|Home | About | Journals | Submit | Contact Us | Français|
Immune evasion is required for Mycobacterium tuberculosis to survive in the face of robust adaptive CD4+ T-cell responses. We have previously shown that M. tuberculosis can indirectly inhibit CD4+ T cells by suppressing the major histocompatibility complex class II antigen-presenting cell function of macrophages. This study was undertaken to determine if M. tuberculosis could directly inhibit CD4+ T-cell activation. Murine CD4+ T cells were purified from spleens by negative immunoaffinity selection followed by flow sorting. Purified CD4+ T cells were activated for 16 to 48 h with CD3 and CD28 monoclonal antibodies in the presence or absence of M. tuberculosis and its subcellular fractions. CD4+ T-cell activation was measured by interleukin 2 production, proliferation, and expression of activation markers, all of which were decreased in the presence of M. tuberculosis. Fractionation identified that M. tuberculosis cell wall glycolipids, specifically, phosphatidylinositol mannoside and mannose-capped lipoarabinomannan, were potent inhibitors. Glycolipid-mediated inhibition was not dependent on Toll-like receptor signaling and could be bypassed through stimulation with phorbol 12-myristate 13-acetate and ionomycin. ZAP-70 phosphorylation was decreased in the presence of M. tuberculosis glycolipids, indicating that M. tuberculosis glycolipids directly inhibited CD4+ T-cell activation by interfering with proximal T-cell-receptor signaling.
Aerosolized Mycobacterium tuberculosis infects alveolar and lung parenchymal macrophages, where it replicates unrestrained in the face of innate responses until T-cell immunity controls its growth. Despite robust activation of innate and adaptive immunity, M. tuberculosis survives and persists as a latent infection (15, 18). CD4+ T cells have a central role in controlling M. tuberculosis during acute and latent infections (53). Animal studies have shown that depletion or absence of CD4+ T cells during primary infection results in unchecked M. tuberculosis growth in the lung and decreased survival (37, 38, 41). Depletion of CD4+ T cells during latent infection also worsens disease and survival (46). In humans, loss of CD4+ T cells from progressive human immunodeficiency virus infection is directly responsible for the high rates of tuberculosis in human immunodeficiency virus-infected persons (48).
Much is known about how M. tuberculosis manipulates macrophages for its survival (19, 29, 44). However, the way in which M. tuberculosis interferes with adaptive T-cell immunity is not well understood. Our recent studies have demonstrated that M. tuberculosis can modulate CD4+ T-cell function both indirectly and directly. M. tuberculosis, through Toll-like receptor 2 (TLR-2), inhibits gamma-interferon-regulated genes that result in decreased major histocompatibility complex class II (MHC-II) antigen processing by macrophages for effector and memory CD4+ T cells (22, 23, 40, 43). M. tuberculosis can also induce increased adhesion to fibronectin through α5β1 integrin on CD4+ T cells (45).
M. tuberculosis molecules responsible for modulating CD4+ T-cell function reside in the mycobacterial cell wall and include the lipoproteins LpqH, LprG, and LprA as well as the glycolipid phosphatidylinositol mannoside (PIM). The lipoproteins bind to TLR-2 on macrophages, and PIM binds to VLA-5 (α5β1) on CD4+ T cells. The M. tuberculosis cell wall also contains lipoarabinomannan (LAM); complex lipids, such as phthiocerol dimycocerosate, and cord factor/dimycolytrehalose; and sulfolipids that are both targets of the immune response (e.g., glycolipids presented by CD1 to T cells) and agonists of host cell receptors (9, 10, 32). Although M. tuberculosis bacilli largely reside within macrophages, mycobacterial cell wall components, including glycolipids, can traffic outside infected macrophages through the production of exosomes. These exosomes can then deliver M. tuberculosis molecules to T cells and other host cells that are not directly interacting with M. tuberculosis-infected cells and thereby affect host immune responses (3, 4).
The purpose of this study was to determine if M. tuberculosis can directly (i.e., independently from its effect on MHC-II antigen processing) interfere with CD4+ T-cell activation and, if so, what the M. tuberculosis molecules(s) and mechanism(s) are. Highly purified murine CD4+ T cells devoid of antigen-presenting cells (APCs) were activated by CD3 and CD28 monoclonal antibodies (MAbs) in the presence or absence of M. tuberculosis and biochemical fractions. We have found that M. tuberculosis bacilli directly inhibited CD4+ T-cell activation. Biochemical fractionation identified cell wall glycolipids as potent inhibitors of signaling through the T-cell receptor (TCR) complex by interfering with ZAP-70 phosphorylation.
