extra-pulmonary dissemination TB 12

(1)

REVIEW

The mechanisms and consequences of the extra-pulmonary dissemination

of

Mycobacterium tuberculosis

Nitya Krishnan

*

_{, Brian D. Robertson, Guy Thwaites}

Centre for Molecular Microbiology and Infection, Division of Infectious Diseases, Faculty of Medicine, Imperial College, London SW7 2AZ, United Kingdom

a r t i c l e i n f o

Article history: Received 21 April 2010 Received in revised form 11 August 2010 Accepted 11 August 2010

Keywords: Tuberculosis Dissemination Epithelial cells Phagocyte Granulomas RD1

s u m m a r y

The dissemination ofMycobacterium tuberculosisfrom the primary focus of infection is central to the pathogenesis of tuberculosis. Trafficking of bacteria to the regional lymph nodes is essential to the development of a protective T-cell mediated immune response, but bacteria trafficked within the blood-stream can lead to extra-pulmonary dissemination and some of the most devastating clinical conse-quences of tuberculosis. Yet howM. tuberculosisleaves the lungs is poorly understood. Here, we review the potential pathways and molecular mechanisms behind the dissemination ofM. tuberculosisand consider the consequences to both host and bacteria. To disseminateM. tuberculosismust breach the alveolar epithelium and various bacterial factors have been implicated in this process. Heparin binding haemag-glutinin adhesin (HBHA) enablesM. tuberculosisto bind to sulphated glycoconjugates on epithelial cells; disruption of its synthesis severely impairs the ability of bacteria to disseminate from the lungs to the spleen. Two products of theM. tuberculosisRD1 gene locus, early secretory antigenic target 6 kDa (ESAT-6) and culturefiltrate protein 10 kDa (CFP-10), have been linked to cell lysis and may enable bacteria to invade and spread within the alveolar epithelium. Recent studies in embryonic zebrafish indicate ESAT-6 may also stimulate the trafficking of infected macrophages within granulomas, thereby promoting the early dissemination of bacteria. Thesefindings challenge conventional notions of the protective role of granu-lomas in mycobacterial infection and indicateM. tuberculosishas evolved specific mechanisms which utilise granulomas as foci of macrophage recruitment, infection, and subsequent bacterial dissemination. Further understanding of the pathways, mechanisms and consequences ofM. tuberculosisdissemination could have a major impact in designing novel vaccines and therapeutic strategies.

Introduction

Tuberculosis (TB) most commonly affects the human lung, but some of the most devastating clinical consequences of the disease result from the ability ofMycobacterium tuberculosis(M. tubercu-losis) to spread from the lung to other organs. The trafﬁcking of bacteria from the initial site of infectionethe pulmonary alveoluse

to regional lymph nodes and elsewhere has two conﬂicting conse-quences. First, it facilitates the presentation ofM. tuberculosis anti-gens within the regional lymph nodes, essential for the development of a protective T-cell mediated immune response. Second, it results in the dissemination of bacteria from the lung to other organs and can lead to forms of the disease, such as miliary and meningeal tuberculosis, which are invariably fatal without prompt treatment. Therefore, the mechanisms and pathways of

dissemination lie at the heart of tuberculosis pathogenesis and knowledge of their relationship to the development of adaptive immunity and disease may lead to novel approaches to disease prevention, diagnosis and treatment.

Potential pathways of dissemination

The ability ofM. tuberculosisto migrate from the primary site of infection to the lymphatic system and bloodstream is well estab-lished, but the precise mechanisms by which this occurs remain speculative. Theﬁrst obstacle to the bacilli is the alveolar epithe-lium, which may be breeched by direct invasion and lysis of epithelial cells, translocation across epithelial cells, or by travelling within professional phagocytes (Fig. 1).

Alveolar epithelial cell invasion

InhaledM. tuberculosisare likely to interact with epithelial cells, as they are the major constituents of the lining of the alveolus, and *Corresponding author. Tel.:þ44 2075943094; fax:þ44 2075943095.

E-mail addresses: [email protected] (N. Krishnan), b.robertson@ imperial.ac.uk(B.D. Robertson),[email protected](G. Thwaites).

Contents lists available atScienceDirect

Tuberculosis

j o u r n a l h o m e p a g e : h t t p : / / i n t l . e ls e v i e r h e a lt h . c o m / j o u r n a l s / t u b e

(2)

provide a less hostile niche than macrophages for bacterial repli-cation. It is evident, at leastin vitro, thatM. tuberculosisis able to replicate and undergo cellecell spreading through an epithelial monolayer.1,2 A number of studies suggest cell-to-cell spread is preceded by a period of intra-cellular bacterial growth, followed by lysis and infection of neighbouring cells.1Some have observed that

M. tuberculosisinfection is cytotoxic for epithelial cells leading to necrosis, which results in increased intercellular spread of the bacteria.3,4Indeed, important bacterial factors have been identiﬁed to explain the observed cytotoxicity ofM. tuberculosisin cellular models. Two proteins, early secretory antigenic target 6 kDa (ESAT-6) and cultureﬁltrate protein 10 kDa (CFP-10), encoded by genes

Rv3874and Rv3875of the RD1 gene cluster, have been strongly linked to cell lysis of both macrophages and pneumocytes.5e8 In addition, puriﬁed ESAT-6 was capable of forming pores in the membranes of red blood cells and macrophages in a dose depen-dent manner.7Recently, it was shown that ESAT-6 can bind laminin on the basolateral surface of alveolar epithelial cells and lyse both type 1 and 2 pneumocytes.8Taken together these studies suggest ESAT-6 is a cytotoxic factor which enables the bacteria to invade and spread within cells lining the alveolus through cell lysis.

