The Mycobacterium tuberculosis Early Secreted Antigenic Target of 6 kDa Inhibits T Cell Interferon-γ Production through the p38 Mitogen-activated Protein Kinase Pathway*

We reported previously that the early secreted antigenic target of 6 kDa (ESAT-6) from Mycobacterium tuberculosis directly inhibits human T cell IFN-γ production and proliferation in response to stimulation with anti-CD3 and anti-CD28. To determine the mechanism of this effect, we treated T cells with kinase inhibitors before stimulation with ESAT-6. Only the p38 MAPK inhibitor, SB203580, abrogated ESAT-6-mediated inhibition of IFN-γ production in a dose-dependent manner. SB203580 did not reverse ESAT-6-mediated inhibition of IL-17 and IL-10 production, suggesting a specific effect of SB203580 on IFN-γ production. SB203580 did not act through inhibition of AKT (PKB) as an AKT inhibitor did not affect ESAT-6 inhibition of T cell IFN-γ production and proliferation. ESAT-6 did not reduce IFN-γ production by expanding FoxP3+ T regulatory cells. Incubation of T cells with ESAT-6 induced phosphorylation and increased functional p38 MAPK activity, but not activation of ERK or JNK. Incubation of peripheral blood mononuclear cells with ESAT-6 induced activation of p38 MAPK, and inhibition of p38 MAPK with SB203580 reversed ESAT-6 inhibition of M. tuberculosis-stimulated IFN-γ production by peripheral blood mononuclear cells from subjects with latent tuberculosis infection. Silencing of p38α MAPK with siRNA rendered T cells resistant to ESAT-6 inhibition of IFN-γ production. Taken together, our results demonstrate that ESAT-6 inhibits T cell IFN-γ production in a p38 MAPK-dependent manner.

critical to understand the pathogenesis of tuberculosis and design improved ESAT-6-based vaccines against tuberculosis infection.
In this study, we found that ESAT-6 activated p38 MAPK in T cells without inducing Ca 2ϩ influx, unlike the case for most bacterial pore-forming toxins that activate p38 MAPK. Neutralizing p38 MAPK with a specific chemical inhibitor or with siRNA rendered T cells resistant to ESAT-6-induced inhibition of IFN-␥ production. ESAT-6 did not expand T regulatory cells (Tregs). We conclude that ESAT-6 inhibits human T cell IFN-␥ production in a p38 MAPK-dependent manner.

MATERIALS AND METHODS
Human Subjects-Heparinized venous blood was obtained from eight healthy donors with latent tuberculosis infection and 21 without latent tuberculosis infection, based on results of the QuantiFERON-TB GOLD test. All donors were employees or students at the University of Health Science Center at Tyler, TX. Blood samples were collected after obtaining informed consent, as approved by the Institutional Review Board.
Preparation of Recombinant ESAT-6-The original recombinant plasmid (pET23b) containing Rv3875 was obtained from Colorado State University at Fort Collins through the TB Vaccine Testing and Research Materials Contract. ESAT-6 was cloned from this plasmid and inserted into the same expression plasmid. After confirming the correct sequence of the inserted esat-6 gene by commercial PCR sequencing, the plasmid was transformed into Escherichia coli BL21 (DE3) cells. The cells were cultured in LB medium, and ESAT-6 expression was induced by the addition of isopropyl-D-1-thiogalactopyranoside. Recombinant ESAT-6 was purified, LPS was removed, and its purity was Ͼ95%, based on SDS-PAGE, followed by Coomassie Brilliant Blue staining and Western blotting with anti-ESAT-6 mAb (HYB 76-8), as described previously (14). Recombinant ESAT-6 contained Ͻ50 pg of LPS/mg of protein, by the QCL-1000 limulus amebocyte assay, and was resuspended in Hanks' balanced salt solution (2 mg/ml), aliquoted, and stored at Ϫ70°C until use.
