| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
|
|
|
|||||||
|
|||||||||
el2,*
*
Institut für Klinische Mikrobiologie, Immunologie und Hygiene der Friedrich-Alexander Universität Erlangen-Nürnberg, Erlangen, Germany; and
Klinikum Nürnberg, Medizinische Klinik 3, Nürnberg, Germany
 |
Abstract |
|---|
 Top Abstract IntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
|
Introduction |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
We sought to clarify how TNF participates in the immunopathogenesis of
human infection using tuberculosis as a model. Mycobacterium
tuberculosis, the causative agent of tuberculosis, is a
facultative intracellular bacterium that is capable of surviving and
persisting within host mononuclear cells. Infection with M.
tuberculosis occurs via the respiratory route by the inhalation of
bacilli-containing droplets. The bacilli are taken up by alveolar
macrophages (AM),4
which represent the source of a primary tuberculous focus in which the
pathogen either replicates or is contained. Depletion of AM protected
mice against tuberculosis, indicating that the primary host cell for
inhaled bacilli may cause more harm than protection in vivo
(11). One of the most intriguing questions in tuberculosis
concerns the fate of phagocytosed bacteria. Nonactivated AM are not
equipped with effector mechanisms sufficient to clear the microbial
invader. Ag-specific T cells are attracted to the site of disease by
the action of locally released chemokines (12, 13, 14, 15, 16, 17, 18).
Consequently, T cells secrete macrophage-activating cytokines (e.g.,
TNF and IFN-
) (15) or directly lyse infected cells and
contribute to the containment of mycobacterial infection
(19). Less than 10% of the infected individuals will
develop clinically overt disease due to the efficient interplay of the
specific and nonspecific effector mechanisms of the immune system
(20). Despite successful containment of mycobacterial
spread and prevention of tissue destruction, the pathogen manages to
persist in the host for decades.
Detailed studies of murine models of tuberculosis have succeeded in
unraveling a sequence of immunological events, involving a complex
network of cytokines (e.g., IFN- and TNF) and effector molecules
(oxygen and nitrogen radicals) (21). TNF contributes to
immunity against mycobacteria by synergizing with IFN- to activate
infected macrophages (22) and in recruiting macrophages
and lymphocytes to seal up infectious foci by forming granulomas
(5, 6, 7). Protection is mediated primarily by CD4-positive T
cells, which are supported by MHC class I-restricted CD8-positive T
cells and T cells expressing the 
TCR (23).
In contrast, the immune response against M. tuberculosis in
humans remains poorly defined. Naturally occurring genetic deficiencies
in humans have pointed to a critical role of IFN- (24)
and IL-12 (25) in the protective immune response. Whether
TNF is beneficial or detrimental for the clinical course of disease
remains to be determined. TNF is involved in the development of tissue
damage and can support bacterial multiplication in the lung (26, 27). In addition, high levels of TNF were implicated in clinical
worsening of symptoms shortly after the initiation of tuberculostatic
therapy in tuberculosis patients (28). The in vivo effects
of TNF appear to be dose dependent, with low levels of the cytokine
mediating protection against tuberculosis, whereas high concentrations
provoke tissue damage (26). A critical role for TNF in
maintaining the latent state of dormant bacilli is suggested by the
reactivation of tuberculosis in rheumatoid arthritis patients treated
with anti-TNF Abs (29, 30, 31).
The effect of TNF on human cells infected with mycobacteria varies
widely depending on the source of the cells, the mycobacterial strain,
and the experimental setting (32, 33). TNF and IFN-
treatment does not activate antibacterial effects in human monocytes,
which is in striking contrast to the effect on murine cells
(34). To better understand the functional role of TNF at
the site of human infection, we analyzed its expression and impact on
mycobacterial growth in human AM.
|
Materials and Methods |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Cells were cultured in RPMI 1640 (Biochrom, Berlin, Germany) supplemented with 10% heat-inactivated FCS (Sigma-Aldrich, St. Louis, MO), 2 mM glutamine (Sigma-Aldrich), 10 mM HEPES, 13 mM NaHCO3 (Biochrom), 5.6 µg/ml amphotericin B (Sigma-Aldrich), and 60 µg/ml penicillin (Biochrom). In experiments involving the infection of cells with M. tuberculosis, FCS was replaced by pooled human serum (generated from the blood of healthy volunteers) to optimize the phagocytosis of the bacteria.
AM and monocytes
AM were obtained from the bronchoalveolar lavage fluid of
patients who underwent bronchoscopy for diagnostic purposes. AM from
patients given a diagnosis of an infectious lung disease or a disease
afflicting the alveolar space were discarded. Lavage fluid was filtered
through a cell strainer (40 µm; BD Biosciences, Heidelberg, Germany)
and centrifuged (1200 rpm for 10 min) at 4°C. The pellet was
resuspended in RPMI 1640 supplemented with amphotericin B (5.6 µg/ml;
Sigma-Aldrich), penicillin (60 µg/ml), and 10% human serum. Cells
were plated in six-well plates (Costar, Cambridge, MA) at a density of
1 x 106/ml. AM were cultured at 37°C
overnight to allow adherence of the cells and to minimize the
background production of TNF induced by the adherence to plastic.
Nonadherent cells were removed by vigorous washing with PBS. AM were
harvested by gentle scraping with a cell scraper (Costar), which did
not affect the viability of the cells. Purity of the cells was
confirmed by 
-naphthyl-acetate esterase staining (Sigma-Aldrich) and
was found to be >95% in all experiments. Flow cytometric analysis
confirmed the purity of the population (CD3 < 1%, CD19, CD56,
CD66, CD1 negative). Viability of the cells was examined in
representative samples by trypan blue dye exclusion.
Monocytes were isolated by adhering PBMCs for 2 h in a cell culture flask. Nonadherent cells were removed by rinsing the flask at least five times with PBS. The adherent cells were detached by treatment with EDTA (1 mM for 10 min) or using a cell scraper. The resulting cell population contained at least 95% CD14-positive cells as determined by flow cytometry.
