sign in sign up
<<
Home , CD46, Platelet, Thrombin

Interactions between and complement with implications for the regulation at surfaces

Linnaeus University Dissertations No 83/2012

INTERACTIONS BETWEEN PLATELETS AND COMPLEMENT WITH IMPLICATIONS FOR THE REGULATION AT SURFACES

PER H. NILSSON

LINNAEUS UNIVERSITY PRESS

INTERACTIONS BETWEEN PLATELETS AND COMPLEMENT WITH IMPLICATIONS FOR THE REGULATION AT SURFACES. Doctoral dissertation, School of Natural Sciences, Linnaeus University 2012

Cover: Illustration of a spread on an artificial surface. Illustration by Magdalena Gajewska from an electron microscope image by the research group of Professor Shaun Jackson.

ISBN: 978-91-86983-46-8 Tryck: Ineko AB, Kållered

Abstract

Nilsson, Per, (2012). Interactions between platelets and complement with implications for the regulation at surfaces. Linnaeus University Dissertations No 83/2012. ISBN: 978-91-86983-46-8. Written in English with a summary in Swedish.

Disturbances of host integrity have the potential to evoke activation of innate immunologic and hemostatic protection mechanisms in . Irrespective of whether the activating stimulus is typically immunogenic or thrombotic, it will generally affect both the and platelets to a certain degree. The theme of this thesis is complement and platelet activity, which is intersected in all five included papers. The initial aim was to study the responses and mechanisms of the complement cascade in relation to platelet activation. The secondary aim was to use an applied approach to regulate platelets and complement on model biomaterial and cell surfaces. Complement activation was found in the fluid phase in response to platelet activation in . The mechanism was traced to platelet release of stored chondroitin sulfate-A (CS-A) and classical pathway activation via C1q. C3 was detected at the platelet surface, though its binding was independent of complement activation. The inhibitors and C4-binding (C4BP) were detected on activated platelets, and their binding was partly dependent on surface-exposed CS-A. Collectively, these results showed that platelet activation induces inflammatory complement activation in the fluid phase. CS-A was shown to be a central molecule in the complement-modulatory functions of platelets by its interaction with C1q, C4BP, and factor H. Platelet activation and surface adherence were successfully attenuated by conjugating an ADP-degrading apyrase on a model biomaterial. Only minor complement regulation was seen, and was therefore targeted specifically on surfaces and cells by co-immobilizing a factor H-binding peptide together with the apyrase. This combined approach led to a synchronized inhibition of both platelet and complement activation at the interface of biomaterials/xenogeneic cells and blood. In conclusion, here presents a novel crosstalk-mechanism for activation of complement when triggering platelets, which highlights the importance of regulating both complement and platelets to lower inflammatory events. In addition, a strategy to enhance the biocompatibility of biomaterials and cells by simultaneously targeting ADP-dependent platelet activation and the alternative complement C3-convertase is proposed.

Keywords: complement system (activation, regulation), platelets (activation, regulation), biomaterials, biocompatibility, chondroitin sulfate-A, apyrase, C1q, C3, factor H, C4BP, ADP

POPULÄRVETENSKAPLIG SAMMANFATTNING

I kroppen finns naturligt en rad försvarsmekanismer med syfte att skydda kroppens funktioner och integritet. Hot kan bland annat utgöras av infektiösa mikroorganismer, yttre trauma, men även döda eller skadade kroppsegna celler. En stor del av dessa försvarsmekanismer återfinns i blodet i form av celler och proteiner. I denna avhandling ligger fokus på två centrala skyddsmekanismer och dess samverkan: komplementsystemet och trombocyter. Komplementsystemet är evolutionärt sett ett gammalt skydd som hittas även i många primitiva organismer. Det består centralt av ett flertal proteiner vars uppgift är att identifiera, markera och eliminera allt som är främmande för kroppen, och är på så sätt involverat i den komplexa inflammationsprocessen. Trombocyter är små celler vars centrala uppgift är att i samarbete med koagulationssystemet stoppa blödning från ett skadat blodkärl. Trombocyterna fungerar som väktare och reagerar direkt vid skada genom att klumpa ihop sig och täta det skadade området. Traditionellt har komplementsytemet och trombocyter beskrivits som separata, i princip isolerade händelser, men en rad vetenskapliga arbeten har på senare tid visat att de interagerar på ett flertal punkter. De båda är syftade att snabbt och lokalt reagera på hot, och oavsett om aktiveringen primärt påverkar komplement eller trombocyter så återfinns en mer eller mindre aktivering av de båda systemen.

I denna avhandling ingår fem delprojekt vars syfte i stort var att studera komplement- och trombocytaktivering. Målet med de tre första arbetena var att undersöka hur komplementsystemet påverkas då trombocyter aktiveras. I de efterföljande två var principen att på en för mänskligt blod främmande yta försöka hämma aktiveringen av dels trombocyter, men även samtidigt hämma komplementaktivering. Idén för hur hämningen sker var hämtad från hur ett friskt blodkärl förhindrar att aktivera de båda systemen. De främmande ytorna var en plastyta, samt blodkärlsceller från gris. Dessa modeller exemplifierar två kliniskt viktiga behandlingsmetoder, dels behandling med biomaterial och dels att via blodbanan transplantera celler icke ämnade för kontakt med mänskligt blod. Biomaterial innefattar alla kroppsfrämmande material ämnade att komma i kontakt med mänsklig vävnad (ofta blod) inom sjukvården. Transplantation av celler via blodbanan sker exempelvis då bukspottkörtelceller transplanteras med syfte att behandla diabetes typ I. Både biomaterial och icke-blodceller har i kontakt med blod visat sig vara associerade med aktivering av komplementsystemet och trombocyter. Följden av detta är lidande för patienten då det föreligger risk för , blodproppar samt försämrad funktion av biomaterialet respektive minskad överlevnad av transplanterade celler.

Våra resultat visade att trombocytaktivering ger upphov till inflammatoriska signaler via aktivering av komplementsystemet. Mekanismen för aktivering kunde spåras till frisättning av kondroitinsulfat (en kolhydratbaserad molekyl) från trombocyterna ut i plasma. Komplementaktivering skedde dock inte på ytan av trombocyterna. Orsaken till detta är sannolikt att trombocyten bär på ett flertal komplementinhibitorer. I del två kunde vi, genom att binda upp ett specifikt enzym, hämma trombocytaktivering på en plastyta såväl som på celler. Motsvarande hämning av komplementsystemet sågs på samma modeller efter att specifikt rekryterat en viktig löslig komplementinhibitor. De positiva effekterna bibehölls när de två strategierna kombinerades på samma ytor.

Sammanfattningsvis bidrar denna avhandling till ökad förståelse om hur komplementsystemet och trombocyter interagerar. En helt ny mekanism för hur trombocyter ger upphov till inflammatorisk komplementaktivering beskrivs. Vidare visar vi hur det är möjligt att minska dessa reaktioner, på för blodet främmande ytor, genom att inhibera komplement- och trombocytaktivering via två mekanismer direkt hämtade från våra blodkärl.

LIST OF PUBLICATIONS

This thesis is based on the following articles, which will be referred to in the text by their roman numerals:

I. Hamad, O. A., Ekdahl, K. N., Nilsson, P. H., Andersson, J., Magotti, P., Lambris, J. D., and Nilsson, B. (2008) Complement activation triggered by chondroitin sulfate released by receptor-activated platelets. Journal of and Haemostasis 6, 1413-1421

II. Hamad, O. A., Nilsson, P. H., Wouters, D., Lambris, J. D., Ekdahl, K. N., and Nilsson, B. (2010) Complement component C3 binds to activated normal platelets without preceding proteolytic activation and promotes binding to 1. The Journal of 184, 2686-2692

III. Hamad, O. A*., Nilsson, P. H*., Lasaosa, M., Ricklin, D., Lambris, J. D., Nilsson, B., and Ekdahl, K. N. (2010) Contribution of chondroitin sulfate A to the binding of complement to activated platelets. PLoS One 5, e12889. * Authors contributed equally

IV. Nilsson, P. H., Engberg, A. E., Bäck, J., Faxälv, L., Lindahl, T. L., Nilsson, B., and Ekdahl, K. N. (2010) The creation of an surface by apyrase immobilization. Biomaterials 31, 4484-4491

V. Nilsson, P. H., Ekdahl, K. N., Magnusson, P. U., Ricklin, D., Iwata, H., Qu, H., Lambris, J. D., Hong, J., Nilsson, B., and Teramura, Y. Autoregulation of innate immune attacks on biomaterials and cells to be used in therapeutic medicine. Manuscript

Papers I-IV are reprinted with permission from the publishers. Additional work outside the scope of this thesis

1. Ekdahl, K. N., Lambris, J. D., Elwing, H., Ricklin, D., Nilsson, P. H., Teramura, Y., Nicholls, I. A., and Nilsson, B. (2011) Innate immunity activation on biomaterial surfaces: a mechanistic model and coping strategies. Advanced Delivery Reviews 63, 1042-1050

2. Hamad, O. A., Bäck, J., Nilsson, P. H., Nilsson, B., and Ekdahl, K. N. (2012) Platelets, complement, and contact activation: partners in inflammation and thrombosis. Advances in Experimental Medicine and Biology 946, 185-205

ABBREVIATIONS

ADP aHUS atypical hemolytic uremic syndrome AMP adenosine monophosphate AP alternative pathway of complement AT ATP C1INH C1-inhibitor C3aR receptor C3H2O hydrolyzed C3 C4BP C4b-binding protein C5aR (CD88) CD40L CD40 ligand CFHR complement factor H-related CP classical pathway of complement CR complement receptor CRIg complement receptor of the immunoglobulin family CRP C-reactive protein CS chondroitin sulfate DAF decay acceleration factor (CD55) DAMP danger associated molecular pattern DS ELISA -linked immunosorbent ESI electrospray ionization GAG glycosaminoglycan GP HS heparan sulfate HSA human albumin IC immune complex ICAM-1 intercellular adhesion molecule-1 (CD54) Ig immunoglobulin ITP immune thrombocytopenic LP of complement MAC membrane attack complex MALDI matrix-assisted laser desorption ionization MASP MBL-associated serine MBL mannose-binding lectin MCP membrane protein (CD46) MCP-1 chemotactic protein-1 MS mass spectrometry NO PAEC porcine arterial endothelial cell PAR protease-activated receptor PBS phosphate buffered saline PEG polyethylene glycol PF4 PNH paroxysmal nocturnal hemoglobinuria PPP platelet-poor plasma PRP platelet-rich plasma PSGL-1 P- glycoprotein ligand-1 (CD162) RANTES regulated upon activation, normal T-cell expressed, and secreted sC5b-9 soluble terminal complement complex SCR short consensus repeat SDS sodium dodecyl sulfate SDS-PAGE SDS polyacrylamide gel electrophoresis inhibitor SLE systemic lupus erythematosus SPR surface plasmon resonance TAT thrombin antithrombin complex TF (CD142) TRAP thrombin receptor-activating peptide TXA2 A2 VCAM-1 vascular -1 (CD106) vWF

TABLE OF CONTENTS

1. INTRODUCTION...... 9

2. THE COMPLEMENT SYSTEM ...... 10 2.1 Overview of the complement system...... 10 2.2 Complement activation ...... 11 2.3 Complement effector phase...... 12 2.4 Complement C1q ...... 14 2.5 Complement C3...... 15 2.6 Receptors involved in the complement response...... 16 2.7 Complement regulators ...... 17 2.8 Complement factor H and related proteins ...... 19 2.9 Complement C4b-binding protein...... 20

3. PLATELETS ...... 21 3.1 Overview of platelets ...... 21 3.2 Platelet production and morphology...... 21 3.3 Platelet activation ...... 22 3.4 Effects of platelet activation...... 24 3.5 Platelet regulation...... 25

4. ...... 27 4.1 The coagulation cascade...... 27 4.2 The tissue factor pathway...... 27 4.3 Contact activation ...... 28 4.4 Coagulation control ...... 28

5. PLATELET AND COMPLEMENT INTERACTIONS ...... 30

6. PROTEOGLYCANS AND GLYCOSAMINOGLYCANS...... 32 6.1 Overview of proteoglycans and glycosaminoglycans ...... 32 6.2 Serglycin ...... 34

7. ADP AND APYRASES...... 36 7.1 ADP in platelet activation ...... 36 7.2 Apyrases ...... 37

8. BLOOD-MATERIAL INTERACTIONS ...... 38 8.1 Biomaterial definition ...... 38 8.2 Biomaterials for blood-contact applications ...... 38 8.3 Initial exposure and activation of cascade systems ...... 39 8.4 Cell activation ...... 40 8.5 Modification strategies ...... 40 8.5.1 Material design...... 40 8.5.2 coating...... 41 8.5.3 Autoregulation...... 41

9. CELL THERAPY ...... 42 9.1 Cell therapy and instant-blood-mediated inflammatory reactions ...... 42 9.2 Strategies to increase cell transplant survival ...... 43

10. AIMS AND OBJECTIVES ...... 44

11. METHODOLOGICAL CONSIDERATIONS...... 45 11.1 Reagents...... 45 11.1.1 Blood sampling...... 45 11.1.2 Platelet preparation and activation...... 46 11.1.3 Complement control ...... 46 11.1.4 Other reagents ...... 46 11.2 Central methodology and techniques ...... 47 11.2.1 Quantification of activation markers and protein binding ...... 47 11.2.2 Immobilization of biomolecules ...... 47 11.2.3 Biomaterial testing...... 48 11.2.4 Affinity chromatography and protein identification...... 48 11.2.5 Surface plasmon resonance ...... 48

12. RESULTS...... 49 12.1 CS-A released from activated platelets triggers activation of the CP of complement (paper I) ...... 49 12.2 Binding of C3 to activated platelets occurs independently of complement activation (paper II)...... 50 12.3 CS-A links C1q, C4BP, and factor H to the platelet surface (paper III)...... 51 12.4 Apyrase immobilization lowers the thrombogenecity of a model biomaterial (paper IV)...... 52 12.5 Enhanced overall hemocompatibility of materials and cells with 5C6 and apyrase (Paper V) ...... 53

13. GENERAL DISCUSSION ...... 54

14. CONCLUDING REMARKS...... 61

15. ACKNOWLEDGEMENTS ...... 63

16. REFERENCES...... 66 1. INTRODUCTION

Under resting conditions, approximately five litres of blood are pumped through the human heart into the vascular system every minute. This process is essential to support human tissues with oxygen and nutrients and to discharge their waste products. Within the complex fluid called blood are a number of cells, proteins, and other molecules that are important for maintaining vital functions. Approximately 55 % of the blood volume is plasma, and the remaining fraction is composed of cells, of which more than 99 % are erythrocytes and the rest leukocytes and platelets.

A large fraction of the blood cells and proteins are associated with innate mechanisms that aim to counteract dangers that might damage the vascular system as well as overall host integrity. These dangers include invading microorganisms, damaged self-cells, and trauma affecting the integrity of the vascular wall. These mechanisms have been classified into the innate immunity and hemostatic systems and are traditionally described separately; however, a growing body of evidence has begun to describe intense crosstalk between the two systems.

The focus of this thesis is the biology and interaction between the complement system and the blood platelets. The complement system, also known as the complement cascade, is a part of our innate and has traditionally been viewed as being important for combating invading microorganisms. The platelet is the central cell of the hemostatic system and is also involved in inflammatory events. Together with coagulation proteins and inflammatory cells, this thesis has explored complement and platelet interactions in both the physiological situations and in situations involving contact with biomaterial surfaces.

9 2. THE COMPLEMENT SYSTEM

2.1 Overview of the complement system

The complement system was discovered in the late 1890s and was described as a component that complements the lytic effect of . Today, almost 50 complement proteins have been described. Our appreciation of the function of the cascade has evolved, and numerous interaction points with other molecular pathways have been discovered [1].

The complement system is a central part of our and represents a powerful first line of defense against microbes, as well as against damaged endogenous cells. By recognizing and tagging surfaces exposing danger-associated molecular patterns (DAMPs) (e.g., , cellular debris and immune complexes [ICs]), it can efficiently eliminate unwanted particles. In addition to its immunological function, the complement system has also been shown to participate in other processes, such as synapse maturation [2], tissue regeneration [3] and angiogenesis [4].

The core of the complement system is a proteolytic cascade that begins with pattern recognition molecules binding to DAMPs and the sequential activation of zymogens. Activation of only a few individual molecules can give rise to a sufficient response, thanks to a powerful amplification process built into the system [5]. Throughout the cascade, complement proteins are cleaved to produce activated fragments that directly or indirectly lead to the elimination of particular activating agents. Among the important effects of complement activation are: opsonization for , generation of anaphylactic/chemotactic peptides, lysis of pathogens, and enhancement of adaptive immunity.

10 2.2 Complement activation

The complement system can be activated via three pathways: the classical, lectin, and alternative pathways, all of which form enzymatic complexes (convertases) that cleave C3 and C5, the central components of the system (see Fig. 1).

The classical pathway (CP) is, in general, activated by clusters of IgG or IgM, whose Fc regions are recognized by the recognition molecule C1q. Upon C1q binding, a mechanical stress occurs that leads to the activation of C1r, which subsequently cleaves C1s. C1r and C1s are serine associated with C1q in a -dependent fashion [6]. C1s cleaves C4 into C4b and , with C4b then covalently binding adjacent to the activation site and C4a diffusing into the fluid phase. C2 associates with C4b and is cleaved by C1s into C2a and C2b, leading to the formation of the CP C3 convertase, C4b2a. The C3 convertase cleaves C3 into C3a and . Like C4b, C3b also binds covalently to the surface close to the activation site. Like C4a, C3a is released into the fluid phase, but unlike C4a, it acts as a potent anaphylatoxin.

The lectin pathway (LP) begins with either mannose-binding lectin (MBL) or ficolins, which typically bind to carbohydrates associated with microbes. MBL-associated serine proteases (MASPs) share structural similarities with C1r/C1s and are also associated with MBL/ficolins, by analogy to the C1r/C1s connection to C1q [7]. LP activation is mainly associated with MASP-2, which once activated has been shown to cleave both C4 and C2, generating a CP C3 convertase [8].

Covalently surface-bound C3b, generated by either the CP or the LP, triggers the alternative pathway (AP). Once C3b is attached, it binds factor B, which is cleaved into Bb and Ba by , resulting in the formation of the AP C3 convertase, C3bBb. Every AP C3 convertase molecule can form several new C3 convertases as a result of the deposition of C3b. In this way, the AP constitutes an important amplification loop, multiplying the complement activation initiated by the CP or LP. The AP can also be activated independently of CP and LP by so-called “C3 -over”: C3 is continuously hydrolyzed at a slow rate in the fluid phase. Hydrolyzed C3 is termed C3H2O. Fluid-phase C3H2O, but not C3, can bind factor B, which is then cleaved by factor D as described above, resulting in formation of a soluble AP C3 convertase, C3H2OBb [9]. This convertase cleaves C3 in the fluid phase and generates C3b, which can bind to surfaces and initiate activation of the AP. Endogenous cells carry regulatory molecules that degrade deposited C3b immediately, but the activation is instead amplified on foreign cells. In this way, the AP is only activated on surfaces lacking inhibitory molecules. The

11 AP also includes a pattern recognition molecule called . Properdin binds DAMPs on pathogenic and damaged cells. Once bound, it binds C3b and follows the same sequence described above for factors B and D [10]. Properdin can also stabilize an already formed C3bBb complex by forming C3BbP, which has up to a 10-fold higher stability than that of C3Bb [11].

2.3 Complement effector phase

The effect phase of complement activation begins once C3 is cleaved to C3a and C3b. The small anaphylactic fragment C3a is released into the fluid phase, and C3b is deposited at the cell surface in the vicinity of the convertase. The anaphylactic C3a triggers inflammation by binding to the (C3aR), expressed on many cells. C3b has a dual function, acting as an opsonising agent for phagocytosis and also fueling activation by assembling new AP C3 convertases. As complement activation proceeds, the density of C3b increases at the surface, mediating a convertase substrate shift from C3 to C5 [12]. The assembly and function of the C3 convertase are well known from structural characterizations [13], but the mechanism by which the substrate-specificity shift occurs, as well as the structural mechanisms of the C5 convertase, have not yet been clarified.

Like the C3 convertase, the C5 convertase cleaves C5 into C5a and C5b. C5a is a potent anaphylatoxin that recruits and activates leukocytes by binding to receptors (C5aR, C5aL2) at the surface of a variety of cells [14]. C5b forms a complex with C6 and C7 that is inserted into the plasma membrane. C8 and multiple units of C9 are subsequently inserted through the membrane to form a lytic pore, C5b-9 (also called membrane attack complex [MAC]) [15]. The formation of the MAC is one of the effects that is not mediated by receptor interactions and is particularly important in the host defense against Neisseria species. [16]. MAC insertion at sublytic levels in nucleated host cells is also important for signaling events [17]. The activity of C3a/C4a (but not C5a) is another effect that involves membrane destabilization and does not require receptor interaction; C3a and C4a are believed to kill microbes by destabilizing the microbial membrane [18].

12

Fig. 1. Complement activation, amplification and effect on a microbial cell. Complement is activated by one of the three activation pathways, starting with binding of C1 (CP), MBL (LP) or C3b/C3H2O (AP), which all are combined in the formation of a C3 convertase (C4bC2a/C3bBbP). Once C3b is bound, the AP C3 convertase amplification loop amplifies the response. Increased surface density of C3b switches the substrate-specificity of corresponding convertase towards C5 (C4bC2aC3b/C3bC3bBbP). Elimination is undertaken by insertion of MAC through the membrane of the microbial cell, by recruiting (C5a) and activating (C3a/C5a) inflammatory cells that phagocytose the C3b/iC3b opsonised cell. Activation of AP with properdin as the recognition molecule is not shown in this illustration.

