You are on page 1of 10

Molecular Immunology 44 (2007) 2332

Review

Bacterial complement evasion


Suzan H.M. Rooijakkers, Jos A.G. van Strijp
Experimental Microbiology, UMC Utrecht G04-614, Heidelberglaan 100, 3584 CX Utrecht, The Netherlands Received 1 June 2006; received in revised form 22 June 2006; accepted 27 June 2006 Available online 27 July 2006

Abstract The human complement system is elemental to recognize bacteria, opsonize them for handling by phagocytes, or kill them by direct lysis. However, successful bacterial pathogens have in turn evolved ingenious strategies to overcome this part of the immune system. In this review we discuss the different stages of complement activation sequentially and illustrate the immune evasion strategies that various bacteria have developed to evade each subsequent step. The focus is on bacterial proteins, either surface-bound or excreted, that block complement activation. The underlying molecular mechanism of action and the possible role in pathophysiology of bacterial infections are discussed. 2006 Elsevier Ltd. All rights reserved.
Keywords: Complement; Bacteria; Evasion; Proteins; Inhibitors

1. Introduction The complement system is an essential and effective part of the innate immune system. It can rapidly recognize and opsonize bacteria for phagocytosis by professional phagocytes or kill them directly by membrane perturbations. Therefore it is to no surprise that bacteria have evolved a whole array of highly specic complement-modulating strategies. In this way bacteria can either stop or delay the detrimental effects of an innate immune attack, thereby creating a window of opportunity to divide and create a microenvironment that allows an even better survival. The human complement system consists of more than thirty proteins in plasma and on cells. The complement system is organized in three different initiation pathways that all converge at one step: the cleavage of the central complement protein C3 (Walport, 2001a,b). The three different complement pathways are represented by the classical (CP), the lectin (LP) and the alternative (AP) pathway. These pathways consist of different recognition molecules to sense a foreign substance. After recognition, these pathways use similar activation mechanisms to generate C3 convertases, the enzymes that cleave C3. The attachment of C3b to acceptor cells is necessary to initiate phagocytosis, formation of the membrane attack complex (MAC) and enhancement of humoral responses to antigens (Gasque, 2004).

In this overview we have divided the activation steps of the complement system in the following sequential phases: initial steps (2), convertases (3), C3 and its degradation products (4), the terminal pathway (5), host regulators (6) and receptors (7). In these chapters, the bacterial inhibitors that have been identied for these subsequent steps will be discussed. 2. Modulation of the initial steps The primary step in complement activation is the binding of recognition molecules to the microbe. For the classical pathway this is achieved through the binding of C1q to IgG or IgM that is specically bound to the surface of the microbe (Duncan and Winter, 1988). C1q is complexed with two serine proteases, C1r and C1s (Sim and Laich (2000)). When the globular heads of C1q bind to an activator, the associated C1r undergoes auto activation and activates C1s. Activated C1s then cleaves C4 to generate C4b and the anaphylatoxic peptide C4a. Exposed within the C4b is a highly labile internal thiol ester, which is able to react with hydroxyl groups (creating an ester bond) and amino groups (creating an amide bond) (Law and Dodds, 1997). C2 binds to surface-bound C4b and activated C1s of the nearby C1 complex cleaves C2 to release the small C2b fragment and form the classical pathway C3 convertase, C4b2a. The lectin pathway (LP) is highly analogous to the CP and its activation also results in formation of C4b2a (Matsushita and Fujita, 2001). However, the recognition molecules for the

Corresponding author. Tel.: +31 30 2506528; fax: +31 30 2505863. E-mail address: j.vanstrijp@umcutrecht.nl (J.A.G. van Strijp).

0161-5890/$ see front matter 2006 Elsevier Ltd. All rights reserved. doi:10.1016/j.molimm.2006.06.011

24

S.H.M. Rooijakkers, J.A.G. van Strijp / Molecular Immunology 44 (2007) 2332

lectin pathway recognize microbial sugars instead of immune complexes. The recognition molecules of the LP are mannanbinding lectin (MBL) and colin (L-, H- or M-colin) (Fujita, 2004). These lectins are structurally similar to C1q. MBL binds in a Ca2+ -dependent manner to the target via its C-type lectin domains, which recognize neutral sugars (preferentially mannose, N-acetylglucosamine and fucose) on the surfaces of a range of microorganisms, such as Neisseria and Leishmania species (Holmskov et al., 2003). l-Ficolin binds GlcNAc and peptidoglycan. In circulation, both MBL and colins are associated with several proteases called MBL-associated serine proteases (MASPs) (Matsushita et al., 2000). MASP-1, MASP-2 and MASP-3 are structurally similar to C1r and C1s. However, only MASP-2 is known to cleave complement components C4 and C2 and thereby generate C4b2a. The alternative pathway (AP) mainly functions as an amplication loop of the CP and LP after surface-bound C3b is created. However, it can also be spontaneously activated by hydrolysis of the internal thioester bond in C3 (0.005%/min) (Sahu and Lambris, 2001), forming the so-called C3H2 O. Just like particlebound C3b, this C3H2 O can participate in the formation of the AP C3 convertase. Both surface-bound C3b and C3H2 O can bind factor B. The resulting complexes are recognized by factor D (Xu et al., 2001). Unlike other complement serine proteases that are present as zymogens in the serum, factor D circulates in its active form. Factor D cleaves factor B to release Ba and yield the activated C3H2 O Bb complex, an unstable uid-phase C3 convertase, or a surface-bound C3bBb complex. Bacteria have evolved several modulators of these initial steps in complement activation. So far no inhibitors of the lectin pathway components have been described but, since this pathway has been discovered only recently, we envision that bacterial modulators will emerge in the coming years. Although many biological roles have been attributed to Staphylococcal protein A (SpA), its capacity to bind the Fc part of IgG is still the most established one (Silverman et al., 2005). SpA is a type I membrane protein that is bound to the cell wall of Staphylococcus aureus via its C-terminal cell-wallbinding region X. In the N-terminal half of the protein are its IgG-binding domains E, D, A, B, and C. Through binding of IgG, protein A blocks Fc-receptor mediated phagocytosis but is also a highly efcient complement activation modulator by interfering with binding of C1q (Verhoef et al., 2004; Forsgren et al., 1966; Goward et al., 1993). SpA is surface-bound, but can also be released in the surrounding environment during growth of the peptidoglycan layer to which it is attached. A second IgG-binding protein has been reported in S. aureus called Sbi (Zhang et al., 1998). The protein consists of 436 amino acids and exhibits an immunoglobulin-binding specicity similar to protein A. Its role in complement modulation remains to be established. In other bacteria, proteins with similar functions have been described. Protein G, a bacterial cell wall protein with comparable afnity for IgG, was isolated from group G streptococci. With a molecular weight of 30 kDa, Protein G was found to bind all human IgG subclasses but also rabbit, mouse, and goat IgG. On the IgG molecule, the Fc part appears mainly responsible for the