Cell Preparation, Culture, and Detection of Cytokines in Culture Supernatants-PBMC were isolated by Ficoll-Hypaque density gradient centrifugation of heparinized blood, and untouched CD3 ϩ T cells were purified by immunomagnetic negative selection (human Pan T cell isolation kit, Miltenyi Biotec). The purity of CD3 ϩ cells was Ͼ95-98%, as measured by staining with FITC anti-CD3, followed by flow cytometry analysis with a FACSCalibur (BD Biosciences). PBMC or CD3 ϩ cells from QuantiFERON-negative subjects were suspended at 2 ϫ 10 6 cells/ml in RPMI 1640 medium (Invitrogen), supplemented with 10% heat-inactivated pooled human serum (Atlanta Biologicals), 100 units or 100 g/ml penicillin and streptomycin, 1 mM sodium pyruvate, and 0.1 mM minimum essential medium nonessential amino acids (all from Invitrogen).
Flow Cytometry-For intracellular IFN-␥ staining, purified CD3 ϩ cells were treated with SB203580 for 1 h and ESAT-6 for 1 h followed by stimulation with anti-CD3 and anti-CD28 for 48 h, as outlined above. GolgiStop TM (BD Biosciences) was added during the last 6 h of stimulation. The cells were collected, washed, and stained with PE-anti-CD4 (eBiosciences). After 30 min of surface staining, the cells were washed, fixed, and permeabilized and then incubated with FITC-anti-IFN-␥ (BD Pharmingen) for 30 min at room temperature, according to the manufacturer's protocols. The cells were washed, and IFN-␥-positive cells were identified by flow cytometry with a FACSCalibur.
To measure cell proliferation, purified CD3 ϩ cells were labeled with CFSE, as described previously (14), and treated with PK inhibitors and ESAT-6 prior to stimulation with anti-CD3 and anti-CD28, as outlined above. After 96 h, the cells were stained with PE-anti-CD8 (eBiosciences), and CFSE dilution was analyzed by flow cytometry.
Real-time PCR-Total RNA was extracted from 8 ϫ 10 5 CD3 ϩ cells with TRIzol reagent (Invitrogen), cDNA was synthesized, and IFN-␥ cDNA was quantified by real-time PCR, with 18 S ribosomal RNA as an endogenous control, as described previously (14). The relative quantity of IFN-␥ mRNA was calculated by the ⌬C t method (16).
Kinase Assay for p38 MAPK Activity-Purified human CD3 ϩ cells (6 -8 ϫ 10 6 cells/reaction) were incubated with ESAT-6 (3.3 M) or medium alone for different periods, and functional p38 MAPK activity was measured, using the p38 MAPK assay kit (Cell Signaling). Briefly, cellular protein extracts were prepared in ice-cold cell lysis buffer with phosphatase inhibitors and 1 mM PMSF, and activated p38 MAPK was immunoprecipitated from the cell extracts by incubating with bead-immobilized anti-phospho-p38 MAPK (Thr-180/Tyr-182) mAb at 4°C with gentle rocking. After incubation for 16 h, the beads were washed twice with cell lysis buffer and then twice with kinase buffer. The beads were then resuspended in 50 l of kinase buffer, supplemented with 200 M ATP, and incubated at 30°C with recombinant ATF-2 (encompassing amino acids 19 -96) as substrate. After a 30-min incubation, the kinase reaction was terminated by adding 3ϫ SDS-PAGE loading buffer. The samples were boiled for 5 min, resolved by SDS-PAGE, electroblotted to a nitrocellulose membrane, and blotted with anti-phospho-ATF-2 (Thr-76).
