γδ T cells are early recruited into mycobacterial lesions. Upon microbial Ag recognition, γδ cells secrete cytokines and chemokines and undergo apoptosis via CD95/CD95 ligand (CD95L) interaction, possibly influencing the outcome of infection and the characteristics of the disease. In this paper we show that activated phagocytes acquire, upon challenge with Mycobacterium tuberculosis, the ability to inhibit M. tuberculosis-induced γδ cell apoptosis. Apoptosis protection was due to NO because it correlated with NO synthase (NOS)-2 induction and activity in scavenger cells and was abrogated by NOS inhibitors. Furthermore, the NO donor S-nitrosoacetylpenicillamine mimicked the effect of enzyme induction. NO left unaffected the expression of CD95 and CD95L, suggesting interference with an event ensuing CD95/CD95L interaction. NO was found to interfere with the intracellular accumulation of ceramide and the activation of caspases, which were involved in γδ T cells apoptosis after M. tuberculosis recognition. We propose that NO generated by infected macrophages determines the life span and therefore the function of lymphocytes at the infection site, thus linking innate and adaptive immunity.

Infectious agents activate in the host a highly organized response that controls pathogen dissemination and minimizes tissue damage. Early phases of the response require the immunoregulatory influence of γδ T cells (1, 2) that contribute to the overall ability of infected hosts to eliminate infecting intracellular microbes, i.e., Listeria, Leishmania, or Mycobacterium (3, 4, 5, 6). This effect involves the limitation of lesion size, possibly via control of polymorphonuclear leukocytes homing and functions at the infection site (4, 7). Expansion of the γδ cell compartment differs in healthy subjects or in infected patients: peripheral blood γδ cells expand in healthy hospital workers after contact with tuberculosis patients, whereas a loss of Vγ9/Vδ2+ T cells in the bronchoalveolar lavages correlates with the severity of active pulmonary tuberculosis (8, 9).

The cross talk between T cells and macrophages also contributes to the outcome of the disease (10), possibly influencing individual resistance to infections. The diffusible messenger NO might be a candidate for such a cross talk. The resistance to intracellular pathogens correlates with macrophage expression of the NO synthase isoform 2 (NOS-2),4 which releases NO in a continuous way. Treatments with inhibitors of NOS activity often result in striking exacerbation of experimental infection (11). Mice bearing a genetically disrupted NOS-2 gene are highly susceptible to Mycobacterium tuberculosis infection, with rapid bacterial outgrowth, diffuse granulomatous lung involvement, and death (11). This pattern resembles that of wild-type mice heavily immunosuppressed with corticosteroids, and suggests that NOS-2 gene represents a protective locus against tuberculosis (12).

M. tuberculosis which have escaped intracellular killing multiply, and infected cells actively release vesicles containing lipoarabinomannan (LAM) and other microbial products (13). The arabinose termini of a virulent strain of M. tuberculosis and of avirulent Mycobacterium bovis bacillus Calmette-Guérin (BCG) are capped with mannose residues (14, 15). LAM moieties exert a wide spectrum of immunomodulatory effects (13). In particular, secreted ManLAM, i.e. a LAM with a few additional mannose residues, although leaving unaffected NO generation, rescues macrophages from apoptosis (16). The ability to survive to NO generated by macrophages may be crucial for successful intracellular infections (17, 18, 19, 20).

An enzymatically competent NOS-2 enzyme is expressed in the vast majority of alveolar macrophages of patients with newly diagnosed, untreated pulmonary tuberculosis (21). Although the antimicrobial action of NO possibly relates to its ability to interact with other radicals, and in particular with superoxide, to generate peroxynitrite (18, 22, 23, 24), NO generation by macrophages infected by M. tuberculosis does not result in M. tuberculosis killing per se, but rather exerts a bacteriostatic effect (25). This suggests that a complex series of events underlies the NO antimycobacterial activity in vivo, possibly mediated via other bystander cells, including T cells recognizing infected macrophages.

The recognition of M. tuberculosis-infected macrophages by γδ T lymphocytes expressing the CD95 receptor (Fas, APO-1) results in the synthesis of CD95 physiological ligand (CD95L) (26). Because the outcome of the CD95-CD95L interaction is the apoptotic suicide of sensitive cell (27), M. tuberculosis is likely to exploit this event to cause the suicide of leukocytes involved in the antimycobacterial immune response. NO has been described among the physiological regulators of the CD95 signal transduction (28, 29, 30).

