| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
1 Department of Clinical Sciences Lund, Division of Infection Medicine, Lund University, BMC, B14, SE-221 84 Lund, Sweden
2 Helmholtz Zentrum für Infektionsforschung, Abteilung Mikrobielle Pathogenität, Inhoffenstrasse 7, D-381 24 Braunschweig, Germany
3 Department of Clinical Sciences Lund, Division of Respiratory Medicine and Allergology, Lund University, BMC, B14, SE-221 84 Lund, Sweden
Correspondence
Helena M. Linge
Helena_M.Johansson{at}med.lu.se
 |
ABSTRACT |
|---|
 TOP ABSTRACT INTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
|
INTRODUCTION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
; Honore et al., 1996; Sylvetsky et al., 2002). GGS can cause a spectrum of illnesses, ranging from superficial, e.g. pharyngitis and erysipelas, to deep severe infections, e.g. necrotizing fasciitis and sepsis. Although being a less common cause of disease, the clinical manifestations are sometimes shared with the better-known group A streptococci (GAS; Streptococcus pyogenes) (Cunningham, 2000). GGS infections may be less severe and also mainly found in elderly and immunocompromised individuals (Bowler et al., 2001; Lewthwaite et al., 2002; Vartian et al., 1985).
Surface-associated proteins are important for streptococcal virulence (Navarre & Schneewind, 1999). Clinical isolates of pyogenic streptococci express M/M-like proteins on their surface (Simpson et al., 1987). These coiled-coil proteins are highly homologous in structure and confer resistance to phagocytic killing, but often differ in primary sequence and ability to bind plasma proteins (Cunningham, 2000; Fischetti, 1989; Oster & Bisno, 2000). We recently identified the M-like fibrinogen-binding protein of GGS, denoted FOG (Johansson et al., 2004). Besides binding fibrinogen, protein FOG mediates GGS binding to albumin, immunoglobulin G and collagen type I. Protein FOG acts as an adhesin and inhibits neutrophil function (Johansson et al., 2004; Nitsche et al., 2006). Bacterial release of M/M-like proteins could enable the bacteria to corrupt host defence functions distant from the bacterial surface and the primary site of infection. Fragments of M1 protein are released from the GAS surface by endogenous proteolytic cleavage as well as by released proteases of neutrophils (Berge & Björck, 1995; Herwald et al., 2004). Such release has so far not been shown for M/M-like proteins of GGS.
Serious conditions with GGS and GAS are often preceded by milder superficial infections (Auckenthaler et al., 1983). The skin and mucosa comprises a number of cell types with defence tasks such as synthesis and release of antimicrobial peptides (AMPs). The interferon (IFN) -governed CXC chemokine MIG/CXCL9 is a lymphocyte chemoattractant with antibacterial properties (Cole et al., 2001; Liao et al., 1995). MIG/CXCL9 is expressed in high amounts by inflamed epithelium, as seen in the case of streptococcal tonsillitis (Egesten et al., 2007) and psoriasis (Goebeler et al., 1998). The electrolyte compositions of body fluids prone to streptococcal infection differ, and may influence the activity of AMPs (Goldman et al., 1997). Since GGS are generally regarded as commensals and GAS as being highly virulent, the interplay between the ionic strength of the environment and innate immunity may affect the potential of these bacteria to survive as commensals on epithelial surfaces.
In this study we address the question of the innate immune response and its effect on GGS, by investigating FOG-provoked MIG/CXCL9 expression from a monocytic cell line and the effect of a MIG/CXCL9-derived peptide on bacterial survival. By including electrolyte concentrations relevant to the in vivo situation in MIG/CXCL9 bactericidal assays, we reveal properties that can at least partly explain the persistence of GGS and their success as commensals.
|
METHODS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). The GAS strains AP1 (40/58), AP22 (59/50), AP41 (71/50) and AP49 (1/64) of serotypes M1, M22, M41 and M49, respectively, were from the World Health Organization Collaborating Centre for Reference and Research on Streptococci, Prague, Czech Republic. Streptococci were routinely grown in Todd–Hewitt (TH) liquid medium at 5 % CO2 and 37 °C. Strain G45 and isogenic mutants thereof were grown in the presence of erythromycin at a final concentration of 1 µg ml–1, to maintain a knock-out phenotype (Nitsche-Schmitz et al., 2007). For heat-killing of bacteria, a 2x109 ml–1 bacterial suspension was placed at 80 °C for 20 min, collected by centrifugation and resuspended in sterile-filtered RPMI 1640 (Gibco) with bovine or human serum albumin (BSA or HSA respectively); 2x108 c.f.u. were added to individual wells.
Chemokine-producing cells.
