then measured at 562 nm. Results are expressed as the mean ± SE of two measurements. (A) Increase in absorbance at 562 nm demonstrates release of PF4 from PF4/UFH complexes treated with UHRA.  (B) The amount of PF4 released from PF4/UFH complexes following treatment with UHRA, were determined by interpolating from a standard curve of bovine serum albumin. Results are expressed as the mean ± SE of two independent measurements.  Further, extensive ex vivo (platelet activation, aggregation assays etc.), and in vivo studies using murine HIT models are needed to validate the efficacy of UHRA as drug candidate for treatment of HIT. Very recently, it has been shown that polyP/PF4 complexes are also recognized by HIT antibodies [294]. We have shown that UHRA can bind to polyP.  Therefore, UHRA could be used to disrupt polyP/PF4 complexes. These studies are currently ongoing.  A B 186  5.2.2 Inhibitors for prothrombotic nucleic acids Recent studies have shown that polyanions such as polyphosphates and extracellular nucleic acids can initiate clotting, leading to thrombosis [89]. In fact, high levels of circulating DNA (cDNA) are found in patients with sepsis, myocardial infarction and cancer [92-95]. Therefore, targeting and inhibiting prothrombotic activity of these polyanions would be a viable strategy for prevention and treatment of thrombosis. This would reduce the use of anticoagulants and subsequent bleeding complications.   From our previous report and data presented in the chapter 2, we know that the binding affinity of UHRA to UFH can be modulated by changing the charge density of the core, length of mPEG chains, ionic conditions, pH etc. Currently, UHRA consists of methylated tris(2-aminoethyl)amine ligands (4 nitrogen; figure 2.2A) that acquire +3 charge at pH 7.4. The spatial arrangement of anionic charges on nucleic acid is different from heparins. Hence, novel multivalent ligands could be developed to achieve high affinity binding to nucleic acids. For instance, more dendritic ligand with six protonatable nitrogen (figure 5.3) could be attached to the HPG core using click chemistry.    Figure 5.3: Novel multivalent ligand with more tertiary amine groups. Inhibitor molecule/s that achieves micromolar or higher binding affinity to nucleic acid 187  will be identified by performing ITC. Such molecules could be tested for the neutralization of prothrombotic activity of nucleic acids using clotting assays and thrombin generation assay. Finally, selected inhibitor molecule/s could be evaluated using a cecal ligation puncture model of sepsis.               188  Bibliography 1) Versteeg HH, Heemskerk JW, Levi M, Reitsma PH. New fundamentals in hemostasis. Physiol Rev 2013; 93 (1):327-358. 2) Davie EW, Ratnoff OD. Waterfall Sequence for Intrinsic Blood Clotting. Science 1964; 145(3638):1310-1312. 3) Macfarlane RG. An Enzyme Cascade in the Blood Clotting Mechanism, and it’s Function as a Biochemical Amplifier. Nature 1964; 202:498-499. 4) Gailani D, Renne T. Intrinsic pathway of coagulation and arterial thrombosis. Arterioscler Thromb Vasc Biol 2007; 27(12):2507-2513. 5) Mackman N. The Role of Tissue Factor and Factor VIIa in Hemostasis. Anesth Analg 2009; 108(5):1447-1452. 6) Levy JH, Szlam F, Wolberg AS, Winkler A. Clinical Use of the Activated Partial Thromboplastin Time and Prothrombin Time for Screening: A Review of the Literature and Current Guidelines for Testing. Clin Lab Med 2014; 34(3):453-477. 7) Renne T, Schmaier AH, Nickel KF, Blomback M, Maas C. In vivo roles of factor XII. Blood 2012; 120(22):4296-4303. 8) Hoffman M, Monroe DM. Coagulation 2006: a modern view of hemostasis. Hematol Oncol Clin North Am 2007; 21(1):1-11. 9) Roberts HR, Hoffman M, Monroe DM. A cell-based model of thrombin generation. Semin Thromb Hemost 2006; 32 Suppl 1:32-38. 10)  Chen J, López JA. Interactions of platelets with subendothelium and endothelium. Microcirculation. 2005; 12(3):235-46. 11) Walker RK, Krishnaswamy S. The activation of prothrombin by the prothrombinase complex. The contribution of the substrate-membrane interaction to catalysis. J Biol Chem 1994; 269(44):27441-27450. 12)  Stegner D, Nieswandt B. Platelet receptor signaling in thrombus formation. J Mol Med (Berl). 2011; 89(2):109-21.  13) Kravtsov DV, Matafonov A, Tucker EI, Sun MF, Walsh PN, Gruber A, et al. Factor XI contributes to thrombin generation in the absence of factor XII. Blood 2009; 114(2):452-458. 189  14) Matafonov A, Sarilla S, Sun MF, Sheehan JP, Serebrov V, Verhamme IM, et al. Activation of factor XI by products of prothrombin activation. Blood 2011; 118(2):437-445. 15) Ahmad SS, Rawala-Sheikh R, Walsh PN. Components and assembly of the factor X activating complex. Semin Thromb Hemost. 1992; 18(3):311-23. 16) Monroe DM, Roberts HR, Hoffman M. Platelet procoagulant complex assembly in a tissue factor-initiated system. Br J Haematol 1994; 88:364–71. 17) Rawala-Sheikh R, Ahmad SS, Monroe DM, et al. Role of gamma-carboxyglutamic acid residues in the binding of factor IXa to platelets and in factor-X activation. Blood 1992; 79:398–405. 18) Baird TR, Walsh PN. The interaction of factor XIa with activated platelets but not endothelial cells promotes the activation of factor IX in the consolidation phase of blood coagulation. J Biol Chem 2002; 277(41):38462-38467. 19) Wood JP, Ellery PER, Maroney SA, Mast AE. Biology of tissue factor pathway inhibitor. Blood. 2014; 123(19):2934-2943.  20) Wiebe EM, Stafford AR, Fredenburgh JC, Weitz JI. Mechanism of catalysis of inhibition of factor IXa by antithrombin in the presence of heparin or pentasaccharide. J Biol Chem 2003; 278(37):35767-35774. 21) Mann KG, Butenas S, Brummel K. The dynamics of thrombin formation. Arterioscler Thromb Vasc Biol 2003; 23(1):17-25. 22) Furie B, Furie BC. Mechanisms of thrombus formation. N Engl J Med 2008 28; 359(9):938-949. 23) Dorgalaleh A, Rashidpanah J. Blood coagulation factor XIII and factor XIII deficiency. Blood Rev 2016; 30(6):461-475. 24) Maclean PS, Tait RC. Hereditary and acquired antithrombin deficiency: epidemiology, pathogenesis and treatment options. Drugs 2007; 67(10):1429-1440. 25) Patnaik MM, Moll S. Inherited antithrombin deficiency: a review. Haemophilia 2008; 14(6):1229-1239. 26) Weitz JI. Heparan sulfate: antithrombotic or not? J Clin Invest 2003; 111(7):952-954. 190  27) Jin L, Abrahams JP, Skinner R, Petitou M, Pike RN, Carrell RW. The anticoagulant activation of antithrombin by heparin. Proc Natl Acad Sci U S A 1997; 94(26):14683-14688. 28) Griffin JH, Zlokovic BV, Mosnier LO. Protein C anticoagulant and cytoprotective pathways. Int J Hematol 2012; 95(4):333-345. 29) Aird WC. Natural anticoagulant inhibitors: activated Protein C. Best Practice & Research Clinical Haematology 2004; 17(1):161-182. 30) Esmon CT. The Protein C Pathway. Chest 2003; 124(3, Supplement):26S-32S. 31) Faust SN, Levin M, Harrison OB, Goldin RD, Lockhart MS, Kondaveeti S, et al. Dysfunction of endothelial protein C activation in severe meningococcal sepsis. N Engl J Med 2001; 345(6):408-416. 32) Wood JP, Bunce MW, Maroney SA, Tracy PB, Camire RM, Mast AE. Tissue factor pathway inhibitor-alpha inhibits prothrombinase during the initiation of blood coagulation. Proc Natl Acad Sci USA 2013; 110(44):17838-17843. 33) Longstaff C, Kolev K. Basic mechanisms and regulation of fibrinolysis. J Thromb Haemost 2015; Suppl 1:S98-105. 34) Chapin JC, Hajjar KA. Fibrinolysis and the control of blood coagulation. Blood Rev 2015; 29(1):17-24. 35) Ellis V. Cellular receptors for plasminogen activators recent advances. Trends Cardiovasc Med 1997; (7):227-234. 36) Cesarman-Maus G, Hajjar KA. Molecular mechanisms of fibrinolysis. Br J Haematol 2005; 129(3):307-321. 37) Lijnen HR, Zamarron C, Blaber M, Winkler ME, Collen D. Activation of plasminogen by pro-urokinase. I. Mechanism. J Biol Chem 1986; 261(3):1253-1258. 