Detection and Treatment of Subclinical Tuberculosis

Tuberculosis 92 (2012) 447e452

Contents lists available at SciVerse ScienceDirect

Tuberculosis

journal homepage: http://intl.elsevierhealth.com/journals/tube

Detection and treatment of subclinical tuberculosis

Brian D. Robertson a,*, Danny Altmann b, Clif Barry c, Bill Bishai d, Stewart Cole e, Thomas Dick f, Ken Duncan g, Chris Dye h, Sabine Ehrt i, Hanif Esmail j, JoAnne Flynn k, Richard Hafner l, Gray Handley l, Willem Hanekom m, Paul van Helden n, Gilla Kaplan o, Stefan H.E. Kaufmann p, Peter Kim l, Christian Lienhardt h, Valerie Mizrahi m, Eric Rubin q, Dirk Schnappinger i, David Sherman r, Jelle Thole s, Omar Vandal g, Gerhard Walzl n, Digby Warner m, Robert Wilkinson t, Douglas Young u a Rapporteur, Imperial College London, MRC Centre for Molecular Bacteriology and Infection, Flowers Building, South Kensington Campus, London SW7 2AZ, UK b The Wellcome Trust, London, UK c Tuberculosis Research Section, NIAID/NIH, Bethesda, USA d KwaZulu-Natal Research Institute for Tuberculosis and HIV, Durban, South Africa e École Polytechnique Fédérale De Lausanne, Switzerland f Yong Loo Lin School of Medicine, National University of Singapore, Singapore, g Bill and Melinda Gates Foundation, Seattle, USA h Stop TB Department, WHO, Geneva, Switzerland i Weill Cornell Medical College, New York, USA j Imperial College London, UK k University of Pittsburgh School of Medicine, USA l NIH/NIAID, Bethesda, USA m University of Cape Town, South Africa n Stellenbosch University, South Africa o Public Health Research Institute Center, Newark, USA p Max Planck Institute for Infection Biology, Berlin, Germany q Harvard School of Public Health, USA r Seattle Biomedical Research Institute, Seattle, USA s Tuberculosis Vaccine Initiative, Lelystad, The Netherlands t Imperial College London, MRC National Institute for Medical Research, UK and University of Cape Town, South Africa u Chair, MRC National Institute for Medical Research, and Imperial College London, UK article info summary

Article history: Reduction of active disease by preventive therapy has the potential to make an important contribution Received 24 June 2012 towards the goal of tuberculosis (TB) elimination. This report summarises discussions amongst a Working Accepted 26 June 2012 Group convened to consider areas of research that will be important in optimising the design and delivery of preventative therapies. The Working Group met in Cape Town on 26th February 2012, Keywords: following presentation of results from the GC11 Grand Challenges in Global Health project to discover Detection drugs for latent TB. Treatment Ó 2012 Elsevier Ltd. All rights reserved. Subclinical Tuberculosis

1. The 2050 elimination goal post-second world war Europe and North America. The current global control strategy is guided by Millennium Development Goal 6, The Stop TB Partnership has set the goal of tuberculosis elimi- which is to reduce the incidence of cases by 2015, and the Stop TB nation (defined as an annual incidence of fewer than one case per Partnership target of halving 1990 prevalence and mortality rates by million individuals) by 2050.1 Although this is an ambitious goal in 2015. Historically, control efforts have prioritised identification and the context of global trends over recent decades, the required rate treatment of all cases of active TB, but this has been insufficient to of decline is broadly in line with successful TB control achieved in interrupt transmission and there is a need to combine the existing approach with new interventions that will reduce the development of infectious cases by vaccination and preventive therapy.2 The * Corresponding author. Tel.: þ44 20 7594 3198; fax: þ44 20 7594 3095. framework chosen by the United Nations to replace the Millennium 3 E-mail address: [email protected] (B.D. Robertson). Development Goals after 2015 will be critical for TB control.