The human monocytic cell line THP-1 (ATCC #TIB-202) was routinely grown at 37 °C in RPMI 1640 supplemented with 10 % (v/v) heat-inactivated fetal bovine serum and 50 µg gentamicin ml–1 (all from Gibco). For experiments, cells were washed once in RPMI 1640 and resuspended in RPMI 1640 containing 0.4 or 4 mg HSA ml–1, or 0.4, 4 or 6 mg BSA ml–1. Experiments were performed using 0.5x106 cells ml–1 in cell culture plates (Nunc). The human pharyngeal epithelial cell line Detroit 562 (ATCC #CCL-138) was cultured in Minimal Essential Medium (MEM) with Earle's salts (Invitrogen Life Technologies) supplemented with 2 mM glutamine (Invitrogen), 10 % heat-inactivated fetal bovine serum (Invitrogen), 0.25 µg amphotericin B ml–1 (Fungizone; Invitrogen), penicillin/streptomycin (PEST) (100 U ml–1 and 0.1 ng ml–1 respectively; Invitrogen). Detroit 562 cells were used in experiments when close to confluence in 6 mg BSA ml–1 in MEM. For both cell types, MIG/CXCL9 was induced by IFN (1000 U ml–1) where appropriate.
Proteins, labelling and bacterial binding assay.
Recombinant proteins FOG and M1 were purified as described previously (Åkesson et al., 1994; Johansson et al., 2004) with the exception that protein FOG and constructs thereof lack the GST-tag used for purification. Endotoxin levels varied between 100 and 250 EU ml–1, i.e. 2 ng ml–1 at most, as assessed using the Limulus amoeba lysate technique (a kind service of Dr L. Larsson, Clinical Microbiology, Lund University). Preparations used were more than 95 % pure as judged by SDS-PAGE. FITC-labelling of proteins (EZ-labelling, Pierce) was carried out in accordance with the manufacturer's recommendations. Human IgG and fibrinogen (Sigma) were labelled with 125I using Iodobeads (Pierce). HSA was labelled using the Bolton–Hunter reagent. Assessment of bacterial binding of labelled proteins has been described (Björck & Kronvall, 1984).
Surface release of FOG, SDS-PAGE and blotting methods.
Bacterial overnight culture supernatants from the FOG-expressing strain G41 were recovered by centrifugation, sterile-filtered and precipitated with 5 % trichloracetic acid (Merck). Alternatively, G41 was cultured to mid-exponential phase, washed once and then, in a total volume of 0.5 ml PBS (2.7 mM KCl, 1.5 mM KH2PO4, 137 mM NaCl, 8.1 mM Na2HPO4), incubated with 2.5 mU human neutrophil elastase (Sigma) at 37 °C for 30 min. Strain G148, lacking fog, was used as a control. Bacteria were removed by centrifugation; samples were resolved in SDS loading buffer and then loaded in duplicate onto a 10 % (w/v) acrylamide SDS-PAGE gel. Proteins were separated under reducing conditions. Eurosequencing carried out NH2-terminal sequencing of proteins excised from acrylamide gels. For blotting, proteins were electroblotted onto a PVDF membrane, which was incubated in block buffer (1 %, w/v, BSA, 0.05 % Tween 20 in PBS) for 3x20 min. To detect protein FOG, membranes were incubated for 3 h at room temperature with unlabelled fibrinogen (50 µg ml–1 in 1 %, w/v, BSA in TBST; 0.05 M Tris, 0.15 M NaCl, 0.05 %, v/v, Tween 20 pH 7.5) and probed with a rabbit anti-fibrinogen antibody (diluted 1 : 3000, DAKO). A secondary horseradish-peroxidase-conjugated antibody (dilution 1 : 3000, Bio-Rad) raised in goat was used to detect bound primary antibody. To confirm that the antibodies were not interacting with membrane-bound proteins in a non-immune fashion, the membrane was incubated with radiolabelled fibrinogen in parallel, as a control. In both cases of detection the membrane was washed and developed using an imaging system (Fuji).
Absorption experiments and chemokine response.
In evaluation of the chemokine response, equimolar concentrations (0.14x10–9 mol) of proteins FOG, FOG 1-C and M1 were used. The amount used correlated with the maximum MIG/CXCL9 response by M1 in THP-1 cells. In parallel, an LPS concentration of 2 ng ml–1, corresponding to the maximum levels found in the recombinant protein preparations, was used. Cell supernatants were analysed for MIG/CXCL9 by ELISA according to the manufacturer's instructions (R&D Systems). In a separate set of experiments, G45 and isogenic mutants thereof were added to THP-1 cells in medium containing 0.4 or 4 mg HAS ml–1, or 0.4, 4 or 6 mg BSA ml–1). Control wells were treated with IFN alone. Incubation proceeded at 37 °C for 24 h and supernatants were analysed for MIG/CXCL9 content by ELISA.