Alveolar epithelial cell translocation

Whether cell-to-cell spread within the epithelium involves an extracellular phase remain speculative with conﬂicting evidence concerning the roles of extra- and intra-cellular bacteria. Spread of

M. tuberculosis within an epithelial cell monolayer in vitro was inhibited by the addition of an aminoglycoside antibiotic (which only kill extracellular bacteria),1 whereas others have found

intercellular spread unaffected by these antibiotics, suggesting

M. tuberculosismay travel between epithelial cells without passing into the extracellular milieu.2 A possible route of non-lytic inter-cellular dissemination ofM. tuberculosiswas described using the amoebaDictyosteliumas a model of a basic innate immune cell.9

M. tuberculosisandMycobacterium marinumwere able to exit the cells without lysis via an F-actin rich structure termed the ejecto-some, an activity dependent upon the activity of a GTPase RacH.9

However, bacteria were only able to exit via the ejectosome if they were able to efﬁciently escape the endosomal vacuole, a process which, again, appeared dependent upon the RD1 locus.In vitroinfection of human monocyte dendritic cells with Mycobac-terium bovisBCG and RD1 mutants ofM. tuberculosisdemonstrated the inability of these strains to enter the cytosol thus remaining exclusively in membrane bound phagolysosomes.10 Complemen-tation of the RD1 muComplemen-tations restored the ability of the bacteria to enter the cytosol thus, emphasizing the importance of the RD1 locus in the translocation of the bacteria. Although the phenom-enon of phagolysosome escape byM. tuberculosisremains contro-versial, it may represent an important step towards intercellular dissemination. The ability of mycobacteria to escape via the ejec-tosome additionally provides an elegant novel mechanism by which M. tuberculosis may undergo intercellular spread, but requires conﬁrmation in mammalian cells. Currently, the strongest evidence for the role of alveolar epithelial involvement in

(3)

demonstrated that following pulmonary infection, HBHA disrupted strains showed a 30e200 fold decrease in the number of bacteria recovered from the spleen, in comparison to the parental strain. In addition, fewer bacteria disseminated from the lungs to the spleen if the bacteria were coated with anti-HBHA antibodies before infection.11Gold particles coupled with recombinant HBHA were able to traverse a polarised epithelial cell layer without affecting the tight junctions that link the cells together.13These observations advocate a role for HBHA in the endocytosis ofM. tuberculosisand transcytosis across an epithelial cell layer. Thus, it appears HBHA is an important bacterial factor involved in the dissemination of

M. tuberculosis through interaction, primarily, with alveolar epithelial cells.

Translocation across endothelial cells

In addition to epithelial cells, M. tuberculosis is capable of invading endothelial cells as shown by the ability of mycobacteria to associate and multiply within human lung endothelial cells.14 One of the key events preceding the development of TB menin-gitis is the ability ofM. tuberculosisto breach the blood brain barrier (BBB). The BBB is primarily composed of brain microvascular endothelial cells which are held together by tight junctions. The mechanisms by which the bacteria cross this interface is not well characterised. Using anin vitromodel of human brain microvas-cular endothelial cells, Jain et al. demonstrated thatM. tuberculosis

was able to invade and traverse the monolayers by an active process involving actin cytoskeletal rearrangements, a pre-requisite for cell invasion. The efﬁciency of traversal across these monolayers was an attribute of pathogenic slow growing mycobacteria as opposed to the fast growers likeMycobacterium smegmatis. This model also revealed genes involved in the translocation of the bacteria across the monolayer.15Further studies on the genetic determinants and the interaction between host-bacterial signalling pathways of endothelial cell invasion, may aid in the understanding of the vital step of crossing the BBB and establishing disease at extra-pulmo-nary sites within the host.

