IdeS, a Highly Specific Immunoglobulin G (IgG)-Cleaving Enzyme from Streptococcus pyogenes, Is Inhibited by Specific IgG Antibodies Generate
http://www.100md.com
感染与免疫杂志 2006年第1期
Department of Clinical Science, Lund University, Lund, Sweden
Department of Molecular Biology, Ume University, Ume, Sweden
Department of Community Medicine, Malm University Hospital, Lund University, Malm, Sweden
ABSTRACT
IdeS, a recently discovered cysteine proteinase secreted by the important human pathogen Streptococcus pyogenes, interferes with phagocytic killing by specifically cleaving the heavy chain of immunoglobulin G. The fact that the enzyme targets one of the key molecules of the adapted immune response raised the question of whether an antibody response against IdeS could inhibit, i.e., neutralize, enzyme activity. Paired acute- and convalescent-phase serum samples from patients with pharyngotonsillitis (n = 10), bacteremia (n = 7), and erysipelas (n = 4) were analyzed. Antibodies with the ability to neutralize IdeS enzymatic activity were already found in two-thirds of acute-phase sera. However, patients who seroconverted to IdeS, in particular patients with pharyngotonsillitis and erysipelas, developed specific antibodies during convalescence with an increased capability to efficiently neutralize the enzymatic activity of IdeS. Also, the presence of neutralizing antibodies decreased the ability of IdeS to mediate bacterial survival in human immune blood. In patients with bacteremia, several acute-phase sera contained neutralizing antibodies, but no correlation was found to severity or outcome of invasive infections. Still, the fact that the human immune response targets the enzymatic activity of IdeS supports the view that the enzyme plays an important role during streptococcal infection.
INTRODUCTION
Streptococcus pyogenes is an important human bacterial pathogen that causes a variety of diseases, including pharyngotonsillitis, impetigo, scarlet fever, septicemia, necrotizing fasciitis (NF), and streptococcal toxic shock syndrome (STSS) (6, 7). The survival of S. pyogenes depends on its ability to avoid the various actions of the human immune system. Immunoglobulin G (IgG) plays a key role in the immune defense by specifically recognizing invading microorganisms and mediating their killing by professional phagocytes and the complement system. In order to persist, pathogenic bacteria have to find ways to avoid recognition by immunoglobulins and to interfere with IgFc-mediated phagocytosis. S. pyogenes has evolved a specific enzyme to deal with opsonizing IgG antibodies. This enzyme, designated IdeS or streptococcal Mac-1 (12, 23), is a secreted cysteine proteinase that specifically cleaves the heavy chain of IgG (1, 22, 23). So far, no other substrates for IdeS have been identified, and IgG is the sole substrate of IdeS in plasma samples (22). Due to its early and sustained expression during growth (23) and its highly specific proteolytic activity, IdeS is a tailor-made defense against Fc-mediated phagocytic killing (1, 12, 13, 22-25). Two protein variants of IdeS, complex I and complex II, have been described based on differences in the amino acid sequences of the middle thirds of the proteins (amino acids 112 to 205) (13). Complex II variants have been reported to have weak endopeptidase activity and to interfere with phagocytic killing by blocking the interaction of the FcIIIb receptor with specific antibodies (1, 13), while complex I variants exert their inhibitory function through proteolytic cleavage of IgG (1).
The expression of streptococcal virulence factors in vivo is commonly analyzed by the determination of antibody levels towards particular streptococcal proteins in human blood samples. Paired acute- and convalescent-phase serum samples from children with streptococcal pharyngotonsillitis have been used to study the antibody response against the streptococcal C5a peptidase (17), and acute-phase serum IgG levels towards streptococcal M proteins and pyrogenic exotoxins (Spe's) have been analyzed and correlated to the outcome of disease (4, 10, 18). In another study, acute-phase serum samples of patients suffering from invasive streptococcal infections were analyzed for IgG antibodies towards six streptococcal virulence factors (SclA, SclB, MtsA, Grab, EndoS, and IdeS) (3). Detectable antibody levels against IdeS/Mac were found in this study as well as in two previous studies (12, 13). Other analyses of antistreptococcal antibodies, however, have shown that the total amount of specific antibodies does not necessarily correlate with the level of neutralizing antibodies, and that antibody quality might be clinically more important than antibody quantity (15, 16). The fact that S. pyogenes has evolved an enzyme that specifically targets IgG raises the question of whether specific IgG antibodies are able to neutralize IdeS and the further question of whether the presence of neutralizing antibodies correlates with manifestations of S. pyogenes infection or affects the severity of S. pyogenes-induced disease.
(Part of this work was presented as a scientific poster at the XVIth Lancefield International Symposium on Streptococci and Streptococcal Diseases, 25 to 29 September 2005, Palm Cove, Australia.)
MATERIALS AND METHODS
Patients, serum samples, and bacterial isolates. Serum samples were collected from seven patients treated for S. pyogenes bacteremia at the Clinic for Infectious Diseases, Lund University Hospital, Lund, Sweden. Five of the patients had a nonsevere bacteremia, one patient developed necrotizing fasciitis, and one patient presented with STSS (26). S. pyogenes strains were isolated from blood cultures, and acute-phase serum (days 1 to 4 after onset of symptoms; median, day 1) and convalescent-phase serum (days 18 to 188 after onset; median, day 24) samples were collected from each patient. Sera from 10 patients with pharyngotonsillitis were collected at the Community Health Center Sorgenfrimottagningen, Malm, Sweden. Acute-phase serum samples (taken between days 0 and 4 after onset of symptoms; median, day 1.8) and convalescent-phase serum samples (taken between days 24 and 30 after onset; median, day 26) were collected from each patient. S. pyogenes strains were isolated by throat swab cultures from all patients. Four patients treated for erysipelas at the Clinic for Infectious Diseases, Lund, Sweden, were also included in the study. They had typical signs of a bacterial skin infection, with fever and a rapid spreading of a painful erythema on a lower limb. From these patients, acute-phase sera were collected between days 0 and 5 after onset of symptoms (median, day 2.6), and convalescent-phase serum samples were taken between days 28 and 37 after onset (median, day 31). No bacterial isolate was available from the four erysipelas patients. Acute-phase serum samples from patients with either severe (n = 4) or nonsevere (n = 4) group A streptococcal invasive disease, and with high antibody titers towards IdeS, were from The Netherlands (1994 to 1997) and have been described elsewhere (3, 14).
Immunoglobulins. Human polyclonal IgG and myeloma IgG1() were purchased from Sigma. IgG from patient serum was purified by affinity chromatography using HiTrap protein G Sepharose (Amersham Biosciences) according to the manufacturer's instructions. Myeloma IgG1 was labeled with 125I using the Bolton and Hunter reagent as described by the manufacturer (Amersham Bioscience). Free 125I was separated from labeled protein on a PD10 column (Amersham Bioscience).
Bacterial strains and growth conditions. S. pyogenes strains used in this study are listed in Table 1. Streptococci were routinely grown in Todd-Hewitt broth (TH) (Difco) or TH broth supplemented with 2 to 5% heparinized human plasma at 37°C in 5% CO2. The M serotype of streptococcal strains was determined by sequencing emm gene-specific PCR products as previously described (5). Sequences obtained from emm-specific PCR products were subjected to homology searches in the emm sequence database at the Centers for Disease Control and Prevention (http://www.cdc.gov/ncidod/biotech/strep/emmtypes.htm; last accession, 20 February 2005). The sequence of bases ranging approximately from 65 to 165 obtained with primer 1, TATT(C/G)GCTTAGAAAATTAA, has been implicated mostly as M type specific (5). Sequences that show greater than 95% identity with a reference emm gene sequence were considered to be given emm genes. Allelic forms of IdeS (13) were amplified as previously described (23), and sequences were determined by sequence analysis of nucleotides 121 to 900 of the ideS gene by use of an Applied Biosystems 3100 automated sequencer (Applied Biosystems). Multiple sequence alignment of the internal amino acid sequence of IdeS was conducted with CLUSTAL W (20).