13 2.4 Complement C1q

C1q, the pattern recognition molecule of the classical pathway, is crucial for the efficient clearance of pathogens and apoptotic cells [19]. By utilizing antigen-bound IgG and IgM as its major ligands, C1q indirectly recognizes a broader range of DAMPs than is typical of other pattern recognition molecules. This association with Ig also represents a major connection between the innate and the adaptive immune systems. A fraction of C1q is present in plasma in free form, but the majority forms a calcium-dependent together with C1r and C1s [20]. In addition to antigen-bound IgG and IgM, C1q interacts with several other molecules, including C-reactive protein (CRP) [21], molecular patterns (lipopolysaccharide) [22], abnormal endogenous proteins (, β-amyloid fibers) [23], structures associated with cell death (phosphatidylserine, DNA) [24,25], and glycosaminoglycans (GAGs) / proteoglycans (heparin and chondroitin sulfate proteoglycans) [26,27,28,29]. Specific receptors for C1q have also been described [30,31,32,33].

The molecular weight of C1q is 460 kDa, distributed in hexametric structure that is assembled in a way that resembles a bouquet of tulips, with a gathered N-terminal -like region forming the stalk and the outspread C- terminal globular heads forming the flowers. Each hexamer is composed of three chains, A, B, and C; they are all represented in the six globular heads, which are the major ligand-recognition region mediating avidity for its ligand [34]. It is not exactly known how the globular heads interact with ligands, although a charge recognition pattern has been proposed [35]. However, it is clear that each chain in the head has a different electrostatic potential, with A and C having a combination of basic and acidic residues and B a predominantly positive charge. Different but overlapping regions at the globular head have been suggested to mediate binding to IgG and IgM.

Regarding the role of C1q in removing IC and apoptotic cells, several questions remain to be answered. Binding of C1q to apoptotic cells has been shown to activate complement, but C1q binding may instead clear apoptotic cells via opsonization. C1q has also been discussed in the context of immunological tolerance, in particular regarding the body’s ability to avoid developing against self-structures [36]. A deficiency of C1q is correlated with an inability to eliminate apoptotic cells and with the development of the autoimmune diseases such as systemic lupus erythematosus (SLE) [37] and glomerulonephritis [38].

14

2.5 Complement C3

Present at approximately 1.3 mg/mL in plasma, (C3) is the most abundant complement protein in blood. C3, together with factor B and MASP, is thought to be among the most ancient complement proteins [39]. By displaying a variety of conformations and fragments, it exerts multiple functions in the cascade, including AP activation, convertase formation, opsonization (C3b/iC3b), activation of inflammatory cells (C3a), and B-cell stimulation (C3d).

Recent crystal structure determination of native C3, C3c [40], and C3b in complex with factor B and D [13] has led to new insights into this complex component. Native C3 is a rather inert molecule, consisting of a 110-kDa alpha-chain and a 75-kDa beta-chain held together with a disulfide bond. Upon proteolytic cleavage by a C3 convertase, C3a (9k-Da) is split from the alpha-chain. The remaining C3b undergoes a major conformational change that leads to the exposure of a previously hidden reactive thioester. In the vicinity of a surface, this thioester mediates the covalent attachment of C3 to hydroxyl groups or amines, whereas in the absence of an accepting surface, it hydrolyzes within fractions of a millisecond [41]. Surface-bound C3b is able to interact with factor B in an AP C3 convertase as well as with complement receptor (CR) 1, 3, and 4. Following surface attachment, C3b is prone to inactivation by factor I, in which case the alpha-chain is cleaved into iC3b, which has even higher affinity for the important CR3 phagocytic receptor. iC3b can then be further degraded to C3c/C3dg and C3d/C3g [42]. In addition to being proteolytically activated, C3 can adopt a C3b-like conformation through the spontaneous hydrolysis of the thioester of fluid- phase C3. C3H2O still retains C3a in the alpha-chain but has C3b-like functions, including AP C3 convertase formation and receptor interactions [41].

15 2.6 Receptors involved in the complement response

Most effects from complement activation require interaction with specific receptors (see Fig. 1). Apart from initiating CP activation, C1q exerts its effects by direct binding to receptors, including gC1qR/p33 [31], cC1qR [30], C1qRp (CD93) [33], and CR1 [32]. gC1qR/p33 binds C1q via its globular heads, whereas the three other receptors bind to the collagenous stalk. The physiologic significance of each of these receptors is not known, but they are known to participate in diverse extra- and intracellular events [43].

A major role of the complement system is to mark pathogens and cell debris for phagocytosis by tagging them with C3b. CR1, 3, and 4, together with the complement receptor of the immunoglobulin family (CRIg), are important receptors for binding opsonised particles. CR1 (CD35) is expressed on most blood cells, CR3 (CD11b/CD18) and CR4 (CD11c/CD18) on phagocytic cells, and CRIg on certain tissue . By means of these receptors, opsonized particles are recognized, engulfed, and eliminated by phagocytosis. Erythrocytes, which are incapable of phagocytosis, clear immune complexes (IC) from the bloodstream by binding C3b-tagged particles via CR1 and transporting them to the and for removal and degradation by tissue macrophages [44]. CR3 is the main on and , and its expression increases following complement activation and cell stimulation via the C5aR (CD88) [45]. CR3 binds iC3b, but not C3b [46]. CR2 (CD21) is expressed on B- and shares structural similarities with CR1; CR2 binds C3d with the highest affinity and acts as a co-receptor for B-cell activation, thereby enhancing responsiveness toward antigens [47].

The potent proinflammatory effect of the complement system is mediated via binding of the anaphylatoxins C3a to C3aR and C5a to the C5aR and C5L2. Via corresponding receptors, they attract (C5a) and activate (C3a and C5a) leukocytes at the inflammatory site [48]. The deleterious inflammation of an overactive or inadequately regulated complement system is to a large degree mediated via the proinflammatory effect of C3a and C5a [49].

16 2.7 Complement regulators

The complement response is considered to act rapidly and locally. The complement system is activated within seconds of exposure to a pathogen and normally acts in a limited space. This activation is controlled by complement regulators that are present in both the fluid phase and on healthy endogenous cells (see Fig. 2). The complement system is a potent inflammatory cascade, and inadequate regulation can generate a high concentration of anaphylatoxins and cause consumption of complement proteins and reactivity against self. Loss-of-function mutations in coding for complement regulators are frequently correlated with inflammatory disorders.

The most-characterized regulator of the initial stages of the CP and LP is the C1-inhibitor (C1INH), although others have been identified [50]. C1INH is a serine protease inhibitor that inhibits C1r/C1s as well as the MASPs [51]. C1INH becomes activated by the serine protease, leading to covalent attachment of C1INH and inactivation of the enzyme. Convertase formation is a key target for complement regulation, and several regulators are focused on this process. Regulators control C3 convertase through two different approaches, either by competitive binding/displacement of factor B/Bb from the C3 convertase (decay acceleration), or by serving as a cofactor for the degradation of C3b by the proteolytic protein factor I. Both C4b-binding protein (C4BP) and factor H have these effects and are important regulators at this level, C4BP on the CP convertase and factor H on the AP convertase. Both C4BP and factor H are fluid-phase regulators recruited to endogenous cells by their interaction with surface patterns (e.g., GAGs) and sialic acid [52,53]. Certain microbes have taken advantage of this principle to recruit host fluid-phase regulators to their surfaces in order to evade complement recognition and activation [54].

Under physiological conditions, host cells are also covered by different subsets of membrane regulators. Three of them act on the convertase level: CR1, decay-acceleration factor (DAF) (CD55), and membrane cofactor protein (MCP) (CD46) [55]. CR1 has both decay acceleration and cofactor activity, whereas DAF has only decay acceleration activity, and MCP has only cofactor activity. Regulation of the terminal pathway occurs at the assembly of MAC by the membrane bound CD59 and the fluid-phase regulators and clusterin [17]. They all prevent the insertion of the lytic complex or promote its release from the plasma membrane. The effect of the potent anaphylatoxins C3a and C5a is kept under control by the fact that their half-lives are limited by the action of carboxypeptidase N. This protease cleaves the C-terminal from the anaphylatoxins, thereby preventing the binding of C3a to C3aR and decreasing the affinity of the C5aR for C5a [1].

17

carboxypeptidase N

C3a C4BP factor H C5a

CR1

C1INH clusterin vitronectin factor I factor I C2a DAF MCP C3b Bb C3b iC3b C3d C4b P C4b cofactor activity decay acceleration

CD59

Fig.2. Complement regulation by soluble and membrane bound regulators at various stages on host cells. The soluble serpin C1INH acts on C1r/C1s and MASPs in the initial stage to limit activation via the CP and LP. The C3 convertase is the major site for inhibition excerted through decay acceleration (DAF, C4BP, factor H) and cofactor activity (MCP, C4BP, factor H, CR1) for factor I in the degradation of C3b. CD59, vitronectin and clusterin act in the later stages by preventing the formation of MAC. The half-life of C3a and C5a is limited by the action of carboxypeptidase N.

18

2.8 Complement factor H and related proteins

Factor H is a crucial regulator of the AP of complement. This 150-kDa molecule is present in at 0.3-0.8 mg/mL and is therefore one of the most abundant soluble complement proteins. Like many other complement proteins, factor H is primarily produced in the liver. Individuals with inadequate factor H function are predisposed to developing inflammatory disorders because of insufficient complement regulation [56]. The molecular structure of factor H consists of 20 individually folded domains called short consensus repeats (SCR) that are arranged one after one another in a “string of beads” structure [57]. Other complement regulators possess SCRs, including C4BP and CR1. Each SCR contains about 60 amino acids and forms several subregions with different functions. Factor H is recruited to the surface of host cells through its attraction to GAGs and sialic acid [53]. It controls complement activation by interfering with the factor B-binding portion of C3b/C3H2O and by serving as a cofactor for the degradation of C3b to iC3b, the objective being to stop the AP C3 convertase.

Factor H contains two major functional subregions, one at each end of the molecule: the N-terminal SCRs 1-4 have regulatory activity, and C-terminal SCRs 19-20 are important for surface binding. SCRs 1-4 have affinity for C3b [58], and SCRs 19-20 for C3b/C3d and cell-surface GAGs [59,60]. SCRs 19-20 are known to be an important heparin-binding region, and SCRs 7 [61] and 9 [62] also bind heparin. Heparin binding can be misinterpreted as equal to cell-surface GAG binding, but this correlation has to be carefully considered, since different GAGs have different protein-binding capacities [63], mainly because of differences in their sulfation patterns [64]. In addition to the heparin-binding domains in SCRs 7 and 9, the broad middle region shows an affinity for C3b/C3c [65]. The overall role of this middle region may be to fold factor H into a backbend hairpin structure to organize the elongated structure into a more condensed projection, and thereby optimize the interaction with C3b and the cell-surface GAGs [66].

A number of other molecules closely related to factor H are also present in plasma, in lesser amounts: factor H-related (CFHR) proteins 1-5 and factor H-like protein 1. Factor H-like protein 1 contains SCRs 1-4 of factor H and shares its ancestor factor H’s regulating activity; in contrast, the CFHR proteins lack this region. The exact function of these proteins are not yet identified even if some are implicated in complement regulatory activities [67].

19

2.9 Complement C4b-binding protein

By analogy to factor H’s regulation of C3 in the AP convertase, C4BP is the regulator of C4 in the CP convertase. C4BP regulates the convertase by the same principle of decay acceleration of C4b and C2a and acts as a cofactor for factor I in degradating C4b. Binding to C4b is also a prerequisite for cofactor activity [68]. C4BP, with a molecular size of 570 kDa, is mainly produced in the liver and circulates at a plasma concentration of approximately 0.2 mg/mL. C4BP has a polymeric structure consisting of six to eight identical α-chains and one smaller β-chain, which are composed of seven and three SCRs, respectively. The chains are held together with disulfide bridges. The β-unit has high affinity for the K vitamin-dependent , causing C4BP to circulate in complex with protein S [69]. Protein S can in turn direct C4BP to cell membranes because of its affinity for negatively charged [70]. C4b is a major ligand for binding C4BP to surfaces, even though this large molecule also possesses affinity for C3b [52] and heparin [71]. The interaction sites for these molecules are located in the α-chains of C4BP; thus, the overall binding affinity is based on avidity.

20 3. PLATELETS

3.1 Overview of platelets

Platelets are small cells of great structural and functional complexity. Although sometimes not even referred to as cells, these anucleated, membrane-enclosed discoid structures are essential for maintaining cellular . They have also been shown to be a central component of inflammatory events [72] and to be involved in angiogenesis [73]. Hemostasis is the physiological process of controlling blood fluidity, preventing blood loss after vessel damage and resolving clots after healing. Platelets are a central player in primary hemostasis, which aims to stop by forming an initial . Platelets also promote secondary hemostasis by promoting the coagulation cascade, which leads to the formation of a more stable hemostatic plug. Primary hemostasis is initiated immediately upon vessel damage and continues until the coagulation cascade has been activated to secure the damage by means of a network. Platelets are therefore essential to life but can also be detrimental if activation is uncontrolled in time and space, with regard to both their thrombotic and inflammatory functions.

3.2 Platelet production and morphology

Platelets are produced by cytoplasmic extension and budding from in the marrow via an intermediate, the pro-platelets. Pro- platelet formation is thought to be regulated by the interaction of megakaryocytes with adhesion molecules in the microenvironment, but the details are largely unknown [74]. Approximately 100x109 platelets are produced and released into the circulation each day, maintaining a normal platelet count of 150x109 to 400x109 cells per liter of

21 blood [75]. The resting platelet has a discoid shape with a smooth surface exposing a sensitive “danger-sensing” surface that responds to stress. Upon activation, a dramatic change in platelet shape occurs, involving intracellular contraction and pseudopod protrusion. Simultaneously, intracellular stored granules release their content into the fluid phase. In essence, three kinds of granula are released: alpha-granules containing various proteins, dense granules (with e.g., ATP/ADP, calcium), and containing . secretion is a gradual response, and the extent and rate of secretion is dependent on the potency of the activating agent [76,77]. Many of the functions of platelets are dependent on release of these intracellular stored molecules.

3.3 Platelet activation

The endothelial cells lining the blood vessels expose surfaces covered with anti-thrombotic molecules, including membrane proteins and GAGs, to the lumen of the blood vessels. These surfaces are the perfect blood-compatible surface for keeping platelets in their resting state. Every other surface has a certain risk of activating platelets; this activation is a problem in clinical situations involving treatment with artificial materials that are in contact with blood [78]. Furthermore, injured vessels are highly activating surfaces, with subendothelial matrix molecules such as collagen becoming exposed to the bloodstream and thereby initiating a series of events in platelets (see Fig. 3).

First, collagen binds a protein called von Willebrand factor (vWF), which in turn recruits platelets by interacting with the glycoprotein (GP)Ib-IX-V complex on the platelet surface. This initial link brings platelets to the site of injury, decreases their velocity, and makes them adhere [79]. A more firm adherence of platelets occurs in the next step, in which GPVI interacts with collagen [80]. This interaction causes the platelets to change shape, spread along the subendothelium, and release stored granules [81]. In particular, ADP released from dense granules and thrombin generated at the activated platelet membrane, act through a paracrine feedback loop to further activate platelets in the vicinity. ADP is predicted to be the major amplifier of initial platelet activation [82]. Platelet integrin GPIa/IIa and GPIIb/IIIa become activated, with GPIa/IIa binding collagen with high affinity and GPIIb/IIIa binding [80]. Collagen binding via GPIa/IIa induces a sustained increase in cytosolic calcium, which causes a redistribution of the platelet membrane known as membrane flip-flop. In this process, negatively charged phospholipids, including phosphatidylserine, in the inner leaflet of the platelet membrane become exposed on the outer leaflet [83]. Exposure of negatively

22 charged phosphatidylserine due to membrane flip-flop is a gain for effective thrombin generation in that it promotes the assembly of coagulation factors (F) that cleave prothrombin to active thrombin. Thrombin is a strong platelet activator, binding to proteinase-activated receptors (PARs) 1 and 4 and cleaving an ectodomain of the receptors, leaving a new amino terminus that act as a tethered ligand for autoactivation. PAR 1 responds rapidly to low concentrations of thrombin; PAR 4 requires 10-fold higher concentrations and responds more slowly but is thought to have a more sustained effect [84].

Thrombin is also essential to securing the hemostatic plug that seals the injured site by cleaving fibrinogen to fibrin [85]. Highly procoagulant platelet-derived are released by exocytotic budding of the platelet membrane. This is a consequence of a cytosolic rise in calcium, activation of the proteolytic enzyme calpain, and the loss of platelet cytoskeleton/plasma membrane contact [86]. Platelet microparticles typically have diameters one-tenth that of platelets and therefore have a higher surface- to-volume ratio than do activated platelets. They share similarities with the activated parent platelets but have a much higher procoagulant activity, with an increased surface density of phosphatidylserine, P-selectin, and recruited coagulation [87]. However, the platelet -population is heterogeneous, and the phenotype is dependent on the activation mechanism [88]. Fibrinogen released from alpha-granules causes platelet aggregation by linking two platelets via their GPIIb/IIIa . Once the first layer of platelets is formed, additional platelets attach on top of the first layer, being activated by the release of ADP, thromboxan A2, and generation of thrombin from the first layer of platelets.

The shear rate also affects platelets and is an important determinant of platelet formation [89]. Fibrinogen and vWF link recruited platelets via binding to the GPIIb/IIIa [90]. Stretched platelets express junctions that may induce platelets to form tight junctions with each other [81,91]. P-selectin is a membrane-bound adhesion protein that is stored in the membrane of the alpha-granules of platelets. When granules are secreted, their membranes fuse with the plasma membrane, and P-selectin becomes exposed on the surface of the activated platelets. P-selectin glycoprotein ligand-1 (PSGL-1) on leukocytes is the main ligand for P-selectin, and leukocytes are recruited through this interaction to the site of thrombosis.

23

3.4 Effects of platelet activation

Platelet activation follows a sequence of adherence, activation, and aggregation. The cellular effects involved include contraction, with the release of a variety of molecules, and a dramatic change in the surface phenotype of the platelets, with the expression of P-selectin, CD40 ligand (CD40L), and phosphatidylserine along with exocytotic budding of platelet microparticles. The hemostatic relevance of this process is that it forms an initial platelet plug to stop the bleeding, promotes secondary hemostasis by releasing fibrinogen, and contributes to the activation of thrombin and the subsequent formation of a stable fibrin clot. P-selectin expressed on the platelet surface after activation is important for bringing tissue factor (TF) to the thrombus. Both monocytes and monocyte-derived microparticles express PSGL-1, a receptor that interacts with P-selectin. By binding to P-selectin on activated platelets, circulating TF-bearing monocyte microparticles are brought to the thrombus, where they stimulate coagulation by activating factor VII [92]. Binding of P-selectin to PSGL-1 on monocytes also induces the de novo synthesis of TF in these cells [93].

The proinflammatory effects of platelet activation are complex. These effects originate from the mediators released and expressed upon platelet activation, including adhesion and complement proteins, growth factors, and . Surface expression of P-selectin and CD40 are crucial for the recruitment and activation of leukocytes and endothelial cells near the thrombus [94]. IL-1 and soluble CD40L shed from platelets are important activators of endothelial cells [95,96]. Activated endothelial cells alter their phenotype to express adhesion molecules (ICAM-1, VCAM-1) and release chemotactic mediators (MCP-1, IL-8) that recruit leukocytes [96,97]. Important mediators of the direct recruitment of leukocytes to the thrombi are released platelet factor 4 (PF4) and RANTES. PF4, which is an abundant protein in platelet alpha- granules, attracts monocytes and promotes their differentiation into macrophages [98]. The RANTES also recruits monocytes and cooperates with P-selectin in keeping them at the activation site [99]. Once leukocytes are recruited to the activation site, they are prone to form complexes with activated platelets and platelet microparticles via P-selectin/PSGL-1 and CD40L/CD40. Circulating aggregates of monocytes/ and platelets/platelet microparticles are associated with thrombotic syndromes and are prone to enhance atherosclerotic plaque formation and graft occlusion [100].

24 3.5 Platelet regulation

The optimal platelet hemostatic response is a full prevention of blood loss, without vascular occlusion. To prevent occlusion from occurring, multiple molecules participate in the regulation of platelet thrombus formation. Endothelial cells offer at least three important regulatory mechanisms: production of , generation of nitric oxide (NO), and degradation of ADP (see Fig. 3). The first two are autacoids, released locally upon cell stress and rapidly inactivated. Degradation of ADP is continuously performed by CD39, an ecto-NTPDase expressed at the membrane of healthy endothelial cells [101]. NO and prostacyclin prevent intracellular calcium-mobilization and subsequent events associated with increased intracellular calcium, cytoskeleton rearrangement, and granule release [102]. NO is also associated with attenuation of the TXA2 receptor [103], lowered expression of P-selectin [104], and attenuated fibrinogen binding via GPIIb/IIIa [105]. CD39 acts solely on the prothrombotic ADP by preventing ADP-dependent activation in healthy (CD39 is further described in section 7.1). Aside from the depletion of prothrombotic ADP, CD39 contributes to the formation of adenosine by supplying AMP to CD73. CD73 then degrades AMP into adenosine, a potent platelet inhibitor [106]; CD73 activity has been shown in a mouse model to attenuate platelet activation [107].