Dissemination within professional phagocytic cells

M. tuberculosis may utilise the trafﬁcking of professional phagocytic cells as a means of disseminating from the lung. Macrophages are key players in the control of mycobacterial infection and the ability ofM. tuberculosisto replicate and survive within these cells has long been known.16,17Studies using anin vitro

model of human alveolar epithelial and endothelial cells have found mononuclear phagocytes improved the efﬁciency of bacterial translocation across the alveolar wall.18However, recent investi-gations infecting zebraﬁsh embryos withM. marinumhave thrown new light on the interactions between mycobacteria and macro-phages, and the role of the granuloma in early bacterial dissemi-nation and control.19,20

Macrophages form the major constituents of granulomas, which are conventionally considered to be structures that contain and prevent the spread of mycobacteria. Studies using the zebraﬁsh model have challenged this assumption by demonstrating that macrophages within granulomas are responsible for mycobacterial dissemination during the early stage of infection.19 Following

M. marinuminfection macrophages were rapidly recruited to the site of infection with up-regulation of inﬂammatory cytokines, TNF-

a

and IL-1

b

.20The initial stage of phagocytosis was followed by a period of bacterial growth, recruitment of new uninfected macrophages and formation of macrophage aggregates. Multiple rounds of cell recruitment and phagocytosis of infected apoptotic cells provide an expanding niche for bacterial growth, but also

constitute the early stages of granuloma formation.19_{The ability of} M. marinumto recruit macrophages and induce cellular aggregation appears dependent upon the bacterial genes encoded within the RD1 locus: RD1 mutants recruited fewer uninfected macrophages and also failed to disseminate to distant sites within the embryos as effectively as the parental strains. Theseﬁndings strongly suggest the mycobacterial RD1 locus is essential to initial granuloma formation and, at least in the early phase of infection, the bacteria utilise the granuloma as a focus for recruitment of uninfected macrophages and dissemination of infected cells to distant sites. In embryos devoid of macrophages, the bacteria were unable to disseminate beyond the epithelial barrier of the hindbrain ventricle, providing further evidence for the role of these cells in mycobacterial dissemination.20

Phagocytes other than macrophages may be involved in the internalisation and transport of mycobacteria from the lungs. Dendritic cells are an attractive candidate given their role in transporting antigens from the periphery to secondary lymphoid organs for the priming of the adaptive immune system.21Dendritic cells phagocytose mycobacteria and provide a favourable environ-ment for intra-cellular survival and growth, particularly as they are less efﬁcient at killing mycobacteria than macrophages.22,23 In addition,M. tuberculosisinfected dendritic cells play a crucial role in the priming of IFN-

g

producing T-cells and the cell-mediated response to infection.24,25 Cellular trafﬁcking studies in mice revealed dendritic cells to be the primary subset in the lungs and mediastinal lymph nodes to harbourM. bovisBCG, following intra-nasal infection.26Around 80% ofM. tuberculosisfollowing aerosol infection was associated with myeloid dendritic cells (CD11chigh/ CD11bhigh) in the lymph node, but these cells were poor stimulators of T-cells in comparison to resident lymph node dendritic cells which expressed low or negligible levels of CD11b.27The interac-tion between dendritic cell speciﬁc intercellular adhesion mole-cule-3 grabbing non-integrin (DC-SIGN), a c-type lectin receptor, and mannosylated lipoarabinomannan (ManLAM) on mycobacteria can modify the environment of the dendritic cell, allowing the bacteria to escape the immune surveillance mechanisms of the host.28The binding of DC-SIGN to ManLAM leads to the inhibition of toll-like receptor (TLR) mediated dendritic cell maturation, which can be reversed by the addition of DC-SIGN anti-bodies.29,30Interfering with the maturation of dendritic cells can lead to sub-optimal T-cell priming with immature dendritic cells likely to induce T-cells with a regulatory phenotype.31,32

Further-more, engaging the DC-SIGN receptor by ManLAM can produce IL-10 leading to immune suppression, thus enhancing the survival of the pathogen. These investigations suggest M. tuberculosis may target multiple cell phenotypes, with both macrophages and dendritic cells responsible for the trafﬁcking of bacteria to the lymph nodes and blood.

The timing and consequences of dissemination

(4)

migrating to the lungs and mediastinal lymph nodes, but this was not accompanied by an increase in bacteria in these lymph nodes.35 Only the administration of approximately 10 fold more bacteria provoked earlier dissemination to the lymph nodes, but only by 2 days (8 days vs 10 days).34Immediate bacterial dissemination to the lymph nodes was not achieved even by delivering an increased bacterial inoculum with an inﬂammatory stimulus. Host cells sampling the local environment within alveolar tissues are slow to migrate and prime T-cell responses following M. tuberculosis

infection. This may be because alveolar macrophages are consid-ered to be alternatively activated. Macrophages activated by the alternative pathway by Th-2 cytokines, IL-4 and IL-13 are less efﬁ -cient at clearing bacteria in comparison to classically activated macrophages.36,37

The dissemination ofM. tuberculosisfrom the primary focus of infection also has important consequences for the type of disease suffered by the host. Nearly 100 years ago, Clough demonstrated that the blood of patients with miliary tuberculosis (disseminated tuberculosis of multiple organs) contained viable bacteria which would cause tuberculosis when inoculated into animals.38In the 1930s, Rich and McCordock showed that the development of tuberculous meningitis was dependent upon an initial bacter-aemia to seed the brain; meningitis follows when these granu-loma (or Rich foci) release bacteria into the subarachnoid space.39,40Bacteraemia is, therefore, essential to the development of extra-pulmonary tuberculosis, but it may have an equally important role in the pathogenesis of pulmonary disease. The disconnection between the location of the primary focus of infection, which can be in any part of the lung, and disease, which in humans is usually in the lung apices, has been a source of long-standing controversy.39,41,42The study of guinea pigs infected with a low dose of virulent M. tuberculosis has provided particular insight into this conundrum41,42and has highlighted the ability of

