Anti-hemagglutinin monomeric nanobody provides prophylactic immunity against H1 subtype influenza A viruses

Abstract

Influenza viruses constitute a major threat to human health globally. The viral surface glycoprotein hemagglutinin (HA) is the immunodominant antigen, contains the site for binding to the cellular receptor (RBS), and it is the major target of neutralizing antibody responses post-infection. We developed llama-derived single chain antibody fragments (VHHs) specific for type A influenza virus. Four VHHs were identified and further characterized. VHH D81 bound residues in the proximity of the C-terminal region of HA1 of H1 and H5 subtypes, and showed weak neutralizing activity, whereas VHH B33 bound residues in the proximity of the N-terminal region of the HA's stem domain (HA2) of H1, H5, and H9 subtypes, and showed no neutralizing activity. Of most relevance, VHHs E13 and G41 recognized highly conserved conformational epitopes on the H1 HA's globular domain (HA1) and showed high virus neutralizing activity (ranging between 0.94 to 0.01μM), when tested against several human H1N1 isolates. Additionally, E13 displayed abrogated virus replication of a panel of H1N1 strains spanning over 80 years of antigenic drift and isolated from human, avian, and swine origin. Interestingly, E13 conferred protection in vivo at a dose as low as 0.05 mg/kg. Mice treated with E13 intranasally resulted in undetectable virus challenge loads in the lungs at day 4 post-challenge. The transfer of sterilizing pan-H1 immunity, by a dose in the range of micrograms given intranasally, is of major significance for a monomeric VHH and supports the further development of E13 as an immunotherapeutic agent for the mitigation of influenza infections.

Fig 1. Development and characterization outline of H1N1 specific nanobodies.

A. Schematic outline of the development and characterization of H1N1 specific nanobodies followed in the present report. Repeated immunization of the llama triggered the clonal expansion and selection of B cells expressing specific antibodies. From the mononuclear cells isolated from peripheral blood, the heavy chain Ig specific mRNAs were specifically amplified and cloned in phagemid plasmids, generating a VHH library. Phage display was used for the screening of the candidate VHHs, and followed by in vitro, in silico and in vivo characterizations. B. Summary of the number of clones obtained after each step of the process.

Fig 2. VHH development.

A. Specific anti-H1N1 response in the llama serum. Six doses of monovalent vaccine containing A/California/07/2009 (H1N1) antigen are indicated with light grey arrows, and two doses of the trivalent vaccine (A/California/07/2009 (H1N1), A/Perth/16/2009 (H3N2), B/Brisbane/60/2008 are indicated with dark grey arrows. Black arrow indicates terminal bleeding. Evaluation of specific immune response to H1N1 in serum was assessed by HA ELISA; and H1N1 ELISA; hemagglutination inhibition (HI) assay; and micro neutralization (MN). Results are indicated as the inverse of the highest serum dilution with positive signal. B. VHH specific anti-HA0 reactivity. The type of epitope recognized on HA0 was evaluated by Western blot run either under non-reducing (left panel) vs. reducing conditions (with β-mercaptoethanol) (right panel). A nonrelated VHH against Norovirus was used as negative control (nr-VHH). Anti-VHH rabbit polyclonal serum was included as background control. As positive controls H1N1 rabbit and llama immune sera were included. Bands detected correspond to monomeric (70kD), dimeric (140kD) and trimeric (240kD) conformations of HA0 (70kD), HA1 (40kD), and HA2 (25kD).

Fig 3. Mapping linear epitopes recognized by B33 and D81.

