Comprehensive Summaries of Uppsala Dissertationsfrom the Faculty of Medicine 1253

Complement Activation Triggeredby Biomaterial Surfaces

Mechanisms and Regulation

BY

JONAS ANDERSSON

ACTA UNIVERSITATIS UPSALIENSISUPPSALA 2003

Papers included in the thesis

This thesis is based on the following papers, which are referred to in the text by their Roman numerals:

I Andersson, Jonas ; Nilsson Ekdahl, Kristina ; Nilsson, Bo: Complement activation on a model biomaterial surface: Binding of C3b via the alternative pathway amplification loop to plasma proteins adsorbed to the surface. (manuscript)

II Andersson, Jonas ; Nilsson Ekdahl, Kristina ; Larsson, Rolf ; Nilsson, Ulf R. ; Nilsson, Bo: C3 Adsorbed to a Polymer Surface Can Form an Initiating Alternative Pathway Convertase. Journal of Immunology, 168(2002): no. 11, 5786-5791 (published)

III Andersson, Jonas ; Larsson, Rolf ; Richter, Ralf ; Nilsson Ekdahl, Kristina ; Nilsson, Bo: Binding of a model regulator of complement activation (RCA) to a biomaterial surface: surface-bound factor H inhibits complement activation. Biomaterials, 22(2001): no. 17, 2435-2443 (published)

IV Andersson, Jonas ; Sanchez, Javier ; Nilsson Ekdahl, Kristina ; Elgue, Graciela ; Nilsson, Bo ; Larsson, Rolf: Optimal heparin surface concentration and antithrombin binding capacity as evaluated with human non-antigoagulated blood iin vitrro. Journal of Biomedical Materials Research (accepted)

6

Contents

Introduction........................................................................................................11

Biomaterials .......................................................................................................13Surface characteristics ..................................................................................13Surface modifications ...................................................................................14

Biologically active surface modifications ...............................................14

Blood..................................................................................................................16Cells of the blood..........................................................................................16

Erythrocytes .............................................................................................16Platelets.....................................................................................................16Leukocytes................................................................................................17

Blood plasma and serum...............................................................................17

The Complement System ..................................................................................19Complement Component 3...........................................................................20The Classical Pathway..................................................................................22The Lectin Pathway ......................................................................................22The Alternative Pathway ..............................................................................22The Terminal Pathway..................................................................................23Complement Regulation ...............................................................................23

Factor H....................................................................................................24Complement receptors ..................................................................................25Monitoring Complement Activation ............................................................26Therapeutic Complement Inhibitors.............................................................27

Blood-Biomaterial Interactions .........................................................................29Adsorption of plasma proteins......................................................................29Conformations of adsorbed C3.....................................................................29Cascade Systems of Blood and Biomaterials...............................................30Adsorption of whole blood...........................................................................30

Complement Activation on the Biomaterial Surface........................................32Complement activation in vitro....................................................................32Complement activation upon blood-biomaterial contact in vivo ................33

Extracorporeal circulation........................................................................33Hemodialysis .......................................................................................33

7

Cardiopulmonary Bypass....................................................................34Heparin surfaces in vivo ..........................................................................35

Aims of the Investigation ..................................................................................36

Materials and Methods ......................................................................................37Enzyme Immunoassays ................................................................................37

ß-Thromboglobulin ..................................................................................37Thrombin-Antithrombin complexes........................................................37Soluble C5b-9 complexes ........................................................................37C3a............................................................................................................38

Quartz Crystal Microbalance with Dissipation Monitoring ........................38Factor H Immobilization...............................................................................39The Corline Heparin Surface........................................................................39The Chandler Loop Model ...........................................................................39

Experiments and Results ...................................................................................41Paper I ...........................................................................................................41Paper II ..........................................................................................................42Paper III.........................................................................................................42Paper IV.........................................................................................................43

Discussion ..........................................................................................................44

Conclusions........................................................................................................48

Acknowledgements............................................................................................49

References..........................................................................................................51

8

Abbreviations

AP Alternative pathway of complementAT Antithrombinß-TG ß-ThromboglobulinC Complement componentC1INH C1 inhibitorC3(H2O) Inactivated C3C3b,Bb AP C3 convertaseC3b,Bb,C3b AP C5 convertaseC4b,C2a CP C3 convertaseC4b,C2a,C3b CP C5 convertaseC5b-9 Membrane attack complexcC1q Collagen domain of C1qCD Cluster of differentiationCP Classical pathway of complementD DissipationEIA Enzyme immunoassayf FrequencyFHL Factor H-like proteinFHR Factor H-related proteingC1q Globular domain of C1qHAE Hereditary angioedemaiC3 Inactivated C3LP Lectin pathway of complementMAC Membrane attack complexMASP MBL-associated serine proteaseMBL Mannan-binding lectinP Properdin, factor PPEO Poly(ethylene oxide)PMN Polymorphonuclear granulocytePNH Paroxysmal nocturnal hemoglobinureaPPO Poly(propylene oxide)PRP Platelet-rich plasmaPVC Poly(vinyl chloride)R ReceptorRCA Regulator of complement activationRHP Rheumatoid arthritis protein

9

SCR Short consensus repeatSPDP N-succinimidyl 3-(2-pyridoldithio)

propionateSPR Surface plasmon resonancesC5b-9 Soluble membrane attack complexTAT Thrombin-antithrombin complexQCM-D Quartz crystal microbalance with

dissipation monitoring

10

11

Introduction

Today there are a vast number of medical devices that come in temporary orpermanent contact with tissues of the human body. The purpose of thesedevices is to treat or overcome various disease states, but to be of benefit tothe patient it is of major importance that the advantages of their useoutweigh any adverse reaction produced by their contact with the patient’stissue.

The majority of medical devices make contact with blood either duringimplantation or when functioning. Some examples of common blood-contactapplications are blood pumps, oxygenators and hemodialyzers. Stents,vascular grafts, miniature pumps, sensors, pacemakers and heart valves arepermanently in contact with blood. These devices are constructed fromdiverse types of materials that in many respects have little or nothing incommon. All these materials intended to interact with biological systems are,however, often referred to as biomaterials (1). This term indicates only thatthe particular material is used in contact with the body and says nothingabout whether that material is a good choice for the application.

The term biocompatibility is used to describe the ability of a material toperform with an appropriate host response in a specific application (1). Thisis not an absolute definition, and how biocompatibility should be assesseddiffers from one application to another. For example, what is a highlybiocompatible material in terms of bone integration might be an unsuitablematerial for applications involving contact with blood.

Upon blood-biomaterial contact there is an instantaneous adsorption ofproteins. The protein pattern of this adsorption depends on the particularmaterial, and this event decides the outcome of the contact between bloodand biomaterial. Since blood is important in our defense against foreigninvaders, there is a great risk that a synthetic material will provoke anundesirable response. This response starts with activation of the cascadesystems of blood, the most important ones being the coagulation,complement and contact systems. Activation of these systems will inevitablyresult in recruitment and activation of cells, which in turn are able to releaseinflammatory mediators (2).

The main focus of this work is on the initial blood-biomaterial contactand how the complement system is activated. Both the mechanisms that

12

control this initiation of complement activation and possible ways to regulatethis activation have been studied.

13

Biomaterials

There are many different types of materials that are used as biomaterials.The most important are ceramics, composites, metals and polymers. In bloodcontact-related applications there is also a wide variety of materials in use,although polymers and metals are the most common. For example, polymers,and particularly poly(vinyl chloride), are used in tubing intended forextracorporeal circulation ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !. Membranes in hemodialyzers are often made ofregenerated cellulose, but chemically modified celluloses and completelysynthetic membranes are also used (3-5). Oxygenator membranes are mostoften made from polypropylene (6, 7). Many types of implants are used incontact with blood. Artificial heart valves are typically made of pyrolyticcarbon or titanum (8). Stents used to improve the result from angioplasty aremostly made of stainless steel, titanium and metal alloys (9).

Surface characteristicsThe interaction between blood and biomaterials first takes place at thematerial surface. It is obvious that the properties of the material surfacedecide the outcome of this meeting. The surface can be considered from aphysical, a chemical and a biological point of view.

The mechanical demands on a biomaterial vary widely among the variousapplications. Propertis like hardness, tribology and ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! roughness must alwaysbe taken into account when choosing the proper material. ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !In terms ofchemical properties, hydrophobicity is important, but the presence ofspecific chemical groups is also significant.

There are not many surfaces that are biologically active, but this is agrowing field of great interest. By binding biologically active molecules tothe surface it is possible to prevent unfavorable reactions from taking place.This strategy might also allow specific cell types, such as endothelial cells,to bind to ligands immobilized on the biomaterial surface.

14

Surface modificationsThere are many established methods for optimizing the hardness andtribology of biomaterials. During the last decade there has been a focus onconstructing materials with nanostructures, i.e., patterns on the nm scale.Patterns on the surface include pits, holes, stripes, pyramids and manyothers. There is also the possibility of combining this physical pattern with achemical pattern (e.g., hydrophobic patches on a hydrophilic surface). Theaim of this strategy is to achieve surfaces that will encourage proteins toadhere in a desirable conformation and cells to grow in a controlled manner.

One way to decrease protein binding is to make a hydrophobic surfacemore hydrophilic by binding poly(ethylene oxide) (PEO) to it (10). PEO is aflexible polymer, and when long PEO chains are effectively bound to asurface, they create a steric barrier that prevents proteins from reaching thesurface . An additional advantage of this approach is that proteins on thesurface tend to remain in the native state rather than becoming partlydenatured. This protection from denaturation can be achieved by covalentlybinding PEO to the surface or by adsorption of block copolymers containingone or more PEO blocks and at least one poly(propylene oxide) (PPO) block(11). The use of block copolymers results in adsorption of the PPO block tothe hydrophobic surface and leaves PEO blocks pointing into the aqueoussolution. Copolymers of this type are called Pluronics, and previous studieshave shown that Pluronic F108 is the most effective for reducing proteinadsorption (12). The copolymer approach is advantageous because it can beapplied to a number of different materials, and there is no need to createfunctional groups on the material to bind the PEO chains. Furthermore, ityields an even distribution of PEO chains. A limitation of this approach isthe risk of desorption of the copolymer when the surface is exposed tocertain solutions, such as blood, which contain lipids and lipoproteins (13).

Biologically active surface modificationsThe most familiar biologically active surface modification is the binding

of heparin to a surface. Heparin is present in the body as a proteoglycanwhere heparin is covalently attached to a protein core. Prepared in its solubleform, free from protein, this carbohydrate has been shown to be an efficientanticoagulant agent !(14). Antithrombin (AT) is a known serine proteaseinhibitor and inactivates a broad range of coagulation factors. Binding of ATto a specific pentasaccharide motif in heparin induces a conformationalchange accelerating inactivation of coagulation enzymes !(15). Immobilizedheparin has been demonstrated to modulate activation of coagulation at thesurface, but there is also an inhibitory effect on complement activation !(16-

15

18). Possibly, this is due to the binding of factor H, a soluble regulator ofcomplement activation, known to possess heparin-binding sites!(15, 19, 20).

The aim of immobilized heparin is to create a surface that mimics that ofendothelium in regard of blood compatibility. When considering heparinsurfaces you must keep in mind that there are many different methods ofimmobilization ! (21). It is of great importance that the AT-binding sequenceis not compromised in the immobilization procedure. Every type of heparinsurface must be tested for stability and biological activity.

Other types of bioactive surfaces are less commonly used. The end groupsof Pluronic F108, for example, can be chemically activated to bind anyprotein (22), allowing one to bind an inhibitor of complement or coagulationto a surface while simultaneously preventing nonspecific protein adsorption.

16

Blood

Although it is a fluid, blood is considered to be a body tissue. About 45% ofits volume is made up of cells, and the remainder is an aqueous solution,plasma, with a protein concentration of 80 mg/ml. The blood system fulfillsmany essential functions, such as distributing oxygen and nutrients. It alsoincludes systems that provide defense against foreign invaders and maintainthe integrity of the blood system (23, 24).

Cells of the bloodTo be able to fulfill all the requirements of the blood system, many differentcell types are required, all of which are specialized for different functions.

ErythrocytesThe main function of these cells, which lack a nucleus, is to transport oxygento the tissues of the body and to remove harmful carbon dioxide (24).Although often considered to be an inert cell type in other respects,erythrocytes have been shown to be of importance in the amplification ofcoagulation. In platelet-rich plasma (PRP), which is basically blood lackingerythrocytes and leukocytes, activation of coagulation is much harder totrigger than in whole blood or in PRP supplemented with erythrocytes (25).

PlateletsPlatelets also lack nuclei and are a crucial component in the formation of theclots induced by coagulation (23). Receptors for collagen and fibrin allowplatelets to attach to wounded vessel walls and the fibrin fibers of a clot (26,27). Degranulation of these bound platelets releases substances that are ableto activate other platelets in the vicinity (28-30). In addition, phospholipidsin the platelet cell membrane are able to act as cofactors for clotting factors(31).

17

LeukocytesThe main types of leukocytes are lymphocytes, monocytes and granulocytes.

Lymphocytes are involved in both innate and adaptive immunity: NKcells, involved in innate immunity, are able to fight tumors. T cells either killcells that they recognize as foreign or act as helper cells to stimulate effectorcells. B cells produce antibodies, the humoral arm of specific immunity.

Monocytes are large phagocytic cells that form part of the innate immunesystem. Upon migration from the blood into tissues they can be transformedinto macrophages.

