Impaired Opsonization with C3b and Phagocytosis of Streptococcus ...

2 downloads 35 Views 466KB Size Report
Mar 4, 2008 - subjects, C3b-iC3b deposition and total anti-S. pneumoniae antibody activity ... C3b and iC3b, which stimulate phagocytosis; the lysis of mainly.
INFECTION AND IMMUNITY, Aug. 2008, p. 3761–3770 0019-9567/08/$08.00⫹0 doi:10.1128/IAI.00291-08 Copyright © 2008, American Society for Microbiology. All Rights Reserved.

Vol. 76, No. 8

Impaired Opsonization with C3b and Phagocytosis of Streptococcus pneumoniae in Sera from Subjects with Defects in the Classical Complement Pathway䌤 Jose Yuste,1 Ashwin Sen,1 Lennart Truedsson,2 Go ¨ran Jo ¨nsson,2,3 Liang-Seah Tay,4 Catherine Hyams,1 5 6 Helen E. Baxendale, Fiona Goldblatt, Marina Botto,7 and Jeremy S. Brown1* Centre for Respiratory Research, Royal Free and University College Medical School, Rayne Institute, London WC1E 6JJ, United Kingdom1; Department of Laboratory Medicine, Section of Microbiology, Immunology and Glycobiology, Lund University, Lund, Sweden2; Department of Infectious Diseases, Lund University Hospital, Lund, Sweden3; Department of Infectious Diseases, Faculty of Medicine, Imperial College, Hammersmith Campus, London W12 0NN, United Kingdom4; Infectious Diseases and Microbiology Unit, Institute of Child Health, University College London, 30 Guilford Street, London WC1N 1EH, United Kingdom5; Centre for Rheumatology Research, Royal Free and University College Medical School, 46 Cleveland Street, London W1T 4JF, United Kingdom6; and Molecular Genetics and Rheumatology Section, Faculty of Medicine, Imperial College, Hammersmith Campus, London W12 0NN, United Kingdom7 Received 4 March 2008/Returned for modification 17 April 2008/Accepted 2 June 2008

Results from studies using mice deficient in specific complement factors and clinical data on patients with an inherited deficiency of the classical complement pathway component C2 suggest that the classical pathway is vital for immunity to Streptococcus pneumoniae. However, the consequences of defects in classical pathway activity for opsonization with C3b and the phagocytosis of different S. pneumoniae serotypes in human serum are not known, and there has not been a systematic analysis of the abilities of sera from subjects with a C2 deficiency to opsonize S. pneumoniae. Hence, to investigate the role of the classical pathway in immunity to S. pneumoniae in more detail, flow cytometry assays of opsonization with C3b and the phagocytosis of three capsular serotypes of S. pneumoniae were performed using human sera depleted of the complement factor C1q or B or sera obtained from C2-deficient subjects. The results demonstrate that, in human serum, the classical pathway is vital for C3b-iC3b deposition onto cells of all three serotypes of S. pneumoniae and seems to be more important than the alternative pathway for phagocytosis. Compared to the results for sera from normal subjects, C3b-iC3b deposition and total anti-S. pneumoniae antibody activity levels in sera obtained from C2ⴚ/ⴚ subjects were reduced and the efficiency of phagocytosis of all three S. pneumoniae strains was impaired. Anticapsular antibody levels did not correlate with phagocytosis or C3b-iC3b deposition. These data confirm that the classical pathway is vital for complement-mediated phagocytosis of S. pneumoniae and demonstrate why subjects with a C2 deficiency have a marked increase in susceptibility to S. pneumoniae infections. gram-negative bacteria through the formation of the membrane attack complex from components of the terminal complement pathway on the target cell membranes; and the release of the proinflammatory mediators C3a and C5a (43). In addition, complement activity is essential for the development of optimal adaptive immune responses to S. pneumoniae and other pathogens (43). Experiments using genetically engineered mice with defects in specific complement pathways have demonstrated previously that the classical and alternative pathways, and probably to a lesser extent the MBL pathway, are essential for immunity to S. pneumoniae (7, 42, 47). As the alternative pathway does not require specific acquired antibodies (Abs) for activation and amplifies C3b deposition initiated by the classical and MBL pathways, it is often considered to be the main complement pathway contributing to innate immunity. Conversely, as the classical pathway is activated by acquired Abs, it is generally considered to be an effector of the acquired immune response. However, data from experimental infections in mice have suggested that the classical pathway can also be activated by various innate immune mediators, including natural immu-

Invasive disease due to Streptococcus pneumoniae is common and often severe and may be fatal despite appropriate antibiotic treatment, and the characterization of the immune response to this organism is a high priority. One important component of immunity to S. pneumoniae is the complement system, a series of serum and cell surface proteins that have vital roles for both innate and acquired immunity (43). The complement system can be activated mainly through three protein kinase cascades termed the classical, mannan binding lectin (MBL), and alternative pathways (43). Each pathway leads to the formation of a C3 convertase that cleaves the central complement component C3, which aids innate immunity by three main mechanisms: the coating of pathogens with C3b and iC3b, which stimulate phagocytosis; the lysis of mainly

* Corresponding author. Mailing address: Centre for Respiratory Research, Department of Medicine, Royal Free and University College Medical School, Rayne Institute, 5 University St., London WC1E 6JJ, United Kingdom. Phone: 44 20 7679 6008. Fax: 44 20 7679 6973. E-mail: [email protected]. 䌤 Published ahead of print on 9 June 2008. 3761

3762

YUSTE ET AL.

noglobulin M (IgM), C-reactive protein (CRP), serum amyloid P protein (SAP) (7, 41, 49), and the lectin receptor SIGN-R1 (29). As a consequence, in mice the classical pathway seems to be the dominant complement pathway required for innate immunity to S. pneumoniae, with mice deficient in C1q, the first component of the classical pathway activity, having greater susceptibility to infection than mice deficient in factor B (Bf), a central component of the alternative pathway (7). Data obtained with human serum by using magnesium dichlorideEGTA chelation to inhibit classical pathway activity also suggest that the classical pathway has an important role in C3b opsonization and the phagocytosis of S. pneumoniae (17, 19, 35). Clinical data on complement-deficient subjects support these experimental data, with deficiencies affecting the classical pathway particularly associated with S. pneumoniae infections (10, 28). Of 40 subjects with a homozygous deficiency of the classical pathway component C2 described in a recent report, 13 had had documented invasive S. pneumoniae infections (meningitis or septicemia) and 18 had had pneumonia (28), probably due to S. pneumoniae in many cases (32). Furthermore, these infections were frequently recurrent, and there were also high incidences of acute otitis media and sinusitis, infections commonly caused by S. pneumoniae. These data suggest that the loss of classical pathway activity leads to increased susceptibility to S. pneumoniae infections, possibly due to impaired complement-dependent opsonization of S. pneumoniae (12, 19, 20). However, the relative importance of complement activation by the classical or alternative pathway for the phagocytosis of S. pneumoniae in human samples has not been assessed, and whether the classical pathway is as important in humans as it is in mice needs further clarification. The mouse experiments were performed with a single capsular serotype 2 (ST2) S. pneumoniae strain (D39) (7), but as the relative importance of different complement pathways for immunity to bacteria can vary among strains of the same pathogen and among sera from different sources (37, 48), it is important to replicate the results for S. pneumoniae with more than one strain. Furthermore, a systematic analysis of the effects of C2 deficiency on the interactions of S. pneumoniae with complement and phagocytes in sera from multiple C2⫺/⫺ subjects has not been performed. Although the phagocytosis of S. pneumoniae is likely to be affected by a deficiency of C2, existing data on S. pneumoniae phagocytosis in human serum have emphasized the importance of Abs to capsular antigens. As Ab responses to S. pneumoniae antigens are highly dependent on complement activity (21), C2⫺/⫺ subjects may also be susceptible to S. pneumoniae infection due to impaired acquired immunity as a consequence of the failure of previous S. pneumoniae infections to stimulate optimal levels of protective Abs. In the present study, we have used commercially available human sera depleted of single complement factors to investigate the relative importance of the classical and alternative pathways for C3b deposition onto cells of three S. pneumoniae strains and for phagocytosis. In addition, in order to assess in detail possible mechanisms underpinning the high incidence of S. pneumoniae infections in C2⫺/⫺ subjects, we have characterized C3b deposition onto S. pneumoniae cells and phagocytosis in sera obtained from several of these patients.

