Vaccination of BALB/c mice with Escherichia coli-expressed vaccinia virus proteins A27L, B5R, and D8L protects mice from lethal vaccinia virus challenge

Abstract

The potential threat of smallpox use in a bioterrorist attack has heightened the need to develop an effective smallpox vaccine for immunization of the general public. Vaccination with the current smallpox vaccine, Dryvax, produces protective immunity but may result in adverse reactions for some vaccinees. A subunit vaccine composed of protective vaccinia virus proteins should avoid the complications arising from live-virus vaccination and thus provide a safer alternative smallpox vaccine. In this study, we assessed the protective efficacy and immunogenicity of a multisubunit vaccine composed of the A27L and D8L proteins from the intracellular mature virus (IMV) form and the B5R protein from the extracellular enveloped virus (EEV) form of vaccinia virus. BALB/c mice were immunized with Escherichia coli-produced A27L, D8L, and B5R proteins in an adjuvant consisting of monophosphoryl lipid A and trehalose dicorynomycolate or in TiterMax Gold adjuvant. Following immunization, mice were either sacrificed for analysis of immune responses or lethally challenged by intranasal inoculation with vaccinia virus strain Western Reserve. We observed that three immunizations either with A27L, D8L, and B5R or with the A27L and B5R proteins alone induced potent neutralizing antibody responses and provided complete protection against lethal vaccinia virus challenge. Several linear B-cell epitopes within the three proteins were recognized by sera from the immunized mice. In addition, protein-specific cellular responses were detected in spleens of immunized mice by a gamma interferon enzyme-linked immunospot assay using peptides derived from each protein. Our data suggest that a subunit vaccine incorporating bacterially expressed IMV- and EEV-specific proteins can be effective in stimulating anti-vaccinia virus immune responses and providing protection against lethal virus challenge.

FIG. 1.

Reactivities of recombinant-protein immune sera with potential linear B-cell epitopes within the vaccinia virus A27L (A), B5R (B), and D8L (C) proteins. BALB/c mice received a total of three (weeks 0, 3, and 5) subcutaneous vaccinations of 10 μg of GST or a combination of A27L-GST, D8LΔ-GST, and B5RΔ-GST proteins in MPL-TDM (MT) or TiterMax Gold adjuvant. Serum samples were obtained from these mice 5 weeks after the last immunization or from mice immunized with individual proteins twice and sacrificed 2 weeks after the second vaccination. Twenty-amino-acid biotinylated peptides were added to NeutrAvidin-coated plates and used to screen the serum samples by ELISA. Data are presented as the increase in the absorbance (at 450 nm) of each peptide over that of an irrelevant peptide, OVA_323-339. Serum reactivity levels fourfold (indicated by horizontal line) or more higher than that with the OVA peptide were considered positive.

FIG. 2.

Sera from mice immunized with E. coli-expressed vaccinia virus recombinant proteins inhibit EEV spread to distal cells and comet tail formation in vitro. Serum samples were obtained from mice vaccinated with either GST, the two-protein combination of A27L and B5R, or the three-protein combination of A27L, D8L, and B5R in TiterMax Gold adjuvant as described in the legend to Fig. 1. For comparison, mice were vaccinated once with 8 × 10⁶ PFU of VV-COP by dermal scarification, and serum samples were collected 2 or 4 weeks later. Monolayers of BSC-40 cells were infected with ∼100 PFU of VV-IHD-J, aspirated to remove the viral inoculum, and overlaid with liquid medium containing a 1:25 final dilution of the indicated serum. After incubation for an additional 35 h, the cells were fixed and stained with 0.1% crystal violet to visualize plaques and comet tails. Wells containing cells infected with virus but overlaid with methylcellulose were used as controls for complete inhibition of EEV dissemination.

FIG. 3.

T-cell responses induced by vaccination with recombinant proteins (A) or VV-COP (B to D). Mice were immunized with A27L, D8LΔ, and B5RΔ proteins in TiterMax adjuvant or with VV-COP, as described in the legends to Fig. 1 and 2, respectively. Five weeks after the last dose of protein immunization (A) or 2 (B and C) or 4 (D) weeks after VV-COP dermal scarification, spleens were obtained, and unfractionated splenocytes or positively selected CD4⁺ or CD8⁺ T cells were stimulated with 10 μg/ml of the indicated peptide. Mitomycin C-treated naïve splenocytes were used as antigen-presenting cells for the stimulation of CD4⁺ or CD8⁺ T cells. Data in panels A to C represent mean numbers of IFN-γ-positive spots in triplicate wells. Error bars, standard errors of the means. *, P < 0.05; **, P < 0.01; +, P = 0.059; ++, P = 0.055. For ICCS (D), unfractionated splenocytes from mice vaccinated with VV-COP (4 weeks earlier) were stimulated for 13 h with no peptide or with overlap peptides in the presence of brefeldin A. The cells were then surface stained with anti-CD4 or anti-CD8 antibodies and intracellularly stained with anti-IFN-γ antibody. The percentage of CD4⁺ or CD8⁺ T cells responding to the stimulation by producing IFN-γ (given above each plot) was quantitated by flow cytometry.

FIG. 4.

Immunization of mice with A27L, D8LΔ, and B5RΔ proteins provides full protection from lethal vaccinia virus challenge. Mice (five per group) were vaccinated with recombinant proteins administered in MPL-TDM (A to D) or TiterMax (E to H) adjuvant, as described for Fig. 1. One group of five mice was immunized with a control protein, GST, while another group was vaccinated with VV-COP by dermal scarification. At week 10 for GST- or recombinant-protein-immunized mice or at week 2 for VV-COP-vaccinated mice, the mice were challenged intranasally with 20 LD₅₀ of VV-WR (∼2.4 × 10⁶ PFU) in 20 μl of PBS. Survival (A and E), weight loss (B and F), ear temperature (C and G), and level of illness (D and H) were recorded for each mouse and were plotted as group averages ± standard deviations. Symbols: ○, GST; ▵, A27L plus B5R; ▿, A27L, B5R, and D8L; □, VV-COP.

FIG. 5.

Passive transfer of protein immune serum, but not of splenocytes, protects mice from lethal vaccinia virus challenge. Sera or splenocytes prepared from mice vaccinated with recombinant protein or VV-COP, as described for Fig. 1 and 2, respectively, were used. A group of 5 to 10 naïve mice were given 2 × 10⁷ splenocytes intravenously 24 h prior to virus challenge. Mice that received serum were administered 500 μl of serum 3 h before and 24 h after virus challenge. Virus challenge was performed as described for Fig. 4. Survival (A), weight loss (B), ear temperature (C), and level of illness (D) were recorded for each mouse and were plotted as group averages ± standard deviations. Open symbols, serum; solid symbols, splenocytes; circles, GST; inverted triangles, A27L, B5R, and D8L; squares, VV-COP.