Expression and purification of recombinant IdeS. The expression of recombinant IdeS in Escherichia coli has been described previously (23). Fusion proteins were purified on glutathione-Sepharose (Amersham Biosciences) according to standard protocols. IdeS protein of the complex II family was subcloned from the M8 serotype strain Se4 and purified as described above (for IdeS variants used in this study, see Fig. 2C). IdeS-304 and IdeS-218 were created by PCR with primers I (23) and III (5'-TTCTTTGAATTCTTAAGCAGAAATAGCTACTTTTCC-3') for IdeS-304 and with primers IV (5'-GGTGATCAAAGGATCCTATTGACAAGTC-3') and V (5'-CGGAATTCTTAATTGGTCTGATTCCAAC-3') for IdeS-218 by using IdeS inserted into a pGEX-5X-3 vector as the template (23). PCR fragments were cloned into the corresponding sites of plasmid pGEX-5X-3 (Amersham Biosciences). IdeS-232 and IdeS-47-168 were created by digesting full-length ideS in pGEX5-3 with restriction endonuclease AvaI or StyI (Roche). The digested plasmids were purified, religated, and transformed into E. coli strain NovaBlue (Novagen). All variants were purified as glutathione S-transferase (GST) fusion proteins as described previously (23).
ELISA. Microtiter plates (Maxisorb; NUNC) were coated with enzymatically inactive IdeSC94G (24) at 0.4 μg/ml in coating buffer (0.05 M NaHCO3, pH 9.6). Plates were washed with PBST (0.05% Tween in phosphate-buffered saline [PBS]) and blocked with 2% bovine serum albumin (Sigma) in PBST (PBSTA). Dilutions of each patient and control serum sample in PBSTA were added to the wells. Antigen concentrations and serum dilutions were judged as optimal by checkerboard titration. Bound antibodies were detected by incubation with peroxidase-conjugated antibody against human IgG (1/3000) (Bio-Rad), and enzyme-linked immunosorbent assay (ELISA) plates were developed as previously described (3). Blank samples without serum were included in quadruplicate on each plate. Optical density at 420 nm (OD420) was determined as described previously. An ELISA index was calculated by subtracting OD420 values of blank control samples from the values obtained with serum samples and by dividing the mean OD420 value for each sample by the mean OD420 values for a standard positive serum sample; these were determined in quadruplicate on each plate.
IdeS activity assays. For standard IdeS activity assays in bacterial growth medium, bacteria were grown to an OD620 of 0.4 in TH broth supplemented with 2% heparinized human plasma and 40 μM of the cysteine proteinase inhibitor trans-epoxysuccinyl-L-leucylamido-(4-guanidino) butane (E64; Sigma) to block potential SpeB activity. Cleavage of IgG was determined by analyzing supernatant samples on 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels, and IdeS activity was determined by the presence or absence of a diagnostic 31-kDa IgG cleavage product (1, 22, 23).
Quantification of IdeS activity. Quantification of the endopeptidase activity of IdeS was done as previously described for microbial IgA proteinases (9). Briefly, various amounts of the enzyme were incubated with 105 cpm of 125I-IgG and 4 μg of unlabeled myeloma IgG1 in PBS. Samples were kept at 37°C for 60 min, and the reactions were stopped by the addition of SDS sample buffer and incubation at 95°C for 5 min. IgG chains and IgG cleavage products were separated on 12% SDS-PAGE gels and visualized by staining with Coomassie blue (R-250; U.S. Biochemicals). Protein bands corresponding to the IgG heavy chain, IgG light chain, and the IdeS cleavage product (CP) were excised from the gel for determination of radioactivity. Samples were counted in an LKB Wallac Compugamma counter, and IdeS activity was calculated as the ratio of radioactivity in CP to total heavy chain radioactivity (CP plus heavy chain).
Detection of neutralizing antibodies in patient serum samples. Generally, IdeS was incubated with 5 to 10 μl of acute-phase or convalescent-phase serum samples at 25°C for 10 min. The reaction volume was adjusted to 28 μl with PBS, and 105 cpm of 125I-IgG was added as the substrate. Incubation was continued at 37°C for 45 min. The reactions were stopped by adding SDS sample buffer, and samples were analyzed by 12% SDS-PAGE. IdeS activity was determined as described above and correlated to enzyme activity in serum depleted of IgG. The IdeS endopeptidase activity was 27.5% cleavage of the IgG heavy chain under these experimental conditions, and this value was arbitrarily set as 100%. Activity in patient serum samples is expressed relative to this standard value.
Bactericidal assay. Purified IdeS (5 μg) was incubated with a three- to fivefold molar excess of either human polyclonal IgG (Sigma) or total patient IgG at 25°C for 30 min. As a control, IdeSC94S (25) was incubated with a similar amount of human polyclonal IgG (Sigma). The suspension was added to 1 ml of heparinized human blood containing antibodies against S. pyogenes, and incubation was continued for 60 min at 37°C with rotation. Approximately 500 CFU of a mid-log-phase culture of strain AP1 (OD620 of 0.15) was used to inoculate the pretreated blood. The tubes were incubated with rotation at 37°C for 60 min, and bacterial survival was monitored by plating dilutions on blood agar plates.
RESULTS
Characterization of bacterial isolates. The M serotypes of the 17 clinical S. pyogenes isolates were determined by emm sequence analysis (5) and are listed in Table 1. Eight of the 11 different serotypes identified are among the most frequently identified M serotypes from clinical isolates (21). PCR analysis demonstrated the presence of the ideS gene in all isolates (data not shown). Strains of the same M serotype generally carried the two previously described (13) ideS alleles. Sequence comparison of the complex II IdeS variants found in the present clinical strains confirmed the reported high degree of sequence homology between these variants (13), with only a few amino acid substitutions. Two changes, however, appear to be of particular interest. (i) Strains So3 and So4 express an IdeS protein that is characterized by a four-amino-acid deletion in positions 243 to 246, distinguishing these proteins from all other IdeS variants. (ii) IdeS of the M28 serotype carries a cysteine residue in position 255. All other proteins, including complex I variants, carry a tyrosine residue at this position (Fig. 1).
IdeS antibody levels in acute-phase and convalescent-phase patient serum samples. IdeS-specific antibodies were detectable in serum samples of all patients (Table 2). Seroconversion was defined as an increase in the ELISA index between acute- and convalescent-phase sera of more than 40%. This was found in 10 of the 21 patients investigated, and in 5 of these the increase exceeded 100% (Table 2). For two patients, PS1 and PS4, high antibody levels towards IdeS had already been detected in acute-phase serum samples.
Specific IgG antibodies towards IdeS inhibit enzymatic activity. In a patient (PS2) who seroconverted to IdeS, no IdeS-mediated cleavage of IgG could be detected in convalescent-phase serum samples containing high levels of anti-IdeS antibodies (Fig. 2A; >90% inhibition of enzyme activity towards myeloma IgG1 in sample PS2C). However, when patient IgG was depleted from the sample and replaced by nonspecific polyclonal IgG, the myeloma IgG1 was efficiently cleaved, and no IdeS inhibitory activity could be detected (Fig. 2A, bar 3; 90% cleavage of myeloma IgG1). The serum regained its neutralizing ability when patient IgG was added back to the sample (Fig. 2A, bar 4; >78% inhibition). Thus, IgG antibodies towards IdeS clearly mediate the neutralization of enzymatic activity in serum samples.
In a second experiment, neutralizing convalescent-phase serum was incubated with an excess of inactive IdeSC94S (25) prior to the addition of active IdeS. The absorption of specific IdeS antibodies to inactive IdeSC94S efficiently abolished the neutralizing activity of convalescent-phase serum samples (Fig. 2B, lane 3; presence of 31-kDa cleavage product). Thus, the neutralizing activity in convalescent-phase serum samples is mediated by the binding of specific IgG antibodies to IdeS. Different enzymatically inactive truncation variants of IdeS (Fig. 2C and data not shown) were used to analyze the binding of neutralizing antibodies (Fig. 2B). Although the IdeS variants have retained the ability to bind IgG (data not shown; 11), only IdeS-304, containing both Cys94 and His262 active-site residues, absorbed neutralizing antibodies and mediated cleavage of IgG by active IdeS (Fig. 2B, GST-IdeS-304; presence of 31-kDa cleavage product). Thus, neutralizing activity in convalescent-phase serum samples is likely to be mediated by the binding of specific IgG antibodies to the catalytic region of IdeS.