Research is an essential driver towards TB elimination,4 and the aim a novel approach with potential application in this context.13,14 The of the present document is to outline a research strategy that will third could be novel imaging modalities that reflect bacterial load optimise the impact of therapeutic intervention during the phase of and early disease activity, which may provide important tools to infection prior to development of clinical TB. track hostepathogen interactions and to discover biomarkers for Latent tuberculosis infection (LTBI) is currently defined by different stages of infection.5 With an increasing knowledge of immunological sensitisation to Mycobacterium tuberculosis anti- systems immunology, it may be possible to identify immunological gens in the absence of clinical symptoms of disease. The relation- biomarkers that are precursors of future pathology.15,16 In order to ship of the immune response to the continued presence of live achieve the desired levels of specificity it may be necessary to bacilli is unclear. This description is applicable to one third of the combine multiple biomarkers, using mathematical modelling global population and encompasses a diverse spectrum of indi- approaches to determine the potential utility of combining tests viduals at widely differing risk of developing active TB.5 While the with lower levels of sensitivity and specificity in point-of-care tests. broad concept of a clinically “latent” phase of infection provides the Finally, we should anticipate that, as the prevalence of infection rationale underlying preventive therapy, effective implementation declines as TB control moves towards the elimination goal, we may will depend on stratification of individuals on the basis of disease need to translate increasingly sophisticated diagnostic tools from risk rather than the presence or absence of an immune response. research into practical application. We propose to use the term “subclinical tuberculosis” to denote at- In summary, in order to identify individuals who would receive risk populations who would receive the greatest benefit from maximum benefit from preventive therapy, we need to identify preventive therapy. novel biomarkers, or combinations thereof, that provide informa- There is extensive evidence that preventive therapy can reduce tion about: the incidence of future disease, and current WHO guidelines strongly recommend that HIV-infected persons in whom active TB Bacterial load has been excluded should receive isoniazid for at least 6 months, Early pathology and preferably 9 months, as part of a comprehensive package of HIV Phenotypic state of antigen-specific T cells care.6 Addition of rifapentine allows an important reduction to a three-month treatment schedule7 in HIV-negative individuals, 3. Bacterial physiology though the risk of reinfection imposes an obvious limitation on the utility of preventive therapy in areas with high rates of trans- An intuitive expectation is that clearance of the small numbers mission.8 Current preventive therapy involves prolonged treatment of bacteria present prior to disease onset should require less of large populations in order to prevent a relatively small number of extensive treatment than active disease. The need for prolonged cases. For tractable integration into global strategies for TB control, preventive therapy is thought to reflect either sequestration of we need short, safe preventive treatments targeted at high-risk bacteria in sites of low drug penetration, or transient periods of populations. persistence involving a physiological state that confers phenotypic The overall strategic requirements are: tolerance.5 Phenotypic tolerance can be induced in a variety of in vitro culture systems in which bacterial replication is inhibited by Strategies to identify groups with subclinical TB at high risk of starvation, oxygen deprivation, or other means, and it is likely that developing symptomatic infectious disease such systems are broadly reflective of the in vivo situation. Short, simple therapeutic regimens to treat subclinical TB and Stochastic persisters that express a drug-tolerant phenotype inde- prevent the development of active disease pendent of environmental signals have also been observed within Integration of preventive therapy with pre- and post-exposure an otherwise logarithmically-growing population in vitro, although vaccination strategies their relevance to in vivo infection is unclear.17,18 Non-replicating persistence under oxygen-starved conditions has been well- 2. Widening the diagnostic net characterised, and the observation that metronidazole e a drug that has in vitro activity only under hypoxic conditions e can be The effectiveness