Mycobacterial Dormancy Systems and Host Responses in Tuberculosis (original) (raw)

Immunology of Dormant Mtb

Evading Immune Detection

Masking: An Immunological Disguise

Mycobacteria adopt multiple strategies to avoid the attack from macrophages. They express surface lipids such as phthiocerol dimycoceroserate, which can mask the pathogen-associated molecular patterns (PAMPs), thereby going “unnoticed” by the innate immune system (10). The phenolic glycolipid produced by these bacilli induces the production of the chemokine CCL2 to recruit macrophages for further infection (10). In the upper airway where a constant and heavy recruitment of macrophages occurs due to the presence of TLR stimulating bacteria, thus posing a very hostile environment, Mtb adopts a different immune evasion strategy by forming small infection droplets that allow them to be delivered directly into the alveolar spaces of the lower lung, which anchorages a few microbicidal macrophages (11).

Manipulating the TLR Responses

In the macrophages, which are the crucial niche for replication, Mtb interacts with various receptors to initiate phagocytosis. Despite the bactericidal properties of the macrophages, Mtb employs phagocytosis as a primary mode of gaining entry to establish the niche. The opsonization of the bacillus by the complement or antibodies determines the nature of receptors engaged and also the nature of events that are involved in the outcome of the infection. Recognition of Mtb through its cell wall glycolipids involves the formation of TLR heterodimers (12). The importance of TLR-mediated signaling during Mtb infection is well proven in various TLR knockout animal models (13). Mycobacterial components such as lipomannan, lipoarabinomannan (LAM), 38- and 19-kDa mycobacterial glycoproteins, and phosphatidylinositol mannoside (PIM) induce the formation of TLR1/6 heterodimer (12). The 38- and 19-kDa mycobacterial glycoproteins, PIM, and triacylated lipoproteins favor the formation of TLR2/TLR1, whereas the diacylated lipoprotein induces TLR2/TLR6 dimerization (13). The susceptibility to Mtb infection is also due to genetic polymorphisms in the host genes (14). It is well established that Mtb has the ability to modulate the immune responses to its advantage. Exposure of THP-1 cells to Mtb cell wall components results in the de novo synthesis of TLR4, thereby decreasing the production of Th1 cytokines (15). Induction of apoptosis in bystander cells during Mtb infection of macrophages is a classic example of how this pathogen causes immunosuppression in infected individuals, thereby gaining the survival advantage (16). Interaction of Mtb cell wall components with TLRs modulates a number of events that include antigen presentation (17), phagolysosomal fusion (13), apoptosis of macrophages (12), and production of reactive oxygen and nitrogen intermediates (18).

Although Myd88-dependent signaling of TLRs is well established in mycobacterial pathogenesis, recent studies indicate independent roles for Mal (the TLR adaptor) and Myd88. Individuals with the single-nucleotide polymorphisms, D96N and S180L, in the TIRAP gene (that codes for Mal) display differential susceptibility to Mtb. Heterozygous genotype was associated with a protection toward TB, whereas the homozygous genotype was related to susceptibility (19). Using a murine model that carried the human equivalent mutation in TIRAP gene, it was demonstrated that in the homozygous genotype for the mutation, the mycobacterial load was higher and this was independent of the macrophage cytokine production (20). Further, in vitro studies indicated that mutation in TIRAP gene affected phagosome maturation and intracellular killing of Mtb (20).

Antigen Presentation by MHC

The TLR2-dependent surface expression of MHC class II receptor and their antigen-presenting ability was found to be inhibited by either Mtb infected or 19-kDa lipoprotein (LpqH) exposed macrophages (12, 21). Class II transactivator (CIITA), a TLR-2-dependent regulator of MHC class II a, b, invariant chains contributes to antigen processing and its expression was found to be decreased during Mtb infection (2224). Its importance is further strengthened with the observation that CIITA knockout mice could not survive Mtb infection (25). Mtb also inhibits the expression of genes involved in MHC class II processing and presentation (26, 27) and the posttranslational function of these molecules. Another interesting feature by which Mtb evades TLR-mediated immune response is by the differential antigen-presenting ability of MHC class II. It is demonstrated in _Mtb_-infected lung DCs of mice that despite normal levels of MHC class II molecules, antigen presentation capability was decreased, whereas in macrophages, both the MHC class II molecules expression and antigen presentation capacity were found to be decreased (2831).

