Download Eliminating latent tuberculosis - Institute of Infectious Disease and

TIMI-633; No of Pages 6
Opinion
Eliminating latent tuberculosis
Douglas B. Young1,2, Hannah P. Gideon3 and Robert J. Wilkinson1,3,4
1
National Institute for Medical Research, Mill Hill, London, NW7 1AA, UK
Division of Investigative Science, Imperial College London, SW7 2AZ, UK
3
Institute of Infectious Diseases and Molecular Medicine, University of Cape Town, Cape Town 7925, South Africa
4
Division of Medicine, Imperial College London, W2 1PG, UK
2
Tuberculosis is recognized as the world’s leading bacterial cause of death. Yet 95% of infection is believed to
exist in an asymptomatic ‘latent’ form that is defined not
by the identification of bacteria, but by the host immune
response in the form of reactivity to tuberculosis
proteins in the tuberculin skin test. It seems likely that
clinically defined latent tuberculosis actually represents
a spectrum that runs from elimination of live bacilli to
subclinical disease: hence, it might be unhelpful to use a
single term to describe all these conditions. To support
this view, here we focus on recent increased understanding of the heterogeneity in both bacillary physiology and
host immune response that potentially illuminates new
therapeutic and diagnostic approaches to this condition.
Tuberculosis: a persistent challenge
Tuberculosis (TB) is a leading bacterial cause of death: in
2006, there were 9.2 million new cases and an estimated
1.5 million people died [1]. An increasing proportion of
disease is caused by organisms insensitive to first line
chemotherapeutic agents (multi-drug resistant TB
[MDR-TB]) [2]. Furthermore, it is estimated that one-third
of the world’s population harbours drug sensitive, and
perhaps increasingly drug resistant, TB in the form of
latent infection [3]. Thus, understanding latent TB infection is an important question: even if all cases of active
disease could be detected and successfully treated, an
immense future reservoir of disease exists.
Mycobacterium tuberculosis (the bacterium causing the
disease) is transmitted as a highly infectious aerosol from
the lungs of individuals with active disease (Figure 1). An
encounter with M. tuberculosis is classically regarded to
give rise to three outcomes. First, a minority of perhaps 5%
(higher in infants or immunocompromised persons)
develop rapidly progressive disease that is sometimes
confirmed by isolation of M. tuberculosis from clinical
samples (evidence of bacterial replication) and referred
to as ‘active’ primary TB. Although an acquired immune
response is almost invariably detectable, in these circumstances it is manifestly incompletely effective. By contrast,
the majority of infected persons show no disease symptoms
but develop an effective acquired immune response – they
are referred to as having ‘latent’ TB, on the basis of a
positive result in either a tuberculin skin test (TST), or an
in vitro assay of T-cell function (interferon gamma release
assay, IGRA [4]). As a third possible outcome, a proportion
of those ‘latently’ infected (maybe 1%–2% per year) might
Corresponding author: Young, D.B. ([email protected]).
reactivate infection and develop the disease (known then
as ‘postprimary’ TB). Its onset can also be triggered by
immunosuppression of which the greatest identified cause
is HIV-1 infection [5].
Although clearly a useful clinical concept, we consider
that the current view of ‘latent’ TB as a single homogeneous entity is inadequate and potentially limits the
basis for the rational development of drugs, vaccines and
associated biomarkers. Here, we suggest that this concept
needs to be re-assessed and perhaps replaced (or complemented) by alternative terminologies that more accurately
demarcate important biological features and thereby
research goals. We expect that the new conceptual framework will facilitate further research towards improved
detection methods and treatments for latent TB.
Rethinking ‘latent’ TB
What is the basis for considering most TB infection in the
world to be latent? By analogy, until the mid 1990s, Human
Immunodeficiency Virus (HIV-1) infection was characterized as having three phases. After acute seroconversion
there was resolution to a minimally symptomatic ‘latent’
phase. Then, after a variable interval of 4–15 years and for
reasons that remained vague, an overt clinical AIDS phase
of terminal profound immunodeficiency ensued, characterized by serious opportunistic conditions, of which the
commonest worldwide today remains extrapulmonary
and disseminated TB. The triple advent of technology to
easily determine circulating viral load (quantitative
reverse transcriptase polymerase chain reaction [qRTPCR]), highly effective multidrug treatment for HIV-1
and improvement in methods to quantitate HIV-1 antigen-specific T cells, however, led to a paradigm shift in
understanding [6,7]. These studies and many subsequent
analyses revealed that the asymptomatic latent phase of
HIV-1 infection was actually characterized by extraordinarily high rates of both virion generation and turnover of
T cells that destroyed virally infected cells: a highly
dynamic equilibrium that the host, in time, was destined
to lose [8]. The terminology ‘latent’ was recognized to
misrepresent pathogenesis and disappeared rapidly from
both literature and thinking on the subject.