Quality of patient antibodies. Studies of antistreptococcal antibodies have suggested that the quality of the antibodies, i.e., the neutralizing activity, may be clinically more relevant than the quantity of antibodies (15, 16). In an attempt to correlate the levels of antibodies towards IdeS with the degree of inactivation of IdeS, patient sera were incubated with IdeS, and the enzymatic activity was determined by measuring the cleavage of radiolabeled 125I-labeled myeloma IgG1. This monoclonal IgG did not inhibit IdeS activity (data not shown). The results for two patients with a marked immune response (>100% increase in ELISA index; PT15 and PS2) and for one patient with only a moderate response (19% increase of ELISA index; PS5) are shown in Fig. 3A to C. Convalescent-phase serum samples from all three patients clearly inhibit IdeS activity (83%, 83%, and 23% inhibition compared to acute-phase serum at a 100-fold serum dilution). The inhibitory capacity of convalescent-phase serum samples diminished upon serum dilution, while IdeS activity remained unaffected in diluted acute-phase serum samples (Fig. 3A to C). Relative enzyme activity levels and the degrees of inhibition were determined for all patient serum samples (Table 3). In sera from patients with pharyngotonsillitis and erysipelas, an increase in the ELISA index correlated well with an increase in enzyme inhibition (Table 3). In contrast, seroconversion was not consistently reflected by an increase of enzyme inhibition when analyzing serum samples from bacteremia patients (Table 3; see data for PS3, PS6, and PS7). Furthermore, two-thirds of the analyzed sample pairs had already shown neutralizing activity in the acute-phase serum samples (Table 3). Still, serum samples from patients that did not seroconvert to IdeS showed no increase in neutralizing activity (Table 3), with the exception of the serum sample pair from patient PS1, which had already exhibited very high antibody titers in the acute-phase sample. It should be noted that the degree of inhibition, determined by analyzing the extent of 125I-labeled myeloma IgG1 cleavage, is most accurate when comparing acute-phase and convalescent-phase serum samples from the same patient. Comparisons of inhibition levels between patients or patient groups are not practicable, as for instance 54.9% enzyme inhibition was sufficient to block patient IgG cleavage in serum sample PT1C, while 71.1% inhibition was not sufficient to block patient IgG cleavage in sample PT22A (data not shown).
Antibody quality and invasive streptococcal disease. Since acute-phase serum samples from sepsis patients notably differ in their abilities to neutralize IdeS activity, we speculated that the quality of preexisting antibodies could be important for the outcome of invasive streptococcal diseases. Acute-phase serum samples from patients with severe invasive infections (n = 4) and nonsevere invasive infections (n = 4), all with high levels of IdeS-specific antibodies (Table 4), were tested for their abilities to inactivate IdeS. Seven out of eight samples neutralized IdeS activity (data not shown), and there was no difference between the groups of samples from severe and nonsevere cases (Table 4). Thus, the presence of neutralizing antibodies towards IdeS does not appear to directly affect the severity of invasive streptococcal disease.
Neutralizing antibodies interfere with IdeS-mediated survival in human blood. IdeS mediates streptococcal survival in whole human immune blood by cleaving opsonizing IgG antibodies (23, 24). Although the presence of neutralizing antibodies towards IdeS does not affect the overall outcome of invasive streptococcal disease, such antibodies still could be detrimental for bacterial survival. The effect of neutralizing antibodies on streptococcal survival was assayed by using a human immune blood bactericidal assay. IdeS was incubated with either polyclonal or neutralizing IgG and added to immune blood prior to inoculation with S. pyogenes as previously described (23). Although the concentration of IdeS used in these experiments exceeded the amount of the enzyme used in the neutralizing experiments (25-fold), streptococcal survival was reduced by 58% in the presence of neutralizing antibodies (Fig. 4). Thus, neutralizing antibodies clearly affected the ability of IdeS to interfere with phagocytic killing of S. pyogenes.
DISCUSSION
S. pyogenes has to employ specific strategies to avoid the action of opsonizing IgG antibodies during infection. The streptococcal cysteine proteinase IdeS represents a tailor-made defense against specific IgG antibodies (1, 12, 13, 22-25) and is widely expressed by clinically important serotypes (e.g., M1, M3, M4, and M12), as well as in newly validated serotypes, e.g., M122 (2) suggesting an important role for IdeS during S. pyogenes infections.
The main focus of this study was to characterize the immune response towards the IgG endopeptidase IdeS in acute-phase and convalescent-phase serum samples from patients suffering from streptococcal disease. In accordance with previous data (3, 12, 13), detectable levels of antibodies against IdeS were found in all samples investigated. The neutralizing abilities of these antibodies varied notably between different patient samples, but consistent results were obtained for patients suffering from pharyngotonsillitis and erysipelas, revealing that patients who seroconverted to IdeS developed IgG antibodies with the ability to neutralize IdeS enzymatic activity. In contrast, patients with no or weak seroconversion have only poor neutralizing capability (Table 3). For patients treated for streptococcal sepsis, less consistent results were obtained, and seroconversion was not necessarily reflected by an increase of enzyme inhibitory activity (see results for PS3, PS6, and PS7). However, IdeS is expressed as one of two distinct protein variants (13), and neutralizing antibodies might be specific for one of the two IdeS variants. This fact might explain the lack of inhibitory activity in samples from patients PS6 and PS7 (see below). For two-thirds of the patients, neutralizing activities could already be detected in acute-phase serum. It appears unlikely that these specific antibodies are the result of a newly acquired IgG immune response this early after onset of infection (median, day 1). Instead, the ability of acute-phase serum to inactivate IdeS most likely reflects a preexisting immune response caused by previous S. pyogenes infections or colonization. This view is consistent with the fact that anti-IdeS antibodies are widely distributed in the healthy population (3).
The finding that neutralizing antibodies against IdeS are present early during infection raised the question of whether these could be of importance for the outcome of invasive streptococcal diseases. This hypothesis was tested by analyzing IdeS-neutralizing activity of acute-phase serum samples from patients with severe or nonsevere invasive streptococcal infections. However, no such correlation between the presence of neutralizing antibodies against IdeS and the severity of invasive disease could be observed, as the capabilities of neutralizing IdeS enzymatic activity were indistinguishable between the different samples (Table 4). Thus, although neutralizing antibodies are involved in the host defense towards streptococci, the presence of antibodies against IdeS appears not to directly affect the outcome of invasive disease. However, the importance of IdeS endopeptidase activity for streptococcal survival in the human host is supported by the finding that neutralizing antibodies interfere with IdeS-mediated streptococcal survival ex vivo (Fig. 4). In contrast to results from experiments addressing the neutralizing capacity of patient IgG, IdeS activity is not completely blocked by patient antibodies, which is due to the relative decrease in the ratio of total patient antibodies present in whole blood to the amount of IdeS added (240- to 400-fold). Still, neutralizing antibodies significantly affected the ability of IdeS to interfere with Fc-mediated phagocytosis and clearly promoted survival of the bacteria in whole blood (Fig. 4).
Pharyngotonsillitis and erysipelas are streptococcal diseases with a high tendency to recur. In recurrent menstruation-associated Staphylococcus aureus toxic shock syndrome, the subsequent development of specific IgG antibodies against toxic shock syndrome toxin 1 protects from recurrent disease (19). The fact that patients develop antibodies with neutralizing activity against IdeS raises the questions of whether these patients are protected from recurrent disease and whether the presence of neutralizing antibodies might protect from recurrent mucocutaneous S. pyogenes infections rather than from invasive disease. This hypothesis is currently under investigation.
Sequence alignment of complex II IdeS variants revealed that strains of the M28 serotype express an IdeS variant with a cysteine residue in position 257 (Fig. 1) (13), while all other variants, including complex I proteins, carry a tyrosine residue in this position (Fig. 1) (13). The role of this second cysteine residue has not yet been investigated, but the recently published three-dimensional structure of IdeS (25) shows that the cysteine in position 257 is part of a flexible loop close to the catalytic site of the enzyme (25). The potential interaction of cysteine 257 and the catalytic site might explain the reported weak enzymatic activity of this variant (1, 13) and indicates that the previous characterization of complex II protein (1, 13) might not generally apply to other complex II variants. This view is supported by the fact that the complex II variants secreted by the clinical streptococcal strains exhibit enzymatic activity. However, neutralizing antibodies might be specific for one of the two IdeS variants, but only 4 of the 17 patient samples were found to contain neutralizing antibodies specific for complex II IdeS, and complex specificity correlated with the IdeS variant expressed by the infecting strain (data not shown).