25 PMP

ADP/ATP

P-selectin

Alpha granules Prothrombin P2Y1 CD40L Dense granules Lysosomes P2X1 FXa/FVa

Thrombin PAR1 TXA2 PAR4

FIXa/FVIIa Fibrinogen Serglycin αIIbβ3 GPIb-IX-V GPVI α2β1

Collagen vWF CD40 PSGL-1 CD40 PSGL-1

Monocyte

Monocyte NO/PGI2 heparan sulfate ADP AMP TF CD39

vWF

Collagen healthy endothelium Injured vessel

Fig. 3. Platelet regulation and activation on healthy and damaged endothelium. Healthy endothelium expose e.g., heparan sulfate and CD39 that prevent activation of platelets. Once the vessel wall is damaged, vWF and collagen promote adherence and activation of platelets at the damaged site. Various stimuli promote a series of events leading to release of granule content, activation of platelets in the vicinity and formation of a platelet aggregate. Damaged endothelial cells and recruited monocytes express TF that promotes coagulation.

26 4. COAGULATION

4.1 Overview of the coagulation cascade

As the secondary hemostatic system, coagulation acts as a relief to platelet- derived primary hemostasis. Like the complement system, the coagulation pathway is organized in a cascade that involves the formation of enzyme complexes and cleavage of zymogens to activate new , some of which are calcium-dependent. Two different branches exist for coagulation activation: the TF pathway and the contact activation pathway. They converge in activated FXa, which forms a common enzyme complex for the activation of thrombin. Thrombin subsequently forms insoluble fibrin from soluble fibrinogen. As discussed in section 3.3, coagulation and platelets are intimately connected; activated platelets and platelet microparticles provide a surface for the assembly of coagulation enzyme complexes. Platelets also recruit TF to the thrombotic site as well as promote a negatively charged surface that can initiate contact activation.

4.2 The tissue factor pathway

The major physiologic initiator of coagulation is TF. Monocytes and endothelial cells both show inducible expression of TF [93,108]. When stimulated by proinflammatory mediators, TF is expressed at the cell surfaces of endothelial cells and monocytes, as well as on monocyte- and platelet- derived microparticles [109,110]. TF is also constitutively expressed in the vascular adventitia and in extravascular tissue [111]. FVII binds membrane- bound TF, which facilitates the activation and formation of the extrinsic complex TF-FVIIa. TF-FVIIa cleaves FIX and FX to its representative active forms (FIXa and FXa).

27 4.3 Contact activation

Contact activation is initiated by the adsorption and autoactivation of FXII on negatively charged surfaces, e.g., upon contact with an anionic artificial material. Even when negatively charged surfaces are also physiologically present in the bloodstream, the relevance of contact activation in vivo is still rather indefinite. However, anionic macromolecules such as collagen [112] and RNA [113] are thought to be involved, and endothelial cells [114] and activated platelets have been shown to initiate contact activation. Platelets can initiate activation both by offering a negatively charged outer leaflet [115] and by releasing , which is prone to activate FXII [116]. Activation of FXIIa leads to a stepwise activation of FXI and FIX and the formation of the intrinsic tenase complex FIXa-FVIIIa, which cleaves FX into FXa. FXIIa can also convert into . Kallikrein contributes to the inflammatory response by generating the proinflammatory peptide by cleaving high molecular weight [117].

Once FXa is formed, it initiates the common pathway by assembling the complex with FVa. Prothrombin is cleaved to form thrombin, which cleaves fibrinogen to generate fibrin. Thrombin also activates the FXIII into FXIIIa, which stabilizes the clot by crosslinking the fibrin network.

4.4 Coagulation control

The contact/coagulation cascade is assembled by serine proteases and hence derives its major regulation from the serine protease inhibitors (): antithrombin (AT), heparin cofactor II, , and C1INH, all of which represent important coagulation inhibitors [118]. They all act at multiple points in the cascade, although the level of thrombin and thrombin activation (FXa) is of major importance because of thrombin’s role as a key molecule in the cascade. AT is a major and acts by forming a complex with the target molecules (e.g., thrombin and FXa), thereby blocking their active sites. The rate of inhibition by AT and other serpins is influenced by heparin and heparan sulfate (HS) [118,119]. Heparin binds both AT and serine protease and facilitates their interaction. Heparin accelerates AT’s interaction by 1000-fold for thrombin and 10,000-fold for FXa [118]. The level of thrombin-antithrombin (TAT) complexes formed is a representative marker for the propagation of coagulation [120]. Interestingly, the regulation of the contact activation serine proteases (FXIIa, FXIa, kallikrein) differs according

28 to the initiating agent. Contact activation as a result of platelet activation or contact with a material surface is regulated by AT and C1INH, respectively [121]. TF pathway inhibitor is an early regulator of the extrinsic pathway and acts by inhibiting the extrinsic pathway tenase complex (TF-FVIIa) [122].

29 5. PLATELET AND COMPLEMENT INTERACTIONS

Both platelets and the complement cascade are constantly present to guard against the stress that prompts hemostasis. Following rapid activation, they affect the local milieu and also influence each other. Vascular damage caused by external trauma accompanied by a septic condition leads to synchronized activation of the thrombotic and complement systems. However, aseptic trauma associated with thrombotic events is also, to some greater or lesser extent, accompanied by complement activation. -reperfusion injury has been shown to activate complement in a variety of tissues [123], and all three complement pathways have been suggested to be involved [124,125,126]. Much of this work has been carried out in thrombotic heart tissue in the course of studying complement activation resulting from myocardial [127,128,129].

Thrombotic events are also a consequence of disturbances of complement control. Atypical-familial hemolytic uremic syndrome (aHUS) and paroxysmal nocturnal hemoglobinuria (PNH), both of which are associated with deficiencies in complement regulation, show a clinical picture of [130,131] and platelet hyper-reactivity [132,133]. aHUS patients with mutations in factor H showed complement activation on platelets that can be abrogated by the addition of normal factor H [134]. The multifactorial disease systemic lupus erythematosus (SLE), which exhibits a clinical picture involving disturbances in complement function [37,135], is reported to feature thrombocytopenia as a correlate of a higher severity of disease [136,137].

The molecular details of the interactions between the complement and platelet systems are still unclear today and are therefore the subject of this thesis. These interactions are, however, rather difficult to evaluate, not least because of the need to use more or less purified systems, , and inhibitors. The phenotype of the activated platelets is also dependent on the

30 concentration/strength of the agonist used, as well as on the intrinsic variation between platelets [138].

Complement activation elicited via both the AP and the CP has been postulated to occur at the surface of activated platelets. In 2005, del Conde et al. showed deposition of C3 and C5b-9 as a consequence of platelet activation by ADP, thrombin receptor-activating peptide (TRAP), or shear stress [139]. They proposed a mechanism whereby P-selectin acts as an acceptor molecule for the binding of C3b and the assembly of the alternative pathway convertase. Following this study, Peerschke et al. reported CP activation with C4d deposition at the surface of activated platelets [140]. Binding of C4d to platelets was demonstrated on platelets in suspension as well as on immobilized platelets that had been induced in response to TRAP, or to TRAP or shear stress, respectively. Initiation of complement activation was proposed to occur by the binding of C1q via its globular C1q head to the receptor gC1qR/p33. The gC1qR/p33 C1q receptor shows weak expression in resting platelets, but its expression is induced after stimulation with various agonists [141,142]. Complement activation was also found on platelet microparticles [143]. In 1978, Polley et al. showed a complement-dependent deposition of C3 on platelets, in the absence of C2 and in heat-inactivated serum. These authors proposed a mechanism for thrombin-dependent C3 convertase formation at the platelet surface [144]. Thrombin has recently been shown to cleave C3 and C5 in a purified system [145] and has been suggested to cleave C5 in a C3-/- knockout mouse model [146], but if these findings are relevant in the in vivo human situation, it has yet to be established. Comple- ment activation on the surface of activated platelets is, however, counteracted by a number of expressed and recruited complement regulators: DAF [147], MCP [148], and CD59 [149] are all expressed at the platelet membrane; C1INH [150] and factor H [151] are stored in platelet alpha-granules and released and expressed on the platelet surface after activation; clusterin is abundantly transcribed in platelets and expressed at the surface [152].

Platelet activation has been shown to occur subsequent to the binding of complement proteins and activation fragments. Sims and coauthors have thoroughly investigated the effects of the insertion of C5b-9 into the platelet membrane. They found that sublytic levels of C5b-9 induce platelet activation, with release of stored granules [153] and shedding of microparticles [154]. This induction, however, was counteracted by the presence of CD59 on the platelet surface [155]. Also, direct binding of C1q multimers to specific C1q receptors induces platelet activation [156]. Polley et al. in 1983 demonstrated platelet responsiveness to C3a and C3-des-Arg [157]. High concentrations were able to directly cause platelet activation, whereas lower concentrations decreased the activation threshold needed for ADP-dependent platelet activation.

31 6. PROTEOGLYCANS AND GLYCOSAMINOGLYCANS

6.1 Overview of proteoglycans and glycosaminoglycans

Proteoglycans are formed from one or several GAG units covalently connected to a common protein core (see Fig. 4). Generally, the GAG chains tend to dominate the specific features of the proteoglycans, but the protein core can influence binding by directly interacting with ligands or by tethering the GAG chains in a specific organization that promotes high-avidity binding [158]. Proteoglycans have since long been known as an important structural unit in the and for the role of heparin as an anticoagulant, but in addition they have been progressively implicated in other physiologic [159,160] and pathologic situations [161]. Proteoglycans are crucial for normal hemostasis and inflammatory modulation [162]. Intriguingly, platelets and all leukocytes show the presence of some type of intracellular GAG [163].

The common GAG backbone is a linear carbohydrate chain of multiple repeating disaccharide units consisting of an amino sugar (N- acetylglucosamine or N-acetylgalactosamine) and a glucuronic acid or galactose. By combining and modifying the disaccharide building blocks in distinct ways, four families of GAGs are assembled: chondroitin/dermatan sulfate (DS), HS/heparin, keratan sulfate, and hyaluran. The last two are of lower significance in this thesis and will not be considered in the following text. Important modifications of CS/DS and HS/heparin chains include sulfation, deacetylation, and epimerization, which create a large diversity of different GAGs. Carboxylic acids, and in particular sulfate groups, supply negative charges that are important for molecular interactions [164]. Not only

32 is the overall negative charge important, but the specific pattern of sulfate groups is also important for recognition and function [64].

The Golgi apparatus is the main site of GAG synthesis. CS/DS and HS/heparin share the same biosynthetic pathway. A tetrasaccharide linker region is first attached to serine residues on the protein core via sequential addition of monosaccharides, followed by stepwise addition of N- acetylgalactosamine/N-acetylglucosamine and glucuronic acid units to the growing chain [165]. Finally, the GAG chain is modified at specific locations by reactions that include O-sulfation (all), N-sulfation (HS/heparin), epimerization (DS, HS/heparin) and deacetylation (HS/heparin) [165]. Of these four GAGs, CS has the lowest degree of sulfation, followed by DS, HS, and heparin. Heparin is highly sulfated throughout the chain, whereas HS is highly sulfated in discrete blocks [166] (see Fig. 4).

CS is composed of repeating units of N-acetylgalactosamine and glucuronic acid. The disaccharide units in CS are most often monosulfated at carbon 4 (CS-A) or 6 (CS-C) of the N-acetylgalactosamine unit, but they also exist in non- di- or tri-sulfated forms [167]. DS has similarities to CS and was formerly referred to as CS-B, but it differs from CS in its frequent epimerization of the glucuronic acid to iduronic acid [168]. This epimerization increases its molecular flexibility and promotes ligand interactions [169].

All the GAGs except heparin are found in the bloodstream. Heparin, which is an effective anticoagulant that potentiates AT, is exclusively found in connective tissue mast cells [163]. HS proteoglycan is abundantly present at healthy vascular endothelial surface but can be shed upon stress [170]. By interacting with various ligands such as cytokines, complement proteins, and adhesion molecules, it modulates vascular homeostasis (e.g., inflammatory events at the endothelium) [162]. A minor fraction of the endothelial HS carries the AT potentiating sequence and is thought to be sufficient to maintain blood fluidity [171].

33

6.2 Serglycin

DS, and particularly CS, make up the major pool of intracellular GAGs in hematopoietic cells [163]. Many of the GAGs are found in secretory granules, the vast majority connected to a protein core rich in serine and residues that form the proteoglycan serglycin [172]. Serglycin in secretory granules has mainly been implicated in promoting efficient storage of various granule components destined for regulated secretion. It may also be important for facilitating the delivery/presentation of the secreted compounds [172]. Serglycin has been implicated in immune regulation; serglycin (core protein) expression has been shown to be highly upregulated in a canine model after bacterial infection [173]. A serglycin knockout mouse model has shown a reduced capacity to combat Klebsiella pneumoniae infection [174] as well as delayed infiltration during Toxoplasma gondii infection [175]. These reports implicating serglycin in promoting the innate immune response are, however, not in line with other studies that show classical pathway complement inhibition by serglycin (CS-A) in B-cell lines [28,176].

The proteoglycan in platelets is a serglycin [177,178,179] carrying a fully sulfated CS-A as the GAG chain [180]. Serglycin is stored in the platelet alpha-granules and, once activated, is rapidly released into the fluid phase as well as exposed on the surface of platelets [181,182,183]. CS is also present in low concentrations in blood plasma, but unlike platelet CS-A is predominately unsulfated [184]. Along with serglycin in other blood cells, the platelet variant has been shown to promote the storage of granule components. PF4 has shown to be released from activated platelets in complex with serglycin [178]. Woulfe et al. recently shed on the function of serglycin in mouse platelets [179]. In a serglycin-/- knockout mouse model, they found platelets with altered morphology and reduced levels of granule components: PF4, β-thromboglobulin, and platelet-derived . The platelets also had defective granule secretion and aggregation, and the in vivo phenotype showed reduced thrombus formation.

34 - OSO3 OH H - COO O O H H O H O CS-A O H - H OSO3 - HO OH H H OSO3 NHCOCH COO- H H 3 O H O H H O H H O H O CS-E HO OH H NHCOCH H H H 3 s

OH - s OSO3 H - COO O O H H O H CS-C H O - H O OSO3 OH H H HO OH H O NHCOCH 3 O H H H H H O H H O DS COO- O HO OH H NHCOCH H H H 3 H OH H - COO O O H O H H O H H - HS H H OSO3 HO OH O H major units HO NHCOCH O 3 O H H H H O H H Heparin H COO- O major units HO - - O OSO3 HO NHOSO3 s H H H

H - H - OSO3 COO O s O H O H H HS H H O minor units HO - O OSO3 HO - H NHOSO3 H OH H H H - COO O O H Heparin O H H H minor units H O O HO OH HO NHCOCH3 H H H

Xylose Glucoronic acid N-acetylgalacosamine

Galactose Iduronic acid N-acetylglucosamine

Fig. 4. Hypothetical serglycin carrying different GAG chains. All GAGs are connected to a serine residue in the protein core via a common tetrasaccaride linker of xylose-galactose-galactose- glucoronic acid. The different GAG chains shown here represent three types of chondroitin sulfate (A, C and E), dermatan sulfate, heparan sulfate and heparin. The major part of heparan sulfate and heparin is composed of the “major units” disaccharides, but they can carry any degree of modifications represented in the “minor units” disaccharides.

35 7. ADP AND APYRASES

7.1 ADP in platelet activation

Two phosphate groups are what set ADP apart from adenosine, distinguishing a platelet activator from an inhibitor. Degradation of ADP via adenosine occurs continuously in healthy vascular endothelium and is mediated by CD39 and CD73 lining the endothelial surface [185,186]. ADP, as well as ATP is degraded by CD39 into AMP, which is subsequently degraded to the platelet inhibitor adenosine by the action of CD73. When platelets are activated, ADP and ATP are released from platelet-dense granules and act as paracrine activators of platelets in the vicinity, a function that is essential for the recruitment and aggregation of platelets. Two ADP receptors and one ATP receptor are present at the platelet surface: P2Y1, P2Y12, and P2X1. The first two bind ADP and are responsible for early and late aggregation, whereas the last binds ATP and mediates the change in platelet shape [187,188]. ADP is crucial for efficient platelet aggregation and is an obvious therapeutic target for platelet inhibition [189]. and are pharmacologic inhibitors that irreversibly block P2Y12 receptors. Both inhibitors have proved beneficial in patients undergone insertion and those with a history of arterial thromboembolic disease [190,191]. However, blockade of the platelet ADP receptors incapacitates the circulating platelets, and treatment is associated with an increased risk of bleeding [192].

36

7.2 Apyrases

Ecto-apyrases, or apyrases, are enzymes that share the same catalytic mechanism for degrading triphospho- and diphosphonucleosides to their equivalent monophosphonucleosides, and the term most often refers to ATPases and ADPases. Related apyrases are found in a variety of eukaryotic organisms [193]. They have homologous regions that have been conserved throughout evolution, particularly the regions related to the catalytic mechanism [194]. The use of apyrases as antiplatelet agents is an intriguing approach; these agents target the activating agents ATP and ADP and do not interfere with the platelets per se. Apart from endothelium-expressed CD39, ADP-dependent platelet activation is also targeted by several soluble apyrases expressed in the mouthparts of hematophagous arthropods. By excreting apyrases, bloodsucking arthropods can feed on their hosts’ blood without having blood-clotting problems. The ADP depletion is postulated to be an essential factor in host-parasite interplay [195]. Both the CD39 and arthropod variants of apyrase have been expressed in recombinant form as soluble enzymes and successfully used as platelet inhibitors in various settings [196,197,198,199]. Extensive work has also been done to overexpress CD39 in organs for (xeno-)transplantation. Platelet activation as a result of the shedding of endothelial surface CD39 in response to ischemia-reperfusion stress is a factor impairing the outcome of transplanted organ [200,201]. Transgenic overexpression of CD39 has been shown to reduce damage after ischemia-reperfusion stress [202,203], as well as to improve the outcome of transplanted organ [204,205,206].

37 8. BLOOD-MATERIAL INTERACTIONS

8.1 Biomaterial definition

The only true hemocompatible surface is the healthy endothelium; every other surface will to some extent exert stress that affects the various blood constituents. This problem is highlighted in clinical situations that encompass treatments involving biomaterials. “Biomaterial” is a rather broad concept and has recently been defined as follows:

“A biomaterial is a substance that has been engineered to take a form which, alone or as part of a complex system, is used to direct, by control of interactions with components of living systems, the course of any therapeutic or diagnostic procedure, in human or veterinary medicine” [207].

The ability of a biomaterial to work together with living systems and blood is referred to as biocompatibility.

8.2 Biomaterials for blood-contact applications

The biomaterials field is a growing industry, and the need of devices is increasing along with our aging population, and improvements in clinical replacement therapy. Frequently used biomaterials in contact with blood include vascular stents and hemodialyzers [208]. A stent is an expandable tube made up of metal mesh that is inserted into an artery to hold the lumen free after reopening an occluded vessel. Complications after insertion of a bare metal stent are associated with thrombosis in the short term and in the long term [209]. Thrombosis occurring in response to stent insertion is

38 counteracted by the obligatory administration of antiplatelet . Hemodialysis refers to the extracorporeal removal of waste products in patients suffering from renal failure. Anticoagulant is administered, and blood is led into an extracorporeal circuit that includes the dialyzer, a semi-permeable membrane permitting low molecular weight molecules to leave the blood. Hemodialysis is associated with chronic inflammation and high morbidity/mortality risk from progressive atherosclerotic cardiovascular disease [210,211]. These two devices represent different types of blood-material interactions, the stent being an implantable metal device that exposes only a small surface, and the hemodialyzer exposing the blood to a very large polymer surface a large number of times.

8.3 Initial exposure and activation of cascade systems

When a material comes in contact with the blood, plasma proteins rapidly adsorb to the solid surface [212]. The details of how this plasma protein adsorption occurs have not yet been clarified [213]. The composition of the protein layer is prone to change over a period of time, with generally smaller, highly abundant proteins being displaced by larger, higher-affinity proteins; this effect is referred to as the Vroman effect [214]. The nature of the adsorbed proteins, as well as the conformation of the proteins after the interaction is determined by the surface characteristics of the material (e.g., their charge, hydrophobicity, and porosity) [215,216]. The nature and conformational changes in the adsorbed proteins will in turn largely influence the biological effect and are a driving force for the activation of the various activation systems, such as the complement cascade [217]. The degree of complement activation is also influenced by the relative amounts of recruited activators and regulators. Direct binding of C3 [218] and IgG/IgM [219] to material surfaces has been shown to evoke complement activation, whereas interfering binding by inert proteins and regulators limits activation [215]. Thrombin is generated on various surfaces such as metals and glass and triggers contact activation via the autoactivation of FXII [220,221]. Contact activation is traditionally thought to occur only on anionic surfaces but has recently been shown to occur on hydrophobic surfaces as well; therefore, it may have an effect on a broader range of materials than traditionally thought [222,223]. Adsorbed fibrinogen has been shown to recruit and activate platelets by binding platelet GPIIb/IIIa receptors [224]. The degree of adsorption-induced unfolding of the fibrinogen structure is a strong determinant of platelet binding, with the extent of platelet binding being directly correlated with the degree of denaturation of adsorbed fibrinogen

39 [225]. A similar pattern was recently shown for human serum albumin (HSA), in which platelets did not bind to either native or moderately changed HSA, but denaturation induced a certain degree of platelet binding that was intensified along with the degree of HSA denaturation [226]. Altogether, these data highlight the importance of surfaces interacting with abundant plasma proteins (e.g., albumin, IgG, fibrinogen, C3) in the proper conformation to augment hemocompatibility.