By analogy, how true is the current paradigm that
classifies TB into ‘active primary’, ‘latent’ and ‘postprimary’? It is acknowledged to be an oversimplification, leading to a plethora of other terminologies in the literature to
qualify tuberculous infection variously as dormant, active,
inactive, subclinical, acute, chronic, persistent, remote and
quiescent. We propose the following spectrum of responses
0966-842X/$ – see front matter ! 2009 Elsevier Ltd. All rights reserved. doi:10.1016/j.tim.2009.02.005 Available online xxxxxx
1
TIMI-633; No of Pages 6
Opinion
Figure 1. Strategies for tuberculosis control. The TB epidemic is driven by
transmission of bacteria from individuals with active disease. This generates an
infected population, some of whom go on to develop active disease and continue
the cycle. Current interventions focus on reduction of transmission by prompt
diagnosis and treatment of infectious cases. Ultimately, effective control will
depend on combining this strategy with interventions that reduce the risk of
progression to disease after infection.
to TB infection: innate immune, acquired immune, quiescent infection, active infection, disease (Figure 2). Innate
immune is proposed for infection that is eliminated without the need for an acquired response, and acquired
immune for infection that is eliminated by a combination
of innate and acquired responses. Quiescent infection
would represent infection that is controlled by the immune
response but in which viable bacteria persist in a nonreplicating form. Active infection would denote the presence of replicating bacilli nevertheless controlled without
symptoms by the immune response, and clinical disease
the development of symptoms and signs of TB caused by
bacteria that replicate despite the ongoing immune
response.
Is there evidence that these states exist? First, it has
long been recognized that a proportion of repeatedly TBexposed healthy people, for example health care workers,
never convert to a positive TST [9–12]. Although this might
simply reflect a sensitivity deficit of the TST it seems also
plausible that exposure might not always result in an
acquired immune response. Non-immune (e.g. via ciliary
washing up and thereby expectoration) or innate mechanisms including natural killer (NK) cells, neutrophils or
macrophages might eliminate bacilli without the need
for an acquired response [13,14]. Even if an acquired
response is triggered, TST conversion is not invariable.
The advent of IGRAs (which rely on a single immune
parameter, unlike the TST [15]) has consistently identified
persons who are IGRA-positive but TST-negative after
exposure to TB [16–18]. In the study by Ewer [16], reversion of a positive IGRA to negative was noted: could this
represent elimination of low-level infection by an acquired
immune response insufficient to trigger a TST reaction?
Conversely, another reason for failure to convert to positive
in the TST is HIV-1 infection, which markedly decreases
the sensitivity of TST testing [17]. Another intriguing
group thrown up by comparisons between the TST and
2
Trends in Microbiology Vol.xxx No.x
Figure 2. A spectrum of responses to tuberculosis infection. Infection with M.
tuberculosis is usually viewed in terms of a binary outcome: as active disease or
latent infection. We propose that a model that includes a spectrum of responses
provides a better representation of the biology of infection and might assist in
formulation of appropriate research questions.
IGRA are persons who are IGRA-negative but strongly
TST-positive. The recognized more dynamic nature of
IGRA responses might reflect recent infection that stimulates effector responses to the antigens secreted, by
actively dividing bacilli upon which these assays depend
[19,20]. Could resolution of such responses and the development of a memory response to other antigens,
expressed at low levels by non-replicating bacilli and only
present in PPD, in highly protected people, result in this
phenotype? Is it also feasible that strain variation in M.
tuberculosis has differential effects on TST conversion and
disease progression? The advent of discriminatory techniques to classify clinical strains of M. tuberculosis has
facilitated correlation of virulence attributes with genotype [21]. An early investigation of an outbreak caused by
strain CDC1551 noted a tendency to marked TST conversion, possibly related to its ability to induce a vigorous
innate immune response [22,23]. Conversely, transmission
of TB detected only by IGRA and not by TST has also been
described [18,24].