38) Ewald GA, Eisenberg PR. Plasmin-mediated activation of contact system in response to pharmacological thrombolysis. Circulation 1995; 91(1):28-36. 39) Foley JH, Walton BL, Aleman MM, O'Byrne AM, Lei V, Harrasser M, et al. Complement Activation in Arterial and Venous Thrombosis is Mediated by Plasmin. EBioMedicine 2016; 5:175-182. 40) Schneider M, Nesheim M. A study of the protection of plasmin from antiplasmin inhibition within an intact fibrin clot during the course of clot lysis. J Biol Chem 2004; 279(14):13333-13339. 191  41) Bouma BN, Mosnier LO. Thrombin activatable fibrinolysis inhibitor (TAFI)--how does thrombin regulate fibrinolysis? Ann Med 2006; 38(6):378-388. 42) Medved L, Weisel JW. Fibrinogen and Factor XIII Subcommittee of Scientific Standardization Committee of International Society on Thrombosis and Haemostasis. Recommendations for nomenclature on fibrinogen and fibrin. J Thromb Haemost 2009; 7(2):355-359. 43) Weisel JW, Litvinov RI. Mechanisms of fibrin polymerization and clinical implications. Blood 2013; 121(10):1712-1719. 44) Wolberg AS. Thrombin generation and fibrin clot structure. Blood Rev 2007; 21(3):131-142. 45) Medved' LV, Gorkun OV, Privalov PL. Structural organization of C-terminal parts of fibrinogen Aα-chains. FEBS Lett 1983; 160(1):291-295. 46) Betts L, Merenbloom BK, Lord ST. The structure of fibrinogen fragment D with the 'A' knob peptide GPRVVE. J Thromb Haemost 2006; (5):1139-1141. 47) Riedel T, Suttnar J, Brynda E, Houska M, Medved L, Dyr JE. Fibrinopeptides A and B release in the process of surface fibrin formation. Blood 2011; 117(5):1700-1706. 48) Gaffney PJ. Fibrin degradation products. A review of structures found in vitro and in vivo. Ann N Y Acad Sci 2001; 936:594-610. 49) Walker JB, Nesheim ME. The molecular weights, mass distribution, chain composition, and structure of soluble fibrin degradation products released from a fibrin clot perfused with plasmin. J Biol Chem 1999; 274(8):5201-5212. 50) Weisel JW. Structure of fibrin: impact on clot stability. J Thromb Haemost 2007; 5 Suppl 1:116-124. 51) Longstaff C, Thelwell C, Williams SC, Silva MM, Szabo L, Kolev K. The interplay between tissue plasminogen activator domains and fibrin structures in the regulation of fibrinolysis: kinetic and microscopic studies. Blood 2011; 117(2):661-668. 52) Matveyev MY, Domogatsky SP. Penetration of macromolecules into contracted blood clot. Biophys J 1992; 63(3):862-863. 53) Lam WA, Chaudhuri O, Crow A, Webster KD, Li TD, Kita A, et al. Mechanics and contraction dynamics of single platelets and implications for clot stiffening. Nat Mater 2011; 10(1):61-66. 192  54) Carvalho FA, Connell S, Miltenberger-Miltenyi G, Pereira SV, Tavares A, Ariens RA, et al. Atomic force microscopy-based molecular recognition of a fibrinogen receptor on human erythrocytes. ACS Nano 2010; 4(8):4609-4620. 55) Cines DB, Lebedeva T, Nagaswami C, Hayes V, Massefski W, Litvinov RI, et al. Clot contraction: compression of erythrocytes into tightly packed polyhedra and redistribution of platelets and fibrin. Blood 2014; 123(10):1596-1603. 56) Aleman MM, Byrnes JR, Wang JG, Tran R, Lam WA, Di Paola J, et al. Factor XIII activity mediates red blood cell retention in venous thrombi. J Clin Invest 2014; 124(8):3590-3600. 57) Hethershaw EL, Cilia La Corte AL, Duval C, Ali M, Grant PJ, Ariens RA, et al. The effect of blood coagulation factor XIII on fibrin clot structure and fibrinolysis. J Thromb Haemost 2014; 12(2):197-205. 58) Ritchie H, Lawrie LC, Mosesson MW, Booth NA. Characterization of crosslinking sites in fibrinogen for plasminogen activator inhibitor 2 (PAI-2). Ann N Y Acad Sci 2001; 936:215-218. 59) He S, Blomback M, Jacobsson Ekman G, Hedner U. The role of recombinant factor VIIa (FVIIa) in fibrin structure in the absence of FVIII/FIX. J Thromb Haemost 2003; 1(6):1215-1219. 60) Wolberg AS, Allen GA, Monroe DM, Hedner U, Roberts HR, Hoffman M. High dose factor VIIa improves clot structure and stability in a model of haemophilia B. Br J Haematol 2005 ;131(5):645-655. 61) Fatah K, Hamsten A, Blombäck B, Blombäck M. Fibrin gel network characteristics and coronary heart disease: relations to plasma fibrinogen concentration, acute phase protein, serum lipoproteins, and coronary atherosclerosis. Thromb Haemost. 1992; 68:130–135. 62) Undas A, Nowakowski T, Cieœla-Dul M, Sadowski J. Altered plasma fibrin clot characteristics are associated with worse clinical outcome in patients with peripheral arterial disease and thrombangiitis obliterans. Atherosclerosis. 2011; 315:481–486. 63) Lisman T, de Groot PG, Meijers JCM, Rosendaal FR. Reduced plasma fibrinolytic potential is a risk factor for venous thrombosis. Blood. 2005; 15:1102–1105. 64) Williams S, Fatah K, Ivert T, Blombäck M. The effect of acetyl salicylic acid on fibrin gel lysis by tissue plasminogen activator. Blood Coagul Fibrinolysis. 1995; 6:718–725. 193  65) Undas A, Celinska-Lowenhoff M, Lowenhoff T, Szczeklik A. Statins, fenofibrate, and quinapril increase clot permeability and enhance fibrinolysis in patients with coronary artery disease. J Thromb Haemost. 2006; 4:1029–1036. 66) Effects on fibrin network porosity of anticoagulants with different modes of action and reversal by activated coagulation factor concentrate. Br J Haematol. 2011; 152:758–765. 67) Raskob GE, Angchaisuksiri P, Blanco AN, Buller H, Gallus A, Hunt BJ, et al. Thrombosis: a major contributor to global disease burden. Arterioscler Thromb Vasc Biol 2014; 34(11):2363-2371.  68) Global and regional mortality from 235 causes of death for 20 age groups in 1990 and 2010: a systematic analysis for the Global Burden of Disease Study 2010. The Lancet 2013; 380 (9859):2095 – 2128.  69) Wendelboe AM, Raskob GE. Global Burden of Thrombosis: Epidemiologic Aspects. Circ Res 2016; 118(9):1340-1347.  70) Mozaffarian D, Benjamin EJ, Go AS, Arnett DK, Blaha MJ, Cushman M, et al. Heart Disease and Stroke Statistics—2016 Update. Circulation 2015 Lippincott Williams & Wilkins.  71) Cohen AT, Agnelli G, Anderson FA, Arcelus JI, Bergqvist D, Brecht JG, et al. Venous thromboembolism (VTE) in Europe. The number of VTE events and associated morbidity and mortality. Thromb Haemost 2007; 98(4):756-764.  72) Heit JA, Cohen AT, Anderson FA. Estimated Annual Number of Incident and Recurrent, Non-Fatal and Fatal Venous Thromboembolism (VTE) Events in the US. Blood 2005; 106(11):910-910.  73) Previtali E, Bucciarelli P, Passamonti SM, Martinelli I. Risk factors for venous and arterial thrombosis. Blood Transfusion. 2011; (2):120-138.  74) Jackson SP. Arterial thrombosis--insidious, unpredictable and deadly. Nat Med 2011; 17(11):1423-1436. 75) Siess W, Zangl KJ, Essler M, Bauer M, Brandl R, Corrinth C, et al. Lysophosphatidic acid mediates the rapid activation of platelets and endothelial cells by mildly oxidized low density lipoprotein and accumulates in human atherosclerotic lesions. Proc Natl Acad Sci U S A 1999; 96(12):6931-6936. 194  76) Mackman N. Triggers, targets and treatments for thrombosis. Nature. 2008; 451(7181):914-8.  77) Piran S, Schulman S. Management of venous thromboembolism: an update. Thromb J. 2016; 14(Suppl 1):23.  78) Woodruff RS, Sullenger B, Becker RC. The many faces of the contact pathway and their role in thrombosis. J Thromb Thrombolysis 2011; 32(1):9-20. 79) Matafonov A, Leung PY, Gailani AE, Grach SL, Puy C, Cheng Q, et al. Factor XII inhibition reduces thrombus formation in a primate thrombosis model. Blood 2014; 123(11):1739-1746. 80) Cushman M, O'Meara ES, Folsom AR, Heckbert SR. Coagulation factors IX through XIII and the risk of future venous thrombosis: the Longitudinal Investigation of Thromboembolism Etiology. Blood 2009; 114(14):2878-2883. 81) Gailani D, Bane CE, Gruber A. Factor XI and contact activation as targets for antithrombotic therapy. J Thromb Haemost 2015; 13(8):1383-1395. 