INFECT. IMMUN. MATERIALS AND METHODS Bacterial strains and growth conditions. S. pneumoniae strains used for this study were M87 (ST4) and Io11697 (ST23F), both gifts from Brian Spratt, Imperial College London (25), and D39 (ST2). Bacteria were cultured at 37°C in 5% CO2 on blood agar plates or in Todd-Hewitt broth supplemented with 0.5% yeast extract to an optical density at 580 nm of 0.4 (approximately 108 CFU/ml) and stored at ⫺70°C in 10% glycerol as single-use aliquots. Human and mouse sera. Human sera with deficiencies of single complement components (C9⫺, C1q⫺, and Bf⫺ sera) (48) were supplied by Calbiochem. When tested using total hemolytic complement and alternative pathway complement kits (The Binding Site), for which results between 80 and 120% of the reference levels are accepted as normal levels of activity, the levels of classical pathway activity in C1q⫺ and Bf⫺ sera were 0 and 100%, respectively, and those of alternative pathway activity were 85 and 0%, respectively. In C1q⫺ sera complemented with 75 ␮g of exogenous C1q (Calbiochem)/ml, the level of classical pathway activity was 100% and the level of alternative pathway activity was 85%. Sera from normal controls and nine C2⫺/⫺ subjects (homozygous for the 28-bp deletion defining C2 deficiency type I) were obtained according to institutional guidelines and stored as single-use aliquots at ⫺70°C. Mouse sera were obtained by terminal exsanguinations from mice (C1q-deficient sera from C1qa⫺/⫺ mice, Bf-deficient sera from Bf⫺/⫺ mice, C3-deficient sera from C3⫺/⫺ mice, and natural-IgM-deficient sera from ␮s⫺/⫺ mice) (2, 15, 34, 45) that had had no prior exposure to S. pneumoniae, and the sera were stored at ⫺70°C as single-use aliquots. For the complementation of depleted sera, purified human C1q or C2 (Calbiochem) was added to samples to a concentration equivalent to 75 or 25 ␮g/ml in 100% serum, respectively. Complement factor, IgG, and IgM binding to S. pneumoniae. Complement factor interactions with S. pneumoniae were assessed using flow cytometry assays as described previously (7, 48). Briefly, C3b-iC3b deposition was investigated by incubating 106 CFU of S. pneumoniae in 10 ␮l of human serum (diluted to 20 or 50% in phosphate-buffered saline [PBS]) for 30 min with or without supplementation with purified C1q, C2, or bovine serum albumin (BSA) and then incubating the bacteria with a fluorescein isothiocyanate (FITC)-conjugated polyclonal goat anti-human C3 Ab (ICN) and characterizing the S. pneumoniae cells by using a FACSCalibur flow cytometer (BD Biosciences) with gating based on the analysis of at least 25,000 bacteria as described previously (7). The very large differences in results for control and C2⫺/⫺ sera were compared using the geometric mean fluorescence intensities (MFIs) of bacteria positive for C3b. The smaller differences obtained with C9⫺, C1q⫺, and Bf⫺ sera were compared using the fluorescence index (FI, defined as the proportion of bacteria positive for C3b multiplied by the geometric MFI) expressed as a relative percentage of the results for C9⫺ serum (16). Assays for the binding of C4b binding protein (C4BP) to S. pneumoniae in human serum used a flow cytometry method similar to that of the C3b deposition assays, with polyclonal sheep anti-human C4BP Ab (Abcam) and a FITC-conjugated donkey anti-goat IgG secondary Ab (Serotec) (both at a 1/300 dilution) (48). The detection of natural IgM binding was investigated using sera from C57BL/6 wild-type mice and ␮s⫺/⫺ mice and a 1/200diluted phycoerythrin-conjugated goat anti-mouse IgM Ab (Jackson ImmunoResearch) as previously described (7). The binding of human IgG and IgM to S. pneumoniae was assessed using assays identical to the C4BP binding assays, but with FITC-conjugated goat anti-human IgM (Sigma) and R-phycoerythrin goat anti-human IgG (Jackson ImmunoResearch), both at a 1/300 dilution, and expressed as FI. Phagocytosis of S. pneumoniae. The phagocytosis of S. pneumoniae in different sera by cells of the human cell line HL-60 (CCL240 promyelocytic leukemia cells; American Type Culture Collection) differentiated into granulocytes was investigated by an established flow cytometry assay (13, 26, 48). In contrast to fresh neutrophils, the HL-60 cell line avoids inconsistency in results due to donor variation, removes the possibility of the contamination of neutrophil samples with human serum products, and is the standard for assessing vaccination-induced effects on the opsonophagocytosis of S. pneumoniae (40). Briefly, S. pneumoniae strains were fluorescently labeled with 6-carboxyfluorescein succinimidyl ester (FAMSE; Molecular Probes) as described previously (26) and stored at ⫺70°C in 10% glycerol. FAMSE-labeled bacteria (106 CFU/10 ␮l) were opsonized with dilutions of human sera deficient in different complement components for 20 min at 37°C in a 96-well plate with horizontal shaking (150 rpm). Differentiated HL-60 cells were washed three times in Hanks’ buffered salt solution (HBSS) and counted, and samples of 105 cells were added to the opsonized bacteria. In control experiments, HL-60 cells were incubated with 10 ␮M cytochalasin D (Sigma) for 15 min before being added to the bacteria. After incubation for 30 min at 37°C with shaking, the suspension was fixed using 3%

VOL. 76, 2008

C2 DEFICIENCY AND STREPTOCOCCUS PNEUMONIAE

paraformaldehyde and a minimum of 6,000 cells were analyzed using a FACSCalibur to identify cells associated with fluorescent bacteria. Measurement of levels of capsule-specific Abs in sera. Levels of capsulespecific Abs in human sera were measured using standardized enzyme-linked immunosorbent assays (ELISAs) and purified capsular antigen as described in detail at http://www.vaccine.uab.edu/ELISA%20Protocol.pdf. Briefly, each well was coated with 100 ␮l of ST-specific pneumococcal polysaccharide antigen (American Type Culture Collection, Manassas, VA) and kept overnight at 4°C. The coated plates were washed in a PBS–0.05% Tween 20 solution, serial dilutions of sera were added to the microtiter plates, and the plates were incubated for 2 h at room temperature, washed in a PBS–0.05% Tween 20 solution, and incubated with goat anti-human IgG-alkaline phosphatase conjugate for a further 2 h. Following further washing as described above, 100 ␮l of 1-mg/ml p-nitrophenylphosphate substrate was added to each well and the plates were incubated for 2 h at room temperature. The reactions were stopped by the addition of 50 ␮l of 3 M NaOH, and the optical density at 405 nm was measured with a reference filter of 690 nm. Optical density data were converted into Ab concentrations with the CDC ELISA computer program and data obtained with the reference serum 89-SF. Statistical analyses. The data presented are representative of results obtained from repeated independent experiments with at least three (usually four) replicas per reaction. The results of phagocytosis and C3b deposition assays were compared to results obtained with control sera by one-way analysis of variance (ANOVA) tests with Dunnett’s posttest analysis. The results of phagocytosis and C3b deposition assays for sera from C2⫺/⫺ subjects with and without the addition of purified C2 were compared by Kruskal-Wallis nonparametric tests.

