Humoral cross-reactivity between Zika and dengue viruses: implications for protection and pathology

Abstract

Zika virus (ZIKV) is a re-emerging mosquito-borne flavivirus that has recently caused extensive outbreaks in Central and South America and the Caribbean. Given its association with Guillain-Barré syndrome in adults and neurological and ocular malformities in neonates, ZIKV has become a pathogen of significant public health concern worldwide. ZIKV shares a considerable degree of genetic identity and structural homology with other flaviviruses, including dengue virus (DENV). In particular, the surface glycoprotein envelope (E), which is involved in viral fusion and entry and is therefore a chief target for neutralizing antibody responses, contains regions that are highly conserved between the two viruses. This results in immunological cross-reactivity, which in the context of prior DENV exposure, may have significant implications for the generation of immune responses to ZIKV and affect disease outcomes. Here we address the issue of humoral cross-reactivity between DENV and ZIKV, reviewing the evidence for and discussing the potential impact of this cross-recognition on the functional quality of antibody responses against ZIKV. These considerations are both timely and relevant to future vaccine design efforts, in view of the existing overlap in the distribution of ZIKV and DENV and the likely spread of ZIKV to additional DENV-naive and experienced populations.

Figure 1

The ZIKV particle and E protein dimer. Cryo-EM surface structures of (A) immature (PDB 5U4W) and (B) mature (PDB 5IRE) ZIKV. The E protein dimer is highlighted in a yellow box. (C) The ZIKV E protein dimer colored by its domain, EDI: red, EDII: yellow and EDIII: blue. The fusion loop is circled in orange. All structural figures in (A–C) were created using PyMol (Schrödinger LLC).

Figure 2

ZIKV and DENV E proteins share considerable sequence identity. (A) Phylogenetic tree, showing relatedness based on E protein sequence, created using MEGA7. The evolutionary history between the viruses was inferred by using the maximum likelihood method based on the JTT matrix-based model. The percentage of trees in which the associated viral sequences clustered together is shown next to the branches. Branches are drawn to scale, with lengths measured in the number of substitutions per site. (B) Heat map showing E protein sequence identity, generated with ggplot2 in R. Sequences were aligned in Geneious version 6.1. For (A) and (B), the ZIKV strains analyzed and their GenBank accession numbers are: PRVABC59: KU501215, MR766: AY632535, H/PF/2013: KJ776791, Yap/2007: EU545988 and SPH2015: KU321639. The DENV strains are DENV1 WestPac: U88535, DENV2 Tonga/72: AY744147.1, DENV3 Sleman/78: AY648961.1 and DENV4 Dominica/814669/1981: AF326573.1. In addition, the YFV strain Asibi: KF769016.1 was also included as an outgroup in the sequence analyses above.

Figure 3

Cross-reactive and ZIKV-specific mAbs binding their E protein epitopes. (A) Murine mAb 2A10G6 binding to the conserved fusion loop on a ZIKV E monomer. (B) Human mAb Z3L1 binding to a ZIKV-specific EDI epitope. (C) Human mAb C10 (ribbon structure) binding to a dimer-dependent epitope on E protein dimer. For (A–C), heavy and light chains of mAbs are colored red and pink, respectively. All residues conserved between DENV2 and ZIKV are colored orange. All figures were created using PyMol (Schrödinger LLC).