Granulocytes, also known as polymorphonuclear cells (PMN), can bedivided into three different cell types: eosinophils, basophils and neutrophils.About 60% of the leukocytes and 95% of the granulocytes are neutrophils.This cell type is, together with monocytes, the most rapid responder uponcomplement activation and possesses a number of complement receptors. Itacts both by phagocytosis and by release of inflammatory mediators (32).

Blood plasma and serumPlasma is the fluid that remains when the cells are removed from blood. Thisfluid contains proteins and nutrients. Plasma has a protein concentration ofabout 80 mg/ml and is made up of thousands of different proteins, most ofthem present in trace amounts. These blood proteins can be divided intoseveral functional groups:

The first group is the proteins that maintain the osmotic pressure of theblood, making certain that water does not leak out of the vessel. All theproteins present in blood contribute to this effect, but since albumin is themost common protein, it is considered to be of greatest significance. Presentat 45 mg/ml in blood, albumin is also important as a transport protein.

Since transporting low-molecular weight compounds is a majorresponsibility of the blood system, it includes a number of transport proteins.Albumin is involved in the transportation of a wide variety of compounds,including fatty acids, bilirubin and calcium, copper and zinc ions. There arealso numerous specific transport proteins with defined ligands.

The bulk of the plasma proteins participate in the defense systems of theblood. These defense proteins are involved in the coagulation, thefibrinolytic, the contact and/or the complement system. Also, theimmunoglobulins are considered defense proteins. The defense proteins thatare part of a specific system do not function on their own: They can also bearranged as part of cascade systems. In a cascade system, proteins areactivated in a sequential manner to allow an initial activation to be amplified.

18

In many cases the term network system seems to be more appropriate, sincethere are more than sequential activations (32).

When working with complement it is most common to use serum insteadof plasma (33, 34). Serum is a plasma-like fluid in which the components ofthe coagulation system have been allowed to clot and have been removedtogether with the cells. This treatment does not significantly activatecomponents of the complement system.

19

The Complement System

The complement system is a part of the body’s innate immune system. Itsaim is to eliminate microorganisms and other foreign substances that enterthe body. About 30 soluble and membrane-bound proteins are part of thecomplement system; these include both effector and regulatory proteins (35-38).Because this part of our immune system is always active and does not needto be adapted before it is able to respond to an invader, it is able to provide arapid response. The defensive functions of the complement system areachieved through several effector mechanisms:

- Labeling of foreign substances (=opsonization);- Release of inflammatory mediators, anaphylatoxins- Mediating chemotaxis and release of anaphylatoxins;- Lysis of cells through pore formation.

The central protein in the complement system is complement component3 (C3). Activation of C3 can be achieved by any of three activationpathways: the classical pathway (CP), the alternative pathway (AP) and themannan-binding lectin pathway (LP). At the point of C3 activation, thesepathways converge and are able to trigger the terminal pathway, which mayresult in cell lysis.

It has been shown that the complement system is ancient in origin and byfar precedes the adaptive immune system. The major components ofadaptive immunity have been found only in the jawed vertebrates, such assharks, and in higher vertebrates (39). Components of the complementsystem have been found as early as in jawless deuterostomes, such as seaurchins and ascidians (40, 41). No complement genes, however, have beenfound in the genomes of Drosophila or Caenorhabditis elegans (42, 43).

20

Complement Component 3C3 is the central and most abundant protein of the complement system. This185-kDa protein is present in plasma at a concentration of 1 mg/ml, but sinceit is an acute-phase protein, levels can be elevated during inflammation (44).C3 consists of two chains, an alpha chain (110 kDa) and a beta chain (75kDa), that are connected by a disulfide bond (45). The main synthesis of C3takes place in the liver, but multiple sources outside the liver are also known(46-48). This multifunctional protein or fragments thereof is known tointeract with a number of different proteins (49)i. These proteins arecomplement receptors, complement components and regulators ofcomplement activation (RCA).

C3 belongs to the alpha2-macroglobulin superfamily, which also includesalpha2-macroglobulin, C4 and C5. All of these proteins except C5 contain aunique concealed thioester bond between a cysteine and a glutamic acidresidue (50). In native C3 the thioester bond is protected, but uponproteolytic activation by a C3 convertase the bond is exposed through aconformational change. The half-life of this metastable thioester bond, ableto bind both hydroxyl and amino groups, is in the range of microseconds inthe presence of water (51). The short half-life explains why binding onlyoccurs close to the site of activation. In fact, only a small fraction of theactivated C3b will bind before it is inactivated by water.

Upon activation of C3, a 9-kDa peptide fragment (C3a) is released fromthe alpha chain as a result of proteolysis by the C3 convertase (52). Thisconvertase is made available by any of the activation pathways. Theremaining part of C3, C3b, is conformationally changed and acquires newactivities as it becomes bound to the surface (53). It is now able to interactwith both components of the alternative pathway of complement and withRCAs (49). Interaction with factor I together with RCAs leads to cleavage,accompanied by additional conformational changes. The resulting iC3b isunable to participate in complement activation but is a known ligand forcomplement receptors (54-57). Further interaction with factor I and RCAsresults in formation of two different fragments, soluble C3c and surface-bound C3dg. Finally, only C3d may remain bound if C3g is released byother serine proteases at the inflammatory site (58).

A surface opsonized with C3b and iC3b is made attractive for phagocyticleukocytes that have receptors for activated C3, mainly neutrophils (55). C3aremains soluble and is a potent anaphylatoxin that attracts leukocytes andinduces release of inflammatory mediators (59).

21

Figure 1. Sequential cleavage of C3 by convertases and factor I.

22

The Classical PathwayThe name of the classical pathway refers to the fact that this was the firstpathway to be discovered. The CP consists of three different complementcomponents: the C1 complex (C1q, C1r2, C1s2), C4 and C2. Activationoccurs upon binding of C1 to the Fc portions of IgM or IgG already bound toan antigen. Activated C1 is able to cleave C4 into C4a and C4b. C4a is aweak anaphylatoxin, and C4b is homologous to C3b and binds to the surfaceclose to the site of activation. C2 binds to C4b and is then cleaved by C1 intoC2a and C2b. C2a remains complexed with C4b, and C2b is released. Thecomplex C4b,C2a forms the CP C3 convertase and promotes the proteolysisof C3. Some of the C3b will bind to C4b and form a new complex(C4b,C2a,C3b), which is called the classical C5 convertase (37, 60).

The Lectin PathwayThe LP is the most recently described pathway and is known to be activatedby certain carbohydrates on microbial surfaces. Mannan-binding lectin(MBL) binds to these carbohydrate motifs. Together with MBL are MBL-associated serine proteases (MASP), which are bound in a complex, byanalogy to the C1 complex. This complex is also able to form the CPconvertase, C4b,C2a. There are also indications of direct cleavage of C3(38).

The Alternative PathwayThe AP was discovered after the classical pathway and was first thought ofas an activation pathway alternative to the classical pathway. This pathwayincludes four complement components: C3, factor B, factor D and properdin(factor P). The AP is triggered either by covalently bound C3b generated bythe CP and LP or by soluble C3b-like C3, also called C3(H2O) or iC3, whichis formed when C3 is spontaneously hydrolyzed in the fluid phase. Thishydrolysis is a very slow process, and the half-life of the thiolester in nativeC3 is about 230 h. Factor B binds to C3b or C3(H2O) and is cleaved byfactor D. Surface-bound C3b and activated B (Bb) form the alternativepathway C3 convertase, C3b,Bb. Properdin binds to the convertase andenhances its stability; it also binds to C3b and enhances its association withfactor B (61). Since C3b participates in the C3 convertase, this associationprovides a powerful feedback loop. C3b that is generated through the AP can

23

bind to the C3b of the convertase and form the C3b,Bb,C3b complex, alsocalled the alternative C5 convertase (36, 62, 63).

The Terminal PathwayComplement can mediate lysis of foreign microorganisms through theformation of a membrane attack complex (MAC), also known as C5b-9. Theclassical or alternative C5 convertase cleaves C5 into C5a and C5b. C5bforms a complex with C6, and when C7 is bound to the complex, ametastable binding site is exposed, making membrane insertion possible. C8and multiple copies of C9 are then sequentially associated with the complexto form a pore that is able to mediate lysis of the cell (64). Not all membraneattack complexes are inserted into the cell membrane: Sublyticconcentrations of the MAC are known to stimulate various cells, includingplatelets (65).

Complement RegulationRegulation of complement is needed to avoid uncontrolled activation ofcomplement on host cells. This control is exerted at the levels of C1activation, C3 convertase formation and, finally, MAC complex formation(35). A common motif in regulators of complement regulation (RCA) is theshort consensus repeat (SCR). One such repeat contains about 60 amino acidresidues, with the structure being stabilized by two disulfide bridges(66, 67).

C1 activation is controlled by C1 inhibitor, C1INH. It circulates incomplex with C1 in the blood, and a deficiency of this regulator results inhereditary angioedema (HAE) (68).

Regulation of convertase formation is mediated by two differentmechanisms: Surface-bound C3b and C4b can be inactivated by proteolyticcleavage, with their inactivated forms being unable to participate in theconvertase complex. This cleavage is mediated by factor I and requires oneof several cofactors (69). Some cofactors are able to participate inproteolytic inactivation of both C3b and C4b; others are able to act as acofactor for only a specific substrate. In addition, proteins with decayaccelerating activity dissociate the convertase or even prevent formation ofthe complex. This is achieved in a competetive manner, with the C3b or C4bbeing prevented from binding its ligand (35).

Finally, the control of MAC formation is mediated by the membrane-bound CD59. Through interference with the C5b-8 complex, C9 is preventedfrom binding in the polymeric fashion that is necessary to form the pore that

24

leads to cell lysis. CD59 is a GPI-anchored protein, and individuals unable toform GPI anchors have a high degree of hemolysis (35). This disorder iscalled paroxysmal nocturnal hemoglobinuria (PNH) and involves the lack ofabout 20 membrane-bound proteins, but the hemolysis seems to be causedby the lack of CD59 and DAF (70).

Membrane-bound regulators are, of course, present only on host cells, andsoluble regulators have a much higher affinity for autologous cells than formicroorganisms. This is one explanation for the ability of complement todistinguish between self and non-self. Soluble factor H, in particular, hasbeen shown to be important for this distinction. There are, however,examples of microorganisms that have developed affinity for RCAs as ameans of escaping complement (71).

Decay activity Cofactor activityRegulator Soluble ormembranebound

SCR

AP CP C3b C4b

C1INH soluble No - - - -DAF membrane Yes + + - -MCP membrane Yes - - + +CR1 membrane Yes + + + +Factor H soluble Yes + - + -C4bp soluble Yes - + - +CD59 membrane No - - - -

Factor HThis RCA with a molecular weight of 155 kDa is present in plasma at 0.2-0.6 mg/ml. Its importance in the complement system may be indicated by thefact that it is the second most common complement protein. The proteinconsists of twenty SCRs linked together as a string of pearls with thedimensions 495x34Å (72). Factor H acts as a cofactor for C3b cleavage andalso possesses decay accelerating activity. The N-terminal part of factor H isresponsible for both of these activities (73).

The higher affinity of factor H for autologous cells is one of the mostimportant factors responsible for complement discrimination between selfand non-self (71). This ability has been assigned to regions of the proteinthat are distinct from the N-terminal portion, which is responsible forcofactor and decay accelerating activity. These regions contain sites withaffinity for heparin and sialic acid, motifs that are found on endothelial cells

25

(19, 20, 74). Some pathogens are able to escape complement by expressingreceptors for complement regulators on their surface, thereby mimicking anautologous surface (75).

Plasma contains several proteins that are related to factor H. Reconectin,or factor H-like protein 1 (FHL-1), contains the seven most N-terminal SCRsof factor H and is able to act as an RCA. There are also a group of proteinscalled factor H-related proteins (FHR) that are unable to act as RCAs and areexpressed by individual genes, in contrast to FHL-1, which is analternatively spliced product of the factor H gene (76).

Figure 3. Factor H with 20 SCR domains. The lines below indicate regionsresponsible for different activities of the protein.

Complement receptorsThere are several different types of receptors for complement molecules.These are typically receptors for particular fragments of bound C3,anaphylatoxins or C1q.

The receptors for C3 fragments are named complement receptor (CR) 1-4.They have specificity for different fragments of C3 and are expressed ondifferent cell types. These differences allow for the proper response to bemediated at a specific moment of complement activation.

Receptors for the anaphylatoxins C5a, C4a and C3a mediate bothchemotaxis and activation of cells. The C5a receptor (C5aR) is the mostthoroughly investigated, since C5a is the most potent complementanaphylatoxin (77).

There are also receptors for C1q. Two different types of C1q receptors arefound in complex on the cell surface. They are directed against the globularand the collagen domains of C1q and are called globular C1q receptor(gC1qR) and collagen C1q receptor (cC1qR), respectively (78).

26

Receptor Receptor type Ligand

CR1 CD35, RCA C3b>C4b>iC3b monocytes, neutrophils, Band T cells, erythrocytes,eosinophils and FDCs

CR2 CD21, RCA C3d/C3dg>iC3b B-cells, FDCsCR3 CD11b/CD18,

IntegriniC3b monocytes, neutrophils,

and NK cellsCR4 CD11c/CD18,

IntegriniC3b monocytes, neutrophils,

and NK cellsC3aR G protein coupled

receptorC3a Neutrophils, eosinophils,

ECsC5aR G protein coupled

receptorC5a Mastcells, Neutrophils,

eosinophils, ECsgC1qR Collagen-like gC1q, HMWK,

fXII, fibrinogenB and T cells, neutrophils,fibroblasts, platelets, ECs

cC1qR - cC1q B and T cells, neutrophils,fibroblasts, platelets, ECs

Monitoring Complement ActivationComplement activity and activation can be measured in a number of ways(79). Activity refers to the presence of a functional complement system inthe blood, and activation refers to the traces of a previous event ofcomplement activation.

