Nature Medicine
7, 1298 - 1305 (2001)
doi:10.1038/nm1201-1298
Evasion of human innate and acquired immunity by a bacterial homolog of CD11b that inhibits opsonophagocytosisBenfang Lei1, 5, Frank R. DeLeo1, 5, Nancy P. Hoe1, Morag R. Graham1, Stacy M. Mackie1, Robert L. Cole1, Mengyao Liu1, Harry R. Hill2, Donald E. Low3, Michael J. Federle4, June R. Scott4
& James M. Musser11 Laboratory of Human Bacterial Pathogenesis, Rocky Mountain Laboratories, National Institute of Allergy and Infectious Diseases, Hamilton, Montana, USA 2 Departments of Pathology, Pediatrics and Medicine, University of Utah School of Medicine, Salt Lake City, Utah, USA 3 Department of Microbiology, Mount Sinai Hospital and University of Toronto, Ontario, Canada 4 Department of Microbiology and Immunology, Emory University School of Medicine, Atlanta, Georgia, USA 5 B.L. and F.R.D. contributed equally to this work.
Correspondence should be addressed to James M. Musser jmusser@niaid.nih.govMicrobial pathogens must evade the human immune system to survive, disseminate and cause disease. By proteome analysis of the bacterium Group A Streptococcus (GAS), we identified a secreted protein with homology to the  -subunit of Mac-1, a leukocyte  2 integrin required for innate immunity to invading microbes. The GAS Mac-1−like protein (Mac) was secreted by most pathogenic strains, produced in log-phase and controlled by the covR-covS two-component gene regulatory system, which also regulates transcription of other GAS virulence factors. Patients with GAS infection had titers of antibody specific to Mac that correlated with the course of disease, demonstrating that Mac was produced in vivo. Mac bound to CD16 (Fc RIIIB) on the surface of human polymorphonuclear leukocytes and inhibited opsonophagocytosis and production of reactive oxygen species, which resulted in significantly decreased pathogen killing. Thus, by mimicking a host-cell receptor required for an innate immune response, the GAS Mac protein inhibits professional phagocyte function by a novel strategy that enhances pathogen survival, establishment of infection and dissemination.The innate immune system is the first line of defense against invading microbial pathogens. The molecular mechanisms used by infectious agents to subvert the innate immune functions of the host, and thereby cause disease, are not well characterized. Polymorphonuclear leukocytes (PMNs) are critical components of the innate immune system that protect the host by killing pathogenic microbes after phagocytosis. The ingestion and killing of pathogens by PMNs depend on the normal function of many proteins acting in an orchestrated molecular cascade. Despite some recent progress1,
2, relatively little information is available about specific factors and mechanisms used by pathogens to disable distinct steps in the protective cascade.
Group A Streptococcus (GAS) is a gram-positive bacterial pathogen that is responsible for human morbidity and mortality globally3. The organism causes infections such as pharyngitis, cellulitis, bacteremia, necrotizing fasciitis and post-infectious sequelae such as acute rheumatic fever (ARF) and acute glomerulonephritis. The success of GAS as a pathogen is dependent on its ability to avoid phagocytosis and killing by PMNs, and complement-mediated effects4.
Recently, Lei et al.5 analyzed the culture supernatant proteome of GAS strains and identified a secreted protein (protein spot 22, designated Mac) with homology to the  -subunit of human Mac-1, a  2 integrin. Mac-1 (also known as M2, CD11b/CD18, complement receptor 3) is a leukocyte integrin involved in diverse biologic processes such as innate immunity6. By binding intercellular adhesion molecule-1 (ICAM-1), C3bi and other ligands, Mac-1 regulates the critical leukocyte functions of adhesion, migration, phagocytosis and oxidative killing6,
7. Here we show that Mac is a secreted virulence factor that inhibits phagocytosis and killing of GAS by PMNs through molecular mimicry.
Homology of GAS Mac and the -subunit of human Mac-1
Mac protein made by serotype M1 strain MGAS5005 has 26% amino-acid (aa) identity and 45% aa similarity to a 173-aa residue region of the -subunit of Mac-1 (CD11b) located between E319 and V491 (Fig. 1). The homologous region in human Mac-1 contains three of seven W repeats (W3, W4 and W5), which form a -propeller domain8. Although the sequence homology shared between the two proteins is located outside of the putative complement-binding domain in Mac-1, streptococcal Mac contains an Arg-Gly-Asp (RGD) integrin-binding motif at aa residues 214−216 (Fig. 1). Moreover, Springer and colleagues have recently proposed that the -propeller of 2-integrins may modulate ligand binding to the I-domain or permit ligand-receptor signal transduction8,
9.
