Rapid IgG heavy chain cleavage by the ... - Wiley Online Library

FEBS Letters 587 (2013) 1818–1822

journal homepage: www.FEBSLetters.org

Rapid IgG heavy chain cleavage by the streptococcal IgG endopeptidase IdeS is mediated by IdeS monomers and is not due to enzyme dimerization Reine Vindebro, Christian Spoerry, Ulrich von Pawel-Rammingen ⇑ Department of Molecular Biology and Umeå Centre for Microbial Research, Umeå University, 90187 Umeå, Sweden

a r t i c l e

i n f o

Article history: Received 4 April 2013 Revised 19 April 2013 Accepted 23 April 2013 Available online 10 May 2013 Edited by Peter Brzezinski Keywords: IgG Cysteine protease Monomer/dimer GAS Virulence factor Streptococcus pyogenes

a b s t r a c t Streptococcus pyogenes employs an IgG specific endopeptidase, IdeS, to counteract the effector functions of specific IgG. The physiological significant step in disarming specific IgG is the cleavage of one IgG heavy chain. So far, characterizations of IdeS enzymatic activity have employed techniques that failed to differentiate between the first and the second cleavage step. The present data demonstrate that IdeS is active as a monomer and that IdeS activity follows classical Michaelis–Menten kinetics arguing against the previously proposed formation of a functional IdeS dimer. Our results show that IdeS inactivates IgG 100-fold faster than previously reported. Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved.

1. Introduction Streptococcus pyogenes is the causative agent of a variety of diseases ranging from relatively uncomplicated conditions e.g. pharyngitis, to severe invasive conditions, e.g. necrotizing fasciitis [1]. A common mechanism to avoid recognition by specific, opsonizing antibodies is the degradation of immunoglobulins (Igs) by Ig-proteases. S. pyogenes employs a highly specific IgG degrading enzyme for this purpose; the IgG degrading enzyme of S. pyogenes, IdeS (sometimes designated Mac1; [2,3]). Immunoglobulin cleavage by streptococcal IdeS is a sequential process, in which single cleaved IgG (scIgG) is generated as an intermediate product prior to the cleavage of the second heavy chain (Fig. 2 [4,5]). Complete cleavage of both heavy chains separates the antigen binding domain of IgG from its effector functions and efficiently prevents signaling to immune effector cells [2]. However, already the proteolytic cleavage of one IgG heavy chain destroys IgG mediator functions [5] in that scIgG has lost its ability to activate complement and to interact with Fcc-receptors [5,6]. Nevertheless, earlier characterizations of IdeS activity employed techniques that either only detected cleavage of the second IgG heavy chain or that could not differentiate between first and second cleavage e.g. Surface ⇑ Corresponding author. E-mail address: [email protected] (U. von Pawel-Rammingen).

Plasmon Resonance (SPR) [7,8] size exclusion chromatography (SEC) [9] reducing SDS–PAGE [2] G-protein coupled SELDI-ToF [10] or isothermal titration calorimetry [11]. Thus, to date a reliable determination on the rate by which IdeS inactivates IgG effector functions, i.e. the rate by which IdeS generates scIgG, is lacking. Previous analyses of IdeS enzymatic activity also revealed a non-Michaelis–Menten velocity curve which has been interpreted as indication for the presence of two cooperative binding sites [9,11]. Presence of a substrate binding site, distinct from the catalytic active site, has been suggested for correct positioning of substrate IgG [9]. As an alternative to cooperative binding sites on an IdeS monomer, a symmetric dimer has been reported for crystal forms of IdeS [12] and dimerization of IdeS was proposed as an explanation for the observed non-Michaelis–Menten shaped velocity curve and as the possible mechanism how IdeS acquires its pronounced specificity [12]. Despite their essential function only few data on the enzymatic efficiency of Ig-proteases are available ([7,9] for IdeS; [13] for Gingipain K). In this study, IdeS enzymatic activity of the physiologically important first heavy chain cleavage was investigated in relation to the proposed dimerization of IdeS. The enzyme kinetic parameters for single and double chain cleavage were determined and it is demonstrated that IdeS enzyme activity follows classical Michaelis–Menten kinetics, both regarding first and the second chain cleavage, and that IdeS is active as a monomer.