The activity of complement in blood is determined by using hemolyticassays specific for the alternative or the classical pathway. It is also possibleto measure the levels of various complement components in blood.

As an indication of complement activation, the generation of activationproducts in the blood plasma is important. The anaphylatoxins C3a, C4a andC5a are such products and are therefore obvious candidates. C5a is the mostpotent anaphylatoxin, 20 times stronger than C3a and 2500 times strongerthan C4a. Still, C5a is not a good indicator of activation, since it is rapidlybound to the C5aR. In contrast, the level of C3a is often assessed as anindicator of total complement activation using enzyme immunoassay (EIA)techniques (80). In addition, the formation of C4a can be used as an indicatorof CP activation. To measure AP activation, the liberated Ba fragment offactor B is a common choice. An alternative approach to assessing totalcomplement activation is to measure the level of the soluble MAC, sC5b-9(81-83). This assay is also an indirect way of following the formation ofC5a.

27

Therapeutic Complement InhibitorsFor many years there has been considerable interest in developingcomplement inhibitors for therapeutic treatment of complement activation.At this point, some complement inhibitors have been used clinically, but ithas only been during the last decade that major efforts have been made.Strategies for inhibiting complement activation can be divided into severalcategories:

The most obvious strategy is to use physiological complement inhibitorsalready present in the blood. C1INH purified from plasma has been used forthe treatment of HAE (68). There are also examples of other applications,such as myocardial infarctions and reperfusion injuries (84). C1INH is aserine protease inhibitor (serpin) that acts by covalently binding to the activesite of the protease (85).

Other natural inhibitors of complement have been tested as recombinantproteins. The first to be produced was soluble CR1, sCR1, which is able toinhibit both the alternative and classical convertases (86). In several studiessCR1 has been shown to be effective in preventing injuries caused byreperfusion, asthma and biomaterial-induced injuries (55, 87, 88).Recombinant and soluble forms of DAF and MCP have also been used toinhibit complement-induced injuries (89, 90). Both proteins were shown tobe active, but since they inhibit complement by different mechanisms, aMCP-DAF hybrid was shown to be more potent (91). Also, a soluble form ofCD59 has been shown to work in vitro (92).

Another strategy for stopping complement activation is to simply blockthe proteins to be activated using antibodies(93). To be of clinical use theseantibodies need to be humanized, and the Fc portion needs to be removed orshown not to activate complement (94). The most obvious targets to blockare C3 or C5, and there are examples of antibodies against both of theseproteins. Today a mAb against C5 developed by Alexion PharmaceuticalsInc. (Cheshire, CT, USA) might be the most promising complement inhibitorin clinical trials (95). Many other complement proteins has been tested astargets, including factor D and properdin. An antibody against properdin hasbeen shown to be efficient both in vivo and in vitro (96).

Production of therapeutics based on proteins is, however, both expensiveand difficult. For this reason there is growing interest in developing small-molecule inhibitors of complement. This group can be further divided basedon the mechanism of complement inhibition. There are anaphylatoxinreceptor antagonists, serine protease inhibitors and blocking molecules (97).Today there are C5aR antagonists shown to be working in vivo, but there isno C3aR antagonist (98, 99). There are also some serine protease inhibitors,many directed against factor D. Factor D is a good target because of its low

28

concentration and the fact that it circulates in its active form in the blood.The problem with serine protease inhibitors, however, is that with theseagents it is difficult to achieve the specificity that is needed (97). The small-molecule C3-blocking peptide compstatin has been shown to be efficientboth in vivo and in vitro.

Compstatin is a 13-residue cyclic peptide that was originally selectedusing a phage-displyed random peptide library (100). Since originallydescribed, this peptide has been modified to improve its stability and binding(101). Further studies have revealed that compstatin binds C3 and inhibitscleavage by the C3 convertase. Compstatin has been shown to be efficient inseveral studies. In an ex vivo model of xenograft rejection, survival oftransplanted kidneys was significantly prolonged (102, 103). In a model ofextracorporeal circulation, generation of C3a and sC5b-9 and binding of C3fragments to the surface were inhibited (104). Also, complement activationcaused by heparin-protamine complexes was blocked in baboons (105). Ithas also been shown that compstatin at low levels preferentially inhibits thealternative pathway of complement (104).

29

Blood-Biomaterial Interactions

Contact with blood is the initial event that occurs in extracorporealcirculation and when most implants are introduced into the body. Therefore,blood-biomaterial contact strongly influences the success of all types ofbiomaterials. This interaction is very complex, involving both proteins andcells.

Adsorption of plasma proteinsIt must be mentioned that before protein adsorption there is a very short timeperiod during which ions and water molecules of the plasma are becomingordered at the surface. This event is, however, immediately followed by theadsorption of proteins.

Even adsorption of a single protein can be a complex situation involvinginitial interaction with the surface, binding and conformational change in theprotein at the surface (106). When a mix of protein is adsorbed to the surfacethere is an additional level of complexity. The initial composition on thesurface depends on the concentration and size of the proteins involved.Following this initial binding there is a period of time in which an exchangeof proteins takes place to adapt the composition on the surface to the actualaffinity of the proteins. Larger proteins most often have a higher affinity thando smaller proteins that bind more rapidly. This process has been termed theVroman effect (107, 108). In plasma there are also other biomolecules suchas lipids and carbohydrates to take into account.

Conformations of adsorbed C3As mentioned previously, C3 changes its conformation when it is activatedin sequential steps by proteolysis (53). These conformational changes areaccompanied by changes in both the function of the protein and itsrecognition by receptors (49). A similar change in conformation can beachieved by denaturing soluble C3 in 3 mM SDS (109). Native C3 expressesboth unique native (N) epitopes and stable (S) epitopes. Denatured C3

30

expresses both unique epitopes that are specific for denatured (D) C3 andalso C3(S) epitopes. Upon denaturation the C3(N) epitopes are interchangedwith C3(D) neoepitopes (110, 111).

There is a considerable epitope similarity between denatured soluble C3and physiologically bound C3b. Bound C3b expresses C3(D), C3(N) andC3(S) epitopes, while SDS-denatured C3b expresses only C3(D) and C3(S)epitopes. Thus, C3b apparently undergoes a considerable conformationalchange upon activation and binding to a physiological surface (53, 112,113).

What is more interesting from the perspective of biomaterials is the factthat C3(D) epitopes also become exposed upon adsorption of both C3b andC3 to the plastic surface of a microtiter plate (114). This fact has led to thesuggestion that adsorbed C3 might be able to participate in an AP C3convertase (114, 115).

Cascade Systems of Blood and BiomaterialsIn addition to passive adsorption of proteins from plasma, there is also activebinding of proteins to the surface by the cascade systems of blood. Thecoagulation and complement systems are both activated by surfaces (17, 55).The coagulation system is propagated in the fluid phase and on cell surfaces,both in blood and bound to the surface. Complement is both activated andpropagated on the biomaterial surface (33).

Adsorption of whole bloodBlood cells express adhesion molecules and receptors on their cell surfaces.The repertoire of receptors differs among the various cell types and is alsodependent on the state of activation of the particular cell ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !. Both adsorbedproteins and proteins actively bound by the cascade systems are able to actas ligands for these receptors on the biomaterial surface (55, 116). Someproteins can act as ligands for these receptors in both in solution and whenadsorbed, but there are also proteins that change their conformation uponbinding and thereby increase their affinity for a particular receptor.

Platelets are activated by binding to a surface and will promote furtheractivation of the coagulation cascade, both by the release of substances thatpromote platelet activation and by binding of further platelets (117, 118).Also, the activated platelet surface itself is pro-coagulatory. Platelets bindfibrinogen already bound to the surface, a reaction that is mainly mediated

31

by the GPIIb/IIIa receptor (CD41/61) (116). The presence of C1qR onplatelets indicates the possibility of activation via C1q or fXII (119).

Polymorphonuclear leukocytes (PMNs) and monocytes bind viacomplement receptors and release inflammatory mediators induced byfrustrated phagocytosis. The main receptor responsible for this binding isCR3, which binds iC3b as well as ICAM-1 and fibrinogen (30, 120). Whencomplement is blocked by sCR1 in a model of cardiopulmonary bypass,binding of both PMNs and monocytes is inhibited. Platelet binding,however, remains unaffected (55).

32

Complement Activation on the BiomaterialSurface

Complement activation in vitroAt present, it is not understood how complement activation is triggered onthe biomaterial surface. Some common explanations are adsorption ofinitiating proteins, a lack of complement regulators on the surface or the C3tick-over process. Absence of complement regulators would explain the lackof control on the amplification loop rather than the actual initiation.

Activation of complement is initiated at certain sites at the surface.Deposition of C3 then spreads peripherally from this nucleus of activation(33). On surfaces pre-coated with IgG it has been shown that this C3deposition covers the surface and eventually masks the expression of IgGand C1q (121, 122). Even if an IgG-coated surface is an extremecomplement activator, this deposition is likely also to take place on a non-coated surface.

The general view of complement activation is that biomaterial surfacesactivate the AP (33, 123). This is supposed to be initiated by the tick-overprocess in the fluid phase (124-126). There is, however, a strong reportshowing that the lack of C4 will delay the start of activation for 15-25minutes upon contact with cuprophane membranes, suggesting theimportance of CP in biomaterial-induced complement activation (4).

Different biomaterial surfaces possess different complement-activatingproperties. Physical properties such as hydrophobicity are important, and ithas been shown that hydrophobic surfaces are more complement-activating(127), perhaps because of the conformation of adsorbed C3 on the surface(114). Also, the presence of different chemical groups on the surface isimportant (128). Htdroxyl and amino groups are known to bind activated C3.Recent studies, however, have shown that the surface is covered with a layerof protein and that C3 is actually bound to this layer (34). The chemicalgroups of the biomaterial surface are then covered and the importance of the

33

surface is rather to govern the composition and conformation of adsorbedproteins.

Complement activation upon blood-biomaterial contactin vivoIn the clinic, blood-biomaterial contact takes place in many differentsituations and involves many different types of materials (2). There is also agreat diversity in the amount of surface area being exposed: areas from somesquare centimeters to areas up to several square meters. The smaller areasinclude stents and artificial heart valves (129-133). These devices are notknown to cause any major complement activation. The risk is rather thatthese surfaces are able to act as initiation sites for thrombus formation (8).Local complement activation could, however, have a significant effect on thereactions of neighboring cells.

Extracorporeal circulationThe greatest challenge in the field of biomaterials is presented byextracorporeal circulation, when blood is circulated outside the body (134).Examples of extracorporeal circulation are cardiopulmonary bypass (CPB),membrane oxygenation and hemodialysis (134, 135). All of theseapplications involve the use of membranes to achieve gas exchange orremoval of low molecular weight substances from the blood. To achieveoptimal exchange, the area of these membranes can be at least two squaremeters. Although these procedures are usually accompanied by surgicaltrauma or a disease state, it has been shown that the complement-mediatedtissue injury is to a great extent a consequence of bioincompatibilityreactions toward the biomaterials used (136, 137).

HemodialysisHistorically, hemodialysis has been associated with severe, even fatal,anaphylactic reactions (138). These events are less frequent today but canstill take place. A concern in uremic patients undergoing maintainedhemodialysis is accelerated (139, 140). The risk of myocardial infarction inthese patients is 5-10 times higher than in healthy individuals (141). Manyfactors contribute to this increased risk: The uremic condition itself leads todisturbed lipoprotein levels, acidosis, toxins, free radicals and infections, allof which could contribute to accelerated arteriosclerosis (142, 143). Chronicinflammation, such as that found in patients with rheumatoid arthritis,systemic lupus erythematosus (SLE) and chronic rejections, has been shown

34

to be associated with accelerated arteriosclerosis and cardiovascular disease(144). Thus, it is likely that the chronic whole body inflammation triggeredby hemodialysis is another factor contributing to accelerated arteriosclerosisin uremic patients.

Dialysis membranes activate complement to varying degrees (137).Cellulosic membranes are known to be strong activators, while polymericmembranes are considered to be less active in activating complement. Thisactivation causes generation of membrane-bound C3 fragments (C3b, iC3band C3dg) as well as soluble peptides (C3a and C5a) (145, 146). All of thesefragments are ligands for leukocyte receptors and can trigger inflammationand release of cytokines. Complement activation also triggers up-regulationof cell-surface receptors on leukocytes (e.g., CD11b/CD18 and CD35),combined with down-regulation of L-selectin, making the cells more stickyand prone to interact with platelets and endothelial cells (104, 147). Also,this cellular activation leads to increased levels of acute-phase proteins suchas CRP, allowing these molecules to bind endothelial cells and activatecomplement. CRP and complement components C3, C4 and C9 are presentin atherosclerotic plaques, together with monocytes and macrophages (148,149).