Production of Mac in vitro and in vivo
The ability of GAS to cause human disease is partly due to production of various secreted virulence factors3. Therefore, we determined whether Mac protein was secreted by multiple GAS strains. In 23 of 37 GAS strains (representing 28 distinct M-protein serotypes), immunoreactive Mac was produced. Strains of five of the six most abundant M-protein serotypes responsible for human invasive infections10 (M1, M3, M12, M11 and M22) secreted Mac in vitro under the conditions studied (Fig. 2b). We next tested the hypothesis that the mac gene was preferentially expressed in the log phase of growth, a time when many other GAS virulence factors are produced. Real-time reverse-transcriptase (RT)-PCR analysis of wild-type GAS indicated that the mac gene was transcribed in log phase and downregulated in stationary phase (Fig. 2c). The RT-PCR data were verified by western-blot analysis of culture supernatant proteins (Fig. 2d).
 | | Figure 2. Mac production by GAS and analysis of Mac antibody titers in patients with GAS disease. | | |  | a, SDS−PAGE of a lysate of E. coli expressing recombinant Mac and purified Mac. Lanes: 1, E. coli with empty (control) vector; 2, E. coli lysate containing Mac; 3, purified recombinant Mac. b, Western-blot analysis of Mac in culture supernatants of GAS strains. Lanes: 1, purified recombinant Mac; 2−11 (serotype/MGAS strain number): M1/5005, M1/7638, M1/8520, M3/315, M3/6256, M3/7361, M49/8557, M56/8716, M11/8219 and ST2967/2105. c, RT-PCR analysis of mac transcription. GAS cultures were collected at mid-log (ML), late-log (LL), and early-stationary phase (ES).  , wild-type M1 strain MGAS5005;  , isogenic mutant strain JRS950 with the covR gene insertionally inactivated. d, Western-blots of culture supernatant proteins taken at the same time points. Mac produced by the covR isogenic mutant was degraded by a constitutively upregulated extracellular protease11. e, Mac antibody titers in sera from patients with invasive GAS infections and ARF. Left, each set of line-connected spots represents the acute and convalescent (Conv.) paired sera from an individual with GAS invasive disease as indicated. Right, each set of line-connected spots represents paired sera from an individual with streptococcal ARF. *, P = 0.004 versus acute sera; **, P = 0.02 versus acute stage (paired Student's t-test).
 Full Figure and legend (36K) |
| |
Inactivation of the covR-covS two-component gene regulatory system enhances resistance of the mutant strain to opsonophagocytic killing by human leukocytes11,
12,
13. Although this phenotype has been attributed to increased capsule synthesis, it might also result from increased Mac production. To test this, we studied a wild-type M1 organism and isogenic covR mutant by RT-PCR assays. Transcription of the mac gene was upregulated in the isogenic covR mutant strain (Fig. 2c), which indicates repression by this two-component regulatory system and suggests that Mac contributes to the enhanced resistance to killing.
To determine if Mac was expressed in vivo, we used an ELISA to measure antibody levels to purified recombinant Mac in human sera. Antibodies against Mac were present in 29% of serum samples from 625 healthy individuals living in Finland, Canada and 3 sites in the United States. This result indicated that individuals from diverse localities have antibodies that react with Mac, presumably as a consequence of previous infection with GAS, such as pharyngitis. To determine if GAS infection directly stimulated production of Mac-specific antibody, serum samples from 27 patients with documented recent GAS infections were studied by ELISA (Fig. 2e). Antibodies against Mac were present in 15 of 17 serum samples from humans recovering from GAS invasive infections (P = 0.004 versus acute sera) and in 8 of 10 individuals at the onset of ARF (P = 0.02 versus +1 y) (Fig. 2e). Sera from patients with GAS pharyngitis also contained antibodies against Mac (data not shown). These results indicate that Mac is synthesized during the course of GAS infections in humans, and suggest that Mac influences GAS-host interactions.
Mac inhibits reactive oxygen species generation by PMNs
Because Mac has homology with human Mac-1 and contains a leukocyte integrin-binding motif, Mac might modulate human PMN function. Generation of reactive oxygen species (ROS) by PMNs and other phagocytic leukocytes is a critical component of a competent innate host defense against invading micro-organisms. We therefore used a fluorescence-based assay16 to determine if purified recombinant Mac could inhibit ROS production by serum-opsonized zymosan (OPZ)-stimulated PMNs (Fig. 3a and b), and found that Mac significantly inhibited ROS production in a dose-dependent manner (Fig. 3a). As phagocytosis of OPZ is only partially mediated by complement and complement receptors such as Mac-1, we tested the efficacy of Mac to block complement C3bi-opsonized immunoglobulin G (IgG) latex beads (C3bi−IgG latex beads) (Fig. 3c). In contrast to the partial inhibition of ROS production by Mac in response to OPZ, Mac almost completely abolished ROS production by PMNs stimulated with C3bi−IgG latex beads ( 95.0% inhibition using 5.0  g/ml Mac). Even more striking was the ability of Mac to block ROS production by PMNs stimulated with IgG latex beads (EC50  0.1 g/ml (2.8 nM)) (Fig. 3d). Finally, we tested the ability of Mac to block ROS production by serum-opsonized GAS (Fig. 3e). Consistent with our findings using OPZ and C3bi−IgG- or IgG latex beads, Mac significantly inhibited ROS generation by GAS-stimulated PMNs (Fig. 3e). We also observed that Mac did not activate PMNs, inhibit PMA-stimulated ROS production, or prime for enhanced release of ROS in response to n-formyl peptide (fMLP) (Fig. 3a and data not shown). Thus, Mac did not block ROS generation per se. Rather, the results suggest that Mac inhibits phagocytosis directly or inhibits molecular processes resulting in ROS production during or after phagocytosis.