interaction with protein G, although a low degree interaction was also recorded for Fab fragments. IgM, IgA, and IgD, however, showed no binding to protein G (Bjorck and Kronvall, 1984). By the same group, Protein L was isolated from the surface of the anaerobic bacterium Peptostreptococcus magnus (Bjorck, 1988). Although this protein binds both IgM and IgG, it seems to have no afnity for the Fc-part of immunoglobulins but rather for Fab parts. A role in complement inhibition could not be demonstrated and its role is more likely as a B-cell superantigen (Genovese et al., 2003). The majority of group A streptococci (GAS) express, next to the well known M proteins, structurally similar M-related proteins, Mrp and Enn, which have been described as IgGand IgA-binding proteins. Subsequent analysis of phagocytosis by ow cytometry indicates that, if present, both mrp and emm gene products contribute to phagocytosis resistance by decreasing bacterial binding to granulocytes (Podbielski et al., 1996). GAS also produce two immunoglobulin-degrading enzymes: the streptococcal cysteine proteinases, IdeS and SpeB, that both cleave IgG specically in the hinge region, and thus removing the entire Fc region from IgG molecules that are attached to the bacterium (Von Pawel-Rammingen and Bjorck, 2003). Similarly, staphylococci hinder IgG-mediated effector functions by the excretion of Staphylokinase (SAK). S. aureus expresses several plasminogen (PLG)-binding receptors at their surface and this surface-bound PLG can be activated into plasmin (PL) by SAK. Surface-bound PL has the ability to cleave both IgG and C3b. Recently we showed that PL, formed by the conversion of PLG by SAK at physiological concentrations, leads to opsonin removal. PL cleaves human IgG from the bacterial cell wall leading to impaired phagocytosis by human neutrophils (Rooijakkers et al., 2005a). PL cleaves IgG at position Lys 222, and thus removes the entire Fc fragment, including the glycosylation site (Asn 297) necessary for recognition by C1q thereby inhibiting the activation of the classical pathway of complement. A similar strategy is employed by a protease from Porphyromonas gingivalis, a pathogen in human periodontitis. The prtH gene encodes a 97-kDa active protease, which degrades C3 and IgG. An allelic exchange mutant of P. gingivalis, in which the prtH gene was inactivated, is less virulent in a mouse model of bacterial invasiveness. Also, in comparison with its parent strain, the mutant strain is less able to degrade C3 and accumulates signicantly greater numbers of molecules of C3b and iC3b on the bacterial surface during complement activation, resulting in increased phagocytosis by human neutrophils as compared to the wild type suggesting a function of the prtH gene product may be important in evasion of host defense mechanisms (Schenkein et al., 1995). Also, bacteria can target C1 itself. The sh pathogen Aeromonas salmonicida, encodes a 40 kDa C1q-binding outer membrane protein. This 40 kDa porin binds C1q in an antibody independent process, and its in vivo role in serum resistance was established. The 40 kDa porin gene and/or protein was present in all the A. salmonicida typical or atypical strains tested (Merino et al., 2005). Two excreted enzymes from

S.H.M. Rooijakkers, J.A.G. van Strijp / Molecular Immunology 44 (2007) 2332

25

Pseudomonas elastase (PaE) and alkaline protease (PaAP), when incubated with highly puried C1q (05 h, 37 C) reveal preferential sensitivity of the 28-kDa A-chain and 24-kDa C-chain, of the C1q molecule, with PaAP being more potent than PaE (Hong and Ghebrehiwet, 1992). 3. Modulation of convertases The C3 convertases are bimolecular complexes that activate C3 (Xu et al., 2001). Two different C3 convertases exist: C4b2a and C3bBb representing the C3 convertases of the CP/LP and AP, respectively. C4b2a and C3bBb are structurally and functionally similar (Sim and Laich, 2000). C4b and C3b are derived from common ancestors and both molecules covalently attach to microbial surfaces upon activation of its internal thioester. Also, fB and C2 have a common precursor gene, share the same domain organization and as part of the C3 convertases act as similar proteases. Although C3 convertase activity resides in one molecule (C2a or Bb), the capacity to cleave C3 is acquired only through complex formation. Amplication of convertases is regulated in several ways: rst of all, the complexes itself are instable, C2a and Bb fall off after a few minutes (Ponnuraj et al., 2004). Secondly, serum and cell-bound regulators are known to cause dissociation of C3 convertases on self surfaces (Kirkitadze and Barlow, 2001). On microorganisms, decay of C3bBb can be delayed by binding of the glycoprotein properdin. Dissociated C2a and Bb are inactive and cannot re-associate with C4b and C3b to form new convertases. The covalent binding of another C3b molecule to C4b or C3b within a C3 convertase leads to the formation of a C5 convertase, C4b2a3b or C3bBb3b (Lambris, 1988). Staphylococcal complement inhibitor (SCIN) is a 10 kDa, excreted protein that blocks all complement pathways: the lectin, classical and alternative pathway. SCIN efciently prevents phagocytosis and killing of staphylococci and C5a production. We found SCIN to specically act on surface-bound C3 convertases, which has two major consequences. First SCIN stabilizes both C3bBb as well as C4b2a at the surface of the bacterium. Since the convertases are normally instable, the stabilization of SCIN prevents generation of additional convertases. Secondly and surprisingly, together with the stabilization, the binding of SCIN to C3bBb and C4b2a impairs the enzymatic activity of the convertases. This contrasts the action of properdin on convertases. SCIN binds activator-bound C3bBb, but not C3bB or C3b. This and the fact that SCIN inactivates the convertase, suggests that SCIN binds the active pocket of Bb. However, earlier studies have clearly indicated that the conformation of Bb induced by its cofactor C3b, is crucial for displaying activity of the protease subunit Bb. So, a conformational change within C3bBb by SCIN cannot be excluded (Rooijakkers et al., 2005b). SCIN is highly human-specic. It does not inhibit complement activation in all tested animals thus far including mice and rats. Many microbial modulators act indirectly on the convertases since they attract human decay-accelerating proteins and thereby destabilize the convertases. These will be discussed below under: Interactions with host regulators.

4. Modulation of C3 and its split products C3 is the most abundant complement protein in serum (1.2 mg/ml) and is comprised of an and chain (110 and 75 kDa, respectively) that are connected covalently by a single disulde bond and associated by non-covalent forces (Janssen et al., 2005). One of the most intriguing features of C3 is its ability to attach covalently to acceptor molecules on cells surfaces. This property is derived from the presence of an intramolecular thioester bond within the C3d region. The thioester bond is protected within a hydrophobic pocket and is exposed only in the C3b fragment upon cleavage of C3 by C3 convertases (Law and Dodds, 1997). Once C3 is cleaved to C3b, the transiently exposed thioester bond in C3b participates in a transacylation reaction with nucleophilic groups present on cell surfaces. In many biological systems the majority of C3b is linked via an ester bond indicating a strong preference for the hydroxylated targets. Proteolytic activation of native C3 by the C3 convertases leads to cleavage and generation of C3b (176 kDa) and C3a (9 kDa). Activation of C3 results in the release of the small chemo-attractant molecule C3a on one hand and in the deposition of C3b molecules on the microbial surface on the other hand. C3b deposition is crucial for eradication of microbes since C3b and the C3b degradation product iC3b mark the microbe for efcient uptake by phagocytes. iC3b can be further degraded by factors H and I to C3dg and nally to C3d, still attached to the microbe (Lambris, 1988). Staphylokinase (SAK) targets PLG to the staphylococcal surface and activates it into PL. PL cleaves human IgG as well as human C3b and iC3b from the bacterial cell wall leading to impaired phagocytosis by human neutrophils. It cleaves C3b in both the - and the -chain. The decrease of C3b molecules will indirectly diminish C3 convertases as well as C5 convertases (Rooijakkers et al., 2005a, c). Also the PrtH-encoded 97 kDa proteases from Porphyromonas gingivalis, degrades C3 (Schenkein et al., 1995). Further, the two Pseudomonas proteases that were described above as C1 degraders also target C3. C3, after incubation with PaE and PaAP, was converted from 190-kDa to a 120-kDa fragment. The 120-kDa piece yielded three distinct bands on SDS-PAGE: an intact 75-kDa beta-chain and two alpha-chain pieces of approximately 41and 26-kDa. NH2-terminal end sequence analysis localized the 26-kDa fragment within the cysteine-rich 41-kDa, COOHterminal piece. The NH2-terminal end of the alpha-chain is completely degraded into small fragments (Hong and Ghebrehiwet, 1992). The extracellular brinogen binding molecule (Efb) is a 15.6 kDa excreted molecule that was described earlier to bind brinogen. The group of Brown found that Efb binds the C3d region of C3 (Lee et al., 2004a). Efb blocks classical pathway dependent opsonization and subsequent phagocytosis. Although the present data on Efb do not demonstrate Efb binding to bacterium-bound C3d, a role for Efb in modulation of C3dmediated recognition by CR2 on B-cells cannot be excluded. The C3b binding site of Efb is distinct from its brinogen-binding site, in fact, Efb can bind both molecules simultaneously (Lee et al., 2004b).