Cytokines and Abs
The following cytokines and Abs were used in this study:
recombinant human (rhu)TNF- (Endogen, Woburn, MA), rhuIL-1,
rhuIL-10, rhuIL-12p40, anti-TNF, goat IgG (all from R&D Systems,
Minneapolis, MN), anti-CD56 (clone C218; Immunotech, Marseilles,
France), anti-CD3-PerCP (clone SK 7),
anti-CD19-allophycocyanin (HIB 19), anti-CD66b-FITC
(clone MOPC 315-43X), anti-TNF-allophycocyanin (MAB11; BD
Biosciences), anti-CD14-PE (UCHM-1), anti-CD1a-FITC (clone
NA1/34-HLK; Serotec, Oxford, U.K.), and goat anti-mouse-FITC
(Jackson ImmunoResearch Laboratories, West Grove, PA). Isotype controls
were all purchased from Cymbus Biotechnology (Chandlers Ford,
U.K.).
Growth of M. tuberculosis
M. tuberculosis (virulent strain H37Rv) was grown in suspension with constant, gentle rotation in roller bottles (Corning, Cambridge, MA) containing Middlebrook 7H9 broth (BD Biosciences) supplemented with 1% glycerol (Roth, Karlsruhe, Germany), 0.05% Tween 80 (Sigma-Aldrich), and 10% Middlebrook OADC enrichment (BD Biosciences). Aliquots from logarithmically growing cultures were frozen in PBS containing 10% glycerol, and representative vials were thawed and enumerated for viable CFU on Middlebrook 7H11 plates. Staining of bacterial suspensions with fluorochromic substrates differentiating between live and dead bacteria (BacLight; Molecular Probes, Leiden, Netherlands) revealed a viability of the bacteria above 90%. Because clumping of mycobacteria is a common problem which can influence the validity and reproducibility of the experiments, we undertook several precautions to minimize clumps: 1) culture conditions (rotation, Tween) were chosen to support the growth of single cell suspensions; 2) before in vitro infection M. tuberculosis bacilli were sonicated to disrupt small aggregates of bacteria; and 3) the multiplicity of infection (MOI) was selected such that there were only 1–2 bacilli per AM.
Infection of AM and monocytes
AM or monocytes were infected with single cell suspensions of M. tuberculosis in six-well culture plates at 1 x 106 cells/ml in a final volume of 3 ml. In selected experiments mycobacteria were killed by exposure to 80°C for 30 min in a water bath. After 4 h of incubation at 37°C extracellular bacteria were removed by intensive rinsing with PBS. In experiments designed to quantitate mycobacterial growth, the adherent cells were harvested by gentle scraping with a cell scraper and replated at a concentration of 1 x 106 cells/ml in a 24-well plate (final volume 500 µl) in complete medium without antibiotics plus 10% human serum. The efficiency of infection, as quantified by staining of control cultures on Permanox chamber slides (Nunc, Naperville, IL) in every experiment was dependent on the MOI. In selected experiments AMs were incubated with FITC-conjugated latex beads (FluoSpheres; Molecular Probes) under identical conditions. The microscopic evaluation of infected macrophages under the fluorescence microscope confirmed the absence of any mycobacterial aggregates. Cell viability of infected AMs was determined by trypan blue exclusion and was >99% in all experiments.
FACS staining of M. tuberculosis
The bacterial suspension was incubated with an equal volume of BacLight viability staining kit (Molecular Probes). Bacteria were incubated for 15 min at room temperature in the dark. Cells were washed seven times in 0.5 ml of PBS, centrifuged at 3200 rpm for 20 min, and sonicated in a preheated water bath for 5 min. The stained mycobacteria were then used to infect AM. Labeled bacteria did not differ in terms of viability, infectivity, intracellular growth, or viability of eukaryotic cells as compared with unstained bacteria (data not shown). This procedure allowed the detection of extracellular bacteria by flow cytometry using the 488-nm laser.
Intracellular FACS staining
AM were infected with stained or unstained mycobacteria (MOI 2.5) for 6 h. The final 2 h of incubation were performed in the presence of brefeldin A (10 µg/ml; Sigma-Aldrich). Infected macrophages were harvested by gentle scraping with a cell scraper. Cells were centrifuged in aerosol-tight tubes and the supernatant was collected and stored at -20°C until further analysis. The pellet was resuspended in 500 µl of PBS/2% FCS/4% paraformaldehyde. An aliquot of the cell suspension was given onto a glass slide and allowed to air dry for auramine-rhodamine staining. After 20 min of incubation cells were washed twice and resuspended in 150 µl of PBS/2% FCS/0.5% saponin (Sigma-Aldrich). Ten minutes later anti-TNF-allophycocyanin (1/150 dilution) was added and incubated for 30 min at 4°C. Acquisition was performed in a FACSCalibur flow cytometer (BD Biosciences) and data were analyzed using CellQuest software (BD Biosciences) or WinMDI.
Measurement of cytokine release
TNF bioactivity was assessed by measuring the lytic effect on
the fibrosarcoma cell line WEHI 164 (35). WEHI cells were
pretreated with actinomycin D (1 µg/ml; Sigma-Aldrich) for 2 h.
