Francisella tularensis Vaccines Elicit Concurrent Protective T- and B-Cell Immune Responses in BALB/cByJ Mice

Abstract

In the last decade several new vaccines against Francisella tularensis, which causes tularemia, have been characterized in animal models. Whereas many of these vaccine candidates showed promise, it remains critical to bridge the preclinical studies to human subjects, ideally by taking advantage of correlates of protection. By combining in vitro intramacrophage LVS replication with gene expression data through multivariate analysis, we previously identified and quantified correlative T cell immune responses that discriminate vaccines of different efficacy. Further, using C57BL/6J mice, we demonstrated that the relative levels of gene expression vary according to vaccination route and between cell types from different organs. Here, we extended our studies to the analysis of T cell functions of BALB/cByJ mice to evaluate whether our approach to identify correlates of protection also applies to a Th2 dominant mouse strain. BALB/cByJ mice had higher survival rates than C57BL/6J mice when they were immunized with suboptimal vaccines and challenged. However, splenocytes derived from differentially vaccinated BALB/cByJ mice controlled LVS intramacrophage replication in vitro in a pattern that reflected the hierarchy of protection observed in C57BL/6J mice. In addition, gene expression of selected potential correlates revealed similar patterns in splenocytes of BALB/cByJ and C57BL/6J mice. The different survival patterns were related to B cell functions, not necessarily to specific antibody production, which played an important protective role in BALB/cByJ mice when vaccinated with suboptimal vaccines. Our studies therefore demonstrate the range of mechanisms that operate in the most common mouse strains used for characterization of vaccines against F. tularensis, and illustrate the complexity necessary to define a comprehensive set of correlates.

Fig 1. Splenocytes from LVS-related vaccinated mice exhibit a hierarchy of control of intramacrophage LVS growth.

BMMΦ from BALB/cByJ mice were infected with LVS (Macs), and co-cultured with splenocytes obtained from naïve or vaccinated BALB/cByJ mice (Panel A), and from naive or vaccinated BKO mice (Panel B), as indicated. After two days of co-culture, BMMΦ were washed, lysed, and plated to evaluate the recovery of intracellular bacteria. Values shown are the mean numbers of CFU/ml ± SD of viable bacteria for triplicate samples. Results shown are from one representative experiment of seven (using splenocytes of BALB/cByJ mice) or four (using splenocytes of BKO mice) independent experiments of similar design and outcome. Brackets indicate a significant difference (P < 0.05) between the recoveries of bacteria in co-cultures. There were no significant differences between the recovery of bacteria from co-cultures using LVS-immune cells and LVS-G-immune cells (Panel A) and the recovery of bacteria from co-cultures using LVS-G-immune cells and LVS-R-immune cells (Panel B).

Fig 2. IFN-γ and NO production exhibit patterns mostly similar to that of in vitro LVS replication.

Supernatants from co-cultures described in Fig 1 using splenocytes of BALB/cByJ mice (Panels A and C) and BKO mice (Panels B and D) were collected after two days of co-culture, and separated from cells for analyses of IFN-γ by ELISA (Panels A and B) and NO by Griess reaction (Panels C and D). Concentrations were calculated using standard curves as reference. Values shown are the mean concentration in ng/ml (IFN-γ) or μmoles (NO) ± standard deviation of triplicate samples. Results shown are from one representative experiment of seven (using splenocytes of BALB/cByJ mice) or four (using splenocytes of BKO mice) independent experiments of similar design and outcome. Brackets indicate a significant difference (P < 0.05) between amounts of IFN-γ or NO produced in co-cultures. There were no significant differences in IFN-γ production between the co-cultures using LVS-immune cells and the co-cultures using LVS-G-immune cells (Panel A), between the co-cultures using LVS-G-immune cells and the co-cultures using LVS-R-immune cells (Panel A), and between the co-cultures using LVS-immune cells and the co-cultures using LVS-G-immune cells or LVS-R-immune cells (Panel B).

Fig 3. Humoral immune responses patterns to LVS-related vaccines differ between BALB/cByJ and C57BL/6J mice.

Pooled sera from five mice for each vaccine group were obtained from BALB/cByJ mice (Panel A) and from C57BL/6J mice (Panel B) six weeks after vaccination, and analyzed for anti-LVS total IgG. Error bars depict standard deviation of the mean of triplicates samples tested in the ELISA. Results shown are from one representative experiment of four independent experiments of similar design and outcome.

Fig 4. Protein staining and reactivity with serum derived from vaccinated mice revealed differences among vaccine extracts.

Twenty μg of protein extracts, prepared from whole LVS, HK-LVS, LVS-G, and LVS-R, were loaded on SDS-PAGE in reducing conditions and stained with Red Ponceau (Panel A). Following Ponceau staining, reactivity of protein extracts was analyzed by blotting the membranes with pooled sera derived from LVS vaccinated (Panel B) and from HK-LVS vaccinated BALB/cByJ mice (Panel C). Results shown are from one representative experiment of four independent experiments of similar design and outcome.

Fig 5. Anti-LVS antibody titers of HK-LVS vaccinated BALB/cByJ mice do not correlate with survival.

Sera from BALB/cByJ mice vaccinated with different doses of HK-LVS were individually analyzed six weeks after vaccination for anti-LVS total IgG. Mice were then challenged IP with a maximal lethal dose of LVS. Error bars depict standard deviation of the mean of triplicates samples tested in the ELISA. * indicates the antibody responses of the mice that eventually survived the lethal challenge. Results shown are from one representative experiment of two independent experiments of similar design and outcome.