点击显示 收起
Sections of Experimental Medicine and Toxicology and Proteomics, Departments of Infectious Diseases and Histopathology
National Heart and Lung Institute, Faculty of Medicine, Imperial College London, London, United Kingdom
Lethal necrotizing fasciitis caused by Streptococcus pyogenes is characterized by a paucity of neutrophils at the site of infection. Interleukin (IL)8, which is important for neutrophil transmigration and activation, can be degraded by S. pyogenes. Blood isolates of S. pyogenes were better able to degrade human IL-8 than throat isolates. Degradation of IL-8 was the result of a single specific cleavage between 59glutamine and 60arginine within the IL-8 C-terminal helix. Cleaved IL-8 reduced neutrophil activation and migration. IL-8cleaving activity was found in partially purified supernatant of a necrotizing fasciitis isolate, and this activity was associated with an 150-kDa fraction containing S. pyogenes cell envelope proteinase (SpyCEP). IL-8cleaving activity corresponded with the presence of SpyCEP in the supernatant. Cleavage of IL-8 by S. pyogenes represents an unprecedented mechanism of immune evasion, effectively preventing IL-8 C-terminusmediated endothelial translocation and subsequent recruitment of neutrophils.
Streptococcus pyogenes causes a spectrum of disease, ranging from pharyngitis to necrotizing soft tissue infections to toxic shock. Lethal necrotizing infections are characterized by an absence of neutrophils at the site of bacterial growth [1, 2]. Interleukin (IL)8 (or CXCL8) is a major chemokine responsible for recruitment of neutrophils to sites of injury or infection. IL-8 is produced in interstitial tissues by epithelial cells, monocytes/macrophages, and fibroblasts as a 72-residue polypeptide that passes from the abluminal to the luminal endothelial surface through interactions between heparan sulfates and the C-terminus of IL-8. IL-8 is presented to neutrophils on the luminal endothelial surface, is anchored by interactions between the C-terminus and glycosaminoglycans, and can induce neutrophil slowing and arrest, which are essential for subsequent neutrophil transmigration [3]. IL-8 activates neutrophils through interaction with the 7 transmembrane domainspanning CXCR1 or CXCR2 cell surface receptor. The complex nature of the CXCRs has hindered full analysis of receptor-ligand interactions, although the IL-8 N-terminus seems to be important [4, 5].
Although several purified S. pyogenes products stimulate production of IL-8 by mononuclear cells [6], IL-8 production is remarkably poor when cells are stimulated with whole S. pyogenes supernatants [7]. Here, we demonstrate that IL-8 is inactivated by S. pyogenes supernatants and demonstrate that inactivation occurs through a specific cleavage at the C-terminus between positions 59Q and 60R of the mature IL-8 polypeptide. This represents a novel and highly specific mechanism by which S. pyogenes can combat the innate immune response to infection.
MATERIALS AND METHODS
Reagents and ELISA.
Carrier-free recombinant human IL-8, murine macrophage inflammatory protein (MIP)2, and other cytokines were obtained from R&D Systems. Recombinant human C5a was obtained from Calbiochem. Monoclonal and polyclonal biotinylated antibodies to human IL-8 were obtained from R&D Systems and Biosource. Chemokines and cytokines were assayed by ELISA using paired antibodies in accordance with the manufacturer's protocols (R&D Systems). All samples were assayed in duplicate, and standards were diluted in buffers that were appropriate to test samples. Protease inhibitors were from Roche Diagnostics.
Bacterial supernatants.
Blood and throat isolates of S. pyogenes were sequentially collected during November 1994March 2000 at the Hammersmith Hospital, using a protocol that was approved by the local research ethics committee. Isolates were frozen as glycerol stocks and were passaged only once. Blood isolates were emm typed. Bacteria from single colonies were cultured in Todd Hewitt broth (TH; Oxoid) or RPMI 1640 with 10% fetal calf serum and 2 mmol/L glutamine (Life Technologies) for 18 h at 37°C without shaking. There was no difference in growth between blood isolates (median, 0.75 × 108 [range, 0.3 × 1081.07 × 108] cfu/mL) and throat isolates (median, 1.07 × 108 [range, 0.41 × 1081.48 × 108] cfu/mL). Cell-free supernatants were obtained by centrifugation (at 800 g) and filtration using 0.2 mol/L filters (Sartorius AG).