0014-5793/$36.00 Ó 2013 Federation of European Biochemical Societies. Published by Elsevier B.V. All rights reserved. http://dx.doi.org/10.1016/j.febslet.2013.04.039

R. Vindebro et al. / FEBS Letters 587 (2013) 1818–1822

2. Materials and methods Purification of IdeS: IdeS was expressed and purified as previously described [8]. IgG substrate preparation: For affinity purification of polyclonal IgG, enzymatically inactive mutant protein IdeSC94S carrying a His-Trx tag was covalently coupled to CNBr-activated Sepharose (GE Biosciences) according to manufacturer’s instructions and washed with 0.1 M Tris–HCl, 500 mM NaCl at pH 6.8 and pH 8.8 [9]. The final concentration of covalently coupled protein was 6 mg per ml of Sepharose. IgG (190 mg), (Sigma) was applied to a PD10 column and eluted with assay buffer (20 mM phosphate buffer; pH 6.5; I = 154 mM adjusted with NaCl). IgG was incubated with 1 ml IdeSC94S-Sepharose at room temperature and the flowthrough was collected and incubated again with 1 ml IdeSC94S-Sepharose. The procedure was repeated twice with incubation times of 6–12 h for all incubations. The flow-through from the last incubation consists of affinity purified IgG lacking IdeS specific antibodies (IgG-depl.) For western blot analysis 4 lg, 1 lg, 0.25 lg and 0.0625 lg of IdeS were run on a denaturing, reducing 12% SDS– PAGE. Western blots were performed using standard techniques using 5% dry milk in PBS + 0.05% Tween-20 as blocking buffer. Commercial polyclonal IgG (0.1 mg/ml) was used as primary antibody and Goat anti-human HRP conjugate (BioRad) as secondary antibody. For ELISA analyses enzymatically inactive mutant protein IdeSC94S at a protein concentration of 0.1–10 lg/ml was coated on a Maxisorp plate (Nunc) in 50 mM carbonate buffer pH 9.6. 0.05% Tween-20 in PBS was used as washing buffer and 2% BSA was used as blocking reagent. IgG or IgG-depl. were used as primary antibody at a concentration of 0.1–100 lg/ml. Goat anti-human HRP conjugate (BioRad) (1/2000) was used as secondary antibody. HRP substrate kit (BioRad) was used for detection and a Tecan Infinite M200 instrument was used for quantification. Endopeptidase assays: To determine the kinetics of IdeS, 2 nM IdeS (256 nM for the detection of the second cleavage) was mixed with 1.67–200 lM IgG-depl. and incubated at 37 °C. Each reaction was sampled at multiple time points and stopped by incubation at 96 °C in pre-warmed non-reducing loading buffer. Buffer was assay buffer with 1.25 lM BSA added. Samples were analyzed by 8% SDS–PAGE under non-reducing conditions. Gels were stained with Coomassie Fluor Orange (Invitrogen). The rate of reaction was determined by densitometrically quantifying the bands. Bands corresponding to IgG and scIgG were used to determine the rate of first cleavage, and bands corresponding to scIgG and F(ab0 )2 were used to determine the rate of second cleavage. Band intensities were quantified using the LAS4000 imaging system and Multi Gauge Software (Fujifilm, Version 3.2). Kinetic constants were calculated using GraphPad Prism non-linear regression tools. All experiments were done at least twice with duplicate samples (n = 4). To determine the effect of neutralizing antibodies on IdeS activity, 66.6 lM IgG were incubated with 5–1280 lM IdeS at 37 °C and reactions were sampled between 5 and 160 min and analyzed as described above. For determination of specific activity dependency on enzyme concentration, IgG-depl. at a concentration of 66.6 lM was incubated with 0.5–32 nM IdeS and analyzed as described above. Dynamic light scattering measurements: Recombinant IdeS in PBS at 180 lM and 18 lM was analyzed by dynamic light scattering (DLS) (Nano Zetasizer equipped with a HeNe-laser with a wave length of 633 nm, Malvern Instruments). 50 ll of sample was measured at room temperature for 480 s and back scatter was detected at an angle of 173°. The DLS data was analyzed using the Dispersion Technology Software v.5.10 (Malvern). Size exclusion chromatography: Recombinant IdeS at 1 mM (1 ml; 37.29 mg/ml) was applied to a HiPrep 16/60 Sephacryl

