YF17D-vectored Ebola vaccine candidate protects mice against lethal surrogate Ebola and yellow fever virus challenge

Abstract

Ebola virus (EBOV) and related filoviruses such as Sudan virus (SUDV) threaten global public health. Effective filovirus vaccines are available only for EBOV, yet restricted to emergency use considering a high reactogenicity and demanding logistics. Here we present YF-EBO, a live YF17D-vectored dual-target vaccine candidate expressing EBOV glycoprotein (GP) as protective antigen. Safety of YF-EBO in mice was further improved over that of parental YF17D vaccine. A single dose of YF-EBO was sufficient to induce high levels of EBOV GP-specific antibodies and cellular immune responses, that protected against lethal infection using EBOV GP-pseudotyped recombinant vesicular stomatitis virus (rVSV-EBOV) in interferon-deficient (Ifnar^-/-) mice as surrogate challenge model. Concomitantly induced yellow fever virus (YFV)-specific immunity protected Ifnar^-/- mice against intracranial YFV challenge. YF-EBO could thus help to simultaneously combat both EBOV and YFV epidemics. Finally, we demonstrate how to target other highly pathogenic filoviruses such as SUDV at the root of the 2022 outbreak in Uganda.

Fig. 1. vaccine design and in vitro characteristics.

a Schematic of YF17D and YF17D-vectored Ebola vaccine candidate (YF-EBO). EBOV GP was inserted into the E/NS1 intergenic region as translational fusion within the YF17D polyprotein inserted in the endoplasmic reticulum (gray). To cope with topological constraints of the fold of both EBOV GP antigen and the polyprotein of the YF17D vector, one extra transmembrane domain (derived from the West Nile virus E protein; light yellow) was added to the C-terminal cytoplasmic domain of the full-length EBOV GP protein. Arrows indicate protease cleavage sites. b Representative images of plaque phenotypes from YF17D and YF-EBO on BHK-21J cells, fixed 6 days post-infection. c Growth kinetics of YF17D and YF-EBO. BHK-21J cells were infected at a multiplicity of infection (MOI) of 0.01 and virus yields were quantified over time by virus titration on BHK-21J cells. Error bars indicate SEM (n = 5) and dashed line represents limit of detection (LOD). d Antigenicity of YF-EBO: confocal immunofluorescent images of BHK-21J cells 2 days post-infection with YF-EBO, staining for YF17D (green) and EBOV GP antigen (red) (nuclei stained with DAPI, blue). Scale bar, 25 μm.

Fig. 2. Attenuation of YF-EBO.

a Weight evolution and b survival curve of Ifnar^-/- mice after intraperitoneal inoculation with 250 PFU of YF-EBO (n = 6, green circles), YF17D (n = 6, yellow circles), or sham (n = 6, white circles) as a control. c Weight evolution and d survival curve of AG129 mice after intraperitoneal inoculation with 250 PFU of YF-EBO (n = 8), YF17D (n = 6) or sham (n = 6) as a control. e Weight evolution and f survival curve of 5-day-old suckling BALB/c mice after intracranial inoculation with 25 PFU of YF-EBO (n = 12), 10 PFU of YF17D (n = 10) or sham (n = 6) as a control. The number of surviving mice at study endpoint are indicated within parentheses in the legends and a log-rank test was applied to compare YF-EBO- with YF17D-vaccinated mice, significant p values < 0.05 are indicated (b, d, f). Error bars indicate SEM (a, c).

Fig. 3. EBOV-specific immunity and protection in mice.