Phagolysosomal Fusion/Fission

During chronic infections, repeated stimulation of TLRs by the Mtb components such as mannosylated-LAM (ManLAM) and PIM causes phagosomal maturation and arrest allowing persistence of mycobacteria inside the phagosome (32) (Figure 1). Among the successful strategies adopted by Mtb to establish a niche in the host, inhibition of macrophage maturation is best characterized. The mycobacterial products (ManLAM, trehalose dimycolate, and sulfolipids), phosphatase SapM, kinase PknG, and early secretory antigenic target-6 (ESAT-6) have been implicated in the inhibition of macrophage maturation (33, 34). ManLAM inhibits Ca+ surge that modulates the calmodulin- and Ca2+/calmodulin-dependent kinase II-dependent delivery of early endosomal autoantigen1 (3537), which in turn is necessary for the delivery of lysosomal hydrolases and vacuolar H+-ATPases into phagosomes. ManLAM blocks ESAT-6 recruitment by inhibiting PI3K hVPS34 to block PIP3 production as well as SapM-mediated dephosphorylation of phosphatidylinositol 3-phosphate (3739). Further, phosphorylation of unknown substrates by the kinase PknG and activation of p38MAPK by LAM to reduce the levels of Rab5 are some other mechanisms by which Mtb inhibits macrophage maturation, though the exact role of ManLAM in these two mechanisms is not clear (4042). Further, Mtb also disrupts the scaffolding of endosomes required for phagosome–endosome interactions leading to delay in phagosomal maturation (43, 44). Because of the abovementioned mechanisms, the phagosomal compartment formed in the macrophage is devoid of acidification and lysosomal enzymes, thereby allowing Mtb to successfully establish a niche for its growth and replication.

Apoptosis of Immune Cells and Redefining the Immune Cell Fate: A Process of Exploitation

Mycobacterial infection leads to various cellular fates such as apoptosis, necroptosis (type of programed necrosis), and autophagy (4547). In the macrophages, apoptosis and autophagy are the natural defense mechanisms that operate to eliminate microbial infection and invasion. The spread of these bacilli is lowered (46) by tumor necrosis factor (TNF)-α activated caspase 8-mediated extrinsic cell death pathway that involves kinases p38, ASK1, and c-Abl (48). Additionally, autophagy also promotes the clearance of Mtb (47, 49), which is supported by the observation that the survival of Mtb depends on the expression pattern of factors predominantly involved in autophagy (50). However, mycobacterial species have evolved mechanisms that prevent apoptosis and the autophagy of immune cells, so that they can survive in these cells and remain dormant for longer durations (51). Interestingly, the extent of alveolar macrophage apoptosis and Mtb virulence are inversely correlated. Mtb H37Rv, a virulent strain, inhibits apoptosis by enhancing the release of membrane-bound TNFR2 receptors (52, 53) and also by upregulating the expression of Mcl-1 protein, a member of antiapoptotic B-cell lymphoma/leukemia 2 family (54). The products encoded by certain Mtb genes also influence the apoptosis of infected macrophages. Overexpression of the Mtb type I NADH dehydrogenase (nuoG) neutralizes NOX2-derived reactive oxygen species (ROS) (55, 56) and thereby inhibits apoptosis in macrophages. Such a role was also observed in PknE and SecA2 genes using knockout strategies (57, 58).

Apoptosis is induced in host cells in response to pathogen infection, which determines the initiation of infection, survival, and escape from the host. An interesting feature during this process is the induction of apoptosis in uninfected cells due to the non-specific activation of cytokines (59). Thus, the induction of apoptosis of T cells (bystander cells) by Mtb is beneficial for its survival (16). In T cells stimulated with non-specific phytohemagglutinin (PHA) or specific culture filtrate protein of Mtb, apoptosis was evident in non-specifically stimulated T-cells that were dependent on Fas–Fas ligand interactions. Further, a significant release of TNF-α indicated its association with specific T-cell apoptosis during Mtb infection of macrophages (16).

Mtb exerts altogether a different kind of action on T-cells (CD4+ and CD8+) to delay T-cell responses by inhibiting apoptosis. Evidence for the ability of Mtb to inhibit apoptosis is indicated by (a) the promotion of development of CD8+ T cell responses by the Sec2A-deficient mutant strain of Mtb; (b) manipulation of eicosanoid metabolism of T cells (60); (c) increased frequency of macrophage apoptosis, accelerated CD4+ and CD8+ T-cell responses, and enhanced control of bacterial burden in Alox5 (5-lipoxygenase, required for generation of LXA4) deficient mice infected with virulent Mtb (61); and (d) enhanced susceptibility to Mtb infection due to polymorphisms in Alox5 and lta4h (62, 63). Another important mechanism by which Mtb enhances its survival in the host is to delay the expansion of Foxp3+ regulatory T (Treg) cells and thereby delay adaptive immunity (64). Although inflammation and cytokines produced during Mtb infection play a role in Treg cell proliferation, Treg cells that recognize Mtb antigens expand preferentially. It is reported that mice infected with wild-type Mtb displayed higher Treg cell proliferation than those infected with a virulent strain of Mtb deficient in expression of the specific antigen, Ag85B (64). Further, increased bacterial load associated with delay in priming of effector T cells was observed in mice receiving _Mtb_-specific Treg cells and such an effect was not observed in mice that received _Mtb_-specific Foxp3− CD4+ T cells (64). These observations clearly indicate that Mtb expresses certain antigens and allows Treg cells specific to these antigens to proliferate rapidly and limit the rate of effector T-cell priming and expansion at this site.