The finding that an increase in the amount of antibodies does not necessarily correlate with an increased neutralizing capacity also supports the earlier notion (15, 16) that measurements of antibody titers in patient samples are not sufficient to assess the clinical and biological importance of antibodies during bacterial infection. However, the fact that seroconversion towards IdeS leads to the inactivation of enzyme activity strongly supports the view that IdeS IgG endopeptidase activity is an important streptococcal virulence determinant.
ACKNOWLEDGMENTS
E. M. Mascini, Eijkman-Winkler Institute for Microbiology, Utrecht University Hospital, is acknowledged for the generous gift of acute-phase serum samples. We thank Anders O. Persson for technical assistance.
This work was supported by grants from the Swedish Research Council (projects 7480, 13419, 14240, 14767, and 14800), the European Union, project Strep-EURO QLK2-CT-2002-01398, the Royal Physiographic Society in Lund, the Foundations of Crafoord and Wiberg, the University Hospital in Lund, and the SJCKMS Foundation and Insamlingsstiftelsen at Ume University.
The study was approved by the Research Ethics Committee, Lund University. We declare that we have no commercial or other relationship that might lead to a conflict of interest.
The first two authors contributed equally to this work.
REFERENCES
1. Agniswamy, J., B. Lei, J. M. Musser, and P. D. Sun. 2004. Insight of host immune evasion mediated by two variants of group A streptococcus Mac protein. J. Biol. Chem. 279:52789-52796.
2. kesson, P., A. Linder, J. Cronqvist, and B. Christensson. 2004. Group A Streptococcus bacteraemia complicated by osteomyelitis in an immunocompetent adult. Scand. J. Infect. Dis. 36:63-65.
3. kesson, P., M. Rasmussen, E. Mascini, U. von Pawel-Rammingen, R. Janulczyk, M. Collin, A. Olsen, E. Mattsson, M. L. Olsson, L. Bjrck, and B. Christensson. 2004. Low antibody titers to cell wall-attached proteins of Streptococcus pyogenes predispose for severe invasive disease. J. Infect. Dis. 189:797-804.
4. Basma, H., A. Norrby-Teglund, Y. Guedez, A. McGeer, D. E. Low, O. El-Ahmedy, B. Schwartz, and M. Kotb. 1999. Risk factors in the pathogenesis of invasive group A streptococcal infections: role of protective humoral immunity. Infect. Immun. 67:1871-1877.
5. Beall, B., R. Facklam, T. Hoenes, and B. Schwartz. 1997. Survey of emm gene sequences and T-antigen types from systemic Streptococcus pyogenes infection isolates collected in San Francisco, California; Atlanta, Georgia; and Connecticut in 1994 and 1995. J. Clin. Microbiol. 35:1231-1235.
6. Bisno, A. L. 2001. Acute pharyngitis. N. Engl. J. Med. 344:205-211.
7. Cunningham, M. W. 2000. Pathogenesis of group A streptococcal infections. Clin. Microbiol. Rev. 13:470-511.
8. Facklam, R. F., D. R. Martin, M. Lovgren, D. R. Johnson, A. Efstratiou, T. A. Thompson, S. Gowan, P. Kriz, G. J. Tyrrell, E. Kaplan, and B. Beall. 2002. Extension of the Lancefield classification for group A streptococci by addition of 22 new M protein gene sequence types from clinical isolates: emm103 to emm124. Clin. Infect. Dis. 34:28-38.
9. Gilbert, J. V., A. G. Plaut, B. Longmaid, and M. E. Lamm. 1983. Inhibition of microbial IgA proteases by human secretory IgA and serum. Mol. Immunol. 20:1039-1049.
10. Holm, S. E., A. Norrby, A. M. Bergholm, and M. Norgren. 1992. Aspects of pathogenesis of serious group A streptococcal infections in Sweden, 1988-1989. J. Infect. Dis. 166:31-37.
11. Kawabata, S., Y. Tamura, J. Murakami, Y. Terao, I. Nakagawa, and S. Hamada. 2002. A novel, anchorless streptococcal surface protein that binds to human immunoglobulins. Biochem. Biophys. Res. Commun. 296:1329-1333.
12. Lei, B., F. R. DeLeo, N. P. Hoe, M. R. Graham, S. M. Mackie, R. L. Cole, M. Liu, H. R. Hill, D. E. Low, M. J. Federle, J. R. Scott, and J. M. Musser. 2001. Evasion of human innate and acquired immunity by a bacterial homolog of CD11b that inhibits opsonophagocytosis. Nat. Med. 7:1298-1305.
13. Lei, B., F. R. DeLeo, S. D. Reid, J. M. Voyich, L. Magoun, M. Liu, K. R. Braughton, S. Ricklefs, N. P. Hoe, R. L. Cole, J. M. Leong, and J. M. Musser. 2002. Opsonophagocytosis-inhibiting Mac protein of group A streptococcus: identification and characteristics of two genetic complexes. Infect. Immun. 70:6880-6890.
14. Mascini, E. M., M. Jansze, J. F. Schellekens, J. M. Musser, J. A. Faber, L. A. Verhoef-Verhage, L. Schouls, W. J. van Leeuwen, J. Verhoef, and H. van Dijk. 2000. Invasive group A streptococcal disease in the Netherlands: evidence for a protective role of anti-exotoxin antibodies. J. Infect. Dis. 181:631-638.
15. Norrby-Teglund, A., K. Pauksens, S. E. Holm, and M. Norgren. 1994. Relation between low capacity of human serum to inhibit streptococcal mitogens and serious manifestation of disease. J. Infect. Dis. 170:585-591.
16. Norrby-Teglund, A., R. Kaul, D. E. Low, A. McGeer, J. Andersson, U. Andersson, and M. Kotb. 1996. Evidence for the presence of streptococcal-superantigen-neutralizing antibodies in normal polyspecific immunoglobulin G. Infect. Immun. 64:5395-5398.
17. Shet, A., E. L. Kaplan, D. R. Johnson, and P. P. Cleary. 2003. Immune response to group A streptococcal C5A peptidase in children: implications for vaccine development. J. Infect. Dis. 188:809-817.
18. Sriprakash, K. S., J. Hartas, and A. White. 2002. Antibodies to streptococcal inhibitor of complement function and M peptides in a post-streptococcal glomerulonephritis endemic region of Australia. J. Med. Microbiol. 51:589-594.
19. Stolz, S. J., J. P. Davis, J. M. Vergeront, B. A. Crass, P. J. Chesney, P. J. Wand, and M. S. Bergdoll. 1985. Development of serum antibody to toxic shock toxin among individuals with toxic shock syndrome in Wisconsin. J. Infect. Dis. 151:883-889.
20. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680.
21. Tyrrell, G. J., M. Lovgren, B. Forwick, N. P. Hoe, J. M. Musser, and J. A. Talbot. 2002. M types of group A streptococcal isolates submitted to the National Centre for Streptococcus (Canada) from 1993 to 1999. J. Clin. Microbiol. 40:4466-4471.
22. Vincents, B., U. von Pawel-Rammingen, L. Bjrck, and M. Abrahamson. 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.
23. von Pawel-Rammingen, U., B. P. Johansson, and L. Bjrck. 2002. IdeS, a novel streptococcal cysteine proteinase with unique specificity for immunoglobulin G. EMBO J. 21:1607-1615.
24. von Pawel-Rammingen, U., B. P. Johansson, H. Tapper, and L. Bjrck. 2002. Streptococcus pyogenes and phagocytic killing. Nat. Med. 8:1044-1045.
25. Wenig, K., L. Chatwell, U. von Pawel-Rammingen, L. Bjrck, R. Huber, and P. Sondermann. 2004. Structure of the streptococcal endopeptidase IdeS, a cysteine proteinase with strict specificity for IgG. Proc. Natl. Acad. Sci. USA 101:17371-17376.