of current control efforts is limited by applied successfully as preventive therapy in a nonhuman primate incomplete recruitment of infectious cases, thereby limiting their model19 provides compelling evidence that hypoxia contributes to impact on TB transmission. Prevalence surveys have shown that an latent TB. Phenotypic tolerance is also observed during active important proportion of infectious individuals with bacteriologi- disease, and the distinction between subclinical infection and cally positive sputum do not experience symptoms of sufficient active TB is best viewed as a quantitative difference in distribution e severity to prompt health-seeking behaviour.9 11 In the context of of bacterial phenotype and load rather than as a strict qualitative HIV co-infection in high burden settings, the prevalence of difference between replicating and non-replicating states.5 A asymptomatic culture-positive TB can be as high as 8.5%.12 consequence of this line of reasoning is that the targeting of Improved recruitment of prevalent cases will require expansion phenotypically tolerant populations is of equal importance for of the diagnostic net by increased social awareness of early symp- improved preventive therapy and for shortening regimens applied toms as well as the incorporation of more flexible diagnostic tests in the treatment of active disease. into health care programmes. While technical advances, particularly in the area of tran- Three broad criteria can be proposed as potential early markers scriptomics, have allowed characterisation of M. tuberculosis in of disease risk. The first is bacterial load. Potential biomarkers mouse tissues and in human sputum,20,21 there is very little include detection of bacterial components in systemic fluids, information about the physiological state of bacteria within human detection of antibodies, and detection of M. tuberculosis-specificT lesions. Studies of mycobacteria isolated directly from human cell responses based on novel phenotypic markers. These strategies sputum surprisingly revealed that the organisms are phenotypi- are being explored in the context of definitive diagnosis of active cally and transcriptionally similar to bacteria adapted to hypoxia disease, but may have alternative application as indicators of future in vitro.20 This conflicts with the conventional view of the origin of risk. A second approach could be based on molecular markers of sputum-borne bacilli as emerging from rapidly replicating bacterial early pathology; for example, metabolic or proteomic disturbances populations at the cavity surface. Also contrary to conventional related to lung damage. Blood-based transcriptional profiling offers wisdom, in sputum (and lesions including cavities) the bacilli are B.D. Robertson et al. / Tuberculosis 92 (2012) 447e452 449 not replicating within macrophages but are primarily extracellular understanding of downstream consequences, rather than on simple or within neutrophils.22 The extensive heterogeneity of cell types genetic evidence of essentiality, is likely to increase the produc- and architecture that make up human TB lesions suggests a diver- tivity of target-based discovery. The availability of medicinal sity of microenvironments that are likely to induce a corresponding chemistry resources and the early use of ‘proof of concept’ or tool heterogeneity in bacterial phenotypes.5 Understanding of myco- compounds in vitro and in appropriate animal models, will be key bacterial physiology in vivo will depend on our ability to interpret in elucidating issues associated with permeability, efflux and target sparse experimental data using systems biology models that inte- cell metabolism. grate transcriptomic, proteomic, lipidomic and metabolomic In the mouse model, combined administration of novel combi- networks.23 This should be complemented by the use of appro- nations such as TMC207 and pyrazinamide have a striking synergy priate experimental models in which methods not easily applied to that allows a significant reduction in the treatment time required humans, such as reporter bacteria and tool compounds to assess for sterilisation.30 These empirical findings highlight the impor- drug penetration and lesion targeting, can be exploited. tance of screening for positive and negative interactions between A characteristic of bacteria in the non-replicating hypoxia model old and new drugs in novel combinations and dosages. Under- is that they survive with a marked reduction in their energy standing which model systems or in vitro tests predict synergies reserves.24 Further reduction of low ATP levels by targeting of ATP that might translate into therapies for human disease will require synthase provides an