82) Woodruff RS, Xu Y, Layzer J, Wu W, Ogletree ML, Sullenger BA. Inhibiting the intrinsic pathway of coagulation with a factor XII-targeting RNA aptamer. J Thromb Haemost 2013; 11(7):1364-1373. 83) Xu Y, Cai TQ, Castriota G, Zhou Y, Hoos L, Jochnowitz N, et al. Factor XIIa inhibition by Infestin-4: in vitro mode of action and in vivo antithrombotic benefit. Thromb Haemost 2014; 111(4):694-704. 84) Geddings JE, Mackman N. New players in haemostasis and thrombosis. Thromb Haemost 2014; 111(4):570-574. 85) Caen J, Wu Q. Hageman factor, platelets and polyphosphates: early history and recent connection. J Thromb Haemost 2010; 8(8):1670-1674. 86) Morrissey JH, Choi SH, Smith SA. Polyphosphate: an ancient molecule that links platelets, coagulation, and inflammation. Blood 2012; 119(25):5972-5979. 87) Smith SA, Choi SH, Davis-Harrison R, Huyck J, Boettcher J, Rienstra CM, et al. Polyphosphate exerts differential effects on blood clotting, depending on polymer size. Blood 2010; 116(20):4353-4359. 88) Morrissey JH. Polyphosphate multi-tasks. J Thromb Haemost 2012; 10(11):2313-2314. 195  89) Smith SA, Morrissey JH. Polyphosphate enhances fibrin clot structure. Blood 2008; 112(7):2810-2816. 90) Mutch NJ, Engel R, Uitte de Willige S, Philippou H, Ariens RA. Polyphosphate modifies the fibrin network and down-regulates fibrinolysis by attenuating binding of tPA and plasminogen to fibrin. Blood 2010; 115(19):3980-3988. 91) Muller F, Mutch NJ, Schenk WA, Smith SA, Esterl L, Spronk HM, et al. Platelet polyphosphates are proinflammatory and procoagulant mediators in vivo. Cell 2009; 139(6):1143-1156. 92) Fuchs TA, Brill A, Duerschmied D, Schatzberg D, Monestier M, Myers DD,Jr, et al. Extracellular DNA traps promote thrombosis. Proc Natl Acad Sci U S A 2010; 107(36):15880-15885. 93) Demers M, Wagner DD. NETosis: a new factor in tumor progression and cancer-associated thrombosis. Semin Thromb Hemost 2014; 40(3):277-283. 94) Sorensen OE, Borregaard N. Neutrophil extracellular traps - the dark side of neutrophils. J Clin Invest 2016; 126(5):1612-1620. 95) Martinod K, Wagner DD. Thrombosis: tangled up in NETs. Blood 2014; 123(18):2768-2776. 96) Brill A, Fuchs TA, Savchenko AS, Thomas GM, Martinod K, De Meyer SF, et al. Neutrophil extracellular traps promote deep vein thrombosis in mice. J Thromb Haemost 2012; 10(1):136-144. 97) Garcia DA, Baglin TP, Weitz JI, Samama MM, American College of Chest Physicians. Parenteral anticoagulants: Antithrombotic Therapy and Prevention of Thrombosis, 9th ed: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines. Chest 2012; 141(2 Suppl):e24S-43S.  98) Shriver Z, Capila I, Venkataraman G, Sasisekharan R. Heparin and Heparan Sulfate: Analyzing Structure and Microheterogeneity. Handbook of experimental pharmacology. 2012; 207:159-176.   99) Crowther MA, Warkentin TE. Bleeding risk and the management of bleeding complications in patients undergoing anticoagulant therapy: focus on new anticoagulant agents. Blood 2008; 111(10):4871-4879.  100) Schulman S, Beyth RJ, Kearon C, Levine MN, American College of Chest Physicians. Hemorrhagic complications of anticoagulant and thrombolytic treatment: 196  American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008; 133(6 Suppl):257S-298S.  101) Alquwaizani M, Buckley L, Adams C, Fanikos J. Anticoagulants: A Review of the Pharmacology, Dosing, and Complications. Current Emergency and Hospital Medicine Reports. 2013; 1(2):83-97.   102) Hirsh J, Anand SS, Halperin JL, Fuster V, American Heart Association. Guide to anticoagulant therapy: Heparin: a statement for healthcare professionals from the American Heart Association. Circulation 2001; 103(24):2994-3018.  103) Olson ST, Richard B, Izaguirre G, Schedin-Weiss S, Gettins PGW. Molecular mechanisms of antithrombin-heparin regulation of blood clotting proteinases. a paradigm for understanding proteinase regulation by serpin family protein proteinase inhibitors. Biochimie. 2010; 92(11):1587-1596.   104) Gray E, Hogwood J, Mulloy B. The anticoagulant and antithrombotic mechanisms of heparin. Handb Exp Pharmacol 2012; 207:43-61.   105) Huntington JA, Read RJ, Carrell RW. Structure of a serpin-protease complex shows inhibition by deformation. Nature 2000 Oct 19; 407(6806):923-926.  106) Izaguirre G, Aguila S, Qi L, Swanson R, Roth R, Rezaie AR, et al. Conformational activation of antithrombin by heparin involves an altered exosite interaction with protease. J Biol Chem 2014; 289(49):34049-34064.  107) Johnson DJ, Langdown J, Huntington JA. Molecular basis of factor IXa recognition by heparin-activated antithrombin revealed by a 1.7-A structure of the ternary complex. Proc Natl Acad Sci U S A 2010; 107(2):645-650.  108) Li W, Johnson DJ, Esmon CT, Huntington JA. Structure of the antithrombin-thrombin-heparin ternary complex reveals the antithrombotic mechanism of heparin. Nat Struct Mol Biol 2004; 11(9):857-862.  109) Mosier PD, Krishnasamy C, Kellogg GE, Desai UR. On the specificity of heparin/heparan sulfate binding to proteins. Anion-binding sites on antithrombin and thrombin are fundamentally different. PLoS One 2012; 7(11):e48632.  110) Mulloy B, Hogwood J, Gray E, Lever R, Page CP. Pharmacology of Heparin and Related Drugs. Pharmacol Rev 2016; 68(1):76-141. 197   111) O'Keeffe D, Olson ST, Gasiunas N, Gallagher J, Baglin TP, Huntington JA. The heparin binding properties of heparin cofactor II suggest an antithrombin-like activation mechanism. J Biol Chem 2004; 279(48):50267-50273.  112) Tovar AM, de Mattos DA, Stelling MP, Sarcinelli-Luz BS, Nazareth RA, Mourão PA. Dermatan sulfate is the predominant antithrombotic glycosaminoglycan in vessel walls: implications for a possible physiological function of heparin cofactor II. Biochim Biophys Acta 2015; 1740:45–53.  113) He L, Vicente CP, Westrick RJ, Eitzman DT, and Tollefsen DM. Heparin cofactor II inhibits arterial thrombosis after endothelial injury. J Clin Invest 2002; 109:213–219.  114) Gandhi NS, Mancera RL. Heparin/heparan sulphate-based drugs. Drug Discov Today 2010; 15(23-24):1058-1069.  115) Bergqvist D, Arcelus JI, Felicissimo P, ETHOS investigators. Post-discharge compliance to venous thromboembolism prophylaxis in high-risk orthopaedic surgery: results from the ETHOS registry. Thromb Haemost 2012; 107(2):280-287.  116) Ansell J, Hirsh J, Hylek E, Jacobson A, Crowther M, Palareti G, et al. Pharmacology and management of the vitamin K antagonists: American College of Chest Physicians Evidence-Based Clinical Practice Guidelines (8th Edition). Chest 2008; 133(6 Suppl):160S-198S.  117) Chan NC, Eikelboom JW, Weitz JI. Evolving Treatments for Arterial and Venous Thrombosis: Role of the Direct Oral Anticoagulants. Circ Res 2016; 118(9):1409-1424.  118) Yeh CH, Fredenburgh JC, Weitz JI. Oral direct factor Xa inhibitors. Circ Res 2012; 111(8):1069-1078.  119) Lassen MR, Gallus A, Raskob GE, Pineo G, Chen D, Ramirez LM, et al. Apixaban versus enoxaparin for thromboprophylaxis after hip replacement. N Engl J Med 2010; 363(26):2487-2498.   120) Hokusai-VTE Investigators, Buller HR, Decousus H, Grosso MA, Mercuri M, Middeldorp S, et al. Edoxaban versus warfarin for the treatment of symptomatic venous thromboembolism. N Engl J Med 2013; 369(15):1406-1415.   198  121) Agnelli G, Buller HR, Cohen A, Curto M, Gallus AS, Johnson M, et al. Oral apixaban for the treatment of acute venous thromboembolism. N Engl J Med 2013; 369(9):799-808.   122) Giugliano RP, Ruff CT, Braunwald E, Murphy SA, Wiviott SD, Halperin JL, et al. Edoxaban versus warfarin in patients with atrial fibrillation. N Engl J Med 2013; 369(22):2093-2104.   123) Patel MR, Mahaffey KW, Garg J, Pan G, Singer DE, Hacke W, et al. Rivaroxaban versus warfarin in nonvalvular atrial fibrillation. N Engl J Med 2011; 365(10):883-891.   