26. The Working Group on Severe Streptococcal Infections. 1993. Defining the group A streptococcal toxic shock syndrome. Rationale and consensus definition. JAMA 269:390-391.(Per kesson, Linnea Moritz)
Department of Molecular Biology, Ume University, Ume, Sweden
Department of Community Medicine, Malm University Hospital, Lund University, Malm, Sweden
ABSTRACT
IdeS, a recently discovered cysteine proteinase secreted by the important human pathogen Streptococcus pyogenes, interferes with phagocytic killing by specifically cleaving the heavy chain of immunoglobulin G. The fact that the enzyme targets one of the key molecules of the adapted immune response raised the question of whether an antibody response against IdeS could inhibit, i.e., neutralize, enzyme activity. Paired acute- and convalescent-phase serum samples from patients with pharyngotonsillitis (n = 10), bacteremia (n = 7), and erysipelas (n = 4) were analyzed. Antibodies with the ability to neutralize IdeS enzymatic activity were already found in two-thirds of acute-phase sera. However, patients who seroconverted to IdeS, in particular patients with pharyngotonsillitis and erysipelas, developed specific antibodies during convalescence with an increased capability to efficiently neutralize the enzymatic activity of IdeS. Also, the presence of neutralizing antibodies decreased the ability of IdeS to mediate bacterial survival in human immune blood. In patients with bacteremia, several acute-phase sera contained neutralizing antibodies, but no correlation was found to severity or outcome of invasive infections. Still, the fact that the human immune response targets the enzymatic activity of IdeS supports the view that the enzyme plays an important role during streptococcal infection.
INTRODUCTION
Streptococcus pyogenes is an important human bacterial pathogen that causes a variety of diseases, including pharyngotonsillitis, impetigo, scarlet fever, septicemia, necrotizing fasciitis (NF), and streptococcal toxic shock syndrome (STSS) (6, 7). The survival of S. pyogenes depends on its ability to avoid the various actions of the human immune system. Immunoglobulin G (IgG) plays a key role in the immune defense by specifically recognizing invading microorganisms and mediating their killing by professional phagocytes and the complement system. In order to persist, pathogenic bacteria have to find ways to avoid recognition by immunoglobulins and to interfere with IgFc-mediated phagocytosis. S. pyogenes has evolved a specific enzyme to deal with opsonizing IgG antibodies. This enzyme, designated IdeS or streptococcal Mac-1 (12, 23), is a secreted cysteine proteinase that specifically cleaves the heavy chain of IgG (1, 22, 23). So far, no other substrates for IdeS have been identified, and IgG is the sole substrate of IdeS in plasma samples (22). Due to its early and sustained expression during growth (23) and its highly specific proteolytic activity, IdeS is a tailor-made defense against Fc-mediated phagocytic killing (1, 12, 13, 22-25). Two protein variants of IdeS, complex I and complex II, have been described based on differences in the amino acid sequences of the middle thirds of the proteins (amino acids 112 to 205) (13). Complex II variants have been reported to have weak endopeptidase activity and to interfere with phagocytic killing by blocking the interaction of the FcIIIb receptor with specific antibodies (1, 13), while complex I variants exert their inhibitory function through proteolytic cleavage of IgG (1).
The expression of streptococcal virulence factors in vivo is commonly analyzed by the determination of antibody levels towards particular streptococcal proteins in human blood samples. Paired acute- and convalescent-phase serum samples from children with streptococcal pharyngotonsillitis have been used to study the antibody response against the streptococcal C5a peptidase (17), and acute-phase serum IgG levels towards streptococcal M proteins and pyrogenic exotoxins (Spe's) have been analyzed and correlated to the outcome of disease (4, 10, 18). In another study, acute-phase serum samples of patients suffering from invasive streptococcal infections were analyzed for IgG antibodies towards six streptococcal virulence factors (SclA, SclB, MtsA, Grab, EndoS, and IdeS) (3). Detectable antibody levels against IdeS/Mac were found in this study as well as in two previous studies (12, 13). Other analyses of antistreptococcal antibodies, however, have shown that the total amount of specific antibodies does not necessarily correlate with the level of neutralizing antibodies, and that antibody quality might be clinically more important than antibody quantity (15, 16). The fact that S. pyogenes has evolved an enzyme that specifically targets IgG raises the question of whether specific IgG antibodies are able to neutralize IdeS and the further question of whether the presence of neutralizing antibodies correlates with manifestations of S. pyogenes infection or affects the severity of S. pyogenes-induced disease.
(Part of this work was presented as a scientific poster at the XVIth Lancefield International Symposium on Streptococci and Streptococcal Diseases, 25 to 29 September 2005, Palm Cove, Australia.)
MATERIALS AND METHODS
Patients, serum samples, and bacterial isolates. Serum samples were collected from seven patients treated for S. pyogenes bacteremia at the Clinic for Infectious Diseases, Lund University Hospital, Lund, Sweden. Five of the patients had a nonsevere bacteremia, one patient developed necrotizing fasciitis, and one patient presented with STSS (26). S. pyogenes strains were isolated from blood cultures, and acute-phase serum (days 1 to 4 after onset of symptoms; median, day 1) and convalescent-phase serum (days 18 to 188 after onset; median, day 24) samples were collected from each patient. Sera from 10 patients with pharyngotonsillitis were collected at the Community Health Center Sorgenfrimottagningen, Malm, Sweden. Acute-phase serum samples (taken between days 0 and 4 after onset of symptoms; median, day 1.8) and convalescent-phase serum samples (taken between days 24 and 30 after onset; median, day 26) were collected from each patient. S. pyogenes strains were isolated by throat swab cultures from all patients. Four patients treated for erysipelas at the Clinic for Infectious Diseases, Lund, Sweden, were also included in the study. They had typical signs of a bacterial skin infection, with fever and a rapid spreading of a painful erythema on a lower limb. From these patients, acute-phase sera were collected between days 0 and 5 after onset of symptoms (median, day 2.6), and convalescent-phase serum samples were taken between days 28 and 37 after onset (median, day 31). No bacterial isolate was available from the four erysipelas patients. Acute-phase serum samples from patients with either severe (n = 4) or nonsevere (n = 4) group A streptococcal invasive disease, and with high antibody titers towards IdeS, were from The Netherlands (1994 to 1997) and have been described elsewhere (3, 14).
Immunoglobulins. Human polyclonal IgG and myeloma IgG1() were purchased from Sigma. IgG from patient serum was purified by affinity chromatography using HiTrap protein G Sepharose (Amersham Biosciences) according to the manufacturer's instructions. Myeloma IgG1 was labeled with 125I using the Bolton and Hunter reagent as described by the manufacturer (Amersham Bioscience). Free 125I was separated from labeled protein on a PD10 column (Amersham Bioscience).
Bacterial strains and growth conditions. S. pyogenes strains used in this study are listed in Table 1. Streptococci were routinely grown in Todd-Hewitt broth (TH) (Difco) or TH broth supplemented with 2 to 5% heparinized human plasma at 37°C in 5% CO2. The M serotype of streptococcal strains was determined by sequencing emm gene-specific PCR products as previously described (5). Sequences obtained from emm-specific PCR products were subjected to homology searches in the emm sequence database at the Centers for Disease Control and Prevention (http://www.cdc.gov/ncidod/biotech/strep/emmtypes.htm; last accession, 20 February 2005). The sequence of bases ranging approximately from 65 to 165 obtained with primer 1, TATT(C/G)GCTTAGAAAATTAA, has been implicated mostly as M type specific (5). Sequences that show greater than 95% identity with a reference emm gene sequence were considered to be given emm genes. Allelic forms of IdeS (13) were amplified as previously described (23), and sequences were determined by sequence analysis of nucleotides 121 to 900 of the ideS gene by use of an Applied Biosystems 3100 automated sequencer (Applied Biosystems). Multiple sequence alignment of the internal amino acid sequence of IdeS was conducted with CLUSTAL W (20).