26

S.H.M. Rooijakkers, J.A.G. van Strijp / Molecular Immunology 44 (2007) 2332

5. Modulation of the terminal pathway The terminal complement pathway is the nal cytolytic step in the complement cascade. The killer molecule is the membrane attack complex (MAC), a lytic assembly of C5b, C6, C7, C8 and multiple molecules of C9 (Ramm et al., 1982). The terminal pathway starts when C5 is split into the chemoattractant C5a and C5b by the C5 convertase. C5b forms a soluble bimolecular complex with C6 and the subsequent binding of C5b6 to C7 induces it to express a metastable site through which C5b7 is inserted into target lipid bilayer membranes. Subsequent incorporation of C8 and multiple C9 molecules allows the complex to penetrate lipid bilayers creating complete transmembrane channels resulting in osmotic lysis of the cell. The MAC directly kills Gram-negative organisms that have an outer lipid membrane. However, Gram-positive bacteria resist this attack by the simple fact that their cell membrane is shielded by a thick cell wall (Joiner et al., 1983). With the cleavage of C5, the important chemo-attractant C5a is released. Together with C3a and bacterial formylated peptides, C5a attracts phagocytes to the site of infection (Ramm et al., 1982). In S. aureus, 11 different staphylococcal superantigen-like (SSLs) proteins were identied. Although these molecules are closely related to the superantigens, they have different biological functions. SSL7 (23 kDa) is a complement inhibitor that specically binds to C5 (KD = 18 nM) and thereby prevents complement-mediated lysis of erythrocytes or Escherichia coli cells (Langley et al., 2005). SSL7 is not specic for the human host, since it also binds primate, sheep, pig and rabbit C5. Although the SSL7 binding site on C5 remains to be elucidated, the authors propose the C5 cleavage site at the C5 -chain as a logical site. In this case, SSL7 would prevent cleavage of C5 into both C5b and C5a. From a bacterial point of view, inhibition of C5a production is probably a more important complement evasion strategy since S. aureus is resistant to C5b-9 cytolysis. Next to its role in complement evasion, SSL7 also binds monomeric forms of human IgA1/IgA2 and the secretory form of IgA. SSL7 prevented binding of serum IgA to Fc R1 (CD89) on myeloid cells. All SSL genes were found to be located on a 19-kb genetic cluster of the S. aureus pathogenicity island SaPIn2. Streptococcal inhibitor of complement (SIC) is a 31 kDa excreted protein that fulls many different roles in immune evasion by Group A Streptococci (GAS). SIC, exclusively found in the highly virulent M1 type, was initially identied as a terminal complement pathway inhibitor since it binds the soluble C5b-7 complex and thereby prevents its insertion into cell membranes (Akesson et al., 1996; Fernie-King et al., 2001). SIC functions similarly to the human MAC regulators, clusterin and S-protein. Since streptococci are resistant to complementmediated cytolysis, complement inhibition was suggested not to be the sole function of SIC. Next to its role in complement inhibition, SIC also counteracts the antibacterial actions of secretory leukocyte proteinase inhibitor (sLPI) and lysozyme (Fernie-King et al., 2002), inactivates human neutrophil alphadefensin (HNP-1) and LL-37 (Frick et al., 2003) and alters

cellular processes by binding the intracellular proteins Ezrin and Moesin in epithelial cells and neutrophils (Hoe et al., 2002). A very unique property of SIC is that it is extraordinarily polymorphic (Stockbauer et al., 1998). A population-based surveillance of GAS infections revealed that epidemic waves were composed of strains expressing a heterogeneous array of SIC variants (Hoe et al., 1999). From 892 different isolates, a total of 162 different sic alleles were identied. Two variants of SIC have been described in GAS. In serotype M57, the gene for closely related to SIC (CRS) was found to be located outside the mga regulon. CRS shares many characteristics of SIC, including its ability to bind C6 and C7. The protein that is distantly related to SIC (DRS) has a limited sequence similarity with the C-proximal half and its biological function is unknown. DRS is present among M12 and M55 serotypes (Hartas and Sriprakash, 1999). The plasmid-encoded outer membrane protein TraT from E. coli K12 strongly prevents the terminal pathway. TraT inhibited serum haemolytic activity at the step downstream of C5 activation and upstream of C7. TraT probably inhibits the assembly of C5b6 or causes structural changes that inactivate C5b6 (Pramoonjago et al., 1992). Borrelia burgdorferi confers resistance to the terminal pathway by a 80 kDa surface protein that shares both antigenic and functional similarities with human CD59, a natural membranebound inhibitor of MAC. Both CD59 and Borrelial CD59-like inhibit cell lysis by preventing the polymerisation of C9 and the formation of MAC. Despite its similarities, the CD59-like molecule exhibits a number of structural and functional differences from human CD59; CD59-like is much larger than CD59, indicating that it is not an acquired regulator. Furthermore, CD59-like interacts with native C8 and C9 while human CD59 binds the same molecules only in context of the assembling MAC. Inhibition of MAC formation by a membrane bound CD59-like represents an important survival mechanism in B. burgdorferi (Pausa et al., 2003). Both Group A (GAS) and group B streptococci (GBS) encode a cell wall anchored C5a peptidase (scpA and scpB) (Wexler and Cleary, 1985; Chmouryguina et al., 1996; Navarre and Schneewind, 1999). This peptidase is an established virulence factor as determined in animal models for scpA (Ji et al., 1996). Mutant strains were better cleared than wild-type strains. Interestingly, ScpB is highly specic for human C5a (Bohnsack et al., 1993) and an identical gene in Group G streptococci has been shown to be restricted to strains that are capable of infecting humans (Cleary et al., 1991). These C5a peptidases are unable to cleave the whole C5 protein but do cleave the C5a fragment (Cleary et al., 1992). The 56-kilodalton protease (56 kDa protease) from Serratia marcescens signicantly and dose-dependently decreases the chemotactic activity of activated human serum. Furthermore, treatment of human recombinant C5a with 56 K protease at a dose of 1 g/ml resulted in a complete loss of chemotactic activity. In mice, the magnitude of inltration of neutrophils into the peritoneal cavity was much lower than that caused by a low protease producing strain (Oda et al., 1990).

S.H.M. Rooijakkers, J.A.G. van Strijp / Molecular Immunology 44 (2007) 2332

27

6. Interactions with host regulators Since complement activation is potentially harmful to host tissues, it is tightly regulated. At virtually every step of the complement cascade, uid-phase or cell bound regulators control complement activation. The rst proteolytic step in the classical and lectin pathway is controlled by the serine protease inhibitor C1 Inhibitor (C1-INH) (Sim and Laich, 2000). Further downstream, C3 convertases are regulated by the so-called regulators of complement activation (RCA) including the plasma proteins C4-binding protein (C4BP), factor H (FH), FH-like protein (FHL-1) and the membrane proteins CR1, CR2, CD46 (MCP) and CD55 (DAF) (Kirkitadze and Barlow, 2001). RCA proteins accelerate the decay of C3 convertases and furthermore act as cofactors in the proteolytic degradation of C4b and/or C3b by the soluble protease factor I (fI). Only one positive regulator of complement is known; properdin is a multimeric protein that enhances the stability of the alternative pathway C3 convertase on microbial surfaces (Hourcade, 2006). The terminal complement pathway is regulated by CD59, vitronectin and clusterin that all prevent insertion of C5b-7 into lipid membranes (Davies and Lachmann, 1993). Secreted protease of C1 esterase inhibitor from Enterohemorrhagic E. coli O157:H7 (StcE) is a 98 kDa zinc metalloprotease that is encoded by the large virulence plasmid pO157. Lathem et al. initially showed that StcE specically cleaves C1-INH from its full length of 105 kDa into 6065 kDa fragments (Lathem et al., 2002). Later on the authors described the functional consequences of StcE interacting with C1-INH and indicated that instead of inactivating C1-INH, StcE enhanced the ability of the serpin to down-regulate the classical complement pathway (Lathem et al., 2004). StcE was found to trap C1-INH to the cellular surface by binding both to the surface and the serpin at the same time. This increases the local concentration of C1-INH and strengthens its inhibitory capacity. The proteolysis of C1-INH by StcE was not necessary to increase its inhibitory actions. StcE was found to interact with C1-INH at its glycosylated amino terminal domain, leaving the serpin domain free to inactivate its target serine proteases. Next to inhibition of classical complement-dependent lysis of sheep erythrocytes, StcE interacting with C1-INH also offers increased serum resistance to E. coli K-12. In summary, by recruiting C1-INH to cell surfaces, StcE may protect E. coli O157:H7 from complement mediated lysis and inammatory events. Serratia marcescens also produces a protease that cleaves C1INH (Molla et al., 1989). The 56-kDa protease cleaves C1-INH, but also alpha 2-antiplasmin, and antithrombin III into molecular weights of approximately 810 kDa. The 56-kDa protease also inactivated serum complement within 26 h. A number of bacterial complement regulators evade complement attack by sequestration of human regulators to the bacterial surface. Well-described examples are bacterial surface proteins that bind human C4BP and FH/FHL-1. These regulators are captured in such a way that they are still able to interact with C3 convertases and function as cofactors in factor I cleavage of C3b/C4b.