124) Beyer-Westendorf J, Lutzner J, Donath L, Radke OC, Kuhlisch E, Hartmann A, et al. Efficacy and safety of rivaroxaban or fondaparinux thromboprophylaxis in major orthopedic surgery: findings from the ORTHO-TEP registry. J Thromb Haemost 2012; 10(10):2045-2052.  125) Hylek EM, Held C, Alexander JH, Lopes RD, De Caterina R, Wojdyla DM, et al. Major bleeding in patients with atrial fibrillation receiving apixaban or warfarin: The ARISTOTLE Trial (Apixaban for Reduction in Stroke and Other Thromboembolic Events in Atrial Fibrillation): Predictors, Characteristics, and Clinical Outcomes. J Am Coll Cardiol 2014; 63(20):2141-2147 . 126) Prins MH, Lensing AW, Bauersachs R, van Bellen B, Bounameaux H, Brighton TA, et al. Oral rivaroxaban versus standard therapy for the treatment of symptomatic venous thromboembolism: a pooled analysis of the EINSTEIN-DVT and PE randomized studies. Thromb J 2013; 11(1):21-9560-11-21.  127) Davies EC, Green CF, Taylor S, Williamson PR, Mottram DR, Pirmohamed M. Adverse drug reactions in hospital in-patients: a prospective analysis of 3695 patient-episodes. PLoS One 2009; 4(2):e4439.  128) Shepherd G, Mohorn P, Yacoub K, May DW. Adverse drug reaction deaths reported in United States vital statistics, 1999-2006. Ann Pharmacother 2012; 46(2):169-175.  129) Budaj A, Eikelboom JW, Mehta SR, Afzal R, Chrolavicius S, Bassand JP, et al. Improving clinical outcomes by reducing bleeding in patients with non-ST-elevation acute coronary syndromes. Eur Heart J 2009 ;30(6):655-661 . 199  130) Eikelboom JW, Quinlan DJ, O'Donnell M. Major bleeding, mortality, and efficacy of fondaparinux in venous thromboembolism prevention trials. Circulation 2009; 120(20):2006-2011.  131) Woodman RC, Harker LA. Bleeding complications associated with cardiopulmonary bypass. Blood 1990; 76(9):1680-1697.  132) Crowther MA, Berry LR, Monagle PT, Chan AK. Mechanisms responsible for the failure of protamine to inactivate low-molecular-weight heparin. Br J Haematol 2002; 116(1):178-186.  133) Bromfield SM, Wilde E, Smith DK. Heparin sensing and binding - taking supramolecular chemistry towards clinical applications. Chem Soc Rev 2013 7; 42(23):9184-9195.  134) Horrow JC. Protamine: a review of its toxicity. Anesth Analg 1985; 64(3):348-361.  135)  Ni Ainle F, Preston RJ, Jenkins PV, Nel HJ, Johnson JA, Smith OP, et al. Protamine sulfate down-regulates thrombin generation by inhibiting factor V activation. Blood 2009; 114(8):1658-1665.  136) Porsche R, Brenner ZR. Allergy to protamine sulfate. Heart Lung 1999; 28(6):418-428.  137)  Singla A, Sullivan MJ, Lee G, Bartholomew J, Kapadia S, Aster RH, et al. Protamine-induced immune thrombocytopenia. Transfusion 2013; 53(10):2158-2163.  138) Wakefield TW, Andrews PC, Wrobleski SK, Kadell AM, Tejwani S, Hulin MS, et al. A [+18RGD] protamine variant for nontoxic and effective reversal of conventional heparin and low-molecular-weight heparin anticoagulation. J Surg Res 1996; 63(1):280-286.  139) Byun Y, Chang LC, Lee LM, Han IS, Singh VK, Yang VC. Low molecular weight protamine: a potent but nontoxic antagonist to heparin/low molecular weight protamine. ASAIO J 2000; 46(4):435-439.  140) Chang LC, Lee HF, Chung MJ, Yang VC. PEG-modified protamine with improved pharmacological/pharmaceutical properties as a potential protamine substitute: synthesis and in vitro evaluation. Bioconjug Chem 2005; 16(1):147-155.  200  141) Dehmer GJ, Lange RA, Tate DA, Pirwitz M, Daniel W, Fisher M, et al. Randomized trial of recombinant platelet factor 4 versus protamine for the reversal of heparin anticoagulation in humans. Circulation 1996; 94(9 Suppl):II347-52.  142) Mixon TA, Dehmer GJ. Recombinant platelet factor 4 for heparin neutralization. Semin Thromb Hemost 2004; 30(3):369-377.  143) Michelsen LG, Kikura M, Levy JH, Lee MK, Lee KC, Zimmermann JJ, et al. Heparinase I (neutralase) reversal of systemic anticoagulation. Anesthesiology 1996; 85(2):339-346.  144) Stafford-Smith M, Lefrak EA, Qazi AG, Welsby IJ, Barber L, Hoeft A, et al. Efficacy and safety of heparinase I versus protamine in patients undergoing coronary artery bypass grafting with and without cardiopulmonary bypass. Anesthesiology 2005; 103(2):229-240.  145) Bianchini EP, Fazavana J, Picard V, Borgel D. Development of a recombinant antithrombin variant as a potent antidote to fondaparinux and other heparin derivatives. Blood 2011; 117(6):2054-2060.  146) Schick BP, Maslow D, Moshinski A, San Antonio JD. Novel concatameric heparin-binding peptides reverse heparin and low-molecular-weight heparin anticoagulant activities in patient plasma in vitro and in rats in vivo. Blood 2004; 103(4):1356-1363.  147) Kalaska B, Kaminski K, Miklosz J, Yusa SI, Sokolowska E, Blazejczyk A, et al. Heparin-binding copolymer reverses effects of unfractionated heparin, enoxaparin, and fondaparinux in rats and mice. Transl Res 2016; 177:98-112.e10.  148) Kalaska B, Kaminski K, Sokolowska E, Czaplicki D, Kujdowicz M, Stalinska K, et al. Nonclinical evaluation of novel cationically modified polysaccharide antidotes for unfractionated heparin. PLoS One 2015; 10(3):e0119486.  149) Valimaki S, Khakalo A, Ora A, Johansson LS, Rojas OJ, Kostiainen MA. Effect of PEG-PDMAEMA Block Copolymer Architecture on Polyelectrolyte Complex Formation with Heparin. Biomacromolecules 2016; 17(9):2891-2900.  150) Lorkowska-Zawicka B, Kamiński K, Ciejka J, et al. Inactivation of Heparin by Cationically Modified Chitosan. Marine Drugs. 2014; 12(7):3953-3969.   201  151) Quinlan DJ, Eikelboom JW, Weitz JI. Four-factor prothrombin complex concentrate for urgent reversal of vitamin K antagonists in patients with major bleeding. Circulation 2013; 128(11):1179-1181.  152) Dickneite G, Hoffman M. Reversing the new oral anticoagulants with prothrombin complex concentrates (PCCs): what is the evidence? Thromb Haemost 2014; 111(2):189-198.  153)  Ansell JE, Bakhru SH, Laulicht BE, Steiner SS, Grosso M, Brown K, et al. Use of PER977 to reverse the anticoagulant effect of edoxaban. N Engl J Med 2014; 371(22):2141-2142.  154) Laulicht B, Bakhru S, Lee C, et al. Small Molecule Antidote for Anticoagulants. Circulation. 2012; 126:A11395.  155) Sullivan DW,Jr, Gad SC, Laulicht B, Bakhru S, Steiner S. Nonclinical Safety Assessment of PER977: A Small Molecule Reversal Agent for New Oral Anticoagulants and Heparins. Int J Toxicol 2015; 34(4):308-317.  156) Bakhru S, Laulicht B, Jiang X, et al. Reversal of Anticoagulant-induced Bleeding in External and Internal Bleeding Models by PER977, a Small Molecule Anticoagulant Antidote . Circulation. 2014; 130:A19361.  157) Bakhru S, Laulicht B, Jiang X, et al. A synthetic small molecule antidote for anticoagulants. European Heart Journal Aug 2013, 34 (suppl 1) 1078.  158) Costin JC, Laulicht B, Bakhru S, Steiner S. PER977 reverses low molecular weight heparin in addition to IIa and Xa new oral anticoagulants. J Am Coll Cardiol. 2015; 65(10_S).  159)  Milling TJ,Jr, Kaatz S. Preclinical and Clinical Data for Factor Xa and "Universal" Reversal Agents. Am J Med 2016; 129(11S):S80-S88.  160) Lu G, DeGuzman FR, Hollenbach SJ, Karbarz MJ, Abe K, Lee G, et al. A specific antidote for reversal of anticoagulation by direct and indirect inhibitors of coagulation factor Xa. Nat Med 2013; 19(4):446-451.  161) Crowther M, Lu G, Conley PB, et al. Reversal of factor Xa inhibitors-induced anticoagulation in healthy subjects by andexanet alfa. Crit Care Med 2014; 42 p. A1469 [Suppl., Abstract 455]. 202   162) Crowther M, Levy CG, Lu G, et al. A Phase 2 Randomized, Double-Blind, Placebo-Controlled Trial Demonstrating Reversal of Edoxaban-Induced Anticoagulation in Healthy Subjects By Andexanet Alfa (PRT064445), a Universal Antidote for Factor Xa (fXa) Inhibitors. Blood 2014 124:4269.  