Eight- to 10-week-old female C57BL/6 mice were purchased from Charles River Laboratories (Wilmington, MA). D011.10 TCR transgenic mice that express TCRs specific for the OVA323-339 peptide presented in the context of I-Ad (39) were a gift from Alan Levine (Case Western Reserve University, Cleveland, OH). TLR2−/− and MyD88−/− mice were generously provided by Shizuo Akira (Research Institute for Microbial Disease, Osaka University, Osaka, Japan) and were backcrossed to C57BL/6 mice a minimum of eight times. Mice were housed under specific-pathogen-free conditions. Studies were approved by the Institutional Animal Care and Use Committee at Case Western Reserve University.
The following MAbs were purchased for analysis of surface marker expression: phycoerythrin (PE)-conjugated CD3 (145-2C11), fluorescein isothiocyanate-conjugated CD4 (GK1.5) from BD Biosciences (San Jose, CA), APC-conjugated CD28 (37.1), PE-conjugated interleukin 2Rα (IL-2Rα; PC61.5), PE-Cy7-conjugated CD62L (MEL-14) from eBioscience (San Diego, CA), PeCy5-conjugated CD69 (H1.2F3), PE-conjugated CD44 (IM7.8.1) from Invitrogen (Carlsbad, CA), and conjugated isotype controls. For T-cell activation, hamster anti-mouse CD3 (145-2C11), anti-mouse CD28(L3T4), and mouse anti-hamster secondary immunoglobulin G1 (IgG1) were purchased from BD Biosciences. Rabbit MAbs against ZAP-70 and phosphorylated ZAP-70 (Tyr319) were purchased from Cell Signaling Technologies (Danvers, MA). Horseradish peroxidase (HRP)-conjugated secondary MAb (Jackson ImmunoResearch, West Grove, PA) was used for detection. Mannose-capped LAM (ManLAM) and PIM from M. tuberculosis H37Rv were obtained through the Tuberculosis Vaccine Testing and Research Materials Contract (NIAID HHSN266200400091C) at Colorado State University.
M. tuberculosis H37Ra (American Type Culture Collection, Manassas, VA) was grown to log phase in Middlebrook 7H9 medium (Difco, Detroit, MI) with albumin, dextrose, and catalase enrichments (Difco); harvested; and frozen at −80°C as previously described (40). Bacterial titer was determined by counting CFU on 7H10 medium (Difco). M. tuberculosis lysate was prepared by passing bacterial pellets resuspended in deionized water containing 7.5 mM EDTA, 0.7 μg/ml leupeptin (Sigma, St. Louis, MO), 0.2 mM phenylmethylsulfonyl fluoride (Sigma), 0.7 μg/ml pepstatin A (Sigma), 10 U/ml DNase (Sigma), and 25 U/ml RNase (Sigma) through a French press four times (40). Intact cells and cellular particulate were removed by centrifuging the suspension for 15 min at 3,000 rpm. Protein concentrations were estimated by bicinchoninic acid protein assay (Thermo Scientific, Rockford, IL). Lysate was stored at −80°C.
Purification of glycolipids from M. tuberculosis H37Ra was performed using a modified protocol previously described (22, 23, 43). M. tuberculosis lysate was extracted with 6% Triton X-114 (Sigma). The detergent layer was washed three times with Tris-buffered saline (50 mM Tris, 150 mM NaCl, pH 7.5) and once with cold water and then precipitated by overnight incubation at −20°C with cold acetone (Fischer Scientific, Pittsburgh, PA). The pellet was washed once with cold acetone and resuspended in phosphate-buffered saline (PBS). Acetone precipitate was then mixed with an equal volume of PBS-saturated phenol (Fischer Scientific) and vigorously shaken overnight at room temperature. The mixture was then centrifuged at 15,000 × g for 30 min and the aqueous layer removed and replaced with fresh PBS. Following multiple washes with PBS, the organic phase was mixed with PBS and vigorously shaken overnight. The aqueous layer was once again removed and the organic phase washed multiple times with PBS. The organic phase was dialyzed (1-kDa cellulose acetate dialysis membrane; Fischer Scientific) against running water for 48 h. After dialysis, the supernatant was lyophilized, weighed, reconstituted in PBS, stored at −80°C, and designated M. tuberculosis glycolipids.
