Stimulation of protective immune responses against intracellular pathogens is difficult to achieve using non-replicating vaccines. BALB/c mice immunized by intramuscular injection with killed Francisella tularensis (live vaccine strain) adjuvanted with preformed immune stimulating complexes admixed with CpG, were protected when systemically challenged with a highly virulent strain of F. tularensis (Schu S4). Serum from immunized mice was used to probe a whole proteome microarray in order to identify immunodominant antigens. Eleven out of the top 12 immunodominant antigens have been previously described as immunoreactive in F. tularensis. However, 31 previously unreported immunoreactive antigens were revealed using this approach. Twenty four (50%) of the ORFs on the immunodominant hit list belonged to the category of surface or membrane associated proteins compared to only 22% of the entire proteome. There were eight hypothetical protein hits and eight hits from proteins associated with different aspects of metabolism. The chip also allowed us to readily determine the IgG subclass bias, towards individual or multiple antigens, in protected and unprotected animals. These data give insight into the protective immune response and have potentially important implications for the rational design of non-living vaccines for tularemia and other intracellular pathogens.

The pathogenesis of infection with Yersinia pestis, the causative agent of plague, was examined following subcutaneous infection of BALB/c mice with a fully virulent strain expressing green fluorescent protein. Plate culturing, flow cytometry, and laser confocal microscopy of spleen homogenates throughout infection revealed three discernible stages of infection. The early phase was characterized by the presence of a small number of intracellular bacteria mostly within CD11b؉ macrophages and Ly-6G ؉ neutrophils. These bacteria were not viable, as determined by plate culturing of spleen homogenates, until day 2 postinfection. Between days 2 and 4 postinfection, a plateau phase was observed, with bacterial burdens of 10 3 to 10 4 CFU per spleen. Flow cytometric analysis revealed that there was even distribution of Y. pestis within both CD11b؉ macrophage and Ly-6G ؉ neutrophil populations on day 2 postinfection. However, from day 3 postinfection onward, intracellular bacteria were observed exclusively within splenic CD11b ؉ macrophages. The late phase of infection, between days 4 and 5 postinfection, was characterized by a rapid increase in bacterial numbers, as well as escape of bacteria into the extracellular compartment. Annexin V staining of spleens indicated that a large proportion of splenic neutrophils underwent rapid apoptosis on days 1 and 2 postinfection. Fewer macrophages underwent apoptosis during the same period. Our data suggest that during the early stages of Y. pestis infection, splenic neutrophils are responsible for limiting the growth of Y. pestis and that splenic macrophages provide safe intracellular shelters within which Y. pestis is able to grow and escape during the later stages of infection. This macrophage compliance can be overcome in vitro by stimulation with a combination of gamma interferon and tumor necrosis factor alpha.

Supplementary data are available at Bioinformatics online.