Although the definition of latency under a clinical point of view seems clear, the bacterial biology behind that clinical situation remains poorly understood. While dormant, the tubercle bacilli are considered to be under non-replicating (NR) stage. In such a condition, bacilli are not only difficult to be detected but also refractory to the standard treatments avoiding their clearance from the infected tissues. The proposal has been built with the intend of providing tools to understand the bacterial mechanisms that leads to metabolic stage of M. tuberculosis during dormancy as the basis of sorting out the detection and treatment of latent infection. Several models of analysis have been developed trying to characterize dormant tubercle bacilli. Those models are ranging from in vitro conditions, such as hypoxia or starvation, to in vivo analysis, such as the animal model. We propose not only to study a complete range of those previously tested conditions, but also checking some other putative newly described, such as the recently described adipocytes ex vivo model, the new developments of the classical model of hypoxia, and the use of guinea-pigs as more adequate animal model of latent infection. Moreover, a set of drug combinations will be applied to determine their capability of clearance of the bacterial load. Due to its central role in the bacterial metabolism and growth we will analyze the non-replicating stage of the M. tuberculosis bacilli by determining the pre-rRNA synthesis. Finally, the cellular and tissue distribution of dormant bacilli will be also tested during in vivo latent infection both in the animal model and in clinical human samples. Research on the metabolic conditions of the dormant bacilli inside hosts, will provide important and invaluable insight into the biology of M. tuberculosis. That knowledge may lead to the development of novel strategies targeted at the control of the latent infection.