| | Figure 3. Mac-dependent inhibition of ROS production during phagocytosis. | | |  | a, Mac inhibits ROS production by OPZ-stimulated PMNs. ROS production was measured in unstimulated human PMNs ( ,  ) or those stimulated with OPZ (, ) in the presence (, ) or absence (, ) of Mac. b, Rate of Mac-dependent inhibition of ROS production by PMNs during phagocytosis. Rates of ROS production were determined for unstimulated PMNs ( ), OPZ-stimulated PMNs () or OPZ-stimulated PMNs incubated with 5.0 g/ml Mac ( ). c and d, Mac inhibits PMN ROS production during receptor-mediated phagocytosis. PMNs were stimulated with C3bi/IgG latex beads (c) or IgG latex beads (d) in the presence () or absence () of Mac (in g/ml). e, Mac inhibits ROS production by PNs stimulated with opsonized GAS. PMNs were stimulated with opsonized GAS serotype M1 strain MGAS 5005 in the presence () or absence () of Mac (in g/ml). c and e show the effect of the MacRGE mutant protein (). GAS PGK was used to control for non-specific inhibitory effects in the assay (, in c−e). Insets in a, c, d and e show kinetic plots of ROS production by stimulated PMNs in the absence (black solid line) or presence (blue solid line) of 5.0 g/ml Mac (a, c and e) or indicated concentration (d). Dotted line, fluorescence in unstimulated PMNs. Results for all panels are expressed as the mean  s.d. of 3−6 separate experiments. *, P < 0.01 versus stimulated PMNs in the absence of Mac.
Full Figure and legend (67K) |
| |
Mac inhibits phagocytosis and killing of GAS by human PMNs
To determine whether Mac inhibited phagocytosis of GAS, human PMNs were incubated with serotype M1 or M3 GAS opsonized with serum containing antibody against GAS in the presence of recombinant Mac (Fig. 4a). Robust ingestion of strains of both GAS serotypes occurred after opsonization with GAS-specific antibody (Fig. 4b and c). In contrast, few bound or ingested GAS were observed in the assays that used unopsonized GAS or bacteria opsonized with serum lacking anti-GAS antibody (Fig. 4b & c, white and cross-hatched bars, respectively). Because non-immune serum was ineffective at mediating opsonization of GAS, we hypothesized that sera containing anti-GAS antibody would promote complement deposition on GAS and mediate binding and/or ingestion by PMNs. Antibodies specific for CD11b (Mac-1) significantly inhibited the phagocytosis of GAS by PMNs (Fig. 4b & c, green bar), indicating that GAS ingestion is in part mediated by complement-Mac-1 binding. Consistent with its ability to inhibit ROS production by PMNs, Mac significantly decreased binding or phagocytosis of GAS (P < 0.01, ANOVA, with Dunnett's correction for multiple comparisons) and this effect was abrogated with anti-Mac antibody (Fig. 4b). Phagocytosis of GAS was not completely inhibited by Mac, likely due to ingestion facilitated by GAS surface components other than those targeted by Mac. Together, the data indicate that the inhibition of ROS production was caused by the ability of Mac to block phagocytosis, an effect mediated by Mac's disruption of an antibody and/or complement mediated interaction with receptor(s) on the PMN cell surface.
| | Figure 4. Mac inhibits phagocytosis and killing of GAS by human PMNs. | | |  | a, Representative micrographs of GAS ingested by PMNs in the absence (left) or presence (right) of 5.0 g/ml Mac. Top, bright-field micrographs; bottom, Nomarski DIC micrographs with GAS pseudocolorized in red. Black arrowheads indicate the position of bound or ingested GAS. Magnification, x500. b and c, Dose-dependent inhibition of phagocytosis by Mac. PMNs were stimulated with serotype M1 (b) or M3 (c) GAS pre-opsonized with immune serum (all bars except Unop and Ab- op) in the presence of Mac at the indicated concentrations (in g/ml) and phagocytosis was determined. Unop, assays containing PMNs stimulated with unopsonized GAS; Ab- op, GAS treated with non-immune serum; CD11b, antibody specific for CD11b used at 2 g/ml; Isotype Ab, IgG2b; PGK, GAS PGK used at 50 g/ml; Mac, affinity-purified anti-Mac antibody used at 20 g/ml. d, Dose-dependent inhibition of PMN bactericidal activity by Mac. PMNs were incubated with serotype M1 or M3 GAS pre-opsonized with immune serum in the presence of Mac at the indicated concentrations (in g/ml) and bactericidal activity was determined. For all panels, the number of experiments performed for each condition is indicated within the bars. *, P < 0.01 versus Ab+ op.
Full Figure and legend (116K) |
| |
The ability of Mac to block phagocytosis and ROS production by PMNs suggested that the protein also inhibits killing of GAS by PMNs. To test this, we measured the bactericidal activity of PMNs toward GAS in the presence of Mac (Fig. 4d). Mac significantly reduced the ability of human PMNs to kill serotype M1 and M3 GAS (P  0.01, ANOVA, with Dunnett's correction for multiple comparisons) (Fig. 4d). Thus, the inhibition of PMN phagocytosis and ROS production by Mac directly results in significantly increased survival of GAS.
