Low invasiveness of pneumococcal serotype 11A is linked to ficolin-2 recognition of O-acetylated capsule epitopes and lectin complement pathway activation

Abstract

Background: The divergent epidemiological behavior of Streptococcus pneumoniae serotypes suggests that serotype-specific features such as capsule O-acetylation influence the propensity of a strain to cause invasive pneumococcal disease (IPD). We hypothesize that innate host factors mediate the observed negative association between IPD and the serotype 11A (ST11A) capsule O-acetyltransferase gene, wcjE.

Methods: We evaluated the ability of ficolin-2, an initiator of the lectin complement pathway that was previously shown to bind ST11A pneumococci, to recognize and mediate complement-dependent opsonophagocytosis of different pneumococcal serotypes. We supplemented findings with an epidemiological meta-analysis comparing invasiveness of the 30 most prevalent pneumococcal serotypes.

Results: Ficolin-2 bound ST11A capsule polysaccharide and other wcjE-containing pneumococcal serotypes, except ST9V and ST20B. Ficolin-2 did not bind wcjE-null serotypes, including the wcjE-null variant of ST11A, ST11E. We observed C1q-independent complement deposition and phagocytic killing of pneumococci expressing ST11A but not those expressing ST11E. Inhibition of ficolin-2 binding abrogated ST11A-associated complement deposition and phagocytosis. In children, invasiveness of ST11A was the lowest among serotypes tested in our meta-analysis, while ST9V was among the highest.

Conclusions: Ficolin-2 mediates serum protection by recognizing specific O-acetylated epitopes of pneumococcal capsule polysaccharides, exemplifying a novel host-microbe interaction that innately offers serotype-specific immunity to IPD.

Keywords: Streptococcus pneumoniae; complement; ficolin-2; invasiveness; serotype 11A.

Figure 1.

Flow-cytometric detection of ficolin-2 binding to ST11A pneumococci. Data are representative of at least 3 experiments. A, Flow-cytometric histograms depict Hyp11AM1 (first column), Hyp11AG2 (second column), serum ficolin-2 (third column), and recombinant ficolin-2 (fourth column) binding to the ST11A strain JC03 (top panels) and ST11E strain JC04 (bottom panels). Hyp11AM1 and Hyp11AG2 are murine monoclonal antibodies that bind to the polysaccharide (PS) of both ST11A and ST11E or only ST11A, respectively. Gray-shaded curves depict bacteria incubated with isotype control antibodies and represent negative controls. B, Ficolin-2 bound to JC03 (11A; black diamonds) but not to JC04 (11E; grey circles) after incubation in serial dilutions of serum, quantified as a mean fluorescence intensity (MFI). C, Flow-cytometric histograms depict ficolin-2 binding to ST11A clinical isolates (MNZ273, MNZ272, and MNZ271) but not to ST11E clinical isolates (MNZ741, MNZ266, and MNZ264).

Figure 2.

Specificity of ficolin-2 binding to ST11A bacteria was examined by flow cytometry and expressed as mean fluorescence intensity (MFI). All graphs are representative of at least 3 experiments. A, Serum ficolin-2 binding to ST11A pneumococci in the presence of serially diluted purified polysaccharide (PS) preparations: ST11A PS (solid squares), ST11E PS (open squares), teichoic acid with 1 phosphocholine per repeat unit (TA 1PC; solid circles), teichoic acid with 2 PCs per repeat unit (TA 2PC; open circles), or lipoteichoic acid (LTA; solid diamonds). B, Serum ficolin-2 binding to JC03 incubated in 2.5% serum in the presence of specified concentrations (mM) of ethylenediaminetetraacetic acid (EDTA) and CaCl₂. C, Serum ficolin-2 binding to ST11A pneumococci in the presence of inhibitors: glucose (Glc; open squares), bovine serum albumin (BSA; open circles), N-acetylglucosamine (GlcNAc; solid squares), or acetylated-BSA (acBSA; solid circles). Inhibitor concentrations are mM for Glc and GlcNAc and mg/L for BSA and acBSA. D–E, Flow cytometric detection of ficolin-2 binding to ST11A polysaccharide (PS)–coated latex beads. Data are representative of at least 3 experiments. D, Latex beads were conjugated with purified ST11A PS (solid black curves), ST11E PS (dotted black curves), or 15B PS (gray curves) and were examined for ST11A/11E PS, using Hyp11AM1 monoclonal antibody. E, The same PS-conjugated latex beads used in panel D were incubated in 0.17% normal human serum and examined for ficolin-2 binding, using anti-ficolin-2 mAb GN5.

Figure 3.