8.4 Cell activation

The first phase of protein adsorption/triggering of clotting and complement activation is followed by an amplification of the response and crosstalk between the various cascades as well as the involvement of a variety of cells, as discussed in previous sections. One important feature of hemodialysis has been shown to be complement-dependent . Complement has shown activated in extracorporeal circuits, where it induces the expression of TF on monocytes and neutrophils via C5a. TF expression fuels the generation of thrombin and clotting progression [227].

8.5 Modification strategies

Different approaches have been applied to decrease the reactivity of blood against material surfaces: altering the design of the materials (their composition, structure), coating strategies, and attachment/targeting of specific molecules.

8.5.1 Material design Producing materials with specific nanostructures is a promising approach in regenerative medicine [228] and may also have a specific impact on materials that are in permanent contact with blood. The material pore size has been shown to influence the pattern of proteins that bind as well as the ability to activate the complement cascade [215,229]. A limited pore size may sterically decrease the area accessible for assembly of the complement convertases [217]. As compared to hydrophilic surfaces, hydrophobic surfaces generally show greater protein binding capacities and are, importantly, also prone to induce more prominent conformational changes [230]. Therefore, strategies have been focused on increasing the hydrophilicity of biomaterials. Introducing hydrophilic polymer networks that penetrate the hydrophobic polymer, or

40 simply coating surfaces with hydrophilic polymers, reduces the hydrophobic material interface with the blood [231]. Coatings with hydrophilic polyethylene glycol (PEG) have been tested on biomaterial surfaces. PEG provides the material with a surface that exhibits low protein adsorption and low platelet activation [232]; however, PEG surfaces are still prone to activate the complement cascade [233].

8.5.2 Heparin coating Another coating strategy is the attachment of heparin. Unlike PEG, which is protein-repellant, heparin is believed to specifically recruit inhibitory molecules to the surface. Immobilized heparin, which resembles endothelial surface HS, has been shown to control both coagulation/platelet activation and the complement cascade when coated on biomaterials [234,235]. Much of the coagulation inhibition is certainly due to the potentiating effect of the AT upon heparin binding, and to the complement-regulatory effect of the binding of factor H and perhaps other complement regulators [234,236]. The overall binding pattern and conformation of plasma proteins after recruitment to the heparin surface may also explain the high blood compatibility of heparin [233]. Heparin-coated biomaterials are routinely used in the clinic. Cardiopulmonary bypass circuits carrying heparin coatings are associated with a reduced inflammatory response and an overall improved clinical outcome during extracorporeal circulation [237,238]. Also, implanted heparin-coated vascular grafts have recently been shown to significantly increase graft patency when compared to uncoated grafts [239].

8.5.3 Autoregulation A more specifically targeted approach for increasing hemocompatibility is that of directly immobilizing regulatory molecules or specifically targeting them to the biomaterial surface. Factor H and C4BP are, as important complement regulators, interesting targets. Factor H direct immobilized onto a surface has been shown to reduce surface-dependent complement activation [240,241], as has also been shown for DAF [242]. Direct immobilization of purified proteins is, however, a rather expensive method and is associated with contagion risk. Engberg et al. and Wu et al. have recently presented a further development of this approach [243,244]. By using synthetic peptides with an affinity for C4BP and factor H, respectively, they have succeeded in recruiting these regulators from plasma in the fluid phase onto the material surface, with a resulting attenuation in CP and AP complement activation. This approach may be more applicable because it involves shorter synthetic peptides and is therefore less expensive in terms of preparation and can withstand sterilization. Various microorganisms utilize this approach of recruiting complement regulators to their surface to evade complement attack [245].

41 9. CELL THERAPY

9.1 Cell therapy and instant-blood-mediated inflammatory reactions

Clinical cell therapies have emerged as a treatment of various conditions, such as the transplantation of islets of Langerhans for the treatment of diabetes type I and of for the treatment of hepatic insufficiency. By analogy with the blood-biomaterial contact discussed in the previous section, cell-therapy treatment is prone to evoke similar activation signals in response to blood contact. The administration route for islet transplantation is to infuse islets into the circulation. The rate of islet survival is largely determined by the events taking place in the bloodstream from the time of infusion until the islets have reached their destination in the liver [246]. Islets are not intended to be in contact with blood, and they elicit reactions referred to as instant- blood-mediated inflammatory reactions (IBMIR). These reactions are mainly a result of initial complement and coagulation activation [247], since islets lack some of the regulatory molecules that hematological cells carry [248]. Preformed natural antibodies can bind islets in the bloodstream, triggering complement activation via the binding of C1q [249]. The complement response is followed by islet-cell stress and an expression of TF that initiates a clotting reaction [250]. These initial mechanisms are then amplified to involve platelets and inflammatory leukocytes, which may contribute to inflammation, thrombotic reactions, and sequestration of transplanted cells [251]. The IBMIR has also been seen when porcine islet are transplanted to primates in vivo [252]. In this xenogenic setting, IgM and IgG immediately bind to the xenogenic islets, with subsequent complement activation.

42 9.2 Strategies to increase cell transplant survival

Some of the strategies that have been used to increase the blood compatibility of biomaterials have also been employed at cell surfaces. PEG has been directly introduced into the membrane of cells or polymerized as a thin capsule surrounding the cells. The latter approach protects the system from both humoral and cellular immunity, but the cells suffer from an insufficient diffusion of nutrients and oxygen [251]. Heparin coatings, which have shown to not interfere with islet function or the release of insulin, have shown promising results in attenuating complement as well as the clotting reactions when exposed to blood in vitro [253,254].

43 10. AIMS AND OBJECTIVES

The overall aim of this thesis has been to explore the intersecting activation and crosstalk between platelets and the complement cascade under conditions of stress. The first aim had a basic research purpose: to study the effects of platelet activation on the complement cascade, with specific emphasis on CS- A. The second aim was an applied aim: to increase the biocompatibility of model biomaterial surfaces by specifically targeting platelet and complement activation.

The specific objectives of:

Paper I were to study the response of complement to blood clotting, platelet activation, and the release of CS-A, and to pinpoint the mechanism by which CS-A activates complement.

Paper II were to characterize the binding pattern of complement proteins to the surface of activated platelets, and to determine whether this binding is a result of complement activation at the surface.

Paper III were to identify plasma proteins that interact with CS-A, and to characterize the contribution of CS-A to the binding of identified complement proteins to the platelet surface.

Paper IV were to increase the biocompatibility of a model biomaterial via the immobilization of a functional apyrase targeting ADP-dependent platelet activation.

Paper V were to simultaneously target ADP-dependent platelet activation and the alternative pathway on the surface of a model biomaterial and cells, thereby increasing their overall biocompatibility.

44 11. METHODOLOGICAL CONSIDERATIONS

The goal of this section is to justify and briefly describe the reagents, central methodology, and techniques employed in the thesis. A more detailed description of the materials and methods used is presented in each original paper (I-V).

11.1 Reagents

11.1.1 Blood sampling Nearly all of the work in this thesis is based on the use of human whole blood or blood products: whole blood/plasma/serum, or purified components of cells and proteins. Blood was collected by venous puncture of the forearm vein. Blood donors were healthy volunteers who had not received any for the last 10 days. The aim was to use as unprocessed blood as possible: whole blood rather than plasma, plasma rather than serum, and serum rather than purified components. However, to allow certain experimental setups and facilitate interpretation of data, processing of blood was necessary. To prevent blood coagulation, sampling of blood was done in the presence of EDTA, , or heparin. For particular experiments, blood was collected without soluble anticoagulants, using blood-sampling equipment with a Corline heparin surface [255]. The particular anticoagulant used was selected to minimize interference with the blood compartment to be studied: The divalent cation chelator EDTA was selected for the study of inactivated cascade systems. The thrombin inhibitor lepirudin was used for the study of complement function and platelets because it does not interfere with the complement cascade [256] or with platelets per se, but only thrombin [257]. Soluble heparin, which can interact with both the

45 complement and coagulation cascades in addition to cells, was used in low concentration when the purpose was to study the coagulation cascade alone or coagulation together with complement/platelets.

11.1.2 Platelet preparation and activation All five papers involved platelets in some form. Platelets were studied in whole blood, in platelet-rich plasma (PRP), and as purified, washed cells. Platelets were isolated from whole blood by centrifugation of PRP. Washing and dilution was carried out in Tyrode’s buffer with bovine serum albumin, E1 and heparin, or lepirudin in the washing steps to keep the platelets from being activated. Specific activation of platelets via PAR-1 was performed with the thrombin receptor agonist TRAP-6 in the presence of lepirudin. Platelet activation was verified by the expression of P-selectin.

11.1.3 Complement control Inhibition of the complement system was employed at certain stages in the cascade to identify mechanisms of complement activation and surveillance responses dependent on an active cascade. C1q-depleted serum with subsequent C1q reconstitution was used in paper I to ascertain activation via the classical pathway. Anti-C1q-85 is an that specifically blocks the globular heads of C1q [258] and was used in paper II for evaluating C1q recruitment to CS-A. Compstatin (Ac-ICV(1-MeW)QDWGAHRCT-NH2), employed in papers I and II, inhibits complement activation at the level of C3, binds C3, and prevents its cleavage into C3b and C3a [259]. C5aR antagonist (C5aRA) was used in paper I to block /monocyte activation via the C5aR [260]. EDTA was employed to stop the activation after incubations. EGTA, which preferentially binds calcium, was used together with Mg2+ to displace the equilibrium of the complement activation toward the AP, which is calcium-independent (in contrast to the CP and LP, which are not) [261].

11.1.4 Other reagents Chondroitin sulfate-A (CS-A), used in papers I-III, was from bovine trachea, from a batch containing 77 % CS-A and the rest CS-C. To evaluate the effect of CS, a CS-degrading enzyme (chondroitinase ABC) was employed in paper I, and specific antibodies (CS-56 and 2H6) recognizing different epitopes of CS-A [262] were used in paper III. Apyrase in paper IV-V was obtained from potato and had an activity of ≥200 U/mg, with a low ATPase/ADPase activity ratio.

46 11.2 Central methodology and techniques

11.2.1 Quantification of activation markers and protein binding Established enzyme-linked immunosorbent assays (ELISAs) were the main methods used to quantitatively estimate fluid-phase activation markers. C3a, C5a, and sC5b-9 in the fluid phase and C3 fragments on surfaces were used as markers for complement activation. A hemolytic assay of the AP of complement was also used in paper V. Erythrocyte lysis was calculated as y/(1-y), where y is the percentage lysed [263]. TAT-complexes were analyzed as a marker for coagulation, and complexes between AT/C1INH and FXIa/FXIIa were measured to detect contact activation. Flow cytometry was the corresponding method for measuring activation and direct protein binding to cells. Expression of CD11b was employed as an activation marker of granulocytes and monocytes. Specific antibodies were employed in flow cytometry to detect direct protein binding to platelets.

11.2.2 Immobilization of biomolecules All five papers involved some type of biomolecule immobilization onto surfaces: CS-A was immobilized in paper I-III, apyrase in papers IV and V, and a factor H binding peptide (5C6) in paper V. CS-A possesses two possible sites for immobilization, carboxylic acids in the repetitive disaccharide units and the primary amine in the protein core. The first was employed in paper I and the second in papers II and III. Conjugation via the carboxylic acid may interfere with CS-A-protein interactions, as has been shown for heparin [264], whereas utilizing the primary amine ensures a one- site attachment and a presentation closer to the native presentation. Immobilization involved both direct covalent attachment and conjugation of a biotin and subsequent binding to streptavidin/avidin. Apyrase was immobilized to surfaces in paper IV by biotinylation of primary amines in the protein and attachment to an avidin layer on polystyrene surfaces. Initial experiments in paper V utilized biotin-avidin chemistry for conjugation of biotinylated peptides to polystyrene surfaces. To avoid the use of avidin, peptides and apyrase in paper V were conjugated onto glass surfaces and cell membranes through a PEG-lipid anchor. Peptides synthesized with a terminal cysteine and apyrase chemically modified with thiol groups were utilized for conjugation to a maleimide group of the PEG chain. The conjugate was connected to the glass surfaces, which had been modified with amino groups, or anchored in the cell membranes via the PEG-conjugated .

47 11.2.3 Biomaterial testing Biocompatibility testing of material surfaces in papers IV and V was done either in wells of microtiter plates or in the slide chamber model [265]. In the case of microtiter plates, the wells were directly covered with the molecules to be tested, and a small volume of blood/plasma (100 µL/well) was incubated in the wells. The slide chamber model permitted the testing of a larger blood volume (1.3-1.45 mL/well). This device consisted of two cylinders fixed onto a microscope slide on one end, with the test surface acting as a lid on the other end. The test surface could be uncoated or carrying molecules/cells to test. The device was vertically rotated in a water bath at 37 °C. The chamber surface, as well as all other material surfaces intended to come in contact with blood, were coated with the Corline heparin surface to reduce background activation. 11.2.4 Affinity chromatography and protein identification A CS-A affinity column was used to find plasma proteins with affinity for CS-A. Both normal serum and serum depleted of C1q were applied to the column. Washing was done in reduced-salt PBS in accordance with previously reported recommendations [63]. For the identification of abundant proteins, the eluate was applied to SDS-PAGE for separation and in-gel digestion. After in-gel digestion, visible proteins were ionized with matrix-assisted laser desorption ionization (MALDI) or electrospray ionization (ESI) and detected in MS/MS mode. 11.2.5 Surface plasmon resonance Interaction analysis of CS-A and complement proteins was carried out in real time on a Biacore X biosensor, which uses the principle of surface plasmon resonance (SPR). This method enables the study of binding kinetics between two unlabeled biomolecules in real time, in this case CS-A immobilized onto a surface and purified complement proteins in the fluid phase. The physical principle of SPR is rather complex; in a simplified view, the instrument utilizes an optical device to detect changes in refractive index in close proximity to a sensor surface. The biomolecule of interest is passed over the sensor surface with microfluidics. Changes in the refractive index occur when protein in the fluid phase interacts with ligands immobilized onto the surface. The detected response is direct proportional to the mass of the protein molecules [266]. Binding kinetics were analyzed with ClampXP software [267] and fitted to a “surface heterogeneity model”, a model describing different forms of immobilized ligand that are able to interact with the analyte with separate rate constants. This model best represented the multivalent binding of the analyzed proteins to CS-A and corresponded to a certain heterogeneity of immobilized CS-A, with multiple sites for interaction and possibly featuring rebinding phenomena.

48 12. RESULTS

Inflammation and thrombosis in the blood compartment are not isolated events, but are intertwined processes involving cells and proteins that have evolved to act together to maintain blood homeostasis and overall host integrity. This section presents the findings from three papers (I-III) illustrating how the complement response is influenced by the activation of platelets via thrombin receptors. The remaining two papers in this thesis (IV- V) deal with analyses of platelet and complement regulation on model biomaterials and cells, with the goal of increasing their biocompatibility.

12.1 CS-A released from activated platelets triggers activation of the CP of complement (paper I)

Both in vitro studies and clinical situations have highlighted the increase in complement activation that occurs as a secondary response to thrombosis. We found that complement activation was correlated with thrombin generation, which was elevated in the presence of platelets. The degree of C3a and sC5b-9 formation was similar regardless of whether the activation of platelets was accomplished by TRAP in PRP or by the addition of supernatant from activated platelets to PPP, indicating that platelet-dependent complement activation occurred in the fluid phase. CS-A was hypothesized to modify the complement response, and soluble CS-A was added to PPP to evoke a dose- dependent complement activation, which was found to be abrogated by pre- incubating CS-A with chondroitinase ABC. A similar scenario regarding complement activation was seen after the addition of chondroitinase ABC to the platelet supernatant. Immobilized CS-A was found to specifically bind C1q and thereby trigger complement activation. CS-A-dependent activation of complement was pinpointed to the classical pathway. Activation was found to be absent in C1q-depleted serum but was fully restored by the addition of purified C1q. Activation was also absent in the presence of Mg2+/EGTA.

49 Finally, TRAP-mediated activation of platelets in whole blood was shown to cause up-regulation of CD11b on granulocytes and monocytes, as well as to increase complex formation between platelets and granulocytes/monocytes. Adding compstatin and/or C5aR attenuated these effects.

Altogether, these results showed that thrombin activation of platelets causes fluid-phase activation of the complement system that is dependent on the release of CS-A and activation of C1q. Given this background, the focus in paper II shifted to complement activity on the surface of activated platelets.

12.2 Binding of C3 to activated platelets occurs independently of complement activation (paper II)

There is a lack of consistency in the existing literature regarding complement activation at the surface of activated platelets. The platelet surface is decorated with abundant complement regulators, but complement activation has still been reported recently at the platelet surface.

Binding of C1q, C3, C4, and C9 was significantly increased to the platelet surface after TRAP-mediated activation of platelets in PRP and whole blood. Purified C1q and C3 showed a similar degree of binding to washed platelets and to PRP and whole blood, indicating direct interaction with the platelet surface without support from other fluid-phase plasma proteins. Binding of C3, as well as C1q, C4, and C9, in PRP was independent of an active complement cascade, since the binding was unaffected by the presence of complement-inhibitory EDTA or Mg2+/EGTA. No C3 was covalently attached to the platelet surface, but it could be eluted from the surface with Tween. Characterization of platelet C3 using specific antibodies in flow cytometry and western blotting led to the identification of C3H2O and the factor I-cleaved form iC3H2O as the forms of C3 present on activated platelets. C3H2O is an activated form of C3 and, like C3b, is able to bind complement receptors. Soluble CR1 was indeed able to bind washed, activated platelets (as detected by flow cytometry), but only when co-incubated with purified C3.

Conclusively, this paper demonstrated the binding of early to late complement components without complement activation. C3 was bound in the form of nonproteolytically activated C3H2O, a conformation that was able to bind sCR1 and possibly act to link platelets to CR-bearing leukocytes.

50 12.3 CS-A links C1q, C4BP, and factor H to the platelet surface (paper III)

In paper II, complement activation was found to be absent at the platelet surface, presumably because of the presence of complement regulators. CS-A, which is present on activated platelets, has previously been shown to interact with complement regulators and might also do so on activated platelets.

Passing serum through a column with Sepharose-immobilized CS-A led to the identification of C1q, , plasminogen, TAT, and HSA as CS-A-interacting proteins. Depleting C1q from the serum led to the specific recruitment of as well, and in general, to overall higher-quantity binding profiles of other interacting proteins. Special emphasis was placed on complement regulators: C4BP, factor H, and C1INH were all shown by western blotting to be present in the affinity column eluate, although in lower abundance than the previously mentioned proteins. Binding of factor H was to a large degree competed out by C1q in serum, whereas C1INH binding was instead associated with the presence of C1q. SPR analysis confirmed that C1q, C4BP, and factor H all specifically interacted with CS-A, whereas C1INH did not. All four regulators were also shown to bind platelets after TRAP-mediated activation, and all except C1INH also bound activated platelets in the purified system. Blocking platelet-exposed CS-A with a specific antibody reduced the binding of C4BP in particular, but also inhibited that of factor H and C1q, indicating that CS-A also represents a ligand responsible for recruiting these complement regulators and C1q to the platelet surface. Binding of C1q to immobilized CS-A and platelets was shown to link model ICs to their respective surfaces, possibly implicating C1q in linking ICs to activated platelets also in in vivo pathologic situations.

CS-A exposed on activated platelets may link soluble complement regulators to the platelet surface, thereby balancing complement-triggering signals.

51

12.4 Apyrase immobilization lowers the thrombogenecity of a model biomaterial (paper IV)

With the aim of enhancing overall biocompatibility, platelet activation at a model biomaterial surface was targeted in paper IV.

ATP/ADP-degrading apyrase was immobilized onto a model biomaterial. Even though the procedure of surface attachment was associated with an extensive loss of apyrase activity, the immobilized variant was still shown to be active. When compared to a corresponding HSA surface, the apyrase- immobilized surface showed a reduction in platelet surface adherence in the presence of PRP pre-stimulated with ADP. The integrity of the apyrase surface was also damaged, with platelet deposition occurring at the damaged site; this did not happen with the intact apyrase. When PRP without anticoagulant was incubated on apyrase and control surfaces (HSA, avidin, and polystyrene), the apyrase surface provoked a lower level of platelet consumption and also induced significantly lower levels of coagulation (TAT) and contact activation markers (FXIa-AT, FXIIa-AT). The levels of the complement activation fragment C3a were somewhat lower on the apyrase surfaces than on the protein-coated control surfaces, even though the changes were not statistically significant. Staining of surface-deposited leukocytes and platelets on apyrase and control surfaces (HSA, polystyrene, glass) after whole-blood incubation demonstrated a lower total cellular adhesion on protein-coated surfaces.

Surface-degradation of ADP by immobilized apyrase was associated with lower platelet activation as well as lower activation of platelet-dependent coagulation and contact activation.

52

12.5 Enhanced overall hemocompatibility of materials and cells with 5C6 and apyrase (Paper V)

A reduction (though not statistically significant) in complement activation was observed after surface immobilization of apyrase. Paper V simultaneously targeted platelet and complement activation on surfaces and cells.