IL-8 degradation.
To compare isolates, S. pyogenesTH supernatants (1 : 50 dilution) were incubated with 2 ng/mL IL-8 for 18 h at 37°C. S. pyogenes strain H292TH supernatant (1 : 50) was incubated with 1 ng/mL IL-8 for 2, 4, 6, 8, and 24 h at 37°C, before measurement of residual IL-8 by ELISA. Degrading activity was expressed as nanograms of IL-8 digested per hour per microliter of H292 supernatant. For in vitro experiments using human neutrophils, IL-8 was cleaved with H292-RMPI supernatant (1 : 50). For some experiments, IL-8 was incubated with a purified H292 enzyme preparation (described below).
SDS-PAGE and Western blotting.
Proteins were separated by SDS-PAGE using 17.5% (for chemokines) or 10% gels and were stained with either silver or Coomassie blue. Separated proteins were immunoblotted using biotinylated antiIL-8 antibodies and streptavidin-conjugated horseradish peroxidase and were visualized with the ECL system (Amersham Pharmacia Biotech).
Peptide and protein identification.
Protein identification was performed after reduction, carbamidomethylation, and digestion with trypsin. Peptides were separated on a Waters CapLC chromatograph using an LC Packings Nan 75-15-OS-C18-PM column eluting at 200 nL/min with a linear gradient of acetonitrile : water : formic acid (5 : 95 : 0.1 to 95 : 5 : 0.1). During liquid chromatography (LC), the flow was directed into the positive ion nanospray ion source of a Waters QtoF-2 mass spectrometer operated in the survey mode. Proteins were identified using the Mascot search engine (Matrix Science) [8]. IL-8, MIP-2, C5a, and fragments were chromatographed on a 5-m Vydac C4MS column (0.3 × 5 mm) eluting at 5 L/min with a 30-min linear gradient of acetonitrile : water : formic acid (5 : 95 : 0.1 to 60 : 40 : 0.1). The LC flow was directed into the nanospray source of the QtoF-2 mass spectrometer for analysis by positive ion electrospray ionization mass spectrometry (ESI-MS).
Neutrophil response to cleaved IL-8.
IL-8 was cleaved with 1 : 50 H292-RPMI supernatant overnight at 37°C; cleavage was confirmed by SDS-PAGE. The controls were IL-8 and H292-RPMI supernatant alone. Purified H292 enzyme preparation was used for in vivo studies. Neutrophils were purified from peripheral blood of volunteers by Percoll density gradient centrifugation with dextran sedimentation and were resuspended at a density of 2 × 106 cells/mL in RPMI 1640. Neutrophils (80 L) were coincubated for 30 min at 37°C with 80 L of sample (IL-8, cleaved IL-8, or H292-RPMI supernatant) and then were stained with phycoerythrin-labeled antiL-selectin (CD62L, clone Dreg 56; eBiosciences) or nonbinding isotype control murine IgG1 (eBiosciences). A FACScan flow cytometer (BD Biosciences) was used to collect 10,000 events/tube. Results were expressed as the change in mean fluorescence intensity (MFI), compared with that in unstimulated cells. Control antibody binding was insignificant under all conditions. To examine migration across 5-m transwells (Corning), 1 × 106 neutrophils were incubated in the upper chamber, and 80 ng/mL IL-8 was placed in the lower 0.5-mL chamber; transmigrated neutrophils were counted after 2h, and the number was compared with the number in wells containing cleaved IL-8 or H292-RPMI supernatant. For examination of in vivo neutrophil migration, female C57BL/6 mice received intraperitoneally 300 L of sterile saline containing either 1 g of IL-8, 1 g of cleaved IL-8, or 2 L of diluted H292 purified enzyme preparation. Control mice received no stimulus. Mice were killed after 3 h, and peritoneal lavage was performed using 6 mL of saline containing 10% fetal calf serum and 10 U/mL heparin. Total cell counts were performed on 500 L of cells, and differential counts were performed by counting a minimum of 200 cells after cytospins and staining (Diff-Quik; Dade Behring). Mouse studies were performed in accordance with UK Home Office guidelines.
Characterization of IL-8 protease.
