Intranasal delivery of a protein subunit vaccine using a Tobacco Mosaic Virus platform protects against pneumonic plague

Abstract

Yersinia pestis, one of history's deadliest pathogens, has killed millions over the course of human history. It has attributes that make it an ideal choice to produce mass casualties and is a prime candidate for use as a biological weapon. When aerosolized, Y. pestis causes pneumonic plague, a pneumonia that is 100% lethal if not promptly treated with effective antibiotics. Currently, there is no FDA approved plague vaccine. The current lead vaccine candidate, a parenterally administered protein subunit vaccine comprised of the Y. pestis virulence factors, F1 and LcrV, demonstrated variable levels of protection in primate pneumonic plague models. As the most likely mode of exposure in biological attack with Y. pestis is by aerosol, this raises a question of whether this parenteral vaccine will adequately protect humans against pneumonic plague. In the present study we evaluated two distinct mucosal delivery platforms for the intranasal (IN) administration of LcrV and F1 vaccine proteins, a live bacterial vector, Lactobacillus plantarum, and a Tobacco Mosaic Virus (TMV) based delivery platform. IN administration of L. plantarum expressing LcrV, or TMV-conjugated to LcrV and F1 (TMV-LcrV+TMV-F1) resulted in the similar induction of high titers of IgG antibodies and evidence of proinflammatory cytokine secretion. However, only the TMV-conjugate delivery platform protected against subsequent lethal challenge with Y. pestis. TMV-LcrV+TMV-F1 co-vaccinated mice had no discernable morbidity and no mortality, while mice vaccinated with L. plantarum expressing LcrV or rLcrV+rF1 without TMV succumbed to infection or were only partially protected. Thus, TMV is a suitable mucosal delivery platform for an F1-LcrV subunit vaccine that induces complete protection against pneumonic infection with a lethal dose of Y. pestis in mice.

Keywords: Mucosal vaccination; Pneumonic plague; Tobacco Mosaic Virus; Yersinia pestis.

Figure 1

Oral administration of Lip-LcrV L. plantarum is not protective. a, 50% maximum binding titers were calculated for total LcrV specific Ig levels in mice vaccinated and boosted with Lip-LcrV L. plantarum(n=10), L. plantarum transfected with plasmid lacking LcrV (empty vector) (n=10), or IM with rLcrV in alum(n=5). b, survival of mice orally vaccinated and boosted with Lip-LcrV L. plantarum, empty vector or vaccinated and boosted IM with rLcrV in alum after challenge with 10× LD50 or 100× LD50 of Y. pestis CO92pgm−. These data are representative of two experiments. *-p<0.01 rLcrV IM 10 or 100 LD50 vs all other groups.

Figure 2

IN administration of Lip-LcrV L. plantarum is not protective. a, LcrV specific IgG1, IgG2a, IgG2b, and IgG3 levels following vaccination and boost with Lip-LcrV L. plantarum (Lip-LcrV) (n=5), empty vector (n=5), rLcrV in alum (rLcrV IM) (n=5), or unvaccinated (negative) (n=5). b, 50% maximal binding titers were calculated for the indicated LcrV specific isotypes in Lip-LcrV and rLcrV IM vaccinated mice. Titers could not be generated empty vector or negative mice. c, LcrV specific IgA in BAL washes of vaccinated and control mice. d, survival of mice after challenge with 10× LD50 of Y. pestis CO92pgm−. #-p<0.01 rLcrV IM vs. Empty Vector, Lip-LcrV, rLcrV IM survival. These data are representative of five experiments.

Figure 3

IN vaccination and boost with TMV-LcrV+TMV-F1 induces higher anti-LcrV and anti-F1 antibody production compared to IM vaccination with rLcrV+rF1. Anti-LcrV and anti-F1 antibody binding curves (left column) and 50% maximal binding titers (right column) for IgG1 (a), IgG2a (b), and IgG2b (c) in mice vaccinated with TMV-LcrV (n=5), TMV-F1(n=8), TMV-LcrV+TMV-F1(n=7), rLcrV IN(n=5), rF1 IN(n=5), rLcrV+rF1 IN(n=5), rLcrV+rF1 IM(n=5), or unvaccinated mice(n=5). Titers could not be calculated for anti-F1 antibodies in TMV-LcrV or rLcrV IN vaccinated mice or unvaccinated mice, and anti-LcrV antibodies in TMV-F1 or rF1 IN vaccinated mice or unvaccinated mice; these groups are not therefore not represented in the 50% binding titer graphs. These data are representative of 5 experiments for TMV-LcrV and 3 experiments for TMV-F1 and TMV-LcrV+TMV-F1.

Figure 4

TMV-LcrV + TMV-F1 completely protects against morbidity and mortality associated with pneumonic infection with 10× LD50 Y. pestis CO92pgm−. Mice were challenged on day 0 and then weighed daily until completion of the experiment. These data are representative of 5 experiments for TMV-LcrV and 3 experiments for TMV-F1 and TMV-LcrV+TMV-F1. *-p<0.05, TMV-LcrV IN vs. rLcrV IN, TMV-F1 IN vs. rF1 IN, TMV-LcrV+TMV-F1 IN vs. rLcrV+rF1 IN, rLcrV+rF1 IM vs. rLcrV+rF1 IN.