Previous Article | Next Article ![]()
Infection and Immunity, June 2005, p. 3783-3786, Vol. 73, No. 6
0019-9567/05/$08.00+0 doi:10.1128/IAI.73.6.3783-3786.2005
Copyright © 2005, American Society for Microbiology. All Rights Reserved.
Institute of Microbiology,1 Infectious Diseases Service, University Hospital, Lausanne, CH-1011, Switzerland,2 Department of Medicine and Pathology, Section of Rheumatology, Yale University School of Medicine, New Haven, Connecticut3
Received 1 July 2004/ Returned for modification 1 October 2004/ Accepted 11 February 2005
|
|
|---|
|
|
|---|
) are involved in the innate immune response to Mycobacterium tuberculosis (16). The ability of the host response to increase the mycobactericidal activity of M
is important for the development of an early effective response and for the containment of M. tuberculosis infection. The infection of M
with M. tuberculosis induces the release of early proinflammatory cytokines, like tumor necrosis factor alpha (TNF-
) and interleukin-1ß (24). However, at least in humans, the exact mechanisms by which cytokines act to contain mycobacterial growth remain unclear (17). Macrophage migration inhibitory factor (MIF) was first discovered as a proinflammatory T-cell cytokine (5, 14) but was then shown to be released in substantial quantities by M
as well (8). MIF has recently emerged as a key effector molecule of the innate immune responses against bacteria (3, 7, 9, 10, 22, 25, 26) as well as intracellular pathogens (2, 18, 20, 29). In the present study, we provide evidence for an important role for MIF in the innate immune response against M. tuberculosis.
MIF is released by human M
after incubation with M. tuberculosis antigens or infection with virulent M. tuberculosis.
MIF induction by M
was determined in parallel experiments by incubating cells with immunogenic M. tuberculosis antigens or with a virulent M. tuberculosis strain (H37Rv).
Human M
were derived from blood monocytes obtained from healthy volunteers and prepared by centrifugation over a Ficoll-Hypaque (Seromed, Biochrom, Berlin, Germany) gradient followed by a fibronectin adherence step. Monocytes were resuspended in RPMI 1640 (Life Technologies, Gaithersburg, MD) with 2 mM L-alanyl-L-glutamine (Life Technologies) and 10% fetal calf serum (PAA, Linz, Austria) and plated at 5 x 105 cells/well on 24-well Falcon Primaria plates (Becton Dickinson, Lincoln Park, NJ).
Recognition of mycobacterial products is a crucial step of the host defense response. Host-mycobacterial interactions are primarily mediated by antigens that are found on the M. tuberculosis cell wall. These particularly include lipoarabinomannan (LAM) (31) and 19-kDa lipoprotein (6). ManLAM (derived from the virulent Erdman M. tuberculosis strain) and purified M. tuberculosis 19-kDa lipoprotein were a kind gift from John T. Belisle (Department of Microbiology, Colorado State University, Fort Collins, CO). ManLAM was solubilized in phosphate-buffered saline. Lipopolysaccharide (LPS) contamination was <5 ng/mg as determined by Limulus assay.
H37Rv (American Type Culture Collection, Manassas, VA) was used as the virulent M. tuberculosis strain. Mycobacteria were grown as previously described (23). M
were infected for 12 and 24 h with a suspension of H37Rv strain at a density of 5 x 106 CFU/ml. Cell culture supernatants were then collected and centrifuged in Spin-X filter columns (0.22 µm) for 5 min at 800 x g. MIF concentrations were quantified by enzyme-linked immunosorbent assay (ELISA) (11).
Compared to unstimulated cells, M
incubated with ManLAM and 19-kDa lipoprotein released significant amounts of MIF (Fig. 1). Macrophage MIF induction by M. tuberculosis antigens was found to be concentration dependent (0.1 to 10 µg/ml, peaking at 10 µg/ml) and time dependent (6 to 24 h, peaking at 12 h) (data not shown). Indeed, the virulent H37Rv M. tuberculosis strain also appeared to be a potent inducer of MIF production (Fig. 1). No differences were observed, in terms of amount of MIF released by M
, at 12 and 24 h after infection (data not shown).