Cardiopulmonary BypassCPB is associated with side effects triggered by a number of events,

including surgical trauma, blood-biomaterial contact, blood-gas contact,neutralization of heparin with protamine and also ischemia/reperfusion in thetissue disconnected during the procedure (heart and lungs) (150-152). Theincidence of stroke during or after CPB is 1-4% ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! (153). Another neurologicalproblem often associated with CPB is prolonged cognitive dysfunction(154) ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !. Many of these problems are apparently caused by complementactivation, a conclusion supported by in vivo studies using anti-C5antibodies. Complement inhibition achieved by blocking C5 activationattenuates postoperative myocardial infarction, cognitive deficits and bloodloss ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !. There are several causes of complement activation during CPB. In,particular, contact with both the biomaterial and the gas surface is known toactivate complement ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !. Complement is also activated by platelets in clottingblood, and a final activation occurs when soluble heparin is neutralized withprotamine !(155-157). This complement activation affects both leukocytesand platelets. In an in vitro study it has been shown that anti-C5 antibodiesblock both P-selectin expression by platelets and the formation ofplatelet/leukocyte complexes during extracorporeal circulation (158).Complement activation products, in particular C5a, are known to up-regulatereceptor expression on both PMNs and monocytes (159). Other signs of aninflammatory response are the release of cytokines and chemokines (160).

35

Heparin surfaces in vivoThe most widely used surface modification of biomaterial surfaces in vivo isthe heparin surface. In vitro this surface has been shown to be effective inreducing both complement and coagulation activation (16, 17, 161, 162).The most common application of heparin surfaces is extracorporealcirculation and in particular CPB (163). In many cases improvedperformance can be shown, but there are also reports from investigatorsfailing to observe any benefit (164-167). This further underlines the need ofusing a well characterized heparin surface.

There are a number of reports showing reduced complement activation inCPB when using heparinized equipment. This is also accompanied by adecreased activation of granulocytes. When examining the mechanismsbehind down-regulation of complement during CPB using immobilizedheparin an increase in C1rs-C1INH complexes was observed. This increasepresumably reflects the fact that heparin binds C1q. Formation of C3bc andsC5b-9 was, however, significantly lowered when a heparinized system wasused (165).

36

Aims of the Investigation

This study was initiated to investigate the mechanisms of complementactivation triggered by biomaterial surfaces and to develop techniques tospecifically down-regulate complement activation at the surface. The AP isgenerally considered to be responsible for most of the complement activationthat occurs at the biomaterial surface; in this project we wanted toinvestigate if this is indeed the case or whether the AP is a reflection ofcomplement amplification.

The aim of paper I was to investigate whether complement activationoccurs at the biomaterial surface or on top of the adsorbed protein layer. Wealso wanted to identify which mechanism initially triggers complementactivation on a biomaterial surface.

Paper II was initiated to test whether adsorbed C3 was able to participatein an AP C3 convertase. This possibility had been suggested in previousreports, but no conclusive evidence had been presented.

The aim of paper III was to determine whether factor H can beimmobilized to a biomaterial surface with retained activity.

The aim of the last paper was to test the influence of heparin surfaceconcentration on both complement and coagulation parameters. It was alsodesigned to test the influence of blood flow velocity.

37

Materials and Methods

The techniques in this work are mainly based on the use of enzymeimmunoassys (EIAs) and quartz crystal microbalance with dissipationmonitoring (QCM-D). In paper III a method to immobilize factor H to thesurface was developed.

Enzyme Immunoassays

ß-Thromboglobulinß-Thromboglobulin (ß-TG) was measured using the commercialAsserachrome™ ß-TG EIA kit from Diagnostica Stago, Asnieres-sur-Seine,France. ß-TG was captured in wells coated with specific rabbit anti-humanß-TG F(ab’)2 fragments. Peroxidase-coupled rabbit anti-human ß-TGantibody was used for detection. The values are given as IU/ml.

Thrombin-Antithrombin complexesThrombin-Antithrombin (TAT) complexs were measured using acommercial Enzygnost® TAT micro EIA kit from Behringwerke AG,Warburg, Germany. TAT was captured in wells coated with rabbit anti-human thrombin. Peroxidase coupled rabbit anti-human Antithrombinantibody was used for detection. The values are given as µg/ml.

Soluble C5b-9 complexesSoluble C5b-9 (sC5b-9) complexes were measured using a modification ofthe previously described method by Mollnes et al (82). Plasma samplesdiluted 1/5 were added to wells coated with anti-neoC9 mAB McaE11 (81).Polyclonal anti-C5 antibodies (Dako, Glostrup, Denmark) diluted 1/500were used for detection. Zymosan-activated serum, defined as containing40,000 arbitrary units per ml (AU/ml), was used as a standard.

38

C3aC3a was measured as follows: Plasma samples diluted 1/1000 wereincubated in wells coated with mAb 4SD17.3, which served as captureantibody. C3a was detected using a biotinylated anti-C3a followed byhorseradish peroxidase conjugated streptavidin. Zymosan-activated serumcalibrated against a solution of purified C3a served as the standard, and thevalues are given in ng/ml (80).

Quartz Crystal Microbalance with DissipationMonitoringTo measure the complement activation that takes place on a surface in real-time, a QCM-D instrument from Q-sense AB, Gothenburg, was used.Purified components of AP were used to assemble a C3 convertase. Theability of this convertase to perform proteolytic activation of C3 wasmonitored by the binding of C3b to the surface. The QCM-D techniquerelies on the fact that a mass adsorbed onto the sensor surface of a shear-mode oscillating quartz crystal causes a proportional change in its resonancefrequency, f. Changes in f reflect the amount of mass deposited onto to thesurface of the crystal. For thin, evenly distributed and rigid films, anadsorption-induced frequency shift (∆f) is related to mass uptake (∆m) viathe Sauerbrey relation (ref), ∆f=-n∆mC-1, where C (equivalent to 17.7 ng cm-

2 Hz-1) is the mass sensitivity constant and n (=1,3…) is the overtone number(168). However, for proteins adsorbed from the aqueous phase, one mustalso be aware that water hydrodynamically coupled to the adlayer is includedin the measured mass uptake (169). In addition, when the adsorbed materialis non-rigid, additional energy dissipation (viscous loss) is also induced. Thedissipation factor (D) reflects frictional (viscous) losses induced bydeposited materials such as proteins adsorbed on the surface of the crystal.Hence, changes in the viscoelastic properties of adlayers (e.g., those inducedby conformational changes) as well as differences between various protein-surface interactions can be monitored (170-172).

Analysis of adsorption kinetics by simultaneous measurement of both thefrequency, f, and the energy dissipation, D, was performed using a quartzcrystal microbalance with dissipation monitoring (QCM-D) instrument (Q-Sense AB, Gothenburg, Sweden), which is described in detail elsewhere(173). The volume of the chamber is 80 µl, and when the liquid in thechamber is exchanged, 0.5 ml is added from a temperature loop; the excessvolume is allowed to overflow. Sensor crystals (5-MHz), spin-coated withhydrophobic polystyrene or sputtered with steel, were used. Changes in D

39

and f were measured on both the fundamental frequency (n=1, i.e. f≈5MHz)and the third (n=3, i.e. f≈15MHz) and fifth harmonic (n=5, i.e. f≈25MHz).Data from the measurements at the third harmonic are presented. Allmeasurements were carried out at 25°C.

Factor H ImmobilizationTo immobilize factor H on a surface, the surface was first coated with apolymer supplying primary amino groups. A heterobifunctional crosslinker,N-succinimidyl 3-(2 pyridoldithio)propionate (SPDP), able to bind to theamino groups was then added (174). SPDP has a molecular size of 312 Daand is 6.8 Å long. In its unreduced state it can bind to thiol groups; if thebound SPDP is reduced, it provides thiol groups to the surface or the protein.Factor H was then able to bind to thiol groups on the surface supplied bySPDP.

The Corline Heparin SurfaceIn this study the heparin surface from Corline Systems AB was used. The

Corline Heparin Surface was prepared by a conditioning layer of polymericamine onto which a macromolecular conjugate of heaprin was attached. Theconjugate was prepared by covalent binding of approximately 70 molesheparin/mole carrier chain (21). This procedure results in a surface with aheparin surface concentration of 0.5 g/cm2 and an AT binding capacity of 2-4 pmol AT/cm2. Higher levels of heparin surface concentration and ATbinding capacity were achieved by repeating the application of polymericamine and heparin conjugate up to three times.

The Chandler Loop ModelIn the Chandler loop model, extracorporeal circulation is mimicked usingtubing loops. Fresh whole blood was drawn from healthy volunteers whohad received no medication for 10 days. The blood was used with or withoutthe addition of anticoagulant, depending of the material being tested. Piecesof PVC tubing furnished with immobilized heparin were used in the loops.Tubing with a total volume of 6.3 mL (inner diameter= 4 mm, length=500mm) was filled with 4.5 mL of fresh whole blood, leaving a gas volume of1.8 mL. The tubing was made into closed circuits using surface-heparinizedconnectors of thin-walled steel (length=20 mm). The ends of the connector

40

were tightly pushed into the lumen of the tubing and secured using copperseals. The tubing was then rotated vertically at desired speed, typically 33rpm, in a 37°C water bath for the desired time. After incubation, the sampleswere collected in citrate or in EDTA, respectively, and centrifuged at3,300xg for 25 min.

41

Experiments and Results

Four different studies have been carried out. In the first study, ways toidentify the initiating pathway behind complement activation on biomaterialsurfaces was explored. The second study investigated the mechanisms ofalternative pathway activation on biomaterial surfaces. The two followingstudies examined the possibility of specifically down-regulating complementactivation by conjugating two active biomolecules, factor H and heparin, to abiomaterial surface.

Paper IIn the first paper the question to be answered was whether complementactivation takes place on the biomaterial surface or on the adsorbed proteinlayer. Also, the mechanism responsible for initiating complement activationtriggered by the model biomaterial surface, polystyrene, was investigated.The initial protein layer on the polystyrene was estimated by QCM-D tomeasure approximately 8 nm. Complement activation following theadsorption of the protein film added 25% more mass to the surface thanwhen complement activation was blocked.

AP activation was selectively blocked in serum with the complementinhibitor compstatin. In an EIA model it was shown that complementactivation was immediately triggered by a self-limiting CP activation. Thisactivation was immediately following the adsorption of the protein film.

Incubation with EGTA, which blocks the CP, indicated that AP activationbegan after a lag phase of approximately 5-10 min but subsequentlyprovided the bulk of the C3b deposition onto the surface. In order to identifythe types of protein film in whose presence AP amplification could takeplace, we assembled AP convertase complexes using purified C3 and factorsB and D and monitored the assembly by QCM-D analysis. In addition toalbumin, IgG, but not fibrinogen, allowed convertase assembly andamplification.

Western blotting of eluted proteins from the material surfacedemonstrated that the C3 fragments were bound to other proteins and thatthe most abundant fragment was iC3b. These results are consistent with a

42

model in which the main C3b binding is mediated by the AP on top of theinitially adsorbed protein film. The activation is triggered by either CP- orAP-initiating convertases assembled from molecules found among theinitially adsorbed proteins.

Paper IIIn paper II the ability of adsorbed C3 to participate in an AP C3 convertasewas investigated. The AP C3 convertase was assembled on a polystyrenesurface and monitored using QCM-D. Adsorbed C3b was able to bind factorB, which in turn could be activated by factor D. This convertase cleaved C3into C3b and C3a (as assessed by EIA). These results indicated that an activeconventional C3 convertase, C3b,Bb, could be assembled on the surface.

When C3 itself was adsorbed to the polystyrene surface, the boundprotein was recognized by monoclonal antibodies specific for bound C3b.Binding and activation of factor B by factor D also showed that the adsorbedC3 had taken on C3b-like properties, indicating the formation of a C3-containing AP C3 convertase, C3,Bb.

In vivo the AP convertase is stabilized by the presence of properdin.When properdin was added together with factor B a C3,Bb,P convertase wasformed, showing binding of C3b to the surface upon addition of C3.

Interestingly, adsorbed C3 was not down-regulated by factors I and H.Taken together, these results showed that adsorbed C3 can form a weakinitiating AP convertase. The relatively weak proteolytic activity of theconvertase is compensated by the fact that the it is not down-regulated byfactors I and H.

Paper IIIThe third study examined the ability to immobilize factor H with retainedactivity on a model surface. Factor H was bound to the surface with bothreduced and non-reduced SPDP.

By QCM-D it was shown that a higher amount of factor H bound to theunreduced surface, but the flexibility of factor H was higher on the reducedsurface, indicating a more native state.

In both whole blood and plasma the surface with factor H generated lessC3a and TCC in the fluid phase, and the binding of C3/C3 fragments wasabrogated. When serum was incubated on the surface without immobilizedfactor H, C1s-C1INH complexes were generated, but no such generation of

43

C1s-C1INH complexes was observed when factor H was immobilized to thesurface.

Paper IVIn paper IV the effect of the surface concentration of heparin on bothcoagulation and complement was studied. Tubing was coated withincreasing concentrations of heparin, which produced antithrombin bindingcapacities of 6, 12 and 19 pmol/cm2. The tubing was filled with blood, andthe ends were connected to form a loop. The tubing loops were allowed torotate for 1h at three different velocities (13, 28 and 43 cm/s).

When the binding capacity for antithrombin was raised from 6 pmol/cm2

to 12 pmol/cm2 at 28 cm/s, the generation of TAT and the release of ß-TGdecreased, reflecting the fact that both coagulation and platelet activationwere inhibited. The complement parameters C3a and TCC decreased whenthe antithrombin binding capacity was raised from 6 pmol/cm2 to 12pmol/cm2. A further increase to 19 pmol/cm2 did not result in any furtherimprovement regarding either coagulation or complement. The decreasedactivation of coagulation activation could be explained by an increasedantithrombin binding capacity. The factor H binding capacity did notincrease in parallel with decreased complement activation, and the increasedcomplement compatibility remains to be explained.

Interestingly, there was an increased generation of C3a when the flow ratewas raised. TCC showed no significant increase, and TAT was lower at 13cm/s than at the two higher velocities.