Unless otherwise specified, all experiments were performed at 37°C in a 5% CO2 atmosphere and in serum-free HL-1 medium (BioWhittaker, East Rutherford, NJ) supplemented with 1 μM 2-mercaptoethanol, 10 mM HEPES buffer, nonessential amino acids, 2 mM l-glutamine, 100 μg of streptomycin, and 100 U of penicillin (complete HL-1; BioWhittaker). Splenocytes from 8- to 10-week-old C57BL/6 mice were isolated, and red blood cells were lysed in hypotonic lysis buffer (10 mM Tris-HCl and 0.83% ammonium chloride). Splenocytes were plated in 100-mm tissue culture plates and allowed to adhere for 1 h at 37°C. Untouched CD4+ T cells were purified from nonadherent splenocytes using a CD4+ T-cell-negative isolation kit (Miltenyi Biotec, Germany) by following the manufacturer's instructions. The purity of CD4+ T cells was confirmed by flow cytometry and ranged between 88 and 95%. CD4+ T cells then were further purified by staining with anti-CD4 monoclonal antibody and positive selection through fluorescence-activated cell sorting using the BD Aria fluorescence-activated cell sorting system (BD Biosciences). The resulting CD4+ T-cell populations were 98 to 99% pure. For purification of naïve CD4+ T cells, negatively selected CD4+ T cells were positively selected for CD4 and then sorted by their CD44 and CD62L surface levels.
To test for CD4+ T-cell viability, cells were assayed for mitochondrial metabolism with 3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium bromide (MTT; Sigma). CD4+ T cells were harvested, counted, and resuspended in HL-1 (1 × 106 cells/ml). One hundred microliters of cells was added to a 96-well, flat-bottomed-well plate (Becton Dickinson, Franklin Lakes, NJ) along with 25 μl of MTT (5 mg/ml in PBS) for 2 h at 37°C. The reaction was ceased by the addition of 100 μl of stop solution (5 g sodium dodecyl sulfate, 25 ml of dimethyl formamide, 1 ml acetic acid, 1.25 ml 1 N HCl in water). The plate was incubated overnight, and the absorbance was read at a 570-nm wavelength using the Versa Max turntable microplate reader and the Soft Max Pro LS computer analysis software. Wells with medium alone were used to subtract background absorbance.
Splenocytes from naïve DO11.10 transgenic mice were isolated, and red blood cells were lysed with a hypotonic lysis buffer. Effector cells were generated by 24 h of stimulation with 100 nM OVA32-339 peptide (AnaSpec, San Jose, CA) followed by a 7-day culture in 100 U of IL-2 in Dulbecco's modified Eagle's medium (DMEM; BioWhittaker) supplemented as indicated above for complete HL-1 with the addition of 10% heat-inactivated fetal bovine serum (HyClone, Logan, UT). Nonviable cells were removed by density gradient centrifugation using Ficoll-Paque (GE Healthcare). Remaining cells rested for 24 h in complete DMEM and were used in T-cell activation assays. As assessed by flow cytometry, ~92% of the cells were CD4+ and 72 to 80% had the high-level CD44 (CD44high) and low-level CD62L (CD62Llow) phenotype.
CD4+ T cells (1 × 105 cells/well) were activated in 96-well, flat-bottomed microtiter plates with 1 μg/ml of soluble CD28 MAb in wells precoated with 1 μg/ml of CD3 MAb. Cells were stimulated in the presence or absence of M. tuberculosis or its fractions for 48 h, when IL-2 and proliferation were measured. Supernatants from activated CD4+ T cells were assayed for IL-2 production using a set of capture and detection MAbs from eBioscience previously described (11). Briefly, diluted supernatants were added to Immulon 4HBX flat-bottom microtiter plates (Thermo) coated with purified IL-2 MAb (1 μg/ml). Biotinylated IL-2 MAb (1 μg/ml) was added, followed by alkaline phosphatase-conjugated streptavidin (Jackson ImmunoResearch) and phosphatase substrate (Sigma). Plates were read using a Versa Max turntable microplate reader and data analyzed with the Soft Max Pro LS computer analysis software.