1819

S-100 HR column (GE Biosciences) controlled by an FPLC system (ÄktaPrime plus, GE Biosciences). Separation was carried out at a flow rate of 0.5 ml/min at room temperature. Fractions of the isolated first and second peak were pooled and concentrated to 1 ml giving concentrations of 1.04 and 12.81 mg/ml respectively and once more subjected to SEC analysis. Peak fractions were analyzed by SDS–PAGE on 16% Novex Tricine gels (Invitrogen) under reducing conditions. Bands of the first peak sample migrating between 27 and 15 kDa were subjected to MALDI-TOF mass spectroscopy at the Umeå Protein Analysis Facility, Umeå University. To assess endopeptidase activity in peak fractions, protein from both peaks (10 ll; 0.05 mg/ml in assay buffer) was incubated with human IgG1 (10 ll) (Sigma) at 8.07 lM in 20 mM Tris-buffered saline pH 8.0 for 5 min at 37 °C and reactions were terminated with iodoacetamide (10 ll, 16.5 mM) (Sigma). Approximately 2 lg protein from each reaction was subjected to 8% or 10% SDS–PAGE under denaturing, non-reducing conditions and stained with Coomassie blue. 3. Results Substrate IgG preparation: Antibodies against IdeS have been found in serum samples of healthy blood donors, as well as in samples from patients with mild or invasive streptococcal disease [3,14–16]. Specific antibodies are also present in commercial sources of polyclonal IgG as determined by Western blot (Fig. 1A) and ELISA (data not shown). Commercial IgG preparations at physiological concentrations were found to completely inhibit 20 nM of IdeS, and significant inhibition of up to 80 nM of IdeS could be detected (data not shown). Due to these relatively high concentrations of neutralizing antibodies it is likely that IdeS antibodies interfered with earlier characterizations of IdeS0 IgG cleaving properties [9,10]. For appropriate determination of IdeS kinetics, specific IdeS antibodies were depleted from IgG preparations by antigen affinity chromatography using the enzymatically inactive IdeSC94S protein as antigen. Affinity purified IgG showed

Fig. 1. Detection of IdeS specific antibodies. (A) Western blot analysis of IdeS using polyclonal human IgG as primary antibody. (B) Comparison of IgG and IgG depleted of IdeS-specific antibodies (IgG-depl.) as substrates for IdeS. IdeS was incubated with either IgG (lane 1) or IgG-depl. (lane 2). IdeS specific antibodies prevent cleavage of IgG, while IgG-depl. is cleaved by IdeS.

1820


Fig. 2. Time course of IgG cleavage by IdeS. (A) IgG1 was cleaved by IdeS and proportions of intact IgG, scIgG and F(ab0 )2 were analyzed by SDS–PAGE. Most IgG is converted to scIgG prior to second chain cleavage. (B) Schematic representation of sequential IgG cleavage by IdeS. scIgG, single cleaved IgG; Fc, receptor binding domain; F(ab0 )2, antigen binding domain.