Migration of DCs from the lungs to the lymph node precedes CD4+ T-cell responses during Mtb infection, and this was demonstrated in interleukin (IL)-12p40-deficient mice (65). Further, the restoration of the ability of DCs and activation of CD4+ T cells upon treatment with IL-12p40 (65) and presence of the same subset of DCs in the lymph nodes that were primed in the lungs (30) provides further evidence for the acquisition of bacilli and trafficking by DCs as a rate limiting step in the initiation of adaptive immunity. An inflammatory stimulus that resulted in migration of DCs from the lung to the lymph node failed to accelerate the delivery of Mtb from lung to lymph nodes, suggesting that DCs infected with Mtb are intrinsically impaired to migrate (65).

Manipulation of Host Cytokine Responses

Mtb infection manipulates host cytokine responses in different directions to create a balance and take advantage for its survival. Prolonged signaling of TLR2 by Mtb cell wall components results in increased production of IL-10, IL-4, and TGF-β, which then inhibits the IFN-γ-mediated activation of macrophages (27, 6669), thereby the immune surveillance of T cells is evaded. Recognition of mycobacterial PAMPs by macrophages stimulates the production of cytokines like type I interferons (IFNs) and TNF (70, 71) through the TLR2 pathway, and these two cytokines promote apoptosis and necroptosis. Host cytokine responses are manipulated by various molecules during Mtb infection (72). It is demonstrated that IFN-γ and ESAT-6 inhibits TNF-α, IL-17 production, and early expression markers on T cells (73). Tim-3 inhibits the expansion of Th1 cells to prevent production of excess pro-inflammatory cytokines (74). Further, programed death-1 (PD-1) inhibits CTL function during Mtb infection (75). Mtb evolved mechanisms to control the expression of ESTAT-6, Tim-3, and PD-1 to control the manipulation of cytokine responses.

Increased production of TNF-α by macrophages induces generation of mitochondrial ROS, which confers the antimicrobial properties and necroptosis of these cells during Mtb infection (7678). TNF-mediated ROS generation is brought through receptor-interacting protein 1 (RIP1), receptor-interacting protein 3 (RIP3), phosphoglycerate mutase family member 5, mixed lineage kinase domain-like protein (MLKL), and dynamin-related protein-1-dependent pathways (77). In brief, binding of TNF-α to its receptor results in the formation a membrane-proximal super-molecular structure complex 1 [containing TNF receptor-associated death domain (TRADD) that binds to RIP1, TNF receptor-associated factor 2/5 (TRAF2/5), and cellular inhibitor of apoptosis 1/2 (cIAP1/2)] followed by polyubiquitination of RIP1 or TRAF2 by cIAPs, which then allows NF-κB translocation into nucleus to initiate transcription of A20 and cylindromatosis (CYLD) (79, 80). Deubiquitination of RIP1 by A20 and CYLD (79) results in complex I getting converted to complex II [containing RIP1, Fas-associated protein with death domain (FADD), caspase-8, and TRADD] (80). Apoptosis is initiated by the activated caspase-8 of this complex and in situations wherein this activity is abrogated, RIP1 and RIP3 come together in complex III also with FADD, caspase-8, and TRADD to form a necrosome, in which RIP1 phosphorylates RIP3 and further engages MLKL, leading to necroptosis (80). Mtb targets the caspase-8 activity and leading thereby the macrophages to undergo necroptosis instead of apoptosis, since the former process is favorable to its survival (81). Inducing a very high level of TNF-α and promoting the secretion of a biological factor that can block caspase-8 activity have been proposed to be reasons for the ability of virulent Mtb to favor occurrence of necroptosis (82).

Other cytokines that are implicated in Mtb pathophysiology are IL-10 and IL-4/IL-13. IL-10, also referred to as “cytokine synthesis inhibitory factor,” produced by Th2 cells regulates macrophage and DC function in response to Mtb infection. The production of IL-10 during Mtb infection is more of an advantage to the pathogen than the host. IL-10 facilitates Mtb survival by inhibiting phagosome maturation thorough a STAT3-dependent and p38-independent mechanism (83), IFN-γ-mediated production of reactive oxygen and nitrogen intermediates (84), blocking antigen presentation by downregulating the expression of major histocompatibility complex molecules (85), DC migration (86), and recruitment of Th1 cells to the lungs by modulating CXCL10 production (87). In humans, IL-10 is responsible for limiting immune responses during Mtb infection (88). Further, an association between IL-10 gene polymorphism and susceptibility to TB was demonstrated (89). On the other hand, the Mtb strains HN878 and CH subvert the immune response via induction of IL-10 (90).

