Development and Characterization of Recombinant Virus Generated from a New World Zika Virus Infectious Clone

Construction of PRVABC59 infectious clone.

The parental virus from which the infectious clone was derived, PRVABC59, was isolated from a 2015 serum specimen from Puerto Rico (9) and passaged in Vero cells. The sequence of the virus was determined by next-generation sequencing (RNA-Seq) and supplemented by rapid amplification of cDNA ends (RACE) to generate 5′ and 3′ untranslated sequences. The sequence obtained using this method matched published sequences for PRVABC59, which includes a variable amino acid at E-330. Published phylogenetic analyses of ZIKV strain PRVABC59 place it within the Asian clade of ZIKV (9) and indicate that it is similar to other strains currently circulating in the Americas.

An infectious clone was constructed for the PRVABC59 ZIKV strain. Preliminary studies indicated that full-length infectious clones were unstable in Escherichia coli (data not shown), likely due to cryptic prokaryotic promoters within the E and NS1 coding sequences (18). Therefore, we developed a two-plasmid system that would allow the 5′ and 3′ ends of the viral genome to be joined through unique ApaLI and SapI restriction enzyme sites in NS1 (Fig. 1). Plasmid pJW231 contained the 5′ end (nucleotides [nt] 1 to 3498), and plasmid pJW232 contained the 3′ end (nt 3109 to 10809) of the ZIKV genome. Plasmid pJW231 was engineered to contain a T7 RNA polymerase promoter sequence that initiates transcription at nt 1 of the viral genome. Plasmid pJW232 was generated to contain a hepatitis D virus (HDV) ribozyme to produce an authentic 3′ end of the viral genome. A single synonymous mutation was identified at nucleotide position 8489 (C to T) and retained in order to allow infectious clone-derived virus to be distinguished from parental PRVABC59 (Table 1). To prepare RNA from the two-plasmid system, pJW231 (ApaLI and PmlI) and pJW232 (SapI and PvuI) were digested, PCR purified, and ligated using Gibson assembly. From the assembled fragments, capped T7 RNA transcripts were generated, and the resulting RNA was electroporated into Vero cells. Eight days after electroporation, 1.25 × 10⁷ PFU/ml were obtained in the culture supernatant. Viral RNA was isolated from recovered virus and sequenced using RNA-Seq. The expected synonymous mutation at position 8489 was observed, indicating that the virus was infectious clone derived.

In vitro growth properties of primary and infectious clone-derived PRVABC59.

In order to compare the in vitro replication kinetics of the PRVABC59 isolate and the ZIKV infectious clone-derived virus, we inoculated multiple mammalian and mosquito cell lines and measured infectious viral titers by plaque assay (Fig. 2A) and reverse transcription-quantitative PCR (qRT-PCR) of viral RNA in the culture supernatant (Fig. 2B). The replication profiles of the isolate and the infectious clone-derived virus were essentially indistinguishable. Peak viral titers occurred on day 3 to 4 postinoculation in mammalian cell lines, including Vero cells and multiple human cell lines, i.e., placental cells (JAR), testicular cells (NTERA2 cl.D1), neuroblastoma cells (SH-SY5Y), and Huh7 cells (liver), and were not statistically different between the parental strain and the infectious clone-derived virus. A difference in infectious titers between parental and infectious clone-derived virus was observed in BHK cells at early time points but resolved at later times, as indicated by qRT-PCR. Titers of infectious virus in the tissue culture supernatant peaked on day 5 to 6 in two Aedes mosquito cell lines, C6/36 and Aag2.

Transmission of PRVABC59 WT and clone-derived virus by Aedes aegypti mosquitoes.

In order to be maintained in nature, arboviruses must infect and replicate within the midguts of mosquito vectors, disseminate to secondary tissues, and ultimately be expectorated in the saliva during bloodfeeding. Therefore, we tested the ability of the infectious clone-derived virus to be transmitted by relevant vector mosquitoes relative to that of the parental virus. Female Aedes aegypti mosquitoes derived from Poza Rica, Mexico, were exposed to 8 × 10⁶ PFU of each virus in an artificial blood meal. Fourteen days post-blood feed, mosquitoes were examined for the presence of virus in bodies, dissemination of virus into legs, and presence of virus in expectorated saliva. These A. aegypti mosquitoes were highly susceptible to both cloned and parental ZIKV by the oral route, with 75 to 100% of individuals becoming infected (Table 2). Dissemination from infected midguts was also highly efficient, with nearly 100% of infected mosquitoes exhibiting dissemination after 14 days of extrinsic incubation. Furthermore, approximately 50% of infected mosquitoes transmitted both parental and clone-derived virus.

