Report

Specificity, cross-reactivity, and function of antibodies elicited by Zika virus infection

See allHide authors and affiliations

Science  19 Aug 2016:
Vol. 353, Issue 6301, pp. 823-826
DOI: 10.1126/science.aaf8505
  • Fig. 1 Specificity and cross-reactivity of NS1-reactive mAbs and T cells derived from ZIKV- and DENV-infected donors.

    (A) Sequence conservation of NS1 proteins as determined by the alignments shown in fig. S2. Conserved residues in NS1 are in purple. (B) Structure of the ZIKV NS1 dimer [Protein Data Bank (PDB) 5IY3]. The two monomers are shown with different shades of gray. The surface residues of each protomer that are conserved between ZIKV and DENV are colored as in (A). (C and D) Heat map of the reactivity of (C) 41 mAbs derived from four ZIKV-infected donors (ZA-ZD) and (D) 12 mAbs derived from two DENV-infected donors. The mAbs were tested for binding to NS1 proteins of ZIKV and DENV1-4 [median effective concentration (EC50), nanograms per milliliter]. A description of the immune status of the donors and the mAbs codes is provided in table S1. The binding curves of NS1-reactive mAbs are shown in figs. S3 and S5, and EC50 values are reported in tables S3 and S4. Data are representative of at least two independent experiments. (E) Plasma from ZIKV-infected (n = 4 donors), DENV-infected (n = 5 donors), and control donors (n = 48 donors) (1/10 dilution) were tested for their capacity to bind NS1 (blue dots) and to inhibit the binding of the biotinylated mAb ZKA35 to NS1 (red dots). (F and G) T cell libraries were generated from CXCR3+ and CXCR3 memory CD4+ T cells of the four ZIKV-infected donors and screened for reactivity against ZIKV or DENV1-4 NS1 proteins by use of 3H-thymidine incorporation. Shown is the estimated number of ZIKV NS1–specific T cells per 106 cells [mean + SEM and color-coded values for individual donors (F)]. Shown is also the proliferation [counts per minute (cpm)] of individual cultures to either ZIKV or DENV1-4 NS1 proteins (G). The cultures from different donors are color-coded, and their numbers are shown in parentheses. Dotted lines represent the cut-off value.

  • Fig. 2 Specificity and cross-reactivity of E-reactive mAbs and T cells derived from ZIKV- and DENV-infected donors.

    (A) Sequence conservation of E proteins as determined by the alignments of ZIKV isolates and DENV1-4 reference strains, as shown in fig. S8. Conserved residues in EDI, EDII, EDIII, and the hinge regions of the E protein are colored in red, yellow, blue, and green, respectively. FL, fusion loop. (B) Structure of the ZIKV E protein dimer (PDB 5JHM). The two monomers are shown with different shades of gray, and the residues conserved between ZIKV and DENV are colored as in (A). (C and D) Heat map of the reactivity of (C) 78 mAbs derived from four ZIKV donors and (D) 61 mAbs derived from nine DENV donors. NNB, neutralizing non-E-binding mAbs. The mAbs were tested for binding to E-recombinant proteins of ZIKV and DENV1-4 (EC50, nanograms per milliliter). Also shown is the virus-neutralizing activity (IC50, nanograms per milliliter). Strikethrough cells indicate not tested. Binding curves of E-reactive mAbs are shown in figs. S9 and S10. The EC50 and IC50 values of E-reactive and NNB mAbs are shown in tables S5 and S6. Data are representative of at least two independent experiments. (E and F) T cell libraries were generated from CXCR3+ and CXCR3 memory CD4+ T cells of the four ZIKV-infected donors and screened for reactivity against ZIKV or DENV1-4 E proteins by using 3H-thymidine incorporation. Shown is the estimated number of ZIKV E–specific T cells per 106 cells [mean + SEM and color-coded values for individual donors (E)]. Also shown is the proliferation (cpm) of individual cultures to either ZIKV E or DENV1-4 E proteins (F). The cultures from different donors are color-coded, and their numbers are shown in parentheses. Dotted lines represent the cut-off value. (G) CD4+ CXCR3+ T cells from peripheral blood mononuclear cells of a ZIKV-infected (ZB) or DENV-infected (DENV013) donor were labeled with carboxyfluorescein succinimidyl ester (CFSE) and stimulated with ZIKV or DENV1-4 E proteins, respectively. Proliferating T cells were isolated by means of cell sorting, relabeled with CFSE, and stimulated with the indicated proteins. Shown are the CFSE profiles on day 5.