Besides the classical activation of macrophages modulated by many cytokines during Mtb infection, IL-4/IL-13 facilitates alternative activation of these cells. Mtb exploits the alternatively activated macrophages to divert the microbial actions of classically activated macrophages. Alternative activation results in induction of Arg1 gene whose protein product competes with iNOS for the substrate l-arginine (91). This results in lower production of NO reactive intermediates. Upregulation of IL-4/IL-13 was observed in patients with progressive pulmonary TB and in PBMCs infected with HN878 Mtb strain (9294). IL-4/IL-13 induces Arg1 in alternatively activated macrophages to subvert the host NO-based mycobactericidal activity and could be a tactic by Mtb to thrive inside classically activated macrophages. On the other hand, the enhanced production of IL-4/IL-13 due to alternate activation of macrophages inhibits autophagy to facilitate the survival of Mtb (95).

Resistance to Reactive Nitrogen Intermediates

Generation of reactive nitrogen intermediates (RNIs) by the macrophages through nitric oxide synthase 2-dependent pathway mediated by IFN-γ is an antimicrobial strategy displayed by these cells and this process has been shown to be vital for the control of TB (96). In the macrophages, the inducible form of NOS is activated by the cytokines produced by Th1 lymphocytes that stimulate the production of nitric oxide (97), which reacts with superoxide radicals to form RNIs. The RNIs thus produced, attack bacterial macromolecules to aid in killing. Although the role of RNIs in the control of TB in humans is not yet clear, some studies provide substantial evidence that RNIs play a role in innate immunity mounted during mycobacterial infection. The susceptibility to TB was found to be associated with genetic alterations in the NOS2A gene (98). Further, the negative correlation of mycobacterial growth and NO production in human alveolar macrophages, elevated expression of NOS2 in the lungs of TB patients, and reactivation of dormant TB due NOS2 inhibition support the role of the RNIs mycobacterial pathogenesis (99). However, Mtb has developed mechanisms that can subvert the antimicrobial actions of the macrophages, which allows these bacilli to establish a niche and remain in the host for a long time. ManLAM was also described to trigger Th2 cytokines such as IL-4 and IL-10 that inhibit the action of inducible NO synthase, an enzyme critical for the production of NO (9).

The alkyl hydroperoxide reductase subunit C encoded by the mycobacterial gene AhpC, in association with peroxidase, peroxinitrite reductase, dihydrolipoamide dehydrogenase (Lpd), dihydrolipoamide succinyltransferase (SucB), and thioredoxin-like AhpD catalyzes the breakdown of RNIs to protect Mtb from the antimicrobial actions of the macrophages (100, 101). Further, the mycobaterial gene MsrA encodes methionine sulfoxide reductase, an enzyme that converts methionine sulfoxide [produced out of a reaction between peroxynitrite (ONOO−) and methionine residues of proteins] to methionine, protects bacteria against RNIs (102). Rv1205, a pupylated proteasome substrate, catalyzes the production of cytokinins and helps Mtb to defend against NO (103). To summarize, TLR agonists present in the thick cell wall of Mtb causes prolonged TLR signaling resulting in various immune evasion mechanisms. Among these mechanisms, inhibition of MHC class II appears to be predominant that allows preventing detection by CD4+ T cells.

Antibiotic Tolerance: An Acquired Fitness Advantage

Besides the fixed genetic mutations that the Mtb undergo to develop antibiotic tolerance, another interesting feature adopted by Mtb to remain dormant in the host is the development of “phenotypic drug resistance” or simply called “drug resistance,” during which a transient resistance to antibiotics is developed (104, 105). This is achieved by induction of specific macrophage-induced efflux pumps (106). Adding to this, the same pumps promote intracellular bacterial growth, thereby providing a double advantage to the Mtb.

The role of toxin–antitoxin (TA) systems in conferring the “non-classical” antibiotic resistance to allow Mtb to remain in a non-replicating phase for longer durations is very interesting. This system encoded by two genes is composed of two proteins, namely, the long-lived protein “toxin” and the short-lived protein “antitoxin.” In Mtb, under normal physiological conditions, the toxin is neutralized by the antitoxin. To remain in dormant stage, Mtb represses the expression of the antitoxin protein resulting in the accumulation of toxin. Under such conditions, the toxin protein acts as a ribonuclease to cleave free and ribosomal bound single-stranded mRNA resulting in inhibition of protein synthesis and bacterial growth. This allows the Mtb to persist in the host without any signs of infection state for a long time. Multiple TA systems, approximately 88, exist in the bacilli and their numbers are ever increasing (107109), with H37Rv strain found to harbor 38 modules of 5 TA systems (3 relBE, 24 vapBC, 8 mazEF, 1 higBA, and 2 parDE) (108). Over expression of these genes belonging to different modules in Mycobacterium smegmatis stopped the growth of the recipient cells to remain in latent phase. In general, the mycobacterial TA proteins are homologous to the proteins of other bacterial species and some of them exhibit certain special features. The MazF protein besides cleaving mRNA can also interact with DNA topoisomerase (110), thus indicating that differential mechanisms may exist in the way the TA proteins act to contribute to persistent infection in the host.

Granuloma Formation: A Finely Negotiated Refuge?