163) Siegal DM, Curnutte JT, Connolly SJ, Lu G, Conley PB, Wiens BL, et al. Andexanet Alfa for the Reversal of Factor Xa Inhibitor Activity. N Engl J Med 2015; 373(25):2413-2424.  164) Connolly SJ, Milling TJ,Jr, Eikelboom JW, Gibson CM, Curnutte JT, Gold A, et al. Andexanet Alfa for Acute Major Bleeding Associated with Factor Xa Inhibitors. N Engl J Med 2016; 375(12):1131-1141.  165) Shenoi RA, Kalathottukaren MT, Travers RJ, Lai BF, Creagh AL, Lange D, et al. Affinity-based design of a synthetic universal reversal agent for heparin anticoagulants. Sci Transl Med 2014; 6(260):260ra150.  166) Hollenbach S, Lu G, DeGuzman F, et al. Andexanet-alfa and PER977 (Aripazine) Correct Blood Loss in a Rabbit Liver Laceration Model - Only Andexanet Reverses Markers of FXa-mediated Anticoagulation. Circulation. 2014; 130:A14657.  167) Jain S, Pitoc GA, Holl EK, et al. Nucleic acid scavengers inhibit thrombosis without increasing bleeding. Proceedings of the National Academy of Sciences of the United States of America. 2012; 109(32):12938-12943.   168) Travers RJ, Shenoi RA, Kalathottukaren MT, Kizhakkedathu JN, Morrissey JH. Nontoxic polyphosphate inhibitors reduce thrombosis while sparing hemostasis. Blood. 2014; 124(22):3183-3190.  169) Jones CF, Campbell RA, Brooks AE, Assemi S, Tadjiki S, Thiagarajan G, et al. Cationic PAMAM dendrimers aggressively initiate blood clot formation. ACS Nano 2012; 6(11):9900-9910.  170) Rabenstein DL. Heparin and heparan sulfate: structure and function. Nat Prod Rep 2002; 19(3):312-331.  171) Viaro F, Dalio MB, Evora PR. Catastrophic cardiovascular adverse reactions to protamine are nitric oxide/cyclic guanosine monophosphate dependent and 203  endothelium mediated: should methylene blue be the treatment of choice? Chest 2002; 122(3):1061-1066.  172) Park KW. Protamine and protamine reactions. Int Anesthesiol Clin 2004; 42(3):135-145.  173) Kalathottukaren MT, Abraham L, Kapopara PR, Lai BF, Shenoi RA, Rosell FI, et al. Alteration of blood clotting and lung damage by protamine are avoided using the heparin and polyphosphate inhibitor UHRA. Blood 2017; 129 (10): 1368-1379.  174) Kannan RM, Nance E, Kannan S, Tomalia DA. Emerging concepts in dendrimer-based nanomedicine: from design principles to clinical applications. J Intern Med 2014; 276(6):579-617.  175) Lane LA, Qian X, Smith AM, Nie S. Physical chemistry of nanomedicine: understanding the complex behaviors of nanoparticles in vivo. Annu Rev Phys Chem 2015; 66:521-547.  176)  Xue F, Liu L, Mi,Y, et al. Investigation the interaction between protamine sulfate and CdTe quantum dots with spectroscopic techniques RSC Adv., 2016; 6:10215-10220.  177) Chang LC, Lee HF, Yang Z, Yang VC. Low molecular weight protamine (LMWP) as nontoxic heparin/low molecular weight heparin antidote (I): preparation and characterization. AAPS PharmSci 2001; 3(3):E17.  178) Sogawa M and Mohammed SF. Involvement of fibrinogen in protamine-induced pulmonary hypertension. J Surg Res. 1997; 73: 80-84.  179) Stewart GJ, Niewiarowski S. Nonenzymatic polymerization of fibrinogen by protamine sulfate. An electron microscope study. Biochim Biophys Acta. 1969; 194(2):462-469.  180) Protamine reversal of heparin affects platelet aggregation and activated clotting time after cardiopulmonary bypass. Anesth Analg. 1998; 87(4):781-5.  181) Duncan R. Polymer therapeutics as nanomedicines: new perspectives. Curr Opin Biotechnol 2011; 22(4):492-501.  182) Knop K, Hoogenboom R, Fischer D, Schubert US. Poly(ethylene glycol) in drug delivery: pros and cons as well as potential alternatives. Angew Chem Int Ed Engl 2010 Aug; 49(36):6288-6308. 204   183) Overbeek J.Th.G. On the electrostatic interaction in macroionic solutions and suspensions. J Chem Phys. 1987; 87: 4406-4408.  184) Ruckenstein E. Attraction between identical colloidal particles caused by collective electrostatic repulsion. Adv. Colloid Interface Sci. 1998; 75: 169-180.   185) Sogami I and Ise N. On the electrostatic interaction in macroionic solutions. J. Chem. Phys.1984; 81: 6320-6332.  186) Ohshima H. Electrostatic Interaction between Two Dissimilar Spheres with Constant Surface Charge Density. J. Colloid Interface Sci. 1995; 170: 432-439.  187) Derjaguin B. V., Landau L. Theory of the stability of strongly charged lyophobic sols and the adhesion of strongly charged particles in solutions of electrolytes Acta. Physicochim. USSR, 1941; 14:633–662.  188) Verwey E.J.W., Overbeek J.Th.G. Theory of the Stability of Lyophobic Colloids. Elsevier Press, Amsterdam. 1948; 1-203.  189) Quemada D, Berli C. Energy of interaction in colloids and its implications in rheological modeling. Adv Colloid Interface Sci. 2002; 98(1):51-85.  190) Maurer J, Haselbach S, Klein O, Baykut D, Vogel V, Mantele W. Analysis of the complex formation of heparin with protamine by light scattering and analytical ultracentrifugation: implications for blood coagulation management. J Am Chem Soc 2011; 133(4):1134-1140.  191) Mutsuhito K, Mi K L. Jerrold HL. Heparin neutralization with methylene Blue, hexadimethrine, or vancomycin after cardiopulmonary bypass. Anesth Analg 1996; 83:223-7.  192) Hughes JP, Rees S, Kalindjian SB, Philpott KL. Principles of early drug discovery. Br J Pharmacol 2011; 162(6):1239-1249.  193) Ballauff M. Spherical polyelectrolyte brushes. Progress in Polymer Sci. 2007; 32(10): 1135-1151.  205  194) Wittemann A, Ballauff M. Interaction of proteins with linear polyelectrolytes and spherical polyelectrolyte brushes in aqueous solution. Phys Chem Chem Phys 2006; 8(45):5269-5275. 195) Zhang M, Liu L, Zhao H, Yang Y, Fu G, He B. Double-responsive polymer brushes on the surface of colloid particles. J Colloid Interface Sci 2006; 301(1):85-91.  196) Ohshima H. Electrostatic Interaction between a Hard Sphere with Constant Surface Charge Density and a Soft Sphere: Polarization Effect of a Hard Sphere. J. Colloid Interface Sci 1994; 168: 255-265.  197) de Gennes P.G. Scaling Concepts in Polymer Physics. Cornell University Press, Ithaka, NY 1979.  198) Egorov SA. Insertion of nanoparticles into polymer brush under variable solvent conditions. J Chem Phys 2012; 137(13):134905.  199) Steels BM, Leermakers FA, Haynes CA. Analysis of compression of polymer mushrooms using self-consistent field theory. J Chromatogr B Biomed Sci Appl 2000; 743(1-2):31-40.  200) Steels BM, Koska J, Haynes CA. Analysis of brush-particle interactions using self-consistent-field theory. J Chromatogr B Biomed Sci Appl 2000; 743(1-2):41-56.  201) Capila I, Linhardt RJ. Heparin-protein interactions. Angew Chem Int Ed Engl 2002; 41(3):391-412.  202) Wakefield TW, Andrews PC, Wrobleski SK, Kadell AM, Fazzalari A, Nichol BJ, et al. Reversal of low-molecular-weight heparin anticoagulation by synthetic protamine analogues. J Surg Res 1994; 56(6):586-593.  203) Lima MA, de Farias EHC, Rudd TR, et al. Low molecular weight heparins: Structural differentiation by spectroscopic and multivariate approaches. Carbohydrate polymers 2011; 85(4): 903-909.  204) Guerrini M, Guglieri S, Naggi A, Sasisekharan R, Torri G. Low molecular weight heparins: structural differentiation by bidimensional nuclear magnetic resonance spectroscopy. Semin Thromb Hemost 2007; 33(5):478-487.  206  205) Bianchini P, Liverani L, Spelta F, Mascellani G, Parma B. Variability of heparins and heterogeneity of low molecular weight heparins. Semin Thromb Hemost 2007; 33(5):496-502.  206) Khorramian BA, Stivala SS. Small-angle x-ray scattering of high- and low-affinity heparin. Arch Biochem Biophys 1986; 247(2):384-392.  207) Sunder A, Hanselmann R, Frey H, Mülhaupt R. Controlled Synthesis of Hyperbranched Polyglycerols by Ring-Opening Multibranching Polymerization. Macromolecules 1999; 32 (13): 4240–4246.  208) Garamus VM, Maksimova TV, Kautz H, et al. Hyperbranched Polymers:  Structure of Hyperbranched Polyglycerol and Amphiphilic Poly(glycerol ester)s in Dilute Aqueous and Nonaqueous Solution. Macromolecules 2004; 37 (22):8394–8399.  