attractive rationale for the ability of TMC207 significantly more effort. Consideration of combined activities and to kill non-replicating M. tuberculosis.25 Recent evidence suggests synthetic lethality could usefully be incorporated into drug that pyrazinamide, another drug with activity against non- discovery strategies, and early stage screening assays should be replicating populations, disrupts processes involved in trans- developed that mimic in vivo conditions to identify candidates such lation,26 highlighting this as another potential area for drug as pro-drugs with poor in vitro activity. discovery. To accelerate the drug discovery pipeline, there is a need to Improved understanding of the physiology of M. tuberculosis in develop: human tissues will provide an important foundation for the rational design of novel drugs and regimens for preventive therapy Strategies for target selection based on an improved under- and for treatment of active TB. This will depend on: standing of the cascade of biochemical events triggered by the initial drugetarget interaction Technologies that provide direct experimental information Molecular assays for mycobacterial cell death that can be used about bacteria in lesions from patients to monitor realetime activity of drugs and prioritize hits The development of experimental models that accurately Structure-based understanding of penetration and accumula- recapitulate the clinical efficacy of drugs tion of small molecules in tissues, lesions and bacteria Integrative systems biology models that allow reconstruction Understanding of drugedrug interactions and mechanisms of bacterial physiology from partial datasets

4. Drug discovery 5. Moving drugs to the clinic

Renewed efforts to discover novel anti-mycobacterial drugs over The clinical development of new antimycobacterials typically the last decade have followed two approaches. The first is based on involves evaluation in a mouse model followed by assessment of the screening of small molecule libraries for compounds with the early bactericidal activity in humans. In the conventional mouse ability to kill mycobacteria, followed by identification of molecular model, disease is associated predominantly with the presence of targets using genetic tools such as re-sequencing of resistant intracellular bacteria in loosely structured lesions. This represents mutants. The second approach involves the initial application of only a portion of human pathology; in particular, it does not reca- genetic tools to identify essential enzymes that are then used as pitulate the heterogeneous microenvironments and bacterial targets in high-throughput biochemical screens. The first approach phenotypes characteristic of human granulomas or cavities. Early has been more successful, with TMC207 and novel nitroimidazole bactericidal activity (EBA) measures the impact of drugs on bacteria derivatives27 in early stage clinical development for drug-resistant that contribute to sputum positivity and transmission, but provides and drug-susceptible TB, and both showing promising activity no information concerning their effect on closed lesions that are against non-replicating phenotypes. However, whole cell screens likely to contribute to the requirement for prolonged therapy; generate only a small number of potent hits, along with a large measuring EBA over extended times may help to address this number of hits with moderate potency. To exploit the latter cate- limitation. Thus, while it can be anticipated that mainstream drug gory of compounds, bacterial cell death surrogates are needed discovery efforts will generate new compounds with potential which are more tractable and informative than measurement of application to preventive therapy, many opportunities will be colony forming units, and which can be applied in screens that missed without alternative late preclinical and early clinical better model in vivo mycobacterial physiology to prioritize screens. compounds for follow-up by medicinal chemistry. The low-dose non-human primate model reproduces key The failure of target-based drug discovery can be rationalised in features of latent TB in humans (reactivation by anti-TNF anti- part by the difficulty of moving from hits with potent activity bodies, for example) and has been used to compare drug activities against purified enzymes to molecules that efficiently penetrate the during preventive therapy.31 The model is expensive but, by mycobacterial cell envelope. A second limitation is that the incorporating live imaging readouts, provides high value data that conventional model, based on the assumption that inhibition of an closely parallel clinical studies. Extension from cynomolgus individual metabolic