An anti-H5N1 influenza virus FcDART antibody is a highly efficacious therapeutic agent and prophylactic against H5N1 influenza virus infection

Abstract

Highly pathogenic H5N1 avian influenza viruses are associated with severe disease in humans and continue to be a pandemic threat. While vaccines are available, other approaches are required for patients that typically respond poorly to vaccination, such as the elderly and the immunocompromised. To produce a therapeutic agent that is highly efficacious at low doses and is broadly specific against antigenically drifted H5N1 influenza viruses, we developed two neutralizing monoclonal antibodies and combined them into a single bispecific Fc fusion protein (the Fc dual-affinity retargeting [FcDART] molecule). In mice, a single therapeutic or prophylactic dose of either monoclonal antibody at 2.5 mg/kg of body weight provided 100% protection against challenge with A/Vietnam/1203/04 (H5N1) or the antigenically drifted strain A/Whooper swan/Mongolia/244/05 (H5N1). In ferrets, a single 1-mg/kg prophylactic dose provided 100% protection against A/Vietnam/1203/04 challenge. FcDART was also effective, as a single 2.5-mg/kg therapeutic or prophylactic dose in mice provided 100% protection against A/Vietnam/1203/04 challenge. Antibodies bound to conformational epitopes in antigenic sites on the globular head of the hemagglutinin protein, on the basis of analysis of mutants with antibody escape mutations. While it was possible to generate escape mutants in vitro, they were neutralized by the antibodies in vivo, as mice infected with escape mutants were 100% protected after only a single therapeutic dose of the antibody used to generate the escape mutant in vitro. In summary, we have combined the antigen specificities of two highly efficacious anti-H5N1 influenza virus antibodies into a bispecific FcDART molecule, which represents a strategy to produce broadly neutralizing antibodies that are effective against antigenically diverse influenza viruses.

Importance: Highly pathogenic H5N1 avian influenza viruses are associated with severe disease in humans and are a pandemic threat. A vaccine is available, but other approaches are required for patients that typically respond poorly to vaccination, such as the elderly and the immunocompromised. The variability of the virus means that such an approach must be broad spectrum. To achieve this, we developed two antibodies that neutralize H5N1 influenza viruses. In mice, these antibodies provided complete protection against a spectrum of H5N1 influenza viruses at a single low dose. We then combined the two antibodies into a single molecule, FcDART, which combined the broad-spectrum activity and protective efficacy of both antibodies. This treatment provides a novel and effective therapeutic agent or prophylactic with activity against highly pathogenic H5N1 avian influenza viruses.

FIG 1

Structure and characterization of FcDART. (A) The V_H and V_L segments of monoclonal antibodies h10C3 and BF1-19 (yellow and blue, respectively) were combined into a single bispecific Fc fusion protein, FcDART. Shown in detail are the heterodimerization domains of FcDART, which were appended to the C terminus of each chain to promote heterodimeric assembly. These domains consisted of coiled coil-forming sequences where the hydrophobic core of the coiled coil interface is flanked by opposite charges (an E negative charge on one coil [E-coil] and a K positive charge on the other [K-coil]), which is believed to favor heterodimerization by repelling a homotypic association and preventing the initiation of zippering of the hydrophobic core. An Fc domain was also added to the C terminus of the coil with an E negative charge, causing two DART units to dimerize and form a tetravalent bispecific structure. (B) Characterization of FcDART by denaturing, reducing SDS-PAGE and size exclusion chromatography analysis. LD, load fraction on the protein A column; NB, nonbound fraction on the protein A column; NaCl, high-salt wash; Elu, material that eluted at low pH; N. Pool, material that eluted at low pH and that was neutralized. The size exclusion chromatography profiles of purified material at two concentrations (red, 2 mg/ml; blue, 24.5 mg/ml) show peak elution at about 12 ml. The slight difference in the peak elution volume is within the range of variability of manual loading and does not indicate a significant elution volume difference between the two samples.

FIG 2

Human monoclonal antibodies bind to conformational epitopes on hemagglutinin. Antibodies were used to probe a blot of lysed Madin-Darby canine kidney cells that were infected with A/Vietnam/03/2004 (H5N1). BF1-1, BF1-10, and BF1-19, which showed reactivity by the HI assay, showed no binding, indicating that they did not recognize linear epitopes of hemagglutinin. BF1-32, which did not show reactivity by the HI assay, recognized linear epitopes on what appeared to be HA₁, as the most prominent bands corresponded to HA₀ (82 kDa) and HA₁ (58 kDa), with a less intense band being found at 78 kDa.