M. tuberculosis to disseminate from the lung to the mediastinal lymph nodes 10 days following infection, with previously unin-fected lobes of the lungs being colonised via the bloodstream after 3 weeks.41Theseﬁndings suggest the dissemination ofM. tuber-culosis within the lung is also dependent upon bacteraemia, although why the lung apices are the preferential sites of disease remains unexplained.

Factors that inﬂuence bacterial dissemination

M. tuberculosisreplication and dissemination isﬁrst constrained by the host innate immune response and then by the development of a T-cell mediated response and the factors that inﬂuence dissemination can be viewed within this framework.

The innate immune response

The innate immune response to M. tuberculosis is mediated primarily by macrophages and is governed by intra-cellular sig-nalling events via pattern recognition receptors such as TLRs. Polymorphisms in the host genes encoding the molecules respon-sible for pathogen recognition and macrophage activation may determine the outcome fromM. tuberculosisinfection. Studies in Vietnamese adults have found disseminated tuberculosis with meningitis occurred more commonly in those with a single nucleotide polymorphism (SNP) in the genes encoding TLR-2 and Toll-interleukin 1 receptor domain containing adaptor protein (TIRAP), a protein that mediates signals from Toll-like receptors activated by mycobacteria.43,44Lipomannan and the 19 kDa lipo-protein are agonists of the TLR-2 signalling pathway leading to the modulation of macrophage activation and pro-inﬂammatory cyto-kine production.45The TLR-2 SNP, T597C, is strongly associated

with both miliary and TB meningitis.43_{Contributing factors to this}

association could be an impaired early, innate pro-inﬂammatory cytokine response,46sub-optimal expression of CD40 and CD8646 or prolonged bacteraemia.43

Cytokines are important host immune effector agents that play a pivotal role in the immunity toM. tuberculosis. One such pleiotropic cytokine is TNF-

a

. Individuals treated with the anti-TNF-

a

agents inﬂiximab, adalimumab, certolizumab pegol and etanercept are at increased risk of TB,47although the risk is greatest with

anti-TNF-a

antibodies, inﬂiximab and adalimumab.48Inﬂiximab has been associated with the development of disseminated TB in humans49,50 and adalimumab lead to severe disseminated TB in a non-human primate TB infection model.51Interestingly, adalimumab had no effect on the overall structure and organisation of the the granu-loma.51The alteration of macrophage function52,53and reduction of CD8þCCR7eCD45RAþeffector memory T-cells54are likely mecha-nisms involved in the promotion of TB disease during TNF neutral-isation treatment.

Bacterial factors also appear to inﬂuence the innate immune response and determine the outcome of infection. Studies of tuberculosis outbreaks have indicated someM. tuberculosisstrains may produce factors which subvert the host innate immune response and augment the ability of the bacteria to replicate and cause disease. For example, macrophages infected with a strain which caused an outbreak in Leicester, UK, were found to produce less protective IL-12p40 and more anti-inﬂammatory IL-10.55A W-Beijing strain, HN878, responsible for an outbreak of tuberculosis in Texas, USA, was found to be hypervirulent in mice and induced low concentrations of TNF-

a

from macrophages.56This phenotype was originally attributed to the production of a speciﬁc phenolic glycolipid (PGL), although it now appears that whilst this molecule can modulate the host cytokine responses it does not itself confer hypervirulence.56,57There is evidence that bacteria from the Beijing lineage, many of which do not produce PGL, are more capable of causing severe, disseminated tuberculosis in humans.58,59 The bacterial factors which determine the different behaviour of the Beijing lineage remain unclear.

The adaptive immune response

The importance of T-cell mediated immunity in the control of

M. tuberculosisinfection and the prevention of bacterial dissemi-nation is illustrated by the strong association between HIV infec-tion and extra-pulmonary tuberculosis,60the protection induced by BCG vaccination against disseminated disease,61 and the clinical consequences of genetic polymorphisms in the IFN-

g

/IL-12 cyto-kine axis.62,63

HIV targets and depletes CD4þT-cells. Humans with advanced HIV infection and low peripheral blood CD4 counts (<50 cells/mm3) are at high risk of persistent bacteraemia withM. tuberculosisand other environmental mycobacteria,64which is powerful evidence for the critical role played by T-cell mediated immunity in the control ofM. tuberculosisdissemination. Mice deﬁcient in MHC-II molecules or CD4þT-cells are highly susceptible toM. tuberculosis

infection, with increased mortality rates, delayed granuloma formation and reduced IFN-

g

expression.65,66

(5)