Interaction of Mac with the I-domain of CD11b
Although literature suggests that RGD motifs do not mediate Mac-1−ligand interactions14,
15, it is possible that the region of -propeller homology in GAS Mac would mimic that present in human CD11b and modulate binding of C3bi to its complement-binding I-domain or, alternatively, directly interact with complement. Thus, binding of Mac to human Mac-1 or complement might block Mac-1−complement mediated events necessary for GAS-host interaction. To determine if Mac binds to the surface of human PMNs, we treated cells with Mac and analyzed by flow cytometry with Mac-specific antibody (Fig. 5a). GAS Mac bound to the cell surface (Fig. 5a) and binding was blocked partially with an antibody specific for human CD11b (Fig. 5a). To further study a possible interaction of Mac with CD11b, we tested whether transfected CHO cells expressing Mac-1 could bind immobilized Mac (Fig. 5b). Mac bound to the transfected cell line in a dose-dependent manner, whereas we observed no binding to the untransfected parental cell line (Fig. 5b). Moreover, binding was inhibited by Mac-specific antibody (Fig. 5b). We next compared the ability of Mac to bind to the I-domain of CD11b with that of purified human C3bi (Fig. 5c). Mac interacted with the I-domain but not with C3bi (Fig. 5c). Moreover, Mac significantly blocked the binding of C3bi with CD11b I-domain in a dose-dependent manner (EC50  54 nM), albeit not completely (Fig. 5d). Although Mac interacted with the I-domain of CD11b in the ELISAs, purified I-domain did not inhibit the binding of Mac with PMNs when analyzed by flow cytometry (Fig. 5e). Moreover, I-domain inhibited production of ROS by neither C3bi−IgG nor IgG latex beads, although it could be argued that the binding and inhibitory regions of Mac are distinct (data not shown). Thus, these results indicate that interaction of Mac with the I-domain of CD11b on the cell surface is of low-affinity and/or requires a coreceptor.
| | Figure 5. Interaction of Mac with host cells and with CD16/Mac-1. | | |  | a, Mac binds to human PMNs. PMNs were incubated with (+) or without (-) 5 g/ml Mac and stained with Mac (blue lines) or with control antibody (dashed lines) and analyzed by flow cytometry. Third panel, PMNs were incubated with anti-CD11b (green line) or control Ab (blue line), and Mac binding was determined. Purple arrow illustrates cells not binding Mac. Bottom panel illustrates binding of purified recombinant MacRGE (red line) to PMNs. b, Binding of CHO cells expressing human Mac-1 to streptococcal Mac. Binding of immobilized Mac with parental, wild-type CHO () or transfected CHO cells expressing Mac-1 () was determined. The assay was also performed in the presence of Mac or control antibody (Ctl Ab). c, Binding of Mac to CD11b I-domain or C3bi. Immobilized CD11b I-domain (solid lines) or C3bi (dashed lines) were incubated with Mac at 0 g/ml (black line), 0.5 g/ml (blue line), 1 g/ml (green line), 2 g/ml (red line), and Mac binding was determined. d, Inhibitory effect of Mac and C3bi on their binding to CD11b I-domain. Immobilized CD11b I-domain was incubated with C3bi () or Mac () and then Mac or C3bi, respectively, and binding of each was determined. Inset panel, purified C3bi and CD11b I-domain used in the ELISAs. e, CD16 mediates binding of Mac to PMNs. PMNs were incubated with (blue line) or without (black dotted line) Mac in the presence of CD11b I-domain (top panel), IgG1 (second panel), antibody against CD16 that does not block antibody binding (third panel), KB61 antibody against CD32 (fourth panel) and mAb 3G8 (bottom panel), and then stained with Mac. f, Mac inhibits binding of antibody to PMNs. PMNs were incubated with polyclonal antibody (top panel, green line) in the presence (green line) or absence (green dotted line) of Mac (middle panel) or mAb 3G8 (bottom panel). Black dotted line, PMN auto fluorescence; gray line, represents binding of the secondary (detection) antibody. Graph illustrates dose-dependent inhibition of antibody binding to human PMNs by Mac and mAb 3G8 (CD16).
Full Figure and legend (89K) |
| |
Binding of Mac to human PMNs is mediated by CD16
Previous studies have demonstrated that Mac-1 and the Fc-receptor, CD16 (Fc RIIIB), are physically and functionally linked17,
18,
19,
20,
21. Because Mac inhibited ROS production mediated by Fc-receptors alone (Fig. 3d), we tested whether antibodies specific for Fc-receptors blocked Mac binding (Fig. 5e). Using six monoclonal antibodies specific for CD16 and CD32 (FcRIIA), we found that a monoclonal antibody specific for the Fc-binding portion of CD16 (antibody 3G8) completely inhibited binding of Mac with PMNs (Fig. 5e). Based on those findings, we hypothesized that Mac inhibits CD16-mediated antibody binding at the cell surface. Mac inhibited the binding of antibody to the surface of human PMNs in a dose-dependent manner and at concentrations that were effective at blocking phagocytosis and ROS production (Fig. 5f). The capacity of Mac to block antibody binding was similar to that of mAb 3G8 (Fig. 5f). Taken together, these results demonstrate that Mac binds to CD16 on the surface of PMNs and sequesters the antibody Fc-binding region of the receptor, thereby blocking receptor−antibody interactions.