Granulomas are formed in response to infectious and non-infectious stimuli and are associated with various diseases (111). They are aggregates of macrophages whose membranes are interlocked and their occurrence is more prevalent in TB worldwide. Although it was initially reported that granulomas are complex protective structures that contain host cells to wall off bacteria (104) and can also sterilize infection (112), many studies have indicated that they are conducive to heavy bacterial burdens in TB (113). In case of active disease, some of the lesions are cleansed by the host even though there is progression of other lesions indicating that lesional heterogeneity persists after the initiation of adaptive immunity where differential killing of the bacteria take place, deciding the outcome of the clinical infection (112). Mtb in fact enhances the formation of granulomas (114) for their expansion and dissemination. This is accomplished by spreading of Mtb from dying macrophages to newly recruited ones. Macrophages undergo apoptosis when the bacterial load reaches a threshold and the nascent granulomas are presented to the uninfected macrophages. Thus, the Mtb numbers are phenomenally expanded by way of macrophage death and re-phagocytosis (114).

In the granuloma lesions, either the infection is cleared or viability is promoted by creating a favorable niche to the bacterium (Figure 2) (115). The immune status of the human host and transcriptional signature of the bacteria directs early granuloma establishment and consequence of the disease (116). However, it is shown that the immunological responses to Mtb by infected individuals vary to a great extent and is dependent on the granulomatous lesion formed. Two kinds of granulomatous lesions were demonstrated in cynomolgus macaques that were infected with low doses of Mtb (117). The classical ones are the caseous granulomas with low oxygen availability that is characterized by fibroblasts surrounding the epithelial macrophages and neutrophils in the periphery and dead macrophages in the center (118), whereas fibrotic lesions are associated with latent TB and exclusively filled with fibroblasts with sparsely distributed macrophages (119). In the event of an immunocompromised condition, granulomas liquefy and the bacteria are released from degraded granuloma to re-infect lung tissue and spread to new hosts (120).

Although granuloma formation is a basic immune response elicited by the host against an infection, it is also promoted by Mtb as part of its virulence program (121, 122). TNF-α plays an important role in mounting a response that is cytotoxic to the pathogen as well as maintaining the structural integrity of the granuloma and this is evident by the fact that neutralization of TNF-α leads to fatal reactivation of the bacterium and increased burden of the bacilli in the lung tissues (123, 124). It is demonstrated that the lipid-rich mycobacterial cell wall composed of trehalose dimycolate induces TNF-α induction and thus the granulomatous inflammatory response (120). Eliciting an excessive TNF-α response is another mechanism of virulence of Mtb to favor its existence in the host without being detected. This is accomplished by increasing the levels of TNF-α in the macrophages by directly interfering with cAMP mediated responses. In the mutant strain of Mtb lacking Rv0386 (encodes an adenylate cyclase), reduced protein kinase A and cAMP response element-binding protein activation was observed, which results in a significant reduction of macrophage TNF-α secretion (125). Although excess production of TNF-α is required for the containment of Mtb, this pathogen, however, uses adenylate cyclase to deliver excess cAMP to macrophage cytoplasm and acts to subvert host cell signal transduction so as to result in a pro-granulomatous response with excess TNF-α secretion (125).

The mechanisms of granuloma formation are not well studied. However, evidence points to the role of matrix metalloproteinases (MMPs) and tissue inhibitors of metalloproteinases (TIMPs) in Mtb for induction of granuloma formation (126). The expression of MMPs is largely dependent on cAMP-mediated signaling in the host cell (127) and thus the levels of these enzymes can be modulated by bacterial mechanisms (for example, the Mtb Rv0386 pathway) that regulate cAMP levels. Secretion of MMP-1, -2, -7, and -9 and decreased expression of TIMP-1, -2, and -3 (126, 128) have been shown in the peripheral blood mononuclear cells and human airway epithelial cells. Governing granuloma formation is also exploited by this bacterium to increase its numbers and dissemination systemically. In the _Mycobacterium marinum_-zebrafish model, it was demonstrated that the infected macrophages attract new cells and they in turn ingest the bacterium and allow them to sustain, thus allowing the granuloma to expand (114). Thus, it is clear that Mtb pathogenesis involves subversion of host signaling in a pro-inflammatory manner to create an environment that is favorable for its dormancy.

