DENV virion binding and neutralization by purified chimeric mouse-human mAb variants (DV1-E50; WNV-E60) or chimpanzee-human mAb variants (1A5 WT, 1A5 ΔA, 1A5 ΔB). MAb DV1-E50 or WNV-E60 variants incorporated a full-length Fc region corresponding to each human IgG subclass whereas mAb 1A5 variants contained focused amino acid substitutions in the 1A5 WT IgG1 Fc portion that conferred IgG2 (1A5 ΔB) or IgG4 (1A5 ΔA) ligand properties for FcγR binding; a 9-amino acid deletion (1A5 ΔD) abrogated FcγR binding entirely. DENV neutralization dose-response and virion binding curves were generated by parallel plaque reduction neutralization assay in Fc receptor-negative Vero cells and virion-capture ELISA assay, respectively. (A) MAb DV1-E50 IgG subclass variants titrated against DENV1. (B, C) MAb WNV-E60 IgG variants titrated against DENV1 and DENV2, respectively. (D) MAb 1A5 IgG variants titrated against DENV2. Plaques were detected by indirect immunostaining with a DENV NS1-specific monoclonal antibody (9A9) and neutralization dose-response curves were generated using GraphpadPrism5 software. Mean values and S.E.M. of triplicate ELISA determinations and duplicate or triplicate neutralization determinations are plotted. ELISA and neutralization results are representative of at least two similar studies.

DENV (200 PFU) was incubated with graded concentrations of purified mAbs and the mixtures were used to generate neutralization dose-response curves in control and FcγRIA/γ or FcγRIIA stably transfected CV-1 cells. (A) mAb DV1-E50 variants titrated against DENV1. (B) mAb WNV-E60 variants titrated against DENV1. (C) mAb 1A5 WT (IgG1), 1A5 ΔB (IgG2), 1A5 ΔA (IgG4) titrated against DENV2. (D) mAb 1A5 ΔD titrated against DENV2. (E) FcγRIA- or FcγRIIA-expressing CV-1 cells were treated with a mixture of DENV1 immune complexes formed with DENV1 (200 PFU) and mAb DV1-E50 IgG2 (1 μg/ml), and graded concentrations (0.01 to 10 μg/ml) of purified human IgG1, IgG2, IgG3 myeloma proteins or control human serum albumin (HSA). FcγRIA but not FcγRIIA-modulated neutralization was blocked by the high-affinity IgG1 and IgG3 myeloma proteins only. Plaques were detected by indirect immunostaining with a DENV NS1-specific monoclonal antibody (9A9). Mean values and S.E.M. of triplicate infections are plotted. Results are representative of at least three (A-D) or two (E) similar experiments.

(A) FcγR expression. PE-labeled monoclonal antibodies against FcγRIA (CD64) or FcγRIIA (CD32) were used with corresponding labeled isotype controls to verify FcγR expression by flow cytometry. Human macrophage-like THP-1 cells which express FcγRIA and FcγRIIA (Fleit and Kobasiuk, 1991), served as a staining control. Results are representative of at least 6 independent determinations performed during an approximately 3-month period of continuous culture of the FcγR transfectants. (B) Competition-inhibition by aggregated IgG myeloma proteins. Pre-treatment of FcγRIA and FcγRIIA transfectants with heat-aggregated human IgG1 and IgG3 myeloma proteins inhibited (≥ 90%) rosette formation by rabbit IgG-sensitized sheep red blood cells (SRBC). IgG2 myeloma protein preferentially bound to FcγRIIA-H (~60% rosette inhibition). Mean values of triplicate samples and S.E.M. are plotted. Data are representative of four similar experiments. (C) Efficiency of DENV plaque formation in CV-1 transfectants. CV-1 transfectants were infected with serial two-fold (25 to 200) PFU of DENV1 or DENV2. Plaques were detected by indirect immunostaining with a broad DENV serogroup-reactive NS1-specific monoclonal antibody (9A9). Mean values and S.E.M. are indicated for each condition. Differences in plaque numbers over the range of input DENV PFU were not significant (P>0.92, one-way ANOVA). Graphs were generated from an experiment performed in triplicate for both DENV serotypes and are representative of three similar experiments.