Evaluation of Calcium Influx-Freshly purified CD3 ϩ cells from healthy donors were resuspended at 8 ϫ 10 6 /ml in loading buffer (Hanks' solution with 1 mM CaCl 2 , 1 mM MgCl 2 , and 1% FBS) and incubated with the Ca 2ϩ indicator fluorescent dye, fluo-4 AM (Invitrogen), at 4 g/ml at 37°C in the dark. After 30 min, the cells were washed once with loading buffer and resuspended at 2 ϫ 10 6 /ml in loading buffer, supplemented with 10% FBS. The loaded cells were divided into three Eppendorf tubes with 1 ϫ 10 6 cells/tube and cultured either in loading buffer alone as a negative control, with ESAT-6, or with OKT3 as a positive control. The cells with 3.3 M ESAT-6 in loading buffer were incubated at room temperature for 1 h, and the other cells with OKT3 at 1 g/ml were incubated on ice for 15 min followed by incubation on ice with goat anti-mouse IgG at 20 g/ml to cross-link cell-bound OKT3, as described previously (14). The cells with loading buffer alone were kept at room temperature. At the end of treatment, the cells in all three tubes were resuspended in 100 l of ice-cold loading buffer with 10% FBS after removing the free ESAT-6 and antibodies. All the tubes were incubated in a 37°C water bath for 5 min before recording cellular Ca 2ϩ influx by flow cytometry for 250 s.
Transfection of siRNA-Transfection of siRNA was performed by a modification of previously described methods (17). Briefly, purified CD3 ϩ cells were stimulated with phytohemmagglutinin-L (Sigma) at 1 g/ml for 20 h. The cells were collected, washed, and resuspended at 8 -10 ϫ 10 6 /ml in a Dharmacon Accell siRNA delivery medium (Thermo Scientific), supplemented with 5% heat-inactivated pooled human serum, and plated in a 24-well plate at 400 l/well. The cells were then transfected with scrambled siRNA or different concentrations of SMARTpool siRNA for p38 MAPK␣ (MAPK14 on target plus smart pool) in 3 l of DharmaFECT1 transfection reagent/ well, based on pilot experiments (all reagents from Thermo Scientific). Fresh Dharmacon Accell siRNA delivery medium (2 ml/well) was then added with 10% human serum. Forty-eight h after transfection, the cells were collected, washed with RPMI 1640 without serum, and resuspended at 2 ϫ 10 6 /ml in RPMI 1640 medium with 10% human serum. The cells were treated with ESAT-6 at 3.3 M for 1 h at 37°C and 5% CO 2 . The cells were then plated in a 96-well plate precoated with ␣-CD3 (5 g/ml) and ␣-CD28 (1 g/ml) and incubated at 37°C with 5% CO 2 . After 72 h, the culture supernatants were collected, and IFN-␥ levels were measured by ELISA. To determine the effect of siRNA on p38 MAPK expression, some cells were collected 48 and 72 h after transfection and lysed in 1ϫ SDS-PAGE loading buffer. After boiling for 8 min, the proteins were resolved on a 10% SDS-PAGE gel and electroblotted to a nitrocellulose membrane, and Western blotting was performed with anti-p38␣ MAPK (Santa Cruz Biotechnology). To control for protein loading, the blot was stripped and reblotted with anti-GAPDH.

