A Novel Mechanism for Zika Virus Host-Cell Binding

Abstract

Zika virus (ZIKV) recently emerged in the Western Hemisphere with previously unrecognized or unreported clinical presentations. Here, we identify two putative binding mechanisms of ancestral and emergent ZIKV strains featuring the envelope (E) protein residue asparagine 154 (ASN154) and viral phosphatidylserine (PS). Synthetic peptides representing the region containing ASN154 from strains PRVABC59 (Puerto Rico 2015) and MR_766 (Uganda 1947) were exposed to neuronal cells and fibroblasts to model ZIKV E protein/cell interactions and bound MDCK or Vero cells and primary neurons significantly. Peptides significantly inhibited Vero cell infectivity by ZIKV strains MR_766 and PRVABC59, indicating that this region represents a putative binding mechanism of ancestral African ZIKV strains and emergent Western Hemisphere strains. Pretreatment of ZIKV strains MR_766 and PRVABC59 with the PS-binding protein annexin V significantly inhibited replication of PRVABC59 but not MR_766, suggesting that Western hemisphere strains may additionally be capable of utilizing PS-mediated entry to infect host cells. These data indicate that the region surrounding E protein ASN154 is capable of binding fibroblasts and primary neuronal cells and that PS-mediated entry may be a secondary mechanism for infectivity utilized by Western Hemisphere strains.

Keywords: ASN154; Encephalitis; Flavivirus; Microcephaly; N-acetylglucosamine; Neurotropism; Zika Virus; binding motif.

Figure 1

Envelope protein structure. (A) The predicted E protein structure indicates that the region containing ASN154 (circled) and its N-acetyl glucosamine (NAG) modification (inset) is a linear β strand. Intrinsic disorder probabilities were calculated for each amino acid position in the E protein sequence from strains (B) PRVABC59 and (C) MR_766. Probabilities above 0.5 (blue line) are considered indicative of sites representing disordered regions. The region containing ASN154 is indicated (blue box) for each strain.

Figure 2

Zika virus binding motif (ZVBM) binding and Zika virus (ZIKV) inhibition in Vero cells. Peptides PRV, PRV-N, and MR bound Vero cells significantly (* p < 0.05) above scrambled PRV^Scr and PRV-N^Scr controls (A). Peptides PRV and MR bound Madin–Darby canine kidney (MDCK) cells significantly (* p < 0.05) above PRV-N and scrambled PRV^Scr and PRV-N^Scr controls, and PRV bound with significantly († p < 0.05) higher avidity than MR (B). Two-fold dilutions of ZVBM peptides resulted in proportional reductions in signal, with significant (*, p < 0.05) differences between binding peptides and scrambled controls apparent with 0.05 μmol for Vero cells (C) and 0.0125 μmol for MDCK cells (D). The difference in avidity between PRV and MR became significant († p < 0.05) with 0.0125 μmol of treatment. Error bars indicate standard deviations in all panels.

Figure 3

Peptide binding to dorsal root ganglia (DRG) neurons ex vivo (20x magnification). Primary DRG neurons from C57 black mice (DAPI, blue fluorescence) were exposed to ZVBM peptides (FITC, green fluorescence) from (A) PRVABC59, (B) MR_766, and (C) scrambled (unglycosylated) PRVABC59. Punctate green staining around the DRG nuclei was observed in panels A and B, but was largely absent from panel C.

Figure 4

Refinement of ZVBM functional elements. The carboxyterminal peptide PRV-CTD (blue) bound Vero cells significantly (* p < 0.05) above the scrambled control peptide PRV^Scr (white), and the aminoterminal peptide PRV-NTD (red) did not. PRV-CTD bound Vero cells at equivalent levels to peptides PRV-N, and MR (black), indicating that this refined motif facilitates binding to Vero cells.

Figure 5

Disruption of ZIKV Infectivity. Pretreatment of Vero cells with peptides MR (red), PRV (blue), and PRV-N (green) significantly (* p < 0.05) inhibited cytopathogenic effect (CPE) generation following infection with both strains MR_766 and PRVABC59 relative to pretreatment with scrambled controls PRV^Scr (black), PRV-N^Scr (grey), or PBS alone (white) (A). Pretreatment of ZIKV strain PRVABC59 with the PS-binding protein annexin V (black) prior to Vero cell infection resulted in a significant (* p < 0.05) decrease in CPE generation relative to PBS alone (white). Pretreatment of ZIKV strain MR_766 with annexin V did not impact CPE generation relative to pretreatment with PBS alone (B).