Expression and purification of recombinant IdeS. The expression of recombinant IdeS in Escherichia coli has been described previously (23). Fusion proteins were purified on glutathione-Sepharose (Amersham Biosciences) according to standard protocols. IdeS protein of the complex II family was subcloned from the M8 serotype strain Se4 and purified as described above (for IdeS variants used in this study, see Fig. 2C). IdeS-304 and IdeS-218 were created by PCR with primers I (23) and III (5'-TTCTTTGAATTCTTAAGCAGAAATAGCTACTTTTCC-3') for IdeS-304 and with primers IV (5'-GGTGATCAAAGGATCCTATTGACAAGTC-3') and V (5'-CGGAATTCTTAATTGGTCTGATTCCAAC-3') for IdeS-218 by using IdeS inserted into a pGEX-5X-3 vector as the template (23). PCR fragments were cloned into the corresponding sites of plasmid pGEX-5X-3 (Amersham Biosciences). IdeS-232 and IdeS-47-168 were created by digesting full-length ideS in pGEX5-3 with restriction endonuclease AvaI or StyI (Roche). The digested plasmids were purified, religated, and transformed into E. coli strain NovaBlue (Novagen). All variants were purified as glutathione S-transferase (GST) fusion proteins as described previously (23).
ELISA. Microtiter plates (Maxisorb; NUNC) were coated with enzymatically inactive IdeSC94G (24) at 0.4 μg/ml in coating buffer (0.05 M NaHCO3, pH 9.6). Plates were washed with PBST (0.05% Tween in phosphate-buffered saline [PBS]) and blocked with 2% bovine serum albumin (Sigma) in PBST (PBSTA). Dilutions of each patient and control serum sample in PBSTA were added to the wells. Antigen concentrations and serum dilutions were judged as optimal by checkerboard titration. Bound antibodies were detected by incubation with peroxidase-conjugated antibody against human IgG (1/3000) (Bio-Rad), and enzyme-linked immunosorbent assay (ELISA) plates were developed as previously described (3). Blank samples without serum were included in quadruplicate on each plate. Optical density at 420 nm (OD420) was determined as described previously. An ELISA index was calculated by subtracting OD420 values of blank control samples from the values obtained with serum samples and by dividing the mean OD420 value for each sample by the mean OD420 values for a standard positive serum sample; these were determined in quadruplicate on each plate.
IdeS activity assays. For standard IdeS activity assays in bacterial growth medium, bacteria were grown to an OD620 of 0.4 in TH broth supplemented with 2% heparinized human plasma and 40 μM of the cysteine proteinase inhibitor trans-epoxysuccinyl-L-leucylamido-(4-guanidino) butane (E64; Sigma) to block potential SpeB activity. Cleavage of IgG was determined by analyzing supernatant samples on 12% sodium dodecyl sulfate-polyacrylamide gel electrophoresis (SDS-PAGE) gels, and IdeS activity was determined by the presence or absence of a diagnostic 31-kDa IgG cleavage product (1, 22, 23).
Quantification of IdeS activity. Quantification of the endopeptidase activity of IdeS was done as previously described for microbial IgA proteinases (9). Briefly, various amounts of the enzyme were incubated with 105 cpm of 125I-IgG and 4 μg of unlabeled myeloma IgG1 in PBS. Samples were kept at 37°C for 60 min, and the reactions were stopped by the addition of SDS sample buffer and incubation at 95°C for 5 min. IgG chains and IgG cleavage products were separated on 12% SDS-PAGE gels and visualized by staining with Coomassie blue (R-250; U.S. Biochemicals). Protein bands corresponding to the IgG heavy chain, IgG light chain, and the IdeS cleavage product (CP) were excised from the gel for determination of radioactivity. Samples were counted in an LKB Wallac Compugamma counter, and IdeS activity was calculated as the ratio of radioactivity in CP to total heavy chain radioactivity (CP plus heavy chain).
Detection of neutralizing antibodies in patient serum samples. Generally, IdeS was incubated with 5 to 10 μl of acute-phase or convalescent-phase serum samples at 25°C for 10 min. The reaction volume was adjusted to 28 μl with PBS, and 105 cpm of 125I-IgG was added as the substrate. Incubation was continued at 37°C for 45 min. The reactions were stopped by adding SDS sample buffer, and samples were analyzed by 12% SDS-PAGE. IdeS activity was determined as described above and correlated to enzyme activity in serum depleted of IgG. The IdeS endopeptidase activity was 27.5% cleavage of the IgG heavy chain under these experimental conditions, and this value was arbitrarily set as 100%. Activity in patient serum samples is expressed relative to this standard value.
Bactericidal assay. Purified IdeS (5 μg) was incubated with a three- to fivefold molar excess of either human polyclonal IgG (Sigma) or total patient IgG at 25°C for 30 min. As a control, IdeSC94S (25) was incubated with a similar amount of human polyclonal IgG (Sigma). The suspension was added to 1 ml of heparinized human blood containing antibodies against S. pyogenes, and incubation was continued for 60 min at 37°C with rotation. Approximately 500 CFU of a mid-log-phase culture of strain AP1 (OD620 of 0.15) was used to inoculate the pretreated blood. The tubes were incubated with rotation at 37°C for 60 min, and bacterial survival was monitored by plating dilutions on blood agar plates.
RESULTS
Characterization of bacterial isolates. The M serotypes of the 17 clinical S. pyogenes isolates were determined by emm sequence analysis (5) and are listed in Table 1. Eight of the 11 different serotypes identified are among the most frequently identified M serotypes from clinical isolates (21). PCR analysis demonstrated the presence of the ideS gene in all isolates (data not shown). Strains of the same M serotype generally carried the two previously described (13) ideS alleles. Sequence comparison of the complex II IdeS variants found in the present clinical strains confirmed the reported high degree of sequence homology between these variants (13), with only a few amino acid substitutions. Two changes, however, appear to be of particular interest. (i) Strains So3 and So4 express an IdeS protein that is characterized by a four-amino-acid deletion in positions 243 to 246, distinguishing these proteins from all other IdeS variants. (ii) IdeS of the M28 serotype carries a cysteine residue in position 255. All other proteins, including complex I variants, carry a tyrosine residue at this position (Fig. 1).
IdeS antibody levels in acute-phase and convalescent-phase patient serum samples. IdeS-specific antibodies were detectable in serum samples of all patients (Table 2). Seroconversion was defined as an increase in the ELISA index between acute- and convalescent-phase sera of more than 40%. This was found in 10 of the 21 patients investigated, and in 5 of these the increase exceeded 100% (Table 2). For two patients, PS1 and PS4, high antibody levels towards IdeS had already been detected in acute-phase serum samples.
Specific IgG antibodies towards IdeS inhibit enzymatic activity. In a patient (PS2) who seroconverted to IdeS, no IdeS-mediated cleavage of IgG could be detected in convalescent-phase serum samples containing high levels of anti-IdeS antibodies (Fig. 2A; >90% inhibition of enzyme activity towards myeloma IgG1 in sample PS2C). However, when patient IgG was depleted from the sample and replaced by nonspecific polyclonal IgG, the myeloma IgG1 was efficiently cleaved, and no IdeS inhibitory activity could be detected (Fig. 2A, bar 3; 90% cleavage of myeloma IgG1). The serum regained its neutralizing ability when patient IgG was added back to the sample (Fig. 2A, bar 4; >78% inhibition). Thus, IgG antibodies towards IdeS clearly mediate the neutralization of enzymatic activity in serum samples.
In a second experiment, neutralizing convalescent-phase serum was incubated with an excess of inactive IdeSC94S (25) prior to the addition of active IdeS. The absorption of specific IdeS antibodies to inactive IdeSC94S efficiently abolished the neutralizing activity of convalescent-phase serum samples (Fig. 2B, lane 3; presence of 31-kDa cleavage product). Thus, the neutralizing activity in convalescent-phase serum samples is mediated by the binding of specific IgG antibodies to IdeS. Different enzymatically inactive truncation variants of IdeS (Fig. 2C and data not shown) were used to analyze the binding of neutralizing antibodies (Fig. 2B). Although the IdeS variants have retained the ability to bind IgG (data not shown; 11), only IdeS-304, containing both Cys94 and His262 active-site residues, absorbed neutralizing antibodies and mediated cleavage of IgG by active IdeS (Fig. 2B, GST-IdeS-304; presence of 31-kDa cleavage product). Thus, neutralizing activity in convalescent-phase serum samples is likely to be mediated by the binding of specific IgG antibodies to the catalytic region of IdeS.