Pathogenesis of PRVABC59 primary isolate and infectious clone-derived virus in mice.

To evaluate the pathogenesis of the PRVABC59 isolate and recombinant ZIKV, we inoculated AG129 mice (alpha/beta and gamma interferon receptor deficient), which have previously been shown to be a model for ZIKV pathogenesis (12, 13). Four-week-old AG129 mice (11 per group; 7 males and 4 females) were inoculated with 1,000 PFU of either PRVABC59 or infectious clone virus via intraperitoneal injection. Weight and survival were followed after infection. Disease symptoms, including ruffled fur, hunched posture, tremors, conjunctivitis, ataxia, hind limb paralysis, and recumbency, were observed starting 8 to 9 days postinoculation (dpi). Mortality was observed starting on day 10 postinoculation for both the parental isolate and the recombinant virus, with 100% mortality by dpi 14 (Fig. 3A). Median survival times were 11.5 and 11 days for the parental and infectious clone-derived viruses, respectively, and survival curves did not differ significantly. Both groups of mice showed weight loss starting at 7 dpi, with differences not statistically significant (Fig. 3B).

To examine serum viremia and tissue distribution of infection for the PRVABC59 isolate and the infectious clone-derived virus, serum samples and tissues (brain and testes) were collected from three 4-week-old male AG129 mice on days 1, 3, and 5 postinoculation. Serum titers for both the parental PRVABC59 strain and the infectious clone-derived virus were detectable on dpi 1, peaked at 6 to 7 log₁₀ PFU/ml sera on dpi 3, and dropped to 4 log₁₀ PFU/ml sera by dpi 5 (Fig. 4A). Additionally, both the parental and infectious clone-derived viruses were observed in the testes and brains on dpi 3 and 5, with titers of 4 to 5 log₁₀ PFU/g in the brain (Fig. 4B) and 5 to 6 log₁₀ PFU/g in the testes (Fig. 4C). Differences in mean titers between the parental strain and the infectious clone-derived virus were not statistically significant.

Sequence diversity of clone-derived virus versus parental ZIKV PRVABC59.

RNA-Seq was performed in order to compare the genetic diversities of clone-derived virus and wild-type PRVABC59. Clone-derived virus was less diverse by all measures tested (Fig. 5A). Most of the variants found in the two viruses were present at relatively low frequencies across the genome (Fig. 5B and C). A single synonymous variant at nt 7200 in the clone-derived virus (but not in parental virus) was detected at 28.6%, while all other variants were detected at less than 8% of the population and most were under 2%.

ZIKV isolate.

The PRVABC59 ZIKV strain was originally isolated by passage on Vero cells of sera from an individual returning from travel to Puerto Rico collected in December of 2015 (GenBank accession number KU501215) (9). A third Vero cell-passaged stock was utilized for the generation of amplicons for clone construction as well as for all in vivo and in vitro comparative assessments. The sequence of the isolate was confirmed by amplifying the genome in 8 overlapping fragments by RT-PCR, followed by direct sequencing of amplicons. 5′ and 3′ RACE reactions (Invitrogen) were performed to generate amplification products of the terminal untranslated regions and sequenced directly. Primer sequences are available upon request.

Construction of ZIKV infectious clone.

Briefly, viral RNA was extracted using the DirectZOL RNA extraction kit (Zymo, Irvine, CA) with on-column DNase treatment. cDNA was then produced using random nonamers using the Superscript IV first-strand synthesis system (Invitrogen, Carlsbad, CA). ZIKV fragments were amplified with Q5 high-fidelity polymerase (New England BioLabs) and cloned into the pACYC177 vector using Gibson assembly or circular polymerase extension cloning (CPEC) to generate two plasmids. Plasmid pJW231 containing the sequences for the 5′ untranslated region (UTR) to nt 3498 and plasmid pJW232 containing the sequence from nt 3109 to the end of the 3′ UTR were generated. An overlapping region of the genome is found in both plasmids, allowing the 5′ and 3′ portions of the viral genome to be conjoined following digestion with ApaLI and SapI. The sequence for the T7 promoter and hepatitis D virus ribozyme were previously cloned into a West Nile virus infectious clone (19). Clones with the correct restriction pattern were verified by sequencing, and correct clones were selected for virus rescue.