After washing, cells were plated in a 96-well plate (5 x
104/well) and the supernatants were added in cell
culture medium supplemented with actinomycin D (2 µg/ml). After
overnight incubation methylthiazoltetrazolium (final concentration, 500
µg/ml; Sigma-Aldrich) was added for 4 h. To solubilize the
formazan crystals, the cells were treated with 100 µl of acidic
sodium-dodecyl-sulfate (Sigma-Aldrich) overnight and the optical
density was determined in an ELISA plate reader at a wavelength of 550
nm. Estimates of the concentrations of bioactive TNF in the
supernatants were obtained by comparison with calibration curves
established with a rhuTNF standard. TNF bioactivity in selected AM
supernatant samples was inhibited by anti-TNF (10 µg/ml) but not
by control rabbit IgG. All wells were set up in triplicates. The
sensitivity of the assays was >50 pg/ml in all experiments. For
measurement of cytokine concentrations in the supernatants, a sandwich
ELISA was used and performed using Ab pairs as suggested by the
supplier (TNF and IL-10, Endogen; IL-12 and IL1, R&D Systems). The
sensitivity was 15 pg/ml (TNF and IL-1), 60 pg/ml (IL-10), and 125
pg/ml (IL-12).
Transwell experiment
AM were infected (MOI 2.5) and harvested as described above. Infected cells (1 x 106) were plated in the upper chamber of a transfer system (six-well plates; Costar), which is separated from the lower chamber by a membrane permeable only for particles smaller than 0.4 µm. The diameter of the pores allows cytokines and secreted proteins, but not bacteria, to pass. In the lower chamber 1 x 106 uninfected AM were cultured. After 18 h the supernatants were harvested and examined for TNF.
Quantification of mycobacterial growth
To ensure the reliable quantification of intracellular M. tuberculosis we used three independent methods for measuring mycobacterial growth. First, we used acid-fast stain (auramine-rhodamine; Merck, Darmstadt, Germany). Second, for CFU, infected cells were lysed with 0.3% saponin (Sigma-Aldrich) to release intracellular bacteria. At all time points an aliquot of unlysed, infected cells was harvested and counted. This allowed an exact quantification of cells as well as the determination of cellular viability by trypan blue exclusion. Recovery of cells was >80% in all experiments, with cell viability regularly exceeding 90% of total cells. Lysates of infected cells were resuspended vigorously, transferred into screwcaps, and sonicated in a preheated (37°C) water bath sonicator (Elma, Singen, Germany) for 5 min. Aliquots of the sonicate were diluted 5-fold in 7H9 medium. Four dilutions of each sample were plated in duplicates on 7H11 agar plates and incubated at 37°C and 5% CO2 for 21 days. Third, we measured incorporation of tritium-labeled uracil ([3H]uracil). Uptake of [3H]uracil into the mycobacterial RNA was determined following the method published by Rook et al. (36), with several modifications as detailed previously (37). Briefly, 1 x 106 infected AMs were cultured in duplicates as described above. At the end of the incubation period, cells were lysed using 0.3% saponin, resuspended vigorously, and transferred into screwcaps. Lysates were centrifuged in an aerosol-tight microfuge (3000 rpm for 20 min) and resuspended in 100 µl of 7H9 to allow optimal growth of the released mycobacteria. Lysates were then transferred into 96-well round-bottom plates (Nunc) and incubated in the presence of 3 µCi of [3H]uracil (Amersham-Pharmacia, Freiburg, Germany). After 24 h mycobacteria were killed by treatment with paraformaldehyde (final concentration, 4%) for 30 min. The mycobacteria were harvested onto glass fiber filters (Inotech, Dottikon, Switzerland) and [3H]uracil incorporation was measured in a beta counter (Berthold, München, Germany). Background radioactivity in uninfected cells was below 500 cpm in all experiments.
Flow cytometry
A total of 3 x 105 cells were resuspended in 100 µl of staining buffer (2% FCS, 1% NaN3, PBS without Mg2+/Ca2+) and incubated with unconjugated or conjugated Abs for 30 min on ice. Samples were washed twice in staining buffer and, if necessary, incubated for an additional 30 min on ice with goat anti-mouse-FITC Abs (1/500). Cells were then fixed in 2% paraformaldehyde and stored at 4°C until analysis in a FACScan flow cytometer. Data were analyzed using CellQuest software (BD Biosciences).
Measurement of apoptosis
Two independent methods were used to quantify apoptosis. First, negatively charged phospholipid surfaces, which translocate from the inner to the outer leaflet of the plasma membrane early during apoptosis, were stained using annexin V-PE (BD Biosciences) following exactly the protocol supplied by the manufacturer. For quantification, at least 10,000 cells were analyzed by flow cytometry. Second, cytoplasmic, histone-associated DNA fragments (mono- and oligonucleosomes) which occur during apoptosis were identified using the Cell Death Detection ELISAPlus (Boehringer Mannheim, Mannheim, Germany). Briefly, 1 x 104 cells were lysed and centrifuged (1300 rpm for 10 min), and 20 µl of the supernatant were added to the streptavidin-coated microtiter plate. The detection of nucleosomes was performed according to the protocol supplied by the manufacturer. The conditions for the purification, infection (MOI 0.5), and culture of AM were identical to those used for the quantification of mycobacterial growth (see above).
Statistical analysis
Data are presented as the mean value ± SEM except where stated otherwise. Student’s t test was used to determine statistical significance between two differentially treated cultures. Differences were considered significant if p < 0.05.
|
Results |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Human AM were infected with a virulent strain of M.
tuberculosis (H37Rv) with different bacteria:macrophage ratios for
4 h. The number of infected cells as well as the bacterial load
increased with the MOI as determined by acid-fast stain. Both peaked at
an MOI of 10, at which 32 ± 2% of the AM were infected and had
engulfed 14 ± 2 bacteria (data not shown). Further increase of
the MOI was not informative, as cells rapidly disintegrated. The
majority of AM remained uninfected despite an overwhelming number of
bacteria in the culture. Extension of the pulse infection (up to
18 h) had no impact on these results. These data demonstrate that
only a limited number of AM have the capacity to phagocytose viable
mycobacteria. In contrast, FITC-labeled latex beads were taken up much
more efficiently (Fig. 1
A)
under identical culture conditions. Therefore, AM have a specific
inability to phagocytose live M. tuberculosis and possibly
other bacterial pathogens.