Quality of patient antibodies. Studies of antistreptococcal antibodies have suggested that the quality of the antibodies, i.e., the neutralizing activity, may be clinically more relevant than the quantity of antibodies (15, 16). In an attempt to correlate the levels of antibodies towards IdeS with the degree of inactivation of IdeS, patient sera were incubated with IdeS, and the enzymatic activity was determined by measuring the cleavage of radiolabeled 125I-labeled myeloma IgG1. This monoclonal IgG did not inhibit IdeS activity (data not shown). The results for two patients with a marked immune response (>100% increase in ELISA index; PT15 and PS2) and for one patient with only a moderate response (19% increase of ELISA index; PS5) are shown in Fig. 3A to C. Convalescent-phase serum samples from all three patients clearly inhibit IdeS activity (83%, 83%, and 23% inhibition compared to acute-phase serum at a 100-fold serum dilution). The inhibitory capacity of convalescent-phase serum samples diminished upon serum dilution, while IdeS activity remained unaffected in diluted acute-phase serum samples (Fig. 3A to C). Relative enzyme activity levels and the degrees of inhibition were determined for all patient serum samples (Table 3). In sera from patients with pharyngotonsillitis and erysipelas, an increase in the ELISA index correlated well with an increase in enzyme inhibition (Table 3). In contrast, seroconversion was not consistently reflected by an increase of enzyme inhibition when analyzing serum samples from bacteremia patients (Table 3; see data for PS3, PS6, and PS7). Furthermore, two-thirds of the analyzed sample pairs had already shown neutralizing activity in the acute-phase serum samples (Table 3). Still, serum samples from patients that did not seroconvert to IdeS showed no increase in neutralizing activity (Table 3), with the exception of the serum sample pair from patient PS1, which had already exhibited very high antibody titers in the acute-phase sample. It should be noted that the degree of inhibition, determined by analyzing the extent of 125I-labeled myeloma IgG1 cleavage, is most accurate when comparing acute-phase and convalescent-phase serum samples from the same patient. Comparisons of inhibition levels between patients or patient groups are not practicable, as for instance 54.9% enzyme inhibition was sufficient to block patient IgG cleavage in serum sample PT1C, while 71.1% inhibition was not sufficient to block patient IgG cleavage in sample PT22A (data not shown).
Antibody quality and invasive streptococcal disease. Since acute-phase serum samples from sepsis patients notably differ in their abilities to neutralize IdeS activity, we speculated that the quality of preexisting antibodies could be important for the outcome of invasive streptococcal diseases. Acute-phase serum samples from patients with severe invasive infections (n = 4) and nonsevere invasive infections (n = 4), all with high levels of IdeS-specific antibodies (Table 4), were tested for their abilities to inactivate IdeS. Seven out of eight samples neutralized IdeS activity (data not shown), and there was no difference between the groups of samples from severe and nonsevere cases (Table 4). Thus, the presence of neutralizing antibodies towards IdeS does not appear to directly affect the severity of invasive streptococcal disease.
Neutralizing antibodies interfere with IdeS-mediated survival in human blood. IdeS mediates streptococcal survival in whole human immune blood by cleaving opsonizing IgG antibodies (23, 24). Although the presence of neutralizing antibodies towards IdeS does not affect the overall outcome of invasive streptococcal disease, such antibodies still could be detrimental for bacterial survival. The effect of neutralizing antibodies on streptococcal survival was assayed by using a human immune blood bactericidal assay. IdeS was incubated with either polyclonal or neutralizing IgG and added to immune blood prior to inoculation with S. pyogenes as previously described (23). Although the concentration of IdeS used in these experiments exceeded the amount of the enzyme used in the neutralizing experiments (25-fold), streptococcal survival was reduced by 58% in the presence of neutralizing antibodies (Fig. 4). Thus, neutralizing antibodies clearly affected the ability of IdeS to interfere with phagocytic killing of S. pyogenes.
DISCUSSION
S. pyogenes has to employ specific strategies to avoid the action of opsonizing IgG antibodies during infection. The streptococcal cysteine proteinase IdeS represents a tailor-made defense against specific IgG antibodies (1, 12, 13, 22-25) and is widely expressed by clinically important serotypes (e.g., M1, M3, M4, and M12), as well as in newly validated serotypes, e.g., M122 (2) suggesting an important role for IdeS during S. pyogenes infections.
The main focus of this study was to characterize the immune response towards the IgG endopeptidase IdeS in acute-phase and convalescent-phase serum samples from patients suffering from streptococcal disease. In accordance with previous data (3, 12, 13), detectable levels of antibodies against IdeS were found in all samples investigated. The neutralizing abilities of these antibodies varied notably between different patient samples, but consistent results were obtained for patients suffering from pharyngotonsillitis and erysipelas, revealing that patients who seroconverted to IdeS developed IgG antibodies with the ability to neutralize IdeS enzymatic activity. In contrast, patients with no or weak seroconversion have only poor neutralizing capability (Table 3). For patients treated for streptococcal sepsis, less consistent results were obtained, and seroconversion was not necessarily reflected by an increase of enzyme inhibitory activity (see results for PS3, PS6, and PS7). However, IdeS is expressed as one of two distinct protein variants (13), and neutralizing antibodies might be specific for one of the two IdeS variants. This fact might explain the lack of inhibitory activity in samples from patients PS6 and PS7 (see below). For two-thirds of the patients, neutralizing activities could already be detected in acute-phase serum. It appears unlikely that these specific antibodies are the result of a newly acquired IgG immune response this early after onset of infection (median, day 1). Instead, the ability of acute-phase serum to inactivate IdeS most likely reflects a preexisting immune response caused by previous S. pyogenes infections or colonization. This view is consistent with the fact that anti-IdeS antibodies are widely distributed in the healthy population (3).
The finding that neutralizing antibodies against IdeS are present early during infection raised the question of whether these could be of importance for the outcome of invasive streptococcal diseases. This hypothesis was tested by analyzing IdeS-neutralizing activity of acute-phase serum samples from patients with severe or nonsevere invasive streptococcal infections. However, no such correlation between the presence of neutralizing antibodies against IdeS and the severity of invasive disease could be observed, as the capabilities of neutralizing IdeS enzymatic activity were indistinguishable between the different samples (Table 4). Thus, although neutralizing antibodies are involved in the host defense towards streptococci, the presence of antibodies against IdeS appears not to directly affect the outcome of invasive disease. However, the importance of IdeS endopeptidase activity for streptococcal survival in the human host is supported by the finding that neutralizing antibodies interfere with IdeS-mediated streptococcal survival ex vivo (Fig. 4). In contrast to results from experiments addressing the neutralizing capacity of patient IgG, IdeS activity is not completely blocked by patient antibodies, which is due to the relative decrease in the ratio of total patient antibodies present in whole blood to the amount of IdeS added (240- to 400-fold). Still, neutralizing antibodies significantly affected the ability of IdeS to interfere with Fc-mediated phagocytosis and clearly promoted survival of the bacteria in whole blood (Fig. 4).
Pharyngotonsillitis and erysipelas are streptococcal diseases with a high tendency to recur. In recurrent menstruation-associated Staphylococcus aureus toxic shock syndrome, the subsequent development of specific IgG antibodies against toxic shock syndrome toxin 1 protects from recurrent disease (19). The fact that patients develop antibodies with neutralizing activity against IdeS raises the questions of whether these patients are protected from recurrent disease and whether the presence of neutralizing antibodies might protect from recurrent mucocutaneous S. pyogenes infections rather than from invasive disease. This hypothesis is currently under investigation.
Sequence alignment of complex II IdeS variants revealed that strains of the M28 serotype express an IdeS variant with a cysteine residue in position 257 (Fig. 1) (13), while all other variants, including complex I proteins, carry a tyrosine residue in this position (Fig. 1) (13). The role of this second cysteine residue has not yet been investigated, but the recently published three-dimensional structure of IdeS (25) shows that the cysteine in position 257 is part of a flexible loop close to the catalytic site of the enzyme (25). The potential interaction of cysteine 257 and the catalytic site might explain the reported weak enzymatic activity of this variant (1, 13) and indicates that the previous characterization of complex II protein (1, 13) might not generally apply to other complex II variants. This view is supported by the fact that the complex II variants secreted by the clinical streptococcal strains exhibit enzymatic activity. However, neutralizing antibodies might be specific for one of the two IdeS variants, but only 4 of the 17 patient samples were found to contain neutralizing antibodies specific for complex II IdeS, and complex specificity correlated with the IdeS variant expressed by the infecting strain (data not shown).
