A next generation vaccine against human rabies based on a single dose of a chimpanzee adenovirus vector serotype C

Abstract

Rabies, caused by RNA viruses in the Genus Lyssavirus, is the most fatal of all infectious diseases. This neglected zoonosis remains a major public health problem in developing countries, causing the death of an estimated 25,000-159,000 people each year, with more than half of them in children. The high incidence of human rabies in spite of effective vaccines is mainly linked to the lack of compliance with the complicated administration schedule, inadequacies of the community public health system for local administration by the parenteral route and the overall costs of the vaccine. The goal of our work was the development of a simple, affordable and effective vaccine strategy to prevent human rabies virus infection. This next generation vaccine is based on a replication-defective chimpanzee adenovirus vector belonging to group C, ChAd155-RG, which encodes the rabies glycoprotein (G). We demonstrate here that a single dose of this vaccine induces protective efficacy in a murine model of rabies challenge and elicits strong and durable neutralizing antibody responses in vaccinated non-human primates. Importantly, we demonstrate that one dose of a commercial rabies vaccine effectively boosts the neutralizing antibody responses induced by ChAd155-RG in vaccinated monkeys, showing the compatibility of the novel vectored vaccine with the current post-exposure prophylaxis in the event of rabies virus exposure. Finally, we demonstrate that antibodies induced by ChAd155-RG can also neutralize European bat lyssaviruses 1 and 2 (EBLV-1 and EBLV-2) found in bat reservoirs.

Fig 1. Rabies antigens expression in ChAd155 vectors.

A) Whole cell FACS analysis of MRC5 cells not infected (left panel) and infected with 50 MOI (vp/cell) of ChAd155-RG (right panel). 48h after infection cells were stained with anti-Rabies Glycoprotein antibody and FITC-conjugated anti-mouse secondary antibody. B) Western blot analysis of G protein on lysates of A549 cells infected with 250 MOI of ChAd155-RG and 250 and 1250 MOI of ChAd155-RNG and ChAd155-RGN. 48h after infection cells were harvested and 20 μg and 80 μg of total cell lysates were used for WB analysis. GAPDH was used as a loading control. C) Effect of WPRE on G protein expression. Western blot analysis of G protein on total cell lysates. HeLa cells were infected with 250 MOI of ChAd155-RG +/- WPRE, 48h after infection cells were harvested and 20 μg and 80 μg of total cell lysates were analyzed by WB. GAPDH was used as a loading control.

Fig 2. Immunological potency of ChAd155 and ChAd83 vectors in mice.

A) Groups of CD1 mice (n = 8) were immunized with 1x10⁸ vp of ChAd155-RG, ChAd155-RGN and ChAd155-RNG, encoding different rabies antigen designs. As a control, one group of animals received 1/10th of the human dose of RABIPUR. Sera were collected 2 and 4 weeks after vaccination. Rabies VNA were measured by FAVN assay in sera collected at w4, and expressed as International Units (IU)/ml. B) Anti-G protein IgG measured by a commercially available ELISA kit (Platelia) at week 2 and 4 after immunization C) Groups of BALB/c mice (n = 8) were immunized with either 1x10⁵ or 1x10⁶ vp of ChAd155-RG, or ChAd83-RG. At 4 weeks after vaccination animals were sacrificed and T cell response measured in spleen of individual mice by IFNγ ELISpot assay. Data are expressed as number of IFNγ spot forming cells (SFC)/10⁶ splenocytes. D) BALB/c mice received 1x10⁷ vp of either ChAd155-RG, or ChAd83-RG, and were bled 2, 4 and 8 weeks after. Rabies VNA were measured by FAVN assay, and expressed as IU/ml. In all panels, the red lines represent group geometric mean.

Fig 3. Protective efficacy and VNA magnitude and kinetics upon single vaccination with escalating doses of ChAd155-RG compared to three injections of RABIPUR.

A) Mice (n = 10/group) were bled on day 0, 14, 28, 56 and 67. Serum from each animal was tested for detection of rabies VNA by RFFIT, and expressed as IU/ml. Data are shown as group geometric mean titer with 95% CI. Black and blue arrows indicate ChAd155 or RABIPUR (1/10th of the human dose, HD) vaccinations, while grey arrow indicates rabies challenge. B) VNA titers at day 56, three days before challenge, are shown for individual animals. Black lines correspond to geometric mean. For each group, the percentage of animals surviving 30 days post challenge is indicated at the bottom of the graph. C) Kaplan-Meier survival curves; log-rank Mantel-Cox test of the overall study: Chi square = 16.65, P = 0.0052; for mock vs RABIPUR, 10⁸ or 10⁷ vp comparisons: Chi square = 4.76, P = 0.029.

