ABSTRACT Mouse models have been key to studies of tuberculosis pathogenesis and drug efficacy, but many, such as those employing BALB/c mice, fail to reproduce the full range of heterogenous microniches observed in well-structured human lesions, which feature hypoxic caseous cores of necrotic debris surrounded by infected foamy macrophages. The granuloma presents a variety of environments differing in levels of oxygen, ions, nutrients, and intra versus extracellular residence, which determine the physiological state of the infecting bacillus and its susceptibility to immune or drug control. Recently, alternative mouse strains such as C3HeB/FeJ have allowed the study of infection and treatment in the context of these varied environments but exhibit substantial inconsistency in the development of hypoxic necrotic lesions, both within and between individual mice. Building on the observation that inducible nitric oxide synthase ( Nos2 )-deficient mice consistently develop hypoxic necrotic lesions, we have established two simplified models with infection by the aerosol route. The first uses the slightly attenuated M. tuberculosis R1Rv strain, which produces a progressive infection that is contained at a high stable burden by an adaptive immune response. In the second model, vaccination with the attenuated ∆ RD1 , pantothenate auxotroph mc 2 6230 protects from an otherwise lethal infection with virulent Mycobacterium tuberculosis Erdman. This model reflects most contemporary tuberculosis infections, which take place in the context of a pre-existing immune response from vaccination. Both variations uniformly develop well-structured hypoxic necrotic lesions and respond poorly to chemotherapy. These refined models will be useful in studies of M. tuberculosis infection and treatment.
Makafe et al. (Mon,) studied this question.