Discussion
Several lines of evidence indicate that Mac blocks phagocytosis by disrupting the interaction between human CD16 and immune complexes. Mac inhibited phagocytosis of GAS opsonized with immune serum, and blocked ROS production by PMNs activated with opsonized GAS and C3bi−IgG- and IgG-opsonized latex beads, all of which contain surface-bound antibody. Mac bound to the surface of PMNs, and specifically to CD16 through a region that overlaps the antibody-binding domain. That Mac, a mimetic of CD11b, associates with CD16 is consistent with several previous studies demonstrating that CD11b and CD16 are physically associated17,
18,
19,
20,
21. Based on our data demonstrating that Mac also interacts with human Mac-1 by associating with the I-domain of CD11b and competes with C3bi for I-domain binding (Fig. 5b−d), it is possible that the I-domain of CD11b serves as a low-affinity coreceptor for Mac. It should be noted that ingestion of GAS by PMNs was partially mediated by complement−Mac-1 interaction, as indicated by the inhibition of phagocytosis of GAS by antibody specific for CD11b (Fig. 4; ref. 22). Thus, it is likely that inhibition of GAS phagocytosis by GAS Mac was due, at least in part, to blocking complement-mediated ingestion. However, an interaction between Mac and the I-domain at the PMN plasma membrane is apparently not crucial for its ability to bind PMNs or for its ability to inhibit ROS production. The observation that Mac mimics CD11b but targets CD16 underscores the importance of Fc-receptors in the innate immune response, and provides additional evidence for a physical and functional association of Mac-1 and CD16. This is especially intriguing when one considers that CD16 is a glycosylphosphatidylinositol-anchored receptor with no known signal-transducing component. Thus, it is possible that the ability of Mac to bind CD16 and disrupt antibody and/or complement binding is mediated by its -propeller homology to CD11b, although our current data do not address that issue. Our findings are summarized in a model illustrating how the -propeller homology of GAS Mac blocks antibody and complement binding to a CD16−Mac-1 complex (Fig. 6). In combination, these data demonstrate directly that GAS disrupts host-pathogen interactions by secreting Mac, a host-receptor mimetic.
| | |
GAS produces several other molecules that detrimentally alter host defenses, most notably PMN function. For example, although the mechanism is not known, the hyaluronic acid capsule made by GAS is antiphagocytic11. GAS also expresses an extracellular peptidase that cleaves and inactivates complement protein C5a, thereby reducing recruitment of PMNs and mononuclear phagocytes to the site of infection23,
24. M protein, a well-known virulence factor located on the cell surface, decreases opsonization of GAS by inhibiting activation of the alternative complement pathway25,
26,
27. Inactivation of the covR-covS two-component gene regulatory system enhances resistance of the mutant strain to opsonophagocytic killing by human leukocytes11. Although this phenotype has been attributed to increased capsule synthesis, the enhanced mac expression suggests that Mac contributes to the enhanced resistance to killing. Our finding that Mac blocks the phagocytosis of opsonized GAS extends knowledge about the immune subversion mechanisms available to this important human pathogen. Most significantly, we have identified a secreted protein that functions to block both innate and acquired immune responses of the host. Hence, extracellular GAS virulence factors that subvert every step leading to phagocytosis, namely, phagocyte recruitment, opsonization, and binding and/or ingestion all promote human disease pathogenesis.
Evasion of innate immunity by human pathogens is currently an area of intense study and is only now being investigated with recently developed proteomics and genomics methods28,
29. Many pathogens have devised means to survive within host cells by producing extracellular factors, which disable critical components of the innate immune response (reviewed in ref.1). Alternatively, microorganisms such as GAS and Staphylococcus aureus can disseminate and cause disease by preventing their interaction with phagocytic leukocytes4,
30. The latter mechanism for immune evasion is also exploited by some viral pathogens. For example, Ebola virus secretes a glycoprotein that interacts with CD16 on the surface of PMNs, presumably inhibiting host response to infection31. Both strategies can contribute to human disease, but little is known about the specific mechanisms used for immune evasion by the individual pathogens.
Using proteomics, we have identified a secreted protein made by the human bacterial pathogen GAS that inhibits the normal function of professional phagocytes responsible for protecting the host. Mac has homology with the -subunit of human Mac-1 and this homology modulates host−pathogen interaction by a mechanism involving inhibition of processes mediated by CD16−Mac-1, such as binding and ingestion of bacteria. Thus, this is the first report of a secreted bacterial protein that mimics a host-cell receptor required for a competent innate immune response and directly promotes pathogen survival. The presence of specific antibody in the sera of patients with pharyngitis, acute rheumatic fever and severe invasive disease episodes indicates that this protein is made in the course of all major categories of human infection caused by GAS. Mac production during GAS-human interaction would assist establishment of infection, pathogen survival and dissemination. Analysis of microbial genome databases indicates that genes encoding homologous proteins are present in Streptococcus equi (an important horse pathogen) and Treponema denticola (associated with human periodontal disease), suggesting that this mechanism of inhibition of innate immunity is broadly distributed.
Methods
Bacterial strains and sera.
Sixty-eight GAS strains representing 37 M protein serotypes responsible for 75% of human invasive infections in the US and a large percentage of pharyngitis cases in many countries studied in recent years10 were isolated from 3 countries and 4 states. A covR negative isogenic derivative strain of MGAS 5005 (JRS950) has been described11. Recombinant Mac and PGK were used to generate affinity-purified rabbit polyclonal anti-Mac (Mac) and anti-streptococcal phosphoglycerate kinase (PGK) antibodies (Bethyl Laboratories, Montgomery, Texas).