In summary, by increasing the surface concentration of heparin, bloodcompatiblity can be improved. An increased concentration might also benecessary when the flow rate is increased, since the complement activationseems to be dependent on the flow rate of the blood.

44

Discussion

Contact between blood and biomaterial is known to trigger activation of thecomplement system (175). There are many possible explanations for thisactivation, but there is at present no extensive theory on the matter. Oneexplanation for this phenomenon is that complement activation triggered bybiomaterials has been considered to be of AP nature only (33, 123). Thiscould lead to the conclusion that the “tick-over” process in the fluid phase isresponsible for the initiation of complement activation and that the degree ofactivation depends only upon the amount of binding sites on the surface.There are also numerous publications suggesting that the complement-activating properties are dependent upon the quantities of hydroxyl andamino groups, which are known to be acceptors for covalent binding of C3b,on the surface (128).

The fact is, however, that upon contact with blood, the biomaterialsurface is rapidly coated with adsorbed plasma proteins. Most of thechemically reactive groups on the surface will be hidden beneath this layerof proteins. The biomaterial surface determines the composition of theprotein layer and also the conformation of the adsorbed proteins. Thissituation indicates a more indirect role of the biomaterial surface incomplement activation.

In paper I, complement activation on the polystyrene surface was shownto be initiated by the CP. This finding was also confirmed in paper III, inwhich a generation of C1s-C1INH complexes was shown in serum incubatedon polystyrene. This complex is formed when there is activation of the CP,suggesting that the polystyrene surface possess CP-activating properties.Further support for an important role for the CP has come from a reportshowing a 15-min delay in the initiation of complement activation whenserum lacking C4 is incubated on cuprophane (4).

If the AP is blocked, there is a rapid activation that is, however, unable tobe amplified. On adsorbed albumin there is no such rapid activation. Insteadthere is a slow complement activation that does not take place until about 10min after serum exposure. Blocking the alternative pathway totally inhibitscomplement activation, indicating that the initiation of complementactivation is triggered by either the CP or the AP, depending on the surfacein question. The following amplification of complement initiation is,

45

however, driven by the AP. This also explains why complement activationtriggered by biomaterials has been considered to be of AP origin. Whencomplement is measured in the fluid phase, AP markers predominate.

The triggering mechanism behind the initiation of complement activationis probably related to conformational changes occurring in adsorbed proteinsthat produce a more activating state. CP activation could be triggered byadsorbed IgG, IgM, C1 or even C4. When considering AP-induced initiationfrom the generally accepted point of view, the most likely explanation wouldbe the tick-over process. This view implies that the initiation takes place inthe fluid phase through a slow conversion of C3 into iC3, a more C3b-likestate that is able to participate in a fluid-phase AP C3 convertase ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! (124-126).The activation would then be amplified on the biomaterial surface. Paper IIpresents evidence that an alternative explanation is possible. Theseexperiments have indicated that adsorbed C3, as previously suggested, isable to initiate AP activation (114, 115).

We have shown that adsorption of C3 also converts it into a C3b-likemolecule that is able to participate in a C3,Bb,P convertase. The possibilitythat adsorption is the actual tick-over mechanism cannot be ruled out, sincethere is always a surface present. The most important feature of thisconvertase is that it does not seem to be regulated by factors I and H. This isa strong argument for the importance of this convertase, even though it has amuch lower activity than does the C3b,Bb,P convertase. If there is a lack ofregulation, even a low activity might be able to initiate complementactivation.

The QCM-D technique has been shown to be suitable for monitoring theassembly of AP convertases of purified complement components in realtime. Although its sensitivity was lower than that of surface plasmonresonance (SPR), it was still high enough to allow us to measure the weakinteractions of the C3,Bb,P complex. In the case of QCM-D, it is also easy tomodify the surfaces to be tested. Several studies of complement activationhave previously been carried out using SPR ! (104, 176). The design of theseexperiments, however, did not allow a direct comparison to make it possibleto determine the best technology to test the in vitro assembly of APcomponents.

Factor H is a soluble inhibitor of the AP ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !. Immobilization of functionalfactor H on a surface was postulated to specifically inhibit AP. The factor Hsurface did indeed inhibit complement activation, as indicated by lowerlevels of binding of C3 fragments to the surface. There was also a reducedgeneration of C3a and sC5b-9, and no C1s-C1INH could be detected in thefluid phase. This latter finding suggests that factor H is able to inhibit CP, afeature that had not previously been assigned to factor H. There are,however, reports that rheumatoid arthritis protein (RHP), later identified as

46

factor H, interacts with the globular head of C1q, a finding which supportsthe possibility that factor H really interferes with CP activation ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !. Sincecomplement activation was shown to be initiated by CP on polystyrene, thisability could be most important. Factor H could then interfere already at thelevel of the initiation of complement activation instead of regulating only theAP-mediated amplification.

An already available technique to inhibit both coagulation andcomplement activation is covalent immobilization of heparin onto a surfaceintended for contact with blood ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !. This type of surface works by binding ATand accelerating its inhibitory effect on coagulation factors. Complement isinhibited by factor H, which has a high affinity for heparin ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! ! !. An increasedsurface concentration of heparin further inhibits coagulation andcomplement activation. The activation of complement is increased withincreased flow rates, and increased heparin surface concentrations might be away to meet these more demanding conditions. The finding that increasedcomplement inhibition is not associated with increased binding of factor Hdoes not mean that factor H is not of importance at heparin surfaces. Instead,only the additional compatibility might be mediated by another factor. TheCorline heparin surface used in this study is known to consist ofunfractionated heparin. This unmodified form still possesses all thephysiological properties of heparin, which are far from completelyunderstood!!!!!!!!!!!!!!!!!!!!.

The results of this study indicate that there are some key events takingplace during complement activation at a biomaterial surface. First, plasmaproteins are adsorbed to the surface, with a composition and conformationthat are dependent upon the biomaterial involved. Some of the proteins areadsorbed in a conformation that triggers initiation of complement activation:IgG, IgM, C1 or C4 could trigger the CP, or on surfaces that do not bind anyof these proteins in a proper conformation, C3 could activate the AP. Afterthe initiation step, complement activation is amplified by an AP feedbackloop that generates the bulk of the C3b on the surface and may also triggerthe terminal pathway of complement. The result is the generation of fluidphase anaphylatoxins C3a and C5a, soluble C5b-9 complexes and surface-bound C3b and iC3b. All these products together are able to attract andactivate cells and also trigger further release of inflammatory mediators fromthem (55, 104). This response is a good thing if it takes place at the propersite, directed against a foreign invader of the body. If instead we are dealingwith a large area of a biomaterial surface, this reaction could lead to asystemic release of inflammatory mediators. If not treated properly, it couldlead to systemic inflammation, indirectly causing hemostatic problems oreven organ failure (177, 178).

47

Figure 4. Summary of the complement activation on a biomaterial surface. Exposureof blood to the surface results in a immediate adsorption of proteins that play animportant role in initiation of complement activation. This results in binding of C3bto the surface and is amplified by the AP amplification loop. Formation ofanaphylatoxins results in activation of leukocytes and C3 fragments on the surfaceacts as ligands for leukocyte binding.

48

Conclusions

Initiation of complement activation on a biomaterial surface can be triggeredby either the CP or the AP.

When complement activation is triggered, amplification is mediated by theAP feedback loop.

Adsorbed non-activated C3 can form an initiating convertase that escapesfactor H and factor I regulation. This is achieved by a conformational changein C3 upon adsorption that generates a C3b-like form.

It is possible to immobilize factor H on a polymer surface with retainedbiological activity.

The intial trigger of complement activation can be the CP, LP or AP, even ifthe major activation of complement on a biomaterial surface is mediated bythe AP.

Increased surface concentrations of heparin further increase the inhibition ofcomplement activity and coagulation. Complement activation is increased byincreasing blood flow rates.

Quartz crystal microbalance with dissipation monitoring (QCM-D) is anexcellent technique for monitoring complement activation on a surface inreal time.

49

Acknowledgements

This work was carried out at the section of Clinical Immunology,Department of Oncology, Radiology and Immunology, Faculty of Medicine,Uppsala University. Financial support was generously provided by grantsfrom the Swedish Foundation for Strategic Research, the Swedish ResearchCouncil, the University of Kalmar, the Göran Gustafsson ResearchFoundation, King Gustav V’s Research Foundation, Prof Nanna Svartz’Research Foundation and the Faculty of Medicine at Uppsala University.

There are many people without whom my work would have been lesspleasant or not possible at all.

First of all I whish to express my gratitude to my supervisors. They havetogether introduced me to the fields of biomaterials and complement. BoNilsson for sharing his knowledge in complement and biomaterial researchand skills in scientific writing. Kristina Nilsson Ekdahl for her expertise inbiochemistry, protein chemistry, complement research, proof-reading andmost important for always being around at the department. Rolf Larsson forintroducing me to biomaterials, heparin, heparin surfaces and the Chandlerloop.

I also whish to thank all other co-authors. Ulf Nilsson for sharing his uniqueknowledge and participation in complement history. Graciela Elgue forteaching me about contact activation, analyzing a lot of ELISA:s and makingstiff shoulders relaxed. Javier Sanchez, the most recent coworker, whom inan enthusiastic way teaches me about contact activation and heparinsurfaces. Ralf Richter for introducing me to the technology of QCM-D.

The wizards at the lab, without whom I would have been helpless, deservethe deepest gratitude. Foozieh Ghazi for excellent technical skills and forteaching me the importance of setting the alarm to be able to arrive in themorning. Lillemor Funke for invaluable efforts and skills in the lab, trying toteach me how to say no and discussions concerning Very Important and LessImportant Things in Life. Margita Nilsson for great company, now sharing

50

her time between the lab and her grandchildren. Moving to Kalmar did nothelp Kerstin Sandholm in escaping my questions and need of help.

Deborah McClellan is acknowledged for excellent proof reading of bothmanuscripts and thesis. Any mistakes you can find has been added by meduring last minute changes after proof reading.

The company of my first colleagues, Jaan Hong and Lisa Moberg, has beengreat fun. Jaan, the stoneface, being calm and cold under pressure and Lisaalways being cold, “is it only me that thinks it’s cold in here?”. Jaan savedme from getting lost in Rhodes Town after dark. Lisa has taught me that youshould never try to get the last word when talking to a woman. There is nopoint! Believe me, I have tried.

I also whish to thank the more recent PhD students who have made myeveryday life at the lab more interesting in many different ways, Anna-KarinLidehäll, Helena Johansson, Jenny Danielsson, Linda Mathsson and PeterSchmidt. Heidi Wähämaa also contributed to the good atmosphere.

“My” exam workers, Clare Lamont, Hanna Winquist, Linda Sallén and SaraOhlsson have contributed a lot to this work. “That don’t impress me much”was Clare’s comment on my singing and I have not gotten any better.

I wish to thank everyone at the Section of Clinical Immunology, especiallyMona Persson, and the staff at Immunological Chemistry.

Most importantly I whish to thank my family. My sister Petra, for alwaystelling me her opinion in a straighforward way. My mother, Wanja, and myfather, Jan, for always being supportive.

51

References

1. Williams, D. F. 1987. Definitions in Biomaterials. Elsevier,Amsterdam.

2. Courtney, J. M., N. M. Lamba, S. Sundaram, and C. D. Forbes.1994. Biomaterials for blood-contacting applications. Biomaterials15:737.

3. Keogh, J. R., M. F. Wolf, M. E. Overend, L. Tang, and J. W. Eaton.1996. Biocompatibility of sulphonated polyurethane surfaces.Biomaterials 17:1987.

4. Lhotta, K., R. Wurzner, F. Kronenberg, M. Oppermann, and P.Konig. 1998. Rapid activation of the complement system bycuprophane depends on complement component C4. Kidney Int53:1044.

5. Hoenich, N. A., C. Woffindin, P. J. Cox, M. Goldfinch, and S. J.Roberts. 1997. Clinical characterization of Dicea a new cellulosemembrane for haemodialysis. Clin Nephrol 48:253.

6. Hussain, M. A., C. V. Murali, P. Willi, C. P. Sharma, G. S.Bhuvneshwar, and R. Sreekumar. 1998. Comparative complementactivation study of polypropylene hollow fibres of two differentmakes in static condition. J Biomater Appl 12:300.

7. Schaadt, J. 1998. Fiber manufacturing, membrane classification, andwinding technologies associated with membrane oxygenators. JExtra Corpor Technol 30:30.

8. Phillips, S. J. 2001. Thrombogenic influence of biomaterials inpatients with the Omni series heart valve: pyrolytic carbon versustitanium. Asaio J 47:429.

9. Gotman, I. 1997. Characteristics of metals used in implants. JEndourol 11:383.

10. Amiji, M., and K. Park. 1992. Prevention of protein adsorption andplatelet adhesion on surfaces by PEO/PPO/PEO triblockcopolymers. Biomaterials 13:682.

11. Mosqueira, V. C., P. Legrand, A. Gulik, O. Bourdon, R. Gref, D.Labarre, and G. Barratt. 2001. Relationship between complementactivation, cellular uptake and surface physicochemical aspects ofnovel PEG-modified nanocapsules. Biomaterials 22:2967.

12. Li, J. T., J. Carlsson, S.-C. Huang, and K. D. Caldwell. 1996.Adsorption of poly(ethylene oxide)-containing block copolymers: a

52

route to protein resistance. In Hydrophilic polymers. Performancewith environmental acceptability. J. E. Glass, ed. AmericanChemical Society, Washington D.C., p. 61.

13. Li, J.-T., and K. D. Caldwell. 1996. Plasma protein interactions withPluronic™-treated colloids. Colloids and Surfaces B: Biointerfaces7:9.

