High-Titer Self-Propagating Capsidless Chikungunya Virus Generated in Vero Cells as a Strategy for Alphavirus Vaccine Development

Abstract

In our previous study, we found that a new type of Chikungunya virus particle with a complete capsid deletion (ΔC-CHIKV) is still infectious in BHK-21 cells and demonstrated its potential as a live attenuated vaccine candidate. However, the low yield as well as the disability to propagate in vaccine production cell line Vero of ΔC-CHIKV are not practical for commercial vaccine development. In this study, we not only achieved the successful propagation of the viral particle in Vero cells, but significantly improved its yield through construction of a chimeric VEEV-ΔC-CHIKV and extensive passage in Vero cells. Mechanistically, high production of VEEV-ΔC-CHIKV is due to the improvement of viral RNA packaging efficiency conferred by adaptive mutations, especially those in envelope proteins. Similar to ΔC-CHIKV, the passaged VEEV-ΔC-CHIKV is safe, immunogenic, and efficacious, which protects mice from CHIKV challenge after only one shot of immunization. Our study demonstrates that the utilization of infectious capsidless viral particle of CHIKV as a vaccine candidate is a practical strategy for the development of alphavirus vaccine. IMPORTANCE Chikungunya virus (CHIKV) is one of important emerging alphaviruses. Currently, there are no licensed vaccines against CHIKV infection. We have previously found a new type of Chikungunya virus particle with a complete capsid deletion (ΔC-CHIKV) as a live attenuated vaccine candidate that is not suitable for commercial vaccine development with the low viral titer production. In this study, we significantly improved its production through construction of a chimeric VEEV-ΔC-CHIKV. Our results proved the utilization of infectious capsidless viral particle of CHIKV as a safe and practical vaccine candidate.

Keywords: Chikungunya virus; alphavirus; capsid; vaccine.

FIG 1

Growth characterization of VEEV-ΔC-CHIKV particles in BHK-21 and Vero cells. (A) Schematic illustration of CHIKV-WT, ΔC-CHIKV, and VEEV-ΔC-CHIKV genomes. (B) Viral RNA copies detected both in supernatant and cell lysates of Vero cells infected with ΔC-CHIKV at an MOI of 0.01. (C) Immunostaining of ΔC-CHIKV infected Vero cells with different MOI using CHIKV E2 polyclonal antibodies. (D) Immunostaining of BHK-21 cells transfected with equal amounts of VEEV-ΔC-CHIKV or ΔC-CHIKV RNAs (1 μg) at the indicated times posttransfection with CHIKV E2 polyclonal antibodies. (E) IFA analysis of CHIKV-E2 expression using anti-CHIKV E2 rabbit polyclonal antibodies at different passages of VEEV-ΔC-CHIKV and ΔC-CHIKV viruses in both BHK-21 and Vero cells at 72 h postinfection. (F) Growth curves of the recombinant VEEV-ΔC-CHIKV (P0) and ΔC-CHIKV viruses (P0) as well as WT-CHIKV in both BHK-21 and Vero cells. The recombinant VEEV-ΔC-CHIKV and ΔC-CHIKV viruses obtained from BHK-21 cells were used to infect BHK-21 and Vero cells at an MOI of 0.01, and the cell supernatants were collected at the indicated times for plaque assay using BHK-21 cells. Error bars indicate the standard derivation (SD) of three independent experiments. The dashed line indicated the detection limit.

FIG 2

Generation of higher titer VEEV-ΔC-CHIKV virus in Vero cells by extensive passaging. (A) Growth curves comparison of different passages (P0, P10, P20, P30, P40, P50) of VEEV-ΔC-CHIKV in Vero cells. Vero cells were infected at an MOI of 0.01, and the cell supernatants were collected at the indicated times for determination of virus titers by plaque assay using BHK-21 cells. The data are representative of three independent experiments, and error bars indicate the SD. (B) IFA analysis of CHIKV-E2 and Capsid in WT or the passaged VEEV-ΔC-CHIKV infected Vero cells. (C) Plaque morphology comparison between CHIKV-WT and VEEV-ΔC-CHIKV at P0, P30, and P50. BHK-21 cells were infected with indicated viruses, and plaques were developed after 72 h. (D) Plaque diameter comparison of VEEV-ΔC-CHIKV virus at P0, P30, and P50 in BHK-21 cells at 72 hpi. ns, not significant. (E) Immunogenicity comparison between CHIKV-WT and P50 passaged VEEV-ΔC-CHIKV viruses. Two independent experiments were performed in triplicate. Data represent the mean ± SD of triplicate measurements in a representative experiment. The asterisks denote statistical differences between the indicated groups. ns, not significant. (F) Infectivity of VEEV-ΔC-CHIKV in Huh-7 (hepatoma), A549 (lung adenocarcinoma), and MRC-5 (lung fibroblast cell line) cell lines. The above all cell lines were infected with VEEV-ΔC-CHIKV at an MOI of 1. At 36 h postinfection, CHIKV-E2 expression was detected using anti-CHIKV E2 polyclonal antibodies.

FIG 3

The passaged VEEV-ΔC-CHIKV produces more infectious virus particles. (A) Western blotting of CHIKV E1 expression of different viral loads of CHIKV-WT and VEEV-ΔC-CHIKV at P0, P10, and P50 using CHIKV E1 polyclonal antibodies. (B) Western blotting of sucrose density gradient fractionations of VEEV-ΔC-CHIKV at P0 and P50 produced in Vero cells. 10⁵ PFU P0 and 10⁷ PFU P50 viruses were separated on 20%–60% linear sucrose density gradients. Sixteen fractions were harvested from the top (Fraction 1) to the bottom (Fraction 16) of the gradient. Each fraction from P0 and P50 VEEV-CHIKV was subjected to Western blotting assay using E1-specific antibody. (C) Analysis of the intensity of viral E1 protein bands in each fraction using ImageJ software.