To measure CD4+ T-cell proliferation, cells were pulsed with 1 μCi of [3H]thymidine (MP Biomedicals, Solon, OH) for an additional 16 h. Cells were harvested onto glass fiber filters (Packard, The Netherlands) with a Filtermate 196 Harvester (Packard), and 3[3H]thymidine incorporation was measured with a Matrix 96 direct beta counter (Perkin Elmer, Waltham, MA).
For activation of CD4+ T cells by phorbol 12-myristate 13-acetate (PMA) and ionomycin, T cells were cultured in 96-well, flat-bottomed microtiter plates with increasing concentrations of PMA (0.05 to 5 ng/ml; Sigma) and 0.1 μg/ml of ionomycin (Sigma) for 6 h in the presence or absence of M. tuberculosis glycolipids (10 μg/ml). For overnight activation, CD4+ T cells were activated with 0.05 ng/ml of PMA and 0.01 μg/ml of ionomycin.
For surface receptor expression on CD4+ T cells by flow cytometry, cells (1 × 106 cells/well) were stimulated with plate-bound CD3 and soluble CD28 MAbs in 24-well plates (Falcon) for either 16 h (CD69) or 48 h (CD3, CD28, CD25) with or without M. tuberculosis glycolipids (10 μg/ml). Following stimulation, harvested cells were incubated with the appropriate MAbs, fixed with 2% paraformaldehyde, and analyzed with a FACSCalibur instrument (BD Biosciences).
CD4+ T cells were rested overnight in complete DMEM supplemented with 1% fetal bovine serum. T cells (3 × 107 cells/ml) were resuspended in complete HL-1 medium in 1.5 ml Eppendorf tubes (LPS, Rochester, NY) in a 100-μl total volume. Glycolipid (10 μg/ml) or medium alone was either added at the time of activation or incubated for 1 h at 37°C before activation. The TCR complex was activated by adding a cross-linking mouse anti-hamster IgG1 (10 μg/ml) for 2 min before the addition of anti-CD3 (10 mg/ml) for 2 min. CD4+ T-cell activation was stopped by adding 100 μl of 2× Laemmli sample buffer and boiling the samples for 10 min. Unstimulated cells, incubated with mouse anti-hamster IgG1 alone, served as the control for nonspecific TCR-CD3 activation.
Proteins were separated by sodium dodecyl sulfate-polyacrylamide gel electrophoresis on a 10% gel under reducing conditions and electrotransferred to nitrocellulose membranes (Bio-Rad, Hercules, CA) in a buffer containing 25 mM Tris, 192 mM glycine, and 20% methanol. After transfer, membranes were incubated at room temperature for 1 h in Super Block (Thermo Scientific). Primary and secondary antibodies were diluted in 1% nonfat milk, 0.05% Tween 20 in PBS at the concentrations recommended by the manufacturer. Membranes were incubated with primary antibody at 4°C overnight. Following multiple washes with 0.05% Tween 20, membranes were incubated with HRP-conjugated secondary antibody for 1 h at room temperature. Detection of HRP-conjugated Abs was performed using West Pico Supersignal (Thermo Scientific). Chemiluminescence for all membranes was detected using BioMax film (Kodak). Band intensities were determined using the free ImageJ 1.41 software (http://rsb.info.nih.gov/ij) from the National Institutes of Health (Bethesda, MD). Phosphorylated ZAP-70 band intensities were normalized for loading differences based on corresponding ZAP-70 bands.
All statistical analyses were performed by using a one-tailed Student t test. A P value of <0.05 was considered statistically significant.
To determine whether M. tuberculosis bacilli could directly regulate CD4+ T-cell activation independently of effects on APCs, highly purified CD4+ T-cell populations were required. To this end, CD4+ T cells were first isolated from murine spleens by negative selection with MAb-coated magnetic beads. This was followed by CD4 staining and sorting by flow cytometry, resulting in CD4+ T-cell purities of 98% or greater (Fig. (Fig.1A).1A). Purified CD4+ T cells were stimulated with plate-bound CD3 and soluble CD28 MAbs for 48 h in the presence or absence of M. tuberculosis bacilli. T-cell proliferation (Fig. 1B and C) and IL-2 production (Fig. 1D and E) were measured. M. tuberculosis inhibited CD4+ T and IL-2 production by 25 to 100% at M. tuberculosis bacillus/CD4+ T-cell ratios ranging from 0.2 to 2. Under these conditions, CD4+ T-cell viability at 48 h was not affected by M. tuberculosis, as measured by counting of viable cells through trypan blue exclusion and MTT assay (data not shown). These results indicate that M. tuberculosis can directly inhibit CD4+ T-cell activation.