By binding C4BP (570 kDa), bacteria disturb activation of the classical/lectin pathway. Group A Streptococci bind to C4BP via the M protein family members Arp and Sir (Jarva et al., 2003; Thern et al., 1995). Neisseria gonorrhoeae captures C4BP via both its outer membrane porin (Por) and via type IV pili. Both isoforms of Por, Por1A and Por1B, were shown to bind C4BP (Ram et al., 2001). Moraxella catarrhalis interacts with C4BP via Ubiquitous surface protein A1 and A2 (UspA1 and UspA2) (Nordstrom et al., 2004). Both UspA1 (88 kDa) and UspA2 (62 kDa) bind C4BP in a dose-dependent manner and binding studies to single subunits of C4BP revealed KD values of 13 and 1.1 M for UspA150770 and UspA230539 , respectively. The binding of Outer membrane protein A (OmpA, 35 kDa) of E. coli K1 to human C4BP protects this bacterium from C3b deposition and activation of downstream complement proteins (Wooster et al., 2006). Finally, in clinical isolates of Bordetella pertussis (Berggard et al., 1997) Filamentous Hemagglutinin (FHA) was identied as a C4BP-binding molecule. In contrast to the other bacterial C4BP-molecules, FHA could not protect Bordetella from complement-mediated lysis via C4BP. The binding sites in C4BP were identied for these molecules. C4BP is composed of seven identical alpha-chains (70 kDa) and one beta-chain (45 kDa) (Jenkins et al., 2006). The alpha and beta chains consist of repeating domains of 60aa complement control protein (CCP) domains. Arp and Sir, but also Neisserial type IV pili, bind the rst two CCP modules of the alpha-chain (Jenkins et al., 2006; Ram et al., 2001). Por binding sites in C4BP are exclusively in the CCP1 region of the alpha chain. Both UspA1 and UspA2 bound the CCP2, CCP5 and CCP7 domains of the alpha chain. OmpA binds the CCP3 domain of the alpha-chain (Prasadarao et al., 2002). Many bacteria also capture FH and/or FHL-1 to their surface resulting in AP C3 convertase decay and inactivation of C3b by factor I. FH (150 kDa) is composed of 20 short consensus repeats (SCR). FHL-1 (42 kDa) consists of the rst 7 SCR of FH in combination with four additional amino acids at the C terminus. Both FH (400 g/ml) and FHL-1 (1050 g/ml) regulate the AP C3 convertase, C3bBb, by accelerating its decay and displaying cofactor activity. Next to their interaction with C4BP, gonococcal Por molecules and streptococcal M1 protein, also bind to FH (Ram et al., 1999; Kihlberg et al., 1999). For Por1A, different loops were shown to be involved in binding of C4BP or FH. Class 3 Por molecules in N. meningitidis have also been suggested as a receptor for FH. Factor H binding inhibitor of complement (Hic) of Streptococcus pneumoniae and Beta protein of Group B Streptococci are cell-anchored proteins that bind to FH (Janulczyk et al., 2000). Hic and beta protein are structurally related and were found to bind similar regions in FH (SCR811 and SCR1214) (Jarva et al., 2004). The plasmid-encoded outer membrane protein YadA mediates serum resistance in Yersinia enterolitica (China et al., 1993). The reduction of C3b deposition in YadA+ strains was ascribed to the specic binding of YadA to FH. Omp100 belongs to a family of six major outer membrane proteins of Actinobacillus actinomycetemcomitans, a

28

S.H.M. Rooijakkers, J.A.G. van Strijp / Molecular Immunology 44 (2007) 2332

pathogenic bacterium involved in periodontitis. Omp100 is a versatile virulence factor involved in bacterial adhesion, invasion and serum resistance by trapping factor H (Asakawa et al., 2003). Fba is a surface protein of GAS that binds both FH and FHL-1 (Pandiripally et al., 2003). The binding site was localized in the short consensus repeat 7 (SCR7), a domain common to both regulators. Since FHL-1 also functions in cell adhesion (Zipfel and Skerka, 1999), Pandiripally et al found that Fba mediated binding of FHL-1 also promotes entry of streptococci into human epithelial cells. Borrelia burgdorferi produces several FH-binding proteins in order to resist complement (Kraiczy et al., 2003). Complement regulator-acquiring surface protein 1 (CRASP-1) is the dominant FH and FHL-1 binding protein of B. burgdorferi. Also here, the main binding site was localized in the SCR7 of FH and FHL-1. The Erp (OspE-F related lipoprotein) family members are encoded on members of the 32 kb circular plasmid-like prophage family. Many Erp proteins serve as receptors for FH of numerous vertebrate hosts (Miller and Stevenson, 2006; Hovis et al., 2006). Streptococcal pyrogenic exotoxin B (SPE B) is an excreted cysteine protease, well known as an important virulence factor in GAS. Next to the described digestion of numerous host proteins like immunoglobulin, kininogen, bronectin, Tsao et al. recently showed that SPE B also degrades serum properdin, inhibiting complement activation via the alternative pathway (Tsao et al., 2006). GAS opsonized with SPE B-treated serum was more resistant to neutrophil killing. 7. Modulation of complement receptors Complement receptors (CR) on leukocytes form an important and integral part of the complement system. CR1 (CD35) binds the opsonin fragments C3b and C4b and promotes phagocytosis and clearance of antigen-antibody complexes (Gasque, 2004). A receptor for C1q also promotes immune complex binding to phagocytes. CR2 (CD21) is part of the B cell receptor complex; binding of antigen-complement complexes to CR2 increases the sensitivity of the B cell to antigen by up to a thousand fold. CR3 (CD11b/CD18, MAC-1) and CR4 (CD11c/CD18, p150.95) are integrin molecules that allow monocytes, macrophages, neutrophils, and dendritic cells to adhere to blood vessel walls and move into the tissues at the site of inammation. Next to that, especially CR3 is probably the primary responsible receptor for phagocytosis of opsonized bacteria by neutrophils since CR3 recognizes iC3b. The C5a receptor (CD88) is responsible for chemotaxis and primes neutrophils and macrophages to phagocytose complement-coated antigen even in the absence of IgG. At high concentrations (over 106 M), C5a, upon interaction with the C5aR, can stimulate neutrophils to a metabolic burst. M5 protein of GAS interferes with CR3 (CD11b/CD18)dependent association between GAS and neutrophils, and thereby blocks subsequent ingestion of the bacteria. Isolated human neutrophils killed an M-negative GAS mutant (DeltaM5), but not the wild-type parent strain (M5). Differ-