BCG protects humans from pulmonary tuberculosis is highly vari-able.67 A recent study that investigated BCG induced protective immune responses in mice infected with various strains of

M. tuberculosisfound BCG reduced the bacterial load in the lungs, delayed dissemination of bacteria to the spleen, and was associated with smaller, more mature granulomas in the tissues.68

Abnormalities in the genes encoding the IFN-

g

receptor chain and IL-12 are among the host genetic factors known to be important in determining the susceptibility to disseminated mycobacterial disease. Two clinical studies identiﬁed a mutation in the IFN

g

R1 chain responsible for disseminated environmental mycobacteria and BCG infection.62,69The mutations abolish the expression of the IFN-

g

receptor expression or its ability to bind IFN-

g

.62,63,70 Furthermore, recessive mutations in IL-12 receptor

b

1 chain led to the decreased secretion of IFN-

g

from activated natural killer cells and T-cells.71,72In addition, a mutation found in the gene coding for the signal transducer and activator of transcription-1 (STAT-1) decreased responses in the host to IFN-

g

stimulation.73Mutations in these genes affect IFN-

g

mediated immunity, which is a common downstream signalling pathway of these genes. Furthermore, these naturally occurring mutations highlight the necessity of a T-cell protective immune response in the control of TB disease.

Conclusions

M. tuberculosis remains one of the most successful human pathogens. Dissemination of the bacteria from the primary focus of infection in the lungs, to regional lymph nodes and other organs is one of the key events in tuberculosis pathogenesis. This benefits the pathogen by enabling it to spread to new niches and establish alternative sites of infection but also, as exemplified by studies in mice, dissemination aids in the development of a protective T-cell mediated immune response in the host. The mechanisms of dissemination from the lungs are not well understood, although bacterial genes encoded by the RD1 region as well as HBHA appear to be involved in the intercellular spread and subsequent dissem-ination of mycobacteria. It is evident that genetic differences in the host innate and adaptive immune response can also influence the development of disseminated tuberculosis. Future studies will allow us to determine the immunological and molecular events that govern the ability of genetically different M. tuberculosis

isolates to disseminate and cause various spectrums of clinical disease, as well as identify key points for intervention.

Funding: Wellcome Trust.

Competing interests: None declared.

Ethical approval: Not required.

References

1. Castro-Garza J, King CH, Swords WE, Quinn FD. Demonstration of spread by Mycobacterium tuberculosisbacilli in A549 epithelial cell monolayers.FEMS Microbiol Lett2002;212:145e9.

2. Byrd TF, Green GM, Fowlston SE, Lyons CR. Differential growth characteristics and streptomycin susceptibility of virulent and avirulent Mycobacterium tuberculosisstrains in a novelﬁbroblast-mycobacterium microcolony assay. Infect Immun1998;66:5132e9.

3. Dobos KM, Spotts EA, Quinn FD, King CH. Necrosis of lung epithelial cells during infection withMycobacterium tuberculosisis preceded by cell permeation.Infect Immun2000;68:6300e10.

4. McDonough KA, Kress Y. Cytotoxicity for lung epithelial cells is a viru-lence-associated phenotype ofMycobacterium tuberculosis.Infect Immun1995; 63:4802e11.

5. Gao LY, Guo S, McLaughlin B, Morisaki H, Engel JN, Brown EJ. A mycobacterial virulence gene cluster extending RD1 is required for cytolysis, bacterial spreading and ESAT-6 secretion.Mol Microbiol2004;53:1677e93.

6. Hsu T, Hingley-Wilson SM, Chen B, Chen M, Dai AZ, Morin PM, et al. The primary mechanism of attenuation of bacillus Calmette-Guerin is a loss of

secreted lytic function required for invasion of lung interstitial tissue.Proc Natl Acad Sci U S A2003;100:12420e5.

7. Smith J, Manoranjan J, Pan M, Bohsali A, Xu J, Liu J, et al. Evidence for pore formation in host cell membranes by ESX-1-secreted ESAT-6 and its role inMycobacterium marinumescape from the vacuole.Infect Immun2008;76:5478e87.

8. Kinhikar AG, Verma I, Chandra D, Singh KK, Weldingh K, Andersen P, et al. Potential role for ESAT6 in dissemination ofM. tuberculosisvia human lung epithelial cells.Mol Microbiol; 2009.

9. Hagedorn M, Rohde KH, Russell DG, Soldati T. Infection by tubercular myco-bacteria is spread by nonlytic ejection from their amoeba hosts. Science 2009;323:1729e33.

10. van der Wel N, Hava D, Houben D, Fluitsma D, van Zon M, Pierson J, et al. M. tuberculosis andM. leprae translocate from the phagolysosome to the cytosol in myeloid cells.Cell2007;129:1287e98.

11. Pethe K, Alonso S, Biet F, Delogu G, Brennan MJ, Locht C, et al. The heparin-binding haemagglutinin ofM. tuberculosis is required for extra-pulmonary dissemination.Nature2001;412:190e4.