Strain G45 and the isogenic mutants thereof (2x108 c.f.u. in pellet) were incubated in 100 % saliva, 100 % plasma or HSA solution (4 mg ml–1 in PBS) at room temperature for 1 h, washed extensively in PBS, and bound proteins eluted using 0.2 M glycine pH 2. Eluates were separated by SDS-PAGE under reducing conditions and visualized by Coomassie brilliant blue staining.
Real-time PCR.
Total RNA was prepared from THP-1 cells using the Trizol reagent according to the manufacturer's instructions (Invitrogen Life Technologies). Real-time PCR was performed in triplicates of each sample on a Bio-Rad IQ 5 cycler after reverse transcriptase conversion of 50 ng total RNA to cDNA (TaqMan). The SYBRgreen mastermix and the housekeeping GAPDH gene were from Bio-Rad and the labelled MIG/CXCL9 primers were purchased from Applied Biosystems.
Flow cytometry analysis of protein binding.
THP-1 cells were incubated with equimolar concentrations (9.13 µM) of FITC-labelled proteins for 1 h at 37 °C. Following a wash step in PBS, samples were analysed on a FACSCalibur flow cytometer (Becton-Dickinson). FITC-labelled BSA was first related to autofluorescence and then used as a control for cells incubated with protein FOG or M1.
Bactericidal assays.
Bacterial overnight cultures in TH were reinoculated in fresh medium and further cultured to mid-exponential phase. After washing three times with sterile 10 mM Tris/HCl, pH 7.5, supplemented with 5 mM glucose, bacterial numbers were adjusted to 2x109 c.f.u. ml–1 using an optical density standard curve. Dilutions were made in buffer containing 0, 5, 50, 100 or 150 mM NaCl. Approximately 1x105 c.f.u. in 50 µl were then incubated, with or without increasing amounts of protein MIG/CXCL9 (Peprotech) or the MIG/CXCL9-derived synthetic peptide SAD-27 (amino acid residues 79–105 of the holoprotein), for up to 1 h at 37 °C. Suspensions were diluted in the respective buffers and plated on TH solid medium. C.f.u. were counted following an overnight incubation at 37 °C and the results expressed as percentage killing as compared to bacteria incubated in the respective buffers alone.
Electrostatic potential surface modelling.
Modelling of the electrostatic potential surface of MIG/CXCL9 was carried out using the DelPhi software (http://wiki.c2b2.columbia.edu/honiglab_public/index.php/Software) with the following parameters: the nonlinear Poisson–Boltzmann equation 
R(ei=eo)=80 at +4 kT and –4 kT. CSalt=0 and 150 mM were investigated. Results were depicted in blue surrounding the structural MIG/CXCL9 model reported previously (Egesten et al., 2007).
|
RESULTS |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
). In contrast to GAS, GGS multiply rapidly in cell culture medium and to standardize the number of c.f.u. added to the monocytes, heat-killed bacteria were used. The number of c.f.u. added to the cells was monitored by parallel plating of live bacteria on solid TH medium and did not diverge significantly from the number calculated from a standard curve (data not shown). Moreover, binding of ligands such as albumin, IgG and fibrinogen to GGS and GAS were not affected by heat-killing. A dose–response curve of heat killed bacteria showed a maximum MIG/CXCL9 release using 2x108 c.f.u. (data not shown). For THP-1 cells, the presence of bacteria elicited an overall high MIG/CXCL9 response, with levels ranging from 102 500 pg ml–1 (strain G148) to 275 100 pg ml–1 (strain AP1) (Fig. 1a). No significant difference in induction levels was observed between GAS (mean 203 247±60 353 pg ml–1) and GGS (mean 173 000±47 576 pg ml–1) strains. This indicates that there is no difference between the species with respect to innate immune recognition by THP-1 cells. The lower response comparing strain G148 (which naturally lacks protein FOG) to strain G41 (178 700 pg MIG/CXCL9 ml–1) indicated a role for protein FOG in elicitation of a MIG/CXCL9 response. For the epithelial cell line, levels of MIG/CXCL9 were threefold lower than for THP-1 and ranged from 15 500 pg ml–1 (strain G148) to 72 250 pg ml–1 (strain AP49). As for THP-1, GAS strains and GGS strains were equally potent inducers of MIG/ CXCL9.