ent Abs against CR3 blocked adhesion and killing of DeltaM5 bacteria, whereas the blocking of CR1 and CR4 had no effect (Weineisen et al., 2004). Group A Streptococci also secrete a protein with homology to the alpha-subunit of CR3. The GAS Mac-1-like protein (Mac) was secreted by most pathogenic strains, produced in log-phase and controlled by a two-component gene regulatory system, which also regulates transcription of other GAS virulence factors. Patients with GAS infection had titers of antibody specic to Mac that correlated with the course of disease, demonstrating that Mac was produced in vivo. Although Mac has homology to CR3, its action is not related to complement modulation, Mac binds CD16 (Fc RIIIB). This binding inhibits phagocytosis and production of reactive oxygen species, resulting in decreased pathogen killing (Lei et al., 2001). Interestingly, Mac is identical to IdeS, the IgG degrading enzyme of GAS (Von Pawel-Rammingen and Bjorck, 2003). The very early signs of bacterial invasion, C5a and formylated peptides (e.g. fMet-Leu-Phe), are recognized by the innate immune system through two related receptors on neutrophils, the C5aR and the formyl peptide receptor (FPR). The excreted Chemotaxis inhibitory protein of S. aureus (CHIPS) is a 14.1kDa protein that reduces the neutrophil recruitment toward C5a in a mouse peritonitis model, even though its activity is much more potent on human than on mouse cells. CHIPS binds specifically to the C5aR and FPR and blocks C5a- and fMLP-induced calcium mobilization. The apparent K(d) values of CHIPS for the C5aR is 1.1 nM, similar to that of C5a itself (de Haas et al., 2004). Using monoclonal anti-CHIPS Abs, peptides and site directed mutagenesis it was demonstrated that the rst and third amino acid, both a phenylalanine, are essential for CHIPS blocking the fMLP-induced activation of phagocytes, while the C5aR binding site of CHIPS is located at the C-terminal end (Haas et al., 2004). CHIPS does not affect activation of the C5aR by a peptide mimic of the C5a C-terminus. Moreover, CHIPS was found to bind only the C5aR N terminus. Deletion and mutation experiments within this C5aR N-terminal expression system revealed that the binding site of CHIPS is contained in a short stretch of nine amino acids (amino acids 1018), of which the aspartic acid residues at positions 10, 15, and 18 plus the glycine at position 12 are crucial. Binding studies with C5aR/C3aR- and C5aR/IL8RA-chimeras conrmed that CHIPS binds only to the C5aR N terminus without involvement of its extracellular loops (Postma et al., 2004, 2005). CHIPS fragment consisting of residues 31121 (CHIPS31121) has the same activity in blocking the C5aR compared to full-length CHIPS, but completely lacks FPR antagonism. CHIPS31121 has a compact fold comprising an alpha-helix (residues 3851) packed onto a four-stranded anti-parallel beta-sheet. Comparison of CHIPS31121 with known structures reveals striking homology with the C-terminal domain of staphylococcal superantigen-like proteins (SSLs) 5 and 7, and the staphyloccocal and streptococcal superantigens TSST-1 and SPE-C. Also, the recently reported structures of several domains of the staphylococcal extracellular adherence protein (EAP) show a high degree of structural similarity with CHIPS (Haas et al., 2005).

S.H.M. Rooijakkers, J.A.G. van Strijp / Molecular Immunology 44 (2007) 2332

29

Gingipains from Porphyromonas gingivalis have been implicated as C3 degraders. Lys-gingipain, but not Arg-gingipain, is also active against the C5aR. It cleaves C5aR on human neutrophils. The N-terminal region of C5aR (residues 929, PDYGHYDDKDTLDLNTPVDKT) was readily cleaved by a mixture of proteases from P. gingivalis. Next to Lys-gingipain there appear to be additional proteinase(s) in the vesicles that attack the cell surface molecule C5aR (Jagels et al., 1996). 8. Concluding remarks In this overview we have encountered a vast amount of different proteins, partially overlapping in function and sequence or structure, that all have the ability to divert the complement system. Now we will discuss some general aspects of these proteins; arrangement in the genome, regulation of expression, presumed redundancy, species specicity and the potential use as therapeutics in inammatory disorders. Both in S. aureus and Group A Streptococci, we encounter genomic regions with multiple virulence factors specialized in modulation of complement. In S. aureus, the genes for three complement modulators SCIN, CHIPS and SAK cluster on the conserved 3 end of beta-hemolysin (hlb)-converting bacteriophages, representing an immune evasion cluster (IEC) with enterotoxin A. A total of 90% of S. aureus strains carry this IEC in seven IEC variants. This IEC is easily transferred among S. aureus strains by a diverse group of beta-hemolysin-converting bacteriophages. (van Wamel et al., 2006). The M family of streptococcal surface proteins is composed of the related Emm (class I and II), Mrp (FcrA), and Enn proteins. The so-called Mga regulons contain M-family proteins mrp, emm, and enn followed by sic and scpA gene. The exact order and content of the whole operon can vary but this is the basic architecture (Navarre and Schneewind, 1999). Other complement modulators reviewed here, TraT and StcE, from E. coli, YadA from Yersinia and the Erp family members of Borrelia are encoded on plasmids or on plasmid-like prophages (Zhang and Marconi, 2005). All this indicates that the clustered genomic organization of these factors cannot be a coincidence; probably they are (remnants of) mobile elements making bacteria virulent by adding a complete set of immune evasion genes. The other option is that their genomic co-localization is a result of a common regulatory system that guides the expression timing of these genes. Indeed the expression of these genes is under strict control as can be concluded from the following examples. The streptococcal terminal pathway modulator SIC, like the other genes on the mga regulon, is excreted in high amounts during early and late logarithmic growth phase during in vitro culture. In vivo evidence exists that SIC has a role in the early stages of infection. There is no expression during stationary growth phase (Frick et al., 2003). Similarly, we showed that the staphylococcal complement modulators SCIN and CHIPS are explicitly expressed during early growth stage of in vitro bacterial growth implicated with in vivo colonization phase, while the other complement modulator SAK is expressed much later (Rooijakkers et al., 2006).

The question is often posed whether it is costly redundancy when bacteria make so many immune evasion molecules that attack multiple steps of cascades even within one species. In vitro it is often sufcient to block one step of a certain biological cascade to block the whole outcome of the cascade. In simple animal models it is often the same: blocking one step using a natural inhibitor or an antibody can be sufcient to block the biological effects associated with that cascade. In Staphylococci and Streptococci combined there are more than a dozen complement inhibitors described to date. Many people try to prove that their molecule is the most important virulence factor since that is the best blocker at the most crucial step of the complement cascade. However, there is a lesson to be learned here. Viral immune evasion has already shown us years ago that the best way to block a cascade is to inhibit every subsequent step in that cascade by different molecules. Antigen processing is attacked in that very way by herpesviruses that evade adaptive immunity (Ploegh, 1998; Novak and Peng, 2005; Wiertz et al., 1997). We will identify far more evasion molecules in the years to come and it will become clear that the best evasion strategy takes many attacks at many points, even in a cascade. The species specicity of many of these inhibitors is striking. For various bacteria it is clear that neither the pathogen nor the encoded evasion molecules is host specic and these molecules are easily studied in mouse models. In contrast, there are also many evasion molecules that are highly specic for one host and this is also true for complement evasion. Although the complement system is evolutionary extremely conserved still human specicity was found in complement evasion molecules from human pathogens: CHIPS, C5a-peptidase, SAK, SCIN and more. Since it is extremely difcult to study these evasion molecules in animal models it is hard to predict the role of this species specicity. Again in viral immune evasion strategies it is clear that these molecules contribute to a great extent to the host specicity of the pathogen itself (Dobbelstein, 2003; Ahn et al., 1996). From a therapeutically point of view, all proteins in this overview have a high potential in anti-inammatory therapy. Per denition, molecules that help to evade acute innate immune mechanism are anti-inammatory compounds. Their high specicity correlates with their non-toxic nature. Especially for the soluble molecules, it is tempting to speculate that they could be used as injectables in severe acute inammatory disorders. Pre-existing antibodies against all these proteins will complicate this approach but the targets that these molecules identify, to the amino acid level and below, will prove to be crucial targets in anti-inammatory therapy. Taken together, comprehending complement evasion is a crucial aspect of understanding bacterial pathogenesis. The more proteins along with their exact mechanism of action are discovered, the more we will understand bacteria-host interactions in general. Acknowledgements Discussed work that originated from the authors laboratory was supported by grants from the Technology Foundation STW

