Structural and functional analysis of protective antibodies targeting the threefold plateau of enterovirus 71

Figures

• 

  Fig. 1. Breadth and specificity of EV71-specific antibody response in donor C. a Serological response against wild-type EV71 and its epitope variants. The virus (11-96023_VP3 E81K) was an escape variant selected with neutralizing antibody in vitro13 and the TW-1745_VP3 E81K and TW-1745_VP1 D110G were generated in vitro by introducing the residue substitution into an infectious cDNA clone of EV71 (strain TW-1745, GenBank accession number KT354870.1) via site-directed mutagenesis13. Donor C developed a strong antibody response at day 9 after illness onset which focused on the threefold plateau epitope of EV71 capsid. A post-EV71 infection serum (day 10 of illness onset) from another paediatric donor, a post-echovirus 11 infection serum (day 10 of illness onset) and a post-influenza virus infection serum (day 11 of illness onset) were included in the experiment. The neutralization assay was carried out twice with equivalent results. Source data are provided as a Source Data file. b Characterization of 19 plasmablast-derived EV71-specific monoclonal from donor C, measured by the flow cytometry (FCM), ELISA, immunoprecipitation (IP) and neutralization assays. Those marked by stars bound to…

• 

  Fig. 2. Characterization of threefold plateau-binding neutralizing antibodies. a Neutralizing activities of antibodies 38-1-10A, 38-1-2A and 38-3-11A against a panel of EV71 clinical strains in 1998–2016 and the prototype BrCr strain. The neutralization experiment was conducted three times independently (n = 3). Data are presented as mean ± standard error of the mean. Error bars represent the standard error of mean. The concentration of antibody that inhibited infection by 50% (EC50) was determined using nonlinear regression (log concentration vs. response, variable slope) in GraphPad Prism. A Coxsackievirus A16 was included in the neutralization assay as the control. b Immunogenetic analysis of the heavy and light chain variable regions of neutralizing antibodies using the IMGT tool. c Inhibition of EV71 12-96015 infectivity at pre- and post-attachment stages by neutralizing antibodies. Data are presented as the mean for measurements from two independent experiments (n = 2). The concentration of antibody that inhibited infection by 50% (EC50) was determined using nonlinear regression (log concentration vs. response, variable slope) in GraphPad Prism. Source data are provided as a Source…

• 

  Fig. 3. In vivo protection by threefold plateau-binding neutralizing antibodies. a The in vivo prophylactic effects of neutralizing antibodies 38-1-10A, 38-3-11A, 34-1-6D and 16-3-4D against EV71 12-73 (genotype C4) (upper) and 13-50144 (genotype B5) (below) infection (PBS control, n = 4; IgG control, n = 4; 38-1-10A, n = 4; 38-3-11A, n = 4; 34-1-6D, n = 4; 16-3-4D, n = 4). A single dose of antibody (10 mg/kg) or control was administered 24 h prior to intraperitoneal infection in the prophylactic experiment. The experiments were performed with PBS and anti-influenza human IgG antibody 4A-14 as controls. b The in vivo prophylactic effects of neutralizing antibodies 38-1-10A at doses of 10, 2 and 0.4 mg/kg against EV71 12-73 infection (PBS control, n = 6; 38-1-10A 10 mg/kg, n = 6; 38-1-10A 2 mg/kg, n = 5; 38-1-10A 0.4 mg/kg, n = 6). c The in vivo therapeutic effects of neutralizing antibody 38-1-10A administered at different timepoints against EV71 12-73 infection (PBS control 12 h, n = 5, 38-1-10A 12 h, n = 5; PBS control 24 h, n = 4, 38-1-10A 24 h, n = 5; PBS control 36 h, n = 5, 38-1-10A 36 h, n = 5; PBS control 48 h, n = 5, 38-1-10A 48 h, n = 6; PBS control 72 h, n = 5, 38-1-10A 72 h,…

• 

  Fig. 4. Overall structures of EV71/38-1-10A and EV71/38-3-11A complexes. a The structure of the EV71/38-1-10A complex viewed along an icosahedral twofold axis. The viral capsid proteins are shown as surface representations coloured in light blue, pale green and salmon for VP1, VP2 and VP3, respectively. For clarity, only the Vh and Vl domains of the Fab are shown as ribbons with Vh in magenta and Vl in cyan. One icosahedral asymmetric unit of the viron is outlined. b Close-up of 38-1-10A Fab binding around two threefolds. The colour scheme is as for a, but the virus is shown in grey. c Side view of b. Only one Fab is shown for clarity. d–f EV71/38-3-11A complex. The drawing, colour scheme and view direction of each panel are the same as that of a–c.

• 

  Fig. 5. Details of EV71 and Fab interactions. Interactions of EV71 with H3 (a), H1 and H2 (b), L1 (c) and L2 (d) loops in the EV71/38-1-10A complex. Interactions of EV71 with H3 (e), H1 (f) and L1 and L3 (g) loops in the EV71/38-3-11A complex. The main-chain backbones are drawn as thick sticks and side chains as thinner sticks with Vh in grey, Vl cyan, VP2 green and VP3 from a neighbouring protomer in salmon. Hydrogen bonds and salt bridges are shown as broken yellow sticks. h VP3 BC loop of EV71. The yellow broken sticks represent hydrogen bonds from the key stabilizing residue E81, which is also an escape mutation for both antibodies.

• 

  Fig. 6. 38-1-10A and 38-3-11A interact with the same epitope of EV71. The core region of the epitope that interacts with 38-1-10A (a) and with 38-3-11A (b). Capsid proteins of EV71 are shown as ribbons with VP1, VP2 and VP3 coloured in light blue, pale green and salmon, respectively. Cαs of viral residues that have direct contact with CDR H3 or L3 (≤4.0 Å) are drawn as bright spheres. The core region of one epitope is circled and black lines mark the pentamer boundaries. c, d Roadmaps of the virus surface with colour scheme as in a and an icosahedral subunit outlined in black lines. Residues in contact with 38-1-10A (c) or 38-3-11A (d) are highlighted in the corresponding viral protein colour and escape mutations are shown in bright colours.