209) Mattison KW, Dubin PL, Brittain IJ. Complex Formation between Bovine Serum Albumin and Strong Polyelectrolytes:  Effect of Polymer Charge Density. J. Phys. Chem. B 1998; 102 (19): 3830–3836.  210) Seyrek E, Dubin PL, Tribet C, Gamble EA. Ionic strength dependence of protein-polyelectrolyte interactions. Biomacromolecules 2003; 4(2):273-282.  211) Sharp K.A, Madan B. Hydrophobic effect, water structure, and heat capacity changes. J. Phys. Chem. B 1997; 101: 4343-4348.  212) Matulis D, Rouzina I, Bloomfield VA. Thermodynamics of DNA binding and condensation: isothermal titration calorimetry and electrostatic mechanism. J Mol Biol 2000; 296(4):1053-1063.  213) Sharp K. Entropy-enthalpy compensation: fact or artifact? Protein Sci 2001; 10(3):661-667.  214) Price WS. Pulsed-field gradient nuclear magnetic resonance as a tool for studying translational diffusion: Part 1. Basic theory. Concepts in Magnetic Resonance Part A 1997; 9(5):299-336.  215) Rahme K, Chen L, Hobbs RG. et al. PEGylated gold nanoparticles: polymer quantification as a function of PEG lengths and nanoparticle dimensions. RSC Adv 2013; 3: 6085-6094.  207  216) Perry JL, Reuter KG, Kai MP, Herlihy KP, Jones SW, Luft JC, et al. PEGylated PRINT nanoparticles: the impact of PEG density on protein binding, macrophage association, biodistribution, and pharmacokinetics. Nano Lett 2012 Oct 10; 12(10):5304-5310. 217) Desai UR, Petitou M, Bjork I, Olson ST. Mechanism of heparin activation of antithrombin. Role of individual residues of the pentasaccharide activating sequence in the recognition of native and activated states of antithrombin. J Biol Chem 1998; 273(13):7478-7487.  218) Berry L, Stafford A, Fredenburgh J, O'Brodovich H, Mitchell L, Weitz J, et al. Investigation of the anticoagulant mechanisms of a covalent antithrombin-heparin complex. J Biol Chem 1998; 273(52):34730-34736.  219) Olson ST, Shore JD. Binding of high affinity heparin to antithrombin III. Characterization of the protein fluorescence enhancement. J Biol Chem 1981; 256(21):11065-11072.  220) Bakchoul T, Zollner H, Amiral J, Panzer S, Selleng S, Kohlmann T, et al. Anti-protamine-heparin antibodies: incidence, clinical relevance, and pathogenesis. Blood 2013; 121(15):2821-2827.  221) Suvarna S, Espinasse B, Qi R, Lubica R, Poncz M, Cines DB, et al. Determinants of PF4/heparin immunogenicity. Blood 2007; 110(13):4253-4260.  222) van der Gucht J, Spruijt E, Lemmers M, Cohen Stuart MA. Polyelectrolyte complexes: bulk phases and colloidal systems. J Colloid Interface Sci 2011; 361(2):407-422.  223) Voets IK, de Keizer A, Cohen Stuart MA. Complex coacervate core micelles. Adv Colloid Interface Sci 2009; 147-148:300-318.  224) Levy JH, Spyropoulos AC, Samama CM, et al. Direct oral anticoagulants: new drugs and new concepts. JACC Cardiovasc Interv. 2014; 7(12):1333–1351.  225) Holbrook A, Schulman S, Witt DM, et al. Evidence-based management of anticoagulant therapy: Antithrombotic therapy and prevention of thrombosis, 9th ed: American college of chest physicians evidence-based clinical practice guidelines. Chest. 2012; 141: e152S-184S.  208  226) Hirsh J, Bauer KA, Donati MB, Gould M, Samama MM, and Weitz JI. Parenteral anticoagulants: American college of chest physicians evidence-based clinical practice guidelines (8th edition). Chest. 2008; 133: 141S–159S.  227) Baron TH, Kamath PS, and McBane RD. Management of antithrombotic therapy in patients undergoing invasive procedures. N Engl J Med. 2013; 368:2113-2124.  228) Crowther M and Crowther MA. Antidotes for novel oral anticoagulants: current status and future potential. Arterioscler Thromb Vasc Biol. 2015; 35(8):1736-45.  229) Franchini M, Bonfanti C, and Mannucci PM. Management of bleeding associated with new oral anticoagulants. Semin Thromb Hemost. 2015; 41:788–801.  230) Greinacher A, Thiele T, and Selleng K. Reversal of anticoagulants: an overview of current developments. Thromb Haemost. 2015; 113: 931–942.  231) Butterworth J, Lin YA, Prielipp RC, Bennett J, Hammon JW, and James RL. Rapid disappearance of protamine in adults undergoing cardiac operation with cardiopulmonary bypass. Ann Thorac Surg. 2002; 74: 1589-1595.  232) Sarode R, Milling TJ Jr, Refaai MA, Mangione A, Schneider A, Durn BL, and Goldstein JN. Efficacy and safety of a 4-factor prothrombin complex concentrate in patients on vitamin K antagonists presenting with major bleeding: a randomized, plasma-controlled, phase IIIb study. Circulation. 2013; 128(11): 1234-1243.   233) Pollack CV Jr, Reilly PA, Eikelboom J, et al. Idarucizumab for Dabigatran Reversal. N Engl J Med. 2015; 373: 511-520.  234) Lai BFL, Creagh AL, Janzen J, Haynes CA, Brooks DE, and Kizhakkedathu JN.  The induction of thrombus generation on nanostructured neutral polymer brush surfaces. Biomaterials. 2010; 31(26): 6710-6718.  235) Creagh AL, Koska J, Johnson PE, et al. Stability and Oligosaccharide Binding of the N1 Cellulose-Binding Domain of Cellulomonas fimi Endoglucanase CenC. Biochemistry.  1998; 37(10): 3529-3537.  236) Schneider CA, Rasband WS, and Eliceiri KW. NIH Image to ImageJ: 25 years of image analysis. Nat. Methods 2012; 9 (7): 672-675.  209  237) Lai BFL, Zou Y, Yang X, Yu X, and Kizhakkedathu JN. Abnormal blood clot formation induced by temperature responsive polymers by altered fibrin polymerization and platelet binding. Biomaterials 2014; 35: 2518-2528.   238) Kormos J, Johnson PE, Brun E, Tomme P, et al. Binding site analysis of cellulose binding domain CBDN1 from endoglucanase C of Cellulomonas fimi by site-directed mutagenesis. Biochemistry 2000; 39(30): 8844-8852.  239) Connolly SJ, Milling TJ, Eikelboom JW , et al. Andexanet Alfa for Acute Major Bleeding Associated with Factor Xa Inhibitors. N Engl J Med. 2016; Inpress.  240) Ghadimia K, Dombrowskib KE, Levya JH, and Welsbya IJ. Andexanet alfa for the reversal of Factor Xa inhibitor related anticoagulation. Expert Rev Hematol. 2016; 9(2): 115-122.   241) Klebe G. Applying thermodynamic profiling in lead finding and optimization. Nature Rev. Drug. Discov. 2015; 14: 95-110.  242) Crowther M, Gold A, Lu G, et al. ANNEXA (TM)-A PART 2: a phase 3 randomized, double-blind, placebo-controlled trial demonstrating sustained reversal of apixaban-induced anticoagulation in older subjects by andexanet alfa (PRT064445), a universal antidote for factor XA (FXA) inhibitors. J Thromb Haemost. 2015; 13: 84–85.  243) Lu G, Lin J, Curnutte JT, and Conley PB. Reversal of heparin-induced anticoagulation by andexanet alfa, a universal antidote for Factor Xa inhibitors. Blood 2015; 126(23):2329-2329.  244) Laulicht B, Bakhru S, Jiang X, et al.  Antidote for new oral anticoagulants: mechanism of action and binding specificity of PER977. J Thromb Haemost. 2013; 11(Suppl 2): 1-84.  245) Howard EL, Becker KCD, Rusconi CP, and Becker RC. Factor IXa Inhibitors as novel anticoagulants. Arterioscler Thromb Vasc Biol. 2007; 27: 722-727.  246) Lundblad RL and Roberts HR. The acceleration by polylysine of the activation of factor X by factor IXa. Thromb Res. 1982; 25(4): 319-329.  247) Lu G, Kotha J, Cardenas J, Herr M, Pandey A, Curnutte J, Conley P, Jennings L. In vitro characterization of andexanet alfa (PRT064445), a specific fXa inhibitor antidote 210  versus aripazine (PER977), a non-specific reversal agent. Circulation. 2014; 130: A18218.   248) Hu Ty, Vaidya VR, and Asirvatham SJ. Reversing anticoagulant effects of novel oral anticoagulants: role of ciraparantag, andexanet alfa and idarucizumab. Vasc Health Risk Manag. 2016; 12: 35–44.  249) Zubairova LD, Nabiullina RM, Nagaswami C, et al. Circulating microparticles alter formation, structure, and properties of fibrin clots. Sci. Rep. 2015; 5: 17611.  250) Ansell JE, Laulicht BE, Bakhru SH, et al.  