Attachment of a factor H-binding peptide (5C6) to material surfaces and cells, unlike that of a scrambled control peptide, was able to specifically recruit fluid-phase factor H. Binding of factor H was associated with significantly lowered surface-induced complement activation by the biomaterial or cells. The surfaces carrying the peptide showed lower levels of fluid-phase markers, and the introduction of the peptide onto the membrane of erythrocytes lowered the level of in an AP of complement hemolytic assay. Similarly, apyrase anchored to erythrocytes was shown to suppress platelet consumption and TAT generation, and the addition of erythrocyte-conjugated apyrase lowered ADP-triggered platelet aggregation in PRP. The biocompatibility- enhancing effects of apyrase and 5C6 were retained when they were combined on the same surface. Material surfaces carrying both modifications exhibited significantly lower thrombogenicity and elicited complement activation to a lower degree than did control surfaces coated with PEG. Although PEG was inert against platelet consumption and TAT generation, these parameters were significantly lower at the apyrase/5C6 surface. C3a and sC5b-9 levels were similarly (and significantly) reduced. In a model system, apyrase and 5C6 were analogously combined at the membrane of adherent porcine arterial endothelial cells (PAECs) and exposed to human whole blood in a xenogenic setting. The xenogenic interface elicit incompatibility reactions in human blood, since preformed natural antibodies directly bind the xenogenic cells. These reactions were, however, efficiently counteracted by the apyrase and 5C6, which significantly lowered all four tested parameters (TAT, platelet consumption, C3a, and sC5b-9) when compared to cells not carrying apyrase or 5C6- peptide.

These results showed that combining apyrase and the factor H binding peptide on the surfaces of materials and cells mediates an overall enhanced biocompatibility with regard to the activation of platelets, coagulation, and the complement cascade.

53 13. GENERAL DISCUSSION

In vitro studies as well as clinical conditions have revealed crosstalk between the hemostatic and inflammatory systems in the blood compartment [268]. This work, presented in five papers in this thesis, describes the consequences of thrombin receptor agonist-mediated stimulation of platelets, with a focus on complement response. The effects of specifically inhibiting platelet and complement activation were subsequently studied on biomaterial and cell surfaces.

These findings concerning the response of complement activity to platelet activation can be separated into events in the fluid phase and events at the surface of the activated platelets. When platelets were activated via thrombin receptors, complement was found to be activated as well. The mechanism for eliciting the complement response was traced to the fluid-phase release of CS- A, the binding of C1q to CS-A, and complement activation via the CP. The initiated complement activation, in turn, evoked the activation of inflammatory granulocytes and monocytes. After platelet activation, the platelet surface showed enriched binding of C1q, C3, C4, and C9. This binding was, however, shown to occur independently of complement activation. Surface- bound C3 was shown with monoclonal antibodies to exist as C3H2O and in the factor I-cleaved form iC3H2O, two forms that have been shown to bind soluble CR1 and thus may act to link complement receptor-decorated blood cells to thrombi. The absence of complement activation at the surface of activated platelets is a logical outcome, given the presence of abundant complement regulators at the platelet surface. C1INH, factor H, and C4BP were all found to be recruited to the activated platelet surface. The last two showed an affinity for CS-A, and their binding to activated platelets was significantly reduced when platelet CS-A was blocked; this was also the case for C1q binding to platelets. Binding of C1q to platelets was shown to link model ICs to the platelet surface, a reaction that could cause pathological platelet reactivity during IC diseases.

54 The scenario of complement activation resulting from thrombotic events has also been reported in the clinical situation, in which arterial thromboembolism in vivo has been found to be accompanied by elevated levels of complement activation markers [127]. However, it is difficult to determine the direct contribution of platelets to the complement response in vivo, since ischemia-reperfusion episodes exert extensive stress on the vascular cells that might cause complement activation [123]. In a more controlled environment, Kalowski et al. have showed that complement activation occurs as a result of the injection of thrombin and thromboplastin in a rabbit in vivo model [269]; they were able to partially attenuate the complement activation by inducing thrombocytopenia prior to injection. In this in vitro system, the mechanism for the platelet-dependent activation of the complement system was traced to the interaction between CS-A released from activated platelets and the binding of C1q. In this model, the C1 complex is activated when C1q binds CS-A. C1q has previously been shown to interact with CS-A proteoglycans purified from serum and B-cell cancer lines, although the interaction was then described as complement-inhibitory [28,29,176]. In those studies, the effect of complement response resulting from the C1q-CS-A interaction in normal human serum was evaluated in a CP of complement hemolytic assay with IgG sensitized sheep erythrocytes, and hemolysis was found to be attenuated in the presence of CS-A proteoglycan. Although the assays showed that erythrocytes are protected from lysis in the presence of CS-A, they did not reveal the events following the binding of C1q to CS-A in the fluid phase, which we have now shown to involve complement activation and certainly consumption of complement proteins. Consumption of complement components in the fluid phase would, by definition, reduce the hemolytic activity.

C1q has an affinity for DS, HS, and heparin, in addition to CS species [270], and the consequences of its binding to HS and heparin have been reported to be complement-regulatory [271]. Extracellular peritoneal levels of hyaluronic acid (HA) and CS-A have recently been reported to be highly increased in response to Toxoplasma gondii infection [175], while corresponding levels of HS and heparin were almost undetectable, suggesting a selective release of HA and CS-A in response to infection. Serglycin, which can connect all these four GAGs, might then cause different immunological response, depending on the type of GAG attached. The connected GAGs are diverse in phenotype, not least with regard to sulfation pattern. Gama et al. have shown that CS oligosaccharides with different sulfation patterns recognize specific growth factors depending on the specific sulfation motifs present [64], and Robinson et al. have recently demonstrated considerably different binding characteristics for different GAGs [272]. GAGs that bind C1q with different affinities might by analogy have different effects on C1q activation. Heparin, which possesses binding affinity for both the collagenous stalk of C1q and its globular heads, binds the collagenous stalk with higher affinity than the

55 globular heads [270]. Binding of C1q to CS-A in paper II was almost totally inhibited by blocking the globular heads, indicating that CS-A does not interact with the collagenous stalk of C1q. Binding selectivity for different plasma proteins may also influence complement activation. Yu et al. investigated complement protein binding to heparin and found a direct interaction of C1INH with heparin, as well as a strong affinity for factor H [273]. In contrast, we were unable to detect binding of C1INH to CS-A, and factor H interacted with a markedly weaker affinity, which could explain the differences regarding the complement response.

Although platelet activation caused an activation of the complement system in the fluid phase, the platelet surface per se was shown not to initiate complement activation. This result can be considered to be due to the presence of an efficient complement-regulatory system at the platelet surface. An absence of complement activation at the platelet surface, however, contradicts what has previously been shown. Both the AP and CP of complement have been suggested to elicit activation: AP activation via C3b deposition on P- selectin [139], and CP activation via C1q binding to gC1qR/p33 [140]. P- selectin was hypothesized to attract C3 to the platelet surface, but specific blockage of P-selectin in our system with a polyclonal antibody had no effect on the binding of C3. Therefore, we believe that C3 recruitment to the platelet surface is independent of P-selectin and that C3 utilizes another yet- unidentified ligand. The mechanism of thiolester-mediated disruption of bound C3H2O is also still unknown.

Blockage of CS-A significantly reduced the recruitment of C1q to the platelet surface (by 50 %), proposing that CS-A supports C1q in binding the platelet surface. C1q binding was not associated with complement activation but was able to link model ICs to the platelet surface, in agreement with earlier reports [274]. Binding of model ICs to immobilized CS-A and to platelets was potentiated in the presence of C1q. ICs are potent platelet activators that have been shown to be correlated with platelet consumption in IC-associated diseases [275]. Soluble regulators were found to be recruited to the surface, which is already decorated with the membrane-bound regulators DAF, MCP, and CD59 [148,149,276]. Factor H and preferentially C4BP showed affinity for surface-bound CS-A. Factor H is reported to bind to the platelet surface via GPIIb/IIIa [277] and thrombospondin [278], but binding to platelets was significantly attenuated by blocking CS-A with a . C4BP, which can use protein S for membrane binding [70], showed reduced recruitment to the platelet surface, even at low concentrations of anti-CS-A.

The complement cascade is delicately controlled via a plethora of complement regulators, including those at the platelet surface. Through a slow fluid-phase C3 tick-over, the complement system provides ongoing fuel for activation and

56 is prone to form convertases whenever regulation is insufficient. Altering the presence or function of a single regulator can shift the balance toward complement activation, as in the case of PNH and aHUS. Both disorders are associated with altered function in complement control, as well as episodes of thrombocytopenia and platelet hyperactivity [131,133]. PNH is caused by a somatic mutation in the PIG-A that causes a deficiency in the biosynthesis of the GPI anchor, which in turn decreases the expression of DAF and CD59, both of which are dependent on GPI anchors for membrane insertion [279]. PNH platelets have been shown to have markedly increased levels of C3b fragments (over normal platelets) when triggering complement activation by either the CP or AP [280]. Just blocking CD59 has shown to predispose platelets to MAC insertion and exposure of procoagulatory phospholipids via membrane flip-flop [155]. aHUS is associated with altered activity of the AP, typically loss-of-function mutations in factor H but also mutations in genes encoding other complement regulators or gain-of-function mutations in C3 and factor B [281,282]. Ståhl et al. has highlighted the complement activation that occurs on platelets obtained from aHUS patients. Incubating normal platelets with serum from aHUS patients causes the deposition of C3 and C9, with subsequent platelet activation [134]. This reaction can be abrogated by preincubating the platelets with normal factor H or normal human serum. In parallel with a deficient complement regulation at the platelet surface, excessive triggering signals can disrupt the balance toward activation, as may occur in immune (ITP). ITP is associated with accelerated platelet destruction that is mediated by autoantibodies directed toward the platelet membrane [283]. Platelets from some ITP patients with autoantibodies have been proposed to fix CP complement proteins to the platelet surface [284] and to have increased levels of platelet-associated C3 [285].

ADP is crucial for efficient platelet aggregation; this requirement is highlighted by the evolutionary pressure brought to bear on organisms as diverse as mammals and arthropods, which must express ADP-degrading enzymes to inhibit blot clotting. Intact CD39 on the endothelial surface is well known for maintaining platelets in their resting state, and endothelial stress is correlated with platelet activation and shedding of CD39, along with HS and [200]. Evolution has also forced hematophagous arthropods to express apyrases in their salivary glands to maintain blood fluidity when feeding on vertebrate blood [195]. The clinical approach of inhibiting ADP-dependent platelet activation has instead been irreversibly or reversibly antagonized by platelet ADP receptors. Although efficiently blocking the effect of ADP in order to irreversibly attenuate these receptors will incapacitate platelets, it also increases the risk of hazardous bleeding. In this situation, the approach of targeting ADP would be preferred over ADP receptor antagonism.

57

Apart from its direct thromboregulating properties, apyrase influence anti- inflammatory events through the effects of adenosine. CD39 acts in conjunction with CD73 at the endothelial surface to degrade ATP/ADP into adenosine, which is subsequently degraded to hypoxantine via inosine [286]. Adenosine is anti-inflammatory and was recently implicated in attenuating neutrophil accumulation and inflammation during hypoxia [287].

Platelet activation is an important determinant of the pathophysiolology of biomaterial-associated complications, with respect to thrombotic as well as inflammatory events. The overall approach in this project was to enhance the biocompatibility of materials and cells using the endothelial surface as a template by targeting ADP-dependent platelet activation and AP of complement activation. It was shown that coating of material surfaces with apyrase alone could reduce the adherence of platelets from PRP previously challenged with ADP. Furthermore, when PRP was incubated without anticoagulant, apyrase conjugated to the surface not only reduced direct platelet consumption but even significantly lowered secondary effects such as the activation of the contact and coagulation systems. A negligible effect on complement activation was seen, and this aspect was therefore targeted together with platelets in the following project. The factor H-binding 5C6 peptide alone was shown to regulate complement activation on the surface of materials and cells. Co-immobilization of the 5C6-peptide and apyrase efficiently regulated complement activation at the surface, while retaining the positive effects of apyrase on the thrombotic events. Similar results were shown when 5C6 and apyrase were introduced at the cell membrane.

The results have shown the potential of specifically degrading ADP and ATP at the surface of a foreign material in contact with blood. Not only was a direct effect of platelet inhibition detected, but also a subsequent limited activation of the coagulation cascade appeared. Apyrases of various kinds have been employed in experimental settings as putative anti-platelet drugs. Belayev et al. have made use of a truncated, soluble variant of CD39. This soluble CD39 significantly reduced infarct size in an ischemic model in rats, even when administered 3 hours after stroke induction [197]. A human homologue of a salivary arthropod apyrase has been shown to be effective in reducing thrombosis in a ferric chloride-induced model of thrombosis in mice [288]. Transgenic animals overexpressing CD39 in organs suited for (xeno)- transplantation have shown improved outcomes after transplant, [206] as well as improved resistance to injury following ischemia-reperfusion [202,203]. Apyrase has also been previously employed at the surface of biomaterials with the goal of reducing platelet activation [289,290]. Apyrase was either merely adsorbed [289] or conjugated via glutaraldehyde [290] to the biomaterial surface. Implanted devices carrying the apyrase coating showed reduced

58 thrombogenicity. However, the controls in these studies were uncoated surfaces, and the improved biocompatibility could also be explained by the protein coating “pacifying” the surface. In our system, platelet adherence to both apyrase and control HSA surfaces was negligible. This result would suggest a relative inertness of both surfaces. However, stressing the system with ADP caused an adhesion of platelet aggregates to the HSA surface, but the apyrase surface was left intact. Bakker et al. have adsorbed apyrase onto polyurethane [289]. There is a critical point here, that adsorption of the apyrase might result in an “extended release” of soluble apyrase. The immobilization procedure is critical step, and the use of the PEG linker for attachment is a major achievement in apyrase immobilization. Immobilization of apyrase using biotin/avidin chemistry was associated with a significant loss in activity, which most certainly would also be caused by the glutaraldehyde conjugation method. With the PEG linker, apyrase was immobilized on the surfaces and cells with almost complete retention of activity.

Of concern as well is the source of the apyrase. In the experiments described in this thesis, a commercial apyrase purified from potato was employed. Although well characterized and frequently used (e.g., for the preparation of human platelets in vitro) [291], apyrase from potato is not optimal for use in the human system. A pH optimum around 6.5 and the risk of immunological responses against the plant protein strongly suggest the value of using an enzyme of human origin for continued applications.

The 5C6 peptide binds factor H in its middle region, as does the factor H binding protein from that “hijacks” factor H from the host to the bacterial surface via binding to SCR6 [292]. The ability of 5C6 to bind factor H from plasma was recently thoroughly evaluated, and the peptide was shown to bind factor H as well as to regulate AP complement activation in Mg2+/EGTA-plasma [243]. The data have clearly shown that complement is also significantly regulated when the CP is still active. Factor H is exclusively able to regulate the AP C3 convertase, which suggests the importance of regulating the AP amplification loop, as previously shown [293]. The approach of “cheating” the complement cascade by recruiting already-synthesized large proteins to the surface via smaller peptide sequences has been demonstrated for Streptococcus pyogenes, which attracts C4BP via the M-protein [294]. Engberg et al. have made use of the peptides derived from the M-protein and successfully regulated CP complement activation at a material surface [244]. Peptides of bacterial origin are, however, likely to evoke an immune response, and therefore do not constitute a viable approach for clinical situations. Direct immobilization of factor H has previously been shown to be able to lower complement activation at surfaces [241]. However, the use of a small peptide to recruit plasma factor H, rather than a direct immobilization of the protein, has certain advantages. Binding of factor H via

59 cross-linkers is rather nonspecific and can interfere with the protein’s regulatory domains, whereas binding of the synthetic peptide to factor H is site-specific. Production of a peptide is also cost-effective and can be performed under sterile conditions.

The combined approach of regulating both complement and platelets at a common surface can successfully reduce thrombogenicity and inflammatory events; even in the xenogenic setting with porcine endothelial cells, incompatibility reactions have been significantly decreased. The xenogenic setting is a major stressor for human blood, since preformed natural antibodies bind galactose-α-1,3-galactose-structures on porcine endothelial cells [295], instantly leading to complement activation. However, in our system, specific complement regulation alone was not sufficient to lower overall incompatibility reactions, but it was seen to be effective in the combined approach. One can now propose that a combined approach can be useful both when cells of non-human hematologic origin are transplanted into the circulation system, and when human blood is exposed to biomaterial devices. ADP and the amplification loop of complement activation are essential control points in the activation of platelet and complement, respectively, and are therefore suitable targets for inhibition. Accordingly, it can be proposed that this dual-targeting approach be suitable for inhibiting both thrombotic and inflammatory reactions on materials and cells.

60 14. CONCLUDING REMARKS

Taken together, the data shows that fluid-phase activation of the CP of complement is a result of platelet stimulation via thrombin receptors. The activation is confined to the fluid-phase and is not seen at the platelet surface. This distinction is a logical consequence of efficient complement regulation at the platelet surface. The crosstalk between platelets and complement that is suggested here and has been shown by others, highlights the importance of simultaneously targeting complement and platelet activation in order to enhance the overall blood compatibility of surfaces of non-hematologic origin (see Fig. 5).

It was found that biomaterial-induced platelet activation is efficiently attenuated by immobilized apyrase. This apyrase is further able to reduce platelet-induced contact and coagulation activation. Co-immobilization of apyrase and the factor H-binding peptide 5C6 results in synchronized complement and platelet regulation. Introducing the peptide and apyrase into the membrane and cells results in a similar inhibition, clearly demonstrating that the apyrase and soluble factor H on biomaterials and cells can effectively reduce platelet and complement activation.

61

C1q C4BP C3bBb C3a/C5a Apyrase Act platelet Monocyte

CS-A Factor H C1INH C3H2O ADP/AMP Resting platelet 5C6

Fig. 5. Schematic figure summarizing the major findings of this thesis. The complement system is activated in the fluid phase following the activation of platelets. A number of complement inhibitors are recruited to the surface of the activated cells, and C3 is found as C3H2O/iC3H2O. Apyrase and factor H were succesfully co-immobilized onto surfaces and cells, and the presence of these attenuated platelet and complement activation elicited in the interface between the blood and material/cell-surface.

62 15. ACKNOWLEDGEMENTS

Typing the heading “Acknowledgments” afflicted me with the worst writer’s block ever in this thesis: being afraid of forgetting someone and at the same time being concerned that I really express the gratitude I feel to the people who have meant so much to me.

However, I started to type names on a blank sheet, and with every name, you put a smile on my face. So many memories and so many people who have made these 10 years at Norrgård the best in my life.

I will start with my warmest greeting to Kristina for letting me spend 5 years in your research group. You have all qualities an excellent supervisor should have: dedicated to your research, always interested in my lab results and at the same time caring about my overall well-being. I have nothing but compliments for you.

Kerstin and Anna, you were like a family to me, and I will always remember the time we had in the lab. We had a great lot of fun at work in a very collaborating and relaxed environment. Thanks to Kerstin for your warm personality and always being willing to help; you make up the important backbone in our Kalmar group. And Anna, you were not at all that scary blond-haired girl that I was so afraid of during my undergraduate studies. You bring a lot of joy with you, and it was a privilege to have you as a “big sister”! Thanks also to the former and newer group members I have met: Shan, Sussie, Anette, and Fredrik.

A great big thanks to Yuji and Osama for being wonderful collaboration partners in our projects. It has been enjoyable and easy to work with you despite the physical distance. Thanks also to Bo Nilsson as an unofficial but highly involved co-supervisor of my PhD project. Thanks to Lillemor, Javier, Graciella, Jaan, Jennie, and Susanne and the rest of the Uppsala people I have met. Thanks also to Jonas for sharing your heparin knowledge and for being

63 running partner in Uppsala. Even though you are 1-0 in our internal marathon meetings, I promise I will beat you the next time.

Thanks to my co-supervisors Ian Nicholls and John Lambris. Thanks, John, for letting me spend 3 weeks in your lab, and thanks to the people I met during my lab visit. Special thanks to Daniel and Maria for your excellent help with all Biacore and mass spectrometry work. Thanks also to Rodanthi and Stavros for helping me, as a student from a far-away IKEA country, to find my way in the big city of Philadelphia.

Thanks, Anki, for your dedication to the former BMK program. I have always felt well prepared when entering new labs with my background from Kalmar. Thanks to Mia for letting me work in a very interesting diploma project with you. Ischemia/reperfusion is fascinating and will likely cross my path in the lab again.

Thanks Magda for your help with illustrations.

When I leave Norrgård, I will really miss the coffee breaks. It is not the good-tasting coffee but your company that brings me to the “fika-rum” every day. Added to this are the after-work and dissertation parties. Not to forget floorball at the arena, the church relay race and World cup in table hockey. Thanks to past and present colleagues! To mention some of you: Anna AP, Anna G, Annelie, Camilla, Conny, Elina, Ellinor, Fredrik, Håkan, Kim, Karthik, Lovisa, Marlene, Martina, Nina, Pacho, Sarfraz, Stina, Ullis and more…

Thanks Lasse. You are the most committed and honest person I know, and I admire you for that. Barefoot running, scotch fortune wheel, lost in Trondheim, Reinebringen, moose pizza with bea in Kvikkjokk, praying to the “nåjd” in the wilderness are just a few of the memories that come to my mind. I hope we have not made our last journey into the silent north. And perhaps, sometime we will finish our alternative Olympics. Thanks, Stina, for always being you, honest, fair and with a respectful self-distance. Few people I know have such scientific skills as you possess, even though you might have a hard time admitting it to yourself. Thanks, Malin and Dieter, for hosting a lot of late board-game nights, for plenty of good (and some bad) music suggestions, and especially for your kindness and being there when needed. Thanks Linda, my “cirkelfys”-pal, for your laughter and the joy you bring. Not many are so determined, no matter whether it is Settlers or decisions in your career. I think it will soon be time for a second board meeting. Thanks, Louise, for the energy and happiness you share. Thanks for your ability to “styra-upp” social activities. Thanks, Maria G, for Wednesday running trips, and for your warm

64 personality. Thanks Johanna and Jenny for the beautiful and plenty “NO” notes… and a few “Yes;” I will always carry them with me.