Inhibition of p38 MAPK Reverses ESAT-6-mediated Reduction of IFN-␥ Production by T Cells-
We have shown previously that ESAT-6 of M. tuberculosis binds to human T cells and inhibits T cell IFN-␥ production and proliferation in response to stimulation with anti-CD3 and anti-CD28, without affecting cell viability or T cell receptor proximal signaling pathways that activate Zap70 (14). We hypothesized that ESAT-6 inhibits IFN-␥ production through effects on the major PKs, which control T cell cytokine production and proliferation and are affected by toxins of several bacterial pathogens, including Helicobacter pylori (18,19), Bacillus anthracis (20 -22), and Bordetella pertussis (23). To test this, we pretreated T cells from five healthy donors with different PK inhibitors and then stimulated T cells in the absence or presence of ESAT-6. The concentrations of inhibitors used were based on the literature and on our pilot experiments. Stimulation of T cells with anti-CD3 and anti-CD28 induced high levels of IFN-␥ (14,207 Ϯ 2,739 pg/ml), which were significantly reduced by ESAT-6 (4,147 Ϯ 726 pg/ml), as we reported (14). Pretreatment with the p38 MAPK inhibitor, SB203580, at both concentrations tested restored IFN-␥ production by T cells in the presence of ESAT-6 (11,020 Ϯ 1,469 and 12,710 Ϯ 1,444 pg/ml with SB203580 versus 4,147 Ϯ 726 pg/ml with ESAT-6, p ϭ 0.007 for both comparisons, Fig. 1A). The other PK inhibitors did not affect ESAT-6 inhibition of IFN-␥ production, and inhibitors of PI3K and PKC further reduced IFN-␥ levels, perhaps because these kinases are required for T cell activation (24) and IFN-␥ production (25). To confirm the effects of p38 MAPK inhibition, we evaluated the effect of different concentrations of SB203580 on anti-CD3-plus anti-CD28-stimulated IFN-␥ production by T cells from a larger group of 13 healthy donors (Fig. 1B). SB203580 did not affect IFN-␥ production by anti-CD3-plus anti-CD28-stimulated cells but negated the capacity of ESAT-6 to inhibit anti-CD3-plus anti-CD28-induced IFN-␥ production in a dose-dependent manner. We next determined whether the effect of SB203580 was mediated at the transcriptional level (Fig. 1C). ESAT-6 inhibited IFN-␥ transcription in T cells, as we previously demonstrated (14). SB203580 did not affect anti-CD3-plus anti-CD28-induced IFN-␥ mRNA expression by T cells but completely abrogated the capacity of ESAT-6 to reduce anti-CD3-plus anti-CD28-stimulated IFN-␥ transcription. Thus, the results from this study suggest that inhibition of p38 MAPK abrogates ESAT-6 inhibition of T cell IFN-␥ production at the transcriptional level.
p38 MAPK Inhibition Does Not Affect the Capacity of ESAT-6 to Reduce Production of IL-10 or IL-17-We previously showed that ESAT-6 inhibits T cell production of IL-10 and IL-17, as well as IFN-␥ (14). We treated the T cells with different concentrations of SB203580 and ESAT-6 before stimulation with anti-CD3 and anti-CD28. In the absence of ESAT-6, the addition of increasing concentrations of SB203580 reduced levels of IL-10 induced by anti-CD3 plus anti-CD28 stimulation in a dose-dependent manner with significant reduction at 20 M concentration (Fig. 1D). This confirms a previous study that IL-10 production by T cells in response to T cell receptor stimulation depends in part on p38 MAPK (26). Anti-CD3 and anti-CD28 stimulation induced increased IL-17 production by T cells, and the presence of increased concentrations of SB203580 did not significantly reduce IL-17 production (Fig. 1E), suggesting that T cell production of IL-17 is p38 MAPK-independent. ESAT-6 reduced anti-CD3-stimulated IL-10 ( Fig. 1D) and IL-17 levels (Fig. 1E) significantly, but these levels were not restored by preincubation with SB203580. Therefore, inhibition of p38 MAPK abrogated the inhibitory effect of ESAT-6 on T cell IFN-␥ production but did not affect ESAT-6 inhibition of IL-10 and IL-17 production.
ESAT-6 Inhibits IFN-␥ Production by Both CD4ϩ and CD8ϩ T Cells through p38 MAPK-To determine whether p38 MAPK mediates the effects of ESAT-6 on IFN-␥ production and the number of IFN-␥-producing CD4ϩ and CD8ϩ T cells, we used intracellular staining and flow cytometry analysis to evaluate IFN-␥-producing cells. Stimulation of T cells with anti-CD3 and anti-CD28 markedly increased the percentage of IFN-␥producing CD4ϩ and CD8ϩ cells ( Fig. 2A). ESAT-6 reduced the frequency of IFN-␥ϩ cells by 50 -80%, with similar degrees of reduction for CD4ϩ and CD8ϩ cells (Fig. 2, A and B). Treatment of cells with SB203580 before ESAT-6 largely restored the percentages of IFN-␥ϩCD4ϩ and CD8ϩ cells (Fig. 2, A and B), suggesting that ESAT-6-mediated inhibition of IFN-␥ production and reduction of the frequency of IFN-␥-producing CD4ϩ and CD8ϩ T cells act through the p38 MAPK pathway.