no inhibitory capacity towards IdeS (Fig. 1B) and was used as substrate in subsequent experiments. Determination of heavy chain cleavage by IdeS: First and second chain cleavage steps were detected by denaturing, non-reducing SDS–PAGE that allows for the separation of IgG, scIgG, F(ab0 )2 and ½Fc fragments (Fig. 2 and [4]). Densitometric measurements of the SDS–PAGE gels were used to measure ratio of substrate/ product at each time point. The rate of the reaction was calculated as the number of substrate molecules cleaved per second per IdeS molecule. The obtained velocity curves for first and second chain cleavage show typical Michaelis–Menten kinetics (Fig. 3). Kinetic analyses give a Km value of 7.2 lM and a kcat value of 10.1 s1 for first single chain cleavage and a Km value of 28 lM and a kcat value of 0.10 s1 for second chain cleavage (Fig. 3 and inset). Thus, the kcat for single chain cleavage of IgG, which represents the rate by which IgG becomes immunologically inactivated, is approximately 100 fold higher at Vmax compared to values obtained for complete cleavage of IgG. When monitoring product formation over time, basically all IgG is converted to scIgG before second chain cleavage

Fig. 3. Enzyme kinetics of first chain cleavage. Varying concentrations of IgG-depl. were incubated with IdeS and reactions were sampled continuously. The rate of reaction was determined by quantifying bands corresponding to substrate (IgG) and product (scIgG). Inset: Kinetics of second chain cleavage of IgG. Diamonds correspond to measured rates, solid lines correspond to an ideal Michaelis–Menten curve. The rate of reaction was determined by densitometrically quantifying the bands.

is detectable (Fig. 2 and [4]), demonstrating that the proposed simultaneous cleavage of both IgG heavy chains is not a relevant functional mechanism [12]. IdeS is active as a monomeric enzyme in solution: Previous analyses of IdeS enzyme kinetics demonstrated an aberrant, sigmoidal shaped velocity curve [9,11]. It has been suggested that IdeS has multiple binding sites or functions as a dimeric enzyme [9,12,17]. However, the current results for the formation of scIgG or complete cleavage of IgG by IdeS show no indication of an aberrant velocity curve, but instead demonstrate that IdeS kinetics follows a classical hyperbolic saturation curve typical for Michaelis– Menten kinetics (Fig. 3). Consequently, an enzyme kinetic argument for cooperativity is not evident from these results. Assuming that IdeS is active as a dimer, the proportion of the active dimer complex should be dependent on enzyme concentration [18,19] and therefore the specific activity should increase as a function of enzyme concentration [18,19]. However, when monitoring enzyme activity as a function of enzyme concentration the specific activity of IdeS proved independent from enzyme concentration (Fig. 4), indicating that dimer formation is not a prerequisite for IdeS activity. This view is supported by the finding that covalently immobilized IdeS, which is prevented from oligomerisation, also exhibits efficient IgG cleavage [20] (data not shown). A KD for IdeS dimerization has been determined in the lM range for immobilized IdeS [11,12]. IdeS stoichiometry in solution was investigated by dynamic light scattering (DLS). DLS measures the size of proteins in solution by determining the velocity of Brownian motions. IdeS at 18 and 180 lM was analyzed and the size of IdeS in solution was estimated to approximately 37 kDa with only one single peak detected at both concentrations (Fig. 5). Based on these data we conclude that only monomeric IdeS is present in solution and that formation of a dimeric protein cannot be confirmed. Size exclusion chromatography: The results from DLS analyses are in contrast to previous data from SEC experiments describing the presence of monomeric and dimeric forms of IdeS in solution [12]. IdeS at a concentration of 1 mM was applied to SEC chromatography. The protein eluted in two peaks as previously reported (Fig. 6A, top panel, [12]), but when isolated peak fractions were concentrated to original volume and reloaded on the SEC column, no equilibrium between the potential dimeric and monomeric forms of the enzyme was detected; instead two stable protein peaks were evident (Fig. 6A, mid and bottom panel). Fractions from the first peak contained several smaller protein bands on SDS–

Fig. 4. Determination of specific activity dependency on enzyme concentration. Constant amounts of IgG-depl. were incubated with increasing amounts of IdeS. The specific activity of IdeS is independent from enzyme concentration.