RESULTS C3b-iC3b deposition onto S. pneumoniae cells in human serum is dependent on both classical and alternative pathway activities. The roles of the classical and alternative pathways in complement activity against S. pneumoniae were assessed using a flow cytometry assay of C3b-iC3b deposition, three strains representing the relatively common capsular STs 2, 4, and 23F, and commercially available human sera depleted of C1q (no classical pathway activity) or Bf (no alternative pathway activity). Sera depleted of C9, a terminal complement component not involved in the opsonization of bacteria with C3b-iC3b, was used as a control. For all three strains, C3b-iC3b deposition in C1q⫺ sera was significantly impaired (P, ⬍0.05 for the ST2 strain and ⬍0.01 for the ST4 and ST23F strains), reduced by 38 to 68% relative to that in C9⫺ sera. In the Bf⫺ sera, C3b-iC3b deposition was reduced by 14 to 41% compared to that in C9⫺ sera, with only the results for the ST4 and ST23F strains reaching statistical significance (Fig. 1A). To confirm that impaired C3b-iC3b deposition onto S. pneumoniae cells in C1q⫺ sera was due to the loss of C1q rather than other unexpected differences among the sera, the C3b-iC3b deposition experiments were repeated for the ST2 strain by using C1q⫺ sera that had been complemented by the addition of purified C1q to a final concentration in pure serum within the normal physiological range (75 ␮g/ml). The addition of C1q to C1q⫺ sera completely restored C3b-iC3b deposition to the levels seen in C9⫺ sera, whereas the addition of BSA to C1q⫺ sera or C1q to C9⫺ sera had no effect (Fig. 1B). Therefore, the reduced C3b-iC3b deposition onto S. pneumoniae cells in C1q⫺ sera is likely to be due wholly to the loss of classical pathway activity rather than other differences among sera. To investigate whether differences in Ab levels might explain differences in C3b-iC3b deposition levels between Bf⫺ and C9⫺ sera, which have preserved classical pathway activity and in which, therefore, Ab can mediate complement activation, total IgG and IgM binding and levels of capsular antigen-

3763

FIG. 1. (A) C3b-iC3b deposition onto 106 CFU of S. pneumoniae incubated in 10-␮l samples of 50% human sera depleted of complement component C1q or Bf, expressed as percent fluorescence indices relative to the results for sera deficient in C9. Results are shown for a capsular ST2 strain (D39), an ST4 strain (M87), and an ST23F strain (Io11697). (B) C3b-iC3b deposition onto 106 CFU of an ST2 strain of S. pneumoniae incubated in 10-␮l samples of 50% human sera depleted of C1q, with and without complementation with the equivalent of 75 ␮g of purified C1q or BSA/ml of 100% serum, expressed as percent fluorescence indices relative to the results for sera deficient in C9. For both panels, error bars represent standard errors of the means, and asterisks mark results that are statistically significant compared to those for C9-deficient sera (A) or C1q-deficient sera (B) (one-way ANOVA; **, P ⬍ 0.01; *, P ⬍ 0.05).

specific Abs to the three S. pneumoniae STs were assessed using ELISAs and flow cytometry, respectively (Table 1). Total IgG and IgM binding to S. pneumoniae showed no substantial differences between Bf⫺ and C9⫺ sera, despite Bf⫺ sera’s being associated with a reduced level of C3b-iC3b deposition onto cells of the ST4 and ST23F strains. In all sera, ST-specific anticapsular IgG levels were relatively low and again did not clearly correlate with the results of the C3b-iC3b assays, with higher levels of anticapsular Abs to the ST4 strain and lower levels of anticapsular Abs to the ST23F strain in Bf⫺ sera than in C9⫺ sera, despite similar reductions in levels of C3b-iC3b deposition onto cells of both these strains in Bf⫺ sera. Binding of C4BP and natural IgM to S. pneumoniae. Bacterial pathogens often prevent complement activity by binding fluid-phase complement regulators such as C4BP, an inhibitor of the classical pathway, and factor H, an inhibitor of alternative pathway activity (24, 48). S. pneumoniae is known to bind to factor H, which is thought to inhibit alternative pathway activity against the bacteria (14, 23, 36), but the interaction with C4BP is not known. As the failure of S. pneumoniae to inhibit classical pathway activity by binding C4BP may be one possible explanation for the relative importance of the classical pathway for immunity to S. pneumoniae, we investigated the ability of the three different strains of S. pneumoniae to bind C4BP in human serum by using a flow cytometry assay. For all three strains, there was no significant binding to C4BP, whereas a mean of 86% (standard error of the mean, 7.2%) of the cells of Streptococcus pyogenes strain H372 (which is known to bind to C4BP) (46) were positive for C4BP (data not

3764

YUSTE ET AL.

INFECT. IMMUN.

TABLE 1. Relative levels of IgG and IgM binding (assessed using flow cytometry) and capsule-specific IgG Abs in sera used for this studya Result for ST2 for: Serum type

Total IgG binding (FI ⫾ SD)

Total IgM binding (FI ⫾ SD)

Result for ST4 for:

Anticapsular IgG level (␮g/ml ⫾ SD)

Total IgG binding (FI ⫾ SD)

Total IgM binding (FI ⫾ SD)

Result for ST23F for:

Anticapsular IgG level (␮g/ml ⫾ SD)

Total IgG binding (FI ⫾ SD)

C9 deficient C1q deficient Bf deficient

2,400 ⫾ 120 1,130 ⫾ 140 3,080 ⫾ 460 1,070 ⫾ 190 2,410 ⫾ 850 1,070 ⫾ 190

2.7 ⫾ 0.1 1.7 ⫾ 0.1 1.8 ⫾ 0.1

1,780 ⫾ 230 900 ⫾ 160 800 ⫾ 120 1,250 ⫾ 50 680 ⫾ 120 900 ⫾ 150

1.0 ⫾ 0.02 1.0 ⫾ 0.05 1.5 ⫾ 0.02

NHSb

1,070 ⫾ 130 2,700 ⫾ 100

2.4 ⫾ 0.3c

1,740 ⫾ 90

2,390 ⫾ 45

0.7 ⫾ 0.03

1,290 ⫾ 280

1,040 ⫾ 170

0.5 ⫾ 0.5

1.4 ⫾ 0.08 0.2 ⫾ 0.01 0.3 ⫾ 0.02 0.2 ⫾ 0.02 0.4 ⫾ 0.03

⫺/⫺

All tested C2 serad

C2⫺/⫺ serum from subject: 1 2 3 4 5

470 ⫾ 90

1,190 ⫾ 430

1.3 ⫾ 0.5

780 ⫾ 80

370 ⫾ 20 410 ⫾ 80 430 ⫾ 60 590 ⫾ 30 530 ⫾ 70

1,600 ⫾ 120 1,450 ⫾ 400 1,050 ⫾ 50 1,330 ⫾ 180 500 ⫾ 60

2.1 ⫾ 0.1 0.8 ⫾ 0.02 1.5 ⫾ 0.1 0.9 ⫾ 0.1 1.4 ⫾ 0.1

820 ⫾ 180 1,270 ⫾ 310 880 ⫾ 340 1,040 ⫾ 330 730 ⫾ 80 870 ⫾ 350 800 ⫾ 120 890 ⫾ 280 680 ⫾ 120 1,140 ⫾ 240

Total IgM binding (FI ⫾ SD)

9,580 ⫾ 1490 1,380 ⫾ 320 10,240 ⫾ 950 1,920 ⫾ 300 9,350 ⫾ 540 1,270 ⫾ 230

Anticapsular IgG level (␮g/ml ⫾ SD)

4.9 ⫾ 0.2 5.3 ⫾ 0.4 2.5 ⫾ 0.1

1,050 ⫾ 150

1.3 ⫾ 0.1

730 ⫾ 70

780 ⫾ 140

1.3 ⫾ 0.3

670 ⫾ 70 730 ⫾ 100 660 ⫾ 70 820 ⫾ 70 760 ⫾ 120

690 ⫾ 40 860 ⫾ 120 710 ⫾ 30 650 ⫾ 10 980 ⫾ 180

1.3 ⫾ 0.1 0.7 ⫾ 0.03 1.5 ⫾ 0.1 1.4 ⫾ 0.03 1.4 ⫾ 0.03

a

MFIs are expressed in arbitrary units. Normal human sera (NHS) were the control sera used for experiments with C2⫺/⫺ sera. In heat-treated normal human sera (NHS), levels of IgG Abs to the ST2 capsular antigen were not significantly different from those in untreated normal human sera (2.1 ⫾ 0.3 ␮g/ml). d Mean results for C2⫺/⫺ subjects 1 to 5. b c

