Improved Immune Responses Against Zika Virus After Sequential Dengue and Zika Virus Infection in Humans

Abstract

The high levels of dengue-virus (DENV) seroprevalence in areas where the Zika virus (ZIKV) is circulating and the cross-reactivity between these two viruses have raised concerns on the risk of increased ZIKV disease severity for patients with a history of previous DENV infections. To determine the role of DENV preimmunity in ZIKV infection, we analyzed the T- and B-cell responses against ZIKV in donors with or without previous DENV infection. Using peripheral blood mononuclear cells (PBMCs) from donors living in an endemic area in Colombia, we have identified, by interferon (IFN)-γ enzyme-linked immunospot (ELISPOT) assay, most of the immunodominant ZIKV T-cell epitopes in the nonstructural (NS) proteins NS1, NS3, and NS5. Analyses of the T- and B-cell responses in the same donors revealed a stronger T-cell response against peptides conserved between DENV and ZIKV, with a higher level of ZIKV-neutralizing antibodies in DENV-immune donors in comparison with DENV-naïve donors. Strikingly, the potential for antibody-mediated enhancement of ZIKV infection was reduced in donors with sequential DENV and ZIKV infection in comparison with donors with DENV infection only. Altogether, these data suggest that individuals with DENV immunity present improved immune responses against ZIKV.

Keywords: T-cell epitopes; antibody-dependent-enhancement (ADE); cross-reactive T cells; dengue virus; immunodominance; neutralizing antibodies; zika virus.

Figure 1

Cumulative IFN-γ responses (as spot-forming cells (SFCs) per million cells) for each overlapping peptide spanning the ZIKV proteome. (A) SFCs per million cells are shown for donors from group B; (B) SFCs per million cells are shown for donors from groups C, D, and E. The heat map indicates the number of donors with a positive IFN-γ response to each peptide within each protein (C, capsid; M, membrane; E, envelope, NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5). The numbers below each graph represent percentages of the total response for each protein.

Figure 2

DENV/ZIKV-immune donors reveal a broader T-cell response with a higher magnitude. (A) Breadth and (B) magnitude of responses in donors with a history of ZIKV infection only or DENV and ZIKV infection. Each dot represents one donor (open circles, ZIKV-immune DENV-naïve donors; filled circles, DENV/ZIKV-immune donors) and the bars represent the median value for each group of donors. The p values were calculated using the nonparametric two-tailed Mann–Whitney test. Frequency of responses against individual peptides, per donor, in ZIKV-immune DENV-naïve (C) and DENV/ZIKV-immune (D) donors. Each dot represents one peptide. The bars represent the median response for each donor.

Figure 3

Analysis of ZIKV EDIII (ZEDIII)-binding and ZIKV- and DENV-neutralizing activity of plasma samples from DENV- and/or ZIKV-immune donors. (A) Detection of ZEDIII-binding antibodies by ELISA; (B) quantification of neutralizing activity against ZIKV; (C) quantification of neutralizing activity against DENV1, 2, 3, and 4. Naïve: plasma samples from naïve donors; DENV: plasma samples from DENV-immune ZIKV-naïve donors; ZIKV: plasma samples from ZIKV-immune DENV-naïve donors; DENV/ZIKV: plasma samples from DENV-immune ZIKV-immune donors. Each dot represents one donor and the bars represent the median value for each group of donors. The p values were calculated using the nonparametric two-tailed Mann–Whitney test; n.s.: not significant. The dotted line represents the ELISA cut-off value.

Figure 4

Analysis of Antibody-Dependent Enhancement (ADE) activity of plasma samples on ZIKV and DENV4 infection. (A) Detection of ADE activity on ZIKV infection from plasma samples of DENV-immune ZIKV-naïve (higher-left panel), ZIKV-immune DENV-naïve (higher-right panel), and DENV/ZIKV-immune (lower-left panel) donors. Comparison of the peak enhancement titer of antibodies from DENV-immune ZIKV-naïve and DENV/ZIKV-immune donors (lower-right panel). Each dot represents one donor; (B) detection of ADE activity on DENV4 infection from plasma samples of DENV-immune ZIKV-naïve (higher-left panel), ZIKV-immune DENV-naïve (higher-right panel), and DENV/ZIKV-immune (lower-left panel) donors. Infectivity curves for each donor are shown. The p values were calculated using the nonparametric two-tailed Mann–Whitney test. The dotted line represents the ELISA cut-off value.

Figure 5

Specificity and cross-reactivity of antibody and T-cell responses from donors with DENV and/or ZIKV infection. (A) Heat map of the neutralizing activity of plasma samples from DENV-immune ZIKV-naïve, ZIKV-immune DENV-naïve, and DEN/ZIKV-immune donors on DENV1-4 and ZIKV infection. (B) Heat map of the ADE activity of plasma samples from DENV-immune ZIKV-naïve, ZIKV-immune DENV-naïve, and DENV/ZIKV-immune donors on DENV1-4 and ZIKV infection; (C) Heat map of the mean T-cell response against peptides from the whole ZIKV proteome (TOTAL) or the immunodominant proteins of PBMCs from ZIKV-immune DENV-naïve and DENV/ZIKV-immune donors; each column represents one donor.