14. Lindahl, U., K. Lidholt, D. Spillmann, and L. Kjellen. 1994. More to"heparin" than anticoagulation. Thromb Res 75:1.

15. Björk, I., S. T. Olson, and J. D. Shore. 1989. Molecular mechanismsof the accelerated effect of heparin on the reactions betweenantithrombin and cloting factors. In Heparin, Chemical andbiological properties. Clinical applications. D. A. Lane, and U.Lindahl, eds. Edward Arnold, London, p. 229.

16. Pekna, M., R. Larsson, B. Formgren, U. R. Nilsson, and B. Nilsson.1993. Complement activation by polymethyl methacrylateminimized by end-point heparin attachment. Biomaterials 14:189.

17. Sanchez, J., G. Elgue, J. Riesenfeld, and P. Olsson. 1997. Inhibitionof the plasma contact activation system of immobilized heparin:relation to surface density of functional antithrombin binding sites. JBiomed Mater Res 37:37.

18. Larsson, R., P. Olsson, and U. Lindahl. 1980. Inhibition of thrombinon surfaces coated with immobilized heparin and heparin-likepolysaccharides: a crucial non-thrombogenic principle. Thromb Res19:43.

19. Pangburn, M. K., M. A. Atkinson, and S. Meri. 1991. Localizationof the heparin-binding site on complement factor H. J Biol Chem266:16847.

20. Blackmore, T. K., T. A. Sadlon, H. M. Ward, D. M. Lublin, and D.L. Gordon. 1996. Identification of a heparin binding domain in theseventh short consensus repeat of complement factor H. J Immunol157:5422.

21. van der Giessen, W. J., H. M. van Beusekom, R. Larsson, and P. W.Serruys. 1999. Heparin-coated coronary stents. Curr Interv CardiolRep 1:234.

22. Li, J. T., J. Carlsson, J. N. Lin, and K. D. Caldwell. 1996. Chemicalmodification of surface active poly(ethylene oxide)-poly (propyleneoxide) triblock copolymers. Bioconjug Chem 7:592.

23. Fox, S. I. 1996. Human Physiology. Brown, Dubuque.24. Hoffbrand, A. D., and J. E. Pettit. 1994. Essential Haematology.

Blackwell, Oxford.25. Hong, J., K. Nilsson Ekdahl, H. Reynolds, R. Larsson, and B.

Nilsson. 1999. A new in vitro model to study interaction betweenwhole blood and biomaterials. Studies of platelet and coagulationactivation and the effect of aspirin. Biomaterials 20:603.

26. Schwippert-Houtermans, B., S. Strapatsakis, P. Roesen, and D.Tschoepe. 2001. Evaluation of an antibody-based genotype

53

classification of the platelet fibrinogen receptor (GPIIb/IIIa).Cytometry 46:238.

27. Massberg, S., M. Gawaz, S. Gruner, V. Schulte, I. Konrad, D.Zohlnhofer, U. Heinzmann, and B. Nieswandt. 2003. A crucial roleof glycoprotein VI for platelet recruitment to the injured arterial wallin vivo. J Exp Med 197:41.

28. Harrison, P., and E. M. Cramer. 1993. Platelet alpha-granules. BloodRev 7:52.

29. Kaplan, K. L. 1978. Proteins secreted by platelets: significance indetecting thrombosis. Adv Exp Med Biol 102:105.

30. Ross, G. D., and J. D. Lambris. 1982. Identification of a C3bi-specific membrane complement receptor that is expressed onlymphocytes, monocytes, neutrophils, and erythrocytes. J Exp Med155:96.

31. Weerwind, P. W., N. E. Caberg, C. P. Reutelingsperger, T.Lindhout, and D. S. de Jong. 2002. Exposure of procoagulantphospholipids on the surface of platelets in patients undergoingcardiopulmonary bypass using non-coated and heparin-coatedextracorporeal circuits. Int J Artif Organs 25:770.

32. Abbas, A. K., A. H. Lichtman, and P. S. Jordan. 1994. TheComplement System. In Cellular and molecular immunology. W.B.Saunders, Philadelphia, p. 293.

33. Hed, J., M. Johansson, and M. Lindroth. 1984. Complementactivation according to the alternate pathway by glass and plasticsurfaces and its role in neutrophil adhesion. Immunol Lett 8:295.

34. Wetterö, J., A. Askendal, T. Bengtsson, and P. Tengvall. 2002. Onthe binding of complement to solid artificial surfaces in vitro.Biomaterials 23:981.

35. Meri, S., and H. Jarva. 1998. Complement regulation. Vox Sang74:291.

36. Xu, Y., S. V. Narayana, and J. E. Volanakis. 2001. Structuralbiology of the alternative pathway convertase. Immunol Rev180:123.

37. Kinoshita, T. 1991. Biology of complement:the overture.Immunology Today 12:291.

38. Petersen, S. V., S. Thiel, and J. C. Jensenius. 2001. The mannan-binding lectin pathway of complement activation: biology anddisease association. Mol Immunol 38:133.

39. Kasahara, M., J. Nakaya, Y. Satta, and N. Takahata. 1997.Chromosomal duplication and the emergence of the adaptiveimmune system. Trends Genet 13:90.

40. Smith, L. C., K. Azumi, and M. Nonaka. 1999. Complement systemsin invertebrates. The ancient alternative and lectin pathways.Immunopharmacology 42:107.

41. Nonaka, M., and M. Takahashi. 1992. Complete complementaryDNA sequence of the third component of complement of lamprey.

54

Implication for the evolution of thioester containing proteins. JImmunol 148:3290.

42. Adams, M. D., S. E. Celniker, R. A. Holt, C. A. Evans, J. D.Gocayne, P. G. Amanatides, S. E. Scherer, P. W. Li, R. A. Hoskins,R. F. Galle, R. A. George, S. E. Lewis, S. Richards, M. Ashburner,S. N. Henderson, G. G. Sutton, J. R. Wortman, M. D. Yandell, Q.Zhang, L. X. Chen, R. C. Brandon, Y. H. Rogers, R. G. Blazej, M.Champe, B. D. Pfeiffer, K. H. Wan, C. Doyle, E. G. Baxter, G. Helt,C. R. Nelson, G. L. Gabor, J. F. Abril, A. Agbayani, H. J. An, C.Andrews-Pfannkoch, D. Baldwin, R. M. Ballew, A. Basu, J.Baxendale, L. Bayraktaroglu, E. M. Beasley, K. Y. Beeson, P. V.Benos, B. P. Berman, D. Bhandari, S. Bolshakov, D. Borkova, M. R.Botchan, J. Bouck, P. Brokstein, P. Brottier, K. C. Burtis, D. A.Busam, H. Butler, E. Cadieu, A. Center, I. Chandra, J. M. Cherry, S.Cawley, C. Dahlke, L. B. Davenport, P. Davies, B. de Pablos, A.Delcher, Z. Deng, A. D. Mays, I. Dew, S. M. Dietz, K. Dodson, L.E. Doup, M. Downes, S. Dugan-Rocha, B. C. Dunkov, P. Dunn, K.J. Durbin, C. C. Evangelista, C. Ferraz, S. Ferriera, W. Fleischmann,C. Fosler, A. E. Gabrielian, N. S. Garg, W. M. Gelbart, K. Glasser,A. Glodek, F. Gong, J. H. Gorrell, Z. Gu, P. Guan, M. Harris, N. L.Harris, D. Harvey, T. J. Heiman, J. R. Hernandez, J. Houck, D.Hostin, K. A. Houston, T. J. Howland, M. H. Wei, C. Ibegwam, etal. 2000. The genome sequence of Drosophila melanogaster. Science287:2185.

43. Consortium, T. C. e. S. 1998. Genome sequence of the nematode C.elegans: a platform for investigating biology. The C. elegansSequencing Consortium. Science 282:2012.

44. Kushner, I., T. S. Edgington, C. Trimble, H. H. Liem, and U.Muller-Eberhard. 1972. Plasma hemopexin homeostasis during theacute phase response. J Lab Clin Med 80:18.

45. Fong, K. Y., M. Botto, M. J. Walport, and A. K. So. 1990. Genomicorganization of human complement component C3. Genomics7:579.

46. Einstein, L. P., P. J. Hansen, M. Ballow, A. E. Davis, 3rd, J. S. t.Davis, C. A. Alper, F. S. Rosen, and H. R. Colten. 1977.Biosynthesis of the third component of complement (C3) in vitro bymonocytes from both normal and homozygous C3-deficient humans.J Clin Invest 60:963.

47. Cole, F. S., W. J. Matthews, Jr., T. H. Rossing, D. J. Gash, N. A.Lichtenberg, and J. E. Pennington. 1983. Complement biosynthesisby human bronchoalveolar macrophages. Clin ImmunolImmunopathol 27:153.

48. Alper, C. A., A. M. Johnson, A. G. Birtch, and F. D. Moore. 1969.Human C'3: evidence for the liver as the primary site of synthesis.Science 163:286.

55

49. Lambris, J. D., Z. Lao, T. J. Oglesby, J. P. Atkinson, C. E. Hack, andJ. D. Becherer. 1996. Dissection of CR1, factor H, membranecofactor protein, and factor B binding and functional sites in thethird complement component. Journal of Immunology156(12):4821.

50. Tack, B. F. 1983. The beta-Cys-gamma-Glu thiolester bond inhuman C3, C4, and alpha 2-macroglobulin. Springer SeminImmunopathol 6:259.

51. Sim, R. B., T. M. Twose, D. S. Paterson, and E. Sim. 1981. Thecovalent-binding reaction of complement component C3. Biochem J193:115.

52. Hugli, T. E., and H. J. Muller-Eberhard. 1978. Anaphylatoxins: C3aand C5a. Adv Immunol 26:1.

53. Nilsson, B., K. Nilsson Ekdahl, K. E. Svensson, A. Bjelle, and U. R.Nilsson. 1989. Distinctive expression of neoantigenic C3(D)epitopes on bound C3 following activation and binding to differenttarget surfaces in normal and pathological sera. Molec Immunol26:383.

54. McNally, A. K., and J. M. Anderson. 1994. Complement C3participation in monocyte adhesion to different surfaces. Proc NatlAcad Sci U S A 91:10119.

55. Larsson, R., G. Elgue, A. Larsson, K. N. Ekdahl, U. R. Nilsson, andB. Nilsson. 1997. Inhibition of complement activation by solublerecombinant CR1 under conditions resembling those in acardiopulmonary circuit: reduced up-regulation of CD11b andcomplete abrogation of binding of PMNs to the biomaterial surface.Immunopharmacology 38:119.

56. Ueda, T., P. Rieu, J. Brayer, and M. A. Arnaout. 1994. Identificationof the complement iC3b binding site in the beta 2 integrin CR3(CD11b/CD18). Proc Natl Acad Sci U S A 91:10680.

57. Zhang, W., and P. J. Lachmann. 1996. Neutrophil lactoferrin releaseinduced by IgA immune complexes can be mediated either by Fcalpha receptors or by complement receptors through differentpathways. J Immunol 156:2599.

58. Davis, A. E., 3rd, R. A. Harrison, and P. J. Lachmann. 1984.Physiologic inactivation of fluid phase C3b: isolation and structuralanalysis of C3c, C3d,g (alpha 2D), and C3g. J Immunol 132:1960.

59. Bokisch, V. A., H. J. Muller-Eberhard, and C. G. Cochrane. 1969.Isolation of a fragment (C3a) of the third component of humancomplement containing anaphylatoxin and chemotactic activity anddescription of an anaphylatoxin inactivator of human serum. J ExpMed 129:1109.

60. Takata, Y., T. Kinoshita, H. Kozono, J. Takeda, E. Tanaka, K.Hong, and K. Inoue. 1987. Covalent association of C3b with C4bwithin C5 convertase of the classical complement pathway. J ExpMed 165:1494.

56

61. Medicus, R. G., O. Gotze, and H. J. Muller-Eberhard. 1976.Alternative pathway of complement: recruitment of precursorproperdin by the labile C3/C5 convertase and the potentiation of thepathway. J Exp Med 144:1076.

62. Nilsson, B., and K. Nilsson Ekdahl. 1998. Components of thealternative pathway. In The complement system. K. Rother, G. O.Till, and G. M. Hänsch, eds. Springer-Verlag, Berlin Heidelberg.

63. Kinoshita, T., Y. Takata, H. Kozono, J. Takeda, K. S. Hong, and K.Inoue. 1988. C5 convertase of the alternative complement pathway:covalent linkage between two C3b molecules within the trimolecularcomplex enzyme. J Immunol 141:3895.

64. Hetland, G., E. Johnson, and T. Eskeland. 1987. Formation of themembrane attack complex of complement (MAC) on erythrocytesfrom monocyte-produced terminal complement components. ScandJ Immunol 25:571.

65. Spycher, M. O., and U. E. Nydegger. 1995. Participation of theblood platelet in immune reactions due to platelet-complementinteraction. Infusionsther Transfusionsmed 22:36.

66. Barlow, P. N., D. G. Norman, A. Steinkasserer, T. J. Horne, J.Pearce, P. C. Driscoll, R. B. Sim, and I. D. Campbell. 1992. Solutionstructure of the fifth repeat of factor H: a second example of thecomplement control protein module. Biochemistry 31:3626.

67. Barlow, P. N., A. Steinkasserer, D. G. Norman, B. Kieffer, A. P.Wiles, R. B. Sim, and I. D. Campbell. 1993. Solution structure of apair of Complement modules by nuclear magnetic resonance. J MolBiol 232:268.