shown), suggesting that these S. pneumoniae strains were not able to impair classical pathway activity by interacting with C4BP. Conversely, the recognition of S. pneumoniae by natural IgM, Abs with specificity for microbial surface antigens that arise without prior exposure to the pathogen (1), activates the classical pathway (7). We therefore investigated IgM binding to the three S. pneumoniae strains analyzed in this study by using sera from naı¨ve wild-type C57BL/6 mice (known to contain natural Ab to phosphorylcholine) (3, 7), with sera from ␮s⫺/⫺ mice (able to produce cell surface but not secreted IgM and hence deficient in natural IgM) (15) as a negative control. Mouse sera were used for these experiments because the universal prior exposure of human subjects to S. pneumoniae prevents the investigation of natural IgM in human sera. In wild-type sera, IgM bound to means of 18.6% (standard deviation [SD], 2.3%) of the cells of the ST2 strain, 11.9% (SD, 0.9%) of the cells of the ST4 strain, and 14.0% (SD, 2.3%) of the cells of the ST23F strain, compared to less than 2% of cells of all three strains in the negative control ␮s⫺/⫺ sera. These data confirm results obtained previously with the ST2 strain (7) and suggest that natural IgM may mediate classical pathway complement activation for different S. pneumoniae strains. The phagocytosis of S. pneumoniae is impaired in human serum deficient in C1q. The effects of C1q and the alternative pathway on the phagocytosis of S. pneumoniae were investigated using a flow cytometry assay that analyzes the association of fluorescently labeled bacteria with cells of the human neutrophil cell line HL-60 (30). In 5 and 10% human sera, a 10 ␮M concentration of the inhibitor of actin polymerization, cytochalasin D, prevented 85 and 91%, respectively, of the increase in the association of fluorescent ST2 S. pneumoniae cells with HL-60 cells compared to results for incubation in HBSS alone. These data indicate that the majority of the serum effect on the association of fluorescent S. pneumoniae cells with HL-60 cells

is due to phagocytosis rather than the adhesion of bacteria to the cell surface. The serum effect was mainly complement dependent, as the association of S. pneumoniae with HL-60 cells after incubation in heat-killed sera, in which complement is denatured (0% activity when tested using the Binding Site total hemolytic complement kit) but type-specific Ab levels are not affected (Table 1), was only slightly more efficient than that after incubation in HBSS buffer alone (see Table 3). By using this flow cytometry assay, results for the ST2, ST4, and ST23F strains incubated in C9⫺, Bf⫺, and C1q⫺ sera were obtained. The phagocytosis of S. pneumoniae was serum dependent, with a large increase in the proportion of HL-60 cells associated with fluorescent bacteria of all strains after incubation in C9⫺ sera compared to the proportion of such cells after incubation in HBSS buffer alone (Fig. 2). In C1q⫺ sera, phagocytosis was markedly impaired; at a concentration of 2.5%, C1q⫺ sera gave results similar to those obtained with HBSS alone for all three strains, while at the same concentration, C9⫺ sera increased the efficiency of phagocytosis by approximately 40%. Supplementing C1q⫺ sera with the equivalent of 75 ␮g of purified human C1q/ml of 100% serum restored the phagocytosis of S. pneumoniae to levels close to those obtained with C9⫺ sera, confirming that impaired phagocytosis in C1q⫺ sera was due mainly to the loss of C1q (Fig. 2E). In contrast, despite significant impairment of complement deposition onto cells of the ST4 and ST23F strains in Bf⫺ sera, phagocytosis in Bf⫺ sera was relatively mildly impaired, with differences compared to phagocytosis in C9⫺ sera occurring only at high serum dilutions. In common with the results for C3b-iC3b deposition, the results of the phagocytosis assays for the different complement-deficient sera did not correlate with Ab levels. Total levels of IgG and IgM binding to all STs in C1q⫺ sera were generally either equal to or higher than those in C9⫺ sera, and although the levels of capsule-specific Abs in C1q⫺ sera were

VOL. 76, 2008

C2 DEFICIENCY AND STREPTOCOCCUS PNEUMONIAE

3765

FIG. 2. (A to C) Results of flow cytometry phagocytosis assays for fluorescent capsular ST2 (A), ST4 (B), and ST23F (C) strains incubated in dilutions of human sera depleted of the complement component C9, Bf, or C1q and then mixed with human neutrophil HL-60 cells. Results are presented as the proportions of cells associated with fluorescent bacteria. Bars labeled HBSS mark the results for bacteria incubated in HBSS alone. (D) Example of a flow cytometry histogram for phagocytosis of the ST23F strain incubated in 2.5% sera. The gray shading indicates the results for bacteria incubated in HBSS alone. (E) Phagocytosis results for the ST2 strain incubated in 2.5% human sera depleted of C1q, with and without complementation with the equivalent of 75 ␮g of purified C1q or BSA/ml of 100% serum, compared to results for the same strain incubated in C9-deficient sera. For all panels, error bars indicate the SD, and data marked with asterisks are statistically significantly different from the results for C9-deficient sera (one-way ANOVA; P ⬍ 0.01).

lower than those in C9⫺ sera for the ST2 strain, the phagocytosis of all three strains in C1q⫺ sera was impaired (Table 1). These data suggest that classical pathway activity has a major influence on the efficiency of the phagocytosis of S. pneumoniae and is of greater importance than either the alternative pathway or Ab levels in sera from unvaccinated subjects. The phagocytosis of S. pneumoniae in mice is mainly classical pathway dependent. To further confirm that the abovedescribed results for phagocytosis in human sera from unvaccinated subjects are likely to reflect complement deficiency rather than differences in Abs, the phagocytosis assays were repeated using sera obtained from mice that had had no prior exposure to S. pneumoniae and that did not have IgG specific for S. pneumoniae (7). The phagocytosis assays were performed using sera from mice deficient in C1q (C1qa⫺/⫺ mice), Bf (Bf⫺/⫺ mice), or C3 (C3⫺/⫺ mice, as a negative control) (Fig. 3). The phagocytosis of all three strains in C3⫺/⫺ sera was

no different from the results obtained with HBSS buffer alone, demonstrating the importance of complement for the phagocytosis of S. pneumoniae in mouse serum. For the ST2 and ST23F strains, the loss of C1q markedly reduced phagocytosis in naı¨ve mouse sera. The phagocytosis of the ST4 strain in sera from C1qa⫺/⫺ mice was also reduced, albeit to a lesser extent than that of the ST2 and ST23F strains (Fig. 3). The loss of alternative pathway activity in mouse sera also reduced the phagocytosis of all three strains, but to a lesser extent than the loss of classical pathway activity reduced the phagocytosis of the ST2 and ST23F strains. Overall, the results obtained with mouse sera reinforce the data obtained using human sera and suggest that the loss of classical pathway activity has an important effect on the phagocytosis of S. pneumoniae independent of acquired Ab. C3b-iC3b deposition onto S. pneumoniae cells in sera from patients with a C2 deficiency. To investigate why patients

3766

YUSTE ET AL.

INFECT. IMMUN.

FIG. 3. (A to C) Results of flow cytometry phagocytosis assays for fluorescent capsular ST2 (A), ST4 (B), and ST23F (C) strains incubated in dilutions of sera obtained from wild-type (wt), C1qa⫺/⫺, Bf⫺/⫺, or C3⫺/⫺ mice and then mixed with human neutrophil HL-60 cells. Results are presented as the proportions of cells associated with fluorescent bacteria. The ⫻’s mark the results for bacteria incubated in HBSS alone. Error bars indicate the SDs, and data marked with asterisks are statistically significantly different from the results for wild-type sera (Student’s t test; *, P ⬍ 0.05; **, P ⬍ 0.01). (D) Example of a flow cytometry histogram for phagocytosis of the ST2 strain incubated in 5% mouse sera.