30

S.H.M. Rooijakkers, J.A.G. van Strijp / Molecular Immunology 44 (2007) 2332 Genovese, A., Borgia, G., Bjorck, L., Petraroli, A., de Paulis, A., Piazza, M., Marone, G., 2003. Immunoglobulin superantigen protein L induces IL-4 and IL-13 secretion from human Fc epsilon RI+ cells through interaction with the kappa light chains of IgE. J. Immunol. 170, 18541861. Goward, C., Scawen, M., Murphy, J., Atkinson, T., 1993. Molecular evolution of bacterial cell-surface proteins. Trends Biochem. Sci. 18, 136140. Haas, P.J., de Haas, C.J., Kleibeuker, W., Poppelier, M.J., van Kessel, K.P., Kruijtzer, J.A., Liskamp, R.M., van Strijp, J.A., 2004. N-terminal residues of the chemotaxis inhibitory protein of Staphylococcus aureus are essential for blocking formylated peptide receptor but not C5a receptor. J. Immunol. 173, 57045711. Haas, P.J., de Haas, C.J., Poppelier, M.J., van Kessel, K.P., van Strijp, J.A., Dijkstra, K., Scheek, R.M., Fan, H., Kruijtzer, J.A., Liskamp, R.M., Kemmink, J., 2005. The structure of the C5a receptor-blocking domain of chemotaxis inhibitory protein of Staphylococcus aureus is related to a group of immune evasive molecules. J. Mol. Biol. 353, 859872. Hartas, J., Sriprakash, K.S., 1999. Streptococcus pyogenes strains containing emm12 and emm55 possess a novel gene coding for distantly related SIC protein. Microb. Pathog. 26, 2533. Hoe, N.P., Nakashima, K., Lukomski, S., Grigsby, D., Liu, M., Kordari, P., Dou, S.J., Pan, X., Vuopio-Varkila, J., Salmelinna, S., McGeer, A., Low, D.E., Schwartz, B., Schuchat, A., Naidich, S., De Lorenzo, D., Fu, Y.X., Musser, J.M., 1999. Rapid selection of complement-inhibiting protein variants in group A Streptococcus epidemic waves. Nat. Med. 5, 924929. Hoe, N.P., Ireland, R.M., DeLeo, F.R., Gowen, B.B., Dorward, D.W., Voyich, J.M., Liu, M., Burns Jr., E.H., Culnan, D.M., Bretscher, A., Musser, J.M., 2002. Insight into the molecular basis of pathogen abundance: group A Streptococcus inhibitor of complement inhibits bacterial adherence and internalization into human cells. Proc. Natl. Acad. Sci. U S A 99, 76467651. Holmskov, U., Thiel, S., Jensenius, J.C., 2003. Collectins and Ficolins: Humoral lectins of the innate immune defense. Ann. Rev. Immunol. 21, 547578. Hong, Y.Q., Ghebrehiwet, B., 1992. Effect of Pseudomonas aeruginosa elastase and alkaline protease on serumcomplement and isolated components C1q and C3. Clin. Immunol. Immunopathol. 62, 133138. Hourcade, D.E., 2006. The role of properdin in the assembly of the alternative pathway C3 convertases of complement. J. Biol. Chem. 281, 2128 2132. Hovis, K.M., Tran, E., Sundy, C.M., Buckles, E., McDowell, J.V., Marconi, R.T., 2006. Selective binding of Borrelia burgdorferi OspE paralogs to factor H and serum proteins from diverse animals: possible expansion of the role of OspE in Lyme disease pathogenesis. Infect. Immun. 74, 19671972. Jagels, M.A., Ember, J.A., Travis, J., Potempa, J., Pike, R., Hugli, T.E., 1996. Cleavage of the human C5A receptor by proteinases derived from Porphyromonas gingivalis: cleavage of leukocyte C5a receptor. Adv. Exp. Med. Biol. 389, 155164. Janssen, B.J., Huizinga, E.G., Raaijmakers, H.C., Roos, A., Daha, M.R., NilssonEkdahl, K., Nilsson, B., Gros, P., 2005. Structures of complement component C3 provide insights into the function and evolution of immunity. Nature 437, 505511. Janulczyk, R., Iannelli, F., Sjoholm, A.G., Pozzi, G., Bjorck, L., 2000. Hic, a novel surface protein of Streptococcus pneumoniae that interferes with complement function. J. Biol. Chem. 275, 3725737263. Jarva, H., Jokiranta, T.S., Wurzner, R., Meri, S., 2003. Complement resistance mechanisms of streptococci. Mol. Immunol. 40, 95107. Jarva, H., Hellwage, J., Jokiranta, T.S., Lehtinen, M.J., Zipfel, P.F., Meri, S., 2004. The group B streptococcal beta and pneumococcal Hic proteins are structurally related immune evasion molecules that bind the complement inhibitor factor H in an analogous fashion. J. Immunol. 172, 31113118. Jenkins, H.T., Mark, L., Ball, G., Persson, J., Lindahl, G., Uhrin, D., Blom, A.M., Barlow, P.N., 2006. Human C4b-binding protein, structural basis for interaction with streptococcal M protein, a major bacterial virulence factor. J. Biol. Chem. 281, 36903697. Joiner, K., Brown, E., Hammer, C., Warren, K., Frank, M., 1983. Studies on the mechanism of bacterial resistance to complement-mediated killing. III. C5b-9 deposits stably on rough and type 7 S. pneumoniae without causing bacterial killing. J. Immunol. 130, 845849.

(#UKG-6609), the Netherlands Genomics Initiative (NGI #05071-028), the European Union (#FP6-512093) and the Netherlands Organization for Scientic Research (NWO #9120.6020). References
Ahn, K., Meyer, T.H., Uebel, S., Sempe, P., Djaballah, H., Yang, Y., Peterson, P.A., Fruh, K., Tampe, R., 1996. Molecular mechanism and species specicity of TAP inhibition by herpes simplex virus ICP47. EMBO J. 15, 32473255. Akesson, P., Sjoholm, A.G., Bjorck, L., Protein, 1996. SIC, a novel extracellular protein of Streptococcus pyogenes interfering with complement function. J. Biol. Chem. 271, 10811088. Asakawa, R., Komatsuzawa, H., Kawai, T., Yamada, S., Goncalves, R.B., Izumi, S., Fujiwara, T., Nakano, Y., Suzuki, N., Uchida, Y., Ouhara, K., Shiba, H., Taubman, M.A., Kurihara, H., Sugai, M., 2003. Outer membrane protein 100, a versatile virulence factor of Actinobacillus actinomycetemcomitans. Mol. Microbiol. 50, 11251139. Berggard, K., Johnsson, E., Mooi, F.R., Lindahl, G., 1997. Bordetella pertussis binds the human complement regulator C4BP: role of lamentous hemagglutinin. Infect. Immun. 65, 36383643. Bjorck, L., Kronvall, G., 1984. Purication and some properties of streptococcal protein G, a novel IgG-binding reagent. J. Immunol. 133, 969974, Aug. Bjorck, L., 1988. Protein L, a novel bacterial cell wall protein with afnity for Ig L chains. J. Immunol. 140, 11941197. Bohnsack, J.F., Chang, J.K., Hill, H.R., 1993. Restricted ability of group B streptococcal C5a-ase to inactivate C5a prepared from different animal species. Infect. Immun. 61, 14211426. China, B., Sory, M.P., NGuyen, B.T., De Bruyere, M., Cornelis, G.R., 1993. Role of the YadA protein in prevention of opsonization of Yersinia enterocolitica by C3b molecules. Infect. Immun. 61, 31293136. Chmouryguina, I., Suvorov, A., Ferrieri, P., Cleary, P.P., 1996. Conservation of the C5a peptidase genes in group A and B streptococci. Infect. Immun. 64, 23872390. Cleary, P.P., Peterson, J., Chen, C., Nelson, C., 1991. Virulent human strains of group G streptococci express a C5a peptidase enzyme similar to that produced by group A streptococci. Infect. Immun. 59, 23052310. Cleary, P.P., Prahbu, U., Dale, J.B., Wexler, D.E., Handley, J., 1992. Streptococcal C5a peptidase is a highly specic endopeptidase. Infect. Immun. 60, 52195223. Davies, A., Lachmann, P.J., 1993. Membrane defence against complement lysis: the structure and biological properties of CD59. Immunol. Res. 12, 258275. de Haas, C.J., Veldkamp, K.E., Peschel, A., Weerkamp, F., van Wamel, W.J., Heezius, E.C., Poppelier, M.J., van Kessel, K.P., van Strijp, J.A., 2004. Chemotaxis inhibitory protein of Staphylococcus aureus, a bacterial antiinammatory agent. J. Exp. Med. 199, 687895. Dobbelstein, M., 2003. Viruses in therapyroyal road or dead end? Virus Res. 92, 219221. Duncan, A.R., Winter, G., 1988. The binding site for C1q on IgG. Nature 332, 738740. Fernie-King, B.A., Seilly, D.J., Willers, C., Wurzner, R., Davies, A., Lachmann, P.J., 2001. Streptococcal inhibitor of complement (SIC) inhibits the membrane attack complex by preventing uptake of C567 onto cell membranes. Immunology 103, 390398. Fernie-King, B.A., Seilly, D.J., Davies, A., Lachmann, P.J., 2002. Streptococcal inhibitor of complement inhibits two additional components of the mucosal innate immune system: secretory leukocyte proteinase inhibitor and lysozyme. Infect. Immun. 70, 49084916. Forsgren, A., Sj quist, J., 1966. Protein A from S. aureus. I. Pseudo-immune o reaction with human gamma-globulin. J. Immunol. 97, 822827. Frick, I.M., Akesson, P., Rasmussen, M., Schmidtchen, A., Bjorck, L., 2003. SIC, a secreted protein of Streptococcus pyogenes that inactivates antibacterial peptides. J. Biol. Chem. 278, 1656116566. Fujita, T., 2004. The lectin-complement pathway-its role in innate immunity and evolution. Immunol. Rev. 198, 185202. Gasque, P., 2004. Complement: a unique innate immune sensor for danger signals. Mol. Immunol. 41, 10891098.

