Construction of a Full-length Infectious Clone of Zika Virus Stably Expressing EGFP Marker in a Eukaryotic Expression System

Background: Zika virus is among the most widely transmitted arboviruses in the world and closely associated with diseases, such as encephalitis, fetal microcephaly, and Guillain–Barré syndrome. The pathogenic mechanism of the virus has not been fully elucidated, and there are no vaccines or specific drugs targeting the virus. To address these issues, the application of reverse genetics is needed for viral reconstruction and reproduction. Methods: Polymerase chain reaction (PCR) was used to merge the full-length Zika virus genome, CMV promoter, intron, EGFP, hepatitis delta virus ribozyme, and SV40 terminator sequence for cloning into a pBAC11 vector through recombination to produce recombinant pBAC-ZIKA-EGFP. The ZIKA–EGFP was rescued by transfection of 293T cells with pBAC-ZIKA-EGFP, and at 7-days post-transfection, the supernatant (P0 generation) was passed through a 0.45-μm membrane and used to infect Vero cells (to produce the P1 generation). Fluorescence-based quantitative PCR, 50% tissue culture infectious dose, and plaque assays were used to measure differences in replication ability and pathogenicity relative to the rescue virus (ZIKA–WT), the sequence of which is consistent with that of the wild-type Zika virus. Additionally, caffeic acid phenethyl ester (CAPE), a nuclear factor kappaB (NF-kB) inhibitor, was used to examine its effect on viral replication. Results: The results showed that ZIKA–EGFP could effectively infect Vero cells, SH-SY5Y cells and C6/36 cells, and cause cytopathic effects on them. ZIKA–EGFP exhibited stable replication and EGFP expression during cell passage for at least six generations, with no significant difference in replication ability relative to the ZIKA–WT. Fluorescent cell foci were observed in the plaque assay while the ZIKA–EGFP was in the absence of phage plaque formation. The inhibition of NF-kB inhibitor on ZIKA-EGFP was observed by fluorescence microscopy, which was consistent with the results of fluorescence


Background
Zika virus is among the most widely transmitted arboviruses in the world. The virus was first discovered in the Zika Forest in Uganda in 1947, and an outbreak of Zika virus in the Americas in 2015 caused a global public health emergency [1][2][3][4]. Approximately 80% of Zika infections are asymptomatic, with the most common symptoms including fever, arthralgia, rash, myalgia, edema, vomiting, and non-purulent conjunctivitis [5]. The virus can infect the placenta and blood-brain barrier endothelial cells, neurons, and neural stem cells and is closely associated with diseases, such as encephalitis, fetal microcephaly, and Guillain-Barré syndrome [6][7][8][9][10][11]. Similar to other flaviviruses, the Zika genome encodes an open reading frame containing three structural proteins (C: capsid; prM: precursor of membrane; and E: envelope) that comprise the viral particles and seven genes encoding non-structural (NS) proteins (NS1, NS2A, NS2B, NS3, NS4A, NS4B, and NS5) that participate in RNA replication [12]. Recent studies show that some mutations of the Zika virus are related to host adaptation [13][14][15][16]; however, the pathogenic mechanism of the virus has not been fully elucidated, and there are no vaccines or specific drugs targeting the virus. To address these issues, the application of reverse genetics is needed for viral reconstruction and reproduction [13,17].
This study describes a method for constructing a full-length infectious clone of the Zika virus in a eukaryotic expression system that can be used to produce the virus along with stable expression of the enhanced green fluorescence protein (EGFP) marker. We found that the replication ability of ZIKA-EGFP did not significantly differ from that of ZIKA-WT. These results provide a foundation for future studies on Zika virus mutations, pathogenic mechanisms, and drug screening.