with a C2 deficiency are highly susceptible to S. pneumoniae, the C3b-iC3b deposition and phagocytosis assays were repeated using sera obtained from nine C2⫺/⫺ subjects (median age, 40 years; range, 12 to 51 years) who had not been vaccinated against S. pneumoniae. The results were compared to those for a control serum preparation consisting of pooled sera from three normal unvaccinated subjects. The levels of C3b-iC3b deposition onto cells of the ST2 S. pneumoniae strain confirmed that the control sera were representative of sera obtained from 20 normal subjects (median age, 32 years; range, 23 to 42 years). After incubation in 20% control sera, the mean proportion of bacteria positive for C3b-iC3b was 76.0% (SD, 4.0%) and the MFI indicating C3b-iC3b deposition was 104 (SD, 6.1), compared to a mean proportion of positive bacteria of 70.6% (SD, 9.0%) and an MFI of 134 (SD, 52) for the sera from the 20 normal sub-

jects. C3b-iC3b deposition onto cells of the ST2 strain was strikingly impaired in 50% sera from all the C2⫺/⫺ subjects, with a median MFI of 12.7% of the results for the control sera (Table 2 and Fig. 4A). The effect of complementation of C2⫺/⫺ sera with physiological levels of exogenous C2 was investigated with sera from subjects 1 to 5 (sera from subjects 6 to 9 were not investigated due to a limited supply). In all cases, C3b-iC3b deposition onto cells of the ST2 strain was improved, with a median MFI of 107% of the results for controls after incubation in C2⫺/⫺ sera complemented with exogenous C2 (Table 2 and Fig. 4A), demonstrating that impaired C3b-iC3b deposition in C2⫺/⫺ sera was due to the loss of C2 alone. To confirm that impaired C3b-iC3b deposition onto cells of the S. pneumoniae ST2 strain in C2⫺/⫺ sera represented a defect in complement activity against a range of S. pneumoniae strains, the assays were repeated

TABLE 2. Geometric MFIs for C3b-iC3b deposition onto cells of the S. pneumoniae ST2, ST4, and ST23F strains in 50% sera from up to nine subjects with a homozygous C2 deficiencya Geometric MFI ⫾ SD for: ⫺/⫺

Strain

ST2 ST2 ⫹ C2b ST4 ST23F

PBS

Control sera

20

449 ⫾ 128

15 30

273 ⫾ 32 191 ⫾ 50

⫺/⫺

⫺/⫺

C2 serum from subject 1

C2 serum from subject 2

C2 serum from subject 3

C2⫺/⫺ serum from subject 4

C2⫺/⫺ serum from subject 5

C2⫺/⫺ serum from subject 6

C2⫺/⫺ serum from subject 7

C2⫺/⫺ serum from subject 8

C2⫺/⫺ serum from subject 9

60 ⫾ 3.9 630 ⫾ 45 34 ⫾ 21 47 ⫾ 14

57 ⫾ 6 465 ⫾ 77 16 ⫾ 1 26 ⫾ 5

37 ⫾ 1 617 ⫾ 38 15 ⫾ 1 24 ⫾ 3

57 ⫾ 7 490 ⫾ 53 14 ⫾ 1 27 ⫾ 6

62 ⫾ 5 482 ⫾ 22 20 ⫾ 6 30 ⫾ 5

40 ⫾ 7 NDc ND ND

26 ⫾ 6 ND ND ND

88 ⫾ 2 ND ND ND

32 ⫾ 2 ND ND ND

MFIs are expressed in arbitrary units. For the comparison of all results for uncomplemented C2⫺/⫺ sera to results for control sera, P was ⬍0.001 (ANOVA). The ST2 strain in C2⫺/⫺ sera complemented with 12.5 ␮g of exogenous purified C2/ml. c ND, not determined. a b

VOL. 76, 2008

C2 DEFICIENCY AND STREPTOCOCCUS PNEUMONIAE

3767

FIG. 4. Box-and-whisker summary of results for C3b-iC3b opsonization and phagocytosis for the S. pneumoniae ST2, ST4, and ST23F strains in sera from C2⫺/⫺ subjects. (A) Geometric MFIs for C3b-iC3b deposition onto S. pneumoniae cells in 50% sera from subjects with a C2 deficiency (subjects 1 to 9 for ST2 and subjects 1 to 5 for ST4, ST23F, and the ST2 strain complemented with 12.5 ␮g of exogenous C2/ml [ST2⫹C2]), expressed as percentages of the results for control sera. (B) Example of the flow cytometry histogram for C3b-iC3b deposition onto the ST23F strain in PBS, control sera, and C2⫺/⫺ serum from subject 5. (C) Percentages of HL-60 cells associated with FAMSE-labeled S. pneumoniae in 20% sera from subjects with a C2 deficiency (subjects 1 to 9 for all strains), expressed as percentages of the results for control sera. (D) Example of the flow cytometry histogram for phagocytosis of the FAMSE-labeled ST2 strain in HBSS, control sera, and C2⫺/⫺ serum from subject 5. P values were calculated using the Kruskal-Wallis test.

using sera from C2⫺/⫺ subjects 1 to 5 and the ST4 and ST23F strains. For both strains, C3b-iC3b deposition was strongly reduced in sera from the C2⫺/⫺ subjects, with median MFIs of 5.9 and 14.4% of the results for control sera, respectively (Table 2 and Fig. 4A and B). The phagocytosis of S. pneumoniae in sera from patients with a C2 deficiency is impaired. To investigate whether reduced levels of C3b deposition onto S. pneumoniae cells in sera from C2⫺/⫺ subjects had functional consequences, phagocytosis assays were performed using the C2⫺/⫺ sera. Although the results showed some variation among subjects, overall, the

phagocytosis of the ST2 S. pneumoniae strain in 20% sera from C2-deficient subjects was significantly impaired, with a median level of 60.9% of the results obtained with control sera (Table 3 and Fig. 4C and D). Due to the variation in results, the effects of the complementation of C2⫺/⫺ sera with physiological quantities of exogenous C2 were assessed for all available sera individually rather than a subset of sera. For seven of the nine C2⫺/⫺ sera, complementation with C2 at least partially restored the phagocytosis of the ST2 S. pneumoniae strain, giving a median of 79.6% of the results obtained with control sera (Table 3 and Fig. 4C). The phagocytosis of other S. pneu-

TABLE 3. Proportion of HL-60 cells associated with FAMSE-labeled S. pneumoniae ST2, ST4, and ST23F strains after incubation in 20% sera from nine subjects with a homozygous C2 deficiencya Mean % ⫾ SD of associated cells after incubation in: Strain

ST2 ST2 ⫹ C2b ST4 ST23F

HBSS

Control sera

Heat-treated sera

5 ⫾ 0.4

54 ⫾ 9

12 ⫾ 2

11 ⫾ 0.2 15 ⫾ 1

77 ⫾ 5 58 ⫾ 2

18 ⫾ 2 ND

C2⫺/⫺ serum from subject 1

C2⫺/⫺ serum from subject 2

C2⫺/⫺ serum from subject 3

C2⫺/⫺ serum from subject 4

C2⫺/⫺ serum from subject 5

C2⫺/⫺ serum from subject 6

C2⫺/⫺ serum from subject 7

C2⫺/⫺ serum from subject 8

C2⫺/⫺ serum from subject 9

27 ⫾ 1 29 ⫾ 1 46 ⫾ 2 32 ⫾ 2

34 ⫾ 2 43 ⫾ 3 NDc 30 ⫾ 2

32 ⫾ 3 43 ⫾ 3 51 ⫾ 8 34 ⫾ 2

30 ⫾ 3 43 ⫾ 3 48 ⫾ 1 33 ⫾ 1

28 ⫾ 2 40 ⫾ 6 50 ⫾ 1 36 ⫾ 3

29 ⫾ 1 43 ⫾ 4 76 ⫾ 1 34 ⫾ 1

34 ⫾ 3 46 ⫾ 1 48 ⫾ 1 32 ⫾ 2

39 ⫾ 2 42 ⫾ 2 77 ⫾ 1 35 ⫾ 2

41 ⫾ 5 55 ⫾ 9 52 ⫾ 4 31 ⫾ 1

For the comparison of all results for uncomplemented C2⫺/⫺ sera to results for control sera, P was ⬍0.001 (ANOVA). The ST2 strain in C2⫺/⫺ sera complemented with 5 ␮g of exogenous purified C2/ml. c ND, not determined. a b

3768

YUSTE ET AL.