We therefore investigated whether NO generated by M. tuberculosis-stimulated phagocytes could rescue T lymphocytes from M. tuberculosis-induced, CD95-mediated cell death, possibly influencing their function at the early infection site and the molecular events underlying this protection.

Human γδ T cell clones were established by limiting dilution and propagated by cyclic restimulation as described (31). Unless indicated, chronically activated γδ T cells were used, i.e., cells propagated in vitro for at least 21 days after restimulation (26, 28, 32). Expression of the CD95 receptor was routinely assessed by staining with a murine anti-human CD95 IgG mAb (clone SM1/1, Bender MedSystems, Vienna, Austria). A FITC-labeled goat anti-mouse antiserum (Southern Biotechnology Associates, Birmingham, AL) was used as a second step reagent. Microglial murine N9 cells (33) were kindly provided by Paola Ricciardi-Castagnoli (Milan, Italy) and cultured in IMDM containing 10% FCS and supplemented with 100 U/ml penicillin, 0.1 mg/ml streptomycin, 2 mM l-glutamine. The human monocyte-like cell line U937 was purchased from American Type Culture Collection (Manassas, VA).

M. tuberculosis H37Ra (ATCC 25177) was cultured in liquid Middlebrook 7H9 medium (Difco, Detroit, MI) supplemented with 0.5% glycerol, 10% oleic acid-albumin-dextrose complex, as described (34). Before use bacteria were washed and clumps were mechanically disrupted by serial treatments with glass and magnetic beads (2 μm in diameter; Dynabeads, Dynal, Oslo, Norway) before sonication for 20 s. Bacterial counts were conducted by immunofluorescence microscopy after staining with rhodamine and by CFU assessment of serial dilution of mycobacterial suspension on Middlebrook 7H11 agar. Killed bacteria were prepared by heating thawed aliquots at 80°C for 20 min (34).

Persistent NO generation, which is likely to mimic the physiological situation, is achieved in vitro by activated phagocytes endowed with NOS activity. We used well-characterized scavenger murine microglial clones derived from embryonic mouse brain (N9 cells), whose NOS-2 induction requires, besides IFN-γ, a second signal (33). An extra bonus of the N9 cell system is provided by the limited cross-reactivity between soluble agents released by murine and human cells, with the notable exception of NO. For example, species-specific factors are necessary for signaling through the IFN-γ receptor that we used to trigger NO generation by macrophages (35). To induce NOS-2 expression, N9 cells (200,000/well) were incubated for 24 h in the presence of a 5-fold excess of heat-killed M. tuberculosis cells and of recombinant murine IFN-γ (10 U/ml) with or without the NOS inhibitors l-Nω-arginine methyl ester (l-NAME; 1 mM) or aminoguanidine (500 μM). In selected experiments, l-arginine (1.5 mM) was added before the l-NAME. Phagocytosis of FITC-labeled M. tuberculosis by N9 cells was confirmed in parallel experiments by confocal microscopy. NOS-2 induction was assessed by Western blotting as described (25) using a specific polyclonal Ab reagent (Transduction Laboratories, Lexington, KY). U937 cells were stimulated as above with dead mycobacteria and recombinant human IFN-γ (100 U/ml) to induce NOS-2 expression for 96 h before coculture with γδ T cells. When indicated, the NOS inhibitor aminoguanidine (500 μM) was included at the start of U937 stimulation or experiments were performed in medium devoid of l-arginine (30). NO production was measured by determining the nitrite accumulation from the culture medium of cells using the Griess reaction (36). Standard curves with increasing concentrations of sodium nitrite were run in parallel.