a Ifnar^-/- mice were vaccinated intraperitoneally with 250 (circles) or 2500 PFU (squares) of either YF-EBO, YF17D or sham. A subset of mice vaccinated with 250 PFU and all mice vaccinated with 2500 PFU were sacrificed 4 weeks post-vaccination for spleen collection. Remaining mice vaccinated with 250 PFU were challenged intraperitoneally with 100 PFU of rVSV-EBOV 4–6 weeks post-vaccination after which they were monitored daily for 2 weeks for the development of disease symptoms. b EBOV GP-specific IgG binding antibody (bAb) titers determined by IIFA at 4 weeks post-vaccination with 250 PFU (circles; YF-EBO n = 14; YF17D n = 12; sham n = 12) or 2500 PFU (squares; YF-EBO n = 5; YF17D n = 5). c ELISpot counts of IFNγ-secreting cells after EBOV GP peptide pool stimulation of isolated splenocytes from mice vaccinated with 250 PFU (circles; YF-EBO n = 14; sham n = 6) or 2500 PFU (squares; YF-EBO n = 5; YF17D n = 5). A representative image of an ELISpot well from each group is shown below the x-axis. d Percentage of IFNγ-expressing CD8⁺ cells after EBOV GP peptide pool stimulation of splenocytes from mice vaccinated with 250 PFU (YF-EBO n = 8; YF17D n = 7; sham n = 8). e Mean cumulative disease scores of mice post-challenge (YF-EBO n = 11; YF17D n = 10; sham n = 12), determined based on IACUC parameters including: body weight changes, body condition score, behavior and physical appearance. f Percentage survival post-challenge, the number of surviving mice at study endpoint are indicated within parentheses. g rVSV-EBOV infectious viral loads in different organs at 3 days (YF-EBO n = 3; YF17D n = 4; sham n = 3) and 14 days post-challenge (YF-EBO n = 3) quantified by virus titration on Vero E6 cells. Dashed line indicates limit of quantification (LOQ) or limit of detection (LOD). Data are median ± IQR (b–d, g) and two-tailed Kruskal–Wallis test was applied followed by Dunn’s multiple comparison, significant p values < 0.05 are indicated (b–d).

Fig. 4. YFV-specific immunity and protection in mice.

a Ifnar^-/- mice were vaccinated intraperitoneally with 250 (circles) or 2500 PFU (squares) of either YF-EBO, YF17D or sham. A subset of mice vaccinated with 250 PFU and all mice vaccinated with 2500 PFU were sacrificed 4 weeks post-vaccination for spleen collection. Remaining mice vaccinated with 250 PFU were challenged intracranially with 3000 PFU of YF17D after which they were monitored daily for 2 weeks for the development of disease symptoms. b YF17D-specific neutralizing antibody (nAb) titers at 4 weeks post-vaccination with 250 PFU (circles; YF-EBO n = 8; YF17D n = 8; sham n = 6) or 2500 PFU (squares; YF-EBO n = 5; YF17D n = 5). c ELISpot counts of IFNγ-secreting cells after YFV NS3 peptide stimulation of isolated splenocytes from Ifnar^-/- mice that were vaccinated with 250 PFU (circles; YF-EBO n = 14; sham n = 6) or 2500 PFU (squares; YF-EBO n = 5; YF17D n = 5). A representative image of an ELISpot well from each group is shown below the x-axis. d Percentage of IFNγ-expressing CD8⁺ cells after YFV NS4B peptide pool stimulation of splenocytes from mice vaccinated with 250 PFU (YF-EBO n = 8; YF17D n = 7; sham n = 7). One sham mice with high background staining in ICS was excluded from the analysis because it was detected as an outlier with ROUT method (Q = 0.1%). e Mean weight evolution, error bars indicate SEM (YF-EBO n = 8; YF17D n = 8; sham n = 6), and f survival curves after challenge, number of surviving mice at study endpoint are indicated within parentheses. g YF17D infectious viral loads in the brain at day of euthanasia determined by virus endpoint titration on BHK-21J cells. Dashed line indicates limit of detection (LOD). Data are median ± IQR and two-tailed Kruskal–Wallis test was applied followed by Dunn’s multiple comparison (b–d, g) and a log-rank test was applied to compare survival curves (f), significant p values < 0.05 are indicated.

Fig. 5. Antibody (cross)-reactivity of distinct YF17D-vectored ebolavirus vaccine candidates.

a Schematic of YF17D-vectored ebolavirus vaccine constructs and representative images of their plaque phenotypes on BHK-21J cells, fixed 6 days post-infection. b Heatmap representing the different ebolavirus GPs (EBOV GP, SUDV GP, TAFV GP, BDBV GP and RESTV GP) log₁₀-transformed mean binding antibody titers present in serum pools of Ifnar^-/- mice that were hyperimmunized with either YF-EBO, YF-SUD, YF-TAF, YF-BDB or YF17D determined by IIFA (n = 2). Phylogenetic tree based on the amino acid sequences of the different ebolavirus GPs was generated using the Neighbor-Joining method with 1000 bootstrap replications in MEGA11. Small numbers at the nodes and the scale bar indicate bootstrap values and the number of amino acid substitutions per site, respectively.