Inhibition of p38 MAPK Partially Reverses ESAT-6-mediated Reduction in T Cell Proliferation-ESAT-6 inhibits anti-CD3plus anti-CD28-induced T cell proliferation and IFN-␥ produc-  JULY 8, 2011 • VOLUME 286 • NUMBER 27 tion (14). To determine whether ESAT-6 inhibition of T cell proliferation is mediated through the p38 MAPK, we used CFSE-stained T cells and flow cytometry to analyze T cell proliferation (Fig. 3). ESAT-6 markedly inhibited proliferation of both CD4ϩ and CD8ϩ cells, and pretreatment with SB203580 enhanced proliferation of ESAT-6-treated T cells. However, this effect was not statistically significant and was much less than the effect of SB203580 on IFN-␥ production (compare Figs. 1B and 3B). Therefore, ESAT-6-mediated inhibition of T cell proliferation is not mediated primarily through p38 MAPK, and the increase in IFN-␥-producing cells by p38 MAPK inhibition (Fig. 2) is not merely due to increased T cell proliferation.

Protein Kinases and ESAT-6 Inhibition of T Cell IFN-␥
ESAT-6 Inhibition of T Cell IFN-␥ Production Is AKT Kinase-independent-High concentrations of the p38 MAPK inhibitor, SB203580, can also inhibit AKT in T cells as an offtarget effect to reduce IL-2-dependent proliferation of a murine T cell line (27). To determine whether the same mechanism (AKT) contributed to the effects of SB203580 observed above, we used an AKT-specific inhibitor. The results (Fig. 4) demonstrated that inhibition of AKT did not affect ESAT-6-mediated suppression of IFN-␥ production and proliferation (not shown). This suggests that ESAT-6 inhibits IFN-␥ production through p38 MAPK but not through AKT.
ESAT-6 Inhibition of T Cell IFN-␥ Production Is Not Mediated by Tregs-p38 MAPK is critical for generation of induced Tregs (28) and for Tregs to exhibit suppressor function (29,30). To determine whether the effects of p38 MAPK inhibition observed above were mediated through reducing the capacity of ESAT-6 to enhance Treg function, we studied the effect of ESAT-6 on generation of Tregs by staining for FoxP3 and CD25. ESAT-6 reduced the percentage of anti-CD3-plus anti-CD28-stimulated FoxP3ϩCD25ϩ cells (Fig.  5, A and B), indicating that inhibition of IFN-␥ production by ESAT-6 was not due to increased numbers of Tregs. The addition of SB203580 to ESAT-6-treated cells increased the number of Tregs, perhaps because Tregs typically expand when the number of activated T cells increases. These findings indicate that the effects of SB203580 and ESAT-6 on IFN-␥ production in our experimental system were unlikely to be mediated by Tregs.
ESAT-6 Directly Activates p38 MAPK in T Cells-Because our data demonstrated that ESAT-6 inhibits IFN-␥ production through p38 MAPK, we hypothesized that ESAT-6 directly activates p38 MAPK in T cells. To test this, we incubated freshly isolated T cells with 3.3 M ESAT-6 for different time points and evaluated the activation of p38 MAPK. Treatment of T cells with ESAT-6 for 30 -60 min induced phosphorylation of p38 MAPK, based on Western blotting with anti-phospho-p38 MAPK (Thr-180/Tyr-182), which detects the active form of the kinase (Fig. 6A). ESAT-6 did not phosphorylate ERK or JNK. We next immunoprecipitated the ESAT-6-induced phosphorylated p38 MAPK from T cell extracts and measured its functional kinase activity based on phosphorylation of its specific substrate, ATF-2. Incubation of T cells with ESAT-6 for as little as 15 min induced p38 MAPK activity, and this increased further until 60 min after incubation (Fig. 6B). Western blotting also demonstrated increased ATF-2 phosphorylation in ESAT-6-treated T cells, as compared with untreated cells (not shown), confirming that ESAT-6 induced p38 MAPK activity in T cells. These results strongly suggest that ESAT-6 directly activates p38 MAPK in T cells.