  • Fig. 3 Neutralization and enhancement of ZIKV and DENV infection by mAbs and immune plasma.

    (A) Binding of mAbs isolated from ZIKV-infected donors (blue) or from DENV-infected donors (red) to (left and middle) recombinant ZIKV E and (right) to DENV1 E protein. Open and solid symbols indicate mAbs specific or cross-reactive for ZIKV and DENV, respectively. (B) Neutralization of ZIKV and DENV1 as determined by the percentage of infected Vero cells. (C) ADE of ZIKV and DENV1 infection of nonpermissive K562 cells, as determined by the percentage of infected cells on day 5. (D and E) Inhibition of ADE by LALA mutant mAbs. Serial dilutions of ZIKV- and DENV3-immune plasma were incubated with ZIKV or DENV1 before addition to K562 cells in the absence or presence of a fixed amount (50 μg/ml) of ZKA64-LALA or DV82-LALA mAb. Shown is the percentage of infected K562 cells on day 4. Blue and red symbols indicate plasma derived from ZIKV- or DENV-infected donors, respectively. Data are representative of two independent experiments.

  • Fig. 4 In vivo enhancement of DENV2 infection by an anti-ZIKV cross-reactive mAb and ZIKV therapeutic efficacy of a potent anti-ZIKV EDIII-specific mAb.

    (A) In vivo ADE. MAbs were administered intraperitoneally to AG129 mice 20 to 24 hours before intravenous inoculation with 2.5 × 105 plaque-forming units (PFU) of DENV2 D2S10. Results are representative of two independent experiments, with n = 8 mice for the ZKA78 1 and 2-μg groups; n = 6 for the DV82 and all LALA mAb groups; and n = 4 for phosphate-buffered saline (PBS) group. (Left) A Kaplan-Meier survival curve is shown; significance was determined by using the Mantel-Cox log-rank test. ZKA78 versus PBS, P = 0.0008; ZKA78 versus ZKA78-LALA, P = 0.0081; DV82 versus PBS, P = 0.0187; and DV82 versus DV82-LALA, P = 0.0255. (Right) Mice were monitored over a 10-day period and scored for morbidity and mortality by using a standardized 5-point system (38). (B) Prophylaxis and therapy of ZIKV infection. MAbs (15 mg/kg) were administered intraperitoneally to A129 mice (n = 6 mice) 24 hours before or after subcutaneous inoculation with 102 PFU of ZIKV MP1751. (Top left) A Kaplan-Meier survival curve is shown. (Top right) Mice were monitored over a 13-day period for body weight loss. (Bottom) Viral loads were measured on day 5 in blood of all animals and in blood and indicated tissues when animals were culled at the end of the study or when the humane end points were met. Lines indicate geometric means.

Supplementary Materials

  • Specificity, cross-reactivity and function of antibodies elicited by Zika virus infection

    Karin Stettler, Martina Beltramello, Diego A. Espinosa, Victoria Graham, Antonino Cassotta, Siro Bianchi, Fabrizia Vanzetta, Andrea Minola, Stefano Jaconi, Federico Mele, Mathilde Foglierini, Mattia Pedotti, Luca Simonelli, Stuart Dowall, Barry Atkinson, Elena Percivalle, Cameron P. Simmons, Luca Varani, Johannes Blum, Fausto Baldanti, Elisabetta Cameroni, Roger Hewson, Eva Harris, Antonio Lanzavecchia, Federica Sallusto, Davide Corti

    Materials/Methods, Supplementary Text, Tables, Figures, and/or References

    Download Supplement
    • Figs. S1 to S13
    • Tables S1 to S6
    • Full Reference List

Navigate This Article