Microbiology of latent TB
Is there a correlation between an active TB-specific
immune response and the presence of viable M. tuberculosis? This question has plagued TB researchers for over a
hundred years. A straightforward approach to answer it
involves a comprehensive screen of tissue samples for the
presence of bacteria by microbiological culture or by inoculation into highly susceptible guinea pigs. Extensive
autopsy studies were carried out in the first part of the
20th century, generating an impressive although contradictory and bewildering literature [25–29].
Widely varying rates of positive cultures were obtained
from individuals living in endemic areas but dying from
causes unrelated to TB. Some investigators isolated and
TIMI-633; No of Pages 6
Opinion
cultured bacteria from histologically normal tissues; more
commonly investigators focused on lesions, reporting
different rates of positive cultures from different lesion
types. Confounding factors in such studies include the
existence of ‘dormant’ bacterial phenotypes that require
especially careful and prolonged culture to generate bacterial colonies, and the possibility that, in highly endemic
settings, positive cultures might represent recently
inhaled bacteria rather than persistent infection or progressive disease. A general conclusion might be that individuals with latent TB frequently harbour viable bacteria;
however, it is difficult to draw hard quantitative conclusions from the essentially anecdotal process of postmortem analysis.
With respect to ‘dormancy’, there is clear evidence from
experiments performed on in vitro cultures that M. tuberculosis can transcriptionally adapt to a wide variety of
environmental conditions [30]. A classic system employed
to better understand non-replicating persistence uses oxygen limitation to induce this state (the Wayne model [31]).
Under these circumstances it is recognized that an early
response is the coordinate upregulation of genes under the
control of two sensor kinases (dosS and dosT) and a
response regulator (dosR) [32,33]. The ‘dosR regulon’
includes genes involved in triglyceride synthesis: there
is great interest in this field because part of the adaptation
of M. tuberculosis to persistence clearly involves a shift in
carbon source from glucose to fatty acids [34,35]. More
recently it has been shown that a second wave of genes are
induced by more prolonged hypoxia that only minorly
overlap with the dosR regulon: this has been called the
enduring hypoxic response and includes a large number of
transcriptional regulators [36].
Modern technologies, especially nucleic acid amplification, provide the opportunity to advance our detailed understanding of the status of bacteria present in tissue
samples. Two interesting studies have demonstrated the
presence of M. tuberculosis DNA outside both cells and
granulomas in latently infected tissue [37,38]. Determination of the expression profile of bacilli in latently infected
tissue would be of particular interest. However, we have
not so far been able to move from localized analysis
towards a systemic measure of bacterial load. It seems
feasible that bacterial components could be detectable in
the circulation (for example, cell wall components) and the
search for such biomarkers is an important research
priority. An encouraging precedent is provided by the
detection of phenolic glycolipid in blood and urine from
leprosy patients [39–41], and the detection of mycolic acids
persisting in archaeological specimens [42]. A truly quantitative measure of bacterial load, analogous to measures
of viral load routinely used in the epidemiology and management of HIV, would transform the spectrum proposed
in Figure 2 from a conceptual hypothesis to an experimental framework.
Immunology of latent TB
Can we use the immune response as a surrogate measure
of bacterial load? Careful analysis of low-level antibody
responses, sometimes overlooked as ‘background’ in the
development of serodiagnostics, might be a useful indicator
Trends in Microbiology
Vol.xxx No.x
of the presence of antigen in latent TB [43]. Changes in the
level of T-cell response in persons undergoing treatment of
both active and latent infection indicate some relationship
to antigen load [44,45], although measurements using
peripheral blood are complicated by the sequestration of
reactive cells to disease sites [46,47]. From a population
perspective, the level of T-cell response associated with
asymptomatic infection is not readily separable from that
seen in patients. An indirect indication of variations in
antigen load in asymptomatic infection is provided by
clinical studies of the impact of antiretroviral treatment
of HIV patients. Restoration of T-cell responses in these
individuals can be associated with a sudden conversion
from inapparent TB to an acute unmasking form of the
immune reconstitution inflammatory syndrome (IRIS, a
syndrome arising from rapid restoration of pathogenspecific immune responses causing either the deterioration
of a treated infection or the new presentation of a previously subclinical infection) that seems likely to be related
to the induction of an immune response in the context of a
high pre-existing antigen load [48].