H292-TH supernatant aliquots were removed at 45-min intervals after a 1% inoculation of broth and were incubated separately with 3 ng/mL IL-8; residual IL-8 was measured by ELISA. For initial size fractionation, H292-TH supernatants were centrifuged using Centricon 10-kDa, 30-kDa, 50-kDa, and 100-kDa filters (Millipore). IL-8 degradation was evaluated in the presence of 50 g/mL antipain, 40 g/mL bestatin, 30 g/mL chymostatin, 10 g/mL E64, 5 g/mL leupeptin, 1 g/mL pepstatin, 200 g/mL phosphoromidon, 1 mg/mL Pefabloc, 1 mmol/L EDTA, 10 g/mL aprotonin, and 3 mmol/L phenylmethylsulfonyl fluoride (PMSF). H292 supernatant (1 : 5 dilution) was incubated with other cytokines (0.1 ng/mL and 1 ng/mL tumor necrosis factor , IL-1, interferon , IL-6, or IL-8) in triplicate for 18 h at 37°C, and residual cytokines were measured by ELISA. In separate experiments, 25 pmol of IL-8, MIP-2, or C5a was incubated with 0.041 g of the partially purified H292 enzyme preparation in 75 L of ammonium bicarbonate (100 mmol/L; pH 7.5) for 1 or 18 h. Incubates were analyzed for degradation products by LC-MS.
Purification of IL-8 protease activity.
Eleven liters of H292 was cultured in TH for 18 h at 37°C without shaking. Bacteria were removed by centrifugation, and the supernatant was passed through a 0.2 mol/L filter. Proteins were precipitated at 4°C by incubation with 80% saturated ammonium sulfate. The precipitate was collected by centrifugation, resuspended in 600 mL of PBS, and concentrated by pressure filtration over a 30-kDa cutoff membrane (Sartorius) to a final volume of 75 mL. The concentrate (10 mL) was applied at 2 mL/min to a Sephadex G75 column (18 × 300 mm; Amersham Pharmacia Biotech) equilibrated in 20 mmol/L Tris-HCl (pH 9.0). Protein eluted in the void volume was collected, pooled, and concentrated over a 30-kDa cutoff membrane to a volume of 12 mL and then was subjected to anion exchange chromatography on a Q-fast flow column (15 × 100 mm; Amersham Pharmacia Biotech) at 2 mL/min using a step gradient of increasing concentrations of NaCl (0, 0.1, 0.2, 0.3, 0.5, and 1.0 mol/L). Activity was expressed as nanograms of IL-8 digested per hour per microliter of enzyme preparation. Active fractions were pooled (50 mL) and concentrated to 1.5 mL, and volumes of 0.2 mL were subjected to gel filtration at 0.1 mL/min using a Superose 12 column collecting fractions of 0.5 mL. At each step, protease activity was measured by ELISA, and the complexity of the sample was determined by SDS-PAGE. Gel filtration fractions were then subjected to protein identification.
Antibody to S. pyogenes cell envelope proteinase (SpyCEP).
The peptide FDPETNRFKPEPLKDRG, corresponding to residues 10281044 of the M1 SpyCEP sequence, was coupled to keyhole limpet hemocyanin at the N-terminus and polyclonal antibody raised in rabbits [9]. A total of 15 L of 20× concentrated supernatant from strains that exhibited the strongest (residual IL-8 concentration, <0.1 ng/mL) and the weakest (residual IL-8 concentration, >0.8 ng/mL) IL-8degrading activity were separated by SDS-PAGE (n = 6 strains/group) and were subjected to Western blotting using 1 : 500 diluted antiSpyCEP and to development with ProteinG-peroxidase (Sigma) and the ECL system.
RESULTS
Degradation of IL-8 by invasive S. pyogenes isolates.
S. pyogenes blood isolates were significantly better at inactivating IL-8 than were noninvasive throat isolates (figure 1A). One isolate, strain H292 (emm81), which was cultured from the blood and muscle of a patient who died of necrotizing fasciitis, was chosen for further study, because it was the most potent degrader. Histological examination of muscle sections from this patient revealed that, despite the presence of bacteria, there was limited evidence of neutrophil infiltration at the site of infection (figure 1B). Degradation of IL-8 could be detected after coincubation with H292-TH supernatant for 2 h and was complete by 8 h of coincubation, which corresponds to an IL-8 degradation rate of 0.01ng/h/L of H292 supernatant. Degradation was specific for IL-8, because H292 supernatant had no detectable effect on the immunoreactivity of TNF-, IL-1, IFN-, or IL-6 (data not shown).
