Abstract
Specific antituberculous resistance appears to be induced following inoculation of live but not dead BCG. This dependence on BCG viability may explain the diverse responses in terms of protective immunity in clinical trials of BCG conducted worldwide over the last thirty years. In order to dissect out simple parameters which may differentiate a protective from a non-protective response, we have developed a murine in vivo model of experimental BCG infection to study the immune response in draining lymph nodes following footpad inoculation with either live or killed BCG preparations. In this model, live but not heat-killed BCG efficiently migrate to the draining lymph nodes and stimulate the early accumulation of mononuclear cells. In addition, live and heat-killed BCG stimulate different responses in terms of the level of expression of interferon-gamma, inducible nitric oxide (iNOS), as well as macrophage and dendritic cell markers in the draining lymph nodes. This divergent in vivo response was reproduced in vitro when pure macrophage cultures were infected with BCG and responded differently to live and dead preparations, producing significant levels of TNF and reactive nitrogen intermediates only when infected with live BCG. Taken together, these observations suggest that the differences encountered in vivo may be related to the ability of live BCG to migrate to local lymph node, where they cause the accumulation of cells expressing protective cytokines and therefore inducing an efficient immune response. These findings may have important implications for the design of new anti-tuberculosis vaccines.