reaction will cause cell death, may not be macaques to smaller primates, such as marmosets, would make the a complete description of the mode of action of effective anti- model more tractable by reducing the amount of drug that is bacterials.28 Bactericidal activity is likely to depend on the down- required for testing. stream consequences of an initial drugetarget interaction which Serial live imaging by combined computed tomography (CT) and results in the generation of toxic intermediates that act as the true positron emission tomography (PET) provides a comprehensive effector molecules.29 The selection of targets based on an analysis of the dynamics of TB lesions in patients and experimental 450 B.D. Robertson et al. / Tuberculosis 92 (2012) 447e452 animals.5 Monitoring changes to individual lesions in response to Potential mechanisms include senescence or exhaustion of key T cell drug exposure complements early bactericidal activity by dis- populations, or an imbalance between activating and regulatory tinguishing lesions that resolve quickly or slowly in response to circuits that control immune homeostasis and the balance between conventional or novel treatment regimens. Evidence from prelim- inflammation and antimicrobial mechanisms.38 Aseriesofbooster inary studies shows that lesions can be imaged in individuals with vaccines that enhance the CD4 and CD8 T cell responses that are latent TB, suggesting that PET/CT could be used to define rapid dominant in the natural response to infection are currently in clinical endpoints for early clinical trials of preventive therapy regimens. trials to assess their prophylactic efficacy in combination with Host markers may be a simple way to infer bacterial load as the host BCG.38,39 Initial safety trials have found no detrimental impact responds to this assault, and to monitor the decline in load during associated with delivery of these vaccines to individuals with effective therapy; imaging studies may aid in the discovery of such evidence of prior immune sensitisation, and plans are underway to markers. test their ability to reduce reactivation or reinfection disease. Given To evaluate the potential role of novel drug regimens in the likely complexity of the regulatory networks associated with preventive therapy, it will be necessary: existing immune responses, an intriguing alternative vaccine strategy would be to induce a de novo immune response, for example To extend the use of experimental models that mimic human by targeting antigens that are not well-recognised during the natural subclinical TB immune process40; improved phenotypic characterisation of myco- To exploit live imaging technologies to monitor effectiveness of bacterial subpopulations in human lesions has the potential to drugs in preclinical models and in clinical studies contribute to such an approach. The application of mathematical To identify biomarkers that reflect the declining bacterial load modelling to these immunological systems may provide insights in response to effective therapy that can assist in the prioritization of proposed interventions.41 Opportunities to combat subclinical TB by immune modulation include: 6. Immune modulation Evaluation of current booster vaccines in populations with Modulation of the host response presents an important adjunct subclinical TB or alternative alongside preventive therapy targeting the pathogen. Development of vaccines targeting novel antigens and/or While there are well-characterised examples of immune modulation immune mechanisms that enhance progression to active disease e the use of anti-TNF Development of immunomodulatory vaccines that alter antibodies, HIV-mediated depletion of CD4 cells e there is no clear pathology, prevent disease progression, or achieve sterile proof-of-concept for beneficial immune modulation during pathogen eradication subclinical TB. If protective immunity requires a balanced regulation Molecular definition of immune mechanisms of mycobacterial of multiple immune functions, it would be misleading to assume killing that a negative impact associated with reduction in one component Development of robust, well characterised preclinical models should necessarily imply a positive outcome were this component of subclinical TB for comparative assessment of vaccines and enhanced.16 For example, while the absence of IFNg is detrimental to treatment modalities protection, increasing amounts of IFNg will not necessarily correlate with increased protection. There is some evidence that combining antiretroviral therapy and isoniazid preventative therapy has 7. Coordination and networking synergistic benefits in preventing TB in HIV-infected people.32 Immune reconstitution in HIV-TB patients