Aside from the issue of bacterial load, refinements in
analysis of T-cell responses offer the potential to explore
the proposed latent TB spectrum. Because gene expression
profiles differ widely between cultures of M. tuberculosis,
subject to differing conditions in vitro, we might anticipate
these differences would result in distinct sets of antigens
being available for T-cell recognition in the groups we refer
to in Figure 2 as ‘acquired immune’, ‘quiescent infection’
and ‘active infection’. There is some evidence of infectionstage-specific recognition of M. tuberculosis antigens in
humans, although the overlap between clinical categories
is substantial [49–51]. Having a better understanding of
the dynamics and markers of T-cell memory, effector and
perhaps regulatory [52] responses might also help in the
important distinction of individuals who have actually
cleared the infection. Over time, after initial infection,
one might expect a progressive shift from effector to central
memory phenotypes in the ‘adaptive immune’ group, in
contrast to a predominance of effector cells in groups with
more frequent antigen exposure or ongoing low-level bacterial replication. Again, HIV infection points the way.
Polyfunctional T cells that produce multiple cytokines
and which retain the capacity to divide, correlate with
better viral control [53], a finding that has already influenced what is measured in trials of TB vaccines in humans
[54].
Interventions in latent TB
Current efforts to control TB focus primarily on attempts to
reduce transmission of M. tuberculosis by prompt detection
and effective treatment of infectious patients, and it is
hoped that optimal implementation of this strategy will
achieve the United Nations Millennium Development Goal
of a 50% reduction in the global prevalence and mortality of
TB by the year 2015. To achieve the more ambitious goal of
TB elimination by 2050, it will be necessary to combine the
transmission blocking approach with interventions that
reduce the percent of infected individuals who progress to
clinical disease [55] (Figure 1). The powerful impact of
changes in the risk of disease progression is illustrated
3
TIMI-633; No of Pages 6
Opinion
by the consequences of the enhanced susceptibility of
persons who are co-infected with HIV: the 5%–10% lifetime
risk of disease is increased to a 5%–10% annual risk in
association with HIV infection, resulting in dramatic
increases in TB rates in sub-Saharan Africa [5]. In addition
to efforts to enhance protective immunity by vaccination
before infection, the size and elevated risk of the latent TB
population indicate a crucial role for interventions that
would reduce disease progression in individuals who have
already been exposed to infection.
The focus on development of intervention strategies for
latent TB highlights a need for a more careful definition of
the biological condition, or conditions, that we are attempting to target. The presence of antigen-specific T cells is
common to a spectrum of responses to M. tuberculosis that
range from complete clearance of the pathogen, through
containment in a benign but potentially virulent form, to
progressive pre-clinical and clinical disease (Figure 2). The
disease risk associated with latent TB is a combination of
endogenous reactivation of an initial infection or failure to
control progression of an exogenous reinfection: the
relative importance of the two routes presumably depending on the risk of repeated exposure [56,57].
Treatment of latent TB with antitubercular drugs (‘preventive therapy’), most commonly daily isoniazid for 9
months, typically results in a risk reduction of around
50%, even in HIV-infected persons [58,59]. The duration
of protection seems to be related to the risk of reinfection,
being short in HIV-infected persons in high incidence
environments but very long in the historic studies conducted in North America [59,60]. Amongst researchers
aiming to develop improved treatment for latent TB, the
results of these clinical trials prompt two interrelated
questions. First, why does the treatment take so long?
Treatment of latent TB involves elimination of a very much
smaller bacterial population than treatment of clinical TB,
and it might, therefore, be anticipated that this would
require a much less extensive antibiotic regime. An attractive hypothesis is that the bacteria present in latent TB
persist in a physiological state that renders them tolerant
to the effect of drugs. Isoniazid acts by inhibiting mycobacterial cell wall synthesis and loses its efficacy when
bacteria are not replicating. Second, why does it work at
all? Two potential mechanisms could account for the efficacy of isoniazid in preventive therapy: non-replicating
bacteria might have some residual requirement for lowlevel synthesis of cell wall precursors (for maintenance and
repair, for example), or bacterial populations might cycle
between replicating and non-replicating phases providing
periodic windows of isoniazid sensitivity). The search for
rational development of improved drugs for latent infection
focuses on understanding the physiology of bacteria present in latent lesions, primarily using transcriptional profiling to identify patterns of gene expression by M.
tuberculosis present in latent lesions obtained from lungs
resected from patients with suspected lung cancer.