ESAT-6 Does Not Induce Ca 2ϩ Influx in T Cells-Many bacterial virulence factors are pore-forming toxins that induce extracellular calcium influx (31), which in turn activates p38 MAPK as a cellular protective mechanism (31-33). ESAT-6 can induce cytolysis of alveolar epithelial cells and macrophages (34 -36) and cause pore formation in mammalian cells (37). To determine whether ESAT-6 activates p38 MAPK through pore formation and increased calcium influx in our experimental

Protein Kinases and ESAT-6 Inhibition of T Cell IFN-␥
system, we loaded T cells with a Ca 2ϩ -sensitive fluorescent dye and then treated the cells with ESAT-6 and evaluated the changes in intracellular calcium by flow cytometry. ESAT-6 did not affect intracellular calcium levels (Fig. 7). In contrast, ␣-CD3 cross-linking significantly increased cellular calcium, as compared with that in control cells incubated in medium alone, suggesting that ESAT-6-mediated activation of p38 MAPK is not due to membrane pore formation and increased cellular calcium levels.
Inhibition of p38 MAPK by siRNA Reverses ESAT-6-mediated Reduction of IFN-␥ Production by T Cells-Our results above show that a chemical inhibitor of p38 MAPK reverses ESAT-6mediated reduction of IFN-␥ production by T cells (Fig. 1). To confirm that this was due to inhibition of p38 MAPK, we silenced p38␣ MAPK in T cells with siRNA and then incubated the cells with ESAT-6 and measured anti-CD3-induced IFN-␥ production. In four donors, p38␣ MAPK siRNA reduced p38␣ expression in a dose-dependent manner, with 40 nM lowering expression by Ͼ85%, as compared with scrambled siRNA at the same concentration (Fig. 8A). p38 MAPK siRNA also abrogated the capacity of ESAT-6 to inhibit IFN-␥ production in response to anti-CD3 plus anti-CD28 (Fig. 8B), indicating that p38␣ MAPK is required for this effect. We conclude that ESAT-6 inhibits IFN-␥ production by directly activating p38 MAPK in T cells.
ESAT-6 Inhibits M. tuberculosis-induced IFN-␥ Production through p38 MAPK-Our results above show that ESAT-6 inhibits IFN-␥ production in response to T cell receptor stimulation in a p38 MAPK-dependent manner. To determine whether ESAT-6 also inhibits IFN-␥ production in response to bacterial antigens through this mechanism, we evaluated the response of SB203580-treated PBMC to stimulation with heat-killed M. tuberculosis. Pretreatment of PBMC with SB203580 abrogated the capacity of ESAT-6 to inhibit antigen-induced IFN-␥ production (16,216 Ϯ 3,210 versus 6,806 Ϯ 2,557 pg/ml, p Ͻ 0.05, Fig. 9A), and SB203580 did not affect M. tuberculosis-stimulated IFN-␥ production by PBMC in the absence of ESAT-6. When PBMC were incubated with ESAT-6, Western blotting showed marked accumulation of phosphorylated p38 MAPK over 60 min (Fig.  9B). ERK was activated to a lesser extent, and JNK MAPK was not affected. These findings suggest that ESAT-6 inhibits antigen-induced and T cell receptor-elicited IFN-␥ production through p38 MAPK activation.