Fig 4. Kinetics of VNA induction after a single dose in rabbits.

Groups of five rabbits each received a single intramuscular immunization with 5x10¹⁰vp of ChAd155-RG or 500μl RABIPUR (half human dose). Rabbits were bled on time of vaccination (week 0) and on weeks 1, 3, 8 and 12 post vaccination. Serum from each animal was tested for detection of rabies VNA by FAVN, and titers expressed as IU/ml. The horizontal lines represent geometric mean titers. A two-way analysis of variance was applied having as sources of variation “treatment” (the two vaccines), “time” (subsequent weeks of observation) and as dependent variable “antibody titer”. Statistically significant effect was found for both sources of variation and their interaction (treatment p = 0.0025, F (1, 8) = 18,78; time p = 0.011, F (2,043, 16,34) = 5,969; time x treatment p = 0.0118, F (4, 32) = 3,832).

Fig 5. Rabies vaccine immunogenicity study in NHPs.

Groups of 5 Cynomolgus monkeys were vaccinated intramuscularly as highlighted by arrows on x axis with 5x10¹⁰vp of ChAd155-RG vector (purple circles and red diamonds) and half human dose (500μl) RABIPUR (blue triangle). A: rabies VNA titer kinetics: Rabies VNA titers were measured in serum with a FAVN assay and followed starting at 2 weeks post vaccination and then monthly. Group median +/- IQR are shown. B: magnitude and kinetics of rabies-specific T cell response measured in PBMC at multiple time points by IFNγ ELISpot. Data are expressed as IFNγ SFC per million PBMC. Individual data points represent total rabies G response in each animal, obtained by summing reactivity to each of the 2 peptide pools covering rabies G protein and subtracting 2 times the DMSO background. Black lines represent geometric mean.

Fig 6. ChAd155-RG biodistribution in rats.

ChAd155 genomes detected by qPCR in DNA extracted from indicated organs 2, 8, 29 and 49 days after intramuscular injection. Data are expressed as genome copies per mg of DNA. Limit of detection (LOD) and limit of quantification (LOQ), established within assay validation, were set at 12,5 and 100 copies/well respectively. For graphical representation, LOD and LOQ were adjusted to 1μg of tested DNA (31 and 250 copies respectively) and shown as dotted and dashed lines. Experimental samples resulting below LOD or LOQ were assigned values corresponding to one half of the extrapolated LOD and LOQ/mg of DNA (15 and 125 copies respectively).

Fig 7. Neutralization of pseudoviruses harboring representative classical and non-classical phylogroup I lyssavirus glycoproteins.

Immune sera obtained 5 weeks following a prime immunization of NHPs with either RABIPUR or ChAd155-RG were tested for their ability to neutralize pseudoviruses harboring either CVS-11 (A), EBLV-1 (B) or EBLV-2 (C) glycoproteins. Data represent the mean and standard deviation of combined normalized values for two replicate experiments, with each replicate experiment containing three technical replicates, and where 100% infectivity was set at the mean infectivity (measured in relative light units) observed in the absence of serum. Following non-linear regression analysis, an extra sum-of-squares analysis was performed to compare slopes and IC50 values. IC50 dilution were significantly higher for the ChAd155-RG serum than the RABIPUR serum, as indicated in each panel. RABIPUR (round symbols), ChAd155-RG (square symbols).

Fig 8. Neutralization of pseudoviruses harboring phylogroup II and III lyssavirus glycoproteins.

Immune sera obtained 5 weeks following immunization of NHPs with either RABIPUR or ChAd155-RG, and at week 50 following a ChAd155-RG boost were tested for their ability to neutralize pseudoviruses harboring either MOKV (A), IKOV (B), WCBV (C) or control classical CSV-11 (D) glycoproteins. Data represent the mean and standard deviation of combined normalized values for two replicate experiments, with each replicate experiment containing three technical replicates, and where 100% infectivity was set at the mean infectivity (measured in relative light units) observed in the absence of serum.