The search for improved drugs for preventive therapy,
and the associated exploration of bacterial in vivo phenotypes, is part of a broader programme to identify rational
strategies for shortening of treatment times for active TB.
Although excretion of bacteria in sputum is rapidly
4
Trends in Microbiology Vol.xxx No.x
reduced within days or weeks of initiation of treatment,
a 6-month course is required to eliminate ‘persistent’ M.
tuberculosis and prevent relapse of disease. It is believed
that the persistence phenomenon observed during treatment of TB disease is also because of the presence of
phenotypically drug-tolerant bacterial populations. However, the relationship (if any) of the bacteria that persist in
the face of a primed immune response in latent TB and the
bacteria that persist during chemotherapy of active TB is
unknown. Maintenance in a non-replicating state presents
an attractive hypothesis in both cases. A clearly formulated
version of this hypothesis invokes an important role for
non-replicating persistence in the hypoxic zone of granulomas that have become necrotic (known as caseous, i.e.
cheese-like, necrosis), and substantial research efforts
are currently focused on rigorous assessment of this
hypoxia hypothesis [61]. Most animal models incompletely
reproduce all the pathological features of human TB
and latent TB is particularly hard to recreate. Exciting
recent progress in this respect has been made in a primate
model of TB. A wide spectrum of lesions characterize both
active and inactive ‘latent’ infection in this model that
closely recreates much of the pathology of human disease
[62,63].
Consideration of whether a novel agent (or combinations of existing agents) might be effective in elimination
of latent TB also poses a substantial problem in clinical
evaluation. By definition, latent TB is asymptomatic and
not characterised by gross radiographic abnormality.
Clinical trials would need to be very large and prolonged
unless conducted in a high incidence area, in which case
reinfection might be an important confounder. Attention
is, therefore, being paid to surrogates of potential efficacy,
for example a change in the immune response [45]. Efforts
to shorten treatment times for active disease face a similar
challenge in measuring elimination of bacteria that persist
after sterilisation of the population accessed in the sputum.
Live imaging of lesions using a combination of high resolution computerised axial tomography (CT) and positron
emission tomography (PET) scans is being explored in this
context and might also have application for monitoring of
novel interventions against latent infection.
The alternative to drug therapy would be to boost the
natural immunity associated with TB infection by post
exposure vaccination. Although it is attractive to propose
that boosting of the immune response will clear persisting
infection, there is no clinical proof-of-concept, and very few
examples of success in experimental models. In addition,
inappropriate immune activation in the presence of infection clearly has the potential to exacerbate clinical
symptoms [64]. The most optimistic hypothesis for
vaccines against latent TB is to suggest that infection with
M. tuberculosis subverts the natural immune response in a
way that impairs its optimal efficacy. This could involve
triggering signalling pathways involved in tolerance to
commensal bacteria, for example, misleading the immune
system as to the potential danger of the infection. Theoretically, vaccines and adjuvants that overcome this
immune subversion, rather than the classical approach
of reproducing natural infection, might provide an attractive approach to treatment of latent TB.
TIMI-633; No of Pages 6
Opinion
Box 1. A wish list for future research on latent TB
! Identification of potential mechanisms of resistance to M.
tuberculosis in the absence of T-cell priming
! Distinction of individuals who mount a sterilising immune
response to initial infection
! Identification of latent TB groups at highest risk of progression to
disease
! Understanding of the physiology of M. tuberculosis during
human infection
! Improvement in imaging to determine the metabolic activity
associated with ‘inactive’ or ‘old’ TB lesions
! Overall, an improved scientific platform for rational development
of drugs and vaccines to reduce progression from asymptomatic
infection to active disease
Refining ‘latent TB’ in the research lexicon?
Although the concept of latent TB, defined by the biomarker of antigen-specific T cells in the absence of clinical
symptoms, is useful in a clinical context, it can be detrimental in the context of formulating a research agenda.
Although the term is unlikely to fall into disuse, we would
like to propose that it is subdivided as outlined in Figure 2.