Thanks to Stefan, Bosse, the cleaning ladies and the administrative personnel. Norrgård would not be the same without you, and you deserve all respect.

Thanks to Emma, Maria, Frida, Anna, Pelle-Koch, Lotta, Gunnar, Michelle, Mange, Cissi, Maya, Ann and the rest of the great BMK class of 2006. Thanks also to Jimmy and David, my best friends outside work. Thanks to Gösta for inspiration and great teaching in chemistry and biology in the upper secondary school.

I would like to express my great gratitude to my family: Mum, Dad, my sister and your family. Thanks for always supporting me and being there when needed. You have always been supportive, no matter whether it was driving me to pony-trotting competitions or to the forest at 4:00 am in hunting season. I thank you for always backing up my decisions in life, without trying to impinge. Thanks, Julia, for being a great part of my life; I will always be there for you, whenever you need me.

And finally thanks to my dear and wonderful Maggan. Your smile could turn any terrible day into something great, I don’t know how you do it, but I do not need to know either. You changed my life that day in June, and it will never be the same again. It has been a tough period of writing, but you have supported me as I will support you through every tough period. I love you, and no one could ask for more.

When you're on your own, you walk in the rain Ed Harcourt (Apple of my eye)

65 16. REFERENCES

1. Ricklin D, Hajishengallis G, Yang K, Lambris JD (2010) Complement: a key system for immune surveillance and homeostasis. Nature Immunology 11: 785-797. 2. Stevens B, Allen NJ, Vazquez LE, Howell GR, Christopherson KS, et al. (2007) The classical complement cascade mediates CNS synapse elimination. Cell 131: 1164-1178. 3. Mastellos D, Papadimitriou JC, Franchini S, Tsonis PA, Lambris JD (2001) A novel role of complement: mice deficient in the fifth component of complement (C5) exhibit impaired liver regeneration. Journal of Immunology 166: 2479-2486. 4. Langer HF, Chung KJ, Orlova VV, Choi EY, Kaul S, et al. (2010) Complement-mediated inhibition of neovascularization reveals a point of convergence between innate immunity and angiogenesis. Blood 116: 4395-4403. 5. Lachmann PJ (2009) The amplification loop of the complement pathways. Advances in Immunology 104: 115-149. 6. Gaboriaud C, Thielens NM, Gregory LA, Rossi V, Fontecilla-Camps JC, et al. (2004) Structure and activation of the C1 complex of complement: unraveling the puzzle. Trends in Immunology 25: 368-373. 7. Thiel S, Vorup-Jensen T, Stover CM, Schwaeble W, Laursen SB, et al. (1997) A second serine protease associated with mannan-binding lectin that activates complement. Nature 386: 506-510. 8. Rossi V, Cseh S, Bally I, Thielens NM, Jensenius JC, et al. (2001) Substrate specificities of recombinant mannan-binding lectin-associated serine proteases-1 and -2. The Journal of Biological Chemistry 276: 40880- 40887. 9. Bexborn F, Andersson PO, Chen H, Nilsson B, Ekdahl KN (2008) The tick- over theory revisited: formation and regulation of the soluble alternative complement C3 convertase (C3(H2O)Bb). Molecular Immunology 45: 2370-2379.

66 10. Spitzer D, Mitchell LM, Atkinson JP, Hourcade DE (2007) Properdin can initiate complement activation by binding specific target surfaces and providing a platform for de novo convertase assembly. Journal of Immunology 179: 2600-2608. 11. Fearon DT, Austen KF (1975) Properdin: binding to C3b and stabilization of the C3b-dependent C3 convertase. The Journal of Experimental Medicine 142: 856-863. 12. Rawal N, Pangburn MK (2001) Structure/function of C5 convertases of complement. International Immunopharmacology 1: 415-422. 13. Forneris F, Ricklin D, Wu J, Tzekou A, Wallace RS, et al. (2010) Structures of C3b in complex with factors B and D give insight into complement convertase formation. Science 330: 1816-1820. 14. Manthey HD, Woodruff TM, Taylor SM, Monk PN (2009) (C5a). The International Journal of Biochemistry & Cell Biology 41: 2114-2117. 15. Hadders MA, Beringer DX, Gros P (2007) Structure of C8alpha-MACPF reveals mechanism of membrane attack in complement immune defense. Science 317: 1552-1554. 16. Figueroa J, Andreoni J, Densen P (1993) Complement deficiency states and meningococcal disease. Immunologic Research 12: 295-311. 17. Tegla CA, Cudrici C, Patel S, Trippe R, 3rd, Rus V, et al. (2011) Membrane attack by complement: the assembly and biology of terminal complement complexes. Immunologic Research 51: 45-60. 18. Zipfel PF, Reuter M (2009) Complement Activation Products C3a and C4a as Endogenous Antimicrobial Peptides. International Journal of Peptide Research and Therapeutics 15: 87-95. 19. Kishore U, Reid KB (2000) C1q: structure, function, and receptors. Immunopharmacology 49: 159-170. 20. Ziccardi RJ, Tschopp J (1982) The Dissociation Properties of Native C1. Biochemical and Biophysical Research Communications 107: 618-623. 21. Roumenina LT, Ruseva MM, Zlatarova A, Ghai R, Kolev M, et al. (2006) Interaction of C1q with IgG1, C-reactive protein and pentraxin 3: mutational studies using recombinant globular head modules of human C1q A, B, and C chains. Biochemistry 45: 4093-4104. 22. Roumenina LT, Popov KT, Bureeva SV, Kojouharova M, Gadjeva M, et al. (2008) Interaction of the globular domain of human C1q with Salmonella typhimurium lipopolysaccharide. Biochimica et Biophysica Acta 1784: 1271-1276. 23. Sim RB, Kishore U, Villiers CL, Marche PN, Mitchell DA (2007) C1q binding and complement activation by prions and amyloids. Immunobiology 212: 355-362. 24. Paidassi H, Tacnet-Delorme P, Garlatti V, Darnault C, Ghebrehiwet B, et al. (2008) C1q binds phosphatidylserine and likely acts as a multiligand-

67 bridging molecule in apoptotic cell recognition. Journal of Immunology 180: 2329-2338. 25. Paidassi H, Tacnet-Delorme P, Lunardi T, Arlaud GJ, Thielens NM, et al. (2008) The lectin-like activity of human C1q and its implication in DNA and apoptotic cell recognition. FEBS Letters 582: 3111-3116. 26. Garlatti V, Chouquet A, Lunardi T, Vives R, Paidassi H, et al. (2010) Cutting edge: C1q binds deoxyribose and heparan sulfate through neighboring sites of its recognition domain. Journal of Immunology 185: 808-812. 27. Silvestri L, Baker JR, Roden L, Stroud RM (1981) The Clq Inhibitor in Serum Is a Chondroitin 4-Sulfate Proteoglycan. Journal of Biological Chemistry 256: 7383-7387. 28. Kirschfink M, Blase L, Engelmann S, Schwartz-Albiez R (1997) Secreted chondroitin sulfate proteoglycan of human lines binds to the complement protein C1q and inhibits complex formation of C1. Journal of Immunology 158: 1324-1331. 29. Ghebrehiwet B, Galanakis DK (1993) C1q inhibitor (chondroitin-4-sulfate proteoglycan): structure and function. Behring Institute Mitteilungen: 214-223. 30. Ghebrehiwet B, Silvestri L, McDevitt C (1984) Identification of the Raji cell membrane-derived C1q inhibitor as a receptor for human C1q. Purification and immunochemical characterization. The Journal of Experimental Medicine 160: 1375-1389. 31. Ghebrehiwet B, Lim BL, Peerschke EI, Willis AC, Reid KB (1994) Isolation, cDNA cloning, and overexpression of a 33-kD cell surface glycoprotein that binds to the globular "heads" of C1q. The Journal of Experimental Medicine 179: 1809-1821. 32. Klickstein LB, Barbashov SF, Liu T, Jack RM, Nicholson-Weller A (1997) Complement receptor type 1 (CR1, CD35) is a receptor for C1q. Immunity 7: 345-355. 33. Nepomuceno RR, HenschenEdman AH, Burgess WH, Tenner AJ (1997) cDNA cloning and primary structure analysis of C1qR(p), the human C1q/MBL/SPA receptor that mediates enhanced phagocytosis in vitro. Immunity 6: 119-129. 34. Hughesjones NC, Gardner B (1978) Reaction between Complement Subcomponent C1q, IgG Complexes and Poly-Ionic Molecules. Immunology 34: 459-463. 35. Gaboriaud C, Juanhuix J, Gruez A, Lacroix M, Darnault C, et al. (2003) The crystal structure of the globular head of complement protein C1q provides a basis for its versatile recognition properties. Journal of Biological Chemistry 278: 46974-46982. 36. Botto M, Walport MJ (2002) C1q, and apoptosis. Immunobiology 205: 395-406.

68 37. Cook HT, Botto M (2006) Mechanisms of disease: the complement system and the pathogenesis of systemic lupus erythematosus. Nature Clinical Practice Rheumatology 2: 330-337. 38. Botto M, Dell'Agnola C, Bygrave AE, Thompson EM, Cook HT, et al. (1998) Homozygous C1q deficiency causes glomerulonephritis associated with multiple apoptotic bodies. Nature genetics 19: 56-59. 39. Zarkadis IK, Mastellos D, Lambris JD (2001) Phylogenetic aspects of the complement system. Developmental and comparative Immunology 25: 745-762. 40. Janssen BJ, Huizinga EG, Raaijmakers HC, Roos A, Daha MR, et al. (2005) Structures of complement component C3 provide insights into the function and evolution of immunity. Nature 437: 505-511. 41. Gros P, Milder FJ, Janssen BJ (2008) Complement driven by conformational changes. Nature Reviews Immunology 8: 48-58. 42. Davis AE, Harrison RA, Lachmann PJ (1984) Physiologic Inactivation of Fluid Phase C3b - Isolation and Structural-Analysis of C3c, C3d,g (Alpha-2d), and C3g. Journal of Immunology 132: 1960-1966. 43. Eggleton P, Tenner AJ, Reid KB (2000) C1q receptors. Clinical and Experimental Immunology 120: 406-412. 44. Cornacoff JB, Hebert LA, Smead WL, VanAman ME, Birmingham DJ, et al. (1983) Primate erythrocyte-immune complex-clearing mechanism. The Journal of Clinical Investigation 71: 236-247. 45. Mollnes TE, Brekke OL, Fung M, Fure H, Christiansen D, et al. (2002) Essential role of the C5a receptor in E. coli-induced oxidative burst and phagocytosis revealed by a novel lepirudin-based human whole blood model of inflammation. Blood 100: 1869-1877. 46. Ross GD, Lambris JD (1982) Identification of a C3bi-Specific Membrane Complement Receptor That Is Expressed on Lymphocytes, Monocytes, Neutrophils, and Erythrocytes. Journal of Experimental Medicine 155: 96-110. 47. Dempsey PW, Allison MED, Akkaraju S, Goodnow CC, Fearon DT (1996) C3d of complement as a molecular adjuvant: Bridging innate and acquired immunity. Science 271: 348-350. 48. Ehrengruber MU, Geiser T, Deranleau DA (1994) Activation of human neutrophils by C3a and C5a. Comparison of the effects on shape changes, , secretion, and respiratory burst. FEBS Letters 346: 181-184. 49. Klos A, Tenner AJ, Johswich KO, Ager RR, Reis ES, et al. (2009) The role of the anaphylatoxins in health and disease. Molecular Immunology 46: 2753-2766. 50. Inal JM, Hui KM, Miot S, Lange S, Ramirez MI, et al. (2005) Complement C2 receptor inhibitor trispanning: a novel human complement inhibitory receptor. Journal of Immunology 174: 356-366.

69 51. Davis AE, 3rd, Mejia P, Lu F (2008) Biological activities of C1 inhibitor. Molecular Immunology 45: 4057-4063. 52. Blom AM, Villoutreix BO, Dahlback B (2003) Mutations in alpha-chain of C4BP that selectively affect its factor I cofactor function. The Journal of Biological Chemistry 278: 43437-43442. 53. Meri S, Pangburn MK (1990) Discrimination between activators and nonactivators of the alternative pathway of complement: regulation via a sialic acid/polyanion on factor H. Proceedings of the National Academy of Sciences of the United States of America 87: 3982- 3986. 54. Lambris JD, Ricklin D, Geisbrecht BV (2008) Complement evasion by human pathogens. Nature Reviews Microbiology 6: 132-142. 55. Kim DD, Song WC (2006) Membrane complement regulatory proteins. Clinical Immunology 118: 127-136. 56. de Cordoba SR, Esparza-Gordillo J, de Jorge EG, Lopez-Trascasa M, Sanchez-Corral P (2004) The human complement factor H: functional roles, genetic variations and disease associations. Molecular Immunology 41: 355-367. 57. Schmidt CQ, Herbert AP, Hocking HG, Uhrin D, Barlow PN (2008) Translational mini-review series on complement factor H: structural and functional correlations for factor H. Clinical and Experimental Immunology 151: 14-24. 58. Kuhn S, Skerka C, Zipfel PF (1995) Mapping of the complement regulatory domains in the human factor H-like protein 1 and in factor H1. Journal of Immunology 155: 5663-5670. 59. Jokiranta TS, Cheng ZZ, Seeberger H, Jozsi M, Heinen S, et al. (2005) Binding of complement factor H to endothelial cells is mediated by the carboxy-terminal glycosaminoglycan binding site. The American Journal of 167: 1173-1181. 60. Oppermann M, Manuelian T, Jozsi M, Brandt E, Jokiranta TS, et al. (2006) The C-terminus of complement regulator Factor H mediates target recognition: evidence for a compact conformation of the native protein. Clinical and Experimental Immunology 144: 342-352. 61. Blackmore TK, Sadlon TA, Ward HM, Lublin DM, Gordon DL (1996) Identification of a heparin binding domain in the seventh short consensus repeat of complement factor H. Journal of Immunology 157: 5422-5427. 62. Ormsby RJ, Jokiranta TS, Duthy TG, Griggs KM, Sadlon TA, et al. (2006) Localization of the third heparin-binding site in the human complement regulator factor H1. Molecular Immunology 43: 1624-1632. 63. Saito A, Munakata H (2007) Analysis of plasma proteins that bind to glycosaminoglycans. Biochimica et Biophysica Acta 1770: 241-246. 64. Gama CI, Tully SE, Sotogaku N, Clark PM, Rawat M, et al. (2006) Sulfation patterns of glycosaminoglycans encode molecular recognition and activity. Nature Chemical Biology 2: 467-473.

70 65. Jokiranta TS, Zipfel PF, Hakulinen J, Kuhn S, Pangburn MK, et al. (1996) Analysis of the recognition mechanism of the alternative pathway of complement by monoclonal anti-factor H antibodies: evidence for multiple interactions between H and surface bound C3b. FEBS Letters 393: 297-302. 66. Schmidt CQ, Herbert AP, Kavanagh D, Gandy C, Fenton CJ, et al. (2008) A new map of glycosaminoglycan and C3b binding sites on factor H. Journal of Immunology 181: 2610-2619. 67. Skerka C, Zipfel PF (2008) Complement factor H related proteins in immune diseases. 26: I9-I14. 68. Blom AM, Kask L, Dahlback B (2001) Structural requirements for the complement regulatory activities of C4BP. The Journal of Biological Chemistry 276: 27136-27144. 69. Hardig Y, Dahlback B (1996) The amino-terminal module of the C4b- binding protein beta-chain contains the protein S-binding site. The Journal of Biological Chemistry 271: 20861-20867. 70. Schwalbe R, Dahlback B, Hillarp A, Nelsestuen G (1990) Assembly of Protein-S and C4b-Binding Protein on Membranes. Journal of Biological Chemistry 265: 16074-16081. 71. Hessing M, Vlooswijk RA, Hackeng TM, Kanters D, Bouma BN (1990) The localization of heparin-binding fragments on human C4b-binding protein. Journal of Immunology 144: 204-208. 72. Wagner DD, Burger PC (2003) Platelets in inflammation and thrombosis. Arteriosclerosis Thrombosis and Vascular Biology 23: 2131-2137. 73. English D, Garcia JG, Brindley DN (2001) Platelet-released phospholipids link haemostasis and angiogenesis. Cardiovascular Research 49: 588- 599. 74. Malara A, Balduini A (2012) Blood platelet production and morphology. Thrombosis Research. 75. Daly ME (2011) Determinants of platelet count in humans. Haematologica 96: 10-13. 76. Rendu F, Brohard-Bohn B (2001) The platelet release reaction: granules' constituents, secretion and functions. Platelets 12: 261-273. 77. Coppinger JA, O'Connor R, Wynne K, Flanagan M, Sullivan M, et al. (2007) Moderation of the platelet releasate response by . Blood 109: 4786-4792. 78. Gorbet MB, Sefton MV (2004) Biomaterial-associated thrombosis: roles of coagulation factors, complement, platelets and leukocytes. Biomaterials 25: 5681-5703. 79. Schmugge M, Rand ML, Freedman J (2003) Platelets and von Willebrand factor. Transfusion and Science: Official Journal of the World Apheresis Association: Official Journal of the European Society for Haemapheresis 28: 269-277.

71 80. Chen JM, Lopez KA (2005) Interactions of platelets with subendothelium and endothelium. Microcirculation 12: 235-246. 81. Clemetson KJ (2011) Platelets and Primary Haemostasis. Thrombosis Research. 82. Gachet C (2001) ADP receptors of platelets and their inhibition. Thrombosis and Haemostasis 86: 222-232. 83. Zwaal RFA, Comfurius P, Bevers EM (2005) Surface exposure of phosphatidylserine in pathological cells. Cellular and Molecular Life Sciences 62: 971-988. 84. Covic L, Gresser AL, Kuliopulos A (2000) Biphasic kinetics of activation and signaling for PAR1 and PAR4 thrombin receptors in platelets. Biochemistry 39: 5458-5467. 85. Monroe D, Hoffman M, Roberts HR (2002) Platelets and thrombin generation. Arteriosclerosis Thrombosis and Vascular Biology 22: 1381- 1389. 86. Fox JEB, Austin CD, Boyles JK, Steffen PK (1990) Role of the Membrane Skeleton in Preventing the Shedding of Procoagulant-Rich from the Platelet Plasma-Membrane. Journal of Cell Biology 111: 483- 493. 87. Sinauridze EI, Kireev DA, Popenko NY, Pichugin AV, Panteleev MA, et al. (2007) Platelet microparticle membranes have 50-to 100-fold higher specific procoagulant activity than activated platelets. Thrombosis and Haemostasis 97: 425-434. 88. Perez-Pujol S, Marker PH, Key NS (2007) Platelet microparticles are heterogeneous and highly dependent on the activation mechanism: Studies using a new digital flow cytometer. Cytometry Part A 71A: 38- 45. 89. Nesbitt WS, Westein E, Tovar-Lopez FJ, Tolouei E, Mitchell A, et al. (2009) A shear gradient-dependent platelet aggregation mechanism drives thrombus formation. Nature Medicine 15: 665-U146. 90. Brass LF (2003) Thrombin and platelet activation. Chest 124: 18S-25S. 91. Elrod JW, Park JH, Oshima T, Sharp CD, Minagar A, et al. (2003) Expression of junctional proteins in human platelets. Platelets 14: 247- 251. 92. Falati S, Liu QD, Gross P, Merrill-Skoloff G, Chou J, et al. (2003) Accumulation of tissue factor into developing thrombi in vivo is dependent upon microparticle P-selectin glycoprotein ligand 1 and platelet P-selectin. Journal of Experimental Medicine 197: 1585-1598. 93. Celi A, Pellegrini G, Lorenzet R, Deblasi A, Ready N, et al. (1994) P- Selectin Induces the Expression of Tissue Factor on Monocytes. Proceedings of the National Academy of Sciences of the United States of America 91: 8767-8771. 94. Gawaz M, Langer H, May AE (2005) Platelets in inflammation and atherogenesis. Journal of Clinical Investigation 115: 3378-3384.