γδ T cells were incubated for 6–8 h in a double chamber system (200,000/well) either in the presence or the absence of synthetic mycobacterial Ags like the phosphorylated compound isopentenyl pyrophosphate (IPP) (37) (100 μg/ml; Sigma, St. Louis, MO), originally identified in the culture medium of Mycobacterium smegmatis. In the lower adjacent chamber, separated by a semipermeable membrane (cut-off 0.4 μm; Costar, Cambridge, MA), NOS-2+ or NOS-2 N9 cells were cultured for 24 h. When indicated, NOS-2 was induced in the presence of the NOS inhibitor l-NAME (see above). In selected experiments, before addition of IPP, γδ T cells were preincubated for 15 min with the cell permeant cGMP analogue 8-Br cGMP (1 mM), with the NO donor S-nitrosoacetylpenicillamine (SNAP; 100 and 300 μM) or with the caspase inhibitors acetyl-Tyr-Val-Ala-Asp chloromethylketone (ac-YVAD-CMK) and acetyl-Asp-Glu-Val-Asp aldehyde (ac-DVED-CHO) (range tested, 1–500 μM; Calbiochem-Novabiochem, La Jolla, CA). γδ cells that express NOS-3 upon activation completely down-regulated the enzyme expression when 21 days or more elapsed from in vitro restimulation (28). We relied on this property to assess the contribution of endogenous NO in determining the susceptibility of γδ T cells to M. tuberculosis Ag-induced apoptosis. Both NOS-3+, recently activated (less then 10 days), γδ T cells and NOS-3 γδ T cells were challenged with IPP (100 μg/ml) in the presence or the absence of the NOS inhibitor l-NIO (300 μM).

To verify the ability of γδ cells to recognize other Ags, γδ cells were challenged with recombinant heat shock proteins (HSP) for 48 and 72 h. A total of 65 kDa from M. bovis BCG HSP and 70-kDa HSP from M. tuberculosis were kindly provided by Dr. R. van der Zee (Bilthoven, The Netherlands), whereas mammalian 70 kDa HSP, purified from bovine brain, was purchased from StressGen (Victoria, Canada). Proliferation was then assessed by [3H]thymidine incorporation and apoptosis induction was assessed by flow cytometry (see below).

Cell viability and apoptosis induction were evaluated as described (31). To assess the exposure of phosphatidylserine (PS), γδ cells were incubated at room temperature in PBS containing 0.1 mM MgCl2 and 0.1 mM CaCl2 (PBS2+), FITC-labeled annexin V (0.5 μg/ml) (Bender MedSystems), and propidium iodide (PI, Sigma) (10 μg/ml) and analyzed by flow cytometry (FACStarPlus, Becton Dickinson, Sunnyvale, CA). The membrane pattern of PS exposure was further confirmed by confocal microscopy.

To assess the surface expression of the CD95 Ag, cells treated or not with the NO donor SNAP (300 μM) were stained for 30 min at 4°C with a murine anti-human CD95 IgG mAb (Bender MedSystems). A FITC-labeled goat anti-mouse antiserum (Southern Biotechnology Associates) was used as a second step reagent. CD95L expression was assessed by Western blotting, according to the procedure described (26). Briefly, γδ T cells were incubated for 60 min at 37°C with or without IPP (100 μg/ml) in either the absence or the presence of SNAP (300 μM). Cells were then washed and lysed, and proteins were analyzed by SDS-PAGE and Western blotting using different anti-CD95L mAbs (clone #33, Transduction Laboratories, and clone G247-4, from PharMingen, San Diego, CA). As internal controls, a lysate of human endothelial aortic cells (Transduction Laboratories; 1 mg/ml, 5 μl) and a known amount (20 ng/lane) of recombinant extracellular soluble domain of the human CD95 ligand molecule (residues 103–281; Upstate Biotechnology, Lake Placid, NY) were used.

A total of 2 × 106 cells/sample were incubated in PBS (80 μl) with IPP (100 μg/ml) for 0, 30, 60, or 120 min at 37°C, in the presence or in the absence of SNAP (300 μM) or 8-Br cGMP (1 mM). Incubation was stopped by addition of ice-cold CH3OH/CHCl3 (300 μl; 2/1, vol/vol). Samples were then supplemented with CHCl3 (100 μl) and NaCl (100 μl, 1 M). Phospholipids were extracted, dried under nitrogen, and resuspended in a mixture containing cardiolipin (5 mM), diethylenetriaminepentaacetic acid (1 mM), and octyl-β-glucopyranoside (7.5%) (Sigma). Diacylglycerol kinase assay was performed, and ceramide phosphate was isolated by TLC (Silca Gel 60; Merck, Milan, Italy) using CHCl3/CH3OH/CH3COOH (65/15/5, vol/vol/vol) as solvent (38). Authentic ceramide-1-phosphate was identified by autoradiography at Rf 0.25.

