Novel Approach for Insertion of Heterologous Sequences into Full-Length ZIKV Genome Results in Superior Level of Gene Expression and Insert Stability

Abstract

Zika virus (ZIKV) emerged in the Americas in 2015, presenting unique challenges to public health. Unlike other arboviruses of the Flaviviridae family, it is transmissible by sexual contact, which facilitates the spread of the virus into new geographic areas. Additionally, ZIKV can be transmitted from mother to fetus, causing microcephaly and other severe developmental abnormalities. Reliable and easy-to-work-with clones of ZIKV expressing heterologous genes will significantly facilitate studies aimed at understanding the virus pathogenesis and tissue tropism. Here, we developed and characterized two novel approaches for expression of heterologous genes of interest in the context of full-length ZIKV genome and compared them to two previously published strategies for ZIKV-mediated gene expression. We demonstrated that among the four tested viruses expressing nLuc gene, the virus constructed using a newly developed approach of partial capsid gene duplication (PCGD) attained the highest titer in Vero cells and resulted in the highest level of nLuc expression. Suitability of the PCGD approach for expression of different genes of interest was validated by replacing nLuc sequence with that of eGFP gene. The generated constructs were further characterized in cell culture. Potential applications of ZIKV clones stably expressing heterologous genes include development of detection assays, antivirals, therapeutics, live imaging systems, and vaccines.

Keywords: Zika virus; bioluminescence; flaviviruses; heterologous gene expression; reverse genetics.

Figure 1

Mapping the region containing regulatory elements of the C gene of ZIKV. (A) Schematic representation of ZIKV-NS3m infectious clone (on top of (A)), which was used to construct a panel of viruses with chimeric C gene sequence (on the bottom of (A)). White boxes represent wt sequence of ZIKV-NS3m. Gray boxes (C opt) represent sequences that were mutated by synonymous substitutions. (B) Growth kinetics of viruses with chimeric C gene sequence in Vero cells after plasmid DNA transfection. Mean viral titer ± standard deviations in the samples that were collected daily from duplicate flasks were determined by titration in Vero cells. Dotted line (the one right above the blue C6 line) represents limit of virus detection (0.7 log₁₀ pfu/mL). Differences between growth of ZIKV-NS3m and that of the other constructs were compared using two-way ANOVA (**** p < 0.0001; ns—not significant, p > 0.05).

Figure 2

Identification of a minimal region of C gene that is required for ZIKV growth in Vero cells by deletional analysis. (A) Predicted stem-loop structure of the 5′-end of ZIKV genome (strain Paraiba_01/2015). Colored arrows indicate codon positions at which 5′ terminus of wt C gene sequences were fused with target sequence for mir-124 (red box in panel B). Sequence highlighted in green indicates translation initiation codon AUG of ZIKV polyprotein. Sequence highlighted in magenta indicates 5′ genome cyclization sequence. +1 (A) shows the position of ORF-shifting insertion (+1 nt) of a single A residue. (B) Schematic representation of viral genomes featuring gradual reduction of the wt C gene sequence from the 3′-end. Red and yellow boxes indicate target for mir-124-3p and ubiquitin gene, respectively. Gray boxes (C opt) represent full-length sequences of C gene that were mutated by synonymous substitutions. (+1 A) and (-1nt) are positions of ORF shifting and ORF restoring sequence modifications, respectively. Dotted arrow represents the site of cleavage by ubiquitin. (C) Growth kinetics of viruses in Vero cells after plasmid DNA transfection. Differences between growth kinetics of C67-Ubi (mean virus titers for dpi 1–4) and those of the other five viruses depicted in panel B were compared using two-way ANOVA (**** p < 0.0001; ns—denotes not significant, p > 0.05). Dotted line represents limit of virus detection (0.7 log₁₀ pfu/mL). (D) Plaque morphology of viruses in Vero cell monolayer. Infected cells in 24-well plates were fixed at 5 dpi, and viral plaques were visualized by crystal violet staining.