72 95. Hawrylowicz CM, Howells GL, Feldmann M (1991) Platelet-Derived Interleukin-1 Induces Human Endothelial Adhesion Molecule Expression and Production. Journal of Experimental Medicine 174: 785-790. 96. Henn V, Slupsky JR, Grafe M, Anagnostopoulos I, Forster R, et al. (1998) CD40 ligand on activated platelets triggers an inflammatory reaction of endothelial cells. Nature 391: 591-594. 97. Gawaz M, Brand K, Dickfeld T, Pogatsa-Murray G, Page S, et al. (2000) Platelets induce alterations of chemotactic and adhesive properties of endothelial cells mediated through an interleukin-1-dependent mechanism. Implications for atherogenesis. 148: 75-85. 98. Scheuerer B, Ernst M, Durrbaum-Landmann I, Fleischer J, Grage-Griebenow E, et al. (2000) The CXC-chemokine platelet factor 4 promotes monocyte survival and induces monocyte differentiation into macrophages. Blood 95: 1158-1166. 99. Schober A, Manka D, von Hundelshausen P, Huo YQ, Hanrath P, et al. (2002) Deposition of platelet RANTES triggering monocyte recruitment requires P-selectin and is involved in neointima formation after arterial injury. Circulation 106: 1523-1529. 100. Jurk K, Kehrel BE (2005) Platelets: physiology and biochemistry. Seminars in Thrombosis and Hemostasis 31: 381-392. 101. Kaczmarek E, Koziak K, Sevigny J, Siegel JB, Anrather J, et al. (1996) Identification and characterization of CD39 vascular ATP diphosphohydrolase. Journal of Biological Chemistry 271: 33116-33122. 102. Jin RC, Voetsch B, Loscalzo J (2005) Endogenous mechanisms of inhibition of platelet function. Microcirculation 12: 247-258. 103. Wang GR, Zhu Y, Halushka PV, Lincoln TM, Mendelsohn ME (1998) Mechanism of platelet inhibition by nitric oxide: In vivo phosphorylation of by cyclic GMP-dependent protein kinase. Proceedings of the National Academy of Sciences of the United States of America 95: 4888-4893. 104. Rosen P, Schwippert B, Kaufmann L, Tschope D (1994) Expression of Adhesion Molecules on the Surface of Activated Platelets Is Diminished by PGI(2) Analogs and an NO (EDRF) Donor - a Comparison between Platelets of Healthy and Diabetic Subjects. Platelets 5: 45-52. 105. Gries A, Bode C, Peter K, Herr A, Bohrer H, et al. (1998) Inhaled nitric oxide inhibits human platelet aggregation, P-selectin expression, and fibrinogen binding in vitro and in vivo. Circulation 97: 1481-1487. 106. Johnston-Cox HA, Yang D, Ravid K (2011) Physiological Implications of Adenosine Receptor-Mediated Platelet Aggregation. Journal of Cellular Physiology 226: 46-51. 107. Koszalka P, Ozuyaman B, Huo YQ, Zernecke A, Flogel U, et al. (2004) Targeted disruption of CD73/ecto-5 '-nucleotidase alters

73 and augments vascular inflammatory response. Circulation Research 95: 814-821. 108. Slupsky JR, Kalbas M, Willuweit A, Henn V, Kroczek RA, et al. (1998) Activated platelets induce tissue factor expression on human umbilical vein endothelial cells by ligation of CD40. Thrombosis and Haemostasis 80: 1008-1014. 109. Furie B, Furie BC (2004) Role of platelet P-selectin and microparticle PSGL-1 in thrombus formation. Trends in Molecular Medicine 10: 171-178. 110. Losche W (2005) Platelets and tissue factor. Platelets 16: 313-319. 111. Drake TA, Morrissey JH, Edgington TS (1989) Selective Cellular Expression of Tissue Factor in Human-Tissues - Implications for Disorders of Hemostasis and Thrombosis. American Journal of Pathology 134: 1087-1097. 112. van der Meijden PEJ, Munnix ICA, Auger JM, Govers-Riemslag JWP, Cosemans JMEM, et al. (2009) Dual role of collagen in factor XII- dependent thrombus formation. Blood 114: 881-890. 113. Kannemeier C, Shibamiya A, Nakazawa F, Trusheim H, Ruppert C, et al. (2007) Extracellular RNA constitutes a natural procoagulant cofactor in blood coagulation. Proceedings of the National Academy of Sciences of the United States of America 104: 6388-6393. 114. Joseph K, Shibayama Y, Ghebrehiwet B, Kaplan AP (2001) Factor XII- dependent contact activation on endothelial cells and binding proteins gC1qR and cytokeratin 1. Thrombosis and Haemostasis 85: 119-124. 115. Back J, Sanchez J, Elgue G, Ekdahl KN, Nilsson B (2010) Activated human platelets induce factor XIIa-mediated contact activation. Biochemical and Biophysical Research Communications 391: 11-17. 116. Muller F, Mutch NJ, Schenk WA, Smith SA, Esterl L, et al. (2009) Platelet Are Proinflammatory and Procoagulant Mediators In Vivo. Cell 139: 1143-1156. 117. Kaplan AP, Silverberg M, Dunn JT, Ghebrehiwet B (1982) Interaction of the Clotting, Kinin-Forming, Complement, and Fibrinolytic Pathways in Inflammation. Annals of the New York Academy of Sciences 389: 25- 38. 118. Pike RN, Buckle AM, le Bonniec BF, Church FC (2005) Control of the coagulation system by serpins - Getting by with a little help from glycosaminoglycans. Febs Journal 272: 4842-4851. 119. Cugno M, Bos I, Lubbers Y, Hack CE, Agostoni A (2001) In vitro interaction of C1-inhibitor with thrombin. Blood Coagulation & 12: 253-260. 120. Brummel KE, Paradis SG, Butenas S, Mann KG (2002) Thrombin functions during tissue factor-induced blood coagulation. Blood 100: 148-152.

74 121. Back J, Lang MH, Elgue G, Kalbitz M, Sanchez J, et al. (2009) Distinctive regulation of contact activation by antithrombin and C1-inhibitor on activated platelets and material surfaces. Biomaterials 30: 6573-6580. 122. Broze GJ, Miletich JP (1987) Characterization of the Inhibition of Tissue Factor in Serum. Blood 69: 150-155. 123. Diepenhorst GM, van Gulik TM, Hack CE (2009) Complement-mediated ischemia-reperfusion injury: lessons learned from animal and clinical studies. Ann Surg 249: 889-899. 124. Stahl GL, Xu Y, Hao L, Miller M, Buras JA, et al. (2003) Role for the alternative complement pathway in ischemia/reperfusion injury. Am J Pathol 162: 449-455. 125. Walsh MC, Bourcier T, Takahashi K, Shi L, Busche MN, et al. (2005) Mannose-binding lectin is a regulator of inflammation that accompanies myocardial ischemia and reperfusion injury. J Immunol 175: 541-546. 126. Zhang M, Michael LH, Grosjean SA, Kelly RA, Carroll MC, et al. (2006) The role of natural IgM in myocardial ischemia-reperfusion injury. J Mol Cell Cardiol 41: 62-67. 127. Iltumur K, Karabulut A, Toprak G, Toprak N (2005) Complement activation in acute coronary syndromes. Apmis 113: 167-174. 128. Langlois PF, Gawryl MS (1988) Detection of the Terminal Complement Complex in Patient Plasma Following Acute Myocardial-Infarction. Atherosclerosis 70: 95-105. 129. Yasuda M, Kawarabayashi T, Akioka K, Teragaki M, Oku H, et al. (1989) The Complement-System in the Acute Phase of Myocardial-Infarction. Japanese Circulation Journal-English Edition 53: 1017-1022. 130. Loirat C, Fremeaux-Bacchi V (2011) Atypical hemolytic uremic syndrome. Orphanet J Rare Dis 6: 60. 131. Hillmen P, Lewis SM, Bessler M, Luzzatto L, Dacie JV (1995) Natural- History of Paroxysmal-Nocturnal Hemoglobinuria. New England Journal of Medicine 333: 1253-1258. 132. Gralnick HR, Vail M, McKeown LP, Merryman P, Wilson O, et al. (1995) Activated platelets in paroxysmal nocturnal haemoglobinuria. Br J Haematol 91: 697-702. 133. Noris M, Remuzzi G (2009) Medical Progress Atypical Hemolytic-Uremic Syndrome. New England Journal of Medicine 361: 1676-1687. 134. Stahl AL, Vaziri-Sani F, Heinen S, Kristoffersson AC, Gydell KH, et al. (2008) Factor H dysfunction in patients with atypical hemolytic uremic syndrome contributes to complement deposition on platelets and their activation. Blood 111: 5307-5315. 135. Manzi S, Rairie JE, Carpenter AB, Kelly RH, Jagarlapudi SP, et al. (1996) Sensitivity and specificity of plasma and complement split products as indicators of lupus disease activity. Arthritis and Rheumatism 39: 1178-1188.

75 136. Fernandez M, Alarcon GS, Apte M, Andrade RM, Vila LM, et al. (2007) Systemic lupus erythematosus in a multiethnic US cohort XLIII. The significance of thrombocytopenia as a prognostic factor. Arthritis and Rheumatism 56: 614-621. 137. Kwok SK, Ju JH, Cho CS, Kim HY, Park SH (2009) Thrombotic thrombocytopenic purpura in systemic lupus erythematosus: risk factors and clinical outcome: a single centre study. Lupus 18: 16-21. 138. Munnix ICA, Cosemans JMEM, Auger JM, Heemskerk JWM (2009) Platelet response heterogeneity in thrombus formation. Thrombosis and Haemostasis 102: 1149-1156. 139. del Conde I, Cruz MA, Zhang H, Lopez JA, Afshar-Kharghan V (2005) Platelet activation leads to activation and propagation of the complement system. Journal of Experimental Medicine 201: 871-879. 140. Peerschke EIB, Yin W, Grigg SE, Ghebrehiwet B (2006) Blood platelets activate the classical pathway of human complement. Journal of Thrombosis and Haemostasis 4: 2035-2042. 141. Peerschke EIB, Reid KBM, Ghebrehiwet B (1994) Identification of a Novel 33-Kda C1q-Binding Site on Human Blood-Platelets. Journal of Immunology 152: 5896-5901. 142. Peerschke EIB, Murphy TK, Ghebrehiwet B (2003) Activation-dependent surface expression of gClqR/p33 on human blood platelets. Thrombosis and Haemostasis 89: 331-339. 143. Yin W, Ghebrehiwet B, Peerschke EIB (2008) Expression of complement components and inhibitors on platelet microparticles. Platelets 19: 225- 233. 144. Polley MJ, Nachman R (1978) Human Complement-System in Thrombin- Mediated Platelet-Function. Journal of Experimental Medicine 147: 1713-1726. 145. Amara U, Flierl MA, Rittirsch D, Klos A, Chen H, et al. (2010) Molecular Intercommunication between the Complement and Coagulation Systems. Journal of Immunology 185: 5628-5636. 146. Huber-Lang M, Sarma JV, Zetoune FS, Rittirsch D, Neff TA, et al. (2006) Generation of C5a in the absence of C3: a new complement activation pathway. Nature Medicine 12: 682-687. 147. Nicholson-Weller A, March JP, Rosen CE, Spicer DB, Austen KF (1985) Surface-Membrane Expression by Human-Blood Leukocytes and Platelets of Decay-Accelerating Factor, a Regulatory Protein of the Complement-System. Blood 65: 1237-1244. 148. Yu GH, Holers VM, Seya T, Ballard L, Atkinson JP (1986) Identification of a 3rd Component of Complement-Binding Glycoprotein of Human- Platelets. Journal of Clinical Investigation 78: 494-501. 149. Morgan BP (1992) Isolation and Characterization of the Complement- Inhibiting Protein CD59 Antigen from Platelet Membranes. Biochemical Journal 282: 409-413.

76 150. Schmaier AH, Amenta S, Xiong T, Heda GD, Gewirtz AM (1993) Expression of Platelet C1 Inhibitor. Blood 82: 465-474. 151. Devine DV, Rosse WF (1987) Regulation of the Activity of Platelet-Bound C-3 Convertase of the Alternative Pathway of Complement by Platelet Factor-H. Proceedings of the National Academy of Sciences of the United States of America 84: 5873-5877. 152. Gnatenko DV, Dunn JJ, McCorkle SR, Weissmann D, Perrotta PL, et al. (2003) Transcript profiling of human platelets using microaray and serial analysis of . Blood 101: 2285-2293. 153. Wiedmer T, Ando B, Sims PJ (1987) Complement C5b-9-Stimulated Platelet Secretion Is Associated with a Ca-2+-Initiated Activation of Cellular Protein-Kinases. Journal of Biological Chemistry 262: 13674- 13681. 154. Sims PJ, Faioni EM, Wiedmer T, Shattil SJ (1988) Complement Proteins C5b-9 Cause Release of Membrane-Vesicles from the Platelet Surface That Are Enriched in the Membrane-Receptor for Coagulation Factor-Va and Express Prothrombinase Activity. Journal of Biological Chemistry 263: 18205-18212. 155. Sims PJ, Rollins SA, Wiedmer T (1989) Regulatory Control of Complement on Blood-Platelets - Modulation of Platelet Procoagulant Responses by a Membrane Inhibitor of the C5b-9 Complex. Journal of Biological Chemistry 264: 19228-19235. 156. Peerschke EIB, Reid KBM, Ghebrehiwet B (1993) Platelet Activation by C1q Results in the Induction of Alpha(Iib)/Beta(3) Integrins (Gpiib- Iiia) and the Expression of P-Selectin and Procoagulant Activity. Journal of Experimental Medicine 178: 579-587. 157. Polley MJ, Nachman RL (1983) Human-Platelet Activation by C3a and C3a Des-Arg. Journal of Experimental Medicine 158: 603-615. 158. Herndon ME, Stipp CS, Lander AD (1999) Interactions of neural glycosaminoglycans and proteoglycans with protein ligands: assessment of selectivity, heterogeneity and the participation of core proteins in binding. Glycobiology 9: 143-155. 159. Hacker U, Nybakken K, Perrimon N (2005) Heparan sulphate proteoglycans: The sweet side of development. Nature Reviews Molecular Cell Biology 6: 530-541. 160. Bradbury EJ, Moon LDF, Popat RJ, King VR, Bennett GS, et al. (2002) Chondroitinase ABC promotes functional recovery after spinal cord injury. Nature 416: 636-640. 161. Sasisekharan R, Shriver Z, Venkataraman G, Narayanasami U (2002) Roles of heparan-sulphate glycosaminoglycans in cancer. Nature Reviews Cancer 2: 521-528. 162. Taylor KR, Gallo RL (2006) Glycosaminoglycans and their proteoglycans: host-associated molecular patterns for initiation and modulation of inflammation. Faseb Journal 20: 9-22.

77 163. Kolset SO, Prydz K, Pejler G (2004) Intracellular proteoglycans. Biochemical Journal 379: 217-227. 164. Gandhi NS, Mancera RL (2008) The Structure of Glycosaminoglycans and their Interactions with Proteins. Chemical Biology & Drug Design 72: 455-482. 165. Prydz K, Dalen KT (2000) Synthesis and sorting of proteoglycans - Commentary. Journal of Cell Science 113: 193-205. 166. Gallagher JT, Turnbull JE, Lyon M (1992) Patterns of sulphation in heparan sulphate: polymorphism based on a common structural theme. The International Journal of Biochemistry 24: 553-560. 167. Lamari FN, Karamanos NK (2006) Structure of chondroitin sulfate. Advances in 53: 33-48. 168. Silbert JE, Sugumaran G (2002) Biosynthesis of chondroitin/dermatan sulfate. Iubmb Life 54: 177-186. 169. Casu B, Petitou M, Provasoli M, Sinay P (1988) Conformational Flexibility - a New Concept for Explaining Binding and Biological Properties of Iduronic Acid-Containing Glycosaminoglycans. Trends in Biochemical Sciences 13: 221-225. 170. Ihrcke NS, Platt JL (1996) Shedding of heparan sulfate proteoglycan by stimulated endothelial cells: Evidence for of cell-surface molecules. Journal of Cellular Physiology 168: 625-637. 171. Huntington JA (2003) Mechanisms of glycosaminoglycan activation of the serpins in hemostasis. Journal of thrombosis and haemostasis : JTH 1: 1535-1549. 172. Kolset SO, Pejler G (2011) Serglycin: A Structural and Functional Chameleon with Wide Impact on Immune Cells. Journal of Immunology 187: 4927-4933. 173. Hagman R, Ronnberg E, Pejler G (2009) Canine Uterine Bacterial Infection Induces Upregulation of Proteolysis-Related Genes and Downregulation of Homeobox and Zinc Finger Factors. PloS one 4. 174. Niemann CU, Abrink M, Pejler G, Fischer RL, Christensen EI, et al. (2007) Neutrophil depends on serglycin proteoglycan for localization in granules. Blood 109: 4478-4486. 175. Sawesi O, Spillmann D, Lunden A, Wernersson S, Abrink M (2010) Serglycin-independent Release of Active Proteases in Response to Toxoplasma gondii Infection. Journal of Biological Chemistry 285: 38005-38013. 176. Skliris A, Happonen KE, Terpos E, Labropoulou V, Borset M, et al. (2011) Serglycin inhibits the classical and lectin pathways of complement via its glycosaminoglycan chains: Implications for multiple myeloma. European Journal of Immunology 41: 437-449. 177. Perin JP, Bonnet F, Maillet P, Jolles P (1988) Characterization and N- terminal sequence of human platelet proteoglycan. The Biochemical Journal 255: 1007-1013.

78 178. Barber AJ, Kasergla.R, Jakabova M, Luscher EF (1972) Characterization of a Chondroitin 4-Sulfate Proteoglycan Carrier for Heparin Neutralizing Activity (Platelet Factor 4) Released from Human Blood-Platelets. Biochimica et Biophysica Acta 286: 312-329. 179. Woulfe DS, Lilliendahl JK, August S, Rauova L, Kowalska MA, et al. (2008) Serglycin proteoglycan deletion induces defects in platelet aggregation and thrombus formation in mice. Blood 111: 3458-3467. 180. Okayama M, Oguri K, Fujiwara Y, Nakanishi H, Yonekura H, et al. (1986) Purification and Characterization of Human-Platelet Proteoglycan. Biochemical Journal 233: 73-81. 181. Hagen I (1972) The release of glycosaminoglycans during exposure of human platelets to thrombin and polystyrene latex particles. Biochimica et Biophysica Acta 273: 141-148. 182. Ward JV, Packham MA (1979) Characterization of the Sulfated Glycosaminoglycan on the Surface and in the Storage Granules of Rabbit Platelets. Biochimica et Biophysica Acta 583: 196-207. 183. Donato JL, Nogueira MD, Marcondes S, Antunes E, Nader HB, et al. (1994) The Kinetics of Chondroitin 4-Sulfate Release from Stimulated Platelets and Its Relation to Thromboxane A(2) Formation and Granule Secretion. Brazilian Journal of Medical and Biological Research 27: 2163-2167. 184. Volpi N, Cusmano M, Venturelli T (1995) Qualitative and Quantitative Studies of Heparin and Chondroitin Sulfates in Normal Human Plasma. Biochimica et Biophysica Acta-General Subjects 1243: 49-58. 185. Marcus AJ, Safier LB, Hajjar KA, Ullman HL, Islam N, et al. (1991) Inhibition of Platelet-Function by an Aspirin-Insensitive Endothelial- Cell Adpase - Thromboregulation by Endothelial-Cells. Journal of Clinical Investigation 88: 1690-1696. 186. Zimmermann H (1999) Nucleotides and CD39: Principal modulatory players in hemostasis and thrombosis. Nature Medicine 5: 987-988. 187. Hechler B, Gachet C (2011) P2 receptors and platelet function. 7: 293-303. 188. Mahaut-Smith MP, Jones S, Evans RJ (2011) The P2X1 receptor and platelet function. Purinergic Signalling 7: 341-356. 189. Cattaneo M, Gachet C (1999) ADP receptors and clinical bleeding disorders. Arteriosclerosis Thrombosis and Vascular Biology 19: 2281- 2285. 190. Meadows TA, Bhatt DL (2007) Clinical aspects of platelet inhibitors and thrombus formation. Circulation Research 100: 1261-1275. 191. Savi P, Herbert JM (2005) Clopidogrel and ticlopidine: P2Y(12) adenosine diphosphate-receptor antagonists for the prevention of atherothrombosis. Seminars in Thrombosis and Hemostasis 31: 174-183. 192. Serebruany VL, Malinin AI, Eisert RM, Sane DC (2004) Risk of bleeding complications with antiplatelet agents: Meta-analysis of 338,191 patients

79 enrolled in 50 randomized controlled trials. American Journal of 75: 40-47. 193. Handa M, Guidotti G (1996) Purification and cloning of a soluble ATP- diphosphohydrolase (apyrase) from potato tubers (Solanum tuberosum). Biochemical and Biophysical Research Communications 218: 916-923. 194. Yang F, Hicks-Berger CA, Smith TM, Kirley TL (2001) Site-directed mutagenesis of human nucleoside triphosphate diphosphohydrolase 3: The importance of residues in the apyrase conserved regions. Biochemistry 40: 3943-3950. 195. Champagne DE (2005) Antihemostatic molecules from saliva of blood- feeding arthropods. Pathophysiology of Haemostasis and Thrombosis 34: 221-227. 196. Pinsky DJ, Broekman MJ, Peschon JJ, Stocking KL, Fujita T, et al. (2002) Elucidation of the thromboregulatory role of CD39/ectoapyrase in the ischemic brain. The Journal of Clinical Investigation 109: 1031-1040. 197. Belayev L, Khoutorova L, Deisher TA, Belayev A, Busto R, et al. (2003) Neuroprotective effect of SolCD39, a novel platelet aggregation inhibitor, on transient middle cerebral artery occlusion in rats. Stroke; a Journal of Cerebral Circulation 34: 758-763. 198. Buergler JM, Maliszewski CR, Broekman MJ, Kaluza GL, Schulz DG, et al. (2005) Effects of SolCD39, a novel inhibitor of platelet aggregation, on platelet deposition and aggregation after PTCA in a porcine model. Journal of Thrombosis and 19: 115-122. 199. Dai J, Liu J, Deng Y, Smith TM, Lu M (2004) Structure and protein design of a human platelet function inhibitor. Cell 116: 649-659. 200. Robson SC, Kaczmarek E, Siegel JB, Candinas D, Koziak K, et al. (1997) Loss of ATP diphosphohydrolase activity with endothelial cell activation. The Journal of Experimental Medicine 185: 153-163. 201. Imai M, Takigami K, Guckelberger O, Enjyoji K, Smith RN, et al. (1999) Modulation of nucleotide triphosphate diphosphohydrolase-1 (NTPDase-1)/CD39 in xenograft rejection. Molecular Medicine 5: 743- 752. 202. Cai M, Huttinger ZM, He H, Zhang WZ, Li F, et al. (2011) Transgenic over expression of ectonucleotide triphosphate diphosphohydrolase-1 protects against murine myocardial ischemic injury. Journal of Molecular and Cellular Cardiology 51: 927-935. 203. Wheeler DG, Joseph ME, Mahamud SD, Aurand WL, Mohler PJ, et al. (2012) Transgenic swine: Expression of human CD39 protects against myocardial injury Wheeler, CD39 reduces myocardial injury in swine. Journal of Molecular and Cellular Cardiology. 204. Imai M, Takigami K, Guckelberger O, Kaczmarek E, Csizmadia E, et al. (2000) Recombinant adenoviral mediated CD39 gene transfer prolongs cardiac xenograft survival. Transplantation 70: 864-870.