|
; Herwald et al., 2004), is released by host proteases, the FOG-expressing strain G41 was incubated with purified human neutrophil elastase. Cleavage products were analysed by SDS-PAGE and Western blots. For detection of protein FOG, antibodies directed against human fibrinogen were used, as shown in Fig. 2(a). A fragment with an apparent size of 52 kDa was released from the bacterial surface (Fig. 2a, lane 2). A fragment smaller than the full-length protein FOG (Fig. 2a, lane 1), was also found in TCA-precipitated culture medium obtained from the early stationary phase of bacterial growth (Fig. 2a, lane 3). To avoid false-positive results through non-immune binding of the detection antibodies to e.g. protein G, radiolabelled fibrinogen was also used. These experiments confirmed the results of the previous experiments using antibodies. As the isogenic FOG mutant G45 does not bind fibrinogen (Nitsche-Schmitz et al., 2007), we could exclude other fibrinogen-binding proteins on FOG-expressing bacteria. Strain G148, which naturally lacks protein FOG, was negative in both supernatant sample and following elastase incubation (data not shown). NH2-sequencing of the 52 kDa protein band showed 100 % identity to the NH2-terminus of the full-length protein FOG. By size calculation we could conclude that processing most likely occurs at the C-terminal region between amino acid residues 472 and 520, generating a fragment that corresponds to the recombinant construct FOG1-C (aa 1–493; see schematic representation in Fig. 2a; Johansson et al., 2004). The presence of soluble protein FOG in the context of a GGS infection hence seems plausible and motivated the use of FOG 1-C in this study.
|
), and to examine a possible interaction with other cells of the immune system we analysed binding of FOG1-C to THP-1 cells by flow cytometry (Fig. 2b
). To investigate potential differences in innate immune response in response to GGS and GAS M proteins, we included protein M1, a well-studied M protein of GAS. As a control for background binding of unspecific labelling compound we used FITC-BSA, with a background binding of 6.8 % as compared to autofluorescence. Monocytes incubated with the FITC-labelled proteins FOG or M1 demonstrated a clear binding of the proteins as compared to cells labelled with FITC-BSA. The relative numbers of cells binding the two M proteins were similar (24.7 % and 27.6 % for M1 and FOG1-C, respectively). Addition of IFN did not affect binding of the investigated proteins (data not shown), indicating that the cell surface receptor is not affected by the presence or absence of IFN.
Protein FOG induces an increase of MIG/CXCL9 expression
To examine a possible functional response after cellular binding of proteins FOG and M1, production of MIG/CXCL9 by THP-1 cells was investigated using these proteins as stimulants. In the presence of IFN, soluble FOG1-C and M1 triggered a time-dependent increase in expression of MIG/CXCL9 as determined by ELISA. However, using equimolar concentrations of the proteins, FOG1-C failed to generate chemokine production to the same level as M1, which was a potent inducer (Fig. 3a). After 3 h, the FOG1-C-stimulated cells showed a lower MIG/CXCL9 response than cells stimulated with M1 (1.3-fold compared to 3-fold for FOG1-C and M1, respectively, as compared to the IFN-treated control). At 18 h of stimulation M1 protein generated approximately twice the amount of MIG/CXCL9 as did FOG 1-C (Fig. 3a). These results were confirmed by real-time PCR, showing a number of MIG/CXCL9 transcripts at 3 and 18 h of stimulation corresponding to experiments analysed by ELISA (Fig. 3b). Inter-experimental variations motivated the depiction as fold increase of MIG/CXCL9 release (ELISA) and the showing of one representative real-time PCR experiment (n=3). LPS controls excluded significant effects from the low levels of contaminating endotoxin in the protein preparations (not shown). As in the case of THP-1 cells, protein FOG (1861±127.1 pg ml–1) elicited a significantly lower MIG expression as compared to protein M1 (13 625±1515 pg ml–1) in Detroit 562 cells.
|
; Björck et al., 1987). To further validate the contribution of protein FOG to MIG/CXCL9 expression, the GGS strain G45 (wild-type) and the isogenic mutants G45
pG (devoid of protein G) and G45FOG (devoid of protein FOG) were tested for their ability to induce MIG/CXCL9 expression. Similar amounts of MIG/CXCL9 were induced by G45pG and the wild-type strain G45 (Fig. 1b). In contrast, G45FOG gave rise to a significantly reduced expression of MIG/CXCL9, demonstrating that also at the bacterial surface protein FOG contributes to MIG/CXCL9 induction. Stimulation of THP-1 cells is routinely performed in the presence of BSA (6 mg ml–1). Due to the affinity of GGS for HSA, the expression of MIG/CXCL9 was also investigated in the presence of HSA at concentrations of physiological relevance at sites of infection (0.4 and 4 mg ml–1). In all experiments, significantly decreased levels of MIG/CXCL9 were obtained with G45FOG, while G45pG and wild-type G45 were equally potent MIG/CXCL9 inducers.
