Optimized Production of Virus-like Particles in a High-CHO-Cell-Density Transient Gene Expression System for Foot-and-Mouth Disease Vaccine Development

Abstract

Background/objectives: Foot-and-mouth disease virus (FMDV) poses a continuous threat to livestock health and agricultural economies. Current vaccines require high biosafety standards and are costly to produce. While novel vaccine technologies have been explored, most fail to meet industrial scalability, cost-efficiency, or multiserotype flexibility required for effective FMD control. This study aimed to evaluate the feasibility of using a high-cell density transient gene expression (TGE) system in CHO cells for the production of FMDV virus-like particles (VLPs) as a recombinant vaccine platform.

Methods: VLP expression was optimized by adjusting cDNA and polyethyleneimine (PEI) concentrations. Expression yields were compared at 24 and 48 h post-transfection to determine optimal harvest timing. We further tested the system's capacity to express different serotypes and chimeric constructs, incorporating VP1 sequences from various FMDV strains. Immunogenicity was evaluated in swine using VLPs from the A2001 Argentina strain as a model.

Results: Optimal VLP expression was achieved at 24 h post-transfection. Chimeric constructs incorporating heterologous VP1 regions were successfully expressed. Immunized pigs developed protective antibody titers as measured by a virus neutralization test (VNT, log₁₀ titer 1.43) and liquid-phase blocking ELISA (LPBE, titer 2.20) at 28 days post-vaccination (dpv). Titers remained above protective thresholds up to 60 dpv with a single dose. A booster at 28 dpv further elevated titers to levels comparable to those induced by the inactivated vaccine.

Conclusions: Our results demonstrate the feasibility of using CHO cell-based TGE for producing immunogenic FMDV VLPs. This platform shows promise for scalable, cost-effective, and biosafe development of recombinant FMD vaccines.

Keywords: CHO cells; foot-and-mouth disease virus; virus-like particles.

Figure 1

For plasmids synthesis, FMDV A2001 Argentina strain sequence was used. P12A sequences were codon-optimized for mammalian expression.

Figure 2

Expression levels were quantified by ELISA (a) and protein expression was confirmed by Western blotting (b). Line 1 corresponds to the VP1/VP3 proteins from A2001 virus as positive control and line 4 to untransfected CHO-3E7 cells.

Figure 3

CHO-3E7 cells were transfected with different combinations of pTT5-P12A3C, pTT5-P12A and pTT5-3C plasmids, harvested 24 hpt and analyzed by Western Blot. S supernatant, L lysate. GFP was used for transfection control.

Figure 4

CHO-3E7 cells were transfected with different combinations of cDNA and PEI using GFP as an intracellular model protein. Cells were harvested at 24 hpt. (a) GFP+ cells for cells transfected at a density of 5 × 10⁶ viable cells/mL, (b) optimal predicted cDNA and PEI concentration when cells were transfected at a density of 5 × 10⁶ viable cells/mL, (c) GFP+ cells when cells were transfected at a density of 7 × 10⁶ viable cells/mL, (d) optimal predicted cDNA and PEI concentration when cells were transfected at a density of 7 × 10⁶ viable cells/mL.

Figure 5

(a) CHO-3E7 cells were transfected with GFP at cell densities of 5 × 10⁶ viable cells/mL and 7 × 10⁶ viable cells/mL with the DNA and PEI concentrations that gave the highest GFP+ cell value experimentally and the predicted value based on the DOE results. (b) CHO-3E7 cells were transfected with FMDV VLPs encoding plasmids at densities of 5 × 10⁶ viable cells/mL and 7 × 10⁶ viable cells/mL at day of transfection with the DNA and PEI concentration that gave the highest GFP+ cell value experimentally and analyzed with ELISA. The experiments were repeated three times independently at figure (a) and five times at figure (b). Arrows indicate the DNA and PEI concentrations, and cell density selected for future experiments.

Figure 6

CHO-3E7 cells were transfected with the optimized transfection conditions with FMDV VLPs encoding plasmid from different serotypes and chimeras. Cells were harvested at 24 hpt and cell lysates were analyzed by Western Blot using guinea pig polyclonal antibodies anti A2001 Argentina strain. (experiments were performed in triplicate for A2001 (293-6E cells) and AA, A2001, A/Asia and AO (CHO-3E7) cells and quintuplicate for A2001 (CHO-3E7).

Figure 7

CHO-3E7 and 293-6E cells were transfected with the optimized transfection conditions with FMDV VLPs encoding plasmid and compared yield with ELISA. Also, the viability of cell culture was measured.

Figure 8

Sucrose gradient characterization of CHO cells crude lysates 24 hpt. A2001 Argentina VLPs (CHO API) were loaded onto a 45–15% sucrose gradient. Fractions were collected (X–axis) and analyzed by solid phase ELISA. The OD 450 (Y-Axis) shows the absorbance at the position of empty capsids (99.37%) and pentamers (0.6292%).

Figure 9

Vaccination schedule. Swine were immunized at 0 and 28 dpv with W/O vaccines containing 5 µg, 25 µg or 50 µg of A2001 Argentina VLPs or with a vaccine containing inactivated virus. Serum samples were collected at 0, 28 and 60 days post vaccination.

Figure 10

FMDV antibodies by Liquid-phase blocking ELISA. Conditions and experimental groups are described in Table 3. Cut off A2001 Argentina 2.2 log10 [26].