assessment of complement activation in human

ASSESSMENT OF COMPLEMENT ACTIVATION IN HUMAN DISEASE

ASSESSMENT OF COMPLEMENT ACTIVATION IN HUMAN DISEASE

ACADEMISCH PROEFSCHRIFT

ter verkrijging van de graad van doctor aan de Universiteit van Amsterdam op gezag van de Rector Magnificus prof. dr. D.C. van den Boom ten overstaan van een door het college voor promoties ingestelde commissie, in het openbaar te verdedigen in de Agnietenkapel op donderdag 19 juni 2008, te 14:00 uur door

Diana Wouters geboren te Amsterdam

Promotiecommissie: Promotores:

Prof. dr. C.E. Hack Prof. dr. L.A. Aarden

Overige leden:

Prof. dr. A. Sturk Prof. dr. R.J.M. ten Berge Prof. dr. T.W. Kuijpers Prof. dr. M.R. Daha Prof. dr. D. Roos Dr. A.E. Voskuyl

Faculteit der Geneeskunde

Financial support for the printing of this thesis of the following institutions is gratefully acknowledged: Sanquin Research, Dutch Arthritis Association, Jurriaanse Stichting and University of Amsterdam.

The work described in this thesis was performed at the department of Immunopathology, Sanquin Research, Amsterdam, The Netherlands

ISBN 978-90-9023037-5 ©2008 Diana Wouters Printed by PrintPartners Ipskamp, Enschede, The Netherlands

The most exciting phrase to hear in science, the one that heralds new discoveries, is not 'Eureka!' (I found it!) but 'That's funny ...' Isaac Asimov US science fiction novelist & scholar (1920 - 1992)

Contents Page Chapter 1.

General Introduction

Chapter 2.

Complexes between C1q and C3 or C4: novel and specific markers for

9 29

classical complement pathway activation Journal of Immunological Methods (2005) Chapter 3.

Assessment of complement activation in vivo: which assay when?(review)

47

Manuscript in preparation Chapter 4.

Evaluation of classical complement pathway activation in Rheumatoid

67

Arthritis. Measurement of C1q-C4 complexes as novel activation products Arthritis and Rheumatism (2006) Chapter 5.

Studies on the haemolytic activity of circulating C1q-C3/C4 complexes

83

Molecular Immunology (2008) Chapter 6.

High throughput analysis of the C4 polymorphism by a combination of

99

MLPA and isotype-specific ELISA’s Submitted for publication Chapter 7.

Summarizing Discussion

121

Samenvatting en Discussie

129

Korte samenvatting voor niet-ingewijden

137

Curriculum Vitae

139

Dankwoord

141

General Introduction

Chapter 1 General Introduction

9

Chapter 1 The complement system As a major effector mechanism, the complement system plays a central role in innate immunity. It consists of more than 30 plasma- and membrane bound proteins constituting a first line of defence against many pathogens (1, 2). The importance of complement is illustrated by the fact that patients with complement deficiencies show an increased predisposition for infections and auto-immune disease. Activation of the complement system can occur via three pathways, which are initiated via separate mechanisms and eventually converge in a common terminal pathway (Fig. 1). The classical pathway of complement is activated by binding of the recognition molecule C1q to IgG- or IgM-containing immune complexes or to other structures, such as dsDNA (3, 4), amyloid beta (5) or pentraxins like SAP and CRP (8). Furthermore, it has been shown that C1q can directly bind to apoptotic cells and thereby mediate complement activation (9-11). This implicates that C1q may be involved in the clearance of apoptotic cells. The lectin pathway is initiated by binding of mannan binding lectin (MBL) or ficolins to a wide array of carbohydrate structures on bacterial or viral surfaces (12). Finally, activating structures of the alternative pathway are present on the cell walls of bacteria, viruses and yeasts, whereas IgA-containing immune complexes can also activate the alternative pathway (13, 14). Recently, in mouse models for rheumatoid arthritis (collagen induced arthritis (CIA) model and anti-GPI serum transfer model), evidence was found for IgG mediated complement activation, that was completely dependent on the alternative pathway while bypassing the requirements for the classical pathway (15, 16).

The classical pathway The first component of the classical pathway of complement is C1, which is a complex of three proteins, C1q, C1r and C1s. In vivo, two molecules of each of the pro-enzymes C1r and C1s, forming a C1r2C1s2 tetramer, are associated with one C1q molecule, forming the macromolecular C1 complex (17, 18). C1q is a cationic plasma glycoprotein with a molecular weight of 460 kDa. The molecule consists of six subunits, resembling a bunch of tulips (Fig. 2). The subunits are composed of an N-terminal globular head region and a C-terminal collagenous tail. Each subunit consists of 3 polypeptide chains, designated A, B and C. Thus, the complete molecule comprises 6A, 6B and 6C chains. A and B chains of each subunit are connected to each other by disulfide bonds, whereas C chains are non-covalently linked to an A-B heterodimer and disulfide bonded to a C chain of an adjacent subunit (19-21).

10

General Introduction Alternative Microbial surfaces

Classical

Lectin

Immune complexes (IgG, IgM) CRP SAP DNA

Microbial polysaccharides

C1q C1r C1s

MBL/Ficolins MASP1 MASP2

C3(H2O) fB fD Ba

C1-inh

C4

C3(H2O)Bb

C4 C3a

C4a

C3

Properdin (P)

C3b

C4b C4bp

C2

fH

CR1

fB fI

fI

fD

C2b

C4b2a

CR1

DAF

C3a

fH

MCP

C3bBb(P)

Carboxypeptidase N

C3a

C3

C4b2a3b

C3bBb3b(P)

C5

Terminal pathway C5a Carboxypeptidase N CD59

S-Protein

Clusterin

C5b, C6, C7, C8, C9 (C5b-9) Membrane Attack Complex (MAC)

Figure 1. The complement system can be activated via three different pathways; the classical, lectin and alternative pathway that eventually converge into the terminal pathway. Activation of the terminal pathway results in formation of the lytic membrane attack complex (MAC). The complement system is tightly regulated by fluid phase and membrane bound complement regulators which are indicated in grey boxes.

11

Chapter 1

A A C C A

B

C B

C

A B

A B

C

B

C B

A

Figure 2. Schematic representation of the C1q molecule.

C1q links the humoral immune response to the complement system by binding to the Fc-regions of IgG- or IgM-antibodies within immune complexes, thereby activating the classical pathway of complement. Multivalent binding of the C1q globular heads to the target surface fixes the “arms” of C1q in such a way that intramolecular changes in the pro-enzyme C1r are induced, resulting in functional protease activity. Next, C1r cleaves C1s, thereby inducing its proteolytic activity (22). The natural substrates for the C1s protease are C4 and C2. Upon binding of C1s, C4 is cleaved leading to the release of the C4a anaphylatoxin and deposition of the larger C4b fragment near C1q on the pathogenic surface. Subsequently, C2 binds to C4b on the surface. When C2, attached to C4b, is cleaved by C1s, the smaller fragment C2b will be released while C2a forms a complex with C4b (C4b2a) attached to the pathogenic surface. The C4b2a complex is the classic C3 convertase on the surface of the pathogen. This C3 convertase cleaves C3, which results in release of the anaphylatoxin C3a and binding of C3b, either directly to the pathogenic surface or to the C4bC2a complex (24). A key step in the propagation of complement activation and elimination of pathogens from the body is the covalent binding of C3 and C4 to target surfaces. C4 and C3 both contain a highly reactive thio-ester which is easily hydrolyzed by water molecules and is protected within the native molecule. Therefore these proteins circulate as inert proteins until they are proteolytically activated. Activation of C3 and C4 results in a conformational change and exposure of the internal thio-ester which may react with amino- or hydroxyl-groups to form amide or ester bonds. Only a small proportion of activated C3b and C4b will bind to the target surface that initiated complement activation. The majority of the molecules will be hydrolyzed by surrounding water molecules and remain in the fluid phase. This rapid inactivation limits the efficiency of the reactive proteins to the site of activation (23-26). 12

General Introduction The lectin pathway The lectin pathway is an antibody-independent route of complement activation on bacteria and other micro-organisms. The lectin pathway essentially uses the same molecules as the classical pathway; e.g. C4, C2 and C3, except that its recognition molecules are MBL or ficolins rather than C1q. MBL is structurally related to C1q, being a high molecular weight multimeric molecule with globular binding regions and a collagenous stalk (27-29). Both MBL and ficolins are members of the collectin family of proteins, which are characterized by the presence of both a collagen-like region and a sugar-binding C-type lectin domain. MBL can be associated with the MBL-associated serine proteases (MASPs), MASP-1 (30), MASP-2 (31), MASP-3 (32) and a truncated form of MASP-2, called Map-19 (33). MBL binds to a wide variety of carbohydrate structures such as mannose and fucose on bacterial and viral surfaces, thereby activating the complement system (12). MASP-2 appears to be the major complement-activating enzyme of this pathway. It resembles the classical pathway protease C1s in cleaving both C4 and C2, eventually leading to the formation of the C3 convertase C4b2a (34, 35). Next to MBL, two types of human serum ficolins, L-ficolin and H-ficolin, are associated with MASPs and activate the lectin pathway in a similar manner to MBL (36, 37). A third type of human ficolin, M-ficolin, is a non-serum ficolin which is present in lungs and leukocytes (38). Ficolins may bind to patterns of acetyl-groups, either presented by carbohydrates such as GlcNAc or GalNAc, or non-carbohydrate acetylated compounds. Binding has also been observed towards other bacterial components such as lipoteichoic acid, which is a cell wall component of Gram-positive bacteria (39). Ficolins are, just as MBL and C1q, capable of binding to apoptotic cells and thereby activating complement. This has led to the idea that these recognition molecules are directly involved in the disposal of dying cells (40, 41). The alternative pathway The alternative pathway does not require the action of antibodies to initiate the activation cascade; instead, it can be initiated by spontaneous cleavage of complement component C3. Key to the activation of the alternative pathway is that the internal thio-ester in C3 not only serves to link activated C3b to an activator, but also to stabilize the conformation of native C3. Upon disruption of the thio-ester the conformation of C3b changes to become able to interact with factor B (fB). As indicated in a previous section, the internal thio-ester is sensitive to water molecules and is therefore “hidden” in a hydrophobic region of C3 (42-44). Yet, in time water molecules are able to penetrate some C3 molecules and to hydrolyze the thio-ester. The conformation of C3(H2O) changes to become similar to that of C3b. Hence, intact C3 with a 13

Chapter 1 disrupted thio-ester has also been called “C3b-like C3” (45). Via this mechanism of spontaneous hydrolysis of native C3 small amounts of C3b-like C3 are continuously generated in the fluid phase (“tick over” of C3). This C3(H20) binds to fB, which subsequently can be cleaved by factor D (fD). Cleavage of fB results in formation of the subunits Bb and Ba. The Ba subunit diffuses away and this results in an active alternative pathway C3 convertase (C3(H2O)Bb). C3(H2O)Bb is stabilized by binding to properdin (P), and in its turn can activate native C3 into C3b (Fig. 1). In a similar way as described above for C3(H20), the C3bBb(P) complex is formed, which also is a C3 convertase that can cleave C3, thus forming an amplification loop resulting in more C3b deposition. Amplification of the alternative pathway is thought to take place exclusively on activating surfaces, on which bound C3b is not inactivated by complement inhibitors. In case of fixation of C3b at sites where the breakdown of C3b into C3c and C3d,g is prevented (for example on microbial surfaces), this amplification may result in extensive activation of C3. Target specificity of the alternative pathway is largely determined by the carbohydrate environment of bound C3b, which influences the outcome of the competition between fB and the major alternative pathway inhibitor factor H (fH) for binding to C3b on membranes. On activating surfaces, fB will bind to C3b, but on non-activating surfaces fH will bind which abrogates further activation (46). C3b has affinity for IgG molecules. Two C3b molecules can be sequentially bonded to each other and to one IgG heavy chain, forming C3b2-IgG complexes. In covalent complex with IgG, C3b is partially protected from enzymatic inactivation by the fluid phase inhibitor factor I because of impaired factor H binding. Apparently, IgG provides a protected surface, which enables alternative pathway activation (47). Therefore, C3b2-IgG complexes are very efficient activators of the alternative complement pathway (48, 49). Recently, properdin has been reported to act as recognition molecule within the alternative pathway of complement. By direct binding to microbial targets, properdin provides a platform for the formation of C3 convertases (50). Properdin is a multivalent protein composed of identical subunits (51) which may enable binding to polyvalent ligands clustered on microbial surfaces. This is similar to C1q and MBL, the recognition molecules of the classical and lectin pathway, which are also multivalent proteins that bind polyvalent ligands. The alternative pathway is not only an activation pathway by itself, but also provides an important amplification loop for the classical and lectin pathway of complement. C3 which is activated via these pathways may initiate the alternative pathway. Amplification via this route accounts for more than 80% of initial classical or lectin pathway induced terminal complement activation (52). 14

General Introduction The terminal pathway The three complement activation pathways merge at the level of C3, finally resulting in the formation of the membrane attack complex (MAC) through a common terminal pathway. The terminal pathway is entered when C5 binds non-covalently to a site on C3b in the C4b2a3b complex (C5 convertase). Cleavage of C5 occurs by the C2a fragment and results in C5a and C5b. C5a is the most powerful anaphylatoxin of the complement system, attracting leukocytes expressing a C5a-receptor (e.g. neutrophils and mast cells). C5b contains a labile hydrophobic surface binding site and a binding site for C6. In contrast to the initiation pathways, in the terminal pathway none of the components is enzymatically cleaved. Binding of each following component in this route is achieved by conformational changes in the acceptor molecule. A complex of components C5b, C6, C7, and C8 mediates the polymerization of up to eighteen C9 molecules into a tube-like pore that is inserted into the plasma membrane of unwanted organisms, such as gram-negative bacteria and virally infected cells. This channel through the lipid bilayer is called membrane attack complex (MAC) and finally results in osmotic lysis of the target cell. Effector mechanisms of complement Biological activity of the complement system is not restricted to MAC formation and subsequent cell-lysis. An important action of complement is to facilitate phagocytosis and destruction of pathogens by phagocytes. Pathogens may be opsonized with complement components (C4b, C3b) that are recognized by so called complement receptors on phagocytic cells (CR1, CR2, CR3). Complement receptors are also important for the clearance of immune complexes from the circulation. For example, CR1 present on erythrocytes binds to immune complexes opsonized with the activated complement components C3b and C4b and transports them to the liver and spleen where they are removed by macrophages (53). Activation of complement also contributes to inflammation by the release of anaphylatoxins. C5a and, to a lesser extent, C3a trigger release of histamine and other mediators upon binding to specific receptors on mast cells or basophils, resulting in vasodilatation. C5a, a potent chemoattractant, recruits neutrophils, monocytes, eosinophils and mast cells to the site of inflammation (54). More recently, it has been established that complement collaborates with the adaptive immune system by augmenting the humoral response to antigens and thereby lowering the threshold for B cell activation. In mice, CR2 forms a co-receptor complex with CD19 and CD81. Co-ligation of this complex with the B-cell receptor (BCR) occurs when a C3d-opsonized antigen binds via the 15