68. Caliezi, C., W. A. Wuillemin, S. Zeerleder, M. Redondo, B. Eisele,and C. E. Hack. 2000. C1-Esterase inhibitor: an anti-inflammatoryagent and its potential use in the treatment of diseases other thanhereditary angioedema. Pharmacol Rev 52:91.

69. Fearon, D. T. 1977. Purification of C3b inactivator anddemonstration of its two polypeptide chain structure. J Immunol119:1248.

70. Telen, M. J. 1995. Glycosyl phosphatidylinositol-linked blood groupantigens and paroxysmal nocturnal hemoglobinuria. Transfus ClinBiol 2:277.

71. Pangburn, M. K. 2000. Host recognition and target differentiation byfactor H, a regulator of the alternative pathway of complement.Immunopharmacology 49:149.

72. DiScipio, R. G. 1992. Ultrastructures and interactions ofcomplement factors H and I. J Immunol 149:2592.

73. Gordon, D. L., R. M. Kaufman, T. K. Blackmore, J. Kwong, and D.M. Lublin. 1995. Identification of complement regulatory domainsin human factor H. J Immunol 155:348.

74. Blackmore, T. K., J. Hellwage, T. S. Sadlon, N. Higgs, P. F. Zipfel,H. M. Ward, and D. L. Gordon. 1998. Identification of the second

57

heparin-binding domain in human complement factor H. Journal ofImmunology 160:3342.

75. Meri, T., A. Hartmann, D. Lenk, R. Eck, R. Wurzner, J. Hellwage,S. Meri, and P. F. Zipfel. 2002. The yeast Candida albicans bindscomplement regulators factor H and FHL-1. Infect Immun 70:5185.

76. Zipfel, P. F., T. S. Jokiranta, J. Hellwage, V. Koistinen, and S. Meri.1999. The factor H protein family. Immunopharmacology 42:53.

77. Mollnes, T. E., O. L. Brekke, M. Fung, H. Fure, D. Christiansen, G.Bergseth, V. Videm, K. T. Lappegard, J. Kohl, and J. D. Lambris.2002. Essential role of the C5a receptor in E coli-induced oxidativeburst and phagocytosis revealed by a novel lepirudin-based humanwhole blood model of inflammation. Blood 100:1869.

78. Peterson, K. L., W. Zhang, P. D. Lu, S. A. Keilbaugh, E. I.Peerschke, and B. Ghebrehiwet. 1997. The C1q-binding cellmembrane proteins cC1q-R and gC1q-R are released from activatedcells: subcellular distribution and immunochemical characterization.Clin Immunol Immunopathol 84:17.

79. Mollnes, T. E. 1997. Analysis of in vivi complement activation. InWeir's Handbook of Experimental Immunology. L. A. Herzenberg,D. M. Weir, and C. Blackwell, eds. Blackwell Science, Boston, p.781.

80. Nilsson, B., K. E. Svensson, M. Inganas, and U. R. Nilsson. 1988. Asimplified assay for the detection of C3a in human plasmaemploying a monoclonal antibody raised against denatured C3. JImmunol Methods 107:281.

81. Mollnes, T. E., T. Lea, M. Harboe, and J. Tschopp. 1985.Monoclonal antibodies recognizing a neoantigen of poly(C9) detectthe human terminal complement complex in tissue and plasma.Scand J Immunol 22:183.

82. Mollnes, T. E., T. Lea, S. S. Froland, and M. Harboe. 1985.Quantification of the terminal complement complex in humanplasma by an enzyme-linked immunosorbent assay based onmonoclonal antibodies against a neoantigen of the complex. Scand JImmunol 22:197.

83. Mollnes, T. E., H. Redl, K. Hogasen, A. Bengtsson, P. Garred, L.Speilberg, T. Lea, M. Oppermann, O. Gotze, and G. Schlag. 1993.Complement activation in septic baboons detected by neoepitope-specific assays for C3b/iC3b/C3c, C5a and the terminal C5b-9complement complex (TCC). Clin Exp Immunol 91:295.

84. Horstick, G., A. Heimann, O. Gotze, G. Hafner, O. Berg, P.Boehmer, P. Becker, H. Darius, H. J. Rupprecht, M. Loos, S.Bhakdi, J. Meyer, and O. Kempski. 1997. Intracoronary applicationof C1 esterase inhibitor improves cardiac function and reducesmyocardial necrosis in an experimental model of ischemia andreperfusion. Circulation 95:701.

58

85. Bos, I. G., C. E. Hack, and J. P. Abrahams. 2002. Structural andfunctional aspects of C1-inhibitor. Immunobiology 205:518.

86. Krych-Goldberg, M., R. E. Hauhart, V. B. Subramanian, B. M.Yurcisin, 2nd, D. L. Crimmins, D. E. Hourcade, and J. P. Atkinson.1999. Decay accelerating activity of complement receptor type 1(CD35). Two active sites are required for dissociating C5convertases. J Biol Chem 274:31160.

87. Hill, J., T. F. Lindsay, F. Ortiz, C. G. Yeh, H. B. Hechtman, and F.D. Moore, Jr. 1992. Soluble complement receptor type 1 amelioratesthe local and remote organ injury after intestinal ischemia-reperfusion in the rat. J Immunol 149:1723.

88. Regal, J. F., D. G. Fraser, and C. A. Toth. 1993. Role of thecomplement system in antigen-induced bronchoconstriction andchanges in blood pressure in the guinea pig. J Pharmacol Exp Ther267:979.

89. Moran, P., H. Beasley, A. Gorrell, E. Martin, P. Gribling, H. Fuchs,N. Gillett, L. E. Burton, and I. W. Caras. 1992. Human recombinantsoluble decay accelerating factor inhibits complement activation invitro and in vivo. J Immunol 149:1736.

90. Christiansen, D., J. Milland, B. R. Thorley, I. F. McKenzie, and B.E. Loveland. 1996. A functional analysis of recombinant solubleCD46 in vivo and a comparison with recombinant soluble forms ofCD55 and CD35 in vitro. Eur J Immunol 26:578.

91. Higgins, P. J., J. L. Ko, R. Lobell, C. Sardonini, M. K. Alessi, and C.G. Yeh. 1997. A soluble chimeric complement inhibitory proteinthat possesses both decay-accelerating and factor I cofactoractivities. J Immunol 158:2872.

92. Sugita, Y., K. Ito, K. Shiozuka, H. Suzuki, H. Gushima, M. Tomita,and Y. Masuho. 1994. Recombinant soluble CD59 inhibits reactivehaemolysis with complement. Immunology 82:34.

93. Matis, L. A., and S. A. Rollins. 1995. Complement-specificantibodies: designing novel anti-inflammatories. Nat Med 1:839.

94. Thomas, T. C., S. A. Rollins, R. P. Rother, M. A. Giannoni, S. L.Hartman, E. A. Elliott, S. H. Nye, L. A. Matis, S. P. Squinto, and M.J. Evans. 1996. Inhibition of complement activity by humanizedanti-C5 antibody and single-chain Fv. Mol Immunol 33:1389.

95. Kaplan, M. 2002. Eculizumab (Alexion). Curr Opin Investig Drugs3:1017.

96. Gupta-Bansal, R., J. B. Parent, and K. R. Brunden. 2000. Inhibitionof complement alternative pathway function with anti-properdinmonoclonal antibodies. Mol Immunol 37:191.

97. Sahu, A., and J. D. Lambris. 2000. Complement inhibitors: aresurgent concept in anti-inflammatory therapeutics.Immunopharmacology 49:133.

98. Pellas, T. C., W. Boyar, J. van Oostrum, J. Wasvary, L. R. Fryer, G.Pastor, M. Sills, A. Braunwalder, D. R. Yarwood, R. Kramer, E.

59

Kimble, J. Hadala, W. Haston, R. Moreira-Ludewig, S. Uziel-Fusi,P. Peters, K. Bill, and L. P. Wennogle. 1998. Novel C5a receptorantagonists regulate neutrophil functions in vitro and in vivo. JImmunol 160:5616.

99. Heller, T., M. Hennecke, U. Baumann, J. E. Gessner, A. M. zuVilsendorf, M. Baensch, F. Boulay, A. Kola, A. Klos, W. Bautsch,and J. Kohl. 1999. Selection of a C5a receptor antagonist fromphage libraries attenuating the inflammatory response in immunecomplex disease and ischemia/reperfusion injury. J Immunol163:985.

100. Sahu, A., B. K. Kay, and J. D. Lambris. 1996. Inhibition of humancomplement by a C3-binding peptide isolated from a phage-displayed random peptide library. J Immunol 157:884.

101. Sahu, A., A. M. Soulika, D. Morikis, L. Spruce, W. T. Moore, and J.D. Lambris. 2000. Binding kinetics, structure-activity relationship,and biotransformation of the complement inhibitor compstatin. JImmunol 165:2491.

102. Fiane, A. E., T. E. Mollnes, V. Videm, T. Hovig, K. Hogasen, O. J.Mellbye, L. Spruce, W. T. Moore, A. Sahu, and J. D. Lambris. 1999.Compstatin, a peptide inhibitor of C3, prolongs survival of ex vivoperfused pig xenografts. Xenotransplantation 6:52.

103. Fiane, A. E., T. E. Mollnes, V. Videm, T. Hovig, K. Hogasen, O. J.Mellbye, L. Spruce, W. T. Moore, A. Sahu, and J. D. Lambris. 1999.Prolongation of ex vivo-perfused pig xenograft survival by thecomplement inhibitor Compstatin. Transplant Proc 31:934.

104. Nilsson, B., R. Larsson, J. Hong, G. Elgue, K. N. Ekdahl, A. Sahu,and J. D. Lambris. 1998. Compstatin inhibits complement andcellular activation in whole blood in two models of extracorporealcirculation. Blood 92:1661.

105. Soulika, A. M., M. M. Khan, T. Hattori, F. W. Bowen, B. A.Richardson, C. E. Hack, A. Sahu, L. H. Edmunds, Jr., and J. D.Lambris. 2000. Inhibition of heparin/protamine complex-inducedcomplement activation by Compstatin in baboons. Clin Immunol96:212.

106. Wahlgren, M., and T. Arnebrant. 1991. Protein adsorption to solidsurfaces. TIBTECHTrends Biotechnol 9(6):201.

107. Vroman, L., and A. L. Adams. 1969. Identification of rapid changesat plasma-solid interfaces. J Biomed Mater Res 3:43.

108. Vroman, L., and A. L. Adams. 1969. Identification of absorbedprotein films by exposure to antisera and water vapor. J BiomedMater Res 3:669.

109. Nilsson, B., and U. R. Nilsson. 1986. SDS denaturation ofcomplement factor C3 as a model for the allosteric modificationsassociated with C3b-binding: demonstration of a profound

60

conformational change by means of cicular dichroism andquantitative immunoprecipitation. Immunol Lett 13:11.

110. Nilsson, B., K. Nilsson Ekdahl, D. Avila, U. R. Nilsson, and J. D.Lambris. 1990. Neoantigens in C3 as detected by monoclonalantibodies: mapping of the recognized epitopes by syntheticpeptides. Biochem J 268:55.

111. Alsenz, J., J. D. Becherer, B. Nilsson, and J. D. Lambris. 1990.Structural and functional analysis of C3 using monoclonalantibodies. In The Third Component of Complement, Vol. 153. J. D.Lambris, ed. Springer-Verlag, Berlin, p. 235.

112. Nilsson, U. R., and B. Nilsson. 1982. Analogous antigenicalterations elicited in C3 by physiological binding and bydenaturation in the presence of sodium dodecylsulfate. J Immunol129:2594.

113. Nilsson, B., and U. R. Nilsson. 1985. An assessment of the extent ofantigenic analogy between physiologically bound C3 and C3denatured by sodium dodecyl sulphate. Scand J Immunol 22:703.

114. Elwing, H., B. Nilsson, K.-E. Svensson, A. Askendahl, U. Nilsson,and I. Lundström. 1988. Conformational changes of a model protein(complement factor 3) adsorbed on hydrophilic and hydrophobicsolid surfaces. J Colloid Interface Sci 125(1):139.

115. Nilsson, U. R., K. E. Storm, H. Elwing, and B. Nilsson. 1993.Conformational epitopes of C3 reflecting its mode of binding to anartificial polymer surface. Mol Immunol 30:211.

116. Hong, J., A. Larsson, K. N. Ekdahl, G. Elgue, R. Larsson, and B.Nilsson. 2001. Contact between a polymer and whole blood:sequence of events leading to thrombin generation. J Lab Clin Med138:139.

117. Hunt, B. J., R. Parratt, M. Cable, D. Finch, and M. Yacoub. 1997.Activation of coagulation and platelets is affected by thehydrophobicity of artificial surfaces. Blood Coagul Fibrinolysis8:223.

118. Courtney, J. M., and C. D. Forbes. 1994. Thrombosis on foreignsurfaces. Br Med Bull 50:966.

119. Lu, P. D., D. K. Galanakis, B. Ghebrehiwet, and E. I. Peerschke.1999. The receptor for the globular "heads" of C1q, gC1q-R, bindsto fibrinogen/fibrin and impairs its polymerization. Clin Immunol90:360.

120. Altieri, D. C., F. R. Agbanyo, J. Plescia, M. H. Ginsberg, T. S.Edgington, and E. F. Plow. 1990. A unique recognition site mediatesthe interaction of fibrinogen with the leukocyte integrin Mac-1(CD11b/CD18). J Biol Chem 265:12119.

121. Wälivaara, B., A. Askendal, I. Lundström, and P. Tengvall. 1997.Imaging of the Early Events of Classical Complement ActivationUsing Antibodies and Atomic Force Microscopy. J Colloid InterfaceSci 187:121.