1821

SEC, but since these products are present in sample fractions corresponding to the first peak, it appears that these proteins aggregate in complex with misfolded protein and minor amounts of active enzyme, creating higher molecular weight particles. If this hypothesis is correct, then the specific enzymatic activity of these fractions should be considerably lower compared to fractions derived from monomeric IdeS. Enzymatic activity in peak fractions was analyzed and confirmed that protein samples from the monomer peak contained high IgG cleaving activity converting all IgG substrate into ½Fc and F(ab0 )2 fragments at the time point analyzed (Fig. 6C, lane 3), while identical amounts of protein from the putative dimer-peak exhibited much lower activity with most IgG still in the scIgG form (Fig. 6C, lane 2). We conclude that the first peak eluting in SEC does not contain dimeric IdeS, but rather represents partially degraded – possibly misfolded protein – in complex with minor amounts of active protein and that the dimeric form of IdeS does not represent the enzymatically active state of the protein. Fig. 5. Size determination of IdeS in solution. The size of IdeS, at 180 and 18 lM, in solution was analyzed using dynamic light scattering (DLS). IdeS in solution is detected as a 37 kDa protein.

PAGE (Fig. 6B) These bands were identified by MALDI-TOF massspectrometry as truncated IdeS. Degradation products of IdeS should, due to their smaller size, elute after full length IdeS on

4. Discussion IgG antibodies represent a substantial threat to invading bacteria. The human pathogen S. pyogenes employs a highly specific IgG endopeptidase, IdeS, for this purpose [2,3].

Fig. 6. Analysis of IdeS stoichiometry by size exclusion chromatography (SEC). (A) SEC elution profile of recombinant IdeS. Fractions of the first and second peak were chosen to avoid cross-contamination between peaks and collected fractions were pooled, re-concentrated and reloaded on the same column. (B) Protein content analyzed by SDS– PAGE. Lane 1, fraction collected from the first peak; lane 2, fraction collected from the second peak. (C) Endopeptidase activity of IdeS collected from SEC fractionation. IgG1 degradation was analyzed under non-reducing conditions by SDS–PAGE.

1822


Previous studies indicated that IgG cleavage by IdeS occurs in a sequential manner, generating scIgG molecules [4] lacking effector functions [5]. A renewed analysis of IdeS kinetics, measuring the generation of scIgG, shows that IdeS cleaves IgG at a significantly faster rate than previously reported. A 40–100 fold (Fig. 2 [7,11]) increased kcat value for first single chain cleavage defines IdeS as an even more efficient IgG endopeptidase and adds to the view that IgG hydrolysis by IdeS significantly contributes to streptococcal virulence. ScIgG binds only poorly to Fcc receptors, most likely due to the disruption of key structural factors in the hinge domain [5], which also might explain the observed differences between IgG and scIgG as a substrate for IdeS. Although IgG effector functions are efficiently inhibited after single chain cleavage, second chain cleavage, although slow, is still important for streptococcal pathogenesis. ½Fc fragments, prime neutrophil ROS production, leading to an earlier and more powerful response after activation by immunocomplexes [20]. Therefore, IgG cleavage can be regarded as a two-step process, where cleavage of the first heavy chain of IgG inactivates IgG effector functions and cleavage of the second heavy chain generates ½Fc fragments that can prime and affect neutrophils. Previous studies addressed IdeS enzymatic properties and measured proteolysis of IgG heavy chains [7,9,11]. Results from those analyses presented a sigmoidal shaped velocity curve and were interpreted as cooperativity [7,9,11] that could be achieved either by the presence of two separated substrate recognition motifs on one enzyme monomer or by a dimeric enzyme as the catalytically active form of IdeS [9,12]. Dimer formation has also been proposed as an explanation for the pronounced substrate specificity of IdeS [12]. Although an attractive theory on the mechanism, the data of the current study demonstrate a sequential cleavage of IgG by active IdeS monomers and do not support the hypothesis of dimer formation. Notably, scIgG generation by IdeS follows a classical Michaelis–Menten saturation curve, and does not support the notion of cooperative binding sites. Therefore, dimer formation cannot be a prerequisite for enzymatic activity and in fact no evidence for functional dimers was found (Figs. 2, 4 and 5). Previous mutational studies assigned potential interface residue F129 as important for dimer formation, but in absence of a functional dimer, we suggest that this residue, although important for enzymatic activity, most likely affects the structural integrity of IdeS, which is supported by the finding that mutations in this residue strongly affected the solubility of the enzyme [12]. Putative dimer forms of IdeS have not been observed when IdeS was first crystallized at 1.9 Å resolution [21] and in silico analyses of available crystals (PDB codes 2AU1, 2AVW and 1Y08) using PISA CSS and DiMoVo algorithms [22,23] indicate that IdeS dimer crystals might rather be a result of crystal packaging [24] (http://www.ebi.ac.uk/msd-srv/prot_int/pistart.html; last accession 2012-11-12). In summary, we demonstrate that IdeS, the IgG endopeptidase of S. pyogenes, exhibits a considerable faster proteolytic activity towards its sole substrate IgG than previously reported. The newly determined kcat value of 10.1 s1 makes it feasible that the enzyme actively contributes to the defense of S. pyogenes against specific antibodies and underlines the importance of the IgG protease. In addition, we also demonstrate that the protease is active in its monomeric form. Acknowledgements This work was supported by the Swedish Research Council (project numbers 2006-4522 and 2009-4997 and Insamlingsstiftelsen at Umea University.

