Animal models of tuberculosis: zebrafish

Abstract

Over the past decade the zebrafish (Danio rerio) has become an attractive new vertebrate model organism for studying mycobacterial pathogenesis. The combination of medium-throughput screening and real-time in vivo visualization has allowed new ways to dissect host pathogenic interaction in a vertebrate host. Furthermore, genetic screens on the host and bacterial sides have elucidated new mechanisms involved in the initiation of granuloma formation and the importance of a balanced immune response for control of mycobacterial pathogens. This article will highlight the unique features of the zebrafish-Mycobacterium marinum infection model and its added value for tuberculosis research.

Figure 1.

Development of zebrafish immunity in comparison with the human immune system. Zebrafish possess a complex immune system, similar to that of humans. Development of zebrafish larvae is shown in A and a time line is shown in B. The appearance of components of the immune system is shown in C, and comparison is made with human immune components. Components of the innate immune system are detectable and active in the first days postfertilization (dpf) (e.g., macrophages, neutrophils, eosinophils, and mast cells). Adaptive immunity takes longer to develop and starts with thymus development at 60 hours postfertilization (hpf) and the appearance of the first lymphocytic markers at ∼4 dpf. At 21 dpf, the thymus is fully matured, and the first mature T cells and B cells are detected; humoral immunity is functional at 28 dpf. wpf, weeks postfertilization; TCR, T-cell receptor; NK, natural killer; MHC, major histocompatibility complex; DC, dendritic cell; APC, antigen-presenting cell; TLR, Toll-like receptor; NITR, novel immune-type receptor; TNF-α, tumor necrosis factor-α; IFN, interferon. References cited in figure as follows: a, Herbomel et al. 1999; b, Herbomel et al. 2001; c, Traver et al. 2003; d, Meijer and Spaink 2011; e, Novoa and Figueras 2012; f, Renshaw et al. 2006; g, Le Guyader et al. 2008; h, Renshaw and Trede 2012; i, Bertrand et al. 2007; j, Balla et al. 2010; k, Meeker and Trede 2008; l, Dobson et al. 2008; m, Lam et al. 2002; n, Trede et al. 2004; o, Danilova et al. 2004; p, Schorpp et al. 2006; q, Meeker et al. 2010; r, Laing and Hansen 2011; s, Lam et al. 2004; t, Danilova et al. 2005; u, Page et al. 2013; v, Yoder 2009; w, Yoder et al. 2010; x, van der Sar et al. 2004b; y, van der Vaart et al. 2012; z, Palha et al. 2013; aa, de Jong et al. 2011; bb, de Jong and Zon 2012; cc, Lugo-Villarino et al. 2010.

Figure 2.

Pathology in adult fish compared with human granulomas. Zebrafish granulomas caused by M. marinum show great similarities with human granulomas formed after infection with M. tuberculosis. Panels represent nonnecrotic early granuloma in human (A) and zebrafish (B) and granulomas with a necrotic center in human (C) and zebrafish (D). Human granulomas obtained from a neuropathology study in our Department of Pediatric Infectious Diseases and Immunology (D Zaharie, S Roest, M van der Kulp, AM van Furth, pers. comm.).

Figure 3.

Routes of infection. Zebrafish are infected with M. marinum at different time points and via different inoculation routes. Systemic infection is achieved by injection into the caudal vein at 28 hpf or inoculation via the duct of Cuvier in embryos at 2–3 dpf (Benard et al. 2012). Local injection routes are the hindbrain ventricle, muscle, notochord (Alibaud et al. 2011), or otic vesicle. In addition to intravenous injection at 24–28 hpf, yolk injection can be applied at the one- to four-cell stage in a high-throughput setting (Benard et al. 2012).

Figure 4.

Pathology in embryos. (A) Merged bright-field and fluorescent image of a zebrafish embryo infected with red fluorescent M. marinum and photographed at 5 dpi. (Adapted with permission from Stoop et al. 2011.) Clustering of mycobacteria and early granuloma formation is shown as red spots. (B–D) Higher magnification of an early granuloma at 5 dpi formed after bloodstream infection, derived from analysis using confocal imaging by our research group. (B) M. marinum E11 (in red), (C) phagocytes stained with anti-L-Plastin (in green), (D) merge of B and C confirming the colocalization of these cells in early granulomas in zebrafish embryos. Scale bar, 35 µm.

Figure 5.

Graphical summary. This summary shows bacterial and host characteristics important in mycobacterial pathogenesis derived from and validated in the zebrafish infection model. The different hallmarks of pathogenesis are shown. For each step, red shows identified mycobacterial factors important for the pathogen to survive and cope with the immune system of the host, and green shows host factors required for an appropriate immune response. Three types of granuloma are described and schematically depicted in this summary. The granuloma in the middle is the normal granuloma with a balance between inflammation and infection; at the left and right, granulomas without the right balance are depicted with high infection and high inflammation, respectively. Labels in the figure refer to the following references: a, Stoop et al. 2012; b, van der Vaart et al. 2012; c, Davis et al. 2002; d, van der Woude et al. 2012; e, Cosma et al. 2006; f, Meijer and Spaink 2011; g Alibaud et al. 2011; h, Gao et al. 2006; I, Clay et al. 2007; j, Kanwal et al. 2013; k, Davis and Ramakrishnan 2009; l, Volkman et al. 2010; m, Tobin et al. 2010; n, Volkman et al. 2004; o, van der Woude et al. 2013; p, Stoop et al. 2013; q, van der Vaart et al. 2013; r, Roca and Ramakrishnan 2013; s, van der Sar et al. 2009; t, Elks et al. 2013.