|
The secretion of TNF plays an important role in the orchestration
of the local immune response to intracellular pathogens. Therefore, we
investigated the ability of human AM to secrete TNF in response to
infection with virulent M. tuberculosis. The concentration
of TNF in the supernatant increased with the MOI and the bacterial load
of the AM (Fig. 1
B). TNF produced by AM is bioactive as
determined by lysis of a TNF-sensitive fibrosarcoma cell line (Fig. 1C). The biological activity of TNF was neutralized by an
anti-TNF polyclonal Ab (data not shown). Kinetic analysis
demonstrated that cytokine release peaked after 18 h and dropped
gradually thereafter (Fig. 1D). Phagocytosis alone was not
sufficient to mediate release of preformed or membrane-bound TNF, as
the amount did not exceed background levels for the first 4 h of
incubation (data not shown). In addition, uptake of latex beads did not
induce secretion of TNF (data not shown). To investigate how the
production of TNF relates to the growth of intracellular bacteria we
determined the number of CFU after 1, 18, 28, 48, and 96 h of
infection. After an initial lag phase the bacteria multiplied 4-fold
during the 96-h incubation period. Therefore, the early TNF burst might
be involved in the induction of bacterial growth. These data establish
that virulent M. tuberculosis induces bioactive TNF in human
infection in the critical, early stage of the immune response.
Analysis of TNF production on the single cell level
To allow the identification of TNF-producing AM on the single cell
level we used intracellular flow cytometry using a
fluorochrome-conjugated TNF Ab. Infection of AM resulted in 8 (MOI 1),
14% (MOI 5), and 20% (MOI 10) TNF-expressing cells 6 h after
infection (Fig. 2). Stimulation of AM
with LPS resulted in almost 3-fold higher numbers of TNF-expressing
cells (Fig. 2). This experiment shows that infection of AM with
M. tuberculosis activates only a subset of AM, whereas the
majority of cells, despite producing TNF in response to LPS, remain
immunologically silent.
|
Because the number of TNF-expressing cells was in the same order
of magnitude as the number of infected cells, we reasoned that TNF
production was limited to infected AM. To address this hypothesis we
established a double-staining technique, which permits the simultaneous
detection of M. tuberculosis and TNF on the single-cell
level by flow cytometry. Bacterial DNA was stained with BacLight before
the infection of AM. Labeled extracellular bacteria can easily be
detected (>40-fold increase in fluorescence intensity) in the FITC
channel of the flow cytometer (data not shown). Importantly, infected
AM can be clearly distinguished from noninfected cells in the same
culture (Fig. 3A). The number
of infected cells detected by this method were confirmed by
microscopical evaluation (Table I).
Because the vast majority of infected cells had engulfed only one or
two bacteria, differences in the staining intensity are unlikely due to
the differences in the bacterial burden but reflect the difference
between infected and uninfected cells. The discrete increase of
background staining of AM at a high MOI is probably due to the
translocation of fluorochrome from the prokaryotic to the eukaryotic
DNA. To identify infected cells, which simultaneously produce TNF, we
performed double-staining 6 h after the infection. Confirming our
earlier results, only a subset of AM was infected (23%) with M.
tuberculosis. Similarly, only 8% of the total cell population
produced TNF at an MOI of 2.5 (Fig. 3B). These experiments
revealed that not only infected (4 of 26% infected cells; Fig. 3B, upper right quadrant) but also a fraction of
uninfected AM (4 of 74% uninfected cells; Fig. 3B,
upper left quadrant) expressed TNF. Therefore, uptake of
bacteria promotes, but is not mandatory for, the production of TNF
by AM.
|
|
Because uninfected AM produced TNF we asked whether cytokine
production was mediated by a soluble factor released from infected
cells. We plated infected AM in the upper chamber of a transwell
system, allowing soluble mediators and secreted proteins (smaller than
0.4 µm) but not whole bacteria to pass. Uninfected AM were added to
the lower chamber and intracellular TNF staining was performed after
18 h (Table II). Uninfected AM in
the lower chamber did not up-regulate TNF (2 ± 0.6%) in response
to a soluble factor secreted by the infected cells in the upper
chamber. In contrast, 16 ± 3% of the AM in infected control
cultures expressed TNF. This suggests that TNF production by
noninfected AM is mediated by cell to cell contact or by bacterial
metabolites too large to pass the membrane of the transwell system.
|
Next we sought to determine whether mycobacterial strains differ
in their capacity to induce TNF production, thereby modulating the
outcome of disease early after infection. We investigated two pairs of
genetically closely related mycobacterial strains: 1) M.
tuberculosis H37Rv (virulent) and H37Ra (attenuated strain); and
2) M. bovis (virulent) and M. bovis bacillus
Calmette-Guérin (BCG; attenuated strain). Virulent strains
activated more AM for TNF production than their attenuated counterparts
as determined by intracellular staining (Fig. 4A). Phagocytosis by AM did
not differ significantly between the different mycobacterial strains
(data not shown). The correlation between virulence and TNF production
was confirmed by ELISA for six donors (Fig. 4B). On average,
virulent M. tuberculosis induced 1954 ± 235 pg/ml TNF,
whereas the attenuated strain induced only 623 ± 135 pg/ml.
