Murine Mycobacterium marinum infection as a model for tuberculosis

Mycobacteria are a major human health problem globally. Regarding tuberculosis the situation is worsened by the poor efficacy of current vaccine regimens and by emergence of drug-resistant strains (Manjelievskaia J et al, Trans R Soc Trop Med Hyg 110: 110, 2016; Pereira et al., Lancet Infect Dis 12:300–306, 2012; http://www.who.int/tb/publications/global_report/en/) undermining both disease-prevention and available treatments. Thus, increased basic understanding of mycobacterial—and particularly Mycobacterium tuberculosis —virulence strategies and pathogenesis is of great importance. To this end several in vivo infection models are available (Guirado and Schlesinger, Front Immunol 4:98, 2013; Leung et al., Eur J Immunol 43:2246–2254, 2013; Patel et al., J Lab Physicians 3:75–79, 2011; van Leeuwen et al., Cold Spring Harb Perspect Med 5:a018580, 2015). While these models all have their merits they also exhibit limitations, and none perfectly mimics all aspects of human tuberculosis. Thus, there is a need for multiple models that may complement each other, ultimately allowing us to gain true insight into the pathogenesis of mycobacterial infections. Here, we describe a recently developed mouse model of Mycobacterium marinum infection that allows kinetic and quantitative studies of disease progression in live animals [8]. Notably, this model exhibits features of human tuberculosis not replicated in M. tuberculosis infected mice, and may provide an important complement to the field. For example, granulomas in the M. marinum model develop central caseating necrosis (Carlsson et al., PLoS Pathog 6:e1000895, 2010), a hallmark of granulomas in human tuberculosis normally not replicated in murine M. tuberculosis infection. Moreover, while tuberculosis is heterogeneous and presents with a continuum of active and latent disease, M. tuberculosis infected mice essentially lack this dynamic range and do not replicate latency (Guirado and Schlesinger, Front Immunol 4:98, 2013; Patel et al., J Lab Physicians 3(2):75–79, 2011). In contrast, M. marinum infected mice may naturally develop latency, as suggested by reduced inflammation and healing of the diseased tissue while low numbers of bacteria persist in granulomatous lesions (Carlsson et al., PLoS Pathog 6:e1000895, 2010). Thus, infection with M. marinum may offer a unique murine model for studying granuloma formation as well as latency— and possibly also for studies of disease-reactivation. In addition to the in vivo model, we describe infection of bone marrow-derived murine macrophages, an in vitro platform enabling detailed mechanistic studies of host-pathogen interactions occurring in the principal host target cell for pathogenic mycobacteria.