IL-8 cleavage at the C-terminus.
Analysis of IL-8 that was inactivated by H292-TH supernatant showed that the IL-8 band (mass, 8 kDa) disappeared, yielding a single cleavage product with a mass of 6 kDa (figure 2A). Cleaved IL-8 was recognized very weakly by polyclonal antiIL-8 antibody (figure 2B) but not by any monoclonal antibody used. IL-8 was incubated with H292 purified enzyme preparation (see below), and the product, together with control IL-8, was subjected to high-performance liquid chromatographyMS analysis. IL-8 was eluted after 10.8 min of flow and generated an intense electrospray mass spectrum with multiply charged ions centered at mass-to-charge (m/z) ratios of 763.0, 839.2, 932.3, 1048.7, 1198.3, and 1397.7. Deconvoluting the spectra gave an average molecular weight (MAv) of 8381.7 ± 0.5, which corresponds well to the calculated MAv of 8381.8 for IL-8. Incubation of 25 pmol of IL-8 with the H292 purified enzyme preparation (1 g for 18 h) resulted in complete conversion of IL-8 to a new peptide that had an elution time of 10.2 min and generated a mass spectrum of ions with m/z ratios of 759.3, 854.0, 975.8, 1138.3, and 1365.7, corresponding to an MAv of 6823.9 ± 0.4 (figure 3A and B). After reduction and treatment with iodoacetamide, the fragment generated a new spectrum of ions with m/z ratios of 642.4, 706.5, 785.0, 882.9, 1009.0, and 1177.0, corresponding to an MAv of 7056.2 ± 0.3; this is equivalent to the reduction and carbamidomethylation of 2 S-S bridges and demonstrates that both S-S bridges were retained in the cleavage product. These data are consistent with a complete IL-8 cleavage between 59Q and 60R and the loss of the terminal 13 residues to form 159IL-8, on the basis of the 72-residue mature IL-8 (figure 3C and 3D). The other cleavage product, 6072IL-8, was eluted after 3 min of flow and generated a spectrum of ions with m/z ratios of 526.0 (3+) and 788.5 (2+), confirming that the cleavage of IL-8 represented a single cleavage process. The murine CXC chemokine MIP-2 was cleaved by H292 purified enzyme preparation in the analogous position between 60Q and 61K to form 160MIP-2 (MAv, 6395.5) and the C-terminal fragment 6173MIP-2 (MAv, 1467.8). Incubation of 25 pmol of IL-8 with 0.4 g of H292 purified enzyme preparation cleaved >95% of IL-8 but only 15% of MIP-2 in 1 h. Reducing the amount of H292 purified enzyme preparation to 0.04 g resulted in 20% cleavage of IL-8 and <5% cleavage of MIP-2. Therefore, MIP-2 was 10-fold more resistant than IL-8 to cleavage by H292 purified enzyme preparation. The complement component C5a, which has a Q-R bond, was unaffected by H292 purified enzyme preparation under all conditions examined.
Diminished biological activity of cleaved IL-8.
IL-8 induced dose-dependent neutrophil CD62L (L-selectin) shedding, whereas cleaved IL-8 was unable to induce CD62L shedding at any concentration (figure 4A). Pefabloc was able to completely reverse the effects of H292-RPMI supernatant on IL-8induced CD62L shedding (data not shown). The migration of human neutrophils in response to IL-8 was abrogated in response to cleaved IL-8 (figure 4B). H292-RPMI supernatant did not exert any nonspecific effect on neutrophil migration, because it did not inhibit neutrophil migration in response to C5a (data not shown). Injection of full-length IL-8 induced a significant increase in murine peritoneal neutrophils, compared with the number seen in controls, whereas cleaved IL-8 could not elicit a neutrophil response (figure 4C). Thus, inactivation of IL-8 by H292 resulted in a failure to recruit neutrophils and, in addition, reduced the activation status of the neutrophils.
Characterization of IL-8 protease.
