Complement-mediated bactericidal activity of anti-factor H binding protein monoclonal antibodies against the meningococcus relies upon blocking factor H binding

Abstract

Binding of the complement-downregulating protein factor H (fH) to the surface of the meningococcus is important for survival of the organism in human serum. The meningococcal vaccine candidate factor H binding protein (fHbp) is an important ligand for human fH. While some fHbp-specific monoclonal antibodies (MAbs) block binding of fH to fHbp, the stoichiometry of blocking in the presence of high serum concentrations of fH and its effect on complement-mediated bactericidal activity are unknown. To investigate this question, we constructed chimeric antibodies in which the human IgG1 constant region was paired with three murine fHbp-specific binding domains designated JAR 3, JAR 5, and MAb502. By surface plasmon resonance, the association rates for binding of all three MAbs to immobilized fHbp were >50-fold higher than that for binding of fH to fHbp, and the MAb dissociation rates were >500-fold lower than that for fH. While all three MAbs elicited similar C1q-dependent C4b deposition on live bacteria (classical complement pathway), only those antibodies that inhibited binding of fH to fHbp (JAR 3 and JAR 5) had bactericidal activity with human complement. MAb502, which did not inhibit fH binding, had complement-mediated bactericidal activity only when tested with fH-depleted human complement. When an IgG1 anti-fHbp MAb binds to sparsely exposed fHbp on the bacterial surface, there appears to be insufficient complement activation for bacteriolysis unless fH binding also is inhibited. The ability of fHbp vaccines to elicit protective antibodies, therefore, is likely to be enhanced if the antibody repertoire is of high avidity and includes fH-blocking activity.

Fig. 1.

Binding of anti-fHbp MAbs. (A) ELISA. IgG bound to immobilized fHbp was detected with an anti-human kappa light-chain-specific alkaline phosphatase-conjugated antibody. Error bars represent the range in OD values observed in two independent experiments. (B and C) Flow cytometry. (B) Binding of anti-fHbp MAbs (4 μg/ml) with live bacterial cells of N. meningitidis group B strain H44/76. JAR 3, black dashed line; JAR 5, gray line; MAb502, black line. An irrelevant human MAb (100 μg/ml) served as a negative control (gray filled histogram). The binding curves of the three anti-fHbp MAbs are superimposed. (C) Same MAb concentrations as in panel B in the presence of heat-inactivated 20% IgG-depleted human serum as a source of human fH (∼90 μg/ml). (D and E) Surface plasmon resonance. (D) Representative data for binding of 0.25 μg/ml anti-fHbp MAbs (1.7 nM) to immobilized recombinant fHbp (1,000 RU). Lines are as in panel B. (E) Binding of purified human fH to immobilized recombinant fHbp (1,000 RU). fH concentrations of 12 to 90 μg/ml (71 to 580 nM) are shown: 12 μg/ml, thick black line; 23 μg/ml, dotted black line; 45 μg/ml, dashed black line; and 90 μg/ml, gray solid line. The flow cytometric data were replicated in two or three independent experiments.

Fig. 2.

Complement activation on encapsulated group B bacteria of strain H44/76. (A to D) Flow cytometry. (A) Activation of C4b deposition by 4 μg/ml of MAb and 15% C1q-depleted complement that also had been depleted of IgG. Anti-fHbp MAb JAR 3, black dotted line; JAR 5, gray line; MAb502, black line; irrelevant human MAb (100 μg/ml), gray filled histogram (data for each are superimposed). (B) Same as in panel A except for the addition of 30 μg/ml of purified C1q protein to the reaction mixtures. Data were replicated in two independent experiments. (C) Activation of C4b deposition by 4 μg/ml of MAbs and 15% C6-depleted complement that also had been depleted of IgG. (D) Activation of C3b. The conditions and complement source were the same as for panel C. (E) Bactericidal activity. Survival of bacteria after incubation for 60 min at 37°C with each of the MAbs and complement (20% IgG-depleted human serum) is shown. With complement alone, survival was >195%.

Fig. 3.

Inhibition of fH binding by anti-fHbp MAbs. (A) ELISA. Inhibition of binding of fH (2 μg/ml) to recombinant fHbp immobilized on a microtiter plate is shown. ▿, JAR 3; ×, JAR 5; ○, MAb502. Error bars represent the OD range for two independent experiments. (B and C) Flow cytometry. (B) Inhibition of binding of fH (∼90 μg/ml in 20% IgG-depleted human serum) to live bacterial cells by 50 μg/ml of MAb. Black solid line, MAb502; dotted black line, JAR 3; gray solid line, JAR 5; dark gray filled area, bacteria without fH or MAb as a negative control; light gray filled area, bacteria with fH without a MAb as a positive control. (C) Same conditions as for panel B except that 2 μg/ml of each of the anti-fHbp MAbs was tested instead of 50 μg/ml. The ELISA and flow cytometry data were replicated in at least two independent experiments.

Fig. 4.

Binding of fH to fHbp and NspA KO mutants of group B strain H44/76 by flow cytometry. (A) Binding of anti-PorA MAb (P1.7, 20 μg/ml, positive control). Black line, wild-type strain; gray line, fHbp KO mutant; dashed line, double fHbp and NspA KO mutant; filled gray area, no MAb. The binding curves of the three MAbs are superimposed. (B) Binding of purified human fH (100 μg/ml). Designations are the same as in panel A except that the filled gray area is for no added fH. (C) Binding of fH in human serum (20%, IgG depleted). Designations are as in panel B. The results were replicated in two independent assays.

Fig. 5.

Bactericidal activities of anti-fHbp MAbs measured against a mutant of group B H44/76 with genetic inactivation of NspA expression. (A) Binding of purified human fH (100 μg/ml) to live bacteria determined by flow cytometry. Gray line, NspA KO mutant; black line, wild-type control; dashed line, double fHbp and NspA KO control; light gray filled area, no fH. All three strains showed similar binding with a control anti-PorA P1.7 MAb (data not shown). (B to F) Survival of bacteria after incubation for 60 min at 37°C with each of the MAbs and 20% IgG-depleted human serum as a complement source. Open triangles, NspA KO mutant; closed triangles, control wild-type strain. (B) Chimeric anti-fHbp MAb JAR 3. (C) Chimeric anti-fHbp MAb JAR 5. (D) Chimeric anti-fHbp MAb502. (E) Control murine anti-PorA P1.7 MAb. (F) Control murine MAb, SEAM 12 reactive with group B capsule. Bactericidal activity results are from three independent experiments. Where indicated, survivals of the respective wild-type and NspA KO strains incubated at the MAb dilution were significantly different (*, P ≤ 0.02; **, P < 0.001).