12. Locht C, Hougardy JM, Rouanet C, Place S, Mascart F. Heparin-binding hemagglutinin, from an extra-pulmonary dissemination factor to a powerful diagnostic and protective antigen against tuberculosis. Tuberculosis (Edinb) 2006;86:303e9.

13. Menozzi FD, Reddy VM, Cayet D, Raze D, Debrie AS, Dehouck MP, et al. Mycobacterium tuberculosisheparin-binding haemagglutinin adhesin (HBHA) triggers receptor-mediated transcytosis without altering the integrity of tight junctions.Microbes Infect2006;8:1e9.

14. Mehta PK, Karls RK, White EH, Ades EW, Quinn FD. Entry and intra-cellular replication ofMycobacterium tuberculosisin cultured human microvascular endothelial cells.Microb Pathog2006;41:119e24.

15. Jain SK, Paul-Satyaseela M, Lamichhane G, Kim KS, Bishai WR.Mycobacterium tuberculosis invasion and traversal across an in vitro human bloodebrain barrier as a pathogenic mechanism for central nervous system tuberculosis. J Infect Dis2006;193:1287e95.

16. Houben EN, Nguyen L, Pieters J. Interaction of pathogenic mycobacteria with the host immune system.Curr Opin Microbiol2006;9:76e85.

17. Nguyen L, Pieters J. The Trojan horse: survival tactics of pathogenic myco-bacteria in macrophages.Trends Cell Biol2005;15:269e76.

18. Bermudez LE, Sangari FJ, Kolonoski P, Petrofsky M, Goodman J. The efﬁciency of the translocation ofMycobacterium tuberculosisacross a bilayer of epithelial and endothelial cells as a model of the alveolar wall is a consequence of transport within mononuclear phagocytes and invasion of alveolar epithelial cells.Infect Immun2002;70:140e6.

19. Davis JM, Ramakrishnan L. The role of the granuloma in expansion and dissemination of early tuberculous infection.Cell2009;136:37e49. 20. Clay H, Davis JM, Beery D, Huttenlocher A, Lyons SE, Ramakrishnan L.

Dichotomous role of the macrophage in earlyMycobacterium marinum infec-tion of the zebraﬁsh.Cell Host Microbe2007;2:29e39.

21. Banchereau J, Briere F, Caux C, Davoust J, Lebecque S, Liu YJ, et al. Immu-nobiology of dendritic cells.Annu Rev Immunol2000;18:767e811.

22. Bodnar KA, Serbina NV, Flynn JL. Fate ofMycobacterium tuberculosiswithin murine dendritic cells.Infect Immun2001;69:800e9.

23. Henderson RA, Watkins SC, Flynn JL. Activation of human dendritic cells following infection withMycobacterium tuberculosis.J Immunol1997;159:635e43. 24. Orme IM, Cooper AM. Cytokine/chemokine cascades in immunity to

tubercu-losis.Immunol Today1999;20:307e12.

25. Tascon RE, Soares CS, Ragno S, Stavropoulos E, Hirst EM, Colston MJ. Myco-bacterium tuberculosis-activated dendritic cells induce protective immunity in mice.Immunology2000;99:473e80.

26. Humphreys IR, Stewart GR, Turner DJ, Patel J, Karamanou D, Snelgrove RJ, et al. A role for dendritic cells in the dissemination of mycobacterial infection. Microbes Infect2006;8:1339e46.

27. Wolf AJ, Linas B, Trevejo-Nunez GJ, Kincaid E, Tamura T, Takatsu K, et al. Mycobacterium tuberculosis infects dendritic cells with high frequency and impairs their function in vivo.J Immunol2007;179:2509e19.

28. van Kooyk Y, Geijtenbeek TB. DC-SIGN: escape mechanism for pathogens.Nat Rev Immunol2003;3:697e709.

29. Tailleux L, Schwartz O, Herrmann JL, Pivert E, Jackson M, Amara A, et al. DC-SIGN is the majorMycobacterium tuberculosisreceptor on human dendritic cells.J Exp Med2003;197:121e7.

30. Geijtenbeek TB, Van Vliet SJ, Koppel EA, Sanchez-Hernandez M, Vanden-broucke-Grauls CM, Appelmelk B, et al. Mycobacteria target DC-SIGN to suppress dendritic cell function.J Exp Med2003;197:7e17.

31. Fu CL, Chuang YH, Huang HY, Chiang BL. Induction of IL-10 producing CD4þT cells with regulatory activities by stimulation with IL-10 gene-modiﬁed bone marrow derived dendritic cells.Clin Exp Immunol2008;153:258e68. 32. Jonuleit H, Schmitt E, Schuler G, Knop J, Enk AH. Induction of interleukin

10-producing, nonproliferating CD4(þ) T cells with regulatory properties by repetitive stimulation with allogeneic immature human dendritic cells.J Exp Med2000;192:1213e22.

33. Chackerian AA, Alt JM, Perera TV, Dascher CC, Behar SM. Dissemination of Mycobacterium tuberculosis is inﬂuenced by host factors and precedes the initiation of T-Cell immunity.Infect Immun2002;70:4501e9.