FIG 3

Antibody binding to a hemagglutinin panel from diverse H5N1 influenza virus clades. Binding was assessed in an antigen down format, where HA was immobilized on a plate (A, C), or in an MAb down format, where the MAb was immobilized on a plate (B, D, E). Binding of the FcDART molecule (F) was measured in an antibody down format, with FcDART being captured with anti-human IgG Fc-specific polyclonal antibody. OD, optical density.

FIG 4

Monoclonal antibody escape mutant viruses have amino acid changes located primarily in the globular head of hemagglutinin. (A) Amino acid changes shown on a cartoon of the hemagglutinin trimer of A/Vietnam/1203/04 (29). One subunit is highlighted for clarity. (B). Locations of amino acid changes. All mutations map to known antigenic sites (27). H5 numbering is used.

FIG 5

Antibodies provide 100% protection against challenge when administered prophylactically. Mice received a single intraperitoneal injection of antibody 24 h prior to challenge with 100 MLD₅₀s of A/Vietnam/1203/04 (H5N1) administered intranasally, which caused 100% mortality in untreated mice within 10 days of infection (B, D, and F). BF1-19 provided 100%, 80%, and 0% protection against challenge when it was delivered at dosages of 2.5, 1, and 0.1 mg/kg, respectively (A and B). h10C3 provided 80% protection against challenge when it was delivered as a single 2.5-mg/kg dose (C and D), while FcDART provided 100% protection against challenge when it was delivered as a single 2.5-mg/kg dose 24 h before challenge (E and F). (A, C, and E) Data represent the mean percentage of the starting weight ± standard error of the mean for each mouse group (n = 5 mice per group). **, P < 0.01.

FIG 6

Antibodies provided 100% protection against challenge when administered at 2.5 mg/kg at 24 h postchallenge. Mice received a single intraperitoneal dose of antibody at 2.5 mg/kg 24 or 72 h after challenge with 100 MLD₅₀s of A/Vietnam/1203/04 (H5N1) administered intranasally, which caused 100% mortality in untreated mice within 10 days of infection (B, D, and F). BF1-19 (A and B), h10C3 (C and D), and FcDART (E and F) provided 100% protection against challenge when administered at 24 h postchallenge. BF1-19 and h10C3 provided 80% protection against challenge when administered at 72 h postchallenge (B and D, respectively); however, FcDART provided 100% protection at 72 h postchallenge (F). (A, C, and E) Data represent the mean percentage of the starting weight ± standard error of the mean for each mouse group (n = 5 mice per group). hpi, hours postinfection; **, P < 0.01.

FIG 7

BF1-19 provides 100% protection against challenge by A/Vietnam/1203/04 (H5N1) in ferrets. BF1-19 was administered to ferrets as a single 1-mg/kg dose intraperitoneally at 24 h postchallenge with 3 × 10⁴ EID₅₀s of A/Vietnam/1203/04 (H5N1), which caused 100% mortality in ferrets injected with PBS but no mortality in ferrets that were injected with BF1-19 (A). Ferrets that were injected with BF1-19 also exhibited less weight loss and lower temperatures than ferrets injected with PBS (B and C, respectively); however, these differences were not significant. Three ferrets were administered BF1-19, and two ferrets were administered PBS. *, P < 0.05.

FIG 8

BF1-19 provides 100% protection against challenge from the antigenically drifted strain A/Whooper swan/Mongolia/244/05 (H5N1). A single 2.5-mg/ml intraperitoneal dose of BF1-19 administered at 24 h pre- or postchallenge provided 100% protection against challenge with 100 MLD₅₀s of A/Whooper swan/Mongolia/244/05 (H5N1). When this dose of antibody was administered at 72 h postchallenge, 80% protection was observed (A and B). (A) Data represent the mean percentage of the starting weight ± standard error of the mean for each mouse group (n = 5 mice per group). *, P < 0.05; **, P < 0.01.

FIG 9

BF1-19 provides 100% protection against challenge from A/Vietnam/1203/04 (H5N1) in vitro escape mutants. BF1-19 escape mutants were generated using a virus based on A/Puerto Rico/8/34, generated by reverse genetics, containing the hemagglutinin and neuraminidase genes from A/Vietnam/1203/04 (H5N1). Mice were administered 10⁶ TCID₅₀s of the S145F or K144N escape mutant (in hemagglutinin) intranasally, followed by a single 2.5-mg/ml intraperitoneal dose of BF1-19 administered at 24 h postchallenge. All mice that received BF1-19 survived the challenge, while 80% of mice that did not receive antibody died (A and B). (A) Data represent the mean percentage of the starting weight ± standard error of the mean for each mouse group (n = 5 mice per group). **, P < 0.01.