Recombinant rabies virus particles presenting botulinum neurotoxin antigens elicit a protective humoral response in vivo

Abstract

Botulinum neurotoxins are one of the most potent toxins found in nature, with broad medical applications from cosmetics to the treatment of various neuropathies. Additionally, these toxins are classified as Category A-Tier 1 agents, with human lethal doses calculated at as little as 90 ng depending upon the route of administration. Of the eight distinct botulinum neurotoxin serotypes, the most common causes of human illness are from serotypes /A, /B, and /E. Protection can be achieved by eliciting antibody responses against the receptor-binding domain of the neurotoxin. Our previous research has shown that recombinant rabies virus-based particles can effectively present heterologous antigens. Here, we describe a novel strategy using recombinant rabies virus particles that elicits a durable humoral immune response against the botulinum neurotoxin receptor binding domains from serotypes /A, /B, and /E. Following intramuscular administration of β-propiolactone-inactivated rabies virus particles, mice elicited specific immune responses against the cognate antigen. Administration of a combination of these vectors also demonstrated antibody responses against all three serotypes based on enzyme-linked immunosorbent assay (ELISA) measurements, with minimal decay within the study timeline. Complete protection was achieved against toxin challenge from the serotypes /A and /B and partial protection for /E, indicating that a multivalent approach is feasible.

Figure 1

Assembly of recombinant rabies virus (RABV) expressing botulinum neurotoxin type A, B, or E heavy chain carboxyterminal 50 kDa (H_C50) fragment fused to RABV G ectodomain through cytoplasmic tail. (a) The H_C50 fragment from serotype A was flanked by the RABV G signal sequence (SS) and variable lengths of the membrane proximal extracellular region (MPER, dark grey), transmembrane (TM) and cytoplasmic tail (CT) and restriction sites for insertion into BsiWI and NheI sites in the RABV genome. (b) First-generation H_C50/A expression vector with R333 mutation. (c) Second-generation H_C50 expression vector with R333 mutation.

Figure 2

Recombinant H_C50/RABV-G expression analyses on BSR cells. BSR cells were infected at an multiplicity of infection of 10 for 48 hours with the viruses indicated and probed with (a) anti-H_C50/A1 (NR-9353), (b) anti-H_C50/B1 (NR-9354), or (c) anti-BoNT/E1 (NR-17613). Duplicate blots were also probed with anti-HA-tag monoclonal antibody to detect recombinant H_C50/RABV-G proteins; as infection and loading controls, blots were also probed with anti-RABV-M and anti-β-actin.

Figure 3

FACS analysis of BSR cells infected with noncodon-optimized or codon-optimized H_C50 domains. BSR cells were infected at an multiplicity of infection of 5 for 48 hours with BNSP-333 (control, grey), BNSP-333 expressing noncodon-optimized H_C50 protein (unfilled histograms) or codon-optimized H_C50 (shaded histograms) and analyzed for surface staining (a–c) or total protein (d–f) using an anti-HA specific antibody. (a and d) Experimental samples were infected with BNSP-333-H_C50/A (red histograms) or BNSP-333-coH_C50/A (red, shaded histograms). (b and e) Experimental samples were infected with BNSP-333-H_C50/B (green histograms) or BNSP-333-coH_C50/B (green, shaded histograms). (c and f) Experimental samples were infected with BNSP-333-H_C50/E (blue histograms) or BNSP-333-coH_C50/E (blue, shaded histograms).

Figure 4

Multistep growth kinetics of recombinant viruses on Vero cells. Vero cells were infected at a multiplicity of infection of 0.01 for 2 hours with BNSP-333 (black), BNSP-333-coH_C50/A (red), BNSP-333-coH_C50/B (green), or BNSP-333-H_C50/E (blue). Samples were collected at the time points indicated and viral titers were determined in duplicate. FFU, fluorescent focus units.

Figure 5

Analysis of H_C50 incorporation into sucrose-purified particle preparations. (a) Sucrose-purified RABV particles (4 µg/lane) were fractioned on 10% sodium dodecyl sulfate–polyacrylamide gel electrophoresis and stained with SYPRO ruby protein stain (RABV proteins; L~242kDa, G~65kDa, N-57kDa, P~38-41kDa, M-25kDa). (b) 1 µg of sucrose-purified particles were transferred to polyvinylidene fluoride (PVDF) membranes and probed with anti-HA tag and anti-RABV M-specific antibodies.

Figure 6

Antibody responses against vector antigens. Sera from vaccinated mice were collected at the time points indicated (a) and antibody responses of pooled sera was tittered against (b) RABV-G, (c) H_C50/A, (d) H_C50/B, or (e) H_C50/E antigens. Grey bars correspond to the vector control group receiving BNSP-333; red bars correspond to the group receiving BNSP-333-coH_C50/A; green bars correspond to the BNSP-333-coH_C50/B group; blue bars correspond to the BNSP-333-H_C50/E group; black bars correspond to the group receiving the combined particles (BNSP-333-coH_C50/A, −coH_C50/B, and −H_C50/E). Titers are expressed as EC50 values of the reciprocal dilution. The calculated EC50 values represent the effective concentration 50 of the reciprocal dilution (titer) of the sera against each of the groups’ respective antigen(s).

Figure 7

IgG isotype analysis of antibody responses indicates a Th1-biased response. Sera from immunized mice at weeks 4 and 6 were pooled and tested for their IgG1 and IgG2a responses to (a) RABV-G, (b) H_C50/A, (c) H_C50/B, and (d) H_C50/E, and presented as the ratio of the optical densities corresponding to the IgG2a to IgG1-specific enzyme-linked immunosorbent assay signals for the specific antigen (each panel represents one antigen). Only the groups with immune responses to the specific antigens, as grouped by panel, were assayed.

Figure 8

Survival of mice following intraperitoneal challenge with BoNT. Groups of five mice were immunized as outlined in Figure 6a, then challenged via intraperitoneal administration of 1000 mouse LD50 (a) BoNT/A, (b) BoNT/B, or (c) BoNT/E. Combined vaccine groups (ABE) were challenged sequentially with BoNT/A, /B, then /E.