A total of 1 × 106 cells/sample were incubated at 37°C with IPP (100 μg/ml) for 10 or 120 min, alone or in the presence of SNAP (300 μM) or 8-Br cGMP (1 mM). Control untreated samples were processed in parallel as well as samples treated with anti-TCR Abs (TiγA, a kind gift of Dr. Hercend, Institut National de la Santé et de la Recherche Médicale, Unité 267, Villejuif, France) in the presence or in the absence of SNAP (300 μM). Cells were rinsed in cold saline and lysed in a 25 mM HEPES buffer (pH 7.5) containing 5 mM EDTA, 1 mM EGTA, 5 mM MgCl2, 5 mM DTT, 1% 3-[(3-cholamidopropyl)dimethylammonio]-1-propanesulfonate (CHAPS), 10 μg/ml pepstatin, 10 μg/ml leupeptin, and 1 mM PMSF (all regents were purchased from Sigma). Lysates were cleared for 3 min (5000 rpm in an Eppendorf microfuge) and stored at −80°C. Protein content was assayed by the bicinchonic acid procedure (Pierce, Rockford, IL). Lysates were incubated at 37°C in a 25 mM HEPES (pH 7.5) buffer containing 10% sucrose, 0.1% CHAPS, and 1 mM DTT, supplemented with the fluorogenic substrate ac-DEVD-7-amino-4-methyl-coumarin (amc) (50 μM), that mimics the cleavage site of the well-characterized caspase-3 substrate poly(ADP) ribose polymerase. The fluorescence increase following the cleavage of the amc moiety was monitored for 10 min and quantified in a LS50 Perkin-Elmer fluorometer (excitation, 380 nm; emission, 460 nm). Standard curves using increasing amc concentrations were run in parallel.

Results are expressed as means ± SEM. Statistical analysis was performed by Student’s t test for unpaired data (two-tail). A value of p < 0.05 was considered to be statistically significant.

Chronically activated γδ cells extensively died after a 6-h challenge with the phosphorylated compound IPP, as evaluated by flow cytometry by measuring the percentage of γδ T cells that, upon exposure of the anionic phospholipid PS, are selectively labeled by FITC-conjugated annexin V. (39 ± 9%; see also Ref. 21).

To assess the effect of NO on γδ cell apoptosis we used a double chamber system: a semipermeable membrane separated the upper chamber, containing γδ T lymphocytes, from the lower chamber, in which N9 cells were seeded.

γδ T cells underwent extensive apoptosis when challenged with IPP in the presence of resting N9 cells (Fig. 1,A, left panel). On the contrary, γδ cells did not undergo apoptosis when challenged with IPP when challenged with N9 activated for 24 h with IFN-γ and heat-killed mycobacteria (Fig. 1,A, middle panel), a treatment that induced the generation of high levels of NO (Fig. 1,B, middle panel). Treatment of activated N9 cells with the NOS inhibitor l-NAME, which abolished NO production as assessed by the nitrite assay (Fig. 1,B, right panel), also abolished protection from apoptosis (Fig. 1 A, right panel), demonstrating the crucial role of NO generation. NO specificity was demonstrated by the substantial reversal of the effect of l-NAME in the presence of l-arginine (62 ± 9%, n = 3). Apoptotic cells were identified as PS+ cells. Identical results were obtained when apoptotic cells were identified based on their hypodiploid “sub-G1” DNA content, or on the typical apoptotic morphology (membrane blebbing, chromatin and cytoplasm condensation).

FIGURE 1.