Ciraparantag safely and completely reverses the anticoagulant effects of low molecular weight heparin. Thromb Res. 2016; 146: 113-116.  251) Raskob GE, Angchaisuksiri P, Blanco AN, et al. Thrombosis: A major contributor to global disease burden. Semin Thromb Hemost. 2014; 40 (07): 724-735.  252) Lu G, Lin J, Coffey G, Curnutte JT, Conley PB. Interaction of andexanet alfa, a universal antidote to fXA inhibitors, with tissue factor pathway inhibitor enhances reversal of fXA inhibitor-induced anticoagulation. J Thromb Haemost. 2015; 13: 634–635.  253) Kannemeier  C,   Shibamiya  A,  Nakazawa F, et al. Extracellular RNA constitutes a natural procoagulant cofactor in blood coagulation. Proc Natl Acad Sci USA.  2007; 104 (15) : 6388-6393.  254) Smith SA, Mutch NJ, Baskar D, Rohloff P, Docampo R, Morrissey JH. Polyphosphate modulates blood coagulation and fibrinolysis. Proc Natl Acad Sci USA. 2006;103(4):903-908.  255) Smith SA,  Choi SH, Collins JNR , et al. Inhibition of polyphosphate as a novel strategy for preventing thrombosis and inflammation. Blood. 2012; 120(26): 5103–5110.  256) Jain K, Kesharwani P, Gupta U, Jain NK. Dendrimer toxicity: Let's meet the challenge. Int J Pharm. 2010; 394(1-2):122-142.  257) Zhong D, Jiao Y, Zhang Y, et al. Effects of the gene carrier polyethyleneimines on structure and function of blood components. Biomaterials. 2013; 34: 294-305.  211  258) Nielsen VG. Protamine enhances fibrinolysis by decreasing clot strength: Role of tissue factor-initiated thrombin generation. Ann Thorac Surg .2006; 81: 1720-1727.  259) Longstaff C, Varjú I, Sótonyi P et al. Mechanical Stability and Fibrinolytic Resistance of Clots Containing Fibrin, DNA, and Histones. J Biol Chem. 2013; 288: 6946-6956.  260) Rasband WS. ImageJ. Bethesda, MD: National Institutes of Health, 1997–2008. http://rsb.info.nih.gov/ij/.  261) Jacob RE, Carson JP, Gideon KM et al. Comparison of two quantitative methods of discerning airspace enlargement in smoke –exposed mice. PLoS ONE. 2009; 4(8): 1-7.  262) Min J, Meng-Xia X, Dong Z, et al. Spectroscopic studies on the interaction of cinnamic acid and its hydroxyl derivatives with human serum albumin. J. Mol. Struct. 2004; 692:71-80.  263) Carter AM, Cymbalista CM, Spector TD and Grant PJ. Heritability of Clot Formation, Morphology, and Lysis: The EuroCLOT Study. Arterioscler Thromb Vasc Biol. 2007; 27: 2783-2789.  264) Amelot AA,  Tagzirt M, Ducouret G , Kuen RL and Le Bonniec BF. Platelet Factor 4 (CXCL4) seals blood clots by altering the structure of fibrin.  J Biol Chem. 2007; 282: 710-720.  265) Cobel-Geard and Hassouna HI. Interaction of protamine sulfate with thrombin. American Journal of Hematology.1983; 14: 227-233.  266) Schulman S and Bijsterveld NR. Anticoagulants and their reversal. Transfus Med Rev. 2007 ;21(1):37-48.  267) Radulovic V, Laffin A, Hansson KM, Backlund E, Baghaei F and Jeppsson A. Heparin and Protamine Titration Does Not Improve Haemostasis after Cardiac Surgery: A Prospective Randomized Study. PLoS One. 2015 ;10(7):e0130271.  268) Carr ME Jr and Carr SL.  At high heparin concentrations, protamine concentrations which reverse heparin anticoagulant effects are insufficient to reverse heparin anti-platelet effects. Thromb Res. 1994 ;75(6):617-30.  269) Clark CS. Lung injury after cardiopulmonary bypass. Perfusion. 2006;21:225-228. 212   270) Ng CSH, Wan S, Yim APC and Arifi AA. Pulmonary dysfunction after cardiac surgery. Chest. 2002;121:1269-1277.  271) Urdaneta F and Lobato EB. Noncardiogenic pulmonary edema associated with protamine administration during coronary artery bypass graft surgery. J. Clin. Anesth. 1999;11: 675-681.  272) Ege T, Arar C, Canbaz S, et al. The importance of aprotinin and pentoxifylline in preventing leukocyte sequestartion and lung injury caused by protamine at the end of cardiopulmonary bypass surgery. Thorac Cardiov Surg. 2004; 52: 10-15.  273) Ostuni E, Chapman RG, Holmlin ER, Takayama S and Whitesides GM. A survey of structure-property relationshiprotamine of surfaces that resist the adsorption of protein. Langmuir. 2001; 17: 5605-5620.  274) Jeon SI, Lee JH, Andrade JD, Degennes PG. Protein surface interactions in the presence of polyethylene oxide. 1. Simplified theory .J Colloid and Interface Sci. 1991; 142: 149–158.  275) Adamczyk Z, Cichocki B, Ekiel-Jezewska, Slowicka A, Wajnryb E and Wasilewska M. Fibrinogen conformations and charge in electrolyte solutions derived from DLS and dynamic viscosity measurements. J Colloid and Interface Sci. 2012; 385: 244-257.   276) Scheiner T, Jirouskova M, Nagaswami C, Coller BS and Weisel JW. A monoclonal antibody to the fibrinogen γ-chain alters fibrin clot structure and its properties by producing short, thin fibers arranged in bundles. J Thromb Haemost. 2003; 1: 2594–2602.  277) Carr ME, Powers PL, Jones MR. Effects of poloxamer 188 on the assembly, structure and dissolution of fibrin clots. Thromb Haemost. 1991; 665: 565-568.  278) Carr ME, White GC and Gabriel DA. Platelet factor 4 enhances fibrin fiber polymerization. Thromb Res. 1987; 45: 539-543.  279) Carr ME, Cromartie R and Gabriel DA. Effect of homo poly (L-amino acids) on fibrin assembly: Role of charge and molecular weight. Biochemistry. 1989; 28: 1384-1388.  213  280)  McClung WG, Clapper DL, Hu SP and Brash JL. Lysine-derivatized polyurethane as a clot lysing surface: conversion of adsorbed plasminogen to plasmin and clot lysis in vitro. Biomaterials. 2001; 22: 1919–1924.  281) Woodhouse KA, Weitz JI and Brash JL. Lysis of surface-localized fibrin clots by adsorbed plasminogen in the presence of tissue plasminogen activator. Biomaterials. 1996; 17: 75-77.  282) Butterworth J, Lin YA, Prielipp RC, Bennett J, and James RL. The pharmacokinetics and cardiovascular effects of a single intravenous dose of protamine in normal volunteers. Anesth Analg. 2002; 94: 514-522.   283) Khan NU, Wayne CK, Barker J and Strang T. The effects of protamine overdose on coagulation parameters as measured by the thrombelastograph. Eur J Anaesthesiol. 2010 27(7): 624-7.  284) Brooks JC. Noncardiogenic pulmonary edema immediately following rapid protamine administration. Ann Pharmacother. 1999; 33: 927-930.  285) Chang SW and Voelkel NF. Charge-related lung microvascular injury. Am Rev Respir Dis. 1989; 139: 534-545.   286) Arepally GM, Ortel TL. Heparin-induced thrombocytopenia. Annu Rev Med 2010; 61:77-90.  287) Smythe MA, Koerber JM, Mattson JC. The incidence of recognized heparin-induced thrombocytopenia in a large, tertiary care teaching hospital. Chest 2007; 131(6):1644-1649.  288) Warkentin TE, Sheppard JA, Sigouin CS, Kohlmann T, Eichler P, Greinacher A. Gender imbalance and risk factor interactions in heparin-induced thrombocytopenia. Blood 2006; 108(9):2937-2941.  289) Warkentin TE, Cook RJ, Marder VJ, Sheppard JA, Moore JC, Eriksson BI, et al. Anti-platelet factor 4/heparin antibodies in orthopedic surgery patients receiving antithrombotic prophylaxis with fondaparinux or enoxaparin. Blood 2005 ;106(12):3791-3796.  290) Selleng K, Selleng S, Greinacher A. Heparin-induced thrombocytopenia in intensive care patients. Semin Thromb Hemost 2008; 34(5):425-438. 214   291) Kelton JG, Arnold DM, Bates SM. Nonheparin anticoagulants for heparin-induced thrombocytopenia. N Engl J Med 2013; 368(8):737-744.   292) Kelton JG. The pathophysiology of heparin-induced thrombocytopenia: biological basis for treatment. Chest 2005; 127(2 Suppl):9S-20S.   293) Joglekar MV, Quintana Diez PM, Marcus S, Qi R, Espinasse B, Wiesner MR, et al. Disruption of PF4/H multimolecular complex formation with a minimally anticoagulant heparin (ODSH). Thromb Haemost 2012; 107(4):717-725.  294) Cines DB, Yarovoi SV,1 Zaitsev S, et al. Polyphosphate/platelet factor 4 complexes can mediate heparin-independent platelet activation in heparin-induced thrombocytopenia. Blood advances 2016; 1 (1): 62-74.                         