Chapter 1 antigenic part to the specific BCR and via C3d to CR2 in the co-receptor complex. This results in enhanced B cell responses where lower amounts of antigen are required for proper B cell activation (55, 56). Complement regulation To control the process of complement activation, complement regulatory proteins are present in plasma and on host cell membranes. Membrane-bound complement regulators are expressed on most host cells to prevent innocent bystander killing by complement activation in the neighbourhood of these cells. The complement system is tightly regulated to prevent excessive activation on a single target, fluid phase activation and activation on self-molecules and cells. Foreign surfaces lacking control proteins are attacked by complement, while host cells are protected. Deficiencies of control proteins may lead to excessive complement activation and disease. The initiation step of the classical pathway is inhibited by the soluble regulator C1-inhibitor (C1inh). C1-inh blocks the active site of C1r and C1s and dissociates them from C1q. Hereby, fluid phase C1 activation and excessive activation on a target are prevented (57). C1-inh is also able to inhibit the action of MASP-2, the complement activating protease of the lectin pathway (58). At the level of C3 convertases, several inhibitors (both fluid phase and membrane bound) are known. The fluid phase serine protease factor I (fI) requires a cofactor for proteolytic degradation of C3b and C4b (59). Soluble cofactors for fI are the regulators factor H (fH) and C4 binding protein (C4bp). Membrane bound fI cofactors are the regulators CR1 (CD35) (60) and membrane cofactor protein (MCP, CD46) (61). Proteolytic cleavage of C3b into C3c and C3d,g by factor I prevents C3 convertase formation and decreases its affinity for CR1. C4 is degraded by fI in a similar manner as C3. Factor H is the most important inhibitor of the alternative pathway. Next to cofactor activity for fI in the fluid phase, it also acts on C3b on membranes. It inhibits the formation and accelerates decay of C3 convertases by competing with fB for binding to C3b. A key to the regulation of C3 activation is whether fH binds to C3b deposits on membranes. This is influenced by the carbohydrate environment of deposited C3b; fH has affinity for negatively charged molecules on host cells, such as sialic acid (62). The membrane bound regulator decay accelerating factor (DAF (CD55)) accelerates the decay of C3 and C5 convertases (63, 64). MCP (CD46), which is another membrane bound complement regulator, acts as cofactor for fI in the cleavage of C3b and C4b. The activity of the anaphylatoxins C3a, C4a and C5a is regulated by plasma carboxypeptidase N, which degrades these molecules into less potent variants. 16

General Introduction The terminal pathway of complement is regulated by the fluid phase regulators S-protein (vitronectin) and clusterin. Both S-protein and clusterin bind to the C5b-9 complex, thereby preventing insertion of this complex into the cell membrane (65). CD59 is a membrane bound inhibitor of MAC formation. It inhibits the formation of the MAC by preventing the binding of C9 to the C5b-C8 complex. (66). C4 polymorphism Complement component C4 is the most polymorphic protein of the complement system. It is a large plasma protein (200 kDa) consisting of three chains (α, β and γ) that are connected via disulphide bonds (Fig. 3). C4 is encoded by two closely linked genes that are located within the MHC class III region on chromosome 6 and which give rise to two isotypic variants, C4A and C4B. Although these isotypes only differ in four amino acids, C4A and C4B are functionally different. After proteolytic cleavage by C1s or MASP-2, C4A preferentially binds via its exposed thio-ester to amino groups, whereas C4B reacts preferentially with hydroxyl groups (23, 67).

A

C γ-chain (30 kD)

β-chain (75 kD)

C4b C O

HS O S

α-chain (95 kD)

C

OH

C4b C O

HS

B

NH

NH2

β-chain (75 kD)

γ-chain (30 kD)

α-chain (86 kD)

C4a S

C1s

NH2

C4b

C

HS

O

OH

C O O

OH

C4b

Figure 3. (A) Intact C4 is a 200 kDa protein that consists of three disulphide-bridged polypeptide chains. (B) Proteolytic cleavage results in release of C4a and exposure of the reactive thio-ester in C4b. (C) The thio-ester of C4b may react with surrounding water molecules or with free amino- or hydroxyl groups.

17

Chapter 1 This implicates that C4A may be functionally advantageous to ensure the solubilization of immune complexes, and clearance of immune complexes through binding to CR1 on erythrocytes. C4B however, would be mainly important for the removal of bacteria by propagating the classical and lectin complement activation pathways that eventually lead to the formation of MAC. Most individuals have both C4 isotypes, but partial C4 deficiency is quite common in humans. Heterozygous or homozygous C4A or C4B deficiencies are reported to be associated with a great variety of autoimmune or infectious diseases. An increased prevalence of C4A deficiencies has been found in several populations of patients with systemic lupus erythematosus (SLE), which is a typical immune complex disease (68-70). It has been hypothesized that because C4A is more relevant in clearance of immune complexes, deficiency of C4A results in impaired processing of immune complexes which, as discussed below, may be important in the pathogenesis of SLE. Furthermore, it has been suggested that C4B deficiency predisposes for bacterial infections (71, 72). Complement and disease The complement system plays a pivotal role in human disease. On the one hand, complement activation has many protective functions in immunity and deficiencies within the complement system may lead to increased susceptibility to invasive bacterial infections or development of autoimmune diseases. On the other hand, undesired or excessive complement activation is a major cause of tissue injury in many pathological conditions. Deficiencies in the early components of the classical pathway (C1q, C4, and C2) are strongly associated with development of SLE (73, 74). MBL deficiency is associated with increased susceptibility to infections, particularly when immunity is already compromised: for example, in infants and young children, patients with cystic fibrosis and after chemotherapy and transplantation (75). C3 deficiency increases the risk for recurrent pyogenic infections, because of lack of opsonization and inability to use the membrane attack pathway (76). Moreover, C3 deficiency predisposes for membranoproliferative glomerulonephritis (77). Deficiencies in constituents of the alternative pathway (factor B, factor D and properdin) as well as deficiencies in terminal pathway components (C5, C6, C7, C8 and C9) lead to increased susceptibility to Gram-negative bacteria such as Neisseria, as a result of the inability to attack the outer membrane of these organisms (78-81). Deficiency in C1-inh leads to recurrent angioedema attacks. C1-inh deficiency may result from a genetic defect (hereditary angioedema, HAE) or may be caused by an acquired condition 18

General Introduction (acquired angioedema, AAE), such as the formation of auto-antibodies towards the reactive site of C1-inh. In type I HAE C1-inh protein synthesis is defective, which leads to low serum levels. In type II HAE normal C1-inh quantities are produced, but with functional impairment of the protein. Both HAE and AAE are associated with decreased C4 levels, due to uncontrolled activity of C1s (82). Deficiencies in the fluid phase complement regulators factors H and I may result in a state of acquired, severe C3 deficiency (76). Absence of either of these control proteins leads to uncontrolled cleavage of C3. Therefore, deficiency of either of these regulatory proteins gives rise to similar problems as inherited C3 deficiency, such as increased susceptibility to bacterial infections (83). Moreover, mutations in these complement regulators are associated with atypical haemolytic uremic syndrome (aHUS) (84, 85). More recently, variants of fH have been identified as major risk factor for age-related macular degeneration (AMD) (86-88). These findings have been supported by observations that mutations in fB also predispose to AMD (89). The complement regulators DAF and CD59 are linked to the membrane via GPI-anchors. A mutation in the PIG-A gene leads to incomplete synthesis of GPI-anchors and absent or reduced surface expression of GPI-linked proteins, such as DAF and CD59 (90). This genetic disorder leads to paroxysmal nocturnal hemaglobulinuria (PNH), a disease in which affected erythrocytes and platelets are more vulnerable to complement mediated lysis (91, 92). The critical role of C5 in this disease is illustrated by observations that upon administration of C5 blocking antibody, disease symptoms of PNH can be largely reduced (93). Erythrocytes of individuals who are genetically deficient in DAF (Inab phenotype) but express CD59, are not more susceptible to complement mediated lysis as erythrocytes from healthy controls (94). This demonstrates that susceptibility to lysis is controlled primarily by CD59 on the erythrocytes. Inappropriate complement activation may cause tissue injury. Excessive activation of complement has been implicated in the pathogenesis of a large number of diseases, including cardiovascular, neurological and autoimmune disease. Most often complement activation is not the cause of disease, but it can exacerbate clinical symptoms and sustain the inflammation. The pro-inflammatory effects of complement activation products contribute to host tissue injury. Complement activation leads to the release of anaphylatoxins (C3a and C5a), which have many pro-inflammatory effects such as attraction, activation and degranulation of neutrophils. Furthermore, the production of sublytic amounts of MAC may lead to upregulation of adhesion molecules on endothelial cells and synthesis of pro-inflammatory cytokines (95). Activation of complement plays an important role in development of ischaemia/reperfusion injury, such as in acute myocardial infarction (96, 97). Moreover, it has been shown that complement activation is 19

Chapter 1 associated with severity of disease in autoimmune diseases such as rheumatoid arthritis, systemic lupus erythematosus and multiple sclerosis (98-101) and it plays a role in the onset of Alzheimer’s disease (102). Complement and autoimmunity Complement deficiency is associated with autoimmune disease. Mainly the lack of early classical pathway components (C1q, C4, and C2) strongly predisposes for the development of autoimmune disease, such as SLE. Healthy individuals have natural IgM antibodies directed against neo-epitopes on late apoptotic and necrotic cells (103). Furthermore, the acute phase proteins CRP and SAP are also capable of binding to apoptotic cells (104, 105). Upon binding of these adaptor molecules to damaged cells, the classical pathway of complement is activated which contributes to removal of immune complexes and cellular debris derived from apoptotic cells (9, 106-108). In addition, the recognition molecule of the classical pathway, C1q, has been demonstrated to bind directly to apoptotic cells and activate complement as well (9, 11). Deficiencies in the classical pathway may therefore lead to impaired clearance of apoptotic material and release of intracellular auto-antigens such as nuclear components. Prolonged exposure of these auto-antigens to the immune system increases the risk for auto-antibody formation. This is called the “waste disposal hypothesis” (2). The presence of auto-antibodies against nuclear constituents (anti-nuclear antibodies, ANAs) is a common feature in SLE. Immune complexes consisting of auto-antibodies and auto-antigens may accumulate in highly vascularized tissues such as skin, kidneys and lungs, where they become pathogenic. Deposited immune complexes induce inflammation through local activation of the complement system, which leads to infiltration and activation of neutrophils (56, 74). Next to disposal of apoptotic debris, complement plays a role in the regulation of self-reactive B cells. “The tolerance hypothesis” is an alternative hypothesis that might explain how defects in complement result in development of SLE (109). This hypothesis states that complement-coated self-antigens are delivered to developing self-reactive B cells to enhance their negative selection. Defects in complement might therefore result in failure of negative B cell selection, allowing auto-reactive B cells to survive and propagate when they would normally undergo apoptosis or anergy (109, 110). In conclusion, the combination of increased presence of auto-antigens because of impaired clearance of apoptotic cells and escape of B cell tolerance might explain the relationship between classical pathway deficiencies and development of SLE.

20

General Introduction Scope of this thesis The classical pathway of complement activation plays an important role in autoimmune disease. To evaluate the involvement of complement in disease it is important to be able to measure complement activation in patients. Chapter 2 describes a novel assay to measure classical pathway mediated complement activation in plasma (C1q-C4 ELISA). During classical pathway activation, covalent complexes are formed between activated C4 and the recognition molecule C1q. These complexes appear to be highly specific for the classical pathway and are very stable. Chapter 3 reviews the assays that have been described to measure complement activation in bodily fluids. The most important characteristics of the various available assays are discussed here. The newly developed C1q-C4 ELISA is used in Chapter 4, in which classical complement pathway activation is evaluated in patients with rheumatoid arthritis (RA). C1q-C4 complexes are analyzed in patients with either active or inactive disease. C1q-C4 levels are significantly higher in patients with active RA compared to patients with inactive disease which indicates that C1q-C4 complexes may be useful as novel diagnostic marker for RA disease activity. Chapter 5 demonstrates the characteristics of circulating C1q-complement complexes in more detail. It appears that in healthy individuals C1q-complement complexes circulate as part of the intact C1 complex. Moreover, deposition of complement activation fragments on C1q seems to have a regulatory effect, since it lowers haemolytic activity of the C1q molecule. In chapter 6 the C4 polymorphism is studied on both genetic and protein level in a healthy study population using different methods to discriminate between C4A and C4B, the two major C4 isotypes. Since C4 protein levels correlate very well with the C4 genetic profile, we conclude that the combination of MLPA and isotype-specific ELISAs is a good approach to study the C4 polymorphism. Finally, Chapter 7 contains a general discussion and a summary of the results.