with a higher bacterial Despite the substantial global population of individuals with load demonstrates the pathogenic potential of such an increase in subclinical TB, there are disturbingly few experimental techniques immune response, in the form of immune reconstitution inflam- that generate interpretable data concerning their immune status or matory syndrome (IRIS).33 Therefore, the study of pathological and the physiological condition of the bacteria that they are harbouring. protective immune restoration in TB-infected patients during anti- There is very limited access to material from subclinical TB lesions, retroviral therapy provides fertile ground to understand the differ- and experimental models that reflect the human condition are ence between protective and pathological immunity.34 expensive and available only in highly specialised laboratories. To In principle, improved control of subclinical TB could be ach- promote effective research in this area, there is a need to support ieved by induction of an enhanced immune response to clear per- research networks that link availability of samples from clinical and sisting bacteria, or by boosting the existing immune response to experimental infection with expertise in novel technologies, and to maintain its containment function. Clearance of persisting infection enhance partnerships between clinicians, experimentalists and might be achieved by transient stimulation of innate effector epidemiologists. Clinical trial sites need baseline support to main- functions, using small molecules to modulate intracellular path- tain their operational capacity, and long-term cohorts to support ways or by blocking of regulatory cytokines or co-stimulatory surveillance and related studies, such as biomarker discovery and molecules.35,36 An obvious concern with such an approach is the validation. Addressing the logistic constraints associated with risk that systemic activation of innate immune mechanisms may implementation of disease control programmes in often poorly result in collateral pathological damage. In contrast, it may be resourced societies will be critical to the success of any new possible to induce targeted dampening of innate immune cell intervention to combat subclinical TB; notably, this offers an function to reactivate quiescent lesions by transient immune important opportunity to improve social aspects of TB control such modulation.37 This “wake ‘em and whack ‘em” strategy would have as access to drugs and health care, as well as facilitating operational to be carefully coordinated with delivery of anti-mycobacterial research. The reduction of disease by preventive therapy has the drugs, and may have deleterious consequences in individuals potential to make a significant contribution to the goal of TB with high bacterial load or harbouring a drug resistant strain. elimination. This is not a simple undertaking, and will require The rational design of strategies to boost existing immune cover a vigorous response to the challenge of implementing a short, sharp is frustrated by a lack of understanding of pathways (other than HIV intervention that is targeted to at-risk groups and can be integrated co-infection) that lead to immune breakdown and reactivation. within the TB elimination strategy. B.D. Robertson et al. / Tuberculosis 92 (2012) 447e452 451

This paper is the outcome of a workshop held in February 2012 4. Lienhardt C, Espinal M, Pai M, Maher D, Raviglione MC. What research is in South Africa. The paper captures the consensus views and needed to stop TB? Introducing the TB research movement. PLoS Med 2011;8:e1001135. opinions of the participants and is not an official statement from 5. Barry 3rd CE, Boshoff HI, Dartois V, Dick T, Ehrt S, Flynn J, Schnappinger D, the participants’ host organizations. Wilkinson RJ, Young D. The spectrum of latent tuberculosis: rethinking the biology and intervention strategies. Nat Rev Microbiol 2009;7:845e55. 6. WHO. WHO policy on collaborative TB/HIV activities. In: Guidelines for national Working Group participants programmes and other stakeholders Geneva 2012. 7. Sterling TR, Villarino ME, Borisov AS, Shang N, Gordin F, Bliven-Sizemore E, Hackman J, Hamilton CD, Menzies D, Kerrigan A, Weis SE, Weiner M, Wing D, Brian D. Robertson (Rapporteur, Imperial College London, UK) Conde MB, Bozeman L, Horsburgh Jr CR, Chaisson RE. Three months of rifa- Danny Altmann (The Wellcome Trust, London, UK) pentine and isoniazid for latent tuberculosis infection. N Engl J Med Clif Barry (Tuberculosis Research Section, NIAID/NIH, Bethesda, 2011;365:2155e66. 