Regulation of Granuloma Maintenance

Calcification

Neutrophils on the other hand also display antimycobacterial activity and the mechanism involves generation of ROS (129, 130). Further, infected neutrophils undergo apoptosis (131, 132), thus contributing to an effective adaptive immunity during Mtb infection (133). However, Mtb adopts strategies to inhibit apoptosis and promotes necrosis of these cells by various mechanisms by expressing the region of difference 1 (RD1) and a type VII secretion system (ESX) (134). Apoptotic neutrophils are also subjected to necrosis because of the activation of Calpain, a Ca2+-activated protease (135) due to Ca2+ influx, and this process is generally termed as Ca2+-induced necrosis. The Ca2+ influx and the subsequent necrosis of neutrophils are promoted by ESAT, a mycobacterial leukocidin (136). Further, the dormancy regulator of DosR activates the Ca2+ ATPase of the plasma membrane of the microvesicles of mycobacteria to create hypoxic conditions (137). Thus, the calcification of neutrophils is an important survival strategy adopted by Mtb at the cellular level. On the other hand, it was reported that feeding _Mycobacterium paratuberculosis_-infected rats with low calcium enhanced their ability to clear the infection (138). Besides utilizing Ca2+ for its survival at the cellular level, Mtb infection causes extensive calcification in various host tissues. Calcification is observed in myocardial (139), pulmonary (140), musculoskeletal, central nervous, abdominal (141) and genitourinary systems (142), and in lymphatic tissue and more importantly the granulomas. During the granuloma formation, the parenchymal tissues calcify to form caseous structures, especially during chronic or latent infection (143). Calcification allows the formation of various granuloma structures within a single host, thus creating different microenvironments that are unique among them to allow the survival of Mtb (143). The molecular mechanisms of granuloma calcification remain unclear. However, calcification of granulomas allows a safe environment for Mtb to remain dormant for extended periods of time.

Fibroblasts

Besides macrophages, granulomas also contain other cells such as eosinophils, neutrophils, and fibroblasts. In the mature granulomas, fibroblasts are found in the periphery and embedded within collagen fibers. TB inflammation is associated with increased number of fibroblasts and the fibroblastic activity (144). These cells produce collagen bundles such that the fibrotic capsule separates the bacteria from the surrounding tissue. This process is a consequence of intra-granulomatous cytokine secretion (145). On the other hand, fibroblasts also participate in the burning out of granuloma (in the event of the pathogen completely eliminated) by secreting tissue inhibitors of metalloproteases (145). The dynamics of granuloma formation depends on the number and degree of differentiation of fibroblasts (144). Interferon-γ primed fibroblasts present bacterial antigens to Th1 cells for further processing. However, it was demonstrated that _Mtb_-infected fibroblasts fail to present the antigens (146), thus enabling these bacilli to evade immune response of Th1 cells. Mtb also directly affects the collagen turnover and expression of matrix metalloproteases and tissue inhibitors of matrix metalloproteases in fibroblasts (147). Thus, the manipulation of fibroblast function is one of the key strategies of Mtb for the establishment and sustenance of persistent infection.

Genetics of Host and Pathogen in Mycobacterial Dormancy: A Perfect Storm?

The pathogenesis of Mtb, especially its ability to remain in dormant state and reactivation, depends on both the genetics of the host and the pathogen (148). This is evidenced by the fact that (a) dormant mycobacteria do not replicate at all and their cell divisions occur at a very low rate (149, 150) and (b) a high level of genomic stability was observed in these bacterium isolated from human populations. Hence, studies on the genetic variations of the Mtb and the host have invited a lot of interest in the past few years and the same are discussed in this section.

Host gene–environment interactions play a crucial role in determining the outcome of TB and these are very important to evolve strategies to prevent mycobacterial infection at the genomic level. Interestingly, the ability of the host genetic control depends on the mycobacterial strain encountered and the exposure intensity (151). Two major loci on chromosomal regions, 18q11.2 and 11p13, were found to be associated with incidence of TB (152, 153). TST reactivity and interferon-γ release assay responses during Mtb exposure were found to be hereditarily controlled (154, 155). Since heritability is related to latent tuberculosis infection (LTBI), the phenotype of certain genes seems to govern mycobacterial infection and the associated dormancy. For example, in a Ghanaian population, IL-10 promoter haplotype (−2849G/−1082G/−819C/−592C) in TST non-responders was significantly more compared to TST responders (15.3 vs. 9.7%, OR = 2.09, p = 0.01) (156). In individuals carrying GG genotype at 1082A>G, the prevalence of TST non-response was 1.5 times than those carrying the AA and AG genotypes (156). Further, SNPs in IL4 (−590T>C, p = 0.007) and IFN-γ (+874A>T, p = 0.02) genes are associated with TST response (156). A linkage of persistent TST negativity with chromosomal regions 2q21-2q24, 5p13-5q22 (157), 11p14 (also called TST1 locus), and 5p15 (also called TST2 locus) was identified (158). Thus, the genetics of the host determines the susceptibility to mycobacterial infection and the development of LTBI.

Dormancy in Mtb is regulated by a set of approximately 50 genes, the DosR regulon, under the tight control of the dormancy survival regulator transcription factor (159). The genes of the DosR regulon are distributed in nine blocks in the genome: (block 1) Rv0079–Rv0081, (block 2) Rv0569–Rv0574c, (block 3) Rv1733c–Rv1738, (block 4) Rv1812c–Rv1813c, (block 5) Rv1996–Rv1998c, (block 6) Rv2003c–Rv2007c, (block 7) Rv2028c–Rv2032, (block 8) Rv2623–Rv2631, and (block 9) Rv3126c–Rv3134c (160). Evolutionarily, these genes are conserved across various pathogenic, non-pathogenic, and environmental bacteria of diverse habitats (160), and this could have occurred due to horizontal gene transfer mechanism during adaptation to challenging environmental conditions (161). Functionally, they emerged primarily to assist Mtb to adapt for anaerobic environment, thus enabling its survival in the host granuloma (162). Their expression is induced under hypoxia (163, 164) and under conditions where mycobacterial growth is inhibited by external growth factors both in vitro (macrophages) (165) and in vivo (mice and guinea pigs) (166, 167), suggesting their role in maintaining a low profile of bacterial growth under unfavorable conditions. Interestingly, the protein products of many of these genes seem to be good T-cell antigens and involved in many physiological processes of both the bacilli and the host.