![]() View larger version (19K): [in a new window] |
FIG. 1. MIF production by human M (5 x 105 cells/well) incubated with M. tuberculosis antigens (ManLAM and 19-kDa lipoprotein, both at 10 µg/ml) or infected with a virulent (H37Rv) M. tuberculosis strain (5 x 106 bacteria/well). Cell-conditioned media were harvested at 12 h, and MIF content was measured by ELISA. Data are expressed as the mean ± standard error of the mean (SEM) of eight (for ManLAM and H37Rv) and four (for 19-kDa lipoprotein) separate experiments. *, P < 0.05, unpaired t test.
|
during the course of virulent M. tuberculosis growth.
We then analyzed the production of MIF by human M
during the course of M. tuberculosis growth. An in vitro model of infection with virulent H37Rv strain was used (23). Macrophages (5 x 105 M
per well) were infected for 2 h with a single-cell suspension of H37Rv (at a density of 5 x 106 CFU/ml). The cells then were washed three times with RPMI 1640, incubated with RPMI 1640-10% fetal calf serum (PAA), and cultured for 7 days in 24-well plates. The infecting ratio of H37Rv at the time of inoculation was about 10 bacteria/cell leaving, after washes, was from 0.1 to 1 bacterium per macrophage, therefore ensuring optimal conditions for intracellular mycobacterial growth over the ensuing 7 days. To analyze MIF production, cell culture supernatants were processed, as previously described, at days 2, 4, and 7 of M. tuberculosis growth. As shown in Fig. 2, H37Rv-infected M
released abundant quantities of MIF compared to controls, particularly during the exponential phase of mycobacterial growth (4 to 7 days).
![]() View larger version (15K): [in a new window] |
FIG. 2. MIF production by human M during the course of virulent M. tuberculosis growth. Cells were exposed for 2 h to a single-cell suspension of H37Rv (5 x 106 bacteria/well), washed three times, and then incubated for 7 days. Cell-conditioned media were harvested at days 2, 4, and 7, and MIF content was measured by ELISA. Data are expressed as the mean ± SEM of results from four separate experiments. *, P < 0.05, unpaired t test.
|
.
Given that human M
released significant amounts of MIF upon infection with M. tuberculosis and during the course of mycobacterial growth, we next sought to determine the effect of MIF on M. tuberculosis growth. M. tuberculosis growth was assessed by the CFU assay: at the end of the experiment, both adherent and nonadherent cells were lysed separately by incubation with 200 µl of 0.1% saponin (Sigma) in RPMI 1640 for 20 min, collected, and sonicated for 20 s to disperse residual clumps of bacilli. Serial 10-fold dilutions were made in 7H9 broth and plated on 7H10 agar plates. Plates were sealed in plastic, kept at 37°C, and counted after 11 to 13 days with the aid of a dissecting microscope. Bacterial counts were expressed as the number of CFU per well of adherent and nonadherent cell lysate.
We first neutralized activity of MIF released by M
with anti-MIF antibodies. Polyclonal anti-MIF antibodies were generated by immunizing New Zealand White rabbits (Hare Marland, Hewitt, NJ) with purified recombinant human MIF (rhuMIF) as previously described (8). H37Rv-infected human M
were incubated at the beginning of the experiment (day 0) with 100 µg/ml of rabbit polyclonal control or anti-MIF immunoglobulin G (IgG), and M. tuberculosis growth was quantified by CFU counts at day 7. Figure 3 shows absolute CFU at day 0 and day 7 of infected M
treated with control or anti-MIF IgG: when M
were incubated with anti-MIF antibodies, H37Rv growth was significantly enhanced, by 173% ± 20% (mean ± standard deviation [SD] of results from four independent donors, obtained in triplicate well cultures; P = 0.003, unpaired t test). Of note is that H37 growth in cells exposed to IgG control was comparable to that of controls in which no antibodies were added to cultures of infected cells.
![]() View larger version (12K): [in a new window] |
FIG. 3. Anti-MIF antibody enhances the growth of virulent M. tuberculosis within M . M (5 x 105 cells/well) were treated with 100 µg/ml of anti-MIF IgG or control IgG immediately after infection with virulent H37Rv. Cells were lysed for colony counting at day 7. Data are mean CFU ± SD of triplicate cultures of one representative experiment. *, P < 0.05, unpaired t test.
|
.