moniae strains in the majority of the C2⫺/⫺ sera tested was also impaired, with median values for the ST4 and ST23F strains of 64.0 and 56.5%, respectively, of the results obtained with control sera (Table 3 and Fig. 4C). Experiments using 5 and 2.5% sera gave similar results, with impaired phagocytosis of the ST2, ST4, and ST23F strains in the majority of sera from subjects with a homozygous C2 deficiency (data not shown). The possible effect of Ab levels on the results obtained with C2⫺/⫺ sera was investigated by measuring total IgG and IgM binding and levels of capsule-specific IgG Abs to the three STs in sera from C2⫺/⫺ subjects 1 to 5. Total levels of IgG and IgM binding to the ST2, ST4, and ST23 strains in sera from C2⫺/⫺ subjects were consistently reduced, to mean levels of IgG binding of 44, 45, and 57% of the results obtained with control sera and mean levels of IgM binding of 44, 44, and 74% of the results obtained with control sera, respectively (Table 1). However, although mean levels of capsule-specific Abs to the ST2 strain in the C2⫺/⫺ sera were lower than those in normal human sera, there was considerable variation in results for individual sera. In general, the levels of anticapsular Ab did not correlate with the levels of phagocytosis, with sera containing low and high levels of capsule-specific IgG giving similar results in the phagocytosis assays (for examples, see the results for Ab levels and the phagocytosis of the ST4 strain in C2⫺/⫺ sera from subjects 1 and 4, given in Tables 1 and 3). Hence, subjects with no classical pathway activity have a significantly reduced ability to phagocytose S. pneumoniae, and the data obtained with complemented sera and Ab levels suggest that this reduction is due partially to the loss of complement activity and partially to reduced levels of total, but not capsule-specific, Ab. DISCUSSION Considerable experimental data support a vital role for complement in immunity to S. pneumoniae, with older data tending to emphasize the importance of the alternative pathway (5, 7, 12, 19, 29, 37, 42, 46, 47). However, recent data obtained using mice genetically modified to be deficient in one or more complement components have identified a vital role for classical pathway activity in innate immunity to S. pneumoniae and other pathogens and have demonstrated the classical pathway to be activated by a variety of innate immune mediators, including pentraxins, natural IgM, and the cell surface lectin SIGN-R1 (7, 11, 18, 29, 41, 44, 48, 49). An important role for the classical pathway in immunity to S. pneumoniae is also supported by the limited clinical data available on infections in patients with complement deficiencies, as patients with defects in the classical pathway are at high risk of S. pneumoniae infections (10, 27, 28). In contrast, patients with complement deficiencies affecting the alternative pathway have a massively increased incidence of infections with Neisseria species rather than S. pneumoniae (10). However, although previous studies have partially investigated the role of complement pathways in immunity to S. pneumoniae, data suggesting the importance of the classical pathway in mice need to be supported by data obtained with human sera. Furthermore, although reduced opsonization of S. pneumoniae with C3b-iC3b is likely to result in impaired phagocytosis, this effect may be partially compensated for by other phagocytic mediators (including Ab), and

INFECT. IMMUN.

the increased susceptibility of humans or mice with classical pathway defects to S. pneumoniae infections may be due mainly to the impairment of complement-dependent stimulation of inflammatory and/or Ab responses (7, 21, 43). Hence, to investigate the link between the classical pathway and immunity to S. pneumoniae in more detail, we have now characterized C3b-iC3b deposition onto cells of three strains in human sera deficient in classical or alternative pathway activity, including sera obtained from patients with a homozygous C2 deficiency, and the functional consequences of classical pathway defects for the phagocytosis of S. pneumoniae. The results from the present studies demonstrate that in human serum the classical pathway is essential for C3b-iC3b deposition onto cells of the S. pneumoniae strains investigated, with a role at least as important as that of the alternative pathway. Furthermore, we have shown the clinical importance of these observations obtained with modified normal human sera by demonstrating that C3b deposition onto S. pneumoniae cells was also strikingly impaired in sera from subjects with a homozygous deficiency of the classical pathway component C2. Although it may not be surprising that C2 deficiency results in reduced C3b-iC3b deposition onto S. pneumoniae cells, what is striking is how powerful and consistent the effect was, giving results for all the sera tested that were less than 25% of those obtained with control sera for three different S. pneumoniae STs. The restoration of classical pathway activity by the addition of C2 resulted in the return of C3b-iC3b deposition to normal levels, confirming that the reduction in C3b-iC3b deposition onto S. pneumoniae cells in C2-deficient sera was caused by the loss of complement activity rather than other potential differences in the control sera. These data emphasize the importance of an intact classical pathway for complement deposition onto S. pneumoniae cells in sera from actual patients. The profound effect of C2 deficiency may be due to the loss of classical pathway activation by acquired Ab, by innate immune mediators such as CRP and SAP, or as we believe is most likely, by a combination of both. In addition, C2 is required for MBL pathway activity, and although this pathway seems to have little effect on C3b-iC3b deposition in mouse serum (7), it may have contributed to the results obtained with normal human sera. Which complement pathways are important for complement deposition onto a particular pathogen is probably dictated by the combination of bacterial products that interact with complement (e.g., the capsule or surface proteins like PspA for S. pneumoniae and M protein for S. pyogenes) (24, 48) and which complement activators such as CRP, SAP, and natural IgM can bind to the pathogen (3, 7, 41, 49). For example, S. pyogenes binds to the classical pathway inhibitor C4BP (9, 48), in contrast to our results showing that S. pneumoniae does not bind to C4BP but is recognized by the classical pathway activator natural IgM. These interactions with different aspects of the complement system may partially explain why the classical pathway generally has a weaker role in C3b-iC3b deposition onto S. pyogenes cells than onto S. pneumoniae cells (7, 48), despite both PspA and pneumolysin inhibiting classical pathway activity against S. pneumoniae (31, 38, 42, 50). Additionally, the binding of factor H by the surface protein PspC is thought to inhibit alternative pathway-mediated immunity (13, 35), and this arrangement may explain the relatively weak

VOL. 76, 2008

effect of the alternative pathway on C3b-iC3b deposition onto S. pneumoniae cells. More generally, these data show that the roles of different complement pathways in immunity vary among even relatively closely related pathogens. Therefore, research findings on mechanisms of host immunity to one pathogen cannot necessarily be assumed to apply to other pathogens. For extracellular pathogens such as S. pneumoniae, phagocytosis is thought to be an important immune mechanism for controlling infection (4, 6). Previous publications have shown that the phagocytosis of S. pneumoniae is dependent on complement (20, 35), and C1q deficiency in intravenously infected mice impairs S. pneumoniae clearance from the blood, which is dependent mainly on phagocytosis (4, 7, 22). These data suggest that reduced C3b-iC3b deposition onto S. pneumoniae cells in C1q-deficient serum would result in impaired phagocytosis, and we have confirmed this pattern for the three S. pneumoniae strains investigated. Efficient phagocytosis required serum with intact complement activity, and phagocytosis was impaired in both human and mouse C1q-deficient sera and in sera from most C2-deficient subjects. The loss of Bf had generally weak effects on phagocytosis in human sera, and these results did not closely correlate with the effects on C3biC3b deposition, the reason for which is not clear. The greater effect of the loss of C1q on phagocytosis than of the loss of Bf reflects partially the differences in C3b-iC3b deposition levels in the corresponding sera but also may be influenced by the direct effect of C1q on phagocytosis that has been described previously for other pathogens (39, 48). Exogenous C1q complemented the phagocytosis defect in C1q-deficient sera, strongly suggesting that the lack of classical pathway activity was responsible for the impairment of phagocytosis in these sera. In order to ensure that our data are generally representative of the characteristics of different S. pneumoniae strains, we have presented results for three S. pneumoniae STs. Overall, the patterns of the results were similar among strains, but there were differences that may have important consequences during actual infection. For example, the classical pathway seemed to have a stronger effect on C3b-iC3b deposition onto cells of the ST23F strain than onto those of the other strains, and the ST2 strain was generally more resistant to phagocytosis in both human and mouse sera than the other strains. These differences between capsular STs need to be investigated in more detail and may explain some of the differences in invasive potential that have been described previously for different capsular STs (8). Although human sera contain serum Abs to a variety of S. pneumoniae antigens, including the capsule and several subcapsular proteins, our data for control sera obtained from unvaccinated individuals suggest that complement is the most important serum component driving phagocytosis. Results for opsonization with heat-treated sera with preserved Ab but no complement activity were only slightly higher than those obtained with HBSS buffer alone, suggesting that any effects of Ab are mediated through classical pathway activity rather than through direct interactions with phagocyte Fc gamma receptors. In contrast, phagocytosis data obtained with the C2⫺/⫺ sera do suggest that at least in these sera there is an important role for Ab independent of complement. There were consistently lower levels of total IgG and IgM binding to S. pneu-