Construction of a Full-length Infectious Clone of Zika Virus
As shown in Fig. 1, the CMV promoter sequence synthesized by Guangzhou IGE Biotechnology Co., Ltd. (Guangzhou, China) was fused to the Zika viral sequence of (nt 1-5908) using polymerase chain reaction (PCR) in order to generate the 5' half sequence of the Zika virus infectious clone (ZF). The hepatitis delta virus ribozyme and SV40polyA sequences synthesized by Guangzhou IGE Biotechnology Co., Ltd. were then ligated to the Zika viral sequence (nt 5735-10,807) using PCR to generate the 3' half sequence of the clone (ZR). Recombination was then performed using a tan Infusion kit (Takara, Dalian, China). Two fragments were ligated into the pBAC11 vector (Addgene, Watertown, MA, USA) previously digested with restriction enzymes (SfoI and PacI; New England Biolabs, Ipswich, MA, USA) to generate recombinant pBAC-ZIKA (Fig. 1b).
The intron sequence (derived from the pCI-neo vector plasmid) synthesized by Guangzhou IGE Biotechnology Co., Ltd. was inserted into the ZF sequence of the Zika virus (between nt 3128 and 3129) to generate the sequence of the ZFi fragment (the G-to-A mutation introduced at nt 3128 was used as a marker for subsequent experiments). The ZFi and ZR fragments were recombined using the described restriction-digested vector pBAC11 to generate recombinant pBAC-ZIKAi (Fig. 1c).
The EGFP gene was inserted downstream of the 37 th codon of the zika virus using the described methods, and the full-length coding-region sequence of the Zika virus was inserted to construct pBAC-ZIKA-EGFP (Fig. 1d). Amplification of the constructed plasmids was conducted using DH10B competent cells. Total RNA from Vero cells infected with ZIKA-EGFP was extracted, and EGFP expression was examined by PCR using primer 1 (5' AGTTGTTGATCTGTGTGAATCAGACTG 3') targeting the Zika sequence and primer 2 (5'CTTGTACAGCTCGTCCATGCCGAGAGTG3') targeting the EGFP sequence. Cytopathic changes and expression of green fluorescence were monitored. Western blot was then used to measure levels of EGFP, NS1 and E protein in Vero cells infected with ZIKA-EGFP, ZIKA-WT and wide-5 type Zika virus.

Viral Replication Ability
ZIKA-WT and ZIKA-EGFP, each at a multiplicity of infection (MOI) of 1, were used to infect Vero cells.
Supernatants and cells were collected after 12 h, 24 h, and 48 h, respectively, and total RNA was extracted and reverse transcribed. The replication of viral RNA in cells was measured using fluorescence-based quantitative PCR (qPCR) [18] under the following conditions: one cycle at 95 °C for 2 min, followed by a three-step procedure comprising 30 s at 95 °C, 15 s at 55 °C, and 15 s at 72°C for 40 cycles (with data collection at the end of the 72 °C step at each cycle) and cooling at 37 °C for 10 min. The sequences of primers and probe are provided in Table 1. An assay to determine the 50% tissue culture infectious dose (TCID 50 ) was used to determine viral titer in the supernatant, with TCID 50 values calculated according to the Reed-Muench method.
Plaque assays was used to determine the virulence of ZIKA-WT and ZIKA-EGFP [19]. Vero cells were seeded in 24-well plates at 5 × 10 4 cells/well, and the following day, an appropriate viral titer was used to infect the cells. At 2-h post-infection, 2% carboxymethylcellulose was used to overlay the cells, and after 6 days, Crystal Violet staining was performed to observe plaque sizes.

Inhibitory Effect of Caffeic Acid Phenethyl Ester (CAPE; Selleck, Shanghai, China) on ZIKA-EGFP Viral Replication
ZIKA-EGFP and ZIKA-WT (MOI: 0.1) were used to infect Vero cells, and at 2-h post-infection, the supernatant was discarded and replaced with a media containing 10 μM CAPE. After a 48-h incubation, relative expression levels of the viruses were measured using fluorescence-based qPCR.

Production of Rescued Virus
The pBAC-ZIKA-EGFP clone generated as described in Fig. 1 was used to transfect 293T cells, supernatant from which was collected at 7-days post-transfection and used to infect Vero cells. The cells were passaged every 4 to 6 days for a total of five consecutive passages. We extracted total RNA from the cells after infection and identified EGFP expression by PCR, demonstrating the stability of the ZIKA-EGFP viral sequence following cell passage (Fig. 2a)

Viral Replication Ability
Levels of viral RNA in Vero cells infected with ZIKA-EGFP (MOI: 0.1) for 12 h, 24 h, and 48 h were measured using fluorescence-based qPCR. The results showed that ZIKA-EGFP viral RNA could be amplified in Vero cells within a time period comparable with that of the ZIKA-WT (Fig. 3a). Cell-culture supernatants were then collected and subjected to TCID 50 assay, which showed similar titers of ZIKA-EGFP and ZIKA-WT (Fig. 3b). These findings indicated similar replication abilities between the ZIKA-EGFP and ZIKA-WT.

Viral Pathogenicity
After infection of SH-SY5Y cells with the ZIKA-EGFP and ZIKA-WT (MOI: 1) for 48 h, we observed prominent cytopathic changes, with dead cells seen floating in the culture media. Additionally, CCK-8 assays showed that the 72-h viability of ZIKA-EGFP-infected cells decreased significantly, and that virulence was only slightly weaker than that of the ZIKA-WT (Fig. 4a).
Plaque assay results showed that infection with the ZIKA-WT led to the formation of prominent, perfectly round, and uniform plaques after 6 days (Fig. 4b). By contrast, infection with the ZIKA-EGFP did not result in plaque formation, although pronounced cell clusters were observed by fluorescence microscopy (Fig. 4c).