Similarly, M. bovis-treated AM secreted 2156 ± 303
pg/ml TNF, whereas M. bovis BCG-infected AM secreted only
498 ± 218 pg/ml (Fig. 4B). To determine whether the
release of additional monokines was also dependent on the virulence, we
compared the concentrations of IL-1, IL-12, and IL-10 in the
supernatants of infected AM. Virulent strains induced higher amounts of
all three monokines (Fig. 4C) than attenuated strains. The
absolute amounts of IL-12 and IL-10 were low as compared with TNF and
IL-1 for both strains. Heat-killed mycobacteria only induced IL-10
(1076 ± 34 pg/ml). TNF, IL-1, or IL-12 were not detected,
indicating the requirement of a heat labile component or metabolic
activity of the pathogen for the secretion of these monokines. This
cytokine secretion pattern of AM in response to mycobacterial infection
is reminiscent of other human cell populations, including dendritic
cells (37, 38) and macrophages (39).
|
Because virulent mycobacteria induce high TNF production, we
hypothesized that TNF supports the intracellular replication of the
bacteria. To address this issue, we initially compared the growth of
virulent and attenuated mycobacteria in human AM. While virulent
M. tuberculosis and M. bovis grew more than one
order of magnitude within 7 days (M. tuberculosis, 11.8-fold
increase; M. bovis, 13.2-fold increase), both attenuated
strains failed to demonstrate significant proliferation at all time
points investigated (Fig. 5). These data
demonstrate a positive correlation among the virulence of
mycobacteria, the ability to mediate the release of high amounts of
TNF, and the growth rate in human AM.
|
A). In the presence of
anti-TNF the number of bacteria only increased to 43 ± 2
x 104 (45% growth reduction). Addition of
anti-TNF to M. bovis-infected AM decreased the bacterial
proliferation even more strikingly (67%). To strengthen this
observation we supplemented the infected cultures with exogenous TNF
and monitored the bacterial load for 5 days. TNF accelerated the
multiplication of both mycobacterial strains by 35 ± 3% for
M. tuberculosis and 38 ± 4% for M. bovis
as compared with untreated controls (Fig. 6A). Because
freshly recruited monocytes may also participate in the local
protection against M. tuberculosis, we investigated
mycobacterial growth in monocytes in the presence or absence of TNF and
anti-TNF. Exogenous TNF increased the intracellular growth, whereas
anti-TNF had the opposite effect (Fig. 6B). Finally, TNF
also increased the proliferation of M. tuberculosis in
immature dendritic cells (M. Büttner, M. Röllinghoff, and
S. Stenger, manuscript in preparation). Therefore, TNF alone is
insufficient to activate antibacterial effector mechanisms against
virulent mycobacteria in infected human phagocytes.
|
). First, we determined the metabolic
activity of the bacteria by measuring the uptake of
[3H]uracil. In the presence of anti-TNF the
uptake was reduced from 9487 ± 423 to 4730 ± 301 cpm (50%
reduction). In contrast, treatment with TNF enhanced the metabolic
activity of the bacterial inoculum by 38%. Second, we used acid-fast
staining, which does not discriminate between live or dead bacilli, but
allows quantitation of the number of infected cells. Culture in
the presence of anti-TNF clearly decreased the number of infected
cells as well as the bacterial burden of the individual AM (Table III).
TNF-treated AM harbored significantly more bacilli than control
cultures (12.9 ± 2.7 vs 8.2 ± 2). The number of infected
cells was also significantly increased (68 ± 3% vs 49
± 5%).
|
C) (32) and M. bovis BCG (Fig. 6C), suggesting that the biological activity of this
cytokine varies depending on the virulence of the intracellular
invader.
Because mycobacteria induce apoptosis in human macrophages in a
TNF-dependent manner (40, 41, 42), we considered the
possibility that apoptosis is the underlying mechanism of accelerated
bacterial growth. As a measure for apoptotic cell death we compared the
expression of annexin V (Fig. 7A) and the enrichment of
cytoplasmic nucleosomes (Fig. 7B) in uninfected and infected
AM. In our low-dose model of infection, mycobacteria did not induce
apoptosis after 5 days of incubation. UV-irradiated Jurkat cells, which
we used as a positive control, were readily labeled (Fig. 7).
Therefore, apoptosis is not the mechanism by which TNF supports the
growth of M. tuberculosis in human AM.
|
|
Discussion |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
Our initial experiments revealed that only a fraction of AM phagocytosed M. tuberculosis. Binding of mycobacteria to AM is mediated by complement receptors binding to complement components fixed to mycobacteria (32, 47) and by mannose receptors (48) under the participation of surfactant protein A (48, 49), CD14 (50), and sialoglycoprotein CD43 (51). These molecules serve as pattern recognition receptors that bind conserved bacterial structures (50, 52). Therefore, the interaction between mycobacteria and AM is highly dependent on the expression pattern of cell surface receptors on the phagocyte. The healthy or diseased lung contains a heterogenous population of cells, comprising resident AM, elicited macrophages, and possibly immune-activated macrophages (53, 54). This heterogeneity of AM will most likely pertain to the expression of pattern recognition receptors, thereby influencing the ability to take up airborne pathogens. Ongoing studies are designed to use the double-staining technique described above to identify and characterize subpopulations capable of taking up viable mycobacteria.
The failure of a majority of primary AM to take up mycobacteria does
not correlate with global immunological silence. We show that
uninfected AM contribute to TNF production (Fig. 3B).
Similarly, the attenuated vaccination strain M. bovis BCG
has recently been shown to induce TNF production in uninfected and
infected monocyte-derived macrophages (55). Our initial
hypothesis that mediators released by infected cells (e.g., IL-1, IL-6,
IL-10) stimulate noninfected cells was confuted by transwell
experiments (Table II). Alternatively, TNF production from bystander
cells could be induced by digested and regurgitated bacterial products
of larger size. Despite being incapable of taking up whole microbes, AM
may phagocytose degraded bacterial material by receptor-independent
macropinocytosis. Bacterial products could also be shuttled into APCs
by apoptotic bodies that represent cells that have undergone apoptosis
due to an overwhelming bacterial burden. This mechanism has been shown
to permit Ag presentation to cytolytic T cells (56) and
may also induce the secretion of immunomodulatory cytokines. In our
low-dose model of infection, AM did not undergo apoptosis (Fig. 7),
suggesting that this mechanism is not involved in TNF induction.