IL-8degrading activity appeared in H292-TH supernatant at the start of exponential growth, was maximal by early-midlog phase, and remained detectable throughout late log and stationary phase (figure 5A). IL-8degrading activity was heat sensitive (30 min at 80°C) and was limited to fractions retained by a >100-kDa filter. Activity was inhibited by the serine protease inhibitor Pefabloc but was unaffected by a wide range of protease inhibitors, including other serine protease inhibitors, such as PMSF (figure 5B). Pefabloc is more stable than PMSF in aqueous solutions and is known to inactivate some serine proteases faster than PMSF does. Pefabloc, but not other protease inhibitors, was also able to inhibit degradation of IL-8 by 4 other S. pyogenes blood isolates. Pefabloc did not adduct to IL-8; analysis by ESI-MS demonstrated that incubation with IL-8 (2 : 1 molar ratio) for 2 h did not alter the molecular weight of IL-8 (MAv, 8382). Dialysis of Pefabloc-treated culture supernatant did not restore protease activity, which indicates that the inactivation was irreversible.
The protein responsible for IL-8inactivating activity was purified from H292-TH supernatant by ammonium sulfate precipitation, anion exchange chromatography, and Superose 12 gel filtration as a 120150-kDa protein (figure 6A and 6B). Enzyme purification corresponded to a 4000-fold increase in volume activity, compared with that in crude H292-TH supernatant. The 2 active (IL-8cleaving) fractions contained SpyCEP (Mascot score, >1200), peptides covered a 154-kDa range from residues 183 to 1571, and SDS-PAGE confirmed the mass. Peptides from this protein were not found in any inactive fractions. The only additional peptides that were common to both active fractions were from UDP-N-acetylglucosamine pyrophosphorylase, glyceraldehyde-3-phosphate dehydrogenase, and a putative tyrosine tRNA ligase, none of which is a proteolytic enzyme. Other peptides were identified, but in only 1 of the active fractions rather than in both; these were from chaperone protein dnaK, triosephosphate isomerase, a putative peroxide resistance protein, and a putative cysteine aminopeptidase C, none of which is predicted to have endopeptidase activity.
Polyclonal antibody raised against SpyCEP peptide recognized 2 different 120150-kDa bands in supernatant from the most potent IL-8degrading S. pyogenes strains but not from strains that failed to degrade IL-8, which demonstrates that the IL-8degrading phenotype corresponds to the presence of SpyCEP protein in the supernatant (figure 6C).
DISCUSSION
S. pyogenes has evolved an array of mechanisms to combat phagocytosis, including C5a peptidase activity, complement inhibitory activity, and immunoglobulin binding and cleavage strategies [1012]. The C-terminal cleavage of IL-8 by S. pyogenes represents an additional major mechanism that may enhance this pathogen's ability to evade the innate immune response. C-terminal cleavage of chemokines is unusual, because most chemokines demonstrate susceptibility to cleavage at the N-terminus. Indeed, endothelial IL-8 is subjected to rapid N-terminal cleavage, which renders it more active [13]. Cleavage of the IL-8 C-terminus by S. pyogenes is predicted to undermine the uptake, transit, and presentation of IL-8 to circulating neutrophils [3] and would prevent the interaction of IL-8 with glycosaminoglycan [14], an evolutionary strategy not previously demonstrated by other pathogens. Chemically synthesized analogues of IL-8 that lack the C-terminus demonstrate 50-fold reduction in potency of IL-8induced neutrophil elastase release and chemotaxis; the present work extends those findings in a novel and biologically relevant setting [5].
The IL-8 59Q-60R cleavage site resides within the C-terminal helix of IL-8. IL-8 can exist as a dimer in solution, and the proposed dimeric structure indicates a role for the C-terminal helix in this association [15, 16]. The oligomerization state of IL-8 necessary for function in vivo is unknown and may depend on local IL-8 concentration at the endothelial surface. Removal of the IL-8 C-terminal helix is likely to disrupt interactions between IL-8 molecules, and future studies will explore whether this precludes formation of the dimer and subsequent biological effects. Interestingly, loss of IL-8 immunoreactivity in ELISA and Western blotting was due to loss of the C-terminus of IL-8, and this suggests that available antibodies to IL-8 bind mostly to this region.