The protein products encoded by DosR genes are predicted/reported to be involved in various functions, and on this basis, they were classified into eight groups (160). The functional roles of these genes are outlined in Table 1. Although they were predicted to have diverse functions, immune regulation and modulation of host responses have gained importance in the recent years. Out of the DosR genes expressed during dormancy, 18 are T cell responders that trigger strong IFN-γ response in TB patients. Further, it was demonstrated that Rv1733c, Rv2029c, Rv2627c, and Rv2628 are strong IFN-γ responders in latently infected individuals (168). Further, Rv2032, Rv1998c, Rv2031c, Rv2623, and Rv3132c exhibit strong T-cell response, whereas Rv0079, Rv0080, Rv3127, Rv2626c, and Rv2029c exhibit strong humoral immune response (169, 170). It was identified that Rv2626c is a secretory protein, which binds to macrophage affecting its function and also elicit TNF-α and strong B-cell responses (171). In aerosol-mediated murine TB model, decreased induction of pro-inflammatory cytokines (IL-1 and IL-12) and decreased bacterial load and delayed death was observed when infected with Rv1813c deletion mutant of Mtb (172). It is to be noted that evidence for the role of DosR genes in Mtb dormancy stems out from murine models and no concrete evidence exists in higher animal models. In a recent study, using macaque as the model system, it was demonstrated that during hypoxic conditions DosR regulon modulates the timing and magnitude of adaptive immune responses to favor persistence of infection by Mtb (173).

Predicted functional group Gene Reported/predicted functional role of the encoded protein
Stress response Rv2623 Regulates bacillary growth of Mtb by modulating ATP-binding (174). Biomarker for the diagnosis of latent tuberculosis meningitis (175)
Rv3134c Associates with devRS to form an operon that mediates the expression of DevR target genes (176)
Rv2031c Alpha-crystallin (Hsp-X), a master regulator of Rv-2018-2031 operon (177); blocks the differentiation of monocytes to dendritic cells (178)
Rv2624c, Rv1996, Rv2005c, Rv2028c Not yet characterized
Proteases and transport Rv2625c Metalloprotease (Rip3); required for early phases of pulmonary tuberculosis infection (179)
Rv1997 Not yet characterized
Rv1735c Immunogenic (169)
Rv1733c Enhances humoral and cellular immunity (180)
Host–pathogen interactions Rv2626c Modulates macrophage effector functions and delayed hypersensitivity (171). Diagnostic marker for tuberculosis (181)
Rv2004c Binds specifically to U397 macrophages and A549 epithelial cells to modulate immune responses (182)
Sensor kinases and transcriptional regulation Rv3132c (DosS), Rv2027c (DosT)Rv3133c (DosR) Form a two sensor kinase system (183)
Rv0081 Predicted to involve in encoding of formate Hydrogenylase complex (Rv0081-Rv0088 locus) (184); highly immunogenic in African populations (169)
Cell wall and protein synthesis Rv0079 RafH (ribosome-associated factor during hypoxia); stabilizes ribosome under stress conditions (6, 185); inducer of T-cell responses (186)
Rv0574c A pyroglutamate synthase-like protein, modulates poly-α-l-glutamine content in the cell walls to maintain cell integrity under hostile conditions (187)
Rv1738 Shutdown of ribosomal protein synthesis (188)
Nucleotide metabolism and repair Rv2630 Immunogenic in patients with active pulmonary tuberculosis; exact role in nucleotide metabolism and repair are yet to be proven in vivo (189)
Rv2631 RNA-splicing ligase RtcB; function not yet characterized
Rv0570 Putative vitamin B12 dependent ribonucleoside-diphosphate reductase; immunogenic (134)
Rv0571c Putative phosphoribosyltransferase; function not yet characterized
Nitrogen metabolism Rv3131 Putative NAD(P)H nitroreductase and immunogenic in nature (169, 190)
Rv2032 Putative NADP(H) nitroreductase (191); Potent inducer of host cytokines (192, 193)
Rv1737c, Rv1736c Nark2 and NarkX (encoded by Rv1737c and Rv1736c, respectively) are nitrate/nitrite transporters required during mycobacterial anerobic dormancy (194); immunogenic and potent diagnostic markers of tuberculosis (169)
Rv3127 Not yet characterized
Redox balance Rv0573c Predicted to involve in biosynthesis and recycling of nicotinamide; lmmunogenic (195)
Rv1812c Nitrogen metabolism during stress (172)
Rv3130c Putative Diacylglycerol _O_-acyltransferase; facilitates accumulation of triacylglycerol under stress (196)
Rv2029c Putative 6-Phosphofructokinase (PfkB); induces cytokine production (197)
Rv1734c, Rv2006, Rv1998c, Rv2003c, Rv2007c Not yet characterized
Hypothetical proteins Rv2628 Immune-mediated protection against tuberculosis (198)
Rv2627c Delays mycobacterial growth (199)
Rv2629, Rv3126c, Rv0569, Rv0572c, Rv0080, Rv2030c, Rv3128c, Rv1813c Not yet characterized

Functional role of proteins encoded by the DosR genes.