21

Chapter 1 References 1. Walport, M. J. 2001. Complement. Second of two parts. N. Engl. J. Med. 344: 1140-1144. 2. Walport, M. J. 2001. Complement. First of two parts. N. Engl. J. Med. 344: 1058-1066. 3. Jiang, H., B. Cooper, F. A. Robey, and H. Gewurz. 1992. DNA binds and activates complement via residues 14-26 of the human C1q A chain. J. Biol. Chem. 267: 25597-25601. 4. Rosenberg, A. M., P. A. Prokopchuk, and J. S. Lee. 1988. The binding of native DNA to the collagen-like segment of Clq. J. Rheumatol. 15: 1091-1096. 5. Tacnet-Delorme, P., S. Chevallier, and G. J. Arlaud. 2001. Beta-amyloid fibrils activate the C1 complex of complement under physiological conditions: evidence for a binding site for A beta on the C1q globular regions. J. Immunol. 167: 6374-6381. 6. Bristow, C. L. and R. J. Boackle. 1986. Evidence for the binding of human serum amyloid P component to Clq and Fab gamma. Mol. Immunol. 23: 1045-1052. 7. Ying, S. C., A. T. Gewurz, H. Jiang, and H. Gewurz. 1993. Human serum amyloid P component oligomers bind and activate the classical complement pathway via residues 14-26 and 76-92 of the A chain collagenlike region of C1q. J. Immunol. 150: 169-176. 8. Gewurz, H., S. C. Ying, H. Jiang, and T. F. Lint. 1993. Nonimmune activation of the classical complement pathway. Behring. Inst. Mitt. 138-147. 9. Korb, L. C. and J. M. Ahearn. 1997. C1q binds directly and specifically to surface blebs of apoptotic human keratinocytes: complement deficiency and systemic lupus erythematosus revisited. J Immunol 158: 4525-4528. 10. Navratil, J. S., L. C. Korb, and J. M. Ahearn. 1999. Systemic lupus erythematosus and complement deficiency: clues to a novel role for the classical complement pathway in the maintenance of immune tolerance. Immunopharmacology 42: 47-52. 11. Nauta, A. J., L. A. Trouw, M. R. Daha, O. Tijsma, R. Nieuwland, W. J. Schwaeble, A. R. Gingras, A. Mantovani, C.E. Hack, and A. Roos. 2002. Direct binding of C1q to apoptotic cells and cell blebs induces complement activation. Eur. J. Immunol. 32: 1726-1736. 12. Petersen, S. V., S. Thiel, and J. C. Jensenius. 2001. The mannan-binding lectin pathway of complement activation: biology and disease association. Mol. Immunol. 38: 133-149. 13. Hiemstra, P. S., A. Gorter, M. E. Stuurman, L. A. Van Es, and M. R. Daha. 1987. Activation of the alternative pathway of complement by human serum IgA. Adv. Exp. Med. Biol. 216B: 1297-1302. 14. Pangburn, M. K., R. D. Schreiber, and H. J. Muller-Eberhard. 1983. C3b deposition during activation of the alternative complement pathway and the effect of deposition on the activating surface. J. Immunol. 131: 1930-1935. 15. Banda, N. K., J. M. Thurman, D. Kraus, A. Wood, M. C. Carroll, W. P. Arend, and V. M. Holers. 2006. Alternative complement pathway activation is essential for inflammation and joint destruction in the passive transfer model of collagen-induced arthritis. J. Immunol. 177: 1904-1912. 16. Ji, H., K. Ohmura, U. Mahmood, D. M. Lee, F. M. Hofhuis, S. A. Boackle, K. Takahashi, V. M. Holers, M. Walport, C. Gerard et al. 2002. Arthritis critically dependent on innate immune system players. Immunity. 16: 157-168. 17. Arlaud, G. J., V. Rossi, N. M. Thielens, C. Gaboriaud, B. Bersch, and J. F. Hernandez. 1998. Structural and functional studies on C1r and C1s: new insights into the mechanisms involved in C1 activity and assembly. Immunobiology 199: 303-316.

22

General Introduction 18. Arlaud, G. J., C. Gaboriaud, N. M. Thielens, V. Rossi, B. Bersch, J. F. Hernandez, and J. C. FontecillaCamps. 2001. Structural biology of C1: dissection of a complex molecular machinery. Immunol. Rev. 180: 136-145. 19. Kishore, U. and K. B. Reid. 2000. C1q: structure, function, and receptors. Immunopharmacology 49: 159170. 20. Reid, K. B. and R. R. Porter. 1976. Subunit composition and structure of subcomponent C1q of the first component of human complement. Biochem. J. 155: 19-23. 21. Sim, R. B. 1981. The first component of human complement--C1. Methods Enzymol. 80: 6-16. 22. Sim, R. B. and K. B. Reid. 1991. C1: molecular interactions with activating systems. Immunol. Today 12: 307-311. 23. Dodds, A. W., X. D. Ren, A. C. Willis, and S. K. Law. 1996. The reaction mechanism of the internal thioester in the human complement component C4. Nature 379: 177-179. 24. Law, S. K., N. A. Lichtenberg, and R. P. Levine. 1980. Covalent binding and hemolytic activity of complement proteins. Proc. Natl. Acad. Sci. U. S. A 77: 7194-7198. 25. Law, S. K. and A. W. Dodds. 1997. The internal thioester and the covalent binding properties of the complement proteins C3 and C4. Protein Sci. 6: 263-274. 26. Sim, R. B., T. M. Twose, D. S. Paterson, and E. Sim. 1981. The covalent-binding reaction of complement component C3. Biochem. J. 193: 115-127. 27. Jack, D. L., N. J. Klein, and M. W. Turner. 2001. Mannose-binding lectin: targeting the microbial world for complement attack and opsonophagocytosis. Immunol. Rev. 180: 86-99. 28. Kilpatrick, D. C. 2002. Mannan-binding lectin and its role in innate immunity. Transfus Med 12: 335-352. 29. Turner, M. W. 1996. Mannose-binding lectin: the pluripotent molecule of the innate immune system. Immunol. Today 17: 532-540. 30. Matsushita, M. and T. Fujita. 1992. Activation of the classical complement pathway by mannose-binding protein in association with a novel C1s-like serine protease. J. Exp. Med. 176: 1497-1502. 31. Thiel, S., T. Vorup-Jensen, C. M. Stover, W. Schwaeble, S. B. Laursen, K. Poulsen, A. C. Willis, P. Eggleton, S. Hansen, U. Holmskov et al. 1997. A second serine protease associated with mannan-binding lectin that activates complement. Nature 386: 506-510. 32. Dahl, M. R., S. Thiel, M. Matsushita, T. Fujita, A. C. Willis, T. Christensen, T. Vorup-Jensen, and J. C. Jensenius. 2001. MASP-3 and its association with distinct complexes of the mannan-binding lectin complement activation pathway. Immunity. 15: 127-135. 33. Stover, C. M., S. Thiel, M. Thelen, N. J. Lynch, T. Vorup-Jensen, J. C. Jensenius, and W. J. Schwaeble. 1999. Two constituents of the initiation complex of the mannan-binding lectin activation pathway of complement are encoded by a single structural gene. J. Immunol. 162: 3481-3490. 34. Matsushita, M., S. Thiel, J. C. Jensenius, I. Terai, and T. Fujita. 2000. Proteolytic activities of two types of mannose-binding lectin-associated serine protease. J. Immunol. 165: 2637-2642. 35. Vorup-Jensen, T., S. V. Petersen, A. G. Hansen, K. Poulsen, W. Schwaeble, R. B. Sim, K. B. Reid, S. J. Davis, S. Thiel, and J. C. Jensenius. 2000. Distinct pathways of mannan-binding lectin (MBL)- and C1complex autoactivation revealed by reconstitution of MBL with recombinant MBL-associated serine protease-2. J. Immunol. 165: 2093-2100. 36. Matsushita, M., Y. Endo, and T. Fujita. 2000. Cutting edge: complement-activating complex of ficolin and mannose-binding lectin-associated serine protease. J Immunol. 164: 2281-2284.

23

Chapter 1

37. Matsushita, M., M. Kuraya, N. Hamasaki, M. Tsujimura, H. Shiraki, and T. Fujita. 2002. Activation of the lectin complement pathway by H-ficolin (Hakata antigen). J Immunol. 168: 3502-3506. 38. Liu, Y., Y. Endo, D. Iwaki, M. Nakata, M. Matsushita, I. Wada, K. Inoue, M. Munakata, and T. Fujita. 2005. Human M-ficolin is a secretory protein that activates the lectin complement pathway. J Immunol. 175: 3150-3156. 39. Lynch, N. J., S. Roscher, T. Hartung, S. Morath, M. Matsushita, D. N. Maennel, M. Kuraya, T. Fujita, and W. J. Schwaeble. 2004. L-ficolin specifically binds to lipoteichoic acid, a cell wall constituent of Grampositive bacteria, and activates the lectin pathway of complement. J Immunol. 172: 1198-1202. 40. Kuraya, M., Z. Ming, X. Liu, M. Matsushita, and T. Fujita. 2005. Specific binding of L-ficolin and Hficolin to apoptotic cells leads to complement activation. Immunobiology 209: 689-697. 41. Honore, C., T. Hummelshoj, B. E. Hansen, H. O. Madsen, P. Eggleton, and P. Garred. 2007. The innate immune component ficolin 3 (Hakata antigen) mediates the clearance of late apoptotic cells. Arthritis Rheum. 56: 1598-1607. 42. Tack, B. F., R. A. Harrison, J. Janatova, M. L. Thomas, and J. W. Prahl. 1980. Evidence for presence of an internal thiolester bond in third component of human complement. Proc. Natl. Acad. Sci. U. S. A. 77: 57645768. 43. Law, S. K., T. M. Minich, and R. P. Levine. 1981. Binding reaction between the third human complement protein and small molecules. Biochemistry 20: 7457-7463. 44. Janssen, B. J., E. G. Huizinga, H. C. Raaijmakers, A. Roos, M. R. Daha, K. Nilsson-Ekdahl, B. Nilsson, and P. Gros. 2005. Structures of complement component C3 provide insights into the function and evolution of immunity. Nature 437: 505-511. 45. Pangburn, M. K., R. D. Schreiber, and H. J. Muller-Eberhard. 1981. Formation of the initial C3 convertase of the alternative complement pathway. Acquisition of C3b-like activities by spontaneous hydrolysis of the putative thioester in native C3. J Exp. Med. 154: 856-867. 46. Pangburn, M. K. 2000. Host recognition and target differentiation by factor H, a regulator of the alternative pathway of complement. Immunopharmacology 49: 149-157. 47. Fries, L. F., T. A. Gaither, C. H. Hammer, and M. M. Frank. 1984. C3b covalently bound to IgG demonstrates a reduced rate of inactivation by factors H and I. J Exp. Med. 160: 1640-1655. 48. Jelezarova, E., A. Luginbuehl, and H. U. Lutz. 2003. C3b2-IgG complexes retain dimeric C3 fragments at all levels of inactivation. J Biol. Chem 278: 51806-51812. 49. Lutz, H. U. and E. Jelezarova. 2006. Complement amplification revisited. Mol. Immunol. 43: 2-12. 50. Spitzer, D., L. M. Mitchell, J. P. Atkinson, and D. E. Hourcade. 2007. Properdin can initiate complement activation by binding specific target surfaces and providing a platform for de novo convertase assembly. J Immunol. 179: 2600-2608. 51. Smith, C. A., M. K. Pangburn, C. W. Vogel, and H. J. Muller-Eberhard. 1984. Molecular architecture of human properdin, a positive regulator of the alternative pathway of complement. J Biol. Chem 259: 45824588. 52. Harboe, M., G. Ulvund, L. Vien, M. Fung, and T. E. Mollnes. 2004. The quantitative role of alternative pathway amplification in classical pathway induced terminal complement activation. Clin. Exp. Immunol. 138: 439-446. 53. Nielsen, C. H., E. M. Fischer, and R. G. Leslie. 2000. The role of complement in the acquired immune response. Immunology 100: 4-12.

24

General Introduction 54. Fernandez, H. N., P. M. Henson, A. Otani, and T. E. Hugli. 1978. Chemotactic response to human C3a and C5a anaphylatoxins. I. Evaluation of C3a and C5a leukotaxis in vitro and under stimulated in vivo conditions. J. Immunol. 120: 109-115. 55. Barrington, R., M. Zhang, M. Fischer, and M. C. Carroll. 2001. The role of complement in inflammation and adaptive immunity. Immunol. Rev. 180: 5-15. 56. Carroll, M. C. 2004. The complement system in B cell regulation. Mol. Immunol. 41: 141-146. 57. Cooper, N. R. 1985. The classical complement pathway: activation and regulation of the first complement component. Adv. Immunol. 37: 151-216. 58. Ambrus, G., P. Gal, M. Kojima, K. Szilagyi, J. Balczer, J. Antal, L. Graf, A. Laich, B. E. Moffatt, W. Schwaeble et al. 2003. Natural substrates and inhibitors of mannan-binding lectin-associated serine protease-1 and -2: a study on recombinant catalytic fragments. J. Immunol. 170: 1374-1382. 59. Masaki, T., M. Matsumoto, I. Nakanishi, R. Yasuda, and T. Seya. 1992. Factor I-dependent inactivation of human complement C4b of the classical pathway by C3b/C4b receptor (CR1, CD35) and membrane cofactor protein (MCP, CD46). J. Biochem. (Tokyo) 111: 573-578. 60. Lublin, D. M. and J. P. Atkinson. 1989. Decay-accelerating factor: biochemistry, molecular biology, and function. Annu. Rev. Immunol. 7: 35-58. 61. Oglesby, T. J., C. J. Allen, M. K. Liszewski, D. J. White, and J. P. Atkinson. 1992. Membrane cofactor protein (CD46) protects cells from complement-mediated attack by an intrinsic mechanism. J. Exp. Med. 175: 1547-1551. 62. Meri, S. and M. K. Pangburn. 1990. Discrimination between activators and nonactivators of the alternative pathway of complement: regulation via a sialic acid/polyanion binding site on factor H. Proc. Natl. Acad. Sci. U. S. A 87: 3982-3986. 63. Fujita, T., T. Inoue, K. Ogawa, K. Iida, and N. Tamura. 1987. The mechanism of action of decayaccelerating factor (DAF). DAF inhibits the assembly of C3 convertases by dissociating C2a and Bb. J. Exp. Med. 166: 1221-1228. 64. Nicholson-Weller, A. and C. E. Wang. 1994. Structure and function of decay accelerating factor CD55. J. Lab Clin. Med. 123: 485-491. 65. Podack, E. R., K. T. Preissner, and H. J. Muller-Eberhard. 1984. Inhibition of C9 polymerization within the SC5b-9 complex of complement by S-protein. Acta Pathol. Microbiol. Immunol. Scand. Suppl 284: 89-96. 66. Sugita, Y. and Y. Masuho. 1995. CD59: its role in complement regulation and potential for therapeutic use. Immunotechnology. 1: 157-168. 67. Law, S. K., A. W. Dodds, and R. R. Porter. 1984. A comparison of the properties of two classes, C4A and C4B, of the human complement component C4. EMBO J 3: 1819-1823. 68. Traustadottir, K. H., A. Sigfusson, K. Steinsson, and K. Erlendsson. 2002. C4A deficiency and elevated level of immune complexes: the mechanism behind increased susceptibility to systemic lupus erythematosus. J Rheumatol 29: 2359-2366. 69. Welch, T. R., C. Brickman, N. Bishof, S. Maringhini, M. Rutkowski, M. Frenzke, and N. Kantor. 1998. The phenotype of SLE associated with complete deficiency of complement isotype C4A. J Clin Immunol 18: 48-51. 70. Yang, Y., E. K. Chung, Y. L. Wu, S. L. Savelli, H. N. Nagaraja, B. Zhou, M. Hebert, K. N. Jones, Y. Shu, K. Kitzmiller et al. 2007. Gene copy-number variation and associated polymorphisms of complement component C4 in human systemic lupus erythematosus (SLE): low copy number is a risk factor for and high copy number is a protective factor against SLE susceptibility in European Americans. Am J Hum. Genet. 80: 1037-1054.