A and E, ELISA with overlapping peptides corresponding to hemagglutinin HA of A/California/04/2009 (H1N1) pdm09. A concentration of 4 ng/well of peptide individually detected by 20 ng/well of VHH. Results are expressed as the absorbance measured at 450 nm. B and F, Epitope sequence showing the variability of the HA sequences analyzed by LOGOS for viruses belonging to group 1 ((H1N1) hu/Ca/09ma; hu/Arg/09; hu/Arg/09ma; hu/PR8/34; (H5N1) hu/Viet/04, and (H9N1) gf/HK/99, and group 2 (H3N2) hu/Perth/09 used in the ELISA. C and G, Ribbon representation, D and H. Surface representation with detail of the residues involved in the specific binding. Molecular modelling of the VHH and docking analysis against hu/Ca/09 HA (PDB structure 3LZG, at <5-Å resolution). The crystal structures used as templates for 3D modelling of B33 were 5F1K_C, 6GS1_H, 4WEU, and for D81 were 5VXK_B, 6UL4_B, 5JAB_B, 6GWN_C. Relevant amino acid residues for interaction are color coded: CDR1 green, CDR2 blue, and CDR3 red.

Fig 4. Deducing the structure of the conformational epitopes recognized by E13 and G41.

A, E, Peptide array analysis. B, F, VHH deduced epitope footprint on an HA monomer. Detail of the amino acid diversity in the domain of residues involved in the VHH-HA binding, analyzed by LOGOS. The H1N1 viruses used in neutralizing assays were: hu/Ca/09; hu/Ca/09ma; hu/Arg/09; hu/Arg/09ma; hu/PR8/34; hu/Bri/07; rt/Del/09; and sw/SD/18. Residues involved in the interaction with E13 and G41 are indicated in pink or green boxes, correspondingly. Relevant antigenic domains and RBS structures are labeled. C, G, Ribbon representation for HA interacting with the VHH. Side and top views. Residues on positions 152, 216, 306, 315 are labeled. D, H, Surface representation with detail of the residues involved in the specific binding. The shape of the putative domain of interaction is indicated by colored broken lines. Amino acid residues critical for the interaction are indicated in color code: CDR1 is colored in green, CDR2 in blue, and CDR3 in red. For the 3D modelling of E13 crystal structures of 6H7L_C, 5HVG_B, 6H7J_C were used as templates, and for G41 structures of 5XLV_C,5JQH_C, 4H3J_A. The model with the highest probability for E13 showed 106 members, while in the case of G41 the cluster showed 232 members.

Fig 5. Prophylactic efficacy against H1N1 virus challenge.

A. Experimental design. B, C, D, Screening of VHH candidates using hu/Arg/09 ma (2LD₅₀). E, F, G challenge with hu/Arg/ 09 ma (4LD₅₀); H, I, J, challenge with hu/Arg/09 (viral dose 50,000 TCID₅₀); K, L, M challenge with hu/PR8/34 ma (4LD₅₀). B, E, H, K, Survival curve. C, F, I, L, Body weight represented as the mean percentage of the initial body weight in each group. D, G, J, M, viral titer determined by TCID₅₀ on MDCK cell monolayers. Female mice BALB/c were treated by intranasal route with 5 mg/kg of specific VHH, non-related VHH (nr-VHH), hu/Arg/09 mouse immune serum (dil 1/6) (α- H1N1), or saline, 4 h before viral challenge with virus according to the label. Viral titer in lung homogenates was analyzed on day 4 after inoculation (n = 5). Body weight was monitored for 12 days after infection. Error bars represent standard deviations. Symbols are the same for all panels. One-way ANOVA; Dunnett’s Multiple Comparison Test vs Saline group; *: p<0,05; **: p<0,01; ***: p<0,001.

Fig 6. Effective dose of E13 and G41.

A, Kaplan-Meier survival curve for the analysis of effective dose. B, Body weight after infection is represented as the mean percentage of initial body weight for all mice in each group. C, Viral titer in lung tissue was analyzed on day 4 post viral challenge. Error bars represent standard deviations of the mean. BALB/c mice (n = 5) were given saline, or doses of 100, 25, 10, 5, 1 μg of E13, and 200 or 100 μg of G41, or mouse immune serum (α-H1N1); 4 h later mice were challenged with hu/PR8/34 virus (4LD₅₀). Body weight was monitored for 12 days. Symbols are the same for panels a, b, and c. One-way ANOVA; Dunnett’s Multiple Comparison Test vs saline group; *: p<0,05; **: p<0,01; ***: p<0,001.