It is very interesting to note that mutations in dosR does not induce Mtb death under hypoxic conditions, indicating that other factors beyond dosR are important for the dormancy and survival of Mtb in the host. Further, the massive expression of dosR-independent genes during hypoxic conditions and the variation in the expression profiles of dosR genes in different strains of mycobacteria of different virulence indicates that further studies are required to warrant the exclusive role of these genes in mycobacterial dormancy (200202).

Environmental Signals of Sleep

The complex life cycle of Mtb involves adaptation to various stresses and to accomplish this it encodes about 190 regulatory proteins among which 11 form the two-component signal transduction systems (TCSSs) (134, 203). TCSSs found in Mtb are conserved in other closely related mycobacterial species in terms of genetic arrangement and location (134, 204). However, the number of functional TCSSs seems to vary between the species, wherein the Mycobacterium leprae had only four TCSSs (204). The TCSSs identified in Mtb are phoP-phoR, regX3-senX3, dosR-dosS (dosT), Rv0600c-Rv0601-ctcrA, narL-Rv0845, tcrX-tcrY, mprA-mprB, prrA-prrB, trcR-trcS, pdtaR-pdtaS, mtrA-mtrB, and kdpD-kdpE (205).

A typical TCSS comprises a histidine sensor kinase and a response regulator that are localized in the plasma membrane and cytoplasm, respectively. Both of them have specific domains through which they sense environmental cues. The sensor kinase comprises a sensor domain, one or more transmembrane domains, and a cytoplasmic transmitter containing a dimerization motif and a kinase domain, and the latter can be again divided into two subdomains possessing a histidine phosphorylation box and an ATP-binding pocket (made of N, D, F, and G boxes that have highly conserved amino acids); signal recognition results in dimerization and auto-phosphorylation followed by the transfer of this phosphate to the response regulator, thus enabling it to promote transcriptional, translational, and functional aspects (206209). The environmental cues that activate the TCSSs are not yet defined for all of them. Inorganic phosphate (regX3-senX3); SDS, triton X-100 alkaline pH, and nutrient limitation (mprA-mprB); low oxygen, nitric oxide, carbon monoxide, and ascorbate (DosS-DosR) are some of the signals that are identified to activate the TCSSs in model organisms such as M. smegmatis. However, whether the same signals also activate these TCSSs in Mtb is not yet clear. The signaling mechanisms of some of the characterized TCSSs remain more or less similar as described above for a typical TCSS, with some variations. One of the downstream events of TCSS activation is gene regulation. SenX3-RegX3 activation leads to the upregulation of several genetic determinants such as phoA (alkaline phosphatase) (210), pstSCAB and phnDCE (encodes phosphate-specific transporter systems) (210, 211), and phnF (encodes a negative regulator of phnDCE) (211) and SenX3-RegX3 itself (210, 212). PhoP-PhoR regulates about 150 genes that are involved in general and lipid metabolism, and respiration (213, 214). Further, a number of genes that code for membrane proteins, genes of dosRS regulon, genes of the PE/PPE/PE-PGRS protein families, and genes of the virulence-associated RD1 such as espB and espR (214, 215). The narL-Rv0845 TCSS regulates genes involved in nitrate metabolism during anaerobic respiration (216). mprA-mprB differentially regulates about 200 genes (217, 218). Examples include expression of its own gene, pepD and moaB2 and Acr2 (alpha-crystallin-like protein) (219, 220). The trcR-trcS TCSS regulates about 50 genes (221). dosR-dosS (dosT), one of the well-characterized TCSS, regulates about 48 genes (collectively called the DosR regulon), predominantly having a role in hypoxia. MtrA-MtrB seems to be an important TCSS since it regulates genes involved in DNA replication and cell wall integrity (222, 223). The gene expression regulated by pdtaR binding to RNA involves prevention of stem-loop structure formation by acting as an anti-terminator (224). The huge variety of genes regulated by TCSSs gives an edge for Mtb to create favorable conditions for its survival in the host in a dormant condition for longer durations.

Evidence to strengthen the crucial role of TCSS is demonstrated in animal models of TB infected with mycobacterial strains that harbor mutation in the TCSS genes. Attenuation of bacterial growth in lung, liver, and spleen and delayed time to death of _Mtb_-infected animals are some of the phenotypes described in studies that used TCSS mutants [reviewed in Ref. (205)].