References [1] Cunningham, M.W. (2000) Pathogenesis of group A streptococcal infections. Clin. Microbiol. Rev. 13, 470–511. [2] von Pawel-Rammingen, U., Johansson, B.P. and Bjorck, L. (2002) IdeS, a novel streptococcal cysteine proteinase with unique specificity for immunoglobulin G. EMBO J 21, 1607–1615. [3] 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. and Musser, J.M. (2001) Evasion of human innate and acquired immunity by a bacterial homolog of CD11b that inhibits opsonophagocytosis. Nat. Med. 7, 1298–1305. [4] Ryan, M.H., Petrone, D., Nemeth, J.F., Barnathan, E., Bjorck, L. and Jordan, R.E. (2008) Proteolysis of purified IgGs by human and bacterial enzymes in vitro and the detection of specific proteolytic fragments of endogenous IgG in rheumatoid synovial fluid. Mol. Immunol. 45, 1837–1846. [5] Brezski, R.J., Vafa, O., Petrone, D., Tam, S.H., Powers, G., Ryan, M.H., Luongo, J.L., Oberholtzer, A., Knight, D.M. and Jordan, R.E. (2009) Tumor-associated and microbial proteases compromise host IgG effector functions by a single cleavage proximal to the hinge. Proc. Natl. Acad. Sci. USA A106, 17864–17869. [6] Brezski, R.J. and Jordan, R.E. (2010) Cleavage of IgGs by proteases associated with invasive diseases: an evasion tactic against host immunity? mAbs 2, 212–220. [7] Agniswamy, J., Lei, B., Musser, J.M. and Sun, P.D. (2004) Insight of host immune evasion mediated by two variants of group a Streptococcus Mac protein. J. Biol. Chem. 279, 52789–52796. [8] Berggren, K., Vindebro, R., Bergstrom, C., Spoerry, C., Persson, H., Fex, T., Kihlberg, J., von Pawel-Rammingen, U. and Luthman, K. (2012) 3-Aminopiperidine-based peptide analogues as the first selective noncovalent inhibitors of the bacterial cysteine protease IdeS. J. Med. Chem. 55, 2549–2560. [9] Vincents, B., von Pawel-Rammingen, U., Bjorck, L. and Abrahamson, M. (2004) Enzymatic characterization of the streptococcal endopeptidase, IdeS, reveals that it is a cysteine protease with strict specificity for IgG cleavage due to exosite binding. Biochemistry 43, 15540–15549. [10] Hess, J.L., Porsch, E.A., Shertz, C.A. and Boyle, M.D. (2007) Immunoglobulin cleavage by the streptococcal cysteine protease IdeS can be detected using protein G capture and mass spectrometry. J. Microbiol. Methods 70, 284–291. [11] Vincents, B., Vindebro, R., Abrahamson, M. and von Pawel-Rammingen, U. (2008) The human protease inhibitor cystatin C is an activating cofactor for the streptococcal cysteine protease IdeS. Chem. Biol. 15, 960–968. [12] Agniswamy, J., Nagiec, M.J., Liu, M., Schuck, P., Musser, J.M. and Sun, P.D. (2006) Crystal structure of group A streptococcus Mac-1: insight into dimermediated specificity for recognition of human