MIP-2, the murine homologue of IL-8, was subject to a single cleavage by S. pyogenes in the analogous position, at 60Q-61K, whereas several other Q-K bonds within MIP-2 were preserved. Interestingly, MIP-2 was more resistant to cleavage than IL-8, and this may reflect species-specific preferential cleavage of the Q-R sequence typical of primate but not rodent CXC chemokines. Cleavage of CXC chemokines by S. pyogenes is therefore likely to be dependent on both specific residues and position.
Blood isolates of S. pyogenes demonstrated greater IL-8cleaving activity than throat isolates, although this distinction was not absolute, and we cannot exclude IL-8cleaving activity at the bacterial surface. Alteration in virulence of S. pyogenes isolates that have entered the blood is well recognized and may be coordinated by several global regulators [17]. IL-8cleaving activity could not be attributed to specific emm sequence types, because the IL-8cleaving phenotype varied between strains of the same emm type (data not shown).
Chemokine degradation by S. pyogenes has been noted by Hidalgo-Grass et al. [18], although the mechanism by which this occurs and the causative enzyme were not reported. Hidalgo-Grass et al. reported that administration of a peptide (silCR) could attenuate virulence of an M14 S. pyogenes infection in a mouse model and attributed this attenuation to inhibition of MIP-2 degradation. They postulated that the lack of a functional sil locus in certain emm types may confer a virulent phenotype to S. pyogenes, although disruption of the sil locus was reported elsewhere to attenuate virulence of the same M14 strain in mice [19]. The observations reported by Hidalgo-Grass [18] differ from our findings, which show that the IL-8cleaving phenotype varies within single emm types, that MIP-2 is relatively resistant to cleavage, and that the causative protease is inhibited by Pefabloc only.
Our study showed that SpyCEP is the most likely candidate for the IL-8cleaving enzyme. Confirmation of the full role that SpyCEP plays in IL-8 cleavage and necrotizing infection will require mutagenesis of a transformable strain, ideally using IL-8 transgenic mice. Western blotting demonstrated that the phenotypic difference between IL-8degrading isolates and IL-8nondegrading isolates is closely associated with a difference in detectable levels of SpyCEP. The gene prtS, which encodes SpyCEP, is present in all sequenced S. pyogenes strains but is subject to allelic variation. Studies are underway to determine whether genotypic differences underlie the observed phenotypic variation or whether the differences relate solely to transcriptional/posttranslational regulation.
SpyCEP has a predicted N-terminal signal sequence, 3 subtilisase domains, and a peptidoglycan LPXTG (LPXAG) anchoring motif. SpyCEP is similar to other gram-positive CEPs [20], C5a peptidase, and subtilisins of Bacillus species. Recombinant expression of SpyCEP is complicated by the size and complexity of its multidomain structure but will enable structure-function analysis. Demonstration of the 130150-kDa SpyCEP is novel, although a 31-kDa fragment of the same protein was recently identified in an M3 S. pyogenes supernatant [21]. Despite having an LPXAG motif, soluble SpyCEP was released into the supernatant, either through reduced efficiency of the sortase system or through the actions of streptococcal extracellular enzymes. Studies with an M49 speb mutant show that the cysteine protease (SPEB) is unlikely to contribute (data not shown). The precise structure of SpyCEP isoforms detected by Western blotting are unknown, although they were not recognized by antibody to a C-terminus SpyCEP peptide, which confirms that the released SpyCEP is C-terminus truncated.
In summary, S. pyogenes can cleave the C-terminus of human IL-8 at 59Q-60R to yield a fragment, 159IL-8, that lacks the necessary biological activity to promote neutrophil transmigration to infected tissues. This represents a novel and highly specific enzymatic process by which invasive pathogens can evade the human innate immune response and may contribute to the pathogenesis of necrotizing group A streptococcal disease.
Acknowledgments
We thank Karen McGregor, for emm sequencing, and Andreas Podbielski, for the M49 speb cysteine protease mutant.