In vitro transcription and transfection.

Virus was rescued by RNA transcription from ligated full-length infectious clone DNA. pJW231 was prepared by digestion with ApaLI and PmlI and pJW232 was prepared by digestion with SapI and PvuI, leaving a 52-bp overlapping region that could be joined using Gibson assembly. Digested products were PCR purified (Macherey-Nagel, Germany) and ligated using the NEBuilder HiFi DNA Assembly master mix (New England BioLabs, Ipswich, MA). After a 1-h incubation at 50°C, the reaction mixture was purified and used in a T7 RNA transcription reaction using the ARCA 2× T7 master mix (NEB). RNA was electroporated into Vero cells (170 V, 0 resistance, and 950 μF; 1 pulse with the BTX ECM 630 electroporation system), and cells were subsequently monitored for cytopathic effect (CPE) daily. On day 8 postelectroporation, significant CPE was observed and supernatant was harvested, centrifuged to remove cellular debris, and supplemented to a final concentration of 20% fetal bovine serum (FBS) and 10 mM HEPES before being frozen in single-use aliquots. The titer was measured using the plaque assay on Vero cells.

Recovered infectious clone virus was sequenced using amplicon-based RNA sequencing with next-generation sequencing (NGS). For NGS, RNA was isolated from recovered virus, cDNA was produced, and PCR was performed to amplify amplicons for sequencing. DNA was then fragmented with double-stranded DNA (dsDNA) fragmentase (NEB) for 15 min at 37°C. Fragmented DNA was then purified with AMPure XP beads (1.2 bead/DNA ratio) before library preparation with the NEBNext Ultra II DNA library prep kit for Illumina sequencing in accordance with the manufacturer's protocol. Sequencing was subsequently performed using the Illumina NextSeq system. After demultiplexing and trimming of adapters and removal of low-quality reads, fastq files were aligned to the PRVABC59 reference file using Mosaik2 (20). Variants were then called using LoFreq with standard settings (21). Genetic diversity was assessed using distance (i.e., sum of the variant frequencies), complexity (i.e., Shannon entropy, the uncertainty of randomly sampling a variant at each site), and nucleotide diversity (i.e., the sum of the variant frequencies per genome length).

Cell culture and in vitro virus studies.

Vertebrate cell lines were maintained at 37°C with 5% CO₂. Mosquito cell lines were maintained at 28°C and 5% CO₂. The JAR human placental cell line (ATCC HTB-144) was grown in RPMI medium supplemented with 10% FBS and 1% penicillin-streptomycin (Pen-Strep). The SH-SY5Y human neuroblastoma cell line (ATCC CRL-2266) was a gift from D. Beckham and was maintained in DMEM (Dulbecco's modified Eagle's medium)/F12 medium supplemented with 10% FBS and 1% Pen-Strep. The NTERA-2 cl.D1 human testicular cell line (ATCC CRL-1973), Vero cells, and C6/36 mosquito cells (all obtained from ATCC) were grown in DMEM supplemented with 10% FBS and 1% Pen-Strep. Huh7 cells (a gift from R. Kuhn) were grown in DMEM supplemented with 10% FBS and 1% Pen-Strep. Aag2 mosquito cells were grown in Schneider's medium supplemented with 10% FBS and 1% Pen-Strep.

All viral inoculations were performed in 12-well plates. Cells were grown to 70 to 90% confluence before inoculation, and infections were performed at a multiplicity of infection (MOI) of 0.1 in triplicate. Culture supernatant was harvested every 24 h, and virus was titrated on Vero cells by plaque assay as previously described (22), with the modification that the second overlay was added 4 days postinoculation. The lower limit for this assay was 1.7 log₁₀ PFU/ml. Significance was determined by an unpaired t test with a Holm-Sidak correction for multiple comparisons (Prism).

Real-time RT-PCR.