So far it remained controversial whether virulent or attenuated
mycobacteria induce higher levels of TNF in human macrophages. Results
published in the past show higher induction by attenuated strains in
macrophages (57), a comparable up-regulation in AM
(40), and finally higher TNF induction by virulent strains
(58, 59) (Fig. 4). TNF induction in human primary
macrophages appears to be highly dependent on the experimental setting
and the origin and treatment of the cells under investigation, as well
as the bacterial strains. Therefore, when studying TNF production, it
is mandatory to use well-defined experimental procedures under
stringent control of variables including purity of the cells,
efficiency of infection, and culture conditions. In our studies we
infected the cultures with a low number of bacteria to exclude loss of
cell viability, we adjusted the bacterial inoculum such that different
strains infected the AM with nearly equal efficiencies, and we
routinely stained an aliquot of the cultures with
-naphthyl-acetate-esterase to document the purity of the population.
In addition, we confirmed our results by three independent methods
(intracellular flow cytometry, ELISA, and bioactivity) and used two
pairs of virulent/attenuated bacteria (M. tuberculosis and
M. bovis).
The key finding of our study is that TNF supports the growth of
virulent M. tuberculosis. This observation suggests that TNF
may not only harm the infected individual by contributing to the
immunopathology of tuberculosis (28, 58, 60, 61) but also
directly supports the survival of intracellular pathogens. Of note, TNF
has opposite effects on the growth of virulent and attenuated strains.
While virulent mycobacteria appear to have developed efficient evasion
mechanisms for protection against the attack of the host immune system,
attenuated strains are susceptible to treatment of the host cell with
TNF (32) (Fig. 6). The question remains: how does TNF
interfere with the antibacterial effector mechanisms of human
AM? Possible mechanisms include 1) the deprivation of
intracellular iron, 2) modulation of the maturity of cells, 3)
regulation of chemokine/cytokine release, and 4) induction of apoptosis
of infected cells.
The deprivation of intracellular iron
TNF is known to modulate the iron metabolism of phagocytes (62). The activity of many mycobacterial enzymes is dependent on the presence of iron (63), and virulent mycobacteria can be attenuated by expression of an iron repressor (64). By inference, it is conceivable that TNF increases the availability of iron for the bacilli, thereby facilitating their growth. This hypothesis is strengthened by a study showing that TNF promotes mycobacterial growth in monocytes by interfering with the iron metabolism (33).
Modulation of the maturity of cells
The AM used in our experiments were enriched from the
bronchoalveolar lavage by adherence to plastic. Despite staining
positively for the macrophage-specific enzyme -naphthyl-acetate
esterase, the macrophage population is likely to comprise different
stages of maturation. One possible effect of TNF could be to modulate
the maturity of the macrophages, thereby influencing antibacterial
activity. Antimycobacterial activity has been shown to be dependent on
the maturity of the host cells; e.g., immature dendritic cells kill
M. tuberculosis more efficiently than mature dendritic cells
(37). Therefore, the growth-promoting effect might be a
result of the modulation of the maturity of infected cells rather than
the deactivation of antibacterial effector mechanisms.
Regulation of chemokine/cytokine release
TNF could induce the synthesis of immunomodulatory cytokines or
chemokines by infected macrophages. Candidates that have been shown to
down-regulate macrophage functions in macrophages include IL-10 and
TGF (65). Both cytokines are expressed in human AM in
response to infection with M. tuberculosis (41, 66), and their secretion could be promoted by TNF.
Induction of apoptosis of infected cells
Finally, TNF could mediate its growth-promoting effect by the regulation of apoptosis. Infection with M. tuberculosis increases the susceptibility of fibroblasts to undergo TNF-induced apoptosis (67). In addition, infection of human AM with attenuated mycobacteria promotes apoptosis of the host cell by the induction of TNF (40). Tuberculosis is a chronic infection, which suggests that macrophage apoptosis might favor the host by depriving the pathogen of its intracellular sanctuary. Apoptosis, as opposed to necrosis, has been shown to induce killing of mycobacteria (68). Alternatively, apoptosis could result in the spread of mycobacteria to uninfected macrophages via the uptake of apoptotic bodies, thereby offering a more favorable environment for the bacilli than highly infected cells (69).
Recent observations suggest that reactivation of tuberculosis is a
potential side effect of treatment with anti-TNF (Infliximab)
(29, 30, 31). The incidence of tuberculosis in patients
undergoing anti-TNF therapy for rheumatoid arthritis or Crohn’s
disease was higher than that of other infections with opportunistic
pathogens (e.g., listeriosis, Pneumocystis carinii
pneumonia, aspergillosis, histoplasmosis, legionellosis). Because TNF
is an inflammatory cytokine that induces a broad spectrum of biological
effects, the mechanisms underlying this observation are most likely
multifactorial and complex. In mice, the lack of TNF results in the
failure to form (5, 6, 7) or maintain granulomas
(70), possibly due to decreased expression of adhesion
molecules and chemokines (71). Alternatively, apoptosis of
infected cells could be diminished in the absence of TNF, thereby
supporting mycobacterial survival (41). The present (Fig. 6) and earlier studies (22, 33) show that TNF alone does
not activate antimycobacterial activity in murine or human macrophages.
Therefore, the reactivation of tuberculosis in Infliximab-treated
patients is probably not due to a loss of antimycobacterial activity,
but reflects the disintegration of granulomas, which allows dormant
mycobacteria to convert into metabolically active and destructive
pathogens (31).
It is undebatable that TNF is critical for protection in murine
tuberculosis. This applies to in vivo models (5, 6, 7, 72)
and in vitro studies demonstrating antimycobacterial activity in
purified macrophage populations stimulated with TNF and IFN-
(22, 73, 74) but not TNF alone (22). In
contrast, TNF and IFN- did not induce killing of M.
tuberculosis in human monocytes (34). In addition,
effector molecules in mice and humans are distinct, as illustrated by
the recently characterized antibacterial effector pathway initiated by
the binding of bacterial lipopeptides to Toll-like receptor (TLR)2. In
mouse macrophages, bacterial lipoprotein activation of TLR2 leads to
NO-dependent killing of intracellular tubercle bacilli. In human
monocytes and AM, bacterial lipoproteins similarly activated TLR2 to
kill intracellular M. tuberculosis, but by an antimicrobial
pathway that is NO independent. This suggests that similar
antibacterial effector pathways are conserved among different species
but the executing molecules are distinct. Therefore, it is meaningful
to investigate human cells to gain comprehensive insight into the
interactions of the constituents of the immune system.