The finding that an increase in the amount of antibodies does not necessarily correlate with an increased neutralizing capacity also supports the earlier notion (15, 16) that measurements of antibody titers in patient samples are not sufficient to assess the clinical and biological importance of antibodies during bacterial infection. However, the fact that seroconversion towards IdeS leads to the inactivation of enzyme activity strongly supports the view that IdeS IgG endopeptidase activity is an important streptococcal virulence determinant.
ACKNOWLEDGMENTS
E. M. Mascini, Eijkman-Winkler Institute for Microbiology, Utrecht University Hospital, is acknowledged for the generous gift of acute-phase serum samples. We thank Anders O. Persson for technical assistance.
This work was supported by grants from the Swedish Research Council (projects 7480, 13419, 14240, 14767, and 14800), the European Union, project Strep-EURO QLK2-CT-2002-01398, the Royal Physiographic Society in Lund, the Foundations of Crafoord and Wiberg, the University Hospital in Lund, and the SJCKMS Foundation and Insamlingsstiftelsen at Ume University.
The study was approved by the Research Ethics Committee, Lund University. We declare that we have no commercial or other relationship that might lead to a conflict of interest.
The first two authors contributed equally to this work.
REFERENCES
1. Agniswamy, J., B. Lei, J. M. Musser, and P. D. Sun. 2004. Insight of host immune evasion mediated by two variants of group A streptococcus Mac protein. J. Biol. Chem. 279:52789-52796.
2. kesson, P., A. Linder, J. Cronqvist, and B. Christensson. 2004. Group A Streptococcus bacteraemia complicated by osteomyelitis in an immunocompetent adult. Scand. J. Infect. Dis. 36:63-65.
3. kesson, P., M. Rasmussen, E. Mascini, U. von Pawel-Rammingen, R. Janulczyk, M. Collin, A. Olsen, E. Mattsson, M. L. Olsson, L. Bjrck, and B. Christensson. 2004. Low antibody titers to cell wall-attached proteins of Streptococcus pyogenes predispose for severe invasive disease. J. Infect. Dis. 189:797-804.
4. Basma, H., A. Norrby-Teglund, Y. Guedez, A. McGeer, D. E. Low, O. El-Ahmedy, B. Schwartz, and M. Kotb. 1999. Risk factors in the pathogenesis of invasive group A streptococcal infections: role of protective humoral immunity. Infect. Immun. 67:1871-1877.
5. Beall, B., R. Facklam, T. Hoenes, and B. Schwartz. 1997. Survey of emm gene sequences and T-antigen types from systemic Streptococcus pyogenes infection isolates collected in San Francisco, California; Atlanta, Georgia; and Connecticut in 1994 and 1995. J. Clin. Microbiol. 35:1231-1235.
6. Bisno, A. L. 2001. Acute pharyngitis. N. Engl. J. Med. 344:205-211.
7. Cunningham, M. W. 2000. Pathogenesis of group A streptococcal infections. Clin. Microbiol. Rev. 13:470-511.
8. Facklam, R. F., D. R. Martin, M. Lovgren, D. R. Johnson, A. Efstratiou, T. A. Thompson, S. Gowan, P. Kriz, G. J. Tyrrell, E. Kaplan, and B. Beall. 2002. Extension of the Lancefield classification for group A streptococci by addition of 22 new M protein gene sequence types from clinical isolates: emm103 to emm124. Clin. Infect. Dis. 34:28-38.
9. Gilbert, J. V., A. G. Plaut, B. Longmaid, and M. E. Lamm. 1983. Inhibition of microbial IgA proteases by human secretory IgA and serum. Mol. Immunol. 20:1039-1049.
10. Holm, S. E., A. Norrby, A. M. Bergholm, and M. Norgren. 1992. Aspects of pathogenesis of serious group A streptococcal infections in Sweden, 1988-1989. J. Infect. Dis. 166:31-37.
11. Kawabata, S., Y. Tamura, J. Murakami, Y. Terao, I. Nakagawa, and S. Hamada. 2002. A novel, anchorless streptococcal surface protein that binds to human immunoglobulins. Biochem. Biophys. Res. Commun. 296:1329-1333.
12. Lei, B., F. R. DeLeo, N. P. Hoe, M. R. Graham, S. M. Mackie, R. L. Cole, M. Liu, H. R. Hill, D. E. Low, M. J. Federle, J. R. Scott, and J. M. Musser. 2001. Evasion of human innate and acquired immunity by a bacterial homolog of CD11b that inhibits opsonophagocytosis. Nat. Med. 7:1298-1305.
13. Lei, B., F. R. DeLeo, S. D. Reid, J. M. Voyich, L. Magoun, M. Liu, K. R. Braughton, S. Ricklefs, N. P. Hoe, R. L. Cole, J. M. Leong, and J. M. Musser. 2002. Opsonophagocytosis-inhibiting Mac protein of group A streptococcus: identification and characteristics of two genetic complexes. Infect. Immun. 70:6880-6890.
14. Mascini, E. M., M. Jansze, J. F. Schellekens, J. M. Musser, J. A. Faber, L. A. Verhoef-Verhage, L. Schouls, W. J. van Leeuwen, J. Verhoef, and H. van Dijk. 2000. Invasive group A streptococcal disease in the Netherlands: evidence for a protective role of anti-exotoxin antibodies. J. Infect. Dis. 181:631-638.
15. Norrby-Teglund, A., K. Pauksens, S. E. Holm, and M. Norgren. 1994. Relation between low capacity of human serum to inhibit streptococcal mitogens and serious manifestation of disease. J. Infect. Dis. 170:585-591.
16. Norrby-Teglund, A., R. Kaul, D. E. Low, A. McGeer, J. Andersson, U. Andersson, and M. Kotb. 1996. Evidence for the presence of streptococcal-superantigen-neutralizing antibodies in normal polyspecific immunoglobulin G. Infect. Immun. 64:5395-5398.
17. Shet, A., E. L. Kaplan, D. R. Johnson, and P. P. Cleary. 2003. Immune response to group A streptococcal C5A peptidase in children: implications for vaccine development. J. Infect. Dis. 188:809-817.
18. Sriprakash, K. S., J. Hartas, and A. White. 2002. Antibodies to streptococcal inhibitor of complement function and M peptides in a post-streptococcal glomerulonephritis endemic region of Australia. J. Med. Microbiol. 51:589-594.
19. Stolz, S. J., J. P. Davis, J. M. Vergeront, B. A. Crass, P. J. Chesney, P. J. Wand, and M. S. Bergdoll. 1985. Development of serum antibody to toxic shock toxin among individuals with toxic shock syndrome in Wisconsin. J. Infect. Dis. 151:883-889.
20. Thompson, J. D., D. G. Higgins, and T. J. Gibson. 1994. CLUSTAL W: improving the sensitivity of progressive multiple sequence alignment through sequence weighting, position-specific gap penalties and weight matrix choice. Nucleic Acids Res. 22:4673-4680.
21. Tyrrell, G. J., M. Lovgren, B. Forwick, N. P. Hoe, J. M. Musser, and J. A. Talbot. 2002. M types of group A streptococcal isolates submitted to the National Centre for Streptococcus (Canada) from 1993 to 1999. J. Clin. Microbiol. 40:4466-4471.
22. Vincents, B., U. von Pawel-Rammingen, L. Bjrck, and M. Abrahamson. 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.
23. von Pawel-Rammingen, U., B. P. Johansson, and L. Bjrck. 2002. IdeS, a novel streptococcal cysteine proteinase with unique specificity for immunoglobulin G. EMBO J. 21:1607-1615.
24. von Pawel-Rammingen, U., B. P. Johansson, H. Tapper, and L. Bjrck. 2002. Streptococcus pyogenes and phagocytic killing. Nat. Med. 8:1044-1045.
25. Wenig, K., L. Chatwell, U. von Pawel-Rammingen, L. Bjrck, R. Huber, and P. Sondermann. 2004. Structure of the streptococcal endopeptidase IdeS, a cysteine proteinase with strict specificity for IgG. Proc. Natl. Acad. Sci. USA 101:17371-17376.
26. The Working Group on Severe Streptococcal Infections. 1993. Defining the group A streptococcal toxic shock syndrome. Rationale and consensus definition. JAMA 269:390-391.(Per kesson, Linnea Moritz)