25

Chapter 1

71. Bishof, N. A., T. R. Welch, and L. S. Beischel. 1990. C4B deficiency: a risk factor for bacteremia with encapsulated organisms. J. Infect. Dis. 162: 248-250. 72. Rowe, P. C., R. H. McLean, R. A. Wood, R. J. Leggiadro, and J. A. Winkelstein. 1989. Association of homozygous C4B deficiency with bacterial meningitis. J. Infect. Dis. 160: 448-451. 73. Morgan, B. P. and M. J. Walport. 1991. Complement deficiency and disease. Immunol. Today 12: 301-306. 74. Pickering, M. C. and M. J. Walport. 2000. Links between complement abnormalities and systemic lupus erythematosus. Rheumatology. (Oxford) 39: 133-141. 75. Turner, M. W. 2003. The role of mannose-binding lectin in health and disease. Mol. Immunol. 40: 423-429. 76. Reis, S., D. A. Falcao, and L. Isaac. 2006. Clinical aspects and molecular basis of primary deficiencies of complement component C3 and its regulatory proteins factor I and factor H. Scand. J Immunol. 63: 155168. 77. Roord, J. J., van Diemen-van Steenvoorde RA, H. J. Schuurman, G. T. Rijkers, B. J. Zegers, F. H. Gmelig Meyling, and J. W. Stoop. 1989. Membranoproliferative glomerulonephritis in a patient with congenital deficiency of the third component of complement: effect of treatment with plasma. Am J Kidney Dis. 13: 413-417. 78. Nielsen, H. E. and C. Koch. 1987. Meningococcal disease in congenital absence of the fifth component of complement. Scand. J Infect. Dis. 19: 635-639. 79. Nurnberger, W., H. Pietsch, R. Seger, T. Bufon, and V. Wahn. 1989. Familial deficiency of the seventh component of complement associated with recurrent meningococcal infections. Eur. J Pediatr. 148: 758760. 80. Petersen, B. H., T. J. Lee, R. Snyderman, and G. F. Brooks. 1979. Neisseria meningitidis and Neisseria gonorrhoeae bacteremia associated with C6, C7, or C8 deficiency. Ann. Intern. Med. 90: 917-920. 81. Schlesinger, M., Z. Nave, Y. Levy, P. E. Slater, and Z. Fishelson. 1990. Prevalence of hereditary properdin, C7 and C8 deficiencies in patients with meningococcal infections. Clin. Exp. Immunol. 81: 423-427. 82. Carugati, A., E. Pappalardo, L. C. Zingale, and M. Cicardi. 2001. C1-inhibitor deficiency and angioedema. Mol. Immunol. 38: 161-173. 83. Genel, F., A. G. Sjoholm, L. Skattum, and L. Truedsson. 2005. Complement factor I deficiency associated with recurrent infections, vasculitis and immune complex glomerulonephritis. Scand. J Infect. Dis. 37: 615618. 84. Dragon-Durey, M. A., V. Fremeaux-Bacchi, C. Loirat, J. Blouin, P. Niaudet, G. Deschenes, P. Coppo, F. W. Herman, and L. Weiss. 2004. Heterozygous and homozygous factor h deficiencies associated with hemolytic uremic syndrome or membranoproliferative glomerulonephritis: report and genetic analysis of 16 cases. J Am Soc Nephrol. 15: 787-795. 85. Kavanagh, D., E. J. Kemp, E. Mayland, R. J. Winney, J. S. Duffield, G. Warwick, A. Richards, R. Ward, J. A. Goodship, and T. H. Goodship. 2005. Mutations in complement factor I predispose to development of atypical hemolytic uremic syndrome. J Am Soc Nephrol. 16: 2150-2155. 86. Klein, R. J., C. Zeiss, E. Y. Chew, J. Y. Tsai, R. S. Sackler, C. Haynes, A. K. Henning, J. P. SanGiovanni, S. M. Mane, S. T. Mayne et al. 2005. Complement factor H polymorphism in age-related macular degeneration. Science 308: 385-389. 87. Edwards, A. O., R. Ritter, III, K. J. Abel, A. Manning, C. Panhuysen, and L. A. Farrer. 2005. Complement factor H polymorphism and age-related macular degeneration. Science 308: 421-424.

26

General Introduction 88. Haines, J. L., M. A. Hauser, S. Schmidt, W. K. Scott, L. M. Olson, P. Gallins, K. L. Spencer, S. Y. Kwan, M. Noureddine, J. R. Gilbert et al. 2005. Complement factor H variant increases the risk of age-related macular degeneration. Science 308: 419-421. 89. Gold, B., J. E. Merriam, J. Zernant, L. S. Hancox, A. J. Taiber, K. Gehrs, K. Cramer, J. Neel, J. Bergeron, G. R. Barile et al. 2006. Variation in factor B (BF) and complement component 2 (C2) genes is associated with age-related macular degeneration. Nat. Genet. 38: 458-462. 90. Takeda, J., T. Miyata, K. Kawagoe, Y. Iida, Y. Endo, T. Fujita, M. Takahashi, T. Kitani, and T. Kinoshita. 1993. Deficiency of the GPI anchor caused by a somatic mutation of the PIG-A gene in paroxysmal nocturnal hemoglobinuria. Cell 73: 703-711. 91. Nicholson-Weller, A., J. P. March, S. I. Rosenfeld, and K. F. Austen. 1983. Affected erythrocytes of patients with paroxysmal nocturnal hemoglobinuria are deficient in the complement regulatory protein, decay accelerating factor. Proc. Natl. Acad. Sci. U. S. A 80: 5066-5070. 92. Parker, C. J. 2007. The pathophysiology of paroxysmal nocturnal hemoglobinuria. Exp. Hematol. 35: 523533. 93. Hillmen, P., N. S. Young, J. Schubert, R. A. Brodsky, G. Socie, P. Muus, A. Roth, J. Szer, M. O. Elebute, R. Nakamura et al. 2006. The complement inhibitor eculizumab in paroxysmal nocturnal hemoglobinuria. N. Engl. J Med. 355: 1233-1243. 94. Reid, M. E., G. Mallinson, R. B. Sim, J. Poole, V. Pausch, A. H. Merry, Y. W. Liew, and M. J. Tanner. 1991. Biochemical studies on red blood cells from a patient with the Inab phenotype (decay-accelerating factor deficiency). Blood 78: 3291-3297. 95. Tedesco, F., M. Pausa, E. Nardon, M. Introna, A. Mantovani, and A. Dobrina. 1997. The cytolytically inactive terminal complement complex activates endothelial cells to express adhesion molecules and tissue factor procoagulant activity. J Exp. Med. 185: 1619-1627. 96. Griselli, M., J. Herbert, W. L. Hutchinson, K. M. Taylor, M. Sohail, T. Krausz, and M. B. Pepys. 1999. Creactive protein and complement are important mediators of tissue damage in acute myocardial infarction. J Exp. Med. 190: 1733-1740. 97. Nijmeijer, R., W. K. Lagrand, Y. T. Lubbers, C. A. Visser, C. J. Meijer, H. W. Niessen, and C. E. Hack. 2003. C-reactive protein activates complement in infarcted human myocardium. Am. J. Pathol. 163: 269275. 98. Makinde, V. A., G. Senaldi, A. S. Jawad, H. Berry, and D. Vergani. 1989. Reflection of disease activity in rheumatoid arthritis by indices of activation of the classical complement pathway. Ann. Rheum. Dis. 48: 302-306. 99. Manderson, A. P., M. Botto, and M. J. Walport. 2004. The role of complement in the development of systemic lupus erythematosus. Annu. Rev. Immunol. 22: 431-456. 100. Johns, T. G. and C. C. Bernard. 1997. Binding of complement component Clq to myelin oligodendrocyte glycoprotein: a novel mechanism for regulating CNS inflammation. Mol. Immunol. 34: 33-38. 101. Rus, H., C. Cudrici, and F. Niculescu. 2005. C5b-9 complement complex in autoimmune demyelination and multiple sclerosis: dual role in neuroinflammation and neuroprotection. Ann. Med. 37: 97-104. 102. Yasojima, K., C. Schwab, E. G. McGeer, and P. L. McGeer. 1999. Up-regulated production and activation of the complement system in Alzheimer's disease brain. Am J Pathol. 154: 927-936. 103. Zwart, B., C. Ciurana, I. Rensink, R. Manoe, C. E. Hack, and L. A. Aarden. 2004. Complement activation by apoptotic cells occurs predominantly via IgM and is limited to late apoptotic (secondary necrotic) cells. Autoimmunity 37: 95-102.

27

Chapter 1 104. Gershov, D., S. Kim, N. Brot, and K. B. Elkon. 2000. C-Reactive protein binds to apoptotic cells, protects the cells from assembly of the terminal complement components, and sustains an antiinflammatory innate immune response: implications for systemic autoimmunity. J Exp. Med. 192: 1353-1364. 105. Familian, A., B. Zwart, H. G. Huisman, I. Rensink, D. Roem, P. L. Hordijk, L. A. Aarden, and C. E. Hack. 2001. Chromatin-independent binding of serum amyloid P component to apoptotic cells. J Immunol. 167: 647-654. 106. Navratil, J. S., S. C. Watkins, J. J. Wisnieski, and J. M. Ahearn. 2001. The globular heads of C1q specifically recognize surface blebs of apoptotic vascular endothelial cells. J Immunol. 166: 3231-3239. 107. Taylor, P. R., A. Carugati, V. A. Fadok, H. T. Cook, M. Andrews, M. C. Carroll, J. S. Savill, P. M. Henson, M. Botto, and M. J. Walport. 2000. A hierarchical role for classical pathway complement proteins in the clearance of apoptotic cells in vivo. J Exp Med 192: 359-366. 108. Mevorach, D., J. O. Mascarenhas, D. Gershov, and K. B. Elkon. 1998. Complement-dependent clearance of apoptotic cells by human macrophages. J. Exp. Med. 188: 2313-2320. 109. Carroll, M. C. 2004. A protective role for innate immunity in systemic lupus erythematosus. Nat. Rev. Immunol. 4: 825-831. 110. Boackle, S. A. and V. M. Holers. 2003. Role of complement in the development of autoimmunity. Curr. Dir. Autoimmun. 6: 154-168.

28

Complexes between C1q and C3 or C4

Chapter 2 Complexes between C1q and C3 or C4: novel and specific activation markers for classical complement pathway activation Diana Wouters, Hans D Wiessenberg, Margreet Hart, Peter Bruins, Alexandre E Voskuyl, Mohamed R Daha and C Erik Hack

Journal of Immunological Methods (2005) 298: 35-45

29

Chapter 2 Abstract Classical pathway activation is often assessed by measuring circulating levels of activated C4. However, this parameter does not discriminate between activation through the classical or the lectin pathway. We hypothesized that during classical pathway activation, complexes are formed between C1q and activated C4 or C3. Using ELISA, we investigated whether such complexes constitute specific markers for classical pathway activation. In vitro, C1q-C3d/C4d complexes were generated upon incubation of normal recalcified plasma with aggregated IgG or an antiC1q mAb that activates C1 (mAb anti-C1q-130). In contrast, during incubation with C1s or trypsin, C1q-C3d/C4d complexes were not generated, which excludes an innocent bystander effect. Additionally, C1q-C3d/C4d complexes were not generated during activation of the alternative or the lectin pathway. Repeated freezing and thawing did not influence levels of C1qC3d/C4d complexes in recalcified plasma. To measure C1q-complement complexes in plasma samples, we separated unbound complement proteins from C1q-C3d/C4d complexes in the samples prior to testing with ELISA. In samples from patients undergoing cardiopulmonary bypass surgery or suffering from rheumatoid arthritis, we found higher levels of C1q-C4 complexes than in samples from healthy individuals. We conclude that complexes between C1q and C4 or C3 are specific markers of classical complement pathway activation.

30

Complexes between C1q and C3 or C4 Introduction As part of the innate immune system, complement constitutes a first line of defence against many pathogens. The effector pathway of the complement system can be activated through three pathways, the classical, the lectin or the alternative pathway (1, 2). The first component of the classical pathway is C1. This component is composed of C1q and two pro-enzymes, the serine proteases C1r and C1s, which are associated as a Ca2+-dependent tetramer, C1s-C1r-C1r-C1s (3, 4). Binding of C1 to activators such as immune complexes or C-reactive protein (CRP), is mediated by its recognition subunit C1q. Upon binding of C1q, conformational changes trigger the auto activation of C1r. Subsequently, active C1r converts the pro-enzyme C1s into an active protease, which in its turn activates C4 and C2 by limited proteolysis. Activated C4 and C2 form a bimolecular complex, which can function as the C3-convertase of the classical pathway (3-6). The lectin pathway shares some molecules with the classical pathway and is activated by binding of mannose-binding lectin (MBL), a molecule homologous to C1q, to its ligands. Among the latter are mannose or carbohydrate structures on pathogenic surfaces, which can bind MBL in a Ca2+-dependent fashion. Two serine proteases, i.e., the MBL-associated serine protease (MASP)1 and MASP-2, are activated upon binding of MBL. Subsequently activated MASPs cleave C4 and C2 to generate a C3 convertase, which is similar to that formed during classical pathway activation (7). Complement activation occurs in a number of immune and inflammatory diseases. Inhibition of the complement system may even be a target for therapy in some of these (8-10). Hence it may be important to monitor activation of the various complement pathways in patients. To assess activation of the classical pathway, activation of C4 has frequently been measured (11). However, according to current knowledge, activation of this complement protein may result from both classical and lectin pathway activation. Another limitation of measuring activated C4 is that spontaneous hydrolysis of the internal thio-ester in this complement protein (as well as in C3) may result in artificially high levels in a plasma sample, since the resulting iC4 and iC3 expose most of the neo-epitopes of C4b and C3b, respectively (12). As an alternative method to measure classical pathway activation specifically, levels of C1rC1s-C1-inhibitor complexes have been measured (13). The disadvantage of monitoring these C1-inhibitor complexes is, however, their fast clearance in vivo (14). As indicated above, C4 and also C3 contain an internal thio-ester that is exposed during activation of these molecules. This newly exposed thio-ester is highly reactive with amide- or hydroxyl-groups (15-17). The majority of the C4 or C3 molecules exposing their thio-ester will react with surrounding water molecules, but some will bind covalently to molecules in their direct neighbourhood. We hypothesized that fixation of C4 or C3 during complement activation 31

Chapter 2 not only occurs to the activator, but also to C1q thereby generating C1q-C4 and C1q-C3 complexes. In the present study, we have tested this hypothesis and developed an ELISA to detect complexes between C1q and activated C4 or C3. Formation of C1q-C3d/C4d complexes in vitro and in vivo was studied. Our results indicate that these complexes are highly specific for activation of the classical pathway of complement and their measurement may help to monitor classical pathway activation in vivo. Materials and Methods Plasma samples Normal human plasma was obtained by collecting blood in 10 mM EDTA (final concentration) and removing blood cells by centrifugation for 10 min at 1,300 g. Recalcified normal human plasma was obtained by adding a slight excess of CaCl2 (12 mM) to EDTA plasma, whereafter the plasma was allowed to clot. The fibrin clot was then removed by centrifugation (10 min at 1,300 g) and the plasma was stored in aliquots at -70°C until use. EDTA plasma samples from patients undergoing cardiopulmonary bypass (CPB) or suffering from rheumatoid arthritis (RA) were collected as part of ongoing studies (18, 19), and stored in aliquots at -70°C until use. The study was approved by the local institutional ethics review committee.