In parallel, binding experiments using radiolabelled HSA demonstrated that at the bacterial surface protein G serves as the main albumin-recruiting protein (Fig. 1c). Recruitment of albumin to the surface of G45 and the isogenic mutants, from human plasma, saliva and an albumin solution serving as control, further emphasized this finding (data not shown). Taken together, the results demonstrate that on the surface of GGS, protein FOG, but not protein G, clearly contributes to innate immune recognition and a subsequent innate immune response. Additionally, the recruitment of HSA or BSA from the surrounding environment has a limited impact on this result.
GGS show tolerance to the antimicrobial effect of MIG/CXCL9 and the MIG/CXCL9-derived peptide SAD-27
Having established that both GGS and GAS cause release of MIG/CXCL9 from monocytes, we proceeded to investigate the susceptibility of these species to the antibacterial activity of MIG/CXCL9. The fact that GGS can exist as part of the normal flora, whereas GAS more often cause disease, raised the question whether commensal bacteria are more resistant to AMPs. We conducted a number of bactericidal assays using MIG/CXCL9 on strains G41 and AP1. In parallel we tested the synthetic MIG/CXCL9-derived peptide SAD-27, based on the antibacterial part of the MIG/CXCL9 molecule (Egesten et al., 2007). As the effect of many AMPs is reduced by increasing ionic strength (Bals et al., 1998; Goldman et al., 1997; Yang et al., 2003) we also investigated if the killing by MIG/CXCL9 or SAD-27 would be altered by NaCl at the concentrations found in unstimulated saliva (5 mM) and sweat (50 mM).
Strains G41 and AP1 were killed in a dose-dependent manner, with close to 100 % killing at 0.09 µM holoprotein MIG/CXCL9 (Fig. 4a). The same levels of bacterial killing were reached by 0.66 µM of the synthetic peptide SAD-27 (Fig. 4b). Although a higher concentration of the peptide was required for bacterial killing, no additive or stabilizing effect was noted when using the holoprotein. In the presence of 50 mM NaCl, G41 bacteria were resistant to the bactericidal activity of MIG/CXCL9, while 35 % of AP1 bacteria were still killed by the chemokine (Fig. 4a). Similarly the bactericidal activity of SAD-27 was inhibited in the presence of NaCl (Fig. 4b).
|
). The addition of 50 mM NaCl delayed killing of both strains by 5 min and reduced the level of killing of AP1 to 40 % after 30 min. For G41 the presence of NaCl had a greater negative impact on the bactericidal effect of SAD-27-mediated killing, as the maximum level of killing after 1 h was only 17 %.
Next, we investigated the sensitivity of GGS and GAS to ion concentrations corresponding to the levels in body compartments prone to streptococcal infection: saliva (5 mM), sweat (50 and 100 mM), and plasma (150 mM) (van der Merwe et al., 2002). We found that G41 was more tolerant than AP1 to concentrations exceeding 5 mM NaCl (data not shown). At 50 mM, the reduction of c.f.u. (i.e. killing) after 1 h was markedly higher for the AP1 strain as compared to G41 bacteria. At most, G41 showed a decline in c.f.u. of 40 % whereas the AP1 decline approached 80 %. A similar difference in NaCl sensitivity was obtained with other tested isolates of GGS and GAS. In summary, the results suggest that GGS strains are more tolerant than GAS to physiological NaCl concentrations found at sites relevant to streptococcal colonization and infestation, and also more resistant to the bactericidal activity of MIG/CXCL9 expressed at these sites.
Influence of salt on the electrostatic surface potential of MIG/CXCL9
Finally, the effects of increasing ionic strength on the electrostatic surface properties of MIG/CXCL9 were examined. Modelling of the electrostatic surface potential of MIG/CXCL9 revealed a greatly diminished field surrounding the NH2-terminus at 150 mM NaCl compared with absence of salt (Fig. 4d). The C-terminal surface potential was moderately decreased, indicating a retained positive net charge of the bactericidal 
-helical domain. These results demonstrate that although 150 mM NaCl reduces the electrostatic surface potential of the -helical domain of MIG/CXCL9, the bactericidal activity is likely to be retained. Thus, the differences observed in MIG/CXCL9 killing of GGS and GAS point to differences in biophysical properties of their respective surfaces. Further investigations may explain these findings.
|
DISCUSSION |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
) to provoke an innate immune response. During an encounter between neutrophils and FOG-expressing bacteria, neutrophil elastase can release a fragment of protein FOG with retained biological activity. This suggests that also in vivo it is plausible that M/M-like proteins from GGS are cleaved from the bacterial surface, eliciting responses from surrounding immune cells. In the current study we demonstrate that protein FOG1-C, a fragment relevant in size to that found after GGS encounter with neutrophils or directly released from the bacterial surface, binds to monocytes and together with IFN triggers production of the chemokine MIG/CXCL9. Soluble protein FOG differed markedly from the classical virulence factor protein M1 of GAS in inducing production of the chemokine MIG/CXCL9. At equimolar amounts, M1 caused a more pronounced release of MIG/CXCL9 after 18 h. Analysis of gene expression of MIG/CXCL9 confirmed these findings. The less potent induction of MIG/CXCL9 expression by protein FOG may, in an infection, signify a reduced recruitment of T-cells to the site. This may result in a general reduction of inflammatory mediators originating from these cells. Infections with GAS, but traditionally not GGS, are often associated with an exaggerated immune response resulting in tissue damage and severe disease in the host. Our results may signify GGS's ability to act as a commensal and avoid causing a robust immune response.