DISCUSSION
ESAT-6 is a secreted protein of M. tuberculosis that contributes to virulence through multiple mechanisms, including direct suppression of T cell production of IFN-␥, which is central to immune defenses against mycobacteria. In the current study, we demonstrated that inhibition of p38 MAPK with a chemical inhibitor or with siRNA abrogated the ESAT-6-mediated reduction of IFN-␥ production by anti-CD3-stimulated T cells and M. tuberculosis-stimulated PBMC, affecting both CD4ϩ and CD8ϩ T cells. Inhibition of p38 MAPK reversed the  effects of ESAT-6 on IFN-␥ production at the IFN-␥ mRNA expression, and these effects are neither through Tregs nor through AKT. These effects were specific for IFN-␥ as inhibition of p38 MAPK did not affect the capacity of ESAT-6 to reduce production of IL-10 and IL-17. Stimulation of T cells with ESAT-6 induced phosphorylation and increased functional kinase activity of p38 MAPK, but not ERK or JNK. Our results demonstrate that ESAT-6 inhibits T cell IFN-␥ production by inducing functional p38 MAPK activity in T cells, providing a new mechanism by which M. tuberculosis subverts host immune responses.
As a facultative intracellular pathogen that establishes chronic infection in humans, M. tuberculosis has evolved multiple mechanisms to affect intracellular signaling pathways to subvert host immunity (38). Most studies have focused on the effects of M. tuberculosis on mononuclear phagocytes, particularly on the signaling pathways that affect phagosomal maturation and apoptosis. M. tuberculosis activates p38 MAPK to reduce recruitment of early endosome autoantigen 1 to the phagosomal membranes and delay phagosomal maturation (39). In addition, M. tuberculosis activates p38 MAPK in monocytes through complement receptor 3 to reduce CD1 expression on monocyte-derived dendritic cells, limiting their capacity to present mycobacterial antigens to T cells (40). M. tuberculosis also activates p38 MAPK in neutrophils to induce apoptosis and inhibit expression of chemokine receptor, CXCR2 (41)(42)(43). Activation of p38 MAPK in macrophages and neutrophils is thought to be mediated by mannose-capped lipoarabinomannan, the major component of the M. tuberculosis cell wall (43). Thus, these studies suggest that M. tuberculosis activates p38 MAPK in neutrophils and monocytes to attenuate host immunity against M. tuberculosis. Our current results provide the first evidence that the secreted mycobacterial protein, ESAT-6, activates p38 MAPK in human T cells to inhibit production of T cell IFN-␥, which is central for protective immunity against tuberculosis infection. This mechanism is reminiscent of that used by herpes simplex virus, which activates p38 MAPK in infected T cells, inducing expansion of IL-10-producing T cells instead of protective IL-2-and IFN-␥-producing T cells to subvert host immune response (44). The cells were collected, and total protein extracts were prepared. Western blotting was performed with phospho-specific (p) antibodies against p38, ERK, and JNK. The same blot was stripped and blotted for vinculin as a protein loading control. A representative result of experiments performed on T cells from four donors is shown. B, T cells were treated with 3.3 M ESAT-6 for different periods, and total cell protein extracts were prepared. Phosphorylated p38 MAPK was immunoprecipitated with anti-phospho-p38, and the kinase activity in the immunoprecipitates was determined by Western blotting to measure the ability to phosphorylate the p38 MAPK substrate, recombinant ATF-2, as detailed under "Materials and Methods." Total p38 MAPK levels in the protein extracts before immunoprecipitation, measured by Western blotting, were used as input controls. A representative result from experiments with three donors is shown.  