Viral RNA was extracted from cellular supernatant using the Qiagen viral RNA minikit. Quantification of viral RNA was performed by a real-time RT-PCR assay using previously described ZIKV primers and probe (23). To make a standard curve, a fragment of ZIKV from nt 859 to 1278 was amplified by RT-PCR, and the fragment was TOPO cloned into pcDNA3.1 (Invitrogen). The plasmid was linearized with BglII (NEB), in vitro transcribed using the Ampliscribe T7 transcription kit (Epicentre), and DNase treated.

Mosquito infections and analysis.

Infectivity of the primary PRVABC59 isolate and infectious clone-derived virus was determined in A. aegypti mosquitoes (Poza Rica). The mosquitoes used were originally collected from wild populations in Mexico (24). Mosquitoes were maintained on calf blood and given 10% sucrose ad libitum. Larvae were reared and adults were maintained under controlled conditions of temperature (28°C), humidity (70% relative humidity), and light (14-h light–10-h dark diurnal cycle). Experiments involving infectious ZIKV in mosquitoes were performed under biosafety level 3 (BSL3) conditions. Generation F13 female mosquitoes were given an infectious blood meal containing 8 × 10⁶ PFU of either primary PRVABC59 or infectious clone-derived virus in defibrinated calf blood. Following each infectious blood meal, back-titration was performed to ensure that the virus titers to which mosquitoes were exposed during each experimental replicate were comparable. At 14 days postinfection, mosquitoes were incapacitated using triethylamine. The legs were removed and transferred into a tube containing 500 μl mosquito diluent (1× phosphate-buffered saline [PBS] supplemented with 20% FBS [heat-inactivated], 50 μg/ml Pen-Strep, 50 μg/ml gentamicin, and 2.5 μg/ml amphotericin B [Fungizone]) and a stainless steel bead for homogenization. The mosquito proboscis was inserted into a capillary tube containing immersion oil (∼5 μl) and allowed to salivate for 20 min. Following salivation, mosquito bodies were placed in a separate tube containing mosquito diluent and a bead. The ends of capillary tubes containing oil and saliva were broken off into microcentrifuge tubes containing 250 μl mosquito diluent. Mosquito tissues (bodies and legs) were homogenized at 25 cycles/s for 1 min using a Retsch mixer mill MM400 (Germany) and centrifuged at 15,000 × g for 5 min at 4°C. Clarified supernatant was titrated by Vero cell plaque assay to determine the amount of infectious virus present in mosquito bodies. Saliva titers were confirmed by qRT-PCR. The presence of infectious virus in mosquito bodies was defined as infection positive. Dissemination is reported as the percentage of mosquitoes that became infected and contained infectious virus in their legs. The transmission rate is reported as the number of infected mosquitoes that contained infectious virus in the saliva.

In vivo growth and pathogenesis in mice.

Groups of 20 4-week-old alpha/beta and gamma interferon receptor-deficient AG129 mice were inoculated with 1,000 PFU of parental or infectious clone-derived ZIKV in a 0.1-ml inoculum by the intraperitoneal route. Mice were weighed and monitored daily for signs of disease. On days 1, 3, and 5 postinoculation, 3 male mice per group were deeply anesthetized, a blood sample was obtained by cardiac puncture, and mice were perfused with PBS prior to harvesting of tissues. The remaining 11 mice (7 males and 4 females per group) were euthanized when evidence of neurological disease was observed, and blood and tissues were obtained. Euthanasia criteria included hind limb paralysis or weakness, incoordination, and tremors. Additional euthanasia criteria included recumbency or conjunctivitis that precluded vision. Survival curves were compared using a log rank (Mantel-Cox) test (Prism). Serum was separated by centrifugation of whole blood. Tissues were homogenized in BA-1 medium (M199-H, 0.05 M Tris buffer, 1% bovine serum albumin, 2 mM l-glutamine, 4.2 mM sodium bicarbonate, 1% Pen-Strep, 1 μg/ml amphotericin B) by using a pestle and clarified by centrifugation. Serum and tissues were titrated by Vero cell plaque assay. Significance was determined by an unpaired t test with a Holm-Sidak correction for multiple comparisons (Prism). All animal studies were conducted under CDC-approved IACUC protocols.