Blood samples of CPB patients were obtained before and after induction of

anaesthesia, 30 min after the start of CPB, immediately after CPB and after protamine administration. These time points are further referred to as T1-T5, respectively. Proteins and antibodies Trypsin was obtained from Sigma Chem Co, St Louis MO and C1s was from Calbiochem, Darmstadt, Germany. Aggregated human IgG (AHG) was prepared by incubating purified human IgG at a concentration of 10 mg per ml for 20 min at 63°C (12). Monoclonal antibody (mAb) anti-C1q-130, directed against the stem of C1q and able to activate C1 in serum, and mAb anti-C1q-85, which inhibits activation of C1q by immune complexes, have been described before (20). MAb anti-C1q-2 is directed against the stem of C1q. MAb antiC4-1, directed against an activation-dependent neo-epitope exposed on C4b/bi/c, has also been described before (11). MAb anti-C4-4 was obtained from a fusion of spleen cells from a mouse immunized with C4 and is directed against the C4d fragment. The use of this antibody in immunoassays has been described previously (11). MAb anti-C3-9 against an activationdependent epitope on C3b/bi/c, and mAb anti-C3-19 against the C3d fragment have been

32

Complexes between C1q and C3 or C4 published earlier (12). Isolation of C1q from human plasma was described before by Tenner and colleagues (21). Complement activation in plasma Recalcified plasma of healthy donors was incubated for 20 min at 37°C with various complement activators. To this end, one volume of recalcified plasma was incubated with one volume of veronal buffer supplemented with 10 mM CaCl2 and 2 mM MgCl2 (VB++) containing AHG, mAb anti-C1q-130, trypsin or C1s at final concentrations of 0.5 mg/ml, 50 µg/ml, 0.5 mg/ml, and 20 nM, respectively. Activation with AHG and mAb anti-C1q-130 was stopped by adding two volumes of EDTA at a final concentration of 0.2 M in PBS. Trypsin-induced activation was inhibited by the addition of soy bean trypsin inhibitor (SBTI, Sigma) at a final concentration of 3 mg/ml and C1s activation was stopped by adding EDTA and benzamidine (Sigma) at final concentrations of 10 and 100 mM, respectively. Finally, the mixtures were placed on ice. ELISAs to measure overall complement activation Overall complement activation was measured by assessing the amount of C4b/bi/c and C3b/bi/c (abbreviated further as C4b/c and C3b/c, respectively). For the quantification of these activation products, ELISA’s were used that have been described previously (11). Briefly, for the C4b/c ELISA, mAb anti-C4-1, diluted in 0.1 M carbonate buffer, pH 9.6 (2 µg/ml) was coated onto an ELISA plate (Nunc Maxisorp; NalgeNunc Int., Roskilde, Denmark) overnight at room temperature (RT). This and all other incubations were performed in an end volume of 100 µl. The wells were washed five times with PBS/0.02%, w/v, Tween-20 (PT) and subsequently incubated for 30 min at RT with 2%, v/v, cow milk in PBS to block residual binding sites. Hereafter, the wells were washed five times with PT again. Samples, diluted in PT containing 1%, w/v, gelatine (PTG) and 10 mM EDTA, were incubated for 1 h at 4°C. Then the plates were washed five times with PT and incubated with biotinylated polyclonal rabbit anti-human C4c antibodies, diluted in PTG, for 1 h at RT. After five washes with PT, the plates were incubated with 100 µl PTG supplemented with 0.1%, v/v, streptavidin-peroxidase (Amersham/Pharmacia, Uppsala, Sweden) for 30 min at RT. After five washes, the ELISA was developed with 100 µg/ml TMB in 0.11 M sodium acetate (pH 5.5) containing 0.003%, v/v, H2O2. Substrate conversion was stopped by addition of 100 µl H2SO4. Absorbance was measured at 450 nm with a Titertek multiscan.

33

Chapter 2 For measurement of C3b/c levels, a similar ELISA was performed, except that the plates were coated with mAb antiC3-9 (2 µg/ml) and that bound C3b/c was detected with biotinylated polyclonal rabbit anti-human C3c antibodies. ELISAs for C1q-C4d/C3d complexes MAb anti-C1q-85, diluted in 0.1 M carbonate buffer (pH 9.6) to a concentration of 2 µg/ml, was coated overnight at RT onto a 96-well ELISA plate. The wells were washed five times with PT and subsequently all wells were incubated with PBS/0.2% cow milk for 30 min at RT to block the residual binding sites on the plate. The plate was washed five times again with PT. Samples were diluted in PTG-EDTA containing 0.5 M NaCl to prevent non-specific binding of C1q. The plate was then incubated for 1 h at 4°C. After five times washing with PT, the plate was incubated for 60 min at RT with biotinylated mAb against C4d (anti-C4-4) or C3d (anti-C3-19) diluted in PTG, to detect C1q-C4d or C1q-C3d complexes, respectively. After five washes with PT, all wells were incubated with polymerized streptavidin-HRP (Business Unit Reagents, Sanquin Research) diluted 1 to 10000 in PBS/2% cow milk for 30 min at RT. After five washes with PT, ELISAs were developed with TMB as described above. Absorbance at 450 nm was measured with a Titertek multiscan. Plasma incubated with mAb anti-C1q-130 was used as standard (see results). Results obtained with samples were expressed as a percentage of the amount of complexes in the activated plasma sample. Procedure to measure C1q-complement complexes in plasma samples To separate C1q-C4 complexes from unbound C4, one volume of plasma was incubated with one volume of 66% saturated (final concentration 33% or 1.29 M) ammonium sulphate (Merck, Darmstadt, Germany). The ammonium sulphate had been dissolved in PBS containing 10 mM EDTA to prevent complement activation during the precipitation procedure. The mixtures were left on ice for 1 h, and then centrifuged for 30 min at 1,300 g at 4°C. Precipitates were resuspended in ELISA-buffer (high performance ELISA [HPE; Business Unit Immune reagents, Sanquin, Amsterdam, the Netherlands] buffer containing 0.5 M NaCl and 10 mM EDTA to prevent non-specific binding of C1q and in vitro complement activation, respectively). The ELISA was then performed with plates coated with mAb anti-C4-4 (5 µg/ml overnight at RT in 0.11 M sodium acetate buffer, pH 5.5). Plates were washed five times with PT prior to a 1 h incubation of the resuspended precipitates at RT. After five times washing with PT, plates were incubated for 1 h at RT with biotinylated mAb against C1q (anti-C1q-85) to detect C1q-C4 complexes. After five washes with PT, the ELISA was further performed as described above. 34

Complexes between C1q and C3 or C4 Purified C1q, containing complexes, was used as a standard for this assay. Levels of complexes in plasma were expressed as arbitrary units, 100 au being the amount of complexes in the purified C1q sample. SDS-PAGE analysis of C1q complexes MAbs anti-C1q-2, anti-C3-19 and anti-C4-4 were coupled to CNBr-activated Sepharose 4B (Amersham/Pharmacia) at 2-3 mg mAb to 100 mg Sepharose. The Sepharose was suspended in PBS/0.5 M NaCl. For the immune precipitation of C1q, C1q-C3d complexes and C1q-C4d complexes, 30µg purified C1q, diluted in PBS/0.5 M NaCl, was incubated o/n at 4°C with 200 µl anti-C1q-2, anti-C3-19 or anti-C4-4 Sepharose suspension, respectively. After washing the Sepharose five times with PBS, bound proteins were eluted in non-reducing SDS sample buffer, (2% SDS, 62.5 mM Tris-HCl pH 6.8, 10% glycerol and bromophenol blue) by incubation for 5 min at 90°C. The Sepharose beads were removed by centrifugation for 10 min at 3000 rpm and supernatants were electrophoresed on 12% SDS gel under non-reducing conditions. After electrophoresis, proteins were transferred to a PVDF membrane. These blots were first incubated for 1 h at RT with blocking buffer (PBS/5%, v/v, milk powder/0.5%, w/v, BSA/0.1%, w/v, Tween) and then incubated o/n at 4°C with biotinylated mAb anti-C1q-2 or anti-C4-4, diluted in blocking buffer. After three washes with PBS/0.1% Tween the blots were incubated for 45 min at RT with polymerized streptavidin-HRP (Sanquin), diluted 1/1000 in blocking buffer. Subsequently, the blots were washed three times with PBS/0.1% Tween and two times with PBS. Proteins were visualized by chemiluminescence, using ECL (Amersham/Pharmacia). Results ELISA for C1q-C3d/C4d complexes A differential antibody ELISA was used to detect complexes between C1q and the activated complement factors C3 and C4. Antibodies polymerized on the solid phase, potentially can activate complement and, hence, fix C3 and C4 when incubated with fresh serum or recalcified plasma. This may lead to artificially high responses in the assay. To prevent this, samples were incubated in presence of 10 mM EDTA as well as 0.5 M NaCl, to prevent in vitro complement activation. In addition, the mAb against C1q was of the IgG1 mouse isotype which activates complement poorly (22).

35

Chapter 2 Recalcified plasma was activated in the fluid phase with AHG to investigate whether C1q complexes are generated during in vitro classical pathway activation. C3b/c and C4b/c ELISAs showed that complement was substantially activated, as a result of classical pathway activation by IgG aggregates. As depicted in Figure 1A, C1q-C4d complexes were generated in recalcified plasma upon incubation with AHG for 20 min at 37°C. Similar results were obtained for C1qC3d complexes (Fig. 1B). When plates were coated with an irrelevant mAb, no responses were observed in the ELISA when AHG-activated recalcified plasma was tested, indicating the specificity of the assay (data not shown).

1.6

A

A450

1.2 0.8

Figure 1. Generation of C1q-C4d (A) and

0.4

C1q-C3d (B) complexes in recalcified plasma incubated with aggregated human

0 10

100

1000

10000

IgG (AHG). Recalcified plasma was incubated with AHG, final concentration 0.5 mg/ml (♦), for 20 min at 37°C to

1.6

B

A450

1.2

activate

the

complement.

classical C1q

pathway

complexes

of were

measured in the mixtures. As controls, plasma was incubated on ice in the

0.8

presence of 0.2 M EDTA (▲) or at 37°C with VB++ (■). Results represent mean

0.4

and SEM (error bars) of 4 experiments. 0 10

100

1000

10000

sample dilution

Generation of C1q complexes in recalcified plasma by various complement activators To investigate the specificity of C1q-C4d and C1q-C3d complexes for classical pathway activation, various complement activators were tested for their potency to generate these complexes in recalcified plasma. Plasma incubated for 20 min at 37°C with mAb anti-C1q-130, which activates C1 in serum, yielded the highest levels of either C1q-complex. Therefore, we decided to use plasma incubated with this mAb as standard for the amount of C1q complexes.

36

Complexes between C1q and C3 or C4 Levels of C1q-C4d and C1q-C3d complexes in this activated plasma were set at 100 percent, and levels in other samples were related to this standard. As already indicated above, activation with AHG resulted in the formation of C1q-C4d and C1qC3d complexes. Conversely, recalcified plasma activated with trypsin, a protease that amongst others directly cleaves C3 and C4, hardly contained C1q-C4d complexes and only a low level of C1q-C3d complexes. The latter probably resulted from a so-called “innocent-bystander effect”. Also upon incubation with C1s, which activates C4, C1q-C4d or C1q-C3d complexes did not increase in the plasma (Table I). The lack of generation of C1q-C3d/C4d complexes was not due to insufficient activation of complement by any of the activators mentioned, since all activated plasma samples contained high levels of C3b/c and C4b/c (Table I). Additionally, recalcified plasma incubated with mannan or zymosan, activators of the lectin and alternative pathway, respectively, did not contain C1q-C3d/C4d complexes, while these activated plasmas did contain high levels of C3b/c (data not shown). Thus, C1q-C3d/C4d complexes were apparently only generated during classical pathway activation.

C4b/c (nM)

C1q-C4d (%)

C3b/c (nM)

C1q-C3d (%)

MAb-130 (50 µg/ml)

100

100

230

100

AHG (0.5 mg/ml)

1200

8

1200

10

C1s (20 nM)

350

0.1

200

0.16

Trypsin (0.5 mg/ml)

1500

0.2

2400

1.8

VB++

22

0.05

45

0.08

Ice

8.5

0.05

9

0.05

Table I. Generation of C1q-complexes and activated C4 and C3 in plasma by various complement activators.

C1q-C3d/C4d complexes in plasma samples of healthy individuals and RA patients Next, we investigated the presence of C1q-C4 complexes in EDTA plasma samples obtained from healthy individuals or patients to get an impression about the value of these complexes to monitor complement activation in vivo. Unfortunately, we had problems in measuring C1q-C4 in plasma samples, since dose-response curves of plasma samples in the assay described above ere

37

Chapter 2 not parallel with the standard. Furthermore, in gel filtration experiments significant amounts of C1q-C4 in fractions of normal EDTA plasma were detected, whereas without fractionation this plasma seemed to contain hardly any complexes. These results suggested suboptimal detection of C1q complexes due to saturation of the coating Ab with free C1q in the plasma. To improve the detection of C1q-C3d/C4d complexes in plasma, we therefore decided to purify such complexes from plasma samples with a simple one-step procedure, before testing them with ELISA. As we anticipated, the separation of C4 from C1q-C4 was easier than that of C1q from C1q-C4 due to a larger difference in molecular size between the former pair and we adapted the ELISA by using mAb anti-C4-4 as capture Ab and mAb anti-C1q-85 as detector Ab. With purified C1q, which appeared to contain measurable amounts of C1q-C3d/C4d complexes, we established that the reversed assay measured these complexes as well as the initial assay (data not shown). Without separation of C1q-C4 from uncomplexed C4 in normal EDTA plasma, C1q-C4 complexes were detected suboptimally because of saturation of the coat with free C4 in the plasma (Fig. 2).