C2 DEFICIENCY AND STREPTOCOCCUS PNEUMONIAE

3769

moniae in C2⫺/⫺ sera than in the control sera, and impaired complement-independent Ab-mediated phagocytosis would explain why complementation with C2 restored the levels of phagocytosis in C2⫺/⫺ sera to only around 80% of that in control sera despite restoring the levels of C3b-iC3b opsonization to over 100%. In experimental models, complement activity is required for efficient Ab responses, and the low levels of anti-S. pneumoniae IgG and IgM in C2⫺/⫺ subjects despite their high incidence of S. pneumoniae infections suggest that classical pathway activity is also important for Ab responses in humans (43). Although there has been considerable emphasis on the role of anticapsular Ab for immunity to S. pneumoniae in vaccinated individuals, the various levels of anticapsular IgG in the different sera used for this study did not correlate with differences in the efficiency of phagocytosis. Hence, our data support the contention that the main mechanism of natural immunity to S. pneumoniae in unvaccinated individuals is not Ab to capsular antigen (33) and indicate that complement and Abs to noncapsular antigens are likely to be important. In summary, our results confirm that for S. pneumoniae, the classical pathway has a central role in both C3b deposition onto the bacterial surface and phagocytosis in human sera. As a consequence, complement activation and the phagocytosis of S. pneumoniae in sera from subjects with a homozygous C2 deficiency were markedly impaired. In addition, overall anti-S. pneumoniae Ab levels in sera from C2⫺/⫺ subjects were reduced compared to the levels in normal sera. These data provide a biological explanation for the observation that patients with complement deficiencies affecting the classical pathway have a high incidence of S. pneumoniae infections (28) and provide strong support for the importance of classical pathwaymediated innate immunity to S. pneumoniae in humans. ACKNOWLEDGMENTS We acknowledge the major contribution and support for this project given by Anders G. Sjo ¨holm, who unfortunately died before its completion. This work was funded by a Wellcome Trust project grant (reference no. 066335), a British Lung Foundation project grant (grant no. P05/3), a Neil Hamilton Fairley postdoctoral fellowship (National Health & Medical Research Council, Australia) awarded to Fiona Goldblatt, and a grant from the Swedish Research Council (grant no. 15092) to Lennart Truedsson. REFERENCES 1. Boes, M. 2000. Role of natural and immune IgM antibodies in immune responses. Mol. Immunol. 37:1141–1149. 2. Botto, M., C. Dell’Agnola, A. E. Bygrave, E. M. Thompson, H. T. Cook, F. Petry, M. Loos, P. P. Pandolfi, and M. J. Walport. 1998. Homozygous C1q deficiency causes glomerulonephritis associated with multiple apoptotic bodies. Nat. Genet. 19:56–59. 3. Briles, D. E., M. Nahm, K. Schroer, J. Davie, P. Baker, J. Kearney, and R. Barletta. 1981. Antiphosphocholine antibodies found in normal mouse serum are protective against intravenous infection with type 3 Streptococcus pneumoniae. J. Exp. Med. 153:694–705. 4. Brown, E. J., S. W. Hosea, and M. M. Frank. 1981. Reticuloendothelial clearance of radiolabelled pneumococci in experimental bacteremia: correlation of changes in clearance rates, sequestration patterns, and opsonization requirements at different phases of the bacterial growth cycle. J. Reticuloendothel. Soc. 30:23–31. 5. Brown, E. J., S. W. Hosea, and M. M. Frank. 1983. The role of antibody and complement in the reticuloendothelial clearance of pneumococci from the bloodstream. Rev. Infect. Dis. 5:S797–S805. 6. Brown, E. J., S. W. Hosea, and M. M. Frank. 1981. The role of complement in the localization of pneumococci in the splanchnic reticuloendothelial system during experimental bacteremia. J. Immunol. 126:2230–2235. 7. Brown, J. S., T. Hussell, S. M. Gilliland, D. W. Holden, J. C. Paton, M. R.

3770

8.

9.

10. 11.

12.

13.

14.

15.

16.

17. 18.

19.

20.

21.

22.

23.

24. 25.

26.

27.

28.

YUSTE ET AL.

Ehrenstein, M. J. Walport, and M. Botto. 2002. The classical pathway is the dominant complement pathway required for innate immunity to Streptococcus pneumoniae infection in mice. Proc. Natl. Acad. Sci. USA 99:16969– 16974. Brueggemann, A. B., D. T. Griffiths, E. Meats, T. Peto, D. W. Crook, and B. G. Spratt. 2003. Clonal relationships between invasive and carriage Streptococcus pneumoniae and serotype- and clone-specific differences in invasive disease potential. J. Infect. Dis. 187:1424–1432. Carlsson, F., K. Berggård, M. Stålhammar-Carlemalm, and G. Lindahl. 2003. Evasion of phagocytosis through cooperation between two ligandbinding regions in Streptococcus pyogenes M protein. J. Exp. Med. 198:1057– 1068. Carneiro-Sampaio, M., and A. Coutinho. 2007. Immunity to microbes: lessons from primary immunodeficiencies. Infect. Immun. 75:1545–1555. Celik, I., C. Stover, M. Botto, S. Thiel, S. Tzima, D. Kunkel, M. Walport, W. Lorenz, and W. Schwaeble. 2001. Role of the classical pathway of complement activation in experimentally induced polymicrobial peritonitis. Infect. Immun. 69:7304–7309. Chudwin, D. S., D. W. Wara, N. B. Lameris-Martin, and A. J. Ammann. 1983. Effect of antibody concentration on opsonic requirements for phagocytosis in vitro of Streptococcus pneumoniae types 7 and 19. Proc. Soc. Exp. Biol. Med. 172:178–186. Clatworthy, M. R., and K. G. Smith. 2004. Fc␥RIIb balances efficient pathogen clearance and the cytokine-mediated consequences of sepsis. J. Exp. Med. 199:717–723. Dave, S., A. Brooks-Walter, M. K. Pangburn, and L. S. McDaniel. 2001. PspC, a pneumococcal surface protein, binds human factor H. Infect. Immun. 69:3435–3437. Ehrenstein, M. R., T. L. O’Keefe, S. L. Davies, and M. S. Neuberger. 1998. Targeted gene disruption reveals a role for natural secretory IgM in the maturation of the primary immune response. Proc. Natl. Acad. Sci. USA 95:10089–10093. Exley, R. M., J. Shaw, E. Mowe, Y. H. Sun, N. P. West, M. Williamson, M. Botto, H. Smith, and C. M. Tang. 2005. Available carbon source influences the resistance of Neisseria meningitidis against complement. J. Exp. Med. 201:1637–1645. Fine, D. P. 1975. Pneumococcal type-associated variability in alternate complement pathway activation. Infect. Immun. 12:772–778. Garlanda, C., E. Hirsch, S. Bozza, A. Salustri, M. De Acetis, R. Nota, A. Maccagno, F. Riva, B. Bottazzi, G. Peri, A. Doni, L. Vago, M. Botto, R. De Santis, P. Carminati, G. Siracusa, F. Altruda, A. Vecchi, L. Romani, and A. Mantovani. 2002. Non-redundant role of the long pentraxin PTX3 in antifungal innate immune response. Nature 420:182–186. Giebink, G. S., J. Verhoef, P. K. Peterson, and P. G. Quie. 1977. Opsonic requirements for phagocytosis of Streptococcus pneumoniae types VI, XVIII, XXIII, and XXV. Infect. Immun. 18:291–297. Gordon, S. B., G. R. Irving, R. A. Lawson, M. E. Lee, and R. C. Read. 2000. Intracellular trafficking and killing of Streptococcus pneumoniae by human alveolar macrophages are influenced by opsonins. Infect. Immun. 68:2286– 2293. Haas, K., M. Hasegawa, D. Steeber, J. Poe, M. Zabel, C. Bock, D. Karp, D. Briles, J. Weis, and T. Tedder. 2002. Complement receptors CD21/35 link innate and protective immunity during Streptococcus pneumoniae infection by regulating IgG3 antibody responses. Immunity 17:713–723. Hosea, S. W., E. J. Brown, and M. M. Frank. 1980. The critical role of complement in experimental pneumococcal sepsis. J. Infect. Dis. 142:903– 909. Janulczyk, R., F. Iannelli, A. G. Sjo ¨holm, G. Pozzi, and L. Bjorck. 2000. Hic, a novel surface protein of Streptococcus pneumoniae that interferes with complement function. J. Biol. Chem. 275:37257–37263. Jarva, H., T. S. Jokiranta, R. Wurzner, and S. Meri. 2003. Complement resistance mechanisms of streptococci. Mol. Immunol. 40:95–107. Jomaa, M., S. Terry, C. Hale, G. Dougan, and J. S. Brown. 2006. Immunisation with the iron uptake ABC transporter proteins PiaA and PiuA prevents pulmonary infection with Streptococcus pneumoniae. Vaccine 24:5133– 5139. Jomaa, M., J. Yuste, J. C. Paton, C. Jones, G. Dougan, and J. S. Brown. 2005. Antibodies to the iron uptake ABC transporter lipoproteins PiaA and PiuA promote opsonophagocytosis of Streptococcus pneumoniae. Infect. Immun. 73:6852–6859. Jo ¨nsson, G., V. A. Oxelius, L. Truedsson, J. H. Braconier, G. Sturfelt, and A. G. Sjo ¨holm. 2006. Homozygosity for the IgG2 subclass allotype G2M(n) protects against severe infection in hereditary C2 deficiency. J. Immunol. 177:722–728. Jo ¨nsson, G., L. Truedsson, G. Sturfelt, V. A. Oxelius, J. H. Braconier, and A. G. Sjo ¨holm. 2005. Hereditary C2 deficiency in Sweden: frequent occurrence of invasive infection, atherosclerosis, and rheumatic disease. Medicine (Baltimore) 84:23–34.