When stimulating a monocytic cell line with heat-killed whole bacteria, G41 and G148 also induced a lower expression of MIG/CXCL9 as compared to AP1, although no significant difference between GGS and GAS was observed when additional strains were analysed. The same was found for stimulation of Detroit 562 cells. However, for THP-1, the markedly lower MIG/CXCL9 level induced by G148 compared to that of G41 led to the proposition of a role for protein FOG. The plausible role for protein FOG was confirmed using an isogenic mutant lacking protein FOG. Recent testing of isogenic mutants of GAS strain AP1 in our group, to study the contribution of M1 to MIG/CXCL9 expression, found no differences between the wild-type AP1 strain and mutants lacking protein M1 (Eliasson et al., 2007). Seemingly, when anchored to the bacterial surface protein FOG adds to the expression of MIG/CXCL9, whereas in the soluble form it is only a moderate inducer. In addition, the results show that although protein G is the primary albumin-recruiting surface protein of GGS, netiher this protein nor albumin binding contributes to an innate immune response and the subsequent chemokine expression in the cell type investigated here.
Cell walls of Gram-positive bacteria are known to cause chemokine release from monocytes (Heumann et al., 1994) and such a release of TNF, IL-1 and IL-6 was suggested to contribute to the severity of bacterial septicaemia (Whicher & Evans, 1990). Differential host chemokine response has also been implicated as a determinant of severity in GAS infection (Norrby-Teglund et al., 2000). Furthermore, significant amounts of pro-inflammatory cytokines (IL-1 and 
, TNF, IL-6 and IL-8) were secreted from monocytes in response to heat-killed GAS (Müller-Alouf et al., 1994), and these bacteria also induced transcription of the MIG/CXCL9 gene in macrophages (Veckman et al., 2003). However, reports on GGS causing such chemokine release have until now been scarce. Issues concerning how commensals avoid evoking an immunoresponse are interesting and have been discussed (Dieleman et al., 2004; Granucci & Ricciardi-Castagnoli, 2003). The current study adds new findings to this field.
The observation that the investigated GGS strains are more tolerant to salt concentrations typical of various body fluids suggests a structural or metabolic advantage for GGS in the quest for survival and persistence in the colonized host. In NaCl concentrations exceeding that of unstimulated saliva, MIG/CXCL9-mediated killing of GGS was reduced. Also, in concentrations corresponding to that of sweat, GGS were better survivors than GAS. GGS are known to cause severe infections in the immunocompromised host, and socio-economically challenged groups often suffer from GGS infections (Lewthwaite et al., 2002; Vartian et al., 1985). In sweat of malnourished individuals NaCl levels are increased (Rodrigues et al., 1994) and hence malnutrition may be a risk factor for GGS colonization. The resistance to high levels of NaCl is shared by other pathogens (Schmid et al., 1995).
Generally AMPs carry a positive net charge, which enables the interaction with negatively charged surfaces of bacteria. Modelling of the electrostatic surfaces of MIG/CXCL9 showed an altered, but yet intact ability of the C-terminus to exert its bactericidal effect. Plausibly, the presence of NaCl exceeding 5 mM reveals a difference in net surface charge between GGS and GAS. A higher level of lipoteichoic acid (LTA) has been reported for the cell wall of GAS compared to GGS, rendering a more negatively charged and hydrophobic surface (Miörner et al., 1983). Possibly the LTA-related negative surface charge of GAS could explain the susceptibility of these bacteria to AMPs. Both the AMP killing of GAS in NaCl buffer, rather than GGS, and the effect of NaCl alone imply differences in the architecture of the streptococcal cell wall that could greatly affect the degree of persistence during colonization. Recognition of GGS as an emerging human pathogen with distinct survival skills is therefore important.
|
ACKNOWLEDGEMENTS |
|---|
Edited by: R. J. Lamont
|
REFERENCES |
|---|
|
TOPABSTRACTINTRODUCTIONMETHODSRESULTSDISCUSSIONREFERENCES |
|---|
Auckenthaler, R., Hermans, P. E. & Washington, J. A., II (1983). Group G streptococcal bacteremia: clinical study and review of the literature. Rev Infect Dis 5, 196–204.[Medline]
Bals, R., Wang, X., Wu, Z., Freeman, T., Bafna, V., Zasloff, M. & Wilson, J. M. (1998). Human beta-defensin 2 is a salt-sensitive peptide antibiotic expressed in human lung. J Clin Invest 102, 874–880.[Medline]
Berge, A. & Björck, L. (1995). Streptococcal cysteine proteinase releases biologically active fragments of streptococcal surface proteins. J Biol Chem 270, 9862–9867.