Figure 2. C1q-C4 complexes are

1.2

precipitated from normal EDTA plasma

1

by ammonium sulphate. EDTA plasma of a healthy donor was incubated with

A450

0.8

33% saturated ammonium sulphate,

0.6

final

concentration,

precipitate

0.4

C1q-C4

Subsequently,

0.2

in

order

to

complexes.

complexes

were

measured in plasma before precipitation

0

(♦), in the supernatant (▲) and in the 1

10

100 sam ple dilution

1000

10000

precipitate

(■).

This

figure

is

representative for 3 experiments which yielded similar results.

After incubation with ammonium sulphate, C1q-C4 complexes could be measured only in the precipitate and not in the supernatant, indicating that the complexes were precipitated. Moreover, most, if not all, C1q was recovered in the precipitate, whereas most of C4 was in the supernatant (data not shown).

38

Complexes between C1q and C3 or C4

Figure 3. C1q-C4 plasma

5.0

levels in healthy individuals and RA patients. Horizontal

U A

lines indicate the median

2.5

0.0

values. The mean level of

control

RA

C1q-C4

was

elevated

in

compared

to

significantly RA

patients

controls,

as

determined by a two-tailed unpaired t test. (P10% of a given factor needs to be activated). Most complement proteins exhibit a wide normal range and serum levels are always influenced by the balance between synthesis and catabolism. Some complement proteins are acute phase proteins. So, an acute phase increase of complement proteins may mask consumption during inflammatory conditions, thereby leaving serum levels within the normal range. Altogether, a more reliable approach to assess complement activation is detection of activation products which are specifically produced during activation, in biological fluids. In the rest of the paper, we will review the various assays that have been described over the last decades to measure specific complement activation products (Table II). Analysis of complement activation products Complement activation in vivo can be assessed by making use of the unique properties of the complement system, such as the generation of activation fragments, multimolecular proteinprotein complexes and the appearance of neo-epitopes on these fragments and complexes. Multimolecular protein-protein complexes may result from either complement deposition or interaction of a complement protease with its inhibitor. Neo-epitopes on activation products are not present on the intact proteins and may be involved in novel functional activities of the protein obtained upon activation. The use of monoclonal antibodies against neo-epitopes exposed on activation products in assays for activated complement components will minimize interference of native components in the assay. Assays for classical and lectin pathway activation Upon activation of C1, C1-inh binds to the serine proteases C1r and C1s; thereby releasing these components from activator-bound C1q. Early activation of the classical pathway is indicated by the presence of stable C1-inh/C1rC1s complexes in the circulation. These complexes can be measured either by radioimmunoassay (RIA) (25) or ELISA. In both assays, the C1-inh/C1rC1s complexes were originally captured by antibodies against C1s and subsequently detected by C1inh specific antibodies. In these assays, native unbound C1s competes with C1s in the complexes, which influences the sensitivity of the assays. To circumvent this, Fure et al developed a modified assay in which a capturing antibody is used that recognizes a neo-epitope on C1-inh when complexed with its substrates (26). 53

Chapter 3

54

CP: Classical pathway, LP: Lectin pathway, AP: Alternative pathway

Table II. Overview of complement activation parameters

Assessment of complement activation (review)

55

Chapter 3 C1-inh/C1rC1s complexes are rapidly cleared from the circulation, which explains why sometimes low circulating levels of these complexes are found despite evidence of complement activation (27). Since MASP-2 is also inhibited by C1-inh, it might be valuable to develop an ELISA to detect complexes between C1-inh and MASP-2 based on the same principle. This may provide the first assay to specifically measure lectin pathway mediated complement activation, since assays that can monitor activation of this pathway specifically, have not been described up until now. Generation of C4 cleavage products provides evidence of classical or lectin pathway mediated complement activation. At the level of C4, several activation fragments can be measured such as C4a, C4d and C4b/c which are formed upon enzymatic cleavage by C1s or MASP-2. C4a is a small anaphylatoxin, which is released after proteolytic cleavage of C4. In vivo, C4a is inactivated to C4a desArg by serum carboxypeptidase N. Quantification of the C4a desArg fragment comprises a precipitation step prior to a double competitive inhibition RIA performed in the supernatant (28). The reason for this precipitation step is that the antibodies used in the assay for C4a to some extent cross-react with native C4, and hence the latter has to be removed from samples to be tested. One disadvantage of measuring C4a is that as an anaphylatoxin, C4a has a short half life. Moreover, C4a levels may not be appropriate to monitor renal disease, since C4a is cleared by the kidneys and impaired clearance by the kidneys could account for high C4a plasma concentration (29). Using a monoclonal antibody (mAb) that reacts with a neo-epitope exposed on the activation products C4b, C4bi and C4c (abbreviated as C4b/c) but not on native C4, C4 activation can be quantified using a relatively simple ELISA method. This neo-epitope reflects conformational changes resulting from disruption of an internal thio-ester bond. The neo-epitope specific mAb is used as capturing antibody and a polyclonal rabbit anti human C4 for detection (30). However, this ELISA, as well as other assays that detect C4 cleavage products, is influenced by in vitro generation of the activation product. Amongst others, this is due to the fact that the thio-ester within C4 is fragile in physiological fluids and is easily hydrolysed upon temperature changes. Hydrolysed C4 generated in this way may lead to artificially high responses in the assay for C4 activation fragments. In the human circulation, C4b is rapidly inactivated by factor I (using CR1, factor H or C4bp as cofactor) resulting in the formation of C4bi. Further degradation of C4bi by factor I yields the fragments C4c and C4d. C4d can be measured in human serum by rocket immunoelectrophoresis (RIE) (31, 32), ELISA or nephelometry (33). The measurement of C4d is complicated by the polymorphic nature of C4; it is well known that the C4 polymorphism lies within the C4d 56

Assessment of complement activation (review) fragment. Using monoclonal antibodies to detect C4d, one should make sure that the antibody reacts with all C4 allotypes. In general, measurements of C4 degradation fragments are complicated by the allelic variation of C4 and the high frequency of heterozygous C4 deficiency. Therefore, C4 split products given as ratio compared to levels of intact C4, may be preferred. As C4d (and also C3d) contains the structural region that mediates binding to biological surfaces, one can postulate that C4d levels in biological fluid insufficiently reflect total C4 activation. However, although this cannot be denied, in general the amount of C4 fixed to an activator will be less than 10% of the total amount of activated C4. The majority of activated C4 will be hydrolyzed by surrounding water molecules and hence the effect of underestimation will be limited. Activated C4 binds covalently to surrounding molecules via its thio-ester, which becomes exposed after proteolytic cleavage. This results in the deposition of C4 to the activator and to proteins in the vicinity. Recently, our group described that C4 not only binds to its activator, but also to C1q, the recognition unit of the classical pathway (34). C1q-C4 complexes can be measured by ELISA and are specific activation products of the classical pathway. A major advantage of measuring C1q-C4 complexes is that unlike many other complement activation products, these complexes are very stable at different storage conditions and are not generated in vitro, at least provided the samples contain EDTA. Complexes between activated C4 and MBL would provide a specific assay for lectin pathway activation. MBL-C4 levels should be depicted as ratio to total MBL, because of the large variation in MBL concentration due to its polymorphism. However, more research is required to develop such an assay. Elevated levels of C4d deposited on erythrocytes and thrombocytes have been described to be highly specific for SLE and may be a useful diagnostic marker (35, 36). During classical pathway activation, C3b has been demonstrated to bind covalently to C4b attached to the target (37). Meri and Pangburn found that C3b also bound covalently to C4b in the fluid phase (38). For the detection of circulating C4-C3 complexes, Zwirner described an ELISA in which two mAbs are combined with specificities for C3b/iC3b/C3dg and C4/C4b/C4d (39). These complexes appeared to be specific for the classical pathway. In the above mentioned ELISA no discrepancy is made between direct C4-C3 complexes and larger immune complexes containing both activated C4 and C3. During classical and lectin pathway activation complexes are formed between C4b and the inhibitor C4bp. An ELISA has been described in which these C4b-C4bp complexes are measured (40). Since C4b bound to C4bp is easily cleaved by factor I in the serum, C4b-C4bp complexes are rapidly lost, which may be the explanation for the limited use of this assay. 57

Chapter 3 Assays for alternative pathway activation Alternative pathway activation can be measured by nephelometry or ELISA using monoclonal antibodies recognizing neo-antigens on the factor B split products Ba and Bb (41, 42). Comparable to C4a, the Ba fragment is cleared by the kidneys (43), which makes this fragment less suitable for evaluation of complement activation in renal diseases. During alternative pathway activation, activated C3b binds factor B which results in the cleavage of factor B by factor D. The so formed C3bBb complex is stabilized by properdin, resulting in the alternative pathway C3-convertase. Properdin will only bind to C3 in the presence of factor B. Therefore, C3bBb(P) complexes can be measured in an ELISA with a properdin specific antibody as a catching antibody and anti-C3 antibody as detecting antibody (44). Not only direct activation of the alternative pathway by activators such as gram negative bacteria is measured in this assay. Triggering of the amplification loop as result of initial classical or lectin pathway activation is detected as well. Assays for terminal pathway activation C3 is the central protein of all three activation pathways. Thus, measurement of C3 cleavage products provides information about overall complement activation. Various C3 activation products can be measured in plasma (45). C3 is broken down by C3-convertases to the smaller C3a and the larger C3b fragment. C3b is inactivated by factor I to from C3bi. Further degradation leads to C3c and C3d fragments. The biologically active anaphylatoxin C3a is released upon C3 activation and rapidly inactivated to the more stable C3adesArg. Circulating C3a levels may be measured by RIA (46, 47), which requires a pre-assay precipitation step to separate native C3 from the C3a fragment, since the antibody used in this assay recognizes both C3 and C3a. An ELISA method developed later (48, 49) was not affected by the presence of native C3, since it makes use of a neo-epitope specific antibody against C3a. C3b, C3bi and C3c (C3b/c) can be measured by ELISA in which the antigen is captured by a mAb that reacts with a neo-epitope exposed on activated C3, but not on native C3 (30, 50, 51). The thio-ester in C3 is fragile and is easily hydrolysed upon temperature changes. Therefore, due to improper handling of the samples, C3b/c levels may be artificially high. The C3d fragment has a long half-life in circulation. To measure C3d in serum, the activation fragment should first be separated from native C3 by PEG precipitation. The C3d fragment remains in supernatant, in which it can be detected by either nephelometry or ELISA (52, 53).

58

Assessment of complement activation (review) C5a can be used as indicator of terminal pathway activation. C5a levels can be measured by RIA or ELISA. However, C5a levels may not be an appropriate parameter in vivo; C5a is cleared from the circulation very rapidly by binding to high affinity receptors on neutrophils (54). Increased concentrations of sC5b-9 (soluble terminal complement complex, consisting of Sprotein, the components C5b, C6, C7, C8 and polymerized C9), reflect complement activation via each activation pathway. The sC5b-9 complex is only generated when the whole activation route is completed. The sC5b-9 complexes are not detectable in normal serum or plasma, are stable in vitro and have a relatively long half-life in vivo. S-protein levels in serum may be the limiting factor in sC5-9 formation. The complexes can be detected by ELISA with antibodies against the complement components C5 and C9. Particularly a monoclonal antibody against neoepitopes expressed on polymerized C9 has been used to further optimize assays for MAC (55, 56). Discussion Excessive activation of complement may harm the host by mediating inflammatory tissue destruction and likely contributes to the pathogenesis of a number of human diseases. Detection of complement activation is not only important for assessment of disease activity in these diseases but may also help to monitor response to treatment. The preferred approach to evaluate complement activation is detection of complement activation products with assays that are specific for cleavage fragments or multimolecular complexes. Monoclonal antibodies against neo-antigens exposed on these fragments or multimolecular complexes have been shown to be suitable tools in these assays. Before assessing complement activation in vivo, it is important to consider which activation parameter is most appropriate for a given clinical condition. In the next paragraphs we will give some thoughts on aspects of activation products that may help to select an optimal parameter for assessing complement activation in vivo. In normal situation, most complement activation products are only present in trace amounts in vivo, whereas they are rapidly generated in vitro. Therefore, the conditions of sample collection, processing and storage are critical to get results that reliably reflect in vivo activation processes. For most assays, blood should be collected in EDTA containing tubes, to prevent in vitro activation of both the classical and alternative pathway by chelating Ca2+ and Mg2+. The plasma should be processed soon after collection and preferably stored at -70˚C until analysis. Preferably, several aliquots of samples are stored since this may avoid repeated freezing and thawing of samples. Repeated freezing and thawing may result in false-positive results since the thio-ester in both C3 and C4 is fragile and gets easily hydrolyzed (57). Disruption of the thio59

Chapter 3 ester causes conformational changes in C4 and C3 that are similar to the changes occurring upon activation. Nevertheless, samples are often collected and stored under suboptimal conditions, which may lead to artificially high levels of activation products. In these situations measurement of activation products that are stable in vitro, is preferred. C1q-C4 complexes are very stable activation products that are not generated in vitro in the presence of EDTA (34). However, these complexes only reflect classical pathway activation. To assess total complement activation in samples that were not carefully processed, measurement of sC5b-9 complexes is a reliable indicator (56). Next to handling samples correctly, one should realize that complement activation products have different clearance rates. This has an effect on circulating levels of these products. Table III shows a calculation model for C3 cleavage products demonstrating to what extent serum levels of these activation products are influenced by their different half- lives in vivo. Theoretically, one would expect comparable molar increases of the individual C3 activation products after induction of complement activation. However, the concentrations of these products may differ markedly, since small fragments are in general more rapidly removed from the circulation than larger fragments. The median normal C3 concentration in serum is 1,5 g/L, so when all C3 is converted, the molar concentration of each C3 cleavage product would be 8283 nM. Table III shows the molar concentration of C3a, C3b/c and C3d after acute activation of 5%, 10% or 25% of total C3. Half-lives of these products are 7 minutes, 20 minutes and 4 hours respectively. In case of acute conversion of 5% of total serum C3, elevated C3d levels can be measured up until 4 hours after induction of complement activation, whereas C3a and C3b/c levels have returned to normal values already after 1 hour. Notably, the formation of C3d does not occur instantaneously following acute activation as the breakdown of C3bi into C3c and C3d by factor I and cofactors takes some time. At 10% C3 conversion, elevated C3b/c can be detected up to one hour after the event. However, increased C3a levels can only be detected very shortly after the event, due to its shorter half-life. Even at 25% C3 activation, C3a levels are again within the normal range one hour after induction of complement activation. Thus, C3d levels stay longer increased after induction of complement activation than levels of C3b/c and C3a values, which rapidly return into the normal range. The comparison of activation products however can be very useful, as a high concentration of C3d with normal C3b/c may indicate acute activation of short duration, whereas high levels of both parameters point to strong ongoing activation. In case of mild ongoing activation, again C3d levels may be elevated.