We do not envisage these as rigid boxes: individuals might
well move back and forth between categories, and the
concept of host-pathogen interactions mapped along a
spectrum of outcomes seems attractive. A potential unifying theme across the spectrum could be cycling of bacteria
between replicating and non-replicating states. The oscillation frequency might depend on the physiology of individual granulomas, but the net effect might be a relatively
short frequency in active disease as compared to a long
frequency in asymptomatic infection. Whether individual
bacteria do oscillate between replicating states is conjectural, although recent advances in single cell analysis
suggest this possibility (McKinney J: presentation to 7th
International Congress on the pathogenesis of mycobacterial infections, Stockholm, June 2008). We anticipate
that consideration of TB as a spectrum from immunity
to disease will facilitate further research towards improved
detection methods and treatments of latent TB (Box 1).
Acknowledgements
The ideas included in this review have been developed in the context of a
consortium funded by the Bill and Melinda Gates Foundation and the
Wellcome Trust to explore development of drugs for treatment of latent
TB as part of the initiative to address Grand Challenges in Global Health.
R.J.W also receives support from the European Union.
References
1 WHO (2008) Global Tuberculosis Control - Surveillance, Planning,
Financing, WHO
2 Aziz, M.A. et al. (2006) Epidemiology of antituberculosis drug
resistance (the Global Project on Anti-tuberculosis Drug Resistance
Surveillance): an updated analysis. Lancet 368, 2142–2154
3 Dye, C. et al. (1999) Consensus statement. Global burden of tuberculosis:
estimated incidence, prevalence, and mortality by country. WHO Global
Surveillance and Monitoring Project. JAMA 282, 677–686
4 Pai, M. et al. (2008) Systematic review: T-cell-based assays for the
diagnosis of latent tuberculosis infection: an update. Ann. Intern. Med.
149, 177–184
5 Maartens, G. and Wilkinson, R.J. (2007) Tuberculosis. Lancet 370,
2030–2043
6 Ho, D.D. et al. (1995) Rapid turnover of plasma virions and CD4
lymphocytes in HIV-1 infection. Nature 373, 123–126
Trends in Microbiology
Vol.xxx No.x
7 Wei, X. et al. (1995) Viral dynamics in human immunodeficiency virus
type 1 infection. Nature 373, 117–122
8 Douek, D.C. et al. (2003) T cell dynamics in HIV-1 infection. Annu. Rev.
Immunol. 21, 265–304
9 Committee, Tuberculosis Research (1932) Tuberculosis in South
African natives with special reference to the disease amongst the
mine labourers on the Witwatersrand. 430, South African Institute
for Medical Research
10 Stead, W.W. (1995) Management of health care workers after
inadvertent exposure to tuberculosis: a guide for the use of
preventive therapy. Ann. Intern. Med. 122, 906–912
11 Joshi, R. et al. (2006) Tuberculosis among health-care workers in lowand middle-income countries: a systematic review. PLoS Med. 3, e494
12 Morrison, J. et al. (2008) Tuberculosis and latent tuberculosis infection
in close contacts of people with pulmonary tuberculosis in low-income
and middle-income countries: a systematic review and meta-analysis.
Lancet Infect. Dis. 8, 359–368
13 Martineau, A.R. et al. (2007) Neutrophil-mediated innate immune
resistance to mycobacteria. J. Clin. Invest. 117, 1988–1994
14 Clay, H. et al. (2008) Tumor necrosis factor signaling mediates
resistance to mycobacteria by inhibiting bacterial growth and
macrophage death. Immunity 29, 283–294
15 Richeldi, L. (2006) An update on the diagnosis of tuberculosis infection.
Am. J. Respir. Crit. Care Med. 174, 736–742
16 Ewer, K. et al. (2006) Dynamic antigen-specific T Cell responses after
point-source exposure to Mycobacterium tuberculosis. Am. J. Respir.
Crit. Care Med. 174, 831–839
17 Rangaka, M.X. et al. (2007) Effect of HIV-1 infection on T-Cell-based
and skin test detection of tuberculosis infection. Am. J. Respir. Crit.
Care Med. 175, 514–520
18 Anderson, S.T. et al. (2006) Transmission of Mycobacterium
tuberculosis undetected by tuberculin skin testing. Am. J. Respir.
Crit. Care Med. 173, 1038–1042
19 Hill, P.C. et al. (2007) Longitudinal assessment of an ELISPOT test for
Mycobacterium tuberculosis infection. PLoS Med. 4, e192
20 Pai, M. et al. (2006) Serial testing of health care workers for
tuberculosis using interferon-gamma assay. Am. J. Respir. Crit.