Studies have shown that M. tuberculosis cell wall glycolipids have immunomodulatory effects on multiple components of host immunity, and we postulated that they may play a role in our observed inhibition (5, 45, 49). We first verified that intact or viable M. tuberculosis bacilli were not required for inhibition by testing lysates of H37Ra. IL-2 production was significantly inhibited (P < 0.05) at a concentration range between 0.1 and 30 μg/ml, with toxicity not observed at concentrations less than 100 μg/ml (data not shown).
Glycolipids were purified from H37Ra lysate by Triton X-114 and phenol extraction, followed by dialysis of the organic phase. As shown in Fig. Fig.2A,2A, M. tuberculosis glycolipids inhibited IL-2 consistently at concentrations of ≥1 μg/ml; however, in some experiments, significant inhibition was seen at 100 ng/ml. Maximal inhibition was observed between 10 and 30 μg/ml. Thin-layer chromatography and Western blot analyses demonstrated that the M. tuberculosis glycolipid preparation contained both PIM and ManLAM (data not shown). To verify our results and determine whether suppression was specific for either molecule, we tested purified PIM and ManLAM from H37Rv (obtained through the TB Vaccine Testing and Research Materials Contract at Colorado State University). Using concentrations similar to those used with the whole-glycolipid fraction, ManLAM significantly inhibited IL-2 production compared to medium alone (Fig. (Fig.2B).2B). Inhibition was also observed with PIM, however, when both molecules were compared on a molar basis, ManLAM was 10-fold more potent as an inhibitor than PIM (data not shown). These results demonstrated that M. tuberculosis glycolipids can directly inhibit CD4+ T-cell activation. In addition, since the highly purified ManLAM came from a virulent strain of M. tuberculosis, H37Rv, the observed inhibition was not unique to the avirulent H37Ra strain of M. tuberculosis.
Resting splenic CD4+ T cells consist of multiple T-cell subsets (e.g., naïve and effector cells) that express different receptors and have unique mechanisms of activation. To test whether M. tuberculosis glycolipid inhibition was subset specific, we stimulated enriched populations of both naïve and effector CD4+ T cells. CD44 and CD62L surface expression was used to positively select naïve (CD44low CD62Lhigh) from resting CD4+ T cells. Naïve CD4+ T cells behaved similarly to total resting CD4+ T cells, with significant inhibition observed at M. tuberculosis glycolipid concentrations of 0.1 μg/ml and over 60% suppression observed with 10 μg/ml (Fig. (Fig.2C2C).
Since sufficient numbers of effector CD4+ T cells (CD44high CD62Llow) were not present within the resting CD4+ T-cell population, we generated effector CD4+ T cells by stimulating D011.10 splenocytes with the OVA323-339 peptide for 24 h and exogenous IL-2 for 1 week. By flow cytometry, activated spleen cells were >92% positive for CD4, with 72 to 80% of the cells being of the CD44high CD62Llow phenotype. As shown in Fig. Fig.2C,2C, effector CD4+ T cells incubated with M. tuberculosis glycolipid (10 μg/ml) were inhibited to the same degree as naïve CD4+ T cells. These results indicate that inhibition by M. tuberculosis glycolipids is not restricted to a CD4+ T-cell subset and imply a common mechanistic pathway.
Since murine CD4+ T cells express TLRs that modulate T-cell responses (24, 52) and M. tuberculosis glycolipids are known TLR-2 agonists (28), we first determined if TLR-2 had a role in the M. tuberculosis glycolipid-mediated inhibition of T-cell responses. CD4+ T cells from TLR-2 knockout (KO) (Fig. (Fig.3A)3A) and MyD88 KO (Fig. (Fig.3B)3B) mice were activated in the presence of increasing concentrations of M. tuberculosis glycolipids (0.1 to 10 μg/ml), and IL-2 responses were compared to those of wild-type (WT) mice. CD4+ T cells from both strains of mice were inhibited to the same degree as those from WT mice by M. tuberculosis glycolipids as measured by IL-2 production (~80% for 10 μg/ml). Thus, M. tuberculosis glycolipid-mediated inhibition of CD4+ T-cell activation is not dependent on TLR-2 or MyD88 signaling.