8. Smart T. Thibela TB: community-wide IPT did not improve TB control in South USA) Africa’s mines at the population level 2012. Bill Bishai (KwaZulu-Natal Research Institute for Tuberculosis 9. Ayles H, Schaap A, Nota A, Sismanidis C, Tembwe R, De Haas P, Muyoyeta M, and HIV, Durban, South Africa) Beyers N. Prevalence of tuberculosis, HIV and respiratory symptoms in two Stewart Cole (École Polytechnique Fédérale De Lausanne, Zambian communities: implications for tuberculosis control in the era of HIV. PLoS One 2009;4:e5602. Switzerland) 10. Corbett EL, Bandason T, Duong T, Dauya E, Makamure B, Churchyard GJ, Thomas Dick (Yong Loo Lin School of Medicine, National Williams BG, Munyati SS, Butterworth AE, Mason PR, Mungofa S, Hayes RJ. fi University of Singapore) Comparison of two active case- nding strategies for community-based diagnosis of symptomatic smear-positive tuberculosis and control of infectious Ken Duncan (Bill and Melinda Gates Foundation, Seattle, USA) tuberculosis in Harare, Zimbabwe (DETECTB): a cluster-randomised trial. Chris Dye (Stop TB Department, WHO, Geneva, Switzerland) Lancet 2010;376:1244e53. Sabine Ehrt (Weill Cornell Medical College, New York, USA) 11. Hoa NB, Sy DN, Nhung NV, Tiemersma EW, Borgdorff MW, Cobelens FG. National survey of tuberculosis prevalence in Viet Nam. Bull World Health Hanif Esmail (Imperial College London, UK) Organ 2010;88:273e80. JoAnne Flynn (University of Pittsburgh School of Medicine, USA) 12. Oni T, Burke R, Tsekela R, Bangani N, Seldon R, Gideon HP, Wood K, Richard Hafner (NIH/NIAID, Bethesda, USA) Wilkinson KA, Ottenhoff TH, Wilkinson RJ. High prevalence of subclinical tuberculosis in HIV-1-infected persons without advanced immunodeficiency: Gray Handley (NIH/NIAID, Bethesda, USA) implications for TB screening. Thorax 2011;66:669e73. Willem Hanekom (University of Cape Town, South Africa) 13. Maertzdorf J, Repsilber D, Parida SK, Stanley K, Roberts T, Black G, Walzl G, Paul van Helden (Stellenbosch University, South Africa) Kaufmann SH. Human gene expression profiles of susceptibility and resistance in tuberculosis. Genes Immun 2011;12:15e22. Gilla Kaplan (Public Health Research Institute Center, Newark, 14. Berry MP, Graham CM, McNab FW, Xu Z, Bloch SA, Oni T, Wilkinson KA, USA) Banchereau R, Skinner J, Wilkinson RJ, Quinn C, Blankenship D, Dhawan R, Stefan H.E. Kaufmann (Max Planck Institute for Infection Cush JJ, Mejias A, Ramilo O, Kon OM, Pascual V, Banchereau J, Chaussabel D, ’ Biology, Berlin, Germany) O Garra A. An interferon-inducible neutrophil-driven blood transcriptional signature in human tuberculosis. Nature 2010;466:973e7. Peter Kim (NIH/NIAID, Bethesda, USA) 15. Germain RN, Meier-Schellersheim M, Nita-Lazar A, Fraser ID. Systems biology Christian Lienhardt (Stop TB Department, WHO, Geneva, in immunology: a computational modeling perspective. Annu Rev Immunol e Switzerland) 2011;29:527 85. 16. Ritz N, Dutta B, Donath S, Casalaz D, Connell TG, Tebruegge M, Robins- Valerie Mizrahi (University of Cape Town, South Africa) Browne R, Hanekom WA, Britton WJ, Curtis N. The influence of bacille Eric Rubin (Harvard School of Public Health, USA) Calmette-Guerin vaccine strain on the immune response against tuberculosis: e Dirk Schnappinger (Weill Cornell Medical College, New York, a randomized trial. Am J Respir Crit Care Med 2012;185:213 22. 17. Gengenbacher M, Kaufmann SH. Mycobacterium tuberculosis: success through USA) dormancy. FEMS Microbiol Rev 2012;36:514e32. David Sherman (Seattle Biomedical Research Institute, Seattle, 18. Keren I, Minami S, Rubin E, Lewis K. Characterization and transcriptome USA) analysis of Mycobacterium tuberculosis persisters. MBio 2011;2:e00100e11. 19. Flynn JL. Unpublished data. Jelle Thole (TuBerculosis Vaccine Initiative, Lelystad, The 20. Garton NJ, Waddell SJ, Sherratt AL, Lee SM, Smith RJ, Senner C, Hinds J, Netherlands) Rajakumar K, Adegbola RA, Besra GS, Butcher PD, Barer MR. Cytological and Omar Vandal (Bill and Melinda Gates Foundation, Seattle, USA) transcript analyses reveal fat and lazy persister-like bacilli in tuberculosis sputum. PLoS Med 2008;5:e75. Gerhard Walzl (Stellenbosch University, South Africa) 21. Schnappinger D, Ehrt S, Voskuil MI, Liu Y, Mangan JA, Monahan IM, Dolganov G, Digby Warner (University of Cape Town, South Africa) Efron B, Butcher PD, Nathan C, Schoolnik GK. Transcriptional adaptation of Robert Wilkinson (Imperial College London, MRC National Mycobacterium tuberculosis within macrophages: insights into the phagosomal e Institute for Medical Research, UK and University of Cape Town, environment. J Exp Med 2003;198:693 704. 22. Eum SY, Kong JH, Hong MS, Lee YJ, Kim JH, Hwang SH, Cho SN, Via LE, South Africa) Barry 3rd CE. Neutrophils are the predominant infected phagocytic cells in the Douglas Young (Chair, MRC National Institute for Medical airways of patients with active pulmonary TB. Chest 2010;137:122e8. Research, and Imperial College London, UK) 23. Robertson BD, Wren BW, editors. Systems microbiology: current topics and applications. Norfolk: Caister Academic Press; 2012. 