Ficolin-2 binding to various pneumococci was examined by flow cytometry. Bacteria were incubated in 2.5% normal human serum and stained with GN5 (anti-ficolin-2) monoclonal antibody (Ab; black curves) or an isotype control Ab (grey curves). All histograms are representative of at least 3 experiments. A strain's serotype or name is listed in the top right corner of each histogram. A, Ficolin-2 binding to pneumococci expressing serogroup 11 serotypes. The presence or absence of GlcNAc (G+ or G−) or WcjE-mediated βGal6-OAc (W+ or W−) on capsule polysaccharide is noted for each serotype. B, Ficolin-2 binding to serotypes whose cps loci encode an intact copy of wcjE. C, Comparison of ficolin-2 binding between serotypes encoding wcjE (W+; ST11D, ST20A, ST33A) and corresponding wcjE-null (W−) recombinant strains (11D× and 20A×) or clinical isolate (ST33F). D, Ficolin-2 binding to pneumococci expressing capsules containing non-wcjE O-acetylation (10C and 15B), capsules without O-acetylation (6B and D39), or nonencapsulated (cap-) strains (R36A and AMB03). Strains R36A and AMB03 are derivatives of D39 (ST2) and JC03 (ST11A), respectively.

Figure 4.

Complement deposition on isogenic strains expressing ST11A, ST11E, ST9V, and ST9A polysaccharide was examined by flow cytometry and expressed as mean fluorescence intensity (MFI). All data are representative of at least 3 experiments performed in triplicate. A, Deposition of C3 (top panels) and C4 (bottom panels) over time on ST11A (JC03; column 1), ST11E (JC04; column 2), ST9V (AMB15; column 3), and ST9A (AMB16; column 4) bacteria in the presence of 5% normal human serum (NHS) only (open squares), 5% NHS plus 1 mM phosphocholine (PC; open circles), 5% NHS plus 1 mM PC and 10 mg/L bovine serum albumin (BSA; Xs), or 5% NHS plus 1 mM PC and 10 mg/L acetylated BSA (acBSA; closed triangles). B, Deposition of C3 (top panel) and C4 (bottom panel) on ST11A (JC03) and ST11E (JC04) bacteria after 15-minute incubation with 5%, 10%, or 15% NHS with no inhibitor (none), 1 mM PC, 1 mM PC plus 10 mg/L acBSA (PC + acBSA), or 1 mM PC plus 10 mg/L BSA (PC + BSA). C, Deposition of C3 (top panel) and C4 (bottom panel) on ST11A (JC03) and ST11E (JC04) bacteria after 15-minute incubation with 15% NHS with no inhibitor (none), 50 mM N-acetylglucosamine (GlcNAc), 50 mM glucose (Glc), silica clot activator (SCA), or 1 mg/L pneumococcal teichoic acid (TA). D, Deposition of C3 (left panel) and C4 (right panel) on ST11A (JC03) and ST11E (JC04) bacteria after 1-hour incubation in 5% C1q-depleted serum with no inhibitor (none), 50 mM GlcNAc, 50 mM Glc, SCA, 10 mg/L acBSA, 10 mg/L BSA, or 1 mg/L TA.

Figure 5.

Opsonophagocytic killing of ST11A (JC03) and ST11E (JC04). Bacteria were incubated in the presence of normal human serum (NHS) with no inhibitor (None), 1 mM phosphocholine (PC), 1 mM PC plus 10mg/L acetylated bovine serum albumin (PC + acBSA), 1 mM PC plus 10 mg/L bovine serum albumin (PC + BSA), silica clot activator (SCA), 50 µM cytochalasin-B in dimethyl sulfoxide (CytoB + DMSO), or dimethyl sulfoxide alone (DMSO). Percentage survival is calculated as the ratio of colony-forming units (CFUs) in the presence of NHS versus CFUs in the presence of heat-inactivated NHS.

Figure 6.

Meta-analysis of 18 studies for pneumococcal serotype invasiveness in children before the widespread use of pediatric pneumococcal vaccines. Odds ratios (ORs) were determined by comparing the ratio of the number of invasive to carriage isolates expressing a serotype to the corresponding ratio expressing serotype 14 in each study (Supplementary Materials). A, The pooled odds ratios (ORs) with 95% confidence intervals (CIs) for the 30 most prevalent serotypes or serogroups identified in the meta-analysis, listed from highest to lowest ORs. B–E, Study-specific ORs and 95% CIs for serotypes representative of high (ST1; B), medium (ST9V [C]; ST6B [D]), and low (ST11A; E) invasiveness. Data are displayed according to the numbering of respective studies (1–18) in the table in the Supplementary Materials. Open circles represent studies with no data for that serotype.