S.H.M. Rooijakkers, J.A.G. van Strijp / Molecular Immunology 44 (2007) 2332 Ji, Y., McLandsborough, L., Kondagunta, A., Cleary, P.P., 1996. C5a peptidase alters clearance and trafcking of group A streptococci by infected mice. Infect. Immun. 64, 503510. Kihlberg, B.M., Collin, M., Olsen, A., Bjorck, L., 1999. Protein H, an antiphagocytic surface protein in Streptococcus pyogenes. Infect. Immun. 67, 17081714. Kirkitadze, M.D., Barlow, P.N., 2001. Structure and exibility of the multiple domain proteins that regulate complement activation. Immunol. Rev. 180, 146161. Kraiczy, P., Hellwage, J., Skerka, C., Kirschnk, M., Brade, V., Zipfel, P.F., Wallich, R., 2003. Immune evasion of Borrelia burgdorferi: mapping of a complement-inhibitor factor H-binding site of BbCRASP-3, a novel member of the Erp protein family. Eur. J. Immunol. 33, 697707. Lambris, J.D., 1988. The multifunctional role of C3, the third component of complement. Immunol. Today 9, 387393. Langley, R., Wines, B., Willoughby, N., Basu, I., Proft, T., Fraser, J.D., 2005. The staphylococcal superantigen-like protein 7 binds IgA and complement C5 and inhibits IgA-Fc alpha RI binding and serum killing of bacteria. J. Immunol. 174, 29262933. Lathem, W.W., Grys, T.E., Witowski, S.E., Torres, A.G., Kaper, J.B., Tarr, P.I., Welch, R.A., 2002. StcE, a metalloprotease secreted by Escherichia coli O157:H7, specically cleaves C1 esterase inhibitor. Mol. Microbiol. 45, 277288. Lathem, W.W., Bergsbaken, T., Welch, R.A., 2004. Potentiation of C1 esterase inhibitor by StcE, a metalloprotease secreted by Escherichia coli O157:H7. J. Exp. Med. 199, 10771087. Law, S.K., Dodds, A.W., 1997. The internal thioester and the covalent binding properties of the complement proteins C3 and C4. Protein Sci. 6, 263 274. Lee, L.Y., Hook, M., Haviland, D., Wetsel, R.A., Yonter, E.O., Syribeys, P., Vernachio, J., Brown, E.L., 2004a. Inhibition of complement activation by a secreted Staphylococcus aureus protein. J. Infect. Dis. 190, 571579. Lee, L.Y., Liang, X., Hook, M., Brown, E.L., 2004b. Identication and characterization of the C3 binding domain of the Staphylococcus aureus extracellular brinogen-binding protein (Efb). J. Biol. Chem. 279, 5071050716. Lei, B., DeLeo, F.R., Hoe, N.P., Graham, M.R., Mackie, S.M., Cole, R.L., Liu, M., Hill, H.R., Low, D.E., Federle, M.J., Scott, J.R., Musser, J.M., 2001. Evasion of human innate and acquired immunity by a bacterial homolog of CD11b that inhibits opsonophagocytosis. Nat. Med. 7, 12981305. Matsushita, M., Thiel, S., Jensenius, J.C., Terai, I., Fujita, T., 2000. Proteolytic activities of two types of mannose-binding lectin-associated serine proteases. J. Immunol. 165, 26372642. Matsushita, M., Fujita, T., 2001. Ficolins and the lectin complement pathway. Immunol. Rev. 180, 7885. Merino, S., Vilches, S., Canals, R., Ramirez, S., Tomas, J.M., 2005. A C1qbinding 40 kDa porin from Aeromonas salmonicida: cloning, sequencing, role in serum susceptibility and sh immunoprotection. Microb. Pathog. 38, 227237. Molla, A., Akaike, T., Maeda, H., 1989. Inactivation of various proteinase inhibitors and the complement system in human plasma by the 56-kilodalton proteinase from Serratia marcescens. Infect. Immun. 57, 18681871. Miller, J.C., Stevenson, B., 2006. Borrelia burgdorferi erp genes are expressed at different levels within tissues of chronically infected mammalian hosts. Int. J. Med. Microbiol. 296 (Suppl. 1), 185194. Navarre, W.W., Schneewind, O., 1999. Surface proteins of gram-positive bacteria and mechanisms of their targeting to the cell wall envelope. Microbiol. Mol. Biol. Rev. 63, 174229. Nordstrom, T., Blom, A.M., Forsgren, A., Riesbeck, K., 2004. The emerging pathogen Moraxella catarrhalis interacts with complement inhibitor C4b binding protein through ubiquitous surface proteins A1 and A2. J. Immunol. 173, 45984606. Novak, N., Peng, W.M., 2005. Dancing with the enemy: the interplay of herpes simplex virus with dendritic cells. Clin. Exp. Immunol. 142, 405 410. Oda, T., Kojima, Y., Akaike, T., Ijiri, S., Molla, A., Maeda, H., 1990. Inactivation of chemotactic activity of C5a by the serratial 56-kilodalton protease. Infect. Immun. 58, 12691272.

