An encapsulated Yersinia pseudotuberculosis is a highly efficient vaccine against pneumonic plague

Abstract

Background: Plague is still a public health problem in the world and is re-emerging, but no efficient vaccine is available. We previously reported that oral inoculation of a live attenuated Yersinia pseudotuberculosis, the recent ancestor of Yersinia pestis, provided protection against bubonic plague. However, the strain poorly protected against pneumonic plague, the most deadly and contagious form of the disease, and was not genetically defined.

Methodology and principal findings: The sequenced Y. pseudotuberculosis IP32953 has been irreversibly attenuated by deletion of genes encoding three essential virulence factors. An encapsulated Y. pseudotuberculosis was generated by cloning the Y. pestis F1-encoding caf operon and expressing it in the attenuated strain. The new V674pF1 strain produced the F1 capsule in vitro and in vivo. Oral inoculation of V674pF1 allowed the colonization of the gut without lesions to Peyer's patches and the spleen. Vaccination induced both humoral and cellular components of immunity, at the systemic (IgG and Th1 cells) and the mucosal levels (IgA and Th17 cells). A single oral dose conferred 100% protection against a lethal pneumonic plague challenge (33×LD(50) of the fully virulent Y. pestis CO92 strain) and 94% against a high challenge dose (3,300×LD(50)). Both F1 and other Yersinia antigens were recognized and V674pF1 efficiently protected against a F1-negative Y. pestis.

Conclusions and significance: The encapsulated Y. pseudotuberculosis V674pF1 is an efficient live oral vaccine against pneumonic plague, and could be developed for mass vaccination in tropical endemic areas to control pneumonic plague transmission and mortality.

Figure 1. Strain V674pF1 produces an F1 capsule.

A: Strains (a) V674 as negative control, (b) CO92 as positive control, and (c–d) V674pF1 in India ink observed under phase-contrast microscopy. The capsule excludes the India ink and appears as a clear halo around bacterial cells. B: Comparison of F1 production by V674, V674pF1 and CO92 by ELISA. C: F1 production by V674pF1 in vivo: Peyer's patches were taken from animals infected orally 5 days before with 10¹⁰ cfu of V674pF1, and subjected to an immunohistological staining with an anti-F1 mouse mAb, followed by a hematoxylin counterstain. An example of a bacterial focus (brown-red color) is indicated by an arrow.

Figure 2. Gut colonization by Y. pseudotuberculosis mutants.

Bacteria present in the feces (2 pellets) from mice orally vaccinated with strains V674pF1 and V674 at the indicated doses (cfu) were counted eleven and twenty days after. Shown are individual values from 4 to 8 mice per condition. ns: not significant.

Figure 3. Dissemination and tissue lesions caused by Y. pseudotuberculosis mutants.

A: bacteria present in the Peyer's patches and the spleen (cfu/organ) of mice infected orally with strains IP32953 (10⁸ cfu), V674 (10⁸ cfu) or V674pF1 (10⁸ and 10⁹ cfu). Shown are results from individual mice. The median is indicated by a horizontal line. B: quantification of lesions in Peyer's patches and spleen of mice infected orally with strains IP32953 (10⁸ cfu), V674 (10⁸ cfu) or V674pF1 (10⁹ cfu). The limit of detection was 10 cfu per organ. Organs taken at the indicated time were analyzed by histology after hematoxylin-eosin staining. Tissue lesion scores were recorded and shown are means ± s.e.m of four mice per condition. *: p≤0.05, ns: not significant, †: dead mice. C: Examples of lesions observed at Day 5 in the spleen and Peyer's patches from mice inoculated orally with IP32953 or V674pF1 (10⁸ cfu). Arrowheads indicate abscesses, and an N indicate an area of necrosis. In IP32953-infected spleen, hemorrhage is widespread, whereas in V674pF1 infected spleen only subcapsular microhemorrhages (black arrows) and congestioned blood vessels (white arrows) are observed. L: insestinal lumen.

Figure 4. Seric and mucosal humoral immune response of vaccinated mice.

To determine serum IgG titers against Y. pseudotuberculosis antigens (A) or against purified F1 antigen (B), blood was taken on day 21 from mice vaccinated orally with either V674 (10⁸ cfu; 16 mice) or V674pF1 (32 mice received 10⁸ cfu, 8 mice received 10⁹ cfu), or from naive animals (8 mice). To determine anti-Y. pseudotuberculosis (C) or anti-F1 (D) IgA levels in serum, intestine and lungs, groups of 8 mice received 10⁸ cfu or 10⁹ cfu of strain V674pF1 orally, or were not vaccinated, and were sacrificed 3 weeks later to collect blood, intestinal lavage (5 ml) and bronchoalveolar lavage (2 ml). Each dot represents an individual animal and medians (-) are shown. *: p<0.05, **: p<0.005, ***: p<0.001, ns: not significant.

Figure 5. Cellular immune response of vaccinated mice.

Splenocytes isolated from mice vaccinated orally 21 days before with strains V674pF1 (10⁸ cfu) or V674 (10⁸ cfu) or unvaccinated mice (naive) were stimulated in vitro with 5 µg/ml of either a Y. pseudotuberculosis antigenic preparation (Y. ptb Ag.) or purified F1 antigen. The mitogen Concanavalin A (ConA: 1 µg/ml) served as positive control. Supernatants taken 3 days after stimulation were tested by ELISA for the presence of IFNγ (A) and IL-17 (B). Shown are the mean ± s.e.m. of 16 mice per condition (2 pooled experiments). *: p<0.05, **: p<0.005, ***: p<0.001, ns: not significant.

Figure 6. Protection of vaccinated mice against pneumonic plague.

Mice having received a single oral vaccination with strains V674pF1 or V674 at the indicated doses were challenged 4 weeks later by intranasal instillation of the indicated dose of Y. pestis CO92 (A, B) or CO92Δcaf (C). Mouse survival was recorded daily for 21 days. Results from repeated experiments with 7–8 animals per group were pooled and the total number of mice per condition is indicated inside the corresponding bar. *: p≤0.05, **: p≤0.005, ***: p<0.001. ns: not significant.