FIG 4

Reverse genetic analysis of the adaptive mutations in P50 VEEV-ΔC-CHIKV. (A) Nucleotide and amino acid changes in the P50 VEEV-ΔC-CHIKV genome. (B) Diagram illustration of reconstruction of VEEV-ΔC-CHIKV mutants with the mutations in either structural proteins (nsPs_mut), structural proteins (E_mut) or both ((nsPs+E)_mut). The titers at 72 hpt in Vero cells for each construct are given on the corresponding line. (C) Plaque morphology comparison among different recombinant viruses. (D) Growth curves of the recovered constructs. Vero cells were infected with the indicated viruses at an MOI of 0.01 and the cell supernatants were collected at the indicated times for plaque assay in BHK-21 cells. The data are representative of two independent experiments, and error bars indicate the SD.

FIG 5

The whole glycoprotein mutations instead of E2 G82R mutation alone account for the enhanced viral titer. (A) Function of Emut mutations in the background of ΔC-CHIKV. Schematic illustration of ΔC-CHIKV-eGFP and ΔC-CHIKV-Emut-eGFP constructs were presented in the top panel. Equal amounts (1 μg) of in vitro transcribed RNA of ΔC-CHIKV-eGFP and ΔC-CHIKV-Emut-eGFP were transfected into Vero cells, respectively. At indicated time points after transfection, eGFP positive cells were detected using a NIKON upright fluorescence microscope (Tokyo, Japan), as shown in the bottom panel. (B) Effect of E2 G82R mutation on the viral titer. The E2 G82R mutation was engineered into the original VEEV-ΔC-CHIKV constructs, as shown in the top panel. Equal amounts (1 μg) of in vitro transcribed RNAs of VEEV-ΔC-CHIKV and VEEV-ΔC-CHIKV-E2 G82R were transfected into Vero cells, respectively. The expressions of viral E2 protein were detected by IFA at indicated time points posttransfection, as shown in the bottom panel. (C) Viral production of VEEV-ΔC-CHIKV and the E2-G82R mutant. The genomes/mL data and vital titers were determined by triplicate qPCRs and standard plaque assay, respectively. Three independent experiments were performed and the average values are shown.

FIG 6

Pathogenicity of VEEV-ΔC-CHIKV in the C57BL/6 mouse model. C57BL/6 mice (6-week-olds, n = 5 per group) were injected s.c. in the ventral/lateral side of the hind foot with 10⁵ PFU of WT CHIKV (ECSA strain), VEEV-ΔC-CHIKV, and the same volume of PBS (negative control), respectively. (A) Footpad swelling. (B) Viremia. (C) Viral distribution in organs of infected mice. Organs from infected mice were collected and homogenized on days 1, 3, and 5 postinfection. The amounts of viruses were quantified using plaque assay. (D) Expression of cytokine or chemokine in ankle tissue of mice infected with both viruses were analyzed using qRT-PCR. Results were normalized to the level of the housekeeping geneβ-actin and expressed as fold changes compared to the levels in the mock control samples. Student's t test was used for statistical analysis: ***, P < 0.001; ****, P < 0.0001. (E) Histology of feet and muscle from WT CHIKV (ECSA strain), VEEV-ΔC-CHIKV, or PBS infected C57BL/6 mice. Scale bar represents 500 μm for footpad left panel, and 50 μm for footpad right panel and muscle panel. The row labeled “footpad” shows no inflammation in the observed ankle joint of VEEV-ΔC-CHIKV and mock-infected mice, while WT CHIKV (ECSA strain) infected mice exhibited muscle edema (black arrow) and necrosis of numerous muscle cells (red arrow) as well as large amounts of inflammatory cell infiltration in the muscle, synovial membrane, and articular cavity (green arrow). In infected muscle analysis, there was no obvious inflammation or necrosis with VEEV-ΔC-CHIKV and mock-infected mice, whereas a large amount of inflammation (yellow arrow) and necrosis (red arrow) was detected in WT CHIKV infected mice. Also, some fibrous connective tissue hyperplasia was seen (green arrow).

FIG 7

VEEV-ΔC-CHIKV protects the C57BL/6 mouse from CHIKV infection. Four groups of 4-weeks-old C57BL/6 mice (n = 5 per group) were immunized s.c. in the ventral/lateral side of the hind foot once with either 10⁴ PFU of Δ5nsP3 or VEEV-ΔC-CHIKV; PBS was used as mock immunization. On day 30 postimmunization, mice were challenged s.c. in feet with 2.5 × 10⁵ PFU of WT CHIKV (ECSA strain). (A) Total anti-CHIKV IgG antibodies and (B) NAb titers in mouse sera at the indicated days postimmunization. (C) Viremia and (D) footpad swelling postchallenge. All the data represent mean ± SD of group mice, and the horizontal dotted line represents the limit of detection. Student's t test or two-way ANOVA was used to determine statistical differences between groups. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001, respectively.

FIG 8

Safety evaluation of VEEV-ΔC-CHIKV in the IFNAR^−/− mouse model. Four groups of 6-weeks-old IFNAR^−/− mice (n = 5 per group) were infected s.c. in the ventral/lateral side of the hind foot with different doses of WT CHIKV or VEEV-ΔC-CHIKV. The PBS infection was used as mock immunization. Animal survival (A), footpad swelling (B), and weight loss (C) as well as viremia (D) were monitored daily until 14 days postinoculation. Student's t test or two-way ANOVA was used to determine statistical differences between groups. *, P < 0.05; **, P < 0.01; ***, P < 0.001; ****, P < 0.0001, respectively.