Editor: J. N. Weiser

INFECT. IMMUN. 29. Kang, Y.-S., Y. Do, H.-K. Lee, S. H. Park, C. Cheong, R. M. Lynch, J. M. Loeffler, R. M. Steinman, and C. G. Park. 2006. A dominant complement fixation pathway for pneumococcal polysaccharides initiated by SIGN-R1 interacting with C1q. Cell 125:47–58. 30. Lehmann, A., S. Sornes, and A. Halstensen. 2000. Phagocytosis: measurement by flow cytometry. J. Immunol. Methods 243:229–242. 31. Li, J., D. T. Glover, A. J. Szalai, S. K. Hollingshead, and D. E. Briles. 2007. PspA and PspC minimize immune adherence and transfer of pneumococci from erythrocytes to macrophages through their effects on complement activation. Infect. Immun. 75:5877–5885. 32. Lim, W. S., J. T. Macfarlane, T. C. Boswell, T. G. Harrison, D. Rose, M. Leinonen, and P. Saikku. 2001. Study of community acquired pneumonia aetiology (SCAPA) in adults admitted to hospital: implications for management guidelines. Thorax 56:296–301. 33. Lipsitch, M., C. G. Whitney, E. Zell, T. Kaijalainen, R. Dagan, and R. Malley. 2005. Are anticapsular antibodies the primary mechanism of protection against invasive pneumococcal disease? PLoS Med. 2:e15. 34. Matsumoto, M., W. Fukuda, A. Circolo, J. Goellner, J. Strauss-Schoenberger, X. Wang, S. Fujita, T. Hidvegi, D. D. Chaplin, and H. R. Colten. 1997. Abrogation of the alternative complement pathway by targeted deletion of murine factor B. Proc. Natl. Acad. Sci. USA 94:8720–8725. 35. Matthay, K. K., W. C. Mentzer, D. W. Wara, H. K. Preisler, N. B. Lameris, and A. J. Ammann. 1981. Evaluation of the opsonic requirements for phagocytosis of Streptococcus pneumoniae serotypes VII, XIV, and XIX by chemiluminescence assay. Infect. Immun. 31:228–235. 36. Neeleman, C., S. P. Geelen, P. C. Aerts, M. R. Daha, T. E. Mollnes, J. J. Roord, G. Posthuma, H. van Dijk, and A. Fleer. 1999. Resistance to both complement activation and phagocytosis in type 3 pneumococci is mediated by the binding of complement regulatory protein factor H. Infect. Immun. 67:4517–4524. 37. Nilsson, M., N. Weineisen, T. Andersson, L. Truedsson, and U. Sjo ¨bring. 2005. Critical role for complement receptor 3 (CD11b, CD18), but not for Fc-receptors, in killing of Streptococcus pyogenes by neutrophils in immune human serum. J. Immunol. 35:1472–1481. 38. Quin, L. R., Q. C. Moore III, and L. S. McDaniel. 2007. Pneumolysin, PspA, and PspC contribute to pneumococcal evasion of early innate immune responses during bacteremia in mice. Infect. Immun. 75:2067–2070. 39. Rimoldi, M. T., A. J. Tenner, D. A. Bobak, and K. A. Joiner. 1989. Complement component C1q enhances invasion of human mononuclear phagocytes and fibroblasts by Trypanosoma cruzi trypomastigotes. J. Clin. Investig. 84: 1982–1989. 40. Romero-Steiner, S., D. Libutti, L. B. Pais, J. Dykes, P. Anderson, J. C. Whitin, H. L. Keyserling, and G. M. Carlone. 1997. Standardization of an opsonophagocytic assay for the measurement of functional antibody activity against Streptococcus pneumoniae using differentiated HL-60 cells. Clin. Diagn. Lab. Immunol. 4:415–422. 41. Szalai, A. J., D. E. Briles, and J. E. Volanakis. 1996. Role of complement in C-reactive-protein-mediated protection of mice from Streptococcus pneumoniae. Infect. Immun. 64:4850–4853. 42. Tu, A. H., R. L. Fulgham, M. A. McCrory, D. E. Briles, and A. J. Szalai. 1999. Pneumococcal surface protein A inhibits complement activation by Streptococcus pneumoniae. Infect. Immun. 67:4720–4724. 43. Walport, M. J. 2001. Complement. First of two parts. N. Engl. J. Med. 344:1058–1066. 44. Warren, J., P. Mastroeni, G. Dougan, M. Noursadeghi, J. Cohen, M. J. Walport, and M. Botto. 2002. Increased susceptibility of C1q-deficient mice to Salmonella enterica serovar Typhimurium infection. Infect. Immun. 70: 551–557. 45. Wessels, M. R., P. Butko, M. Ma, H. B. Warren, A. L. Lage, and M. C. Carroll. 1995. Studies of group B streptococcal infection in mice deficient in complement component C3 or C4 demonstrate an essential role for complement in both innate and acquired immunity. Proc. Natl. Acad. Sci. USA 92:11490–11494. 46. Winkelstein, J. A. 1981. The role of complement in the host’s defense against Streptococcus pneumoniae. Rev. Infect. Dis. 3:289–298. 47. Xu, Y., M. Ma, G. C. Ippolito, H. W. Schroeder, Jr., M. C. Carroll, and J. E. Volanakis. 2001. Complement activation in factor D-deficient mice. Proc. Natl. Acad. Sci. USA 98:14577–14582. 48. Yuste, J., S. Ali, S. Sriskandan, C. Hyams, M. Botto, and J. S. Brown. 2006. Roles of the alternative pathway and C1q during innate immunity to Streptococcus pyogenes. J. Immunol. 176:6112–6120. 49. Yuste, J., M. Botto, S. E. Bottoms, and J. S. Brown. 2007. Serum amyloid P aids complement-mediated immunity to Streptococcus pneumoniae. PLoS Pathog. 3:1208–1219. 50. Yuste, J., M. Botto, J. C. Paton, D. W. Holden, and J. S. Brown. 2005. Additive inhibition of complement deposition by pneumolysin and PspA facilitates Streptococcus pneumoniae septicaemia. J. Immunol. 175:1813–1819.