Björck, L. & Kronvall, G. (1984). Purification and some properties of streptococcal protein G, a novel IgG-binding reagent. J Immunol 133, 969–974.[Abstract]
Björck, L., Kastern, W., Lindahl, G. & Widebäck, K. (1987). Streptococcal protein G, expressed by streptococci or by Escherichia coli, has separate binding sites for human albumin and IgG. Mol Immunol 24, 1113–1122.[CrossRef][Medline]
Bowler, P. G., Duerden, B. I. & Armstrong, D. G. (2001). Wound microbiology and associated approaches to wound management. Clin Microbiol Rev 14, 244–269.
Cole, A. M., Ganz, T., Liese, A. M., Burdick, M. D., Liu, L. & Strieter, R. M. (2001). Cutting edge: IFN-inducible ELR-CXC chemokines display defensin-like antimicrobial activity. J Immunol 167, 623–627.
Cunningham, M. W. (2000). Pathogenesis of group A streptococcal infections. Clin Microbiol Rev 13, 470–511.
Dieleman, L. A., Hoentjen, F., Qian, B. F., Sprengers, D., Tjwa, E., Torres, M. F., Torrice, C. D., Sartor, R. B. & Tonkonogy, S. L. (2004). Reduced ratio of protective versus proinflammatory cytokine responses to commensal bacteria in HLA-B27 transgenic rats. Clin Exp Immunol 136, 30–39.[CrossRef][Medline]
Egesten, A., Eliasson, M., Johansson, H. M., Olin, A. I., Mörgelin, M., Müeller, A., Pease, J. E., Frick, I. M. & Björck, L. (2007). The CXC chemokine MIG/CXCL9 is important in innate immunity against Streptococcus pyogenes. J Infect Dis 195, 684–693.[CrossRef][Medline]
Eliasson, M., Frick, I. M., Collin, M., Sorensen, O. E., Bjorck, L. & Egesten, A. (2007). M1 protein of Streptococcus pyogenes increases production of the antibacterial CXC chemokine MIG/CXCL9 in pharyngeal epithelial cells. Microb PathogAug 1 [Epub ahead of print]
Fischetti, V. A. (1989). Streptococcal M protein: molecular design and biological behavior. Clin Microbiol Rev 2, 285–314.
Goebeler, M., Toksoy, A., Spandau, U., Engelhardt, E., Brocker, E. B. & Gillitzer, R. (1998). The C-X-C chemokine Mig is highly expressed in the papillae of psoriatic lesions. J Pathol 184, 89–95.[CrossRef][Medline]
Goldman, M. J., Anderson, G. M., Stolzenberg, E. D., Kari, U. P., Zasloff, M. & Wilson, J. M. (1997). Human beta-defensin-1 is a salt-sensitive antibiotic in lung that is inactivated in cystic fibrosis. Cell 88, 553–560.[CrossRef][Medline]
Granucci, F. & Ricciardi-Castagnoli, P. (2003). Interactions of bacterial pathogens with dendritic cells during invasion of mucosal surfaces. Curr Opin Microbiol 6, 72–76.[CrossRef][Medline]
Hashikawa, S., Iinuma, Y., Furushita, M., Ohkura, T., Nada, T., Torii, K., Hasegawa, T. & Ohta, M. (2004). Characterization of group C and G streptococcal strains that cause streptococcal toxic shock syndrome. J Clin Microbiol 42, 186–192.
Herwald, H., Cramer, H., Mörgelin, M., Russell, W., Sollenberg, U., Norrby-Teglund, A., Flodgaard, H., Lindbom, L. & Björck, L. (2004). M protein, a classical bacterial virulence determinant, forms complexes with fibrinogen that induce vascular leakage. Cell 116, 367–379.[CrossRef][Medline]
Heumann, D., Barras, C., Severin, A., Glauser, M. P. & Tomasz, A. (1994). Gram-positive cell walls stimulate synthesis of tumor necrosis factor alpha and interleukin-6 by human monocytes. Infect Immun 62, 2715–2721.
Honore, P. M., Lozana, A., Defalque, D. & Estratiou, A. (1996). Fatal septic shock due to Lancefield group G streptococci. Acta Clin Belg 51, 57–60.[Medline]
Johansson, H. M., Mörgelin, M. & Frick, I. M. (2004). Protein FOG – a streptococcal inhibitor of neutrophil function. Microbiology 150, 4211–4221.