Studies with M. tuberculosis and other pathogens have shown that surface expression of CD3 and CD28 on T cells can be downregulated by microbial products (47, 54). Thus, in order to exclude the possibility that M. tuberculosis glycolipids affect the expression of CD3 and CD28, we measured their surface levels. Expression levels of CD3 and CD28 on CD4+ T cells treated with M. tuberculosis glycolipids (10 μg/ml) were identical to those of cells in medium alone after 48 h of incubation (Fig. 4A and B). Anti-CD3 and -CD28-activated CD4+ T cells also did not express decreased levels of either CD3 or CD28 in the presence of M. tuberculosis glycolipids (data not shown).
In contrast, expression of CD25 (IL-2 receptor α chain) on activated CD4+ T cells was inhibited by M. tuberculosis glycolipids (Fig. (Fig.4C).4C). To determine if M. tuberculosis glycolipids interfered with IL-2 binding or function, we tested the response of CTLL-2 cells to IL-2 in the presence or absence of M. tuberculosis glycolipids. Glycolipids at 10 μg/ml did not affect the CTLL-2 response to IL-2 at concentrations of 1 to 100 ng/ml (data not shown).
The activation inducer molecule CD69 is rapidly expressed upon CD4+ T-cell activation, with peak expression detected after 16 h (50). As shown in Fig. Fig.4D,4D, in the presence of M. tuberculosis glycolipids, CD69 expression was reduced compared to that on activated CD4+ T cells cultured without M. tuberculosis glycolipids. The average decrease in CD69 expression was 68% ± 13% in the presence of glycolipids (10 μg/ml). Since CD69 expression occurs soon after TCR stimulation and is not dependent on the production of IL-2 (58), this suggested that M. tuberculosis glycolipids affect CD4+ T-cell activation through the TCR signaling pathway.
To determine where in the TCR signaling pathway M. tuberculosis glycolipids inhibited CD4+ T-cell activation, we first examined whether the inhibition could be bypassed by activating secondary signaling pathways with PMA and ionomycin. CD4+ T cells were stimulated with PMA (0.005 to 0.05 ng/ml) and ionomycin (1 μg/ml) in the presence of increasing concentrations of M. tuberculosis glycolipids or medium alone for 6 h. Even at high concentrations (>10 μg/ml), M. tuberculosis glycolipids did not inhibit CD4+ T-cell activation (Fig. (Fig.5A).5A). To ensure that the lack of inhibition was not caused by decreasing the incubation period from 16 to 6 h, CD4+ T cells were stimulated with CD3 and CD28 MAbs or with PMA and ionomycin overnight. PMA and ionomycin concentrations were reduced to 0.05 ng/ml and 0.01 μg/ml, respectively, to allow for this longer incubation time without affecting T-cell viability. M. tuberculosis glycolipids did not affect PMA-ionomycin-induced IL-2 production but did inhibit IL-2 production by anti-CD3- and CD28-stimulated CD4+ T cells (Fig. (Fig.5B).5B). This suggested that M. tuberculosis glycolipids inhibit CD4+ T-cell activation by blocking proximal TCR signal transduction.
To determine if M. tuberculosis glycolipids inhibited proximal TCR signaling, we focused on changes in tyrosine phosphorylation. Proximal TCR signaling is driven primarily by tyrosine phosphorylation of a series of proteins, including ZAP-70 and LAT (linker of activated T cells) (31), that can be suppressed by other pathogens (25, 51). Lysates from CD4+ T cells either in medium alone or in the presence of M. tuberculosis glycolipids (10 μg/ml) had no observable changes in their overall tyrosine phosphorylation levels (data not shown). Thus, we postulated that M. tuberculosis glycolipids may need a period of preincubation prior to activation to induce inhibition. Incubation of M. tuberculosis CD4+ T cell glycolipids for up to 1 h prior to stimulation led to reduced tyrosine phosphorylation levels compared to that in cells preincubated in medium alone (data not shown).