H37Rv-infected M
then were exposed to exogenous rhuMIF (4). At day 0 and day 4, rhuMIF (1 to 100 ng/ml; LPS contamination, <5 ng/ml) was added to the medium of M
infected with virulent H37Rv. CFU were quantified at day 0 and day 7 and compared to infected M
treated with medium alone. The addition of rhuMIF was associated with a dose-dependent inhibition of mycobacterial growth inside H37Rv-infected M
(Fig. 4). rhuMIF (10 ng/ml) significantly reduced intracellular M. tuberculosis growth by 67.3% ± 12.3% (mean ± SD of results from four separate donors; P = 0.009).
![]() View larger version (22K): [in a new window] |
FIG. 4. rhuMIF inhibits the growth of virulent M. tuberculosis within M . M (5 x 105 cells/well) infected with H37Rv were treated with rhuMIF (1 to 100 ng/ml). Cells were lysed for colony counting at day 7. Data shown are mean CFU ± SD of triplicate well cultures (representative of four separate donors) from one representative experiment. *, P < 0.05, unpaired t test.
|
exposed to ManLAM and 19-kDa lipoprotein (a TLR2-receptor agonist), derived from virulent M. tuberculosis, released significant amounts of MIF (Fig. 1). These soluble cell wall-associated proteins have important immunoregulatory functions, including macrophage activation, and have been shown to be involved in the initiation of the innate immune response against mycobacteria (6, 31). Our findings that both ManLAM and 19-kDa lipoprotein are able to induce macrophage MIF release suggest a potential role for this cytokine as an immune effector against M. tuberculosis. Indeed, this is further substantiated by our observation that live virulent mycobacteria (H37Rv) are potent inducers of macrophage MIF (Fig. 1 and 2). To test this hypothesis, we analyzed the role of MIF in a well-established human in vitro model of infection that employed the virulent H37Rv strain. First, we showed that the neutralization of macrophage MIF release was associated with a significant enhancement of M. tuberculosis growth (Fig. 3). Second, we were able to demonstrate that a dose-dependent inhibition of mycobacterial growth was observed by adding exogenous MIF to infected cells (Fig. 4). Of note is that the magnitude of inhibition of M. tuberculosis growth by MIF in our assay was comparable to that previously demonstrated by various other cytokines (like TNF-
and gamma interferon) and other activators of phagocytic cells (such as vitamin D) (13, 15, 27, 28). Our data therefore indicate that the release of MIF by human M
during the course of the infection is part of the initiation of the innate immune response and contributes to the containment of M. tuberculosis growth in vitro. One important limitation of our study is related to the in vitro nature of the model, in which exogenous cytokines and their respective antibodies are added to the medium. Although this model has been crucial in elucidating the role of other cytokines in the innate immune response against M. tuberculosis (12), it would be of value to confirm the effect of MIF in another important in vitro model of innate immune response, such as the one that uses the cocultures of both lymphocytes and monocytes (30). However, the in vivo observations that MIF acts as an immune effector in the early stages of pulmonary granulomas in mice (19, 21), and the recently reported findings of high levels of MIF in serum in patients with pulmonary tuberculosis (32), substantially support the concept that MIF is involved in host defense mechanisms against M. tuberculosis.
In conclusion, this study shows that MIF secreted by activated M
plays an important role in innate immune defenses against M. tuberculosis in humans by acting in an autocrine fashion to inhibit the growth of virulent mycobacteria. Further studies are needed to elucidate the precise mechanisms by which MIF acts to contain mycobacterial growth. These findings nevertheless may have important clinical bearing given the recent description of functional, genetic polymorphisms in the human Mif gene (1).
We are grateful to Miguel Munoz and Marlies Knaup for skillful technical assistance.
This work was performed at the Institute of Microbiology, University Hospital, Lausanne CH-1011, Switzerland.
|
|
|---|
This article has been cited by other articles:
| |||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
Copyright © 2009 by the American Society for Microbiology. For an alternate route to Journals.ASM.org, visit: http://intl-journals.asm.org | More Info»