Lewthwaite, P., Parsons, H. K., Bates, C. J., McKendrick, M. W. & Dockrell, D. H. (2002). Group G streptococcal bacteraemia: an opportunistic infection associated with immune senescence. Scand J Infect Dis 34, 83–87.[CrossRef][Medline]
Liao, F., Rabin, R. L., Yannelli, J. R., Koniaris, L. G., Vanguri, P. & Farber, J. M. (1995). Human MIG chemokine: biochemical and functional characterization. J Exp Med 182, 1301–1314.
Miörner, H., Johansson, G. & Kronvall, G. (1983). Lipoteichoic acid is the major cell wall component responsible for surface hydrophobicity of group A streptococci. Infect Immun 39, 336–343.
Müller-Alouf, H., Alouf, J. E., Gerlach, D., Ozegowski, J. H., Fitting, C. & Cavaillon, J. M. (1994). 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 62, 4915–4921.
Navarre, W. W. & Schneewind, O. (1999). Surface proteins of gram-positive bacteria and mechanisms of their targeting to the cell wall envelope. Microbiol Mol Biol Rev 63, 174–229.
Nitsche, D. P., Johansson, H. M., Frick, I. M. & Mörgelin, M. (2006). Streptococcal protein FOG, a novel matrix adhesin interacting with collagen I in vivo. J Biol Chem 281, 1670–1679.
Nitsche-Schmitz, D. P., Johansson, H., Sastalla, I., Reissmann, S., Frick, I. M. & Chhatwal, G. S. (2007). Group G streptococcal IgG binding molecules FOG and protein G have different impact on opsonization by C1q. J Biol Chem 282, 17530–17536.
Norrby-Teglund, A., Chatellier, S., Low, D. E., McGeer, A., Green, K. & Kotb, M. (2000). Host variation in cytokine responses to superantigens determine the severity of invasive group A streptococcal infection. Eur J Immunol 30, 3247–3255.[CrossRef][Medline]
Oster, H. R. & Bisno, A. L. (2000). Group C and G Streptococcal Infections: Epidemiologic and Clinical Aspects. Washington, DC: American Society for Microbiology.
Rodrigues, M. E., Melo, M. C., Reis, F. J. & Penna, F. J. (1994). Concentration of electrolytes in the sweat of malnourished children. Arch Dis Child 71, 141–143.
Schmid, J., Hunter, P. R., White, G. C., Nand, A. K. & Cannon, R. D. (1995). Physiological traits associated with success of Candida albicans strains as commensal colonizers and pathogens. J Clin Microbiol 33, 2920–2926.[Abstract]
Simpson, W. J., Robbins, J. C. & Cleary, P. P. (1987). Evidence for group A-related M protein genes in human but not animal-associated group G streptococcal pathogens. Microb Pathog 3, 339–350.[CrossRef][Medline]
Sylvetsky, N., Raveh, D., Schlesinger, Y., Rudensky, B. & Yinnon, A. M. (2002). Bacteremia due to beta-hemolytic Streptococcus group G: increasing incidence and clinical characteristics of patients. Am J Med 112, 622–626.[CrossRef][Medline]
van der Merwe, D. E., Ubbink, J. B., Delport, R., Becker, P., Dhatt, G. S. & Vermaak, W. J. (2002). Biological variation in sweat sodium chloride conductivity. Ann Clin Biochem 39, 39–43.[CrossRef][Medline]
Vartian, C., Lerner, P. I., Shlaes, D. M. & Gopalakrishna, K. V. (1985). Infections due to Lancefield group G streptococci. Medicine (Baltimore) 64, 75–88.[Medline]
Veckman, V., Miettinen, M., Matikainen, S., Lande, R., Giacomini, E., Coccia, E. M. & Julkunen, I. (2003). Lactobacilli and streptococci induce inflammatory chemokine production in human macrophages that stimulates Th1 cell chemotaxis. J Leukoc Biol 74, 395–402.
Whicher, J. T. & Evans, S. W. (1990). Cytokines in disease. Clin Chem 36, 1269–1281.
Yang, D., Chen, Q., Hoover, D. M., Staley, P., Tucker, K. D., Lubkowski, J. & Oppenheim, J. J. (2003). Many chemokines including CCL20/MIP-3alpha display antimicrobial activity. J Leukoc Biol 74, 448–455.
Received 9 May 2007;
revised 10 July 2007;
accepted 19 July 2007.
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| INT J SYST EVOL MICROBIOL | MICROBIOLOGY | J GEN VIROL |
| J MED MICROBIOL | ALL SGM JOURNALS | |