24. Rao SP, Alonso S, Rand L, Dick T, Pethe K. The protonmotive force is required for Ethical approval: Not required. maintaining ATP homeostasis and viability of hypoxic, nonreplicating Myco- bacterium tuberculosis. Proc Natl Acad Sci U S A 2008;105:11945e50. 25. Andries K, Verhasselt P, Guillemont J, Gohlmann HW, Neefs JM, Winkler H, Van Funding: The GC11 Grand Challenges in Global project was Gestel J, Timmerman P, Zhu M, Lee E, Williams P, de Chaffoy D, Huitric E, Hoffner S, Cambau E, Truffot-Pernot C, Lounis N, Jarlier V. A diarylquinoline jointly funded by the Bill and Melinda Gates Foundation (37882) drug active on the ATP synthase of Mycobacterium tuberculosis. Science and The Wellcome Trust (077381). 2005;307:223e7. 26. Shi W, Zhang X, Jiang X, Yuan H, Lee JS, Barry 3rd CE, Wang H, Zhang W, Zhang Y. Pyrazinamide inhibits trans-translation in Mycobacterium tuberculosis. Competing interests: None declared. Science 2011;333:1630e2. 27. Gler MT, Skripconoka V, Sanchez-Garavito E, Xiao H, Cabrera-Rivero JL, Vargas- Vasquez DE, Gao M, Awad M, Park SK, Shim TS, Suh GY, Danilovits M, Ogata H, References Kurve A, Chang J, Suzuki K, Tupasi T, Koh WJ, Seaworth B, Geiter LJ, Wells CD. Delamanid for multidrug-resistant pulmonary tuberculosis. N Engl J Med 1. Stop-TB-Partnership. The global plan to stop TB 2011e2015. 2012;366:2151e60. 2. Dye C, Williams BG. The population dynamics and control of tuberculosis. 28. Dick T, Young D. How antibacterials really work: impact on drug discovery. Science 2010;328:856e61. Future Microbiol 2011;6:603e4. 3. Editorial. Global health in 2012: development to sustainability. Lancet 29. Singh R, Manjunatha U, Boshoff HI, Ha YH, Niyomrattanakit P, Ledwidge R, 2012;379. Dowd CS, Lee IY, Kim P, Zhang L, Kang S, Keller TH, Jiricek J, Barry 3rd CE. PA- 452 B.D. Robertson et al. / Tuberculosis 92 (2012) 447e452

824 kills nonreplicating Mycobacterium tuberculosis by intracellular NO release. growth is controlled by a kinase network around PKB/AKT1. Nature Science 2008;322:1392e5. 2007;450:725e30. 30. Tasneen R, Li SY, Peloquin CA, Taylor D, Williams KN, Andries K, Mdluli KE, 36. Tobin DM, Roca FJ, Oh SF, McFarland R, Vickery TW, Ray JP, Ko DC, Zou Y, Nuermberger EL. Sterilizing activity of novel TMC207- and PA-824-containing Bang ND, Chau TT, Vary JC, Hawn TR, Dunstan SJ, Farrar JJ, Thwaites GE, regimens in a murine model of tuberculosis. Antimicrob Agents Chemother King MC, Serhan CN, Ramakrishnan L. Host genotype-specific therapies can 2011;55:5485e92. optimize the inflammatory response to mycobacterial infections. Cell 31. Lin PL, Rodgers M, Smith L, Bigbee M, Myers A, Bigbee C, Chiosea I, Capuano SV, 2012;148:434e46. Fuhrman C, Klein E, Flynn JL. Quantitative comparison of active and latent 37. Subbian S, Tsenova L, O’Brien P, Yang G, Koo MS, Peixoto B, Fallows D, tuberculosis in the cynomolgus macaque model. Infect Immun 2009;77:4631e42. Dartois V, Muller G, Kaplan G. Phosphodiesterase-4 inhibition alters gene 32. Golub JE, Chaisson RE, Martinson NA. Additive effects of isoniazid preventive expression and improves isoniazid-mediated clearance of Mycobacterium therapy and HAART. AIDS 2009;23:1446e7. tuberculosis in rabbit lungs. PLoS Pathog 2011;7:e1002262. 33. Meintjes G, Rabie H, Wilkinson RJ, Cotton MF. Tuberculosis-associated immune 38. Kaufmann SH. Future vaccination strategies against tuberculosis: thinking reconstitution inflammatory syndrome and unmasking of tuberculosis by outside the box. Immunity 2010;33:567e77. antiretroviral therapy. Clin Chest Med 2009;30:797e810 [x]. 39. Kupferschmidt K. Infectious disease. Taking a new shot at a TB vaccine. Science 34. Wilkinson KA, Seldon R, Meintjes G, Rangaka MX, Hanekom WA, Maartens G, 2011;334:1488e90. Wilkinson RJ. Dissection of regenerating T-cell responses against tuberculosis 40. Aagaard C, Hoang T, Dietrich J, Cardona PJ, Izzo A, Dolganov G, Schoolnik GK, in HIV-infected adults sensitized by Mycobacterium tuberculosis. Am J Respir Crit Cassidy JP, Billeskov R, Andersen P. A multistage tuberculosis vaccine that Care Med 2009;180:674e83. confers efficient protection before and after exposure. Nat Med 35. Kuijl C, Savage ND, Marsman M, Tuin AW, Janssen L, Egan DA, Ketema M, van 2011;17:189e94. den Nieuwendijk R, van den Eeden SJ, Geluk A, Poot A, van der Marel G, 41. Young D, Stark J, Kirschner D. Systems biology of persistent infection: tuber- Beijersbergen RL, Overkleeft H, Ottenhoff TH, Neefjes J. Intracellular bacterial culosis as a case study. Nat Rev Microbiol 2008;6:520e8.