One of the bands that was markedly decreased by M. tuberculosis glycolipids was in the 70-kDa range, which we postulated was ZAP-70. To test this hypothesis, we analyzed the phosphorylation levels of ZAP-70 using an antibody specific to the Tyr-319 residue required for LAT phosphorylation. As shown in Fig. Fig.6A,6A, ZAP-70 phosphorylation was reduced when CD4+ T cells were incubated with M. tuberculosis glycolipids prior to activation. In addition to our M. tuberculosis glycolipid preparation, purified ManLAM reduced ZAP-70 phosphorylation when administered to CD4+ T cells prior to activation (Fig. (Fig.6A).6A). Densitometry analysis of multiple experiments for both M. tuberculosis glycolipids and ManLAM-treated cells showed no significant difference in their rates of suppression of phosphorylation, with both showing significant decreases in relative intensity of around 60% (Fig. (Fig.6B).6B). These results suggest that M. tuberculosis glycolipids, including ManLAM, inhibit CD4+ T-cell activation by blocking the phosphorylation of ZAP-70.
This report presents the novel observation that M. tuberculosis bacilli, and M. tuberculosis glycolipids, including PIM and ManLAM, can directly inhibit CD4+ T-cell activation as measured by expression of T-cell activation markers, IL-2 production, and proliferation. M. tuberculosis glycolipids did not affect surface expression of either CD3 or CD28 and did not block IL-2 signaling by T cells. Inhibition of CD4+ T-cell activation by M. tuberculosis glycolipids was seen with multiple subsets and was not mediated through TLR-2 or MyD88. M. tuberculosis glycolipids did not affect the downstream T-cell signaling pathways activated by PMA and ionomycin; however, they did suppress ZAP-70 phosphorylation, critical to proximal TCR signal transduction. All studies using our M. tuberculosis glycolipid fraction were reproducible using a highly purified sample of the major cell wall glycolipid ManLAM. To our knowledge this is the first demonstration that major M. tuberculosis glycolipids such as ManLAM can directly inhibit CD4+ T-cell activation by interfering with proximal TCR signaling.
ManLAM, found in the cell wall of M. tuberculosis, readily interacts with host immune cells. ManLAM binds to receptors such as the mannose receptor, dendritic-cell-specific intercellular adhesion molecule 3-grabbing nonintegrin, and TLR-2 in and on macrophages and dendritic cells, resulting in the production of cytokines and chemokines that can affect CD4+ T-cell function (10, 12). Macrophages exposed to ManLAM can produce IL-10 and transforming growth factor β, which can inhibit T-cell activation, along with prostaglandins, which contribute to the expansion of regulatory T cells (20). However, these inhibitory activities of ManLAM are mediated by molecules secreted by APCs and do not require a direct effect of ManLAM on T cells.
Less is known about the direct interactions of M. tuberculosis glycolipids such as ManLAM and PIM with T cells. A study by Shabaana et al. suggested that mycobacterial LAMs can modulate cytokine production by human CD4+ Th1- and Th2-cell clones (49). Non-mannose-capped LAMs inhibited Th1-cytokine production, including IL-2, while ManLAM had little effect. However, in our experimental system, ManLAM had an inhibitory effect on IL-2 production. There are two significant differences that may explain these divergent results. First, our study used primary resting CD4+ T cells, which differ from CD4+ T-cell clones in activation status and distribution in lipid rafts of molecules necessary for T-cell activation, such as CD3 (36). Second, those authors used PMA to activate the CD4+ T-cell clones, whereas we have shown that the inhibition can be bypassed when CD4+ T cells are stimulated with PMA. Thus, it is possible that Shabaana et al. did not observe the inhibition due to their T-cell stimulation strategy. This novel mechanism for modulating CD4+ T-cell function is also distinct from results of our earlier studies that determined that PIM, but not ManLAM, can directly interact with VLA-5 on CD4+ T cells, promoting activation and binding to fibronectin. Overall, these studies suggest that M. tuberculosis glycolipids can directly affect CD4+ T-cell function above and beyond the effects that glycolipids such as ManLAM and PIM have on the APCs required for the activation of M. tuberculosis-specific CD4+ T cells in vivo.
How might M. tuberculosis glycolipids directly access CD4+ T cells? The intracellular localization of M. tuberculosis might preclude direct interaction between mycobacterial glycolipids and CD4+ T cells. However, in infected cells, mycobacterial-glycolipid-containing vesicles can traffic outside phagosomes as exosomes into the extracellular environment and are then available for fusion with plasma membranes of nearby cells, including CD4+ T cells (3, 4). These exosomes are capable of activating T cells but only in the presence of an APC, leaving the possibility that they have additional inhibitory effects directly on T cells (6, 26, 55). Since local concentrations of glycolipids will be highest around infected cells, direct inhibition of CD4+ T-cell activation most likely will occur in the immediate vicinity of these M. tuberculosis-infected APCs.