80 205. Dwyer KM, Robson SC, Nandurkar HH, Campbell DJ, Gock H, et al. (2004) Thromboregulatory manifestations in human CD39 transgenic mice and the implications for thrombotic disease and transplantation. Journal of Clinical Investigation 113: 1440-1446. 206. Crikis S, Lu B, Murray-Segal LM, Selan C, Robson SC, et al. (2010) Transgenic Overexpression of CD39 Protects Against Renal Ischemia- Reperfusion and Transplant Vascular Injury. American Journal of Transplantation 10: 2586-2595. 207. Williams DF (2009) On the nature of biomaterials. Biomaterials 30: 5897- 5909. 208. Ratner BD (2007) A paradigm shift: biomaterials that heal. Polymer International 56: 1183-1185. 209. Doyle B, Rihal CS, O'Sullivan CJ, Lennon RJ, Wiste HJ, et al. (2007) Outcomes of stent thrombosis and restenosis during extended follow-up of patients treated with bare-metal coronary stents. Circulation 116: 2391-2398. 210. Zimmermann J, Herrlinger S, Pruy A, Metzger T, Wanner C (1999) Inflammation enhances cardiovascular risk and mortality in hemodialysis patients. Kidney International 55: 648-658. 211. Yao Q, Axelsson J, Stenvinkel P, Lindholm B (2004) Chronic systemic inflammation in dialysis patients: An update on causes and consequences. Asaio Journal 50: Lii-Lvii. 212. Andersson J, Ekdahl KN, Lambris JD, Nilsson B (2005) Binding of C3 fragments on top of adsorbed plasma proteins during complement activation on a model biomaterial surface. Biomaterials 26: 1477-1485. 213. Vogler EA (2012) Protein adsorption in three dimensions. Biomaterials 33: 1201-1237. 214. Bamford CH, Cooper SL, Tsuruta T (1991) Festschrift in Honor of the 75th Birthday of Dr. Leo Vroman. Foreword. Journal of Biomaterials Science-Polymer Edition 3: R5-R5. 215. Engberg AE, Rosengren-Holmberg JP, Chen H, Nilsson B, Lambris JD, et al. (2011) -polymer adsorption: Implications for understanding complement-mediated hemoincompatibility. Journal of Biomedical Materials Research Part A 97A: 74-84. 216. Lundqvist M, Stigler J, Elia G, Lynch I, Cedervall T, et al. (2008) Nanoparticle size and surface properties determine the protein corona with possible implications for biological impacts. Proceedings of the National Academy of Sciences of the United States of America 105: 14265-14270. 217. Ekdahl KN, Lambris JD, Elwing H, Ricklin D, Nilsson PH, et al. (2011) Innate immunity activation on biomaterial surfaces: A mechanistic model and coping strategies. Advanced Drug Delivery Reviews 63: 1042-1050.

81 218. Andersson J, Ekdahl KN, Larsson R, Nilsson UR, Nilsson B (2002) C3 adsorbed to a polymer surface can form an initiating alternative pathway convertase. Journal of Immunology 168: 5786-5791. 219. Tengvall P, Askendal A, Lundstrom I (2001) Ellipsometric in vitro studies on the activation of complement by human immunoglobulins M and G after adsorption to methylated silicon. Colloids and Surfaces B- Biointerfaces 20: 51-62. 220. Hong J, Azens A, Ekdahl KN, Granqvist CG, Nilsson B (2005) Material- specific thrombin generation following contact between metal surfaces and whole blood. Biomaterials 26: 1397-1403. 221. Arvidsson S, Askendal A, Tengvall P (2007) Blood plasma contact activation on silicon, titanium and aluminium. Biomaterials 28: 1346- 1354. 222. Vogler EA, Siedlecki CA (2009) Contact activation of blood-plasma coagulation. Biomaterials 30: 1857-1869. 223. Zhuo R, Siedlecki CA, Vogler EA (2006) Autoactivation of blood factor XII at hydrophilic and hydrophobic surfaces. Biomaterials 27: 4325- 4332. 224. Shiba E, Lindon JN, Kushner L, Matsueda GR, Hawiger J, et al. (1991) Antibody-Detectable Changes in Fibrinogen Adsorption Affecting Platelet Activation on Polymer Surfaces. American Journal of Physiology 260: C965-C974. 225. Sivaraman B, Latour RA (2010) The relationship between platelet adhesion on surfaces and the structure versus the amount of adsorbed fibrinogen. Biomaterials 31: 832-839. 226. Sivaraman B, Latour RA (2010) The Adherence of platelets to adsorbed albumin by receptor-mediated recognition of binding sites exposed by adsorption-induced unfolding. Biomaterials 31: 1036-1044. 227. Kourtzelis I, Markiewski MM, Doumas M, Rafail S, Kambas K, et al. (2010) Complement anaphylatoxin C5a contributes to hemodialysis- associated thrombosis. Blood 116: 631-639. 228. Yang Y, Leong KW (2010) Nanoscale surfacing for regenerative medicine. Wiley Interdisciplinary Reviews-Nanomedicine and Nanobiotechnology 2: 478-495. 229. Ferraz N, Ott MK, Hong J (2010) Time Sequence of Blood Activation by Nanoporous Alumina: Studies on Platelets and Complement System. Microscopy Research and Technique 73: 1101-1109. 230. Elbert DL, Hubbell JA (1996) Surface treatments of polymers for biocompatibility. Annual Review of Materials Science 26: 365-394. 231. Wang YX, Robertson JL, Spillman WB, Claus RO (2004) Effects of the chemical structure and the surface properties of polymeric biomaterials on their biocompatibility. Pharmaceutical Research 21: 1362-1373.

82 232. Higuchi A, Sugiyama K, Yoon BO, Sakurai M, Hara M, et al. (2003) Serum protein adsorption and platelet adhesion on pluronic-adsorbed polysulfone membranes. Biomaterials 24: 3235-3245. 233. Arima Y, Toda M, Iwata H (2008) Complement activation on surfaces modified with ethylene glycol units. Biomaterials 29: 551-560. 234. Kopp R, Mottaghy K, Kirschfink M (2002) Mechanism of complement activation during extracorporeal blood-biomaterial interaction: Effects of heparin coated and uncoated surfaces. Asaio Journal 48: 598-605. 235. Christensen K, Larsson R, Emanuelsson H, Elgue G, Larsson A (2001) Heparin coating of the stent graft - effects on platelets, coagulation and complement activation. Biomaterials 22: 349-355. 236. Andersson J, Sanchez J, Ekdahl KN, Elgue G, Nilsson B, et al. (2003) Optimal heparin surface concentration and antithrombin binding capacity as evaluated with human non-anticoagulated blood in vitro. Journal of Biomedical Materials Research Part A 67: 458-466. 237. Ovrum E, Tangen G, Tollofsrud S, Skeie B, Ringdal MAL, et al. (2011) Heparinized cardiopulmonary bypass circuits and low systemic anticoagulation: An analysis of nearly 6000 patients undergoing coronary artery bypass grafting. Journal of Thoracic and Cardiovascular 141: 1145-1149. 238. Hsu LC (2001) Heparin-coated cardiopulmonary bypass circuits: current status. Perfusion 16: 417-428. 239. Lindholt JS, Gottschalksen B, Johannesen N, Dueholm D, Ravn H, et al. (2011) The Scandinavian Propaten((R)) trial - 1-year patency of PTFE vascular prostheses with heparin-bonded luminal surfaces compared to ordinary pure PTFE vascular prostheses - a randomised clinical controlled multi-centre trial. European Journal of Vascular and Endovascular Surgery: the Official Journal of the European Society for Vascular Surgery 41: 668-673. 240. Andersson J, Larsson R, Richter R, Ekdahl KN, Nilsson B (2001) Binding of a model regulator of complement activation (RCA) to a biomaterial surface: surface-bound factor H inhibits complement activation. Biomaterials 22: 2435-2443. 241. Andersson J, Bexborn F, Klinth J, Nilsson B, Ekdahl KN (2006) Surface- attached PEO in the form of activated pluronic with immobilized factor H reduces both coagulation and complement activation in a whole-blood model. Journal of Biomedical Materials Research Part A 76A: 25-34. 242. Watkins NJ, Braidley P, Bray CJ, Savill CM, White DJG (1997) Coating of human decay accelerating factor (hDAF) onto medical devices to improve biocompatibility. Immunopharmacology 38: 111-118. 243. Wu YQ, Qu HC, Sfyroera G, Tzekou A, Kay BK, et al. (2011) Protection of Nonself Surfaces from Complement Attack by Factor H-Binding Peptides: Implications for Therapeutic Medicine. Journal of Immunology 186: 4269-4277.

83 244. Engberg AE, Sandholm K, Bexborn F, Persson J, Nilsson B, et al. (2009) Inhibition of complement activation on a model biomaterial surface by streptococcal M protein-derived peptides. Biomaterials 30: 2653-2659. 245. Lambris JD, Ricklin D, Geisbrecht BV (2008) Complement evasion by human pathogens. Nature Reviews Microbiology 6: 132-142. 246. Eich T, Eriksson O, Lundgren T (2007) Visualization of early engraftment in clinical islet transplantation by positron-emission tomography. New England Journal of Medicine 356: 2754-2755. 247. Bennet W, Sundberg B, Groth CG, Brendel MD, Brandhorst D, et al. (1999) Incompatibility between human blood and isolated islets of Langerhans - A finding with implications for clinical intraportal islet transplantation? Diabetes 48: 1907-1914. 248. Bennet W, Bjorkland A, Sundberg B, Brandhorst D, Brendel MD, et al. (2001) Expression of complement regulatory proteins on islets of langerhans: A comparison between human islets and islets isolated from normal and hDAF transgenic pigs'. Transplantation 72: 312-319. 249. Tjernberg J, Ekdahl KN, Lambris JD, Korsgren O, Nilsson B (2008) Acute antibody-mediated complement activation mediates lysis of pancreatic islets cells and may cause tissue loss in clinical islet transplantation. Transplantation 85: 1193-1199. 250. Moberg L, Johansson H, Lukinius A, Berne C, Foss A, et al. (2002) Production of tissue factor by pancreatic islet cells as a trigger of detrimental thrombotic reactions in clinical islet transplantation. Lancet 360: 2039-2045. 251. Nilsson B, Korsgren O, Lambris JD, Ekdahl KN (2010) Can cells and biomaterials in therapeutic medicine be shielded from innate immune recognition? Trends in Immunology 31: 32-38. 252. Goto M, Tjernberg J, Dufrane D, Elgue G, Brandhorst D, et al. (2008) Dissecting the instant blood-mediated inflammatory reaction in islet . Xenotransplantation 15: 225-234. 253. Cabric S, Eich T, Sanchez J, Nilsson B, Ikorsgren O, et al. (2008) A new method for incorporating functional heparin onto the surface of islets of Langerhans. Tissue Engineering Part C-Methods 14: 141-147. 254. Cabric S, Sanchez J, Lundgren T, Foss A, Felldin M, et al. (2007) Islet surface heparinization prevents the instant blood-mediated inflammatory reaction in islet transplantation. Diabetes 56: 2008-2015. 255. Johnell M, Larsson R, Siegbahn A (2005) The influence of different heparin surface concentrations and antithrombin-binding capacity on inflammation and coagulation. Biomaterials 26: 1731-1739. 256. Bexborn F, Engberg AE, Sandholm K, Mollnes TE, Hong J, et al. (2009) versus heparin for use in whole blood in vitro biocompatibility models. Journal of Biomedical Materials Research Part A 89A: 951-959. 257. Stone SR, Hofsteenge J (1986) Kinetics of the Inhibition of Thrombin by Hirudin. Biochemistry 25: 4622-4628.

84 258. Wouters D, Wiessenberg HD, Hart M, Bruins P, Voskuyl A, et al. (2005) Complexes between C1q and C3 or C4: Novel and specific markers for classical complement pathway activation. Journal of Immunological Methods 298: 35-45. 259. Katragadda M, Magotti P, Sfyroera G, Lambris JD (2006) Hydrophobic effect and hydrogen bonds account for the improved activity of a complement inhibitor, compstatin. Journal of Medicinal Chemistry 49: 4616-4622. 260. Paczkowski NJ, Finch AM, Whitmore JB, Short AJ, Wong AK, et al. (1999) Pharmacological characterization of antagonists of the C5a receptor. British Journal of Pharmacology 128: 1461-1466. 261. Des Prez RM, Bryan CS, Hawiger J, Colley DG (1975) Function of the classical and alternate pathways of human complement in serum treated with ethylene glycol tetraacetic acid and MgCl2-ethylene glycol tetraacetic acid. Infection and Immunity 11: 1235-1243. 262. Deepa SS, Yamada S, Fukui S, Sugahara K (2007) Structural determination of novel sulfated octasaccharides isolated from chondroitin sulfate of shark cartilage and their application for characterizing monoclonal antibody epitopes. Glycobiology 17: 631-645. 263. Jackson R, Basinger S, Werth J (1970) Use of a logistic function in expressing kinetic hemolysis data. Infection and Immunity 1: 142-145. 264. Osmond RI, Kett WC, Skett SE, Coombe DR (2002) Protein-heparin interactions measured by BIAcore 2000 are affected by the method of heparin immobilization. Analytical Biochemistry 310: 199-207. 265. Hong J, Ekdahl KN, Reynolds H, Larsson R, Nilsson B (1999) A new in vitro model to study interaction between whole blood and biomaterials. Studies of platelet and coagulation activation acid the effect of aspirin. Biomaterials 20: 603-611. 266. Stenberg E, Persson B, Roos H, Urbaniczky C (1991) Quantitative- Determination of Surface Concentration of Protein with Surface-Plasmon Resonance Using Radiolabeled Proteins. Journal of Colloid and Interface Science 143: 513-526. 267. Myszka DG, Morton TA (1998) CLAMP (c): A biosensor kinetic data analysis program. Trends in Biochemical Sciences 23: 149-150. 268. Markiewski MM, Nilsson B, Ekdahl KN, Mollnes TE, Lambris JD (2007) Complement and coagulation: strangers or partners in crime? Trends in Immunology 28: 184-192. 269. Kalowski S, Howes EL, Jr., Margaretten W, McKay DG (1975) Effects of intravascular clotting on the activation of the complement system: The role of the platelet. The American Journal of Pathology 78: 525-536. 270. Almeda S, Rosenberg RD, Bing DH (1983) The Binding-Properties of Human-Complement Component C1q - Interaction with Mucopolysaccharides. Journal of Biological Chemistry 258: 785-791.

85 271. Weiler JM, Edens RE, Linhardt RJ, Kapelanski DP (1992) Heparin and modified heparin inhibit complement activation in vivo. Journal of Immunology 148: 3210-3215. 272. Robinson DE, Buttle DJ, Short RD, McArthur SL, Steele DA, et al. (2012) Glycosaminoglycan (GAG) binding surfaces for characterizing GAG- protein interactions. Biomaterials 33: 1007-1016. 273. Yu HN, Munoz EM, Edens RE, Linhardt RJ (2005) Kinetic studies on the interactions of heparin and complement proteins using surface plasmon resonance. Biochimica et Biophysica Acta-General Subjects 1726: 168- 176. 274. Peerschke EI, Ghebrehiwet B (1992) Platelet interactions with C1q in whole blood and in the presence of immune complexes or aggregated IgG. Clinical Immunology and Immunopathology 63: 45-50. 275. Puram V, Giuliani D, Morse BS (1984) Circulating Immune-Complexes and Platelet IgG in Various Diseases. Clinical and Experimental Immunology 58: 672-676. 276. Nepomuceno RR, Henschen-Edman AH, Burgess WH, Tenner AJ (1997) cDNA cloning and primary structure analysis of C1qR(p), the human C1q/MBL/SPA receptor that mediates enhanced phagocytosis in vitro. Immunity 6: 119-129. 277. Mnjoyan Z, Li J, Afshar-Kharghan V (2008) Factor H binds to platelet integrin IIb3. Platelets 19: 512-519. 278. Vaziri-Sani F, Hellwage J, Zipfel PF, Sjoholm AG, Iancu R, et al. (2005) Factor H binds to washed human platelets. Journal of Thrombosis and Haemostasis 3: 154-162. 279. Takeda J, Miyata T, Kawagoe K, Iida Y, Endo Y, et al. (1993) Deficiency of the GPI Anchor Caused by a Somatic Mutation of the PIG-A Gene in Paroxysmal-Nocturnal Hemoglobinuria. Cell 73: 703-711. 280. Dixon RH, Rosse WF (1977) Mechanism of complement-mediated activation of human blood platelets in vitro: comparison of normal and paroxysmal nocturnal hemoglobinuria platelets. The Journal of Clinical Investigation 59: 360-368. 281. de Cordoba SR (2010) aHUS: a disorder with many risk factors. Blood 115: 158-160. 282. Sartz L, Olin AI, Kristoffersson AC, Stahl AL, Johansson ME, et al. (2012) A Novel C3 Mutation Causing Increased Formation of the C3 Convertase in Familial Atypical Hemolytic Uremic Syndrome. Journal of Immunology 188: 2030-2037. 283. Cines DB, Bussel JB, Liebman HA, Prak ETL (2009) The ITP syndrome: pathogenic and clinical diversity. Blood 113: 6511-6521. 284. Najaoui A, Bakchoul T, Stoy J, Bein G, Rummel MJ, et al. (2012) -mediated complement activation on platelets is a common finding in patients with immune thrombocytopenic purpura (ITP). European Journal of Haematology 88: 167-174.

86 285. Myers TJ, Kim BK, Steiner M, Baldini MG (1982) Platelet-associated complement C3 in immune thrombocytopenic purpura. Blood 59: 1023- 1028. 286. Heptinstall S, Johnson A, Glenn JR, White AE (2005) Adenine nucleotide metabolism in human blood - important roles for leukocytes and erythrocytes. Journal of Thrombosis and Haemostasis 3: 2331-2339. 287. Eltzschig HK, Thompson LF, Karhausen J, Cotta RJ, Ibla JC, et al. (2004) Endogenous adenosine produced during hypoxia attenuates neutrophil accumulation: coordination by extracellular nucleotide metabolism. Blood 104: 3986-3992. 288. Yang MY, Kirley TL (2008) Engineered human soluble calcium-activated nucleotidase inhibits coagulation in vitro and thrombosis in vivo. Thrombosis Research 122: 541-548. 289. Bakker WW, Vanderlei B, Nieuwenhuis P, Robinson P, Bartels HL (1991) Reduced Thrombogenicity of Artificial Materials by Coating with Adpase. Biomaterials 12: 603-606. 290. Marconi W, Bartoli F, Mantovani E, Pittalis F, Settembri L, et al. (1979) Development of new antithrombogenic surfaces by employing platelet antiaggregating agents: preparation and characterization. Transactions - American Society for Artificial Internal Organs 25: 280-286. 291. Cazenave JP, Ohlmann P, Cassel D, Eckly A, Hechler B, et al. (2004) Preparation of washed platelet suspensions from human and rodent blood. Methods in Molecular Biology 272: 13-28. 292. Schneider MC, Prosser BE, Caesar JJE, Kugelberg E, Li S, et al. (2009) Neisseria meningitidis recruits factor H using protein mimicry of host carbohydrates. Nature 458: 890-U899. 293. Harboe M, Ulvund G, Vien L, Fung M, Mollnes TE (2004) The quantitative role of alternative pathway amplification in classical pathway induced terminal complement activation. Clinical and Experimental Immunology 138: 439-446. 294. Persson J, Beall B, Linse S, Lindahl G (2006) Extreme sequence divergence but conserved ligand-binding specificity in Streptococcus pyogenes M protein. PLoS Pathogens 2: 442-452. 295. Sandrin MS, Vaughan HA, Dabkowski PL, Mckenzie IFC (1993) Anti-Pig IgM Antibodies in Human Serum React Predominantly with Gal(Alpha- 1-3)Gal Epitopes. Proceedings of the National Academy of Sciences of the United States of America 90: 11391-11395.

87