60

Assessment of complement activation (review) % C3 activation at t = 0 h

Parameter

0h

1h

2h

4h

24 h

5%

C3a (nM)*

414

1

nd

nd

nd

C3b/c (nM)**

414

52

7

nd

nd

C3d (nM)***

414

362

311

207

7

C3a

828

2

nd

nd

nd

C3b/c

828

104

13

nd

nd

C3d

828

725

621

414

13

C3a

2070

5

nd

nd

nd

C3b/c

2070

259

32

0.5

nd

C3d

2070

1811

1553

1035

32

10%

25%

Table III. Calculation model for C3 cleavage products *Half-life C3a: 7 min, normal value < 6 nM; **Half-life C3b/c: 20 min, normal value < 57 nM; ***Half-life C3d: 4 hrs, normal value 41-257 nM Total C3 concentration: 1500 µg/ml (8283 nM), nd: not detectable

The effect of different half-lives of C3 cleavage products on the levels that can be measured in patients is illustrated in a study on complement activation after cardiopulmonary bypass surgery (58). In this study, C3a and C3b/c levels were measured as parameters of complement activation. After induction of complement activation during surgery, the molar concentration of C3b/c was almost tenfold higher than that of C3a, whereas initially these products were equally increased. Knowledge of the clearance rates of complement activation products is therefore important to select the most appropriate activation products for clinical and experimental studies. In chronic conditions, complement activation products with a long half-life are preferred like C3d or sC5b9. On the contrary, for acute activation processes products that are rapidly generated and cleared are more appropriate such as C3a, C3b/c and hence elevated levels of these products point to a recent or still ongoing activation process. In this review we gave a short overview of the various assays that are currently available to measure complement activation in vivo. In conclusion, for reliable assessment of complement activation in clinical settings the following characteristics are important to consider: in vitro stability of the activation product, half-life in vivo, acute phase reactivity, influence of renal function, pathway specificity and diagnostic or prognostic value.

61

Chapter 3 References 1. Holmskov, U., S. Thiel, and J. C. Jensenius. 2003. Collections and ficolins: humoral lectins of the innate immune defense. Annu. Rev. Immunol. 21: 547-578. 2. Matsushita, M., S. Thiel, J. C. Jensenius, I. Terai, and T. Fujita. 2000. Proteolytic activities of two types of mannose-binding lectin-associated serine protease. J. Immunol. 165: 2637-2642. 3. Vorup-Jensen, T., S. V. Petersen, A. G. Hansen, K. Poulsen, W. Schwaeble, R. B. Sim, K. B. Reid, S. J. Davis, S. Thiel, and J. C. Jensenius. 2000. Distinct pathways of mannan-binding lectin (MBL)- and C1complex autoactivation revealed by reconstitution of MBL with recombinant MBL-associated serine protease-2. J. Immunol. 165: 2093-2100. 4. Makinde, V. A., G. Senaldi, A. S. Jawad, H. Berry, and D. Vergani. 1989. Reflection of disease activity in rheumatoid arthritis by indices of activation of the classical complement pathway. Ann. Rheum. Dis. 48: 302306. 5. Hietala, M. A., K. S. Nandakumar, L. Persson, S. Fahlen, R. Holmdahl, and M. Pekna. 2004. Complement activation by both classical and alternative pathways is critical for the effector phase of arthritis. Eur. J Immunol. 34: 1208-1216. 6. Petersen, N. E., J. Elmgreen, B. Teisner, and S. E. Svehag. 1988. Activation of classical pathway complement in chronic inflammation. Elevated levels of circulating C3d and C4d split products in rheumatoid arthritis and Crohn's disease. Acta Med. Scand. 223: 557-560. 7. Pickering, M. C. and M. J. Walport. 2000. Links between complement abnormalities and systemic lupus erythematosus. Rheumatology. (Oxford) 39: 133-141. 8. Manderson, A. P., M. Botto, and M. J. Walport. 2004. The role of complement in the development of systemic lupus erythematosus. Annu. Rev. Immunol. 22: 431-456. 9. Nijmeijer, R., W. K. Lagrand, Y. T. Lubbers, C. A. Visser, C. J. Meijer, H. W. Niessen, and C. E. Hack. 2003. C-reactive protein activates complement in infarcted human myocardium. Am. J. Pathol. 163: 269-275. 10. Griselli, M., J. Herbert, W. L. Hutchinson, K. M. Taylor, M. Sohail, T. Krausz, and M. B. Pepys. 1999. Creactive protein and complement are important mediators of tissue damage in acute myocardial infarction. J Exp. Med. 190: 1733-1740. 11. Davis, B. K. and T. Cavallo. 1976. Membranoproliferative glomerulonephritis. Localization of early components of complement in glomerular deposits. Am J Pathol. 84: 283-298. 12. Trouw, L. A., M. A. Seelen, and M. R. Daha. 2003. Complement and renal disease. Mol. Immunol. 40: 125134. 13. Johns, T. G. and C. C. Bernard. 1997. Binding of complement component Clq to myelin oligodendrocyte glycoprotein: a novel mechanism for regulating CNS inflammation. Mol. Immunol. 34: 33-38. 14. Rus, H., C. Cudrici, and F. Niculescu. 2005. C5b-9 complement complex in autoimmune demyelination and multiple sclerosis: dual role in neuroinflammation and neuroprotection. Ann. Med. 37: 97-104. 15. Yasojima, K., C. Schwab, E. G. McGeer, and P. L. McGeer. 1999. Up-regulated production and activation of the complement system in Alzheimer's disease brain. Am J Pathol. 154: 927-936. 16. Bergamaschini, L., S. Canziani, B. Bottasso, M. Cugno, P. Braidotti, and A. Agostoni. 1999. Alzheimer's beta-amyloid peptides can activate the early components of complement classical pathway in a C1qindependent manner. Clin. Exp. Immunol. 115: 526-533. 17. Haines, J. L., M. A. Hauser, S. Schmidt, W. K. Scott, L. M. Olson, P. Gallins, K. L. Spencer, S. Y. Kwan, M. Noureddine, J. R. Gilbert et al. 2005. Complement factor H variant increases the risk of age-related macular degeneration. Science 308: 419-421.

62

Assessment of complement activation (review) 18. Mullins, R. F., N. Aptsiauri, and G. S. Hageman. 2001. Structure and composition of drusen associated with glomerulonephritis: implications for the role of complement activation in drusen biogenesis. Eye 15: 390-395. 19. Choi, G., M. R. Soeters, H. Farkas, L. Varga, K. Obtulowicz, B. Bilo, G. Porebski, C. E. Hack, R. Verdonk, J. Nuijens et al. 2007. Recombinant human C1-inhibitor in the treatment of acute angioedema attacks. Transfusion 47: 1028-1032. 20. Weisman, H. F., T. Bartow, M. K. Leppo, H. C. J. Marsh, G. R. Carson, M. F. Concino, M. P. Boyle, K. H. Roux, M. L. Weisfeldt, and D. T. Fearon. 1990. Soluble human complement receptor type 1: in vivo inhibitor of complement suppressing post-ischemic myocardial inflammation and necrosis. Science 249: 146-151. 21. 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 humanized anti-C5 antibody and single-chain Fv. Mol. Immunol. 33: 1389-1401. 22. Carugati, A., E. Pappalardo, L. C. Zingale, and M. Cicardi. 2001. C1-inhibitor deficiency and angioedema. Mol. Immunol. 38: 161-173. 23. Hillmen, P., N. S. Young, J. Schubert, R. A. Brodsky, G. Socie, P. Muus, A. Roth, J. Szer, M. O. Elebute, R. Nakamura et al. 2006. The complement inhibitor eculizumab in paroxysmal nocturnal hemoglobinuria. N. Engl. J Med. 355: 1233-1243. 24. Hill, A., P. Hillmen, S. J. Richards, D. Elebute, J. C. Marsh, J. Chan, C. F. Mojcik, and R. P. Rother. 2005. Sustained response and long-term safety of eculizumab in paroxysmal nocturnal hemoglobinuria. Blood 106: 2559-2565. 25. Hack, C. E., A. J. Hannema, A. J. Eerenberg-Belmer, T. A. Out, and R. C. Aalberse. 1981. A C1-inhibitorcomplex assay (INCA): a method to detect C1 activation in vitro and in vivo. J. Immunol. 127: 1450-1453. 26. Fure, H., E. W. Nielsen, C. E. Hack, and T. E. Mollnes. 1997. A neoepitope-based enzyme immunoassay for quantification of C1-inhibitor in complex with C1r and C1s. Scand. J. Immunol. 46: 553-557. 27. de Smet, B. J., J. P. de Boer, J. Agterberg, G. Rigter, W. K. Bleeker, and C. E. Hack. 1993. Clearance of human native, proteinase-complexed, and proteolytically inactivated C1-inhibitor in rats. Blood 81: 56-61. 28. Gorski, J. P. 1981. Quantitation of human complement fragment C4ai in physiological fluids by competitive inhibition radioimmune assay. J. Immunol. Methods 47: 61-73. 29. Abou-Ragheb, H. H., A. J. Williams, C. B. Brown, and A. Milford-Ward. 1991. Plasma levels and mode of excretion of the anaphylatoxins C3a and C4a in renal disease. J Clin. Lab Immunol. 35: 113-119. 30. Wolbink, G. J., J. Bollen, J. W. Baars, R. J. ten Berge, A. J. Swaak, J. Paardekooper, and C. E. Hack. 1993. Application of a monoclonal antibody against a neoepitope on activated C4 in an ELISA for the quantification of complement activation via the classical pathway. J. Immunol. Methods 163: 67-76. 31. Nitsche, J. F., E. S. Tucker, III, S. Sugimoto, J. H. Vaughan, and J. G. Curd. 1981. Rocket immunoelectrophoresis of C4 and C4d. A simple sensitive method for detecting complement activation in plasma. Am. J. Clin. Pathol. 76: 679-684. 32. Milgrom, H., J. G. Curd, R. A. Kaplan, H. J. Muller-Eberhard, and J. H. Vaughan. 1980. Activation of the fourth component of complement (C4): assessment by rocket immunoelectrophoresis and correlation with the metabolism of C4. J. Immunol. 124: 2780-2785. 33. Davies, E. T., B. A. Nasaruddin, A. Alhaq, G. Senaldi, and D. Vergani. 1988. Clinical application of new technique that measures C4d for assessment of activation of classical complement pathway. J. Clin. Pathol. 41: 143-147. 34. Wouters, D., H. D. Wiessenberg, M. Hart, P. Bruins, A. Voskuyl, M. R. Daha, and C. E. Hack. 2005. Complexes between C1q and C3 or C4: Novel and specific markers for classical complement pathway activation. J Immunol Methods 298: 35-45.

63

Chapter 3 35. Manzi, S., J. S. Navratil, M. J. Ruffing, C. C. Liu, N. Danchenko, S. E. Nilson, S. Krishnaswami, D. E. King, A. H. Kao, and J. M. Ahearn. 2004. Measurement of erythrocyte C4d and complement receptor 1 in systemic lupus erythematosus. Arthritis Rheum 50: 3596-3604. 36. Navratil, J. S., S. Manzi, A. H. Kao, S. Krishnaswami, C. C. Liu, M. J. Ruffing, P. S. Shaw, A. C. Nilson, E. R. Dryden, J. J. Johnson et al. 2006. Platelet C4d is highly specific for systemic lupus erythematosus. Arthritis Rheum. 54: 670-674. 37. Takata, Y., T. Kinoshita, H. Kozono, J. Takeda, E. Tanaka, K. Hong, and K. Inoue. 1987. Covalent association of C3b with C4b within C5 convertase of the classical complement pathway. J. Exp. Med. 165: 1494-1507. 38. Meri, S. and M. K. Pangburn. 1990. A mechanism of activation of the alternative complement pathway by the classical pathway: protection of C3b from inactivation by covalent attachment to C4b. Eur. J. Immunol. 20: 2555-2561. 39. Zwirner, J., G. Dobos, and O. Gotze. 1995. A novel ELISA for the assessment of classical pathway of complement activation in vivo by measurement of C4-C3 complexes. J Immunol Methods 186: 55-63. 40. Ito, S., T. Fujita, and N. Tamura. 1987. Determination of C4b.C4-bp complex formed by the activation of classical complement pathway using an enzyme-linked immunosorbent assay. J. Immunol. Methods 105: 145150. 41. Senaldi, G., M. Peakman, A. Alhaq, V. A. Makinde, D. E. Tee, and D. Vergani. 1987. Activation of the alternative complement pathway: clinical application of a new technique to measure fragment Ba. J. Clin. Pathol. 40: 1235-1239. 42. Kolb, W. P., P. R. Morrow, and J. D. Tamerius. 1989. Ba and Bb fragments of factor B activation: fragment production, biological activities, neoepitope expression and quantitation in clinical samples. Complement Inflamm. 6: 175-204. 43. Oppermann, M., C. Kurts, R. Zierz, E. Quentin, M. H. Weber, and O. Gotze. 1991. Elevated plasma levels of the immunosuppressive complement fragment Ba in renal failure. Kidney Int. 40: 939-947. 44. Mayes, J. T., R. D. Schreiber, and N. R. Cooper. 1984. Development and application of an enzyme-linked immunosorbent assay for the quantitation of alternative complement pathway activation in human serum. J. Clin. Invest 73: 160-170. 45. Teisner, B., I. Brandslund, N. Grunnet, L. K. Hansen, J. Thellesen, and S. E. Svehag. 1983. Acute complement activation during an anaphylactoid reaction to blood transfusion and the disappearance rate of C3c and C3d from the circulation. J Clin. Lab Immunol. 12: 63-67. 46. Hack, C. E., J. Paardekooper, A. J. Eerenberg, G. O. Navis, M. W. Nijsten, L. G. Thijs, and J. H. Nuijens. 1988. A modified competitive inhibition radioimmunoassay for the detection of C3a. Use of 125I-C3 instead of 125I-C3a. J. Immunol. Methods 108: 77-84. 47. Chenoweth, D. E., S. W. Cooper, T. E. Hugli, R. W. Stewart, E. H. Blackstone, and J. W. Kirklin. 1981. Complement activation during cardiopulmonary bypass: evidence for generation of C3a and C5a anaphylatoxins. N. Engl. J Med. 304: 497-503. 48. Burger, R., G. Zilow, A. Bader, A. Friedlein, and W. Naser. 1988. The C terminus of the anaphylatoxin C3a generated upon complement activation represents a neoantigenic determinant with diagnostic potential. J. Immunol. 141: 553-558. 49. Zilow, G., W. Naser, R. Rutz, and R. Burger. 1989. Quantitation of the anaphylatoxin C3a in the presence of C3 by a novel sandwich ELISA using monoclonal antibody to a C3a neoepitope. J. Immunol. Methods 121: 261-268.

