Avoid What is Strong, Strike What is Weak: MAP Targets Early Host Responses to Prevent Immune Counterattack

Sun Tzu writes ‘Those skilled in war subdue the enemy's army without battle ' (Tzu, 19 71, p. 79). After infection, MAP does not attempt to fight the host immune response head-on, but instead undermines its foundation before it can begin.

Rapidly after phagocytosis, pathogenic my­cobacteria induce critical signal transduction failures, impairing normal phagosomal matu­ration. Normally, Rab5 and phosphatidylinosi­tol 3-phosphate (PI3P) coordinate through the Rab5 effector early endosomal antigen 1 (EEA1) for endosomal sorting and proper phagosomal maturation. Kelley and Schorey (2004) suggest­ed the mycobacterial manipulation of phagocy­tosis is multifaceted and active - after artificially increasing PI3P levels in macrophages, live M. avium hominissuis 104 continued to impair nor­mal phagosomal trafficking, while killed M. avium hominissuis was capable of being pro­cessed normally. Mycobacterium tuberculosis has been shown to secrete lipid phosphatase SapM and degrade PI3P, sabotaging the host process (Vergne et al., 2005). MAP has a putative homo­logue of this gene (K-10: MAP_RS17645), but it has not been investigated to date. Additionally, Rab7 - required for fusion and development of the phagolysosome - is not effectively recruited despite the presence of Rab5 in infected phago­somes. In M. bovis BCG, this recruitment failure was proposed as a key mechanism by which my­cobacterial pathogens freeze the phagosomal maturation process in the earliest stages (Via et al., 199 7). Later research with M. bovis BCG reported a Mycobacterium-secreted factor that inhibited Rab7-interacting lysosomal protein (RILP), interfering with Rab7 function across the entire cell and not merely in the Mycobacterium­containing compartment (Sun et al., 2007).

Many studies have examined gene ex­pression patterns and intracellular signalling inside MAP-infected macrophages relative to uninfected macrophages. In other cases, gene expression, particularly for cytokine genes, has been examined in MAP-infected intesti­nal tissues and draining lymph nodes. Specific studies targeting intracellular signalling inside infected macrophages have been conducted. These investigations have greatly enhanced our understanding of MAP pathobiology, par­ticularly when combined with MAP genome­sequencing efforts and the ability to genetically modify MAP. The effect of MAP on intracel­lular signalling in infected macrophages was highlighted in a study by Murphy et al. (2006), who used a bovine immune microarray. They examined differential gene expression in rest­ing macrophages and in macrophages infected with either MAP or the closely related M. avium hominissuis 104. Although M. avium hominis- suis can produce serious infections in immune- compromised humans, it is non-pathogenic in ruminants and is readily cleared by an efficient immune response. In general, both mycobacte­ria activated gene expression in macrophages 24 h post-infection (Murphy et al., 2006). However, macrophage responses to M. avium hominissuis were consistently more robust than to MAP. Of particular interest, over 41% of the differentially expressed genes in MAP- infected cells were members of, regulators of or regulated by the mitogen-activated protein kinase (MAPK) pathway. In keeping with gene expression patterns, M. avium hominissuis caused a more robust activation of p38 MAPK and extracellular regulated kinase (ERK1/2), two of the major MAPK family members. This response was also longer lived in M. avium hominissuis-infected macrophages relative to MAP-infected macrophages (Murphy et al., 2006). Given that activation of p38 MAPK and ERK1/2 occurred within 15 min of MAP or M.

Several studies suggest that MAP subverts the ability of infected macrophages to react to normal T-cell signalling, resulting in failure of macrophages to activate and destroy MAP. It has been proposed that this disruption may stem from alteration of the host's ability to re­ceive signals through the CD154-CD40 system (Sommer et al., 2009). The resulting cytokine re­sponse favours an inappropriate Th2-like activ­ity, including expression of interleukin (IL)-10, and is linked with a shift towards clinical disease (Hussain et al., 2016). The failure of normal sig­nalling between MAP- infected macrophages and T cells, and the absence of proper macrophage activation, are central to disease progression (Voo et al., 2009). These ideas have been devel­oped into models for immune response to MAP (Coussens, 2001, 2004; de Almeida et al., 2008; Khare etal., 2009, 2012; Arsenault etal., 2014; Khare et al., 2016).

In one study, it was shown that MAP- infected tissues contained high levels of IL-1α and TRAF1 (tumour necrosis receptor- associated factor-1) mRNA and protein (Aho etal., 2003).

This hypothesis is in line with studies dem­onstrating that pathogenic mycobacteria phago- cytosed by macrophages pre-activated with interferon-γ (IFN-γ) or tumour necrosis factor-α (TNFα), are destroyed through normal phagosome maturation (Flynn et al., 1993; Bonecini-Almeida et al., 1998; Florido et al., 1999). However, treat­ment of macrophages with these same agents af­ter infection fails to result in efficient destruction of mycobacteria (Denis et al., 1990; Robertson and Andrew, 1991). Therefore, interfering with normal macrophage activation pathways may be a key element in persistent infection and survival of mycobacteria, including MAP.

Infection of macrophages with MAP leads to dramatic upregulation of IL-10 (Weiss et al., 2005; de Almeida et al., 2008). This is signifi­cant since IL-10 can severely dampen proin- flammatory immune responses, which are critical to clearance of MAP and other intracel­lular infections. Activation of IL-10 gene expres­sion in MAP-infected macrophages is critically dependent upon rapid signalling through p38 MAPK (Hussain et al., 2016). This signal is re­ported to be mediated via interactions between MAP and Toll-like receptor 2 (Souza et al., 2008; Weiss et al., 2008; Weiss and Souza, 2008), and the same interactions have been noted with M. tuberculosis as well (Benson and Ernst, 2009).

From work in vivo, it has been suggested that infected cattle initially develop an early and effective proinflammatory immune response to MAP. However, this response typically declines in cattle that progress to clinical disease, favour­ing a Th2-like peripheral response and antibody production without control of infection (Stabel, 2000; Coussens, 2001, 2004; Tessema et al., 2001). Despite lack of a peripheral proinflam- matory response, MAP-associated lesions con­tinue to present a highly inflamed state. While this appears inconsistent with the loss of IFNγ and TNFα, there may be other factors at play, including perhaps a robust Th17 response and IL-23-induced inflammation. Other diverse dis­coveries on host-pathogen immune dynamics have been made, ranging from MAP-stimulated host prostaglandin E₂ production suppressing Th1 responses (Sajiki et al., 2018) to MAP- secreted extracellular vesicles inducing cytokine releases by macrophages (Wang et al., 2014a). Further complicating the issue, some question whether the Th1 to Th2 paradigm even exists or whether it is simply a product of immune system exhaustion as MAP confounds and defeats the host immune response (Begg et al., 2011).

9.6