(6)

35. Wolf AJ, Desvignes L, Linas B, Banaiee N, Tamura T, Takatsu K, et al. Initiation of the adaptive immune response to Mycobacterium tuberculosisdepends on antigen production in the local lymph node, not the lungs. J Exp Med 2008;205:105e15.

36. von Garnier C, Filgueira L, Wikstrom M, Smith M, Thomas JA, Strickland DH, et al. Anatomical location determines the distribution and function of dendritic cells and other APCs in the respiratory tract.J Immunol2005;175:1609e18. 37. Day J, Friedman A, Schlesinger LS. Modeling the immune rheostat of

macro-phages in the lung in response to infection.Proc Natl Acad Sci U S A2009;106: 11246e51.

38. Clough MC. The cultivation of tubercle bacilli from the circulating blood in miliary tuberculosis.Am Rev Tuberc1917;1:598e621.

39. Rich AR, McCordock HA. The pathogenesis of tuberculous meningitis.Bull John Hopkins Hosp1933;52:5e37.

40. Donald PR, Schaaf HS, Schoeman JF. Tuberculous meningitis and miliary tuberculosis: the rich focus revisited.J Infect2005;50:193e5.

41. Ho RS, Fok JS, Harding GE, Smith DW. Host-parasite relationships in experi-mental airborne tuberculosis. VII. Fate ofMycobacterium tuberculosisin primary lung lesions and in primary lesion-free lung tissue infected as a result of bacillemia.J Infect Dis1978;138:237e41.

42. McMurray DN. Hematogenous reseeding of the lung in low-dose, aerosol-infected guinea pigs: unique features of the host-pathogen interface in secondary tubercles.Tuberculosis (Edinb)2003;83:131e4.

43. Thuong NT, Hawn TR, Thwaites GE, Chau TT, Lan NT, Quy HT, et al. A poly-morphism in human TLR2 is associated with increased susceptibility to tuberculous meningitis.Genes Immun2007;8:422e8.

44. Caws M, Thwaites G, Dunstan S, Hawn TR, Lan NT, Thuong NT, et al. The inﬂuence of host and bacterial genotype on the development of disseminated disease withMycobacterium tuberculosis.PLoS Pathog2008;4:e1000034. 45. Quesniaux V, Fremond C, Jacobs M, Parida S, Nicolle D, Yeremeev V, et al.

Toll-like receptor pathways in the immune responses to mycobacteria.Microbes Infect2004;6:946e59.

46. Quesniaux VJ, Nicolle DM, Torres D, Kremer L, Guerardel Y, Nigou J, et al. Toll-like receptor 2 (TLR2)-dependent-positive and TLR2-independent-negative regulation of proinﬂammatory cytokines by mycobacterial Lipomannans. J Immunol2004;172:4425e34.

47. Harris J, Keane J. How tumour necrosis factor blockers interfere with tuber-culosis immunity.Clin Exp Immunol2010;161:1e9.

48. Tubach F, Salmon D, Ravaud P, Allanore Y, Goupille P, Bréban M, et al, Research axed on tolerance of biotherapies group. Risk of tuberculosis is higher with anti-tumor necrosis factor monoclonal antibody therapy than with soluble tumor necrosis factor receptor therapy: the three-year prospective french research axed on tolerance of biotherapies registry.Arthritis Rheum2009;60: 1884e94.

49. Dimakou K, Papaioannides D, Latsi P, Katsimboula S, Korantzopoulos P, Orphanidou D. Disseminated tuberculosis complicating anti-TNF-alpha treat-ment.Int J Clin Pract2004;58:1052e5.

50. Keane J, Gershon S, Wise RP, Mirabile-Levens E, Kasznica J, Schwieterman WD, et al. Tuberculosis associated with inﬂiximab, a tumor necrosis factor alpha-neutralizing agent.N Engl J Med2001;345:1098e104.

51. Lin PL, Myers A, Smith L, Bigbee C, Bigbee M, Fuhrman C, et al. Tumor necrosis factor neutralization results in disseminated disease in acute and latent Mycobacterium tuberculosis infection with normal granuloma structure in a cynomolgus macaque model.Arthritis Rheum2010;62:340e50.

52. Flynn JL, Goldstein MM, Chan J, Triebold KJ, Pfeffer K, Lowenstein CJ, et al. Tumor necrosis factor-[alpha] is required in the protective immune response againstMycobacterium tuberculosisin mice.Immunity1995;2:561e72. 53. Marino S, Myers A, Flynn JL, Kirschner DE. TNF and IL-10 are major factors in

modulation of the phagocytic cell environment in lung and lymph node in tuberculosis: a next-generation two-compartmental model. J Theor Biol 2010;265:586e98.

54. Bruns H, Meinken C, Schauenberg P, Härter G, Kern P, Modlin RL, et al. Anti-TNF immunotherapy reduces CD8þT cell-mediated antimicrobial activity against Mycobacterium tuberculosisin humans.J Clin Invest2009;119:1167e77.