64

Assessment of complement activation (review) 50. Hack, C. E., J. Paardekooper, R. J. Smeenk, J. Abbink, A. J. Eerenberg, and J. H. Nuijens. 1988. Disruption of the internal thioester bond in the third component of complement (C3) results in the exposure of neodeterminants also present on activation products of C3. An analysis with monoclonal antibodies. J. Immunol. 141: 1602-1609. 51. Hack, C. E., J. Paardekooper, and F. Van Milligen. 1990. Demonstration in human plasma of a form of C3 that has the conformation of "C3b-like C3". J Immunol 144: 4249-4255. 52. Vergani, D., L. Bevis, B. A. Nasaruddin, G. Mieli-Vergani, and D. E. Tee. 1983. Clinical application of a new nephelometric technique to measure complement activation. J. Clin. Pathol. 36: 793-797. 53. Mollnes, T. E. 1985. Quantification of the C3d split products of human complement by a sensitive enzymelinked immunosorbent assay. Scand. J Immunol. 21: 607-613. 54. Oppermann, M. and O. Gotze. 1994. Plasma clearance of the human C5a anaphylatoxin by binding to leucocyte C5a receptors. Immunology 82: 516-521. 55. Gawryl, M. S., M. T. Simon, J. L. Eatman, and T. F. Lint. 1986. An enzyme-linked immunoabsorbent assay for the quantitation of the terminal complement complex from cell membranes or in activated human sera. J. Immunol. Methods 95: 217-225. 56. Mollnes, T. E., T. Lea, S. S. Froland, and M. Harboe. 1985. Quantification of the terminal complement complex in human plasma by an enzyme-linked immunosorbent assay based on monoclonal antibodies against a neoantigen of the complex. Scand. J Immunol. 22: 197-202. 57. Mollnes, T. E., P. Garred, and G. Bergseth. 1988. Effect of time, temperature and anticoagulants on in vitro complement activation: consequences for collection and preservation of samples to be examined for complement activation. Clin. Exp. Immunol. 73: 484-488. 58. Bruins, P., H. te Velthuis., A. P. Yazdanbakhsh, P. G. Jansen, F. W. van Hardevelt, E. M. de Beaumont, C. R. Wildevuur, L. Eijsman, A. Trouwborst, and C. E. Hack. 1997. Activation of the complement system during and after cardiopulmonary bypass surgery: postsurgery activation involves C-reactive protein and is associated with postoperative arrhythmia. Circulation 96: 3542-3548. 59. Buysmann, S., C. E. Hack, F. N. van Diepen, J. Surachno, and I. J. ten Berge. 1997. Administration of OKT3 as a two-hour infusion attenuates first-dose side effects. Transplantation 64: 1620-1623. 60. Pfeifer, P. H., M. S. Kawahara, and T. E. Hugli. 1999. Possible mechanism for in vitro complement activation in blood and plasma samples: futhan/EDTA controls in vitro complement activation. Clin. Chem 45: 11901199.

65

66

C1q-C4 complexes in Rheumatoid Arthritis

Chapter 4 Evaluation of classical complement pathway activation in Rheumatoid Arthritis Measurement of C1q-C4 complexes as novel activation products Diana Wouters, Alexandre E Voskuyl, Esmeralda TH Molenaar, Ben AC Dijkmans, C Erik Hack

Arthritis and Rheumatism (2006) 54: 1143- 1150

67

Chapter 4 Abstract Objective. Novel activation products that are stable and minimally susceptible to in vitro artefacts have recently been described in the classical complement pathway. The present study assessed circulating levels of these products, i.e. covalent complexes between the recognition molecule of the classical pathway (C1q) and activated C4, in plasma samples from patients with rheumatoid arthritis (RA) to establish the relationship between these levels and the clinical and immunological parameters in these patients. Methods. C1q-C4 levels were measured in the plasma of 41 patients with active RA and 43 patients with inactive RA. These levels were related to other complement activation products and to disease activity according to the Disease Aactivity Score in 28 joints (DAS28), using Spearman’s rank correlations. Results. C1q-C4 plasma levels were significantly higher in patients with active RA as compared with patients with RA in clinical remission (median 3.3 arbitrary units [AU], range 0.4-13.4 versus 1.7 AU, range 0.2-5.5; P = 0.0001), suggesting that activation of the classical pathway reflects disease activity. This is supported by a significant correlation between C1q-C4 levels and the DAS28 (r = 0.398, P = 0.0002). Levels of other complement activation products, such as activated C4 (C4b/c), were also significantly elevated in patients with active disease compared with patients with inactive disease (P = 0.03), and were correlated with C1q-C4 levels (r = 0.329, P = 0.002). Levels of C1q-C4 complexes were higher in synovial fluid samples than in plasma from the 4 patients tested. Conclusion. Systemic complement activation via the classical pathway in patients with RA correlates with disease activity. These results indicate that C1q-C4 complexes may be used as a biomarker for RA.

68

C1q-C4 complexes in Rheumatoid Arthritis Introduction Rheumatoid arthritis (RA) is an inflammatory autoimmune disease characterized by chronic inflammation of the joints, eventually leading to bone and cartilage destruction. Although the etiology of RA is unknown, complement activation has been implicated in the pathogenesis of the disease. Various studies identify complement activation as a main event in the inflammatory cascade in RA (1-3). Evidence of complement activation in synovial fluid is abundant. For example, levels of complement proteins are depressed in the synovial fluid of patients with RA, reflecting consumption of complement. Moreover, elevated levels of complement cleavage products, such as sC5b-9, C3a, Bb and C1inh-C1s complexes, have been observed in synovial fluid (4-9). Involvement of complement in the pathogenesis of RA was also confirmed in experimental studies. Collagen-induced arthritis (CIA), an experimental animal model for human RA, was induced in C3- and factor B deficient mice (10). In complement-deficient mice, arthritis was reduced or completely absent, whereas normal mice were susceptible for CIA, indicating an important role of complement in the induction of disease. Consistent with this observation, systemic administration of a monoclonal anti-C5 antibody prevented CIA in susceptible mice (11). RA patients have increased levels of circulating immune complexes (12, 13). Part of these complexes contains rheumatoid factors (RFs), which are autoantibodies against human IgG. RFcontaining immune complexes are capable of activating complement via the classical pathway (14-17), with IgM-RF being considerably more effective in complement activation than IgG-RF (18). Thus, immune complexes, particularly those in the inflamed joints, are often assumed to be the main trigger for complement activation in RA. It is not clear to what extent circulating immune complexes contribute to complement activation in RA (19). However, other activators of complement may also contribute to complement activation in RA. Levels of the acute-phase protein, C-reactive protein (CRP), are elevated in the majority of patients with RA, and are associated with disease activity. CRP bound to a ligand, can activate complement and there is evidence that CRP-mediated activation of complement occurs in RA (20). Both immune complexes and ligand-bound CRP activate complement via the classical pathway (18, 21-23). A parameter that reliably reflects complement activation in vivo might constitute a biomarker in RA. However, most, if not all, activation markers of the complement system are susceptible to in vitro artefacts, resulting in artificially high levels of activation products in plasma samples. We recently described novel activation products of complement that not only are specific for activation of the classical pathway of complement, but also are remarkably stable in plasma (24). 69

Chapter 4 The present study was performed to investigate plasma levels of this new complement parameter, which consists of covalent complexes between the recognition molecule of the classical pathway, C1q, and activated C4, in RA. We also sought to establish the relationship between classical pathway activation and other immunological parameters in patients with active or inactive RA, and we assessed the association of these parameters with disease activity. Patients and Methods Patients We selected patients with active RA (n = 41), and patients with inactive RA (n = 43) from a cohort of 187 patients investigated previously (20). All patients with RA fulfilled the American College of Rheumatology (ACR; formerly, the American Rheumatology Association) 1987 criteria for RA (25). Inactive RA was defined by the ACR criteria for clinical remission (26). Disease activity was assessed by calculating the modified Disease Activity Score in 28 joints (DAS28) (27). In addition, 10 patients treated with anti-tumor necrosis factor (anti-TNF) antibody (infliximab) were studied. All patients received 3 mg/kg infliximab at weeks 0, 2, 6 and 14 and every 8 weeks thereafter. Evaluation of disease activity and collection of plasma samples were carried out in these patients at baseline and at weeks 22 and 52 after the start of therapy. The protocol was approved by the local institutional ethics review committee. For control experiments, human EDTA plasma was obtained from 21 healthy volunteers. Collection of blood samples Blood was collected in 10 mM EDTA (final concentration) and centrifuged for 10 minutes at 1,300g to remove the blood cells. Plasma was stored in aliquots at -70°C until use. Proteins and antibodies Monoclonal antibody (mAb) anti-C1q-85, directed against the globular heads of C1q, has been described previously (28). The mAb anti-C4-4 was obtained from a fusion of spleen cells from a mouse immunized with C4 and is directed against the C4d fragment, recognizing both native and activated C4. The use of this antibody in immunoassays has been described previously (29). The mAb anti-C4-1, directed against an activation-dependent neoepitope, has also been described previously (30). C1q was isolated from human plasma according to the method of Tenner and colleagues (31).

70

C1q-C4 complexes in Rheumatoid Arthritis Enzyme-linked immunosorbent assay (ELISA) for activated C4 Overall complement activation was measured by assessing the amount of C4b/bi/c (abbreviated further as C4b/c). For the quantification of these activation products, an ELISA was used that has been described previously (29). Briefly, a mAb (anti-C4-1) recognizing a neoepitope on activated C4 was used as catching antibody. Biotinylated polyclonal rabbit anti-human anti-C4 antibody was used for detection. Since C4 can be hydrolysed by surrounding water molecules, C4b/c levels in plasma or serum may be artificially high due to temperature changes. Aged human serum, containing a known amount of activated C4, was used for the calibration curve. ELISA for C1q-C4 complexes To measure C1q-C4 complexes, an ELISA that was recently described by our group (24) was used. The catching antibody in this ELISA, anti-C4-4, recognizes both native and activated C4. To prevent saturation of the coating antibody C1q-C4 complexes were separated from unbound C4 by ammonium sulphate precipitation. To this end, 1 volume of plasma was incubated with 1 volume of 66% saturated (final concentration 33% or 1.29M) ammonium sulphate (Merck, Darmstadt, Germany), dissolved in phosphate buffered saline (PBS), pH 7.4 containing 10 mM EDTA. Mixtures were placed on ice for 1 hour, and then centrifuged for 30 minutes at 1,300g at 4°C. Precipitates were dissolved in ELISA-buffer (high-performance ELISA buffer [HPE] [Business Unit Immune Reagents; Sanquin, Amsterdam, the Netherlands] to which 0.5 M NaCl and 10 mM EDTA were added to prevent aspecific binding of C1q and in vitro complement activation, respectively). The ELISA was then performed with plates coated with anti-C4-4 mAb at 5 µg/ml overnight at room temperature (RT), in 0.11M sodium acetate buffer, pH 5.5. Plates were washed 5 times with PBS-0.02% (weight/volume) Tween prior to 1 hour of incubation with dissolved precipitates at RT. After washing 5 times with PBS-0.02% Tween, plates were incubated for 1 hour at RT with biotinylated mAb against C1q (anti-C1q-85) diluted in HPE to detect C1q-C4 complexes. After 5 washes with PBS-0.02% Tween, plates were incubated with polymerized horse radish peroxidase (Business Unit Reagents, Sanquin), diluted 1:10.000 in PBS-2% (v/v) cow milk for 30 minutes at RT. After 5 washes, the ELISA was developed with 100 µg/ml tetramethylbenzidine in 0.11M sodium acetate, pH 5.5, containing 0.003% (v/v) H2O2. Substrate conversion was stopped by addition of 100 µl H2SO4. Absorbance was measured at 450 nm with a Titertek multiscan. Purified C1q, containing C1q-C4 complexes, was used as the calibration curve for this assay. Levels of complexes in plasma were expressed as arbitrary units (AU), based on the amount of complexes in the purified C1q sample. 71

Chapter 4 Absorption of RFs from plasma One volume of plasma was incubated for 30 minutes at RT with 4 volumes of RF neutralization agent (human gamma globulin-coated microparticles; Abbott Laboratories, Abbott Park, IL) and subsequently centrifuged for 10 minutes at 2,000 revolutions per minute. Supernatant was used for testing. Determination of IgM-RFs IgM-RFs were measured in a regular ELISA using human IgG (25 µg/ml) as antigen. A positive plasma sample containing 200 international units (IU) of IgM-RF was used for the calibration curve. IgM-RF levels above 12.5 IU/ml were considered to be positive. Statistical analysis Levels of complement activation products above the upper limit of normal values were considered to be increased. Comparisons between patients with active disease and those with inactive disease were made using the Mann-Whitney, the unpaired t-test or the chi-square test, depending on whether the values were normally distributed. Comparisons between samples before and after freezing and thawing were made using the paired t-test. Correlations between the various complement activation products and disease activity were analyzed using Spearman’s rank correlation coefficients. P values (2-tailed) less than 0.05 were considered statistically significant. Active RA (n = 41)

Inactive RA (n = 43)

Age, years

61 (29-84)

59 (24-86)

0.06

Female, no. (%)

35 (85)

30 (70)

0.09

Disease duration, years

7 (0-50)

8 (2-32)

0.79

DAS28

5.4 (0.9-8.1)

1.7 (0.1-3.6)