Sera from 625 healthy individuals have been described32. Paired sera were obtained from acute and convalescent (9 mo to 14 y following the acute episode) phases of an initial rheumatic fever episode in 10 patients, and from 17 patients with invasive GAS infection upon diagnosis (acute phase) and 11 d to 257 d after diagnosis (convalescent phase). ARF is a post-streptococcal infection sequelae which means that patients with ARF patients generally have high levels of GAS-specific antibodies at the time of diagnosis. Strains from the 17 invasive GAS infections comprised the following serotype strains: ten M1, one M3, two M13, one M22, two M58 and one st2967. Sera were also collected from 8 children and 1 adult with uncomplicated GAS pharyngitis 14 d to 30 d after diagnosis.
Expression and purification of recombinant Mac.
Recombinant Mac was purified from Escherichia coli BL21 (DE3) containing plasmid pSP22 (ref. 5) with DEAE- and phenyl-Sepharose chromatography. Mac was > 98% pure as assessed by SDS−PAGE and identity was confirmed by N-terminal aa sequencing and western-blot analysis with Mac. Contaminating endotoxin was removed with Detoxi-gel (Pierce Chemical, Rockford, Illinois). MacRGE mutant (D216  E216) was generated using a QuickChange XL Site-Directed Mutagenesis Kit (Stratagene, La Jolla, California) and the expressed protein was purified as described above.
The mac gene in MGAS 5005 (serotype M1) is identical to open reading frame Spy0861 in a sequenced M1 GAS genome (http://www.genome.ou.edu/strep.html) and consists of a 1020-bp open reading frame that would encode a 339-aa residue protein with an inferred molecular mass of 38,020 kD. Following cleavage of a typical secretion signal peptide between aa residues 29 (Ala) and 30 (Asp), the mature protein has a predicted molecular mass of 34,939 kD (ref. 5). The gene was identified by PCR and DNA sequencing in all 68 GAS strains studied that together represent 37 distinct M-protein serotypes.
TaqMan assays.
TaqMan assays were performed as described by Chaussee et al.33, with a probe and primers specific for mac. Cultures of GAS were collected at OD600 = 0.3 (mid-log), DO600 = 0.5 (late-log), and OD600 = 0.7 (early stationary phase). Transcription of the mac gene was compared with gyrA, a gene expressed constitutively throughout the GAS growth cycle.
Western-blot and ELISA analyses.
Mac production was assessed in 37 GAS strains with western immunoblotting. GAS were grown in Todd−Hewitt broth containing 0.2% yeast extract (Difco, Detroit, Michigan) at 37 °C with 5% CO2. Proteins from 1 ml of culture supernatant were precipitated with 3 volumes of ethanol on ice, centrifuged and analyzed by SDS−PAGE and western blotting with Mac5. Mac antibody was detected in sera from human patients using recombinant Mac with described ELISA procedures32.
Preparation of purified C3bi and CD11b I-domain.
C3bi was prepared as decribed34. As expected, purified C3bi migrated as 3 protein bands on reducing SDS−PAGE and all were immunoreactive with anti-complement C3 antiserum (Metra Biosystems, Mountain View, California). The CD11b gene fragment encoding E147 to G337 was amplified from the genomic DNA of CHO/Mac-1 cells, cloned and expressed in E. coli BL21. Recombinant CD11b I-domain was purified by cation exchange chromatography. DNA sequencing and N-terminal aa sequencing confirmed the integrity of the CD11b I-domain gene construct and expressed protein, respectively.
PMN isolation, and binding of Mac to PMNs and CHO cells expressing Mac-1.
Venous blood was obtained from healthy individuals in accordance with a protocol approved by the Institutional Review Board for Human Subjects, NIAID. PMNs were isolated as described35 and suspended in Dulbecco's PBS containing 10 mM d-glucose (DPBS/G). All reagents had < 10.0 pg/ml endotoxin. To determine Mac-PMN binding, PMNs (1  106) were incubated with 5 g/ml Mac or MacRGE for 20−30 min, stained with Mac- or PGK-specific antibody and analyzed by flow cytometry (FACsCalibur, Becton−Dickinson, San Jose, California). For blocking experiments, antibodies specific for CD11b (M1/70 and ICRF44, BD Biosciences, San Diego, CA; 2LPM19c, DAKO, Carpinteria, California), CD16 (3G8, BD Biosciences; DJ130c and VIFcRIII, DAKO; and BL-LGL/1, Sigma) and CD32 (FLIR.8, BD Biosciences; and KB61, DAKO) were incubated with PMNs (10−20 g/ml) before or after Mac as indicated. Where indicated, polyclonal antibodies (10 g/ml) were added to PMNs following Mac. Parental CHO-K1 (ATCC) and CHO-K1 cells expressing CD11b/CD18 (CHO/Mac-1) were cultured as described36. Binding between GAS Mac and host cell was studied using described methods37.
Protein-binding assays.
For Mac-C3bi−CD11b binding assays, CD11b I-domain or C3bi was adsorbed to Nunc Maxisorp microtiter plates overnight. Plates were washed, blocked with 0.02% polyvinylpyrrolidone and incubated with Mac. Subsequently, plates were washed, blocked with 2% milk and incubated with Mac and HRP-conjugated anti-rabbit antibody (Bio-Rad, Hercules, California). Bound Mac was detected with a SpectraMax 384 Plus microplate spectrophotometer at 405 nm (Molecular Devices, Sunnyvale, California). For the competition ELISAs, varied Mac or C3bi concentrations were incubated with fixed amounts of C3bi or Mac (1 g/ml), respectively, with immobilized CD11b I-domain (1 g/ml). C3bi binding was detected using the goat anti-human complement C3 antiserum (Metra Biosystems) and HRP-conjugated anti-goat antibody (Santa Cruz Biotechnology, Santa Cruz, California).