55. Newton SM, Smith RJ, Wilkinson KA, Nicol MP, Garton NJ, Staples KJ, et al. A deletion deﬁning a common Asian lineage of Mycobacterium tuberculosis associates with immune subversion. Proc Natl Acad Sci U S A 2006;103: 15594e8.

56. Reed MB, Domenech P, Manca C, Su H, Barczak AK, Kreiswirth BN, et al. A glycolipid of hypervirulent tuberculosis strains that inhibits the innate immune response.Nature2004;431:84e7.

57. Sinsimer D, Huet G, Manca C, Tsenova L, Koo MS, Kurepina N, et al. The phenolic glycolipid ofMycobacterium tuberculosisdifferentially modulates the early host cytokine response but does not in itself confer hypervirulence.Infect Immun2008;76:3027e36.

58. Kong Y, Cave MD, Zhang L, Foxman B, Marrs CF, Bates JH, et al. Association between Mycobacterium tuberculosis Beijing/W lineage strain infection and extrathoracic tuberculosis: insights from epidemiologic and clinical charac-terization of the three principal genetic groups of M. tuberculosisclinical isolates.J Clin Microbiol2007;45:409e14.

59. Caws M, Thwaites G, Stepniewska K, Nguyen TN, Nguyen TH, Nguyen TP, et al. Beijing genotype ofMycobacterium tuberculosisis signiﬁcantly associated with human immunodeﬁciency virus infection and multidrug resistance in cases of tuberculous meningitis.J Clin Microbiol2006;44:3934e9.

60. Yang Z, Kong Y, Wilson F, Foxman B, Fowler AH, Marrs CF, et al. Identiﬁcation of risk factors for extra-pulmonary tuberculosis. Clin Infect Dis 2004;38: 199e205.

61. Fok JS, Ho RS, Arora PK, Harding GE, Smith DW. Host-parasite relationships in experimental airborne tuberculosis. V. Lack of hematogenous dissemination of Mycobacterium tuberculosis to the lungs in animals vaccinated with Bacille Calmette-Guerin.J Infect Dis1976;133:137e44.

62. Levin M, Newport MJ, D’Souza S, Kalabalikis P, Brown IN, Lenicker HM, et al. Familial disseminated atypical mycobacterial infection in childhood: a human mycobacterial susceptibility gene?Lancet1995;345:79e83.

63. Newport MJ, Huxley CM, Huston S, Hawrylowicz CM, Oostra BA, Williamson R, et al. A mutation in the interferon-gamma-receptor gene and susceptibility to mycobacterial infection.N Engl J Med1996;335:1941e9.

64. Hadad DJ, Palaci M, Pignatari AC, Lewi DS, Machado MA, Telles MA, et al. Mycobacteraemia among HIV-1-infected patients in Sao Paulo, Brazil: 1995 to 1998.Epidemiol Infect2004;132:151e5.

65. Caruso AM, Serbina N, Klein E, Triebold K, Bloom BR, Flynn JL. Mice deﬁcient in CD4 T cells have only transiently diminished levels of IFN-gamma, yet succumb to tuberculosis.J Immunol1999;162:5407e16.

66. Scanga CA, Mohan VP, Yu K, Joseph H, Tanaka K, Chan J, et al. Depletion of CD4 (þ) T cells causes reactivation of murine persistent tuberculosis despite continued expression of interferon gamma and nitric oxide synthase 2.J Exp Med2000;192:347e58.

67. Brewer TF. Preventing tuberculosis with bacillus Calmette-Guerin vaccine: a meta-analysis of the literature.Clin Infect Dis2000;31(Suppl. 3):S64e7. 68. Jeon BY, Derrick SC, Lim J, Kolibab K, Dheenadhayalan V, Yang AL, et al.

Mycobacterium bovisBCG Immunization induces protective immunity against nine different Mycobacterium tuberculosis strains in mice. Infect Immun 2008;76:5173e80.

69. Casanova JL, Blanche S, Emile JF, Jouanguy E, Lamhamedi S, Altare F, et al. Idiopathic disseminated bacillus Calmette-Guerin infection: a French national retrospective study.Pediatrics1996;98:774e8.

70. Jouanguy E, Dupuis S, Pallier A, Dofﬁnger R, Fondaneche MC, Fieschi C, et al. In a novel form of IFN-gamma receptor 1 deﬁciency, cell surface receptors fail to bind IFN-gamma.J Clin Invest2000;105:1429e36.

71. Altare F, Durandy A, Lammas D, Emile JF, Lamhamedi S, Le Deist F, et al. Impairment of mycobacterial immunity in human interleukin-12 receptor deﬁciency.Science1998;280:1432e5.

72. de Jong R, Altare F, Haagen IA, Elferink DG, Boer T, van Breda Vriesman PJ, et al. Severe mycobacterial and Salmonella infections in interleukin-12 receptor-deﬁcient patients.Science1998;280:1435e8.