Through their glycosylphosphatidylinositol (GPI) anchor, PIM and ManLAM can integrate into regions of cell membranes rich in GPI (30). These regions include lipid rafts, highly organized structures containing sphingolipids and cholesterol that allow for quick rearrangement of the cell membrane, such as occurs during immune synapse formation (2, 49). Insertion of ManLAM into macrophage lipid rafts is necessary for inhibition of phagosomal maturation (57). In addition to the insertion of ManLAM, we have observed suppression of T-cell activation with both PIM and non-mannose-capped LAM (data not shown), indicating the shared GPI anchor as the likely inducer of inhibition. We postulate that ManLAM exerts its suppressive effect on proximal TCR signaling by inserting into the plasma membrane or lipid raft of CD4+ T cells. The mechanism linking lipid raft insertions with a TCR signaling defect is not known.
While direct inhibition of CD4+ T-cell activation is a novel observation for M. tuberculosis, it has been shown to occur with other pathogens. These include bacteria, viruses, and protozoan parasites. Bacterial toxins, including those from Helicobacter pylori and Bacillus anthracis, can suppress TCR signaling at various points. H. pylori produces the exotoxin VacA, which blocks calcium influx and induces the activation of the stress kinase p38 (7, 21). B. anthracis lethal and edema toxins suppress downstream T-cell signaling through the mitogen-activated protein kinase and calcium signaling pathways, respectively (14, 17, 42). Salmonella enterica may also express a protein(s) that can down-modulate TCR expression (56). Yersinia pestis produces YopH, a tyrosine phosphatase that can dephosphorylate multiple components of the proximal TCR signaling pathway, including Lck (1), LAT, and SLP-76 (25). Whole Neisseria gonorrhoeae and outer membrane vesicles interact with CEACAM1, a coinhibitory receptor that induces SHP-1 and -2 phosphatase activity (8, 33, 34). The herpes family of viruses also can inhibit proximal TCR signaling. Herpes simplex virus reduces LAT phosphorylation (51), while herpesvirus saimiri inhibits ZAP-70 phosphorylation by sequestering Lck (13).
Whereas mechanisms used by pathogens to directly inhibit T-cell function appear to be both diverse and not uncommon, few appear to use glycolipids as mediators. A surface glycolipid, glycoinositolphospholipid on Trypanosoma cruzi, can downregulate T-cell activation (27). While not directly inhibitory of T-cell function, Leishmania spp. produce a glycolipid, lipophosphoglycan, that blocks phagosome maturation (35) and respiratory burst in macrophages by blocking signal transduction (16). As these examples indicate, pathogens do not utilize one universal inhibitory pathway but instead can block T-cell activation at various points in the TCR signaling pathway and can activate more than one suppressive pathway at a time. Collectively, these studies indicate that direct suppression of T cells may play a fundamental role in a pathogen's resistance to adaptive host immunity. The different mechanisms used likely reflect the unique biological niche and distinct evolution of each pathogen.
For a chronic pathogen, such as M. tuberculosis, survival in macrophages and evasion of adaptive T-cell responses are essential for long-term survival in the human host. The ability of M. tuberculosis to directly interfere with CD4+ T-cell activation through the release of its abundant cell wall glycolipids ManLAM and PIM provide it with an additional mechanism to evade host immunity. This novel mechanism is likely one of many used by M. tuberculosis to modulate innate and adaptive immune responses to its advantage as it seeks to survive in macrophages in the face of robust CD4+ T-cell responses.
This work was supported by National Institutes of Health grants AI-27243 and HL-55967 (to W.H.B.); contract no. HHSN266200700022C/NO1-AI-70022 for the Tuberculosis Research Unit (TBRU) (to W.H.B.); NIH grants AI069085, AI034343, and AI035726 to C.V.H.; American Lung Association grants RG48786N to R.E.R. and RG-8407-N to S.A.F.; and grant HHSN26620040091C.
We thank the Cytometry Facility of the Comprehensive Cancer Center at CWRU (P30 CA43703) for their assistance in purifying CD4+ T cells by flow cytometry, Karen Dobos from Colorado State University for thin-layer chromatography analysis of our M. tuberculosis glycolipid sample, and Alan Levine, Mursalin Anis, and Scott Reba for technical advice.
Editor: J. L. Flynn
Published ahead of print on 3 August 2009.