Figure 3

Construction and characterization of nLuc-carrying viruses in Vero cells. (A) Genetic organization of nLuc-containing viruses. ORF shifting and restoring mutations are highlighted as +1nt and −1nt, respectively. 2A: 2A protease sequence from FMDV; Ubi: ubiquitin sequence; gray boxes highlight codon-optimized sequences in the C and E genes of ZIKV. C* in nLuc-25C is a C opt gene with mutations in 14–17 AA codons. Dotted arrows represent the sites of cleavage by 2A protease or ubiquitin. (B) Microscopic evaluation (at 40× magnification) of CPE in Vero cells monolayer observed on day 6 after plasmid DNA transfection (dpt). (C) Growth kinetics of nLuc-carrying viruses in Vero cells after plasmid DNA transfection. Mean virus titer ± standard deviations in the samples that were collected daily from duplicate flasks was determined by titration on Vero cells. Dotted line represents limit of detection of the FFA (0.7 Log₁₀(ffu/mL)). Differences between growth kinetics of nLuc-50C/FrSh and those of the other three constructs were compared using two-way ANOVA (**** p < 0.0001; *** p < 0.001). (D) Plaque morphology of nLuc-carrying viruses in Vero cell monolayer as revealed by immunostaining at 5 dpi. (E) Kinetics of luciferase activity following plasmid DNA transfection into Vero cells.

Figure 4

Evaluation of the insert stability of recombinant nLuc-carrying viruses. To compare genome stability of ZIKV-carrying nLuc, blind passaging was performed for all four viruses in 12.5 cm² flasks of Vero cells in duplicates, and RT-PCR reactions targeting the regions of nLuc insertion were carried out at multiple points during the experiment followed by sequencing. (A) Schematic representation of the positions of primers (represented by arrows) and expected lengths of corresponding RT-PCR fragments. (B,C) Agarose gel electrophoreses of RT-PCR fragments produced using viral RNA extracted from duplicate flasks of Vero cells (Rep#1 and Rep#2) after passage three (panel B) and ten (panel C). For each virus, two RT-PCR reactions were carried out (PCR-C and PCR-E, indicated in red), amplifying the region of nLuc gene insertion and selected region of ZIKV genome. Amplification of the viral regions that do not contain nLuc insertion serves as control of the expected band sizes after a complete deletion of the heterologous sequences in the nLuc-carrying viruses. Negative template control was included in each experiment and is shown in No RNA lane. Passaging of nLuc-25C was terminated after the third passage due to complete loss of nLuc insertion. (D) Kinetics of luciferase activity in Vero cells. Viruses recovered after the tenth passage in Vero cells were used for infection of Vero cells in 24-well plates at an MOI of 0.1. Measurements of nLuc activity in cell lysates were performed daily and data expressed as mean of standardized luciferase units ± standard deviations for the duplicate wells of infected Vero cells.

Figure 5

Characterization of reporter-carrying viruses in different cell lines. (A) Various cell lines of primate, human, and mosquito origin and human primary monocyte derived macrophages (MDM) were infected with nLuc-50C/FrSh in duplicates at an MOI of 0.01 in 24-well plates. Luciferase activity in cell lysates was measured daily for five days. (B–H) Growth kinetics of nLuc-50C/FrSh, GFPc-50C/FrSh, parental Paraiba_01/2015, and recombinant ZIKV-ICD viruses (colors representing each virus are defined in panel B) in Vero (B), C6/36 (C), LLC-MK2 (D), HepG2 (E), MDM (F), BeWo (G), JEG-3 (H). Cells were infected with each virus at a MOI of 0.01 in 12.5 cm² flasks in duplicates. Aliquots were taken every day for five days after infection, and FFA or PFA was performed to determine virus titers. Results are presented as mean values of two biological replicates with SD shown as error bars. Dotted line represents limit of detection of the assay (0.7 Log₁₀(pfu or ffu/mL)).