A replication-incompetent Rift Valley fever vaccine: chimeric virus-like particles protect mice and rats against lethal challenge

Abstract

Virus-like particles (VLPs) present viral antigens in a native conformation and are effectively recognized by the immune system and therefore are considered as suitable and safe vaccine candidates against many viral diseases. Here we demonstrate that chimeric VLPs containing Rift Valley fever virus (RVFV) glycoproteins G(N) and G(C), nucleoprotein N and the gag protein of Moloney murine leukemia virus represent an effective vaccine candidate against Rift Valley fever, a deadly disease in humans and livestock. Long-lasting humoral and cellular immune responses are demonstrated in a mouse model by the analysis of neutralizing antibody titers and cytokine secretion profiles. Vaccine efficacy studies were performed in mouse and rat lethal challenge models resulting in high protection rates. Taken together, these results demonstrate that replication-incompetent chimeric RVF VLPs are an efficient RVFV vaccine candidate.

Fig. 1

Characterization of RVF VLPs. (A) Western blot analysis of chimeric RVF VLPs (chimVLP): concentrated supernatants from 293-gag cells transfected with RVFV G and N expression plasmids analyzed by Western blotting using antibodies specific for RVFV G_N, G_C, N and Moloney murine leukemia virus (MoMLV) gag. (B) Western blot analysis of RVF VLPs: Concentrated supernatants from 293 cells transfected with RVFV G and N expression plasmids analyzed by Western blotting using antibodies specific for RVFV G_N, G_C and N. (C) Negative staining of RVFV G and N transfected 293 cells fixed with glutaraldehyde and stained with uranyl acetate and examined by transmission electron microscopy. Scale bar represents 200 nm. Left panel: arrows point to budding VLPs; right panel: arrows indicate RVFV G spikes protruding from the VLP membrane. (D) Western blot analysis of RVF chimVLPs: Concentrated supernatants from 293-gag cells transfected with RVFV N and wild type (WT) or codon-optimized (CO) RVFV G sequences.

Fig. 2

Optimization of RVF VLP production. (A) Different plasmid ratios to determine optimal chimeric RVF VLP production: 293-gag cells were transfected with 15, 9 or 3 μg of the RVFV G expression plasmid, and co-transfected with 0, 3, 9 or 15 μg RVF N expression plasmid as indicated. Concentrated supernatants of a 60 h post-transfection harvest were analyzed by Western blot using antibodies specific for RVFV G_N. (B) Time course experiment to optimize RVF VLP yields: Western blot analysis was performed using antibodies specific for RVFV G_N and G_C, as indicated at each blot of RVF chimVLPs and VLPs harvested at select times post-transfection. Densitometric analysis of band intensity, displayed as % maximum band intensity for a particular blot, is represented by histograms above each Western blot. Harvest times are indicated.

Fig. 3

Neutralizing antibody titer in RVF VLP-vaccinated mice determined by plaque reduction neutralization tests. Mouse sera were collected after three immunizations with RVF chimVLPs, RVF VLPs with or without RVFV N, Ebolavirus GP-pseudotyped MoMLV (Control 1) and from unimmunized mice (Control 2). The neutralizing antibody titer was determined as the reciprocal of the dilution of five two-fold serial dilutions of sera, respectively. Neutralizing antibody titer is considered positive at the lowest initial serum dilution that results in  > 80% (PRNT₈₀) reduction of the number of plaques as compared to the virus control.

Fig. 4

Antigen-specific cytokine secretion by splenocytes of chimeric RVF VLP-vaccinated mice measured by multiplex analysis. Tissue culture supernatants from splenocytes harvested at 24, 96 or 168 h post-antigen stimulation from mice immunized with RVF chimVLPs, influenza VLPs or an unvaccinated mouse were subjected to multiplex bead analysis to measure select cytokines. Time of harvest and the cytokine measured are described at the top and left of the panels, respectively. Data from splenocytes harvested from the control mouse are indicated by red bars, and data from RVF chimVLP-vaccinated mice are indicated by blue bars. Stimulatory antigens are indicated below the graphs. Insets show the same results as the main graphs except with reduced scales (Y-axis, pg/ml) to show that secretion is detected at many points but is masked by the scale required to illustrate maximal cytokine signal.

Fig. 5

RVF VLP efficacy studies in mice. RVF VLP-vaccinated mice (n = 16) were challenged with 1 × 10³ pfu of RVFV strain ZH501 under BSL-4 conditions. Mice immunized with an Ebolavirus GP-pseudotyped MoMLV vaccine and unimmunized mice served as controls 1 and 2, respectively. Data are shown in a Kaplan–Meier format.

Fig. 6

RVF VLP efficacy studies in rats. Results are shown as Kaplan–Meier survival curves and weights of VLP-immunized rats after RVFV challenge. (A) Rats (n = 6) were inoculated with the RVF chimVLP vaccine candidate and then challenged with 1 × 10⁵ pfu of RVFV strain ZH501. Control rats were immunized with sterile saline. (B) The mean and standard deviation of the weight change for RVF chimVLP vaccinated (circles) and control (squares) rats are shown for each time point.