61

122. Nilsson, U. R. 2001. Deposition of C3b/iC3b leads to theconcealment of antigens, immunoglobulins and bound C1q incomplement-activating immune complexes. Mol Immunol 38:151.

123. Kazatchkine, M. D., and M. P. Carreno. 1988. Activation of thecomplement system at the interface between blood and artificialsurfaces. Biomaterials 9:30.

124. Lachmann, P. J., and L. Halbwachs. 1975. The influence of C3binactivator (KAF) concentration on the ability of serum to supportcomplement activation. Clin Exp Immunol 21:109.

125. Pangburn, M. K., and H. J. Muller-Eberhard. 1980. Relation ofputative thioester bond in C3 to activation of the alternative pathwayand the binding of C3b to biological targets of complement. J ExpMed 152:1102.

126. Pangburn, M. K., R. D. Schreiber, and H. J. Muller-Eberhard. 1981.Formation of the initial C3 convertase of the alternative complementpathway. Acquisition of C3b-like activities by spontaneoushydrolysis of the putative thioester in native C3. J Exp Med 154:856.

127. Nilsson Ekdahl, K., B. Nilsson, C. H. Gölander, H. Elwing, B.Lassen, and U. R. Nilsson. 1993. Complement Activation on RadioFrequency Plasma Modified Polystyrene Surfaces. J ColloidInterface Sci 158:121.

128. Chenoweth, D. E. 1987. Complement activation in extracorporealcircuits. Ann N Y Acad Sci 516:306.

129. Wheatley, D. J., G. M. Bernacca, M. M. Tolland, B. O'Connor, J.Fisher, and D. F. Williams. 2001. Hydrodynamic function of abiostable polyurethane flexible heart valve after six months in sheep.Int J Artif Organs 24:95.

130. Lapeyre, D. M., O. H. Frazier, J. L. Conger, M. P. Macris, P. Perrier,H. Reul, B. Rolland, F. J. Clubb, Jr., S. M. Parnis, J. M. Fuqua, andet al. 1994. In vivo evaluation of a trileaflet mechanical heart valve.Asaio J 40:M707.

131. Okubo, M., and L. N. Benson. 2001. Intravascular and intracardiacstents used in congenital heart disease. Curr Opin Cardiol 16:84.

132. Cwikiel, W., and J. Harnek. 2000. A new stent designed for carotidartery application. An experimental study. Acta Radiol 41:601.

133. Dake, M. D. 2001. Endovascular stent-graft management of thoracicaortic diseases. Eur J Radiol 39:42.

134. Stammers, A. H. 1997. Historical aspects of cardiopulmonarybypass: from antiquity to acceptance. Journal of Cardiothoracic andVascular Anesthesia 11:266.

135. Schmidt, B. 1996. Membranes in artificial organs. Artif Organs20:375.

136. Hinrichs, W. L., H. W. ten Hoopen, G. H. Engbers, and J. Feijen.1997. In vitro evaluation of heparinized Cuprophan hemodialysismembranes. J Biomed Mater Res 35:443.

62

137. Hoenich, N. A., C. Woffindin, J. N. Mathews, and J. Vienken. 1995.Biocompatibility of membranes used in the treatment of renalfailure. Biomaterials 16:587.

138. Schulman, G., R. Hakim, R. Arias, M. Silverberg, A. P. Kaplan, andL. Arbeit. 1993. Bradykinin generation by dialysis membranes:possible role in anaphylactic reaction. J Am Soc Nephrol 3:1563.

139. Burdick, L., M. Periti, A. Salvaggio, S. Bertoli, R. Mangiarotti, D.Castagnone, and G. Anguissola. 1994. Relation between carotidartery atherosclerosis and time on dialysis. A non-invasive study invivo. Clin Nephrol 42:121.

140. Amore, A., and R. Coppo. 2002. Immunological basis ofinflammation in dialysis. Nephrol Dial Transplant 17 Suppl 8:16.

141. Santoro, A., and E. Mancini. 2002. Cardiac effects of chronicinflammation in dialysis patients. Nephrol Dial Transplant 17 Suppl8:10.

142. Morena, M., J. P. Cristol, and B. Canaud. 2000. Why hemodialysispatients are in a prooxidant state? What could be done to correct thepro/antioxidant imbalance. Blood Purif 18:191.

143. Wanner, C., J. Zimmermann, T. Quaschning, and J. Galle. 1997.Inflammation, dyslipidemia and vascular risk factors inhemodialysis patients. Kidney Int Suppl 62:S53.

144. Salmon, J. E., and M. J. Roman. 2001. Accelerated atherosclerosisin systemic lupus erythematosus: implications for patientmanagement. Curr Opin Rheumatol 13:341.

145. Himmelfarb, J., E. McMonagle, D. Holbrook, and C. Toth. 1995.Soluble complement receptor 1 inhibits both complement andgranulocyte activation during ex vivo hemodialysis. J Lab Clin Med126:392.

146. Cheung, A. K., C. J. Parker, and M. Hohnholt. 1994. Solublecomplement receptor type 1 inhibits complement activation inducedby hemodialysis membranes in vitro. Kidney Int 46:1680.

147. Moen, O., K. Hogasen, E. Fosse, E. Dregelid, V. Brockmeier, P.Venge, M. Harboe, and T. E. Mollnes. 1997. Attenuation of changesin leukocyte surface markers and complement activation withheparin-coated cardiopulmonary bypass. Ann Thorac Surg 63:105.

148. Vlaicu, R., F. Niculescu, H. G. Rus, and A. Cristea. 1985.Immunohistochemical localization of the terminal C5b-9complement complex in human aortic fibrous plaque.Atherosclerosis 57:163.

149. Torzewski, M., C. Rist, R. F. Mortensen, T. P. Zwaka, M. Bienek, J.Waltenberger, W. Koenig, G. Schmitz, V. Hombach, and J.Torzewski. 2000. C-reactive protein in the arterial intima: role of C-reactive protein receptor-dependent monocyte recruitment inatherogenesis. Arterioscler Thromb Vasc Biol 20:2094.

150. Bruins, P., H. te Velthuis, A. J. Eerenberg-Belmer, A. P.Yazdanbakhsh, E. M. de Beaumont, L. Eijsman, A. Trouwborst, and

63

C. E. Hack. 2000. Heparin-protamine complexes and C-reactiveprotein induce activation of the classical complement pathway:studies in patients undergoing cardiac surgery and in vitro. ThrombHaemost 84:237.

151. Carlucci, F., A. Tabucchi, B. Biagioli, F. Simeone, S. Scolletta, F.Rosi, and E. Marinello. 2002. Cardiac surgery: myocardial energybalance, antioxidant status and endothelial function after ischemia-reperfusion. Biomed Pharmacother 56:483.

152. Pekna, M., L. Nilsson, K. Nilsson-Ekdahl, U. R. Nilsson, and B.Nilsson. 1993. Evidence for iC3 generation during cardiopulmonarybypass as the result of blood-gas interaction. Clin Exp Immunol91:404.

153. Nussmeier, N. A. 2002. A review of risk factors for adverseneurologic outcome after cardiac surgery. J Extra Corpor Technol34:4.

154. Mojcik, C. F., and J. H. Levy. 2001. Aprotinin and the systemicinflammatory response after cardiopulmonary bypass. Ann ThoracSurg 71:745.

155. Bruins, P., H. te Velthuis, A. J. Eerenberg-Belmer, A. P.Yazdanbakhsh, E. M. de Beaumont, L. Eijsman, A. Trouwborst, andC. E. Hack. 2000. Heparin-protamine complexes and C-reactiveprotein induce activation of the classical complement pathway:studies in patients undergoing cardiac surgery and in vitro. ThrombHaemost 84:237.

156. Cavarocchi, N. C., H. V. Schaff, T. A. Orszulak, H. A. Homburger,W. A. Schnell, Jr., and J. R. Pluth. 1985. Evidence for complementactivation by protamine-heparin interaction after cardiopulmonarybypass. Surgery 98:525.

157. Özmen, L., K. N. Ekdahl, G. Elgue, R. Larsson, O. Korsgren, and B.Nilsson. 2002. Inhibition of thrombin abrogates the instant blood-mediated inflammatory reaction triggered by isolated human islets:possible application of the thrombin inhibitor melagatran in clinicalislet transplantation. Diabetes 51:1779.

158. Rinder, C. S., H. M. Rinder, B. R. Smith, J. C. Fitch, M. J. Smith, J.B. Tracey, L. A. Matis, S. P. Squinto, and S. A. Rollins. 1995.Blockade of C5a and C5b-9 generation inhibits leukocyte andplatelet activation during extracorporeal circulation. J Clin Invest96:1564.

159. Rinder, C. S., H. M. Rinder, K. Johnson, M. Smith, D. L. Lee, J.Tracey, G. Polack, P. Higgins, C. G. Yeh, and B. R. Smith. 1999.Role of C3 cleavage in monocyte activation during extracorporealcirculation. Circulation 100:553.

160. Hill, G. E., C. W. Whitten, and D. F. Landers. 1997. The influenceof cardiopulmonary bypass on cytokines and cell-cellcommunication. J Cardiothorac Vasc Anesth 11:367.

64

161. Gong, J., R. Larsson, K. N. Ekdahl, T. E. Mollnes, U. Nilsson, andB. Nilsson. 1996. Tubing loops as a model for cardiopulmonarybypass circuits: both the biomaterial and the blood-gas phaseinterfaces induce complement activation in an in vitro model. J ClinImmunol 16:222.

162. Christensen, K., R. Larsson, H. Emanuelsson, G. Elgue, and A.Larsson. 2001. Heparin coating of the stent graft — effects onplatelets, coagulation and complement activation. Biomaterials22:349.

163. Videm, V., T. E. Mollnes, P. Garred, and J. L. Svennevig. 1991.Biocompatibility of extracorporeal circulation. In vitro comparisonof heparin-coated and uncoated oxygenator circuits. J ThoracCardiovasc Surg 101:654.

164. Videm, V., J. L. Svennevig, E. Fosse, G. Semb, A. Osterud, and T.E. Mollnes. 1992. Reduced complement activation with heparin-coated oxygenator and tubings in coronary bypass operations. JThorac Cardiovasc Surg 103:806.

165. Videm, V., T. E. Mollnes, K. Bergh, E. Fosse, B. Mohr, T. A.Hagve, A. O. Aasen, and J. L. Svennevig. 1999. Heparin-coatedcardiopulmonary bypass equipment. II. Mechanisms for reducedcomplement activation in vivo. J Thorac Cardiovasc Surg 117:803.

166. Kopp, R., K. Mottaghy, and M. Kirschfink. 2002. Mechanism ofcomplement activation during extracorporeal blood-biomaterialinteraction: effects of heparin coated and uncoated surfaces. Asaio J48:598.

167. Belboul, A., and N. al-Khaja. 1997. Does heparin coating improvebiocompatibility? A study on complement, blood cells andpostoperative morbidity during cardiac surgery. Perfusion 12:385.

168. Sauerbrey, G. 1959. Verwendung von Schwingquarzen zur Wägungdünner Schichten und zur Mikrowägung. Zeitschrift für Physik155:206.

169. Caruso, F., D. N. Furlong, and P. Kingshott. 1997. Characterizationof ferritin adsorption onto gold. J Coll Interf Sci 186:129.

170. Höök, F., B. Kasemo, T. Nylander, C. Fant, K. Sott, and H. Elwing.2001. Variations in coupled water, viscoelastic properties, and filmthickness of a Mefp-1 protein film during adsorption and cross-linking: a quartz crystal microbalance with dissipation monitoring,ellipsometry, and surface plasmon resonance study. Anal Chem73:5796.

171. Höök, F., M. Rodahl, P. Brzezinski, and B. Kasemo. 1998.Measurements using the quartz crystal microbalance technique offerritin monolayers on methyl-thiolated gold: Dependence of energydissipation and saturation coverage on salt concentration. J CollInterf Sci 208:63.

172. Höök, F., M. Rodahl, B. Kasemo, and P. Brzezinski. 1998.Structural changes in hemoglobin during adsorption to solid

65

surfaces: effects of pH, ionic strength, and ligand binding. Proc NatlAcad Sci U S A 95:12271.

173. Rodahl, M., F. Hook, A. Krozer, P. Brzezinski, and B. Kasemo.1995. Quartz-crystal microbalance setup for frequency and Q-factormeasurments in gaseous and liquid. Review of scientific instruments66(7):3924.

174. Carlsson, J., H. Drevin, and R. Axen. 1978. Protein thiolation andreversible protein-protein conjugation. N-succinimidyl-3-(2-pyridyldithio)propionate, a new heterobifunctional reagent.Biochemical Journal 173:723.

175. Mollnes, T. E. 1998. Complement and biocompatibility. Vox Sang74:303.

176. Jokiranta, T. S., J. Westin, U. R. Nilsson, B. Nilsson, J. Hellwage, S.Lofas, D. L. Gordon, K. N. Ekdahl, and S. Meri. 2001. ComplementC3b interactions studied with surface plasmon resonance technique.Int Immunopharmacol 1:495.

177. Kirklin, J. K. 1989. The postperfusion syndrome: Inflammation andthe damaging effects of cardiopulmonary bypass. InCardiopulmonary bypass: Current concepts and controversies. J. H.Tinker, ed. W. B. Saunders, Philadelphia, p. 3026.

178. Greeley, W. J., and R. M. Ungerleider. 1991. Assessing the effect ofcardiopulmonary bypass on the brain [editorial; comment]. AnnThorac Surg 52:417.