Care Med. 174, 349–355
21 Nicol, M.P. and Wilkinson, R.J. (2008) The clinical consequences of
strain diversity in Mycobacterium tuberculosis. Trans. R. Soc. Trop.
Med. Hyg. 102, 955–965
22 Valway, S.E. et al. (1998) An outbreak involving extensive
transmission of a virulent strain of Mycobacterium tuberculosis. N.
Engl. J. Med. 338, 633–639
23 Manca, C. et al. (1999) Mycobacterium tuberculosis CDC1551 induces a
more vigorous host response in vivo and in vitro, but is not more
virulent than other clinical isolates. J. Immunol. 162, 6740–6746
24 Ewer, K. et al. (2003) Comparison of T-cell-based assay with tuberculin
skin test for diagnosis of Mycobacterium tuberculosis infection in a
school tuberculosis outbreak. Lancet 361, 1168–1173
25 Loomis, H. (1890) Some facts in the etiology of tuberculosis, evidenced
by thirty autopsies and experiments upon animals. Medical Record 38,
689–698
26 Opie, E. and Aronson, J. (1927) Tubercle bacilli in latent tuberculous
lesions and lung tissue without tuberculous lesions. Arch. Pathol. 4, 1–21
27 Griffith, A. (1929) Types of tubercle bacilli in human tuberculosis. J.
Pathol. Bacteriol. 32, 813–840
28 Feldman, W. and Bagenstoss, A. (1938) The residual infectivity of the
primary complex of tuberculosis. Am. J. Pathol. 14, 473–490
29 Vandiviere, H.M. et al. (1956) The treated pulmonary lesion and its
tubercle bacillus II. The death and resurrection. Am. J. Med. Sci. 232,
30–37
30 Boshoff, H.I. and Barry, C.E. (2005) Tuberculosis - metabolism and
respiration in the absence of growth. Nat. Rev. Microbiol. 3, 70–80
31 Wayne, L.G. and Lin, K.Y. (1982) Glyoxylate metabolism and
adaptation of Mycobacterium tuberculosis to survival under
anaerobic conditions. Infect. Immun. 37, 1042–1049
32 Roberts, D.M. et al. (2004) Two sensor kinases contribute to the hypoxic
response of Mycobacterium tuberculosis. J. Biol. Chem. 279, 23082–
23087
33 Sherman, D.R. et al. (2001) Regulation of the Mycobacterium
tuberculosis hypoxic response gene encoding alpha -crystallin. Proc.
Natl. Acad. Sci. U. S. A. 98, 7534–7539
5
TIMI-633; No of Pages 6
Opinion
34 McKinney, J.D. et al. (2000) Persistence of Mycobacterium tuberculosis
in macrophages and mice requires the glyoxylate shunt enzyme
isocitrate lyase. Nature 406, 735–738
35 Reed, M.B. et al. (2007) The W-Beijing lineage of Mycobacterium
tuberculosis overproduces triglycerides and has the DosR dormancy
regulon constitutively upregulated. J. Bacteriol. 189, 2583–2589
36 Rustad, T.R. et al. (2008) The enduring hypoxic response of
Mycobacterium tuberculosis. PLoS One 3, e1502
37 Hernandez-Pando, R. et al. (2000) Persistence of DNA from
Mycobacterium tuberculosis in superficially normal lung tissue
during latent infection. Lancet 356, 2133–2138
38 Neyrolles, O. et al. (2006) Is adipose tissue a place for Mycobacterium
tuberculosis persistence? PLoS One 1, e43
39 Kaldany, R.R. et al. (1987) Methods for the detection of a specific
Mycobacterium leprae antigen in the urine of leprosy patients.
Scand. J. Immunol. 25, 37–43
40 Cho, S.N. et al. (1986) Quantitation of the phenolic glycolipid of
Mycobacterium leprae and relevance to glycolipid antigenemia in
leprosy. J. Infect. Dis. 153, 560–569
41 Young, D.B. et al. (1985) Detection of phenolic glycolipid I in sera
from patients with lepromatous leprosy. J. Infect. Dis. 152, 1078–
1081
42 Gernaey, A.M. et al. (2001) Mycolic acids and ancient DNA confirm an
osteological diagnosis of tuberculosis. Tuberculosis (Edinb.) 81, 259–
265
43 Wanchu, A. et al. (2008) Biomarkers for clinical and incipient
tuberculosis: performance in a TB-endemic country. PLoS One 3,
e2071
44 Nicol, M.P. et al. (2005) Enzyme-linked immunospot assay responses to
early secretory antigenic target 6, culture filtrate protein 10, and
purified protein derivative among children with tuberculosis:
implications for diagnosis and monitoring of therapy. Clin. Infect.