Inhibitory Effect of CAPE on Viral Replication
Vero cells infected with the ZIKA-EGFP and ZIKA-WT, respectively, were treated with 10 μM CAPE for 48 h, with fluorescence-based qPCR results revealing that CAPE significantly inhibited viral replication (Fig. 5a). These findings were verified by fluorescence microscopy showing no fluorescence, which agreed with the results of fluorescence-based qPCR (Fig. 5b).

Statistics Analysis
All experiments were repeated at least three times. Data were presented as mean ± SEM. Results were analyzed using Graphpad Prism (GraphPad Software Inc., La Jolla, CA). Comparison between experimental groups was made by Student's t-test and one-way ANOVA. A value of P<0.05 was considered statistically significant.

Discussion
The reverse genetics approach is an important method for studying viral gene function, virulence variation, and drug screening. The presence of cryptic bacterial promoters in flaviviruses has increased the challenges associated with construction of infectious clones [17,20]. The common solution to this issue is mutation of the cryptic promoter or utilization of a single-copy plasmid-based bacterial artificial chromosome. In this study, we used a bacterial artificial chromosome to reduce the toxicity of the viral DNA in order to obtain a full-length clone of Zika virus.
Infectious clones can be constructed in three types of systems. The first is the in vitro transcription system [21][22][23], where a full-length DNA clone is used as a template to transcribe viral mRNA molecules using T7 or SP6 RNA polymerase. The purified viral RNA is then used to transfect cells via electrotransfection or other methods to produce rescued viruses. This method requires highly pure mRNA and often expensive reagents and longer experimental times, because suitable transfection methods and conditions need to be optimized. The second approach involves use of eukaryotic systems [1,24,25], where a clone containing the full-length viral sequence is constructed and placed under the control of a eukaryotic promoter, such as CMV or SV40. Upon transfection of target cells with the plasmid, viral mRNA can be generated using the transcription machinery of the eukaryotic 8 cells, enabling subsequent rescue of the viruses produced in the cells. This method generally does not require expensive reagents or instruments, is simple to perform, and the conditions are easily optimizable. However, the limitation of this method involves the need for transcription inside the nucleus during generation of viral mRNA. Chen Y et al. used this method to construct infectious clones and did not observe incorrect pre-mRNA splicing, which could be due to sequence differences in different viral strains [24]. The third approach is using recombination systems with overlapping fragments [18,26]. In this system, the full-length DNA clone and the transcription-control elements are amplified as three fragments, which are then transfected into target cells via electrotransfection.
The complete viral genome sequence is then generated in the cells by homologous recombination to subsequently produce the rescued viruses. This method can avoid the toxic effect of the cryptic bacterial promoter, however, the transfection conditions need to be optimized, and the efficiency of homologous recombination can greatly affect viral production.
To facilitate investigations of Zika virus pathogenesis, previous studies attempted to include a marker, such as EGFP or luciferase at C25, C33, or other positions, in the Zika viral genome [20,27].
However, most attempts did not lead to stable expression of the markers. Upon analysis of the secondary structure of the viral RNA, we designed the marker-insertion site at the C37 position, which was in the middle of a stem. Our results showed that insertion at this site led to a ZIKA-EGFP viral sequence that remained stable in the genome for at least five generations, with no adverse effect on the range of cells that could be infected (the ZIKA-EGFP was able to infect Vero, SH-SY5Y, C6/36, and endothelial cells). Additionally, we found that the viral-replication speed did not differ significantly from that of the ZIKA-WT. Moreover, ZIKA-EGFP was cytopathogenic and led to cytopathic changes but showed weaker virulence than the ZIKA-WT and no plaque formation. These findings indicated the effectiveness of constructing a full-length infectious clone of the ZIKA-EGFP in a eukaryotic system. Furthermore, we noted that EGFP insertion at positions C36 and C38 also produced stable expression clones, and that the vector constructed using this method contained two cyclization sequences (CS) sites. Introduction of a mutation into the second CS site resulted in no change in viral titer, and insertion of luciferase at position C37 resulted in stable expression for at least three generations (data 9 not shown).
To validate the use of ZIKA-EGFP for monitoring drug effectiveness, we performed an experiment using CAPE, shown to inhibit Zika viral replication in our previous experiments. The results showed that treatment with the drug significantly suppressed fluorescence associated with EGFP levels, which was consistent with results observed following inhibition of RNA replication. These findings suggested that ZIKA-EGFP could be applicable for use in drug screening.

Conclusions
In summary, we successfully constructed an infectious clone of the full-length genome of Zika virus stably expressing an EGFP marker in a eukaryotic expression system. The results presented here provide a foundation for future studies on ZIKA viral gene mutation, function, and drug screening.

Competing interests
The authors declare that they have no competing interests