IgG. Structure 14, 225–235. [13] Vincents, B., Guentsch, A., Kostolowska, D., von Pawel-Rammingen, U., Eick, S., Potempa, J. and Abrahamson, M. (2011) Cleavage of IgG1 and IgG3 by gingipain K from Porphyromonas gingivalis may compromise host defense in progressive periodontitis. FASEB J25, 3741–3750. [14] Akesson, P., Moritz, L., Truedsson, M., Christensson, B. and von PawelRammingen, U. (2006) IdeS, a highly specific immunoglobulin G (IgG)cleaving enzyme from Streptococcus pyogenes, is inhibited by specific IgG antibodies generated during infection. Infect. Immun. 74, 497–503. [15] Lei, B., DeLeo, F.R., Reid, S.D., Voyich, J.M., Magoun, L., Liu, M., Braughton, K.R., Ricklefs, S., Hoe, N.P., Cole, R.L., Leong, J.M. and Musser, J.M. (2002) Opsonophagocytosis-inhibiting mac protein of group a streptococcus: identification and characteristics of two genetic complexes. Infect. Immun 70, 6880–6890. [16] Akesson, P., Rasmussen, M., Mascini, E., von Pawel-Rammingen, U., Janulczyk, R., Collin, M., Olsen, A., Mattsson, E., Olsson, M.L., Bjorck, L. and Christensson, B. (2004) Low antibody levels against cell wall-attached proteins of Streptococcus pyogenes predispose for severe invasive disease. J. Infect. Dis. 189, 797–804. [17] Cornish-Bowden, A. and Cardenas, M.L. (1987) Co-operativity in monomeric enzymes. J. Theor. Biol. 124, 1–23. [18] Schmidt, U. and Darke, P.L. (1997) Dimerization and activation of the herpes simplex virus type 1 protease. J. Biol. Chem. 272, 7732–7735. [19] Darke, P.L., Cole, J.L., Waxman, L., Hall, D.L., Sardana, M.K. and Kuo, L.C. (1996) Active human cytomegalovirus protease is a dimer. J. Biol. Chem. 271, 7445– 7449. [20] Soderberg, J.J. and von Pawel-Rammingen, U. (2008) The streptococcal protease IdeS modulates bacterial IgGFc binding and generates 1/2Fc fragments with the ability to prime polymorphonuclear leucocytes. Mol. Immunol. 45, 3347–3353. [21] Wenig, K., Chatwell, L., von Pawel-Rammingen, U., Bjorck, L., Huber, R. and Sondermann, P. (2004) Structure of the streptococcal endopeptidase IdeS, a cysteine proteinase with strict specificity for IgG. Proc. Natl. Acad. Sci. USA A101, 17371–17376. [22] Krissinel, E. and Henrick, K. (2007) Inference of macromolecular assemblies from crystalline state. J. Mol. Biol. 372, 774–797. [23] Bernauer, J., Bahadur, R.P., Rodier, F., Janin, J. and Poupon, A. (2008) DiMoVo: a Voronoi tessellation-based method for discriminating crystallographic and biological protein–protein interactions. Bioinformatics 24, 652–658. [24] Olsen, J.G., Dagil, R., Niclasen, L.M., Sorensen, O.E. and Kragelund, B.B. (2009) Structure of the mature Streptococcal cysteine protease exotoxin mSpeB in its active dimeric form. J. Mol. Biol. 393, 693–703.