Assays for ROS production, phagocytosis and bactericidal activity.
Intracellular ROS production was monitored for 90 min as described16 and Vmax was calculated as the maximum rate of ROS production over 5−10 min. Latex beads (2.0 m; Polysciences, Warrington, Pennsylvania) coated with complement-C3bi and IgG (C3bi−IgG LB) or IgG alone (IgG LB) were used at an 8:1 bead: PMN ratio. Serotype M1 GAS was opsonized for 30 min at 37 °C with immune serum obtained from an individual with recent GAS-induced pharyngitis and used at a 30:1 GAS: PMN ratio.
Phagocytosis of serotype M1 and M3 GAS by PMNs was performed as described38, but with modifications. Strains MGAS5005 (serotype M1) and MGAS315 (serotype M3) were grown to late exponential phase and opsonized with immune serum. PMNs (2.5 106) and 1 107 opsonized GAS were combined in the presence of Mac, 2 g/ml anti-CD11b M1/70 or control IgG2b (BD Pharmingen), or 50 g/ml purified GAS PGK (control protein similar in size to Mac). Mixtures were rotated for 30 min at 37 °C and phagocytosis was terminated by putting the samples on ice. Smears were prepared on slides and stained with a modified Wright−Giemsa stain (Sigma). Percent bound/ingested GAS was determined by counting 250−350 PMNs per slide and calculating the percentage of PMNs with associated GAS. PMN bactericidal activity was determined from the same experiments. Aliquots of each assay were plated on BHI agar to determine numbers of viable bacteria. Percent relative bactericidal activity was calculated by comparing the ability of each treatment (for example, Mac) to inhibit PMN killing of GAS with that of assays containing anti-CD11b using the equation: [(CFUanti-CD11b treatment-CFUMac treatment) ÷ (CFUanti-CD11b treatment-CFUNo treatment)] 100. Statistics were performed using a one-way ANOVA with Dunnett's correction for multiple comparisons (GraphPad Instat, version 3.01, GraphPad Software, San Diego, California).
Received 20 September 2001; Accepted 24 October 2001
REFERENCES
- Meresse, S. et al. Controlling the maturation of pathogen-containing vacuoles: a matter of life and death. Nature Cell Biol. 1, E183–188 (1999). | Article | PubMed | ISI | ChemPort |
- Ernst, J.D. Bacterial inhibition of phagocytosis. Cell. Microbiol. 2, 379–386 (2000). | Article | PubMed | ISI | ChemPort |
- Musser, J.M. & Krause, R.M. in Emerging Infections (eds. Krause, R.M. & Fauci, A.) 185–218 (Academic Press, San Diego, 1998).
- Cunningham, M.W. Pathogenesis of group A streptococcal infections. Clin. Microbiol. Rev. 13, 470–511 (2000). | Article | PubMed | ISI | ChemPort |
- Lei, B., Mackie, S.M., Lukomski, S. & Musser, J.M. Identification and immunogenicity of group A Streptococcus culture supernatant proteins. Infect. Immun. 68, 6807–6818 (2000). | Article | PubMed | ISI | ChemPort |
- Corbi, A.L., Kishimoto, T.K., Miller, L.J. & Springer, T.A. The human leukocyte adhesion glycoprotein Mac-1 (complement receptor type 3, CD11b) subunit. Cloning, primary structure, and relation to the integrins, von willebrand factor and factor B. J. Biol. Chem. 263, 12403–12411 (1988). | PubMed | ISI | ChemPort |
- Diamond, M.S. et al. ICAM-1 (CD54): a counter-receptor for Mac-1 (CD11b/CD18). J. Cell. Biol. 111, 3129–3139 (1990). | Article | PubMed | ISI | ChemPort |
- Springer, T.A. Folding of the N-terminal, ligand-binding region of integrin -subunits into a -propeller domain. Proc. Natl. Acad. Sci. USA 94, 65–72 (1997). | Article | PubMed | ChemPort |
- Oxvig C., Lu C. & Springer T.A., Conformational changes in tertiary structure near the ligand binding site of an integrin domain. Proc. Natl. Acad. Sci. USA 96, 2215–2220 (1999). | Article | PubMed | ChemPort |
- Beall, B., Facklam, R., Hoenes, T. & Schwartz, B. 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 (1997). | PubMed | ISI | ChemPort |
- Levin, J.C. & Wessels, M.R. Identification of csrR/csrS, a genetic locus that regulates hyaluronic acid capsule synthesis in group A Streptococcus. Mol. Microbiol. 30, 209–219 (1998). | Article | PubMed | ISI | ChemPort |
- Federle, M.J., McIver, K.S. & Scott, J.R. A response regulator that represses transcription of several virulence operons in the group A streptococcus. J. Bacteriol. 181, 3649–3657 (1999). | PubMed | ISI | ChemPort |