31

Pandiripally, V., Wei, L., Skerka, C., Zipfel, P.F., Cue, D., 2003. Recruitment of complement factor H-like protein 1 promotes intracellular invasion by group A streptococci. Infect. Immun. 71, 71197128. Pausa, M., Pellis, V., Cinco, M., Giulianini, P.G., Presani, G., Perticarari, S., Murgia, R., Tedesco, F., 2003. Serum-resistant strains of Borrelia burgdorferi evade complement-mediated killing by expressing a CD59-like complement inhibitory molecule. J. Immunol. 170, 32143222. Podbielski, A., Schnitzler, N., Beyhs, P., Boyle, M.D., 1996. M-related protein (Mrp) contributes to group A streptococcal resistance to phagocytosis by human granulocytes. Mol. Microbiol. 19, 429441. Ponnuraj, K., Xu, Y., Macon, K., Moore, D., Volanakis, J.E., Narayana, S.V., 2004. Structural analysis of engineered Bb fragment of complement factor B: insights into the activation mechanism of the alternative pathway C3convertase. Mol. Cell. 14, 1728. Postma, B., Poppelier, M.J., van Galen, J.C., Prossnitz, E.R., van Strijp, J.A., de Haas, C.J., van Kessel, K.P., 2004. Chemotaxis inhibitory protein of Staphylococcus aureus binds specically to the C5a and formylated peptide receptor. J. Immunol. 172, 69947001. Postma, B., Kleibeuker, W., Poppelier, M.J., Boonstra, M., van Kessel, K.P., van Strijp, J.A., de Haas, C.J., 2005. Residues 1018 within the C5a receptor N terminus compose a binding domain for chemotaxis inhibitory protein of Staphylococcus aureus. J. Biol. Chem. 280, 20202027. Ploegh, H.L., 1998. Viral strategies of immune evasion. Science 280, 248253. Pramoonjago, P., Kaneko, M., Kinoshita, T., Ohtsubo, E., Takeda, J., Hong, K.S., Inagi, R., Inoue, K., 1992. Role of TraT protein, an anticomplementary protein produced in Escherichia coli by R100 factor, in serum resistance. J. Immunol. 148, 827836. Prasadarao, N.V., Blom, A.M., Villoutreix, B.O., Linsangan, L.C., 2002. A novel interaction of outer membrane protein A with C4b binding protein mediates serum resistance of Escherichia coli K1. J. Immunol. 169, 63526360. Ram, S., Mackinnon, F.G., Gulati, S., McQuillen, D.P., Vogel, U., Frosch, M., Elkins, C., Guttormsen, H.K., Wetzler, L.M., Oppermann, M., Pangburn, M.K., Rice, P.A., 1999. The contrasting mechanisms of serum resistance of Neisseria gonorrhoeae and group B Neisseria meningitidis. Mol. Immunol. 36, 915928. Ram, S., Cullinane, M., Blom, A.M., Gulati, S., McQuillen, D.P., Monks, B.G., OConnell, C., Boden, R., Elkins, C., Pangburn, M.K., Dahlback, B., Rice, P.A., 2001. Binding of C4b-binding protein to porin: a molecular mechanism of serum resistance of Neisseria gonorrhoeae. J. Exp. Med. 193, 281295. Ramm, L.E., Whitlow, M.B., Mayer, M.M., 1982. Transmembrane channel formation by complement: functional analysis of the number of C5b6, C7, C8, and C9 molecules required for a single channel. Proc. Natl. Acad. Sci. U S A 79, 47514755. Rooijakkers, S.H.M., van Wamel, W.J.B., Ruyken, M., van Kessel, K.P.M., van Strijp, J.A.G., 2005a. Anti-opsonic properties of staphylokinase. Microbes. Infect. 7, 476484. Rooijakkers, S.H.M., Ruyken, M., Roos, A., Daha, M.R., Presanis, J.S., Sim, R.B., van Wamel, W.J.B., van Kessel, K.P.M., van Strijp, J.A.G., 2005b. Immune evasion by a staphylococcal complement inhibitor that acts on C3 convertases. Nat. Immunol. 6, 920927. Rooijakkers, S.H.M., van Kessel, K.P.M., van Strijp, J.A.G., 2005c. Staphylococcal innate immune evasion. Trends Microbiol. 13, 596601. Rooijakkers, S.H.M., Ruyken, M., van Roon, J., van Kessel, K.P.M., van Strijp, J.A.G., van Wamel, W.J.B., 2006. Early expression of SCIN and CHIPS drives instant immune evasion by Staphylococcus aureus. Cell Microbiol. 8, 12821293. Sahu, A., Lambris, J.D., 2001. Structure and biology of complement protein C3, a connecting link between innate and acquired immunity. Immunol. Rev. 180, 3548. Schenkein, H.A., Fletcher, H.M., Bodnar, M., Macrina, F.L., 1995. Increased opsonization of a prtH-defective mutant of Porphyromonas gingivalis W83 is caused by reduced degradation of complement-derived opsonins. J. Immunol. 154, 53315337. Silverman, G.J., Goodyear, C.S., Siegel, D.L., 2005. On the mechanism of staphylococcal protein A immunomodulation. Transfusion 45, 274280. Sim, R.B., Laich, A., 2000. Serine proteases of the complement system. Biochem. Soc. Trans. 28, 545550.

32

S.H.M. Rooijakkers, J.A.G. van Strijp / Molecular Immunology 44 (2007) 2332 Walport, M.J., 2001b. Complement. Second of two parts. N. Engl. J. Med. 344, 11401144. Weineisen, M., Sjobring, U., Fallman, M., Andersson, T., 2004. Streptococcal M5 protein prevents neutrophil phagocytosis by interfering with CD11b/CD18 receptor-mediated association and signaling. J. Immunol. 172, 379837807. Wexler, D.E., Cleary, P.P., 1985. Purication and characteristics of the streptococcal chemotactic factor inactivator. Infect. Immun. 50, 757 764. Wiertz, E.J., Mukherjee, S., Ploegh, H.L., 1997. Viruses use stealth technology to escape from the host immune system. Mol. Med. Today 3, 116123. Wooster, D.G., Maruvada, R., Blom, A.M., Prasadarao, N.V., 2006. Logarithmic phase Escherichia coli K1 efciently avoids serum killing by promoting C4bp-mediated C3b and C4b degradation. Immunology 117, 482493. Xu, Y., Narayana, S.V., Volanakis, J.E., 2001. Structural biology of the alternative pathway convertase. Immunol. Rev. 180, 123135. Zhang, L., Jacobsson, K., Vasi, J., Lindberg, M., Frykberg, L., 1998. A second IgG-binding protein in Staphylococcus aureus. Microbiology 144, 985991. Zhang, H., Marconi, R.T., 2005. Demonstration of cotranscription and 1methyl-3-nitroso-nitroguanidine induction of a 30-gene operon of Borrelia burgdorferi: evidence that the 32-kilobase circular plasmids are prophages. J. Bacteriol. 187, 79857995. Zipfel, P.F., Skerka, C., 1999. FHL-1/reconectin: a human complement and immune regulator with cell-adhesive function. Immunol. Today 20, 135140.

Stockbauer, K.E., Grigsby, D., Pan, X., Fu, Y.X., Mejia, L.M., Cravioto, A., Musser, J.M., 1998. Hypervariability generated by natural selection in an extracellular complement-inhibiting protein of serotype M1 strains of group A Streptococcus. Proc. Natl. Acad. Sci. U S A 95, 31283133. Thern, A., Stenberg, L., Dahlback, B., Lindahl, G., 1995. Ig-binding surface proteins of Streptococcus pyogenes also bind human C4b-binding protein (C4BP), a regulatory component of the complement system. J. Immunol. 154, 375386. Tsao, N., Tsai, W.H., Lin, Y.S., Chuang, W.J., Wang, C.H., Kuo, C.F., 2006. Streptococcal pyrogenic exotoxin B cleaves properdin and inhibits complement-mediated opsonophagocytosis. Biochem. Biophys. Res. Commun. 339, 779784. van Wamel, W.J.B., Rooijakkers, S.H.M., Ruyken, M., van Kessel, K.P.M., van Strijp, J.A.G., 2006. The innate immune modulators staphylococcal complement inhibitor and chemotaxis inhibitory protein of Staphylococcus aureus are located on beta-hemolysin-converting bacteriophages. J. Bacteriol. 188, 13101315. Verhoef, J., Fluit, A., Schmitz, F.J., 2004. Staphylococci and other micrococci. In: Cohen, J., Powderly, W.G. (Eds.), Infectious Disease. Elsevier, Mosby, pp. 21192132. Von Pawel-Rammingen, U., Bjorck, L., 2003. IdeS and SpeB: immunoglobulin-degrading cysteine proteinases of Streptococcus pyogenes. Curr. Opin. Microbiol. 6, 5055. Walport, M.J., 2001a. Complement. First of two parts. N. Engl. J. Med. 344, 10581066.