215  Appendices Appendix A  A.1 NMR spectra of N-UHRA    Figure A.1: NMR spectrum of N-UHRA.  (A) 1H NMR spectrum. (B) 13C NMR spectrum.   A B 216   A.2 NMR spectra of mPEG750-epoxide               Figure A.2: NMR spectrum of mPEG750-epoxide. (A) 1H NMR spectrum. (B) 13C NMR spectrum. A B 217    A.3 NMR spectra of HPG-mPEG750-36 kDa    Figure A.3: NMR spectrum of HPG-mPEG750-36 kDa . (A) 1H NMR spectrum. (B) 13C NMR spectrum. A B 218    A.4 NMR spectra of mPEG750-UHRA     Figure A.4: NMR spectrum of mPEG750-UHRA. (A)1H NMR spectrum. (B) 13C NMR spectrum. AA B 219   A.5 A Cartoon of UHRA and mPEG750-UHRA molecule    Figure A.5: A cartoon depicting the UHRA and mPEG750-UHRA molecules. Both molecules contain tris(2-aminoethyl)amine ligands on the HPG core. The ligands on average present a +3 cationic charge at physiological pH and collectively bind heparins through an avidity effect. The mPEG chains in mPEG750-UHRA are longer and the core is larger when compared to the UHRA molecule.        220  A.6 GPC profile and NMR spectrum of UHRA   Figure A.6: Characterization of UHRA. (A) Gel-permeation chromatography (GPC-MALLS) elution profile of the 6-kDa HPG core in 0.1M sodium nitrate. (B) 1H-Nuclear magnetic resonance spectrum (CDCl3, 400 MHz) of UHRA showing peaks characteristic of the HPG scaffold, methylated tris(2-aminoethyl)amine ligand, and mPEG350 chains.  A B 221   A.7 Dynamic light scattering profiles of antidote-UFH complexes A (PS-UFH complexes)  B  (UHRA-UFH complexes)  C  (N-UHRA-UFH complexes)  D  (mPEG750-UHRA-UFH complexes)  Figure A.7: Intensity size distributions, volume size distributions and autocorrelation functions obtained for antidote-UFH complexes measured on a Nano ZS Zetasizer using backscatter detection. For all measurements, intercepts > 0.6 (signal-to-noise ratio) were obtained, which indicates acceptable data quality. The intensity size distributions of UHRA-UFH and LPEG-UHRA-UFH complexes show bimodal distribution. The first peak mean is around 20 nm and a second peak mean near 200 nm. The second peak may represent minuscule amounts of larger complexes, as the intensity of scattered light is proportional to the sixth power of the particle diameter. However, conversion of intensity size distribution into volume distribution (using Mie theory run by inbuilt zetasizer software) predicts mono-modal distribution, indicating that 99.9 % of particles are in the size range of 10-20 nm. The values obtained are comparable to the reported z-average hydrodynamic diameters.  222  A.8 Count rate and settings used for DLS   a Obtained from Malvern user’s manual (www.malvern.com).  Derived count rate = (Measured count rate) / (Attenuation factor)  Table A.8: Count rate, derived count rate and attenuator settings for DLS experiments. All measurements were performed in the auto-attenuation mode of the Nano ZS zetasizer. Count rates of samples were in the appropriate range of 200-500 (kcps, kilo counts per second), as suggested in the Malvern user’s manual (count rate number represents scattering intensity which is dependent on particle size and concentrations). Complexes of UHRA-UFH and LPEGUHRA-UFH were prepared by adding 500 µg/mL of each antidote into UFH (20 IU/mL). This provided count rates above 200 kcps at an attenuator setting of 11. No change in derived count rates was observed. Complexes of PS-UFH and N-UHRA–UFH were prepared by adding 62.5 µg/mL of each antidote into UFH (5 IU/mL). In these samples, a change in attenuator setting from 11 to 7 or 8 was required to maintain count rates in the optimal range. This indicates increased scattering intensity in these samples due to formation of larger particles. The high value for derived count rates confirms the presence of large aggregates. Data is the mean ± standard deviation (indicated in parentheses) from two independent experiments.       Complexes  Count rate     (kcps) Derived count rate (kcps) Attenuator  setting Attenuation Factora PS-UFH 279.8 (±31.2)  22181 (±2476) 7 0.012613446 UHRA-UFH 232.4 (±16.4)  232.4 (±16.4) 11 1 N-UHRA-UFH 216.2 (±13.6) 4915 (±310.9) 8 0.044 mPEG750-UHRA-UFH 243.2 (±15.5) 243.25 (±15.5) 11 1 223  Appendix B  B.1 NMR and mass spectrum of PER977   Figure B.1: Characterization of PER977 Structure: (A) 1H NMR spectrum of PER977.HCl (PRT066919-7) in D2O and (B) ESI-MS spectrum (positive mode) of PER977.HCl (PRT066919-7) confirming structure identity. PER977 was synthesized by Portola Pharmaceuticals Inc and kindly provided.      A B 224  Appendix C C.1 GPC profile and NMR spectrum of UHRA   Figure C.1: (A) 1H-Nuclear magnetic resonance spectrum (CDCl3, 300 MHz) of UHRA showing peaks of HPG scaffold, HBG and mPEG chains. (B) Gel-permeation chromatography (GPC-MALLS) elution profile of HPG-PEG polymer of molecular weight 25kDa in 0.1M sodium nitrate. 225   C.2 Characterization of Alexa-Fluor-488 UHRA and protamine  Figure C.2: Purification of Alexa Fluor 488 conjugated protamine by gel filtration chromatography (Sephadex G-25) and heparin neutralizing activity of Alexa Fluor 488 conjugated UHRA and protamine. A column packed with Sephadex G-25 (0.75  38 cm) was used to separate free Alexa Fluor 488 from conjugated protamine. To track the solvent front, the column was illuminated with light from a number of 488 nm LEDs and images were taken with a camera through a 520 nm long pass glass filter. (A) Image illustrating movement of a fluorescent reaction mixture (Protamine + 226  Alexa Fluor 488 succinimidyl (NHS) ester) after loading the mixture into the column. (B) Image shows the separation of protamine conjugates from free Alexa Fluor 488 molecules. (C) Alexa Fluor 488 conjugated UHRA solution was purified by dialyzing the reaction mixture (UHRA + Alexa Fluor 488 succnimidyl (NHS) ester against deionized water using a 3.5 kDa dialysis tubing for 8 h. The excitation and emission fluorescence spectra of Alexa Fluor 488 conjugated UHRA solution shows conjugation of Alexa Fluor 488 to UHRA. (D) To confirm that the Alexa Fluor 488 conjugated UHRA retain the heparin neutralization activity, the neutralization of UFH (2 IU/mL) was measured by aPTT analysis. Sodium citrate–anticoagulated plasma was incubated with UFH and titrated with various concentrations of UHRA or protamine solubilized in HEPES buffered saline (pH 7.4). The percentage of heparin neutralization was calculated from the clotting times. The anticoagulant activity of UFH was completely neutralized by Alexa 488 UHRA and protamine, and is comparable to unlabelled UHRA and protamine. This shows that conjugation of Alexa Fluor 488 did not affect the activity of UHRA and protamine.              

Cite

Citation Scheme:

        

Citations by CSL (citeproc-js)

Usage Statistics

Share

Embed

Customize your widget with the following options, then copy and paste the code below into the HTML of your page to embed this item in your website.
                        
                            <div id="ubcOpenCollectionsWidgetDisplay">
                            <script id="ubcOpenCollectionsWidget"
                            src="{[{embed.src}]}"
                            data-item="{[{embed.item}]}"
                            data-collection="{[{embed.collection}]}"
                            data-metadata="{[{embed.showMetadata}]}"
                            data-width="{[{embed.width}]}"
                            data-media="{[{embed.selectedMedia}]}"
                            async >
                            </script>
                            </div>
                        
                    
IIIF logo Our image viewer uses the IIIF 2.0 standard. To load this item in other compatible viewers, use this url:
https://iiif.library.ubc.ca/presentation/dsp.24.1-0343330/manifest

Comment

Related Items