MAPs, composed of ERK, JNK, and p38 MAPK, play crucial roles in regulating cellular activation, cytokine production, cell death, and proliferation of T cells (45). Functional activity of MAPK requires phosphorylation of critical amino acids, and activation of p38 MAPK depends on dual phosphorylation at Thr-180 and Tyr-182 of its substrate recognition site, through the MAPK activation cascade (46). Several bacterial toxins, including the H. pylori vacuolating toxin (18,19), B. anthracis lethal toxin (20 -22), and B. pertussis adenylate cyclase (23), activate p38 MAPK by increasing Ca 2ϩ influx through forming cell membrane pores in multiple cell types. However, ESAT-6 did not act through this mechanism as it did not affect cell viability (14) or Ca 2ϩ influx (Fig. 7). p38 MAPKs are believed to be downstream signaling molecules that are mobilized after cellular activation by TLRs or cytokines. Although ESAT-6 was reported to act through TLR2 in macrophages to inhibit TLR4stimulated IL-12 production (47), this is unlikely to be the pathway for p38 MAPK activation in T cells as ESAT-6 reduced anti-CD3-plus anti-CD28-stimulated TLR2 expression on T cells and anti-TLR1/2 antibodies had no effect on ESAT-6mediated inhibition of anti-CD3-induced IFN-␥ production. 7 Furthermore, activation induces T cells to express TLR2, which acts as a costimulatory receptor to enhance T cell proliferation and IFN-␥ production (48). ESAT-6-mediated activation of p38 in T cells is unlikely to be the consequence of autocrine effects of ESAT-6-induced cytokines because p38 MAPK activation occurred only 15 min after the addition of ESAT-6. This period is too short for secretion of sufficient concentrations of inflammatory cytokines to activate p38 MAPK. ESAT-6 was shown to stimulate macrophages to produce NO and express surface molecules, possibly through MAPK pathways (49). Thus, the mechanism by which ESAT-6 activates p38 MAPK in T cells remains uncertain but is a fruitful topic for future investigation.
Because M. tuberculosis is an intracellular pathogen that resides in phagosomes surrounded by bilayer membranes, how would a secreted protein, such as ESAT-6, contact T cells to modulate p38 MAPK activity during infection in vivo? Recent studies have shown that exosomes, which are membranecoated vesicles, are released by M. tuberculosis-infected macrophages, contain ESAT-6 and other mycobacterial proteins (50), and stimulate naive T cells (51). These studies support the possibility that ESAT-6 may come in contact with T cells during infection.
The role of p38 MAPK in T cell proliferation and cytokine production depends on the mode of T cell stimulation. When T cells are activated by cytokines such as IL-12 and IL-18, p38 MAPK enhances IFN-␥ gene transcription, in part through activation of the IL-12-responsive transcription factor, STAT4 (52)(53)(54)(55). In contrast, when T cells are activated through the T cell receptor, p38 MAPK inhibits proliferation and does not affect IFN-␥ production (56,57). p38 MAPK activation can also inhibit cytokine production as it reduces IL-2 and IFN-␥ production by natural killer T cells in mice (58). We found that ESAT-6-mediated inhibition of anti-CD3-stimulated production of IFN-␥, but not IL-10 or IL-17, depended on activation of p38 MAPK, suggesting a specific effect of p38 MAPK on ESAT-6 inhibition of IFN-␥ production. p38 MAPK activation can dampen T cell responses by enhancing the expansion and functional activity of Tregs (28,29), but ESAT-6 does not increase the numbers of FoxP3ϩ Tregs (Fig. 5), suggesting that ESAT-6-mediated p38 MAPK activation does not inhibit IFN-␥ production through this mechanism. We considered the possibility that ESAT-6 reduces the supply of p38 MAPK available to T cells. However, this is unlikely because p38 MAPK is necessary for production of 7 B. Samten, unpublished data. Th2 cytokines but dispensable for production of IFN-␥ by T cell receptor-activated primary T cells (56,57) or by T cells from mice with a p38␣ MAPK gene deletion (59). Consistent with these studies, our data demonstrated that inhibition of p38 MAPK did not affect T cell IFN-␥ production in response to either anti-CD3 or M. tuberculosis antigens in the absence of ESAT-6 ( Figs. 1B and 9). The molecular mechanisms for ESAT-6 inhibition of T cell IFN-␥ production via activated p38 MAPK remain to be explored.
In summary, our findings provide the first evidence that a microbial product can activate p38 MAPK to inhibit production of a cytokine that is central to protective immunity against M. tuberculosis and other intracellular pathogens. Further studies to decipher the pathways by which ESAT-6 mediates these effects will provide insight into the mechanisms for pathogen-mediated inhibition of T cell responses and facilitate development of immunomodulatory therapy to reverse these effects.