Dis. 40, 1301–1308
45 Wilkinson, K.A. et al. (2006) Effect of Treatment of Latent Tuberculosis
Infection on the T Cell Response to Mycobacterium tuberculosis
Antigens. J. Infect. Dis. 193, 354–359
46 Barnes, P.F. et al. (1990) Local production of tumor necrosis factor and
IFN-gamma in tuberculous pleuritis. J. Immunol. 145, 149–154
47 Wilkinson, R.J. et al. (1998) Peptide specific response to M.
tuberculosis: clinical spectrum, compartmentalization, and effect of
chemotherapy. J. Infect. Dis. 178, 760–768
48 Lawn, S.D. et al. (2008) Immune reconstitution and ‘‘unmasking’’ of
tuberculosis during antiretroviral therapy. Am. J. Respir. Crit. Care
Med. 177, 680–685
6
Trends in Microbiology Vol.xxx No.x
49 Wilkinson, K.A. et al. (2005) Infection biology of a novel alphacrystallin of Mycobacterium tuberculosis: Acr2. J. Immunol. 174,
4237–4243
50 Wilkinson, R.J. et al. (1998) Human T and B cell reactivity to the 16
kDa alpha crystallin protein of Mycobacterium tuberculosis. Scand. J.
Immunol. 48, 403–409
51 Leyten, E.M. et al. (2006) Human T-cell responses to 25 novel antigens
encoded by genes of the dormancy regulon of Mycobacterium
tuberculosis. Microbes Infect. 8, 2052–2060
52 Guyot-Revol, V. et al. (2006) Regulatory T cells are expanded in blood
and disease sites in patients with tuberculosis. Am. J. Respir. Crit.
Care Med. 173, 803–810
53 Almeida, J.R. et al. (2007) Superior control of HIV-1 replication by
CD8+ T cells is reflected by their avidity, polyfunctionality, and clonal
turnover. J. Exp. Med. 204, 2473–2485
54 Hawkridge, T. et al. (2008) Safety and immunogenicity of a new
tuberculosis vaccine, MVA85A, in healthy adults in South Africa. J.
Infect. Dis. 198, 544–552
55 Dye, C. and Williams, B.G. (2008) Eliminating human tuberculosis in
the twenty-first century. J. R. Soc. Interface 5, 653–662
56 van Rie, A. et al. (1999) Exogenous reinfection as a cause of recurrent
tuberculosis after curative treatment. N. Engl. J. Med. 341, 1174–1179
57 Lillebaek, T. et al. (2003) Stability of DNA patterns and evidence of
Mycobacterium tuberculosis reactivation occurring decades after the
initial infection. J. Infect. Dis. 188, 1032–1039
58 Wilkinson, D. (2000) Drugs for preventing tuberculosis in HIV infected
persons (Cochrane Review). Cochrane Database Syst. Rev. 4
59 Woldehanna, S. and Volmink, J. (2004) Treatment of latent
tuberculosis infection in HIV infected persons. Cochrane Database
Syst. Rev. CD000171
60 Comstock, G.W. et al. (1979) Isoniazid prophylaxis among Alaskan
Eskimos: a final report of the Bethel isoniazid studies. Am. Rev. Respir.
Dis. 119, 827–830
61 Via, L.E. et al. (2008) Tuberculous granulomas are hypoxic in guinea
pigs, rabbits, and nonhuman primates. Infect. Immun. 76, 2333–2340
62 Capuano, S.V., 3rd et al. (2003) Experimental Mycobacterium
tuberculosis infection of cynomolgus macaques closely resembles the
various manifestations of human M. tuberculosis infection. Infect.
Immun. 71, 5831–5844
63 Lin, P.L. et al. (2006) Early events in Mycobacterium tuberculosis
infection in cynomolgus macaques. Infect. Immun. 74, 3790–3803
64 Meintjes, G. et al. (2008) Tuberculosis-associated immune
reconstitution inflammatory syndrome: case definitions for use in
resource-limited settings. Lancet Infect. Dis. 8, 516–523