- Heath, A., DiRita, V.J., Barg, N.L. & Engleberg, N.C. A two-component regulatory system, CsrR-CsrS, represses expression of three Streptococcus pyogenes virulence factors, hyaluronic acid capsule, streptolysin S, and pyrogenic exotoxin B. Infect. Immun. 67, 5298–5305 (1999). | PubMed | ISI | ChemPort |
- Taniguchi-Sidle, A. & Isenman, D.E. Mutagenesis of the Arg-Gly-Asp triplet in human complement component C3 does not abolish binding of iC3b to the leukocyte integrin complement receptor type III (CR3, CD11b/CD18). J. Biol. Chem. 267, 635–643 (1992). | PubMed | ChemPort |
- Van Strijp, J.A. Russell, D.G., Tuomanen, E. Brown, E.J. & Wright, S.D. Ligand specificity of purified complement receptor type three (CD11b/CD18, amb2, Mac-1). Indirect effects of an Arg-Gly-Asp (RGD) sequence. J. Immunol. 151, 3324–3336 (1993). | PubMed | ISI |
- DeLeo, F.R., Allen, L.-A.H., Apicella, M. & Nauseef, W.M. NADPH oxidase activation and assembly during phagocytosis. J. Immunol. 163, 6732–6740 (1999). | PubMed | ISI | ChemPort |
- Brown, E.J., Bohnsack, J.F. & Gresham, H.D. Mechanism of inhibition of immunoglobulin G-mediated phagocytosis by monoclonal antibodies that recognize the Mac-1 antigen. J. Clin. Invest. 81, 365–375 (1988). | PubMed | ISI | ChemPort |
- Sehgal, G., Zhang, K., Todd III, R.F., Boxer, L.A. & Petty, H.R. Lectin-like inhibition of immune complex receptor-mediated stimulation of neutrophils. J. Immunol. 150, 4571–4580 (1993). | PubMed | ISI | ChemPort |
- Zhou, M.J. & Brown, E.J. CR3 (Mac-1,-M 2, CD11b/CD18) and FcRIII cooperate in generation of a neutrophil respiratory burst: requirement for FcRIII and tyrosine phosphorylation. J. Cell. Biol. 125, 1407–1416 (1994). | Article | PubMed | ISI | ChemPort |
- Stocl, J. et al. Granulocyte activation via a binding site near the C-terminal region of complement receptor 3 -chain (CD11b) potentially involved in intramembrane complex formation with glycosylphosphatidylinositol-anchored FcRIIIB (CD16) molecules. J. Immunol. 154, 5452–5463 (1995). | PubMed |
- Galon, J. et al. Soluble Fc receptor type III (FcRIII, CD16) triggers cell activation through interaction with complement receptors. J. Immunol. 157, 1184–1192 (1996). | PubMed | ISI | ChemPort |
- Schnitzler, N., Haase, G., Podbielski, A., Lutticken, R. & Schweizer, K.G. A co-stimulatory signal through ICAM-2 integrin-binding potentiates neutrophils phagocytosis. Nature Med. 5, 231–235 (1999). | Article | PubMed | ISI | ChemPort |
- Wexler, D.E., Chenoweth, D.E. & Cleary, P.P. Mechanism of action of the group A streptococcal C5a inactivator. Proc. Natl. Acad. Sci. USA 82, 8144–8148 (1985). | PubMed | ChemPort |
- Ji, Y., McLandsborough, L., Kondagunta, A. & Cleary, P.P. C5a peptidase alters clearance and trafficking of group A streptococci by infected mice. Infect. Immun. 64, 503–510 (1996). | PubMed | ISI | ChemPort |
- Whitnack, E. & Beachey, E.H. Antiopsonic activity of fibrinogen bound to M protein on the surface of group A Streptococci. J. Clin. Invest. 69, 1042–1045 (1982). | PubMed | ISI | ChemPort |
- Horstmann, R.D., Sievertsen, H.J., Knobloch, J. & Fischetti, V.A. Antiphagocytic activity of streptococcal M protein: selective binding of complement control protein factor H. Proc. Natl. Acad. Sci. USA 85, 1657–1661 (1988). | PubMed | ChemPort |
- Horstmann, R.D., Sievertsen, H.J., Leippe, M. & Fischetti, V.A. Role of fibrinogen in complement inhibition by streptococcal M protein. Infect. Immun. 60, 5036–5041 (1992). | PubMed | ISI | ChemPort |
- Ferrari, G., Langen, H., Naito, M. & Pieters, J. A coat protein on phagosomes involved in the intracellular survival of Mycobacteria. Cell 97, 435–447 (1999). | Article | PubMed | ISI | ChemPort |
- Belcher, C.E. et al. The transcriptional responses of respiratory epithelial cells to Bordetella pertussis reveal host defensive and pathogen counter-defensive strategies. Proc. Natl. Acad. Sci., USA 97, 13847–13852 (2000). | Article | PubMed | ChemPort |
- Thakker, M., Park, J.-S., Carey, V. & Lee, J.C. Staphylococcus aureus serotype 5 capsular polysaccharide is antiphagocytic and enhances bacterial virulence in a murine model. Infect. Immun. 66, 5183–5189 (1998). | PubMed | ISI | ChemPort |
- Yang, Z.-Y. et al. Distinct cellular interactions of secreted and transmembrane Ebola virus glycoproteins. Science 279, 1034–1037 (1998). | Article | PubMed | ISI |
|