References
1. Cockerill FR, Thompson RL, Musser JM, et al. Molecular, serological, and clinical features of 16 consecutive cases of invasive streptococcal disease. Clin Infect Dis 1998; 26:144858. First citation in article
2. Taylor FB, Bryant AE, Blick KE, et al. Staging of the baboon response to group A streptococci administered intramuscularly: a descriptive study of the clinical symptoms and clinical chemical response patterns. Clin Infect Dis 1999; 29:16777. First citation in article
3. Middleton J, Neil S, Wintle J, et al. Transcytosis and surface presentation of IL-8 by venular endothelial cells. Cell 1997; 91:38595. First citation in article
4. Skelton NJ, Quan C, Reilly D, Lowman H. Structure of a CXC chemokine-receptor fragment in complex with interleukin-8. Structure Fold Des 1999; 7:15768. First citation in article
5. Clark-Lewis I, Schumacher C, Baggiolini M, Moser B. Structure-activity relationships of interleukin-8 determined using chemically synthesized analogues. J Biol Chem 1991; 266:2312834. First citation in article
6. Muller-Alouf H, Alouf JE, Gerlach D, Ozegowski J-H, Fitting C, Cavaillon J-M. Comparative study of cytokine release by human peripheral blood mononuclear cells stimulated with Streptococcus pyogenes superantigenic erythrogenic toxins, heat-killed streptococci, and lipopolysaccharide. Infect Immun 1994; 62:491521. First citation in article
7. Soto A, McWhinney PHM, Kibbler CC, Cohen J. Cytokine release and mitogenic activity in the viridans streptococcal shock syndrome. Cytokine 1998; 10:3706. First citation in article
8. Perkins DN, Pappin DJ, Creasy DM, Cottrell JS. Probability-based protein identification by searching sequence databases using mass spectrometry data. Electrophoresis 1999; 20:355167. First citation in article
9. Edwards RJ. Targeting antipeptide antibodies towards cytochrome P450 enzymes. Methods Mol Biol 1998; 107:23949. First citation in article
10. Cunningham MW. Pathogenesis of group A streptococcal infection. Clin Microbiol Rev 2000; 13:470511. First citation in article
11. Collin M, Olsen A. EndoS, a novel secreted protein from Streptococcus pyogenes with endoglycosidase activity on human IgG. EMBO J 2001; 20:304655. First citation in article
12. von Pawel-Rammingen U, Johansson BP, Bjorck L. IdeS, a novel streptococcal cysteine proteinase with unique specificity for immunoglobulin G. EMBO J 2002; 21:160715. First citation in article
13. Mukaida N. Pathophysiological roles of interleukin-8/CXCL8 in pulmonary diseases. Am J Physiol Lung Cell Mol Physiol 2003; 284:L56677. First citation in article
14. Webb LMC, Ehrengruber MU, Clark-Lewis I, Baggiolini M, Rot A. Binding to heparan sulfate or heparin enhances neutrophil responses to interleukin 8. Proc Natl Acad Sci USA 1993; 90:715862. First citation in article
15. Baldwin ET, Weber IT, St. Charles R, et al. Crystal structure of interleukin 8: symbiosis of NMR and crystallography. Proc Natl Acad Sci USA 1991; 88:5026. First citation in article
16. Clore GM, Appella E, Yamada M, Matsushima K, Gronenborn AM. Three-dimensional structure of interleukin 8 in solution. Biochemistry 1990; 29:168996. First citation in article
17. Vicente M, Chater KF, de Lorenzo V. Bacterial transcription factors involved in global regulation. Mol Microbiol 1999; 33:817. First citation in article
18. Hidalgo-Grass C, Dan-Goor M, Maly A, et al. Effect of a bacterial pheromone peptide on host chemokine degradation in group A streptococcal necrotising soft-tissue infections. Lancet 2004; 363:696703. First citation in article
19. Hidalgo-Grass C, Ravins M, Dan-Goor M, Jaffe J, Moses AE, Hanski E. A locus of group A Streptococcus involved in invasive disease and DNA transfer. Mol Microbiol 2002; 46:8799. First citation in article
20. Fernandez-Espla MD, Garault P, Monnet V, Rul F. Streptococcus thermophilus cell wall-anchored proteinase: release, purification and biochemical charcterization. Appl Environ Microbiol 2000; 66:47728. First citation in article
21. Lei B, Mackie S, Lukomski S, Musser JM. Identification and immunogenicity of group A Streptococcus culture supernatant proteins. Infect Immun 2000; 68:680718. First citation in article