Francisella tularensis is a highly virulent intracellular bacterium and cell-mediated immunity is critical for protection, but mechanisms of protection against highly virulent variants, such as the prototypic strain F. tularensis strain SCHU S4, are poorly understood. To this end, we established a co-culture system, based on splenocytes from naïve, or immunized mice and in vitro infected bone marrow-derived macrophages that allowed assessment of mechanisms controlling infection with F. tularensis. We utilized the system to understand why the clpB gene deletion mutant, ΔclpB, of SCHU S4 shows superior efficacy as a vaccine in the mouse model as compared to the existing human vaccine, the live vaccine strain (LVS). Compared to naïve splenocytes, ΔclpB-, or LVS-immune splenocytes conferred very significant control of a SCHU S4 infection and the ΔclpB-immune splenocytes were superior to the LVS-immune splenocytes. Cultures with the ΔclpB-immune splenocytes also contained higher levels of IFN-γ, IL-17, and GM-CSF and nitric oxide, and T cells expressing combinations of IFN-γ, TNF-α, and IL-17, than did cultures with LVS-immune splenocytes. There was strong inverse correlation between bacterial replication and levels of nitrite, an end product of nitric oxide, and essentially no control was observed when BMDM from iNOS−/− mice were infected. Collectively, the co-culture model identified a critical role of nitric oxide for protection against a highly virulent strain of F. tularensis.