Development of a vaccinia virus based reservoir-targeted vaccine against Yersinia pestis

Abstract

Yersinia pestis, the causative organism of plague, is a zoonotic organism with a worldwide distribution. Although the last plague epidemic occurred in early 1900s, human cases continue to occur due to contact with infected wild animals. In this study, we have developed a reservoir-targeted vaccine against Y. pestis, to interrupt transmission of disease in wild animals as a potential strategy for decreasing human disease. A vaccinia virus delivery system was used to express the F1 capsular protein and the LcrV type III secretion component of Y. pestis as a fusion protein. Here we show that a single dose of this vaccine administered orally, generates a dose-dependent antibody response in mice. Antibody titers peak by 3 weeks after administration and remain elevated for a minimum of 45 weeks. Vaccination provided up to 100% protection against challenge with Y. pestis administered by intranasal challenge at 10 times the lethal dose with protection lasting a minimum of 45 weeks. An orally available, vaccinia virus expressed vaccine against Y. pestis may be a suitable vaccine for a reservoir targeted strategy for the prevention of enzootic plague.

Figure 1

Strategy to construct VV-F1-V. F1 (white arrow) and LcrV (black arrow) were amplified and cloned into pCR2.1 at specific restriction sites (marked by dotted gray arrows). A tPA signal sequence (black rectangle) was added to the 5′-end of F1 fragment. The tPA-F1 and LcrV fragments were fused by overlap PCR and ligated into vaccinia virus cloning vector, pRB21. Plasmid pRB21 has a vaccinia virus promoter (black circle), vp37 gene (gray arrow) for selection of recombinants, and VV homologous sites (white box) for recombination with VV-vRB12. pRB21 carrying tPA-F1-V was transfected into VV-vRB12 infected CV-1 cells. Double homologous cross over results in formation of recombinant virus, VV-tPA-F1-V (or VV-F1-V). The recombinant virus gains a copy of vp37 gene during crossing-over event which is essential for plague formation, the feature used for recombinant selection. The purity of recombinants was tested by amplifying gpt sequence (dotted white arrow) (PCR data not shown), which is present only in VV-vRB12 and is lost during recombinant formation.

Figure 2

Expression of F1-V fusion protein by recombinant virus in vitro. Confluent Hela cells were infected with VV-F1-V. Cell lysates were subjected to SDS-PAGE and probed with an antibody to LcrV to test for expression of F1-V fusion protein. Lane 1: Positive control, Yersinia lysate (expressing only LcrV); Lanes 2, 3 and 4 were loaded with VV-F1-V infected cell lysate in increasing 5-fold increments; Lane 5: Negative control, cell lysate infected by a non-specific strain of VV (VV-CD56) which lacks F1-V insert. The predicted sizes of LcrV and F1-V are 37 kDa and 53 kDa respectively.

Figure 3

Dose-dependent antibody response in VV-F1-V vaccinated mice. C57BL/6 mice were administered either 10⁷ (triangle) or 10⁸ (circle) pfu of VV-F1-V by oral gavage. Antibody response in serum obtained at weeks 3 and 10 post-vaccination are shown. Each data point represents antibody titer for one mouse. Black bars represent the geometric mean of antibody response for each group. The difference in antibody titers between the two dosage groups was significant (p<0.05) by Mann-Whitney U test.

Figure 4

Kinetics of antibody response against VV-F1-V. C57BL/6 mice were vaccinated with 10⁸ pfu of VV-F1-V. Serum antibody response against LcrV was measured at various time points by endpoint dilutional titering. Each square represents the mean antibody titer for at least 3 mice. Error bars represent standard deviation.

Figure 5

Determination of LD₅₀ of Y. pestis KIM D27 in C57BL/6 mice by intranasal route of inoculation. Three groups of mice (n=12 in each group) were given 2X10³, 2X10⁴, or 2X10⁵ cfu of Y. pestis KIM D27 by intranasal inoculation. Survival rate of the groups of mice was determined over a period of 14 days. The method of Reed and Münch was used to calculate the LD₅₀ dose [47].

Figure 6

Protective efficacy of VV-F1-V vaccine in mice. C57BL/6 mice were vaccinated with differing amounts of VV-F1-V to generate a diversity of antibody titers. VV-F1-V or control vaccinated mice were administered with a dose of 10XLD₅₀ of Y. pestis KIM D27 by intranasal inoculation at 5 weeks post-vaccination. Three independent experiments were performed (n=3). The survival rates of mice with differing antibody titers from the three experiments are shown. Log-rank test was performed to determine the significance by comparing survival percentage of each titer group to the control vaccinated mice. ‘N’ represents number of mice in each titer group, and asterisks indicate statistical significance obtained by log rank test. The ‘p’ values obtained by log rank test are as follows: For titers>100000 vs. control, p<0.0001; for titers between 32000–64000 vs. control, p=0.02; for titers between 8000–16000 vs. control, p=0.01; for titers less than ≤ 4000 vs. control the difference was not significant (n.s).