This study suggests a scenario in which virulent mycobacteria infect a subset of AM, which secrete TNF and release degraded bacterial products, which activate bystander cells to perpetuate TNF production. This chain of events allows accelerated proliferation of the bacilli and may serve as an evasion mechanism of virulent M. tuberculosis. However, in vivo during the course of a protective immune response T cells will be attracted and activated by APCs. T cells will contribute to the protective immune response by secreting stimulatory cytokines (75) or antibacterial molecules like granulysin (19), thereby counterregulating the growth-permissive state of AM.
|
Acknowledgments |
|---|
|
Footnotes |
|---|
2 M.E. and E.S. contributed equally to this work.
3 Address correspondence and reprint requests to Dr. Steffen Stenger, Institut für Klinische Mikrobiologie, Immunologie und Hygiene der Friedrich-Alexander Universität Erlangen-Nürnberg, Wasserturmstr. 3, D-91054 Erlangen, Germany. E-mail address: steffen.stenger{at}mikrobio.med.uni-erlangen.de
4 Abbreviations used in this paper: AM, alveolar macrophage; MOI, multiplicity of infection; BCG, bacillus Calmette-Guérin; rhu, recombinant human; TLR, Toll-like receptor.
Received for publication June 8, 2001. Accepted for publication November 27, 2001.
|
References |
|---|
|
TopAbstractIntroductionMaterials and MethodsResultsDiscussionReferences |
|---|
plays a central role in immune-mediated clearance of adenoviral vectors. Proc. Natl. Acad. Sci. USA 94:9814. is required in the protective immune response against Mycobacterium tuberculosis in mice. Immunity 2:561.[Medline]
therapy of rheumatoid arthritis: what have we learned?. Annu. Rev. Immunol. 19:163.[Medline]
production and enhances weight gain in patients with pulmonary tuberculosis. Mol. Med. 1:384.[Medline]
interferon, interleukin-4, and tumor necrosis factor. Infect. Immun. 58:2675.-receptor deficiency in an infant with fatal bacille Calmette-Guerin infection. N. Engl. J. Med. 335:1956. in murine mycobacterial infection are dose dependent. Infect. Immun. 68:6954. and concomitant clinical deterioration after initiating therapy in patients with severe tuberculosis. J. Infect. Dis. 178:580.[Medline]
monoclonal antibody) versus placebo in rheumatoid arthritis patients receiving concomitant methotrexate: a randomised phase III trial. ATTRACT Study Group. Lancet 354:1932.[Medline]
treatment. Am. J. Gastroenterol. 96:1665.[Medline]
-neutralizing agent. N. Engl. J. Med. 345:1098.-mediated growth inhibition of Mycobacterium tuberculosis by human alveolar macrophages. J. Immunol. 152:743.[Abstract]
(TNF) promotes growth of virulent Mycobacterium tuberculosis in human monocytes: iron-mediated growth suppression is correlated with decreased release of TNF from iron-treated infected monocytes. J. Clin. Invest. 99:2518.[Medline]
. J. Immunol. 161:2636. in tuberculous pleuritis. J. Immunol. 145:149.[Abstract]
, IL-1 and IL-6 levels in the bronchoalveolar lavage fluid with the upregulation of their mRNA in macrophages lavaged from patients with active pulmonary tuberculosis. Tuber. Lung Dis. 79:279.[Medline]
interferon, interleukin-4, and tumor necrosis factor mRNA in human lung tuberculous granulomas. Infect. Immun. 68:2827. and tumor necrosis factor- synthesis by alveolar macrophages in pulmonary tuberculosis. Am. J. Respir. Crit. Care Med. 161:192.-interferon, vitamin D3 metabolites and tumour necrosis factor in the pathogenesis of tuberculosis. Immunology 62:229.[Medline]
but not with evasion of lymphocyte-dependent monocyte effector functions. Infect. Immun. 66:1190. is a determinant of pathogenesis and disease progression in mycobacterial infection in the central nervous system. Proc. Natl. Acad. Sci. USA 96:5657., interleukin-4, and interleukin-10. Ann. NY Acad. Sci. 685:713.[Medline]
by blood monocytes from patients with active tuberculosis and presence of TGF- in tuberculous granulomatous lung lesions. J. Immunol. 154:465.[Abstract]
on host immune response in chronic persistent tuberculosis: possible role for limiting pathology. Infect. Immun. 69:1847. stimulated macrophages: regulation by endogenous tumor necrosis factor- and by IL-10. Int. Immunol. 6:693.This article has been cited by other articles:
|
|
 |
|
  J. Day, A. Friedman, and L. S. Schlesinger Modeling the immune rheostat of macrophages in the lung in response to infection PNAS, July 7, 2009; 106(27): 11246 - 11251. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 N. R. Patel, J. Zhu, S. D. Tachado, J. Zhang, Z. Wan, J. Saukkonen, and H. Koziel HIV Impairs TNF-{alpha} Mediated Macrophage Apoptotic Response to Mycobacterium tuberculosis J. Immunol., November 15, 2007; 179(10): 6973 - 6980. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 K. C. Wong, W. M. Leong, H. K. W. Law, K. F. Ip, J. T. H. Lam, K. Y. Yuen, P. L. Ho, W. S. Tse, X. H. Weng, W. H. Zhang, et al. Molecular Characterization of Clinical Isolates of Mycobacterium tuberculosis and Their Association with Phenotypic Virulence in Human Macrophages Clin. Vaccine Immunol., October 1, 2007; 14(10): 1279 - 1284. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. I. Rennard, C. Fogarty, S. Kelsen, W. Long, J. Ramsdell, J. Allison, D. Mahler, C. Saadeh, T. Siler, P. Snell, et al. The Safety and Efficacy of Infliximab in Moderate to Severe Chronic Obstructive Pulmonary Disease Am. J. Respir. Crit. Care Med., May 1, 2007; 175(9): 926 - 934. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S. Kurtz, K. P. McKinnon, M. S. Runge, J. P.-Y. Ting, and M. Braunstein The SecA2 Secretion Factor of Mycobacterium tuberculosis Promotes Growth in Macrophages and Inhibits the Host Immune Response Infect. Immun., December 1, 2006; 74(12): 6855 - 6864. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C. Carranza, E. Juarez, M. Torres, J. J. Ellner, E. Sada, and S. K. Schwander Mycobacterium tuberculosis Growth Control by Lung Macrophages and CD8 Cells from Patient Contacts Am. J. Respir. Crit. Care Med., January 15, 2006; 173(2): 238 - 245. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Cho and D. N. McMurray Neutralization of Tumor Necrosis Factor Alpha Suppresses Antigen-Specific Type 1 Cytokine Responses and Reverses the Inhibition of Mycobacterial Survival in Cocultures of Immune Guinea Pig T Lymphocytes and Infected Macrophages Infect. Immun., December 1, 2005; 73(12): 8437 - 8441. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
F. Stegelmann, M. Bastian, K. Swoboda, R. Bhat, V. Kiessler, A. M. Krensky, M. Roellinghoff, R. L. Modlin, and S. Stenger Coordinate Expression of CC Chemokine Ligand 5, Granulysin, and Perforin in CD8+ T Cells Provides a Host Defense Mechanism against Mycobacterium tuberculosis J. Immunol., December 1, 2005; 175(11): 7474 - 7483. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 S Stenger Immunological control of tuberculosis: role of tumour necrosis factor and more Ann Rheum Dis, November 1, 2005; 64(suppl_4): iv24 - iv28. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
M. Buettner, C. Meinken, M. Bastian, R. Bhat, E. Stossel, G. Faller, G. Cianciolo, J. Ficker, M. Wagner, M. Rollinghoff, et al. Inverse Correlation of Maturity and Antibacterial Activity in Human Dendritic Cells J. Immunol., April 1, 2005; 174(7): 4203 - 4209. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
H. Cho, T. M. Lasco, S. S. Allen, T. Yoshimura, and D. N. McMurray Recombinant Guinea Pig Tumor Necrosis Factor Alpha Stimulates the Expression of Interleukin-12 and the Inhibition of Mycobacterium tuberculosis Growth in Macrophages Infect. Immun., March 1, 2005; 73(3): 1367 - 1376. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 V. Rao, N. Fujiwara, S. A. Porcelli, and M. S. Glickman Mycobacterium tuberculosis controls host innate immune activation through cyclopropane modification of a glycolipid effector molecule J. Exp. Med., February 22, 2005; 201(4): 535 - 543. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 R. B. Day, Y. Wang, K. S. Knox, R. Pasula, W. J. Martin II, and H. L. Twigg III Alveolar Macrophages from HIV-Infected Subjects are Resistant to Mycobacterium tuberculosis In Vitro Am. J. Respir. Cell Mol. Biol., March 1, 2004; 30(3): 403 - 410. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 M. B. Drennan, D. Nicolle, V. J. F. Quesniaux, M. Jacobs, N. Allie, J. Mpagi, C. Fremond, H. Wagner, C. Kirschning, and B. Ryffel Toll-Like Receptor 2-Deficient Mice Succumb to Mycobacterium tuberculosis Infection Am. J. Pathol., January 1, 2004; 164(1): 49 - 57. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
T. M. Lasco, T. Yamamoto, T. Yoshimura, S. S. Allen, L. Cassone, and D. N. McMurray Effect of Mycobacterium bovis BCG Vaccination on Mycobacterium-Specific Cellular Proliferation and Tumor Necrosis Factor Alpha Production from Distinct Guinea Pig Leukocyte Populations Infect. Immun., December 1, 2003; 71(12): 7035 - 7042. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
N. M. Price, R. H. Gilman, J. Uddin, S. Recavarren, and J. S. Friedland Unopposed Matrix Metalloproteinase-9 Expression in Human Tuberculous Granuloma and the Role of TNF-{alpha}-Dependent Monocyte Networks J. Immunol., November 15, 2003; 171(10): 5579 - 5586. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
C.-H. Song, J.-S. Lee, H.-J. Kim, J.-K. Park, T.-H. Paik, and E.-K. Jo Interleukin-8 Is Differentially Expressed by Human-Derived Monocytic Cell Line U937 Infected with Mycobacterium tuberculosis H37Rv and Mycobacterium marinum Infect. Immun., October 1, 2003; 71(10): 5480 - 5487. [Abstract] [Full Text] [PDF]
|
|
|
|
 |
|
 G.-S. Shin, B.-H. Lee, S. Lee, S.-Y. Chung, M. Kim, J. Lim, Y. Kim, H. J. Kwon, C. S. Kang, and K. Han Monokine Levels in Cancer and Infection Ann. Clin. Lab. Sci., April 1, 2003; 33(2): 149 - 155. [Abstract] [Full Text] [PDF]
|
|
|
|
|
|
V. Trajkovic, G. Singh, B. Singh, S. Singh, and P. Sharma Effect of Mycobacterium tuberculosis-Specific 10-Kilodalton Antigen on Macrophage Release of Tumor Necrosis Factor Alpha and Nitric Oxide Infect. Immun., December 1, 2002; 70(12): 6558 - 6566. [Abstract] [Full Text] [PDF]
|
|
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