NO generated upon M. tuberculosis phagocytosis rescues γδ T lymphocytes from M. tuberculosis-induced apoptosis. Human γδ cells were incubated for 6 h in the upper chamber of a double chamber system (A and C). The percentage of apoptotic γδ cells either in the absence (lightly hatched bars) or in the presence (densely hatched bars) of the tubercular Ag IPP was evaluated by flow cytometry after staining with FITC annexin V, as described in Materials and Methods. In the lower chamber, separated by a semipermeable membrane, murine N9 cells (B) or human U937 cells (D) were cultured either alone (right panels) or after preactivation with recombinant IFN-γ and phagocytosis of M. tuberculosis cells (middle and left panels). In the left panels, phagocytes were pretreated with NOS inhibitor (l-NAME for N9 cells and aminoguanidine for U937 cells) before challenge with M. tuberculosis. Phagocytosis by N9 cells resulted in NO generation, as assessed determining the nitrite concentration in the medium by the Griess reaction (B), that abated as a consequence of l-NAME treatment. On the contrary, U937 cells did not generate detectable NO (D). In both cases, however, protection was reverted blocking NOS induction. Panels show the results of one of five independent experiments.

FIGURE 1.

NO generated upon M. tuberculosis phagocytosis rescues γδ T lymphocytes from M. tuberculosis-induced apoptosis. Human γδ cells were incubated for 6 h in the upper chamber of a double chamber system (A and C). The percentage of apoptotic γδ cells either in the absence (lightly hatched bars) or in the presence (densely hatched bars) of the tubercular Ag IPP was evaluated by flow cytometry after staining with FITC annexin V, as described in Materials and Methods. In the lower chamber, separated by a semipermeable membrane, murine N9 cells (B) or human U937 cells (D) were cultured either alone (right panels) or after preactivation with recombinant IFN-γ and phagocytosis of M. tuberculosis cells (middle and left panels). In the left panels, phagocytes were pretreated with NOS inhibitor (l-NAME for N9 cells and aminoguanidine for U937 cells) before challenge with M. tuberculosis. Phagocytosis by N9 cells resulted in NO generation, as assessed determining the nitrite concentration in the medium by the Griess reaction (B), that abated as a consequence of l-NAME treatment. On the contrary, U937 cells did not generate detectable NO (D). In both cases, however, protection was reverted blocking NOS induction. Panels show the results of one of five independent experiments.

Close modal

Similar double chamber experiment were also performed using IFN-γ-activated U937 human macrophages challenged with heat-killed mycobacteria. The coincubation protected γδ T cells from IPP-induced apoptosis (Fig. 1,C). NO was undetectable in the supernatant of activated, phagocytosing U937 cells (detection levels of the assay, 0.5 μM) (Fig. 1,D), in agreement with the low efficiency of human macrophages at generating the gaseous messenger (21). However, apoptosis protection was dependent on NO because it was inhibited by treatment with the NOS blocker agent, aminoguanidine (Fig. 1 C). A similar result was obtained when NOS activity was blocked culturing the cells in medium devoid of l-arginine (43 ± 5%; n = 3). Thus, low levels of NO also are sufficient to protect γδ T lymphocytes from M. tuberculosis-induced apoptosis.

In further support, the protective effect of the coincubation with activated macrophages was reconstituted by addition of the NO donor SNAP to γδ cells challenged with IPP (Table I). The effect was partially mimicked by the cell-permeable cGMP analogue, 8-Br cGMP (Table I). Addition of SNAP in the presence of the NO scavenger, hemoglobin, abolished the protective effect mediated by the NO donor, ruling out a protective role of the N-acetyl-penicillamine moiety of the SNAP molecule (Table I).

Table I.

NO protects chronically activated γδ T cells from M. tuberculosis Ag-induced apoptosisa

SNAP (100 μM)SNAP (300 μM)SNAP + Hb (100 μM)8-Br cGMP (1 mM)
Untreated 9 ± 3 11 ± 2 14 ± 6 11 ± 4 7 ± 2 
IPP 39 ± 4 14 ± 2b 18 ± 3b 31 ± 3 28 ± 3 
Anti-γδ TCR mAb 54 ± 6 23 ± 2b 20 ± 4b 61 ± 7 ND 
SNAP (100 μM)SNAP (300 μM)SNAP + Hb (100 μM)8-Br cGMP (1 mM)
Untreated 9 ± 3 11 ± 2 14 ± 6 11 ± 4 7 ± 2 
IPP 39 ± 4 14 ± 2b 18 ± 3b 31 ± 3 28 ± 3 
Anti-γδ TCR mAb 54 ± 6 23 ± 2b 20 ± 4b 61 ± 7 ND 
a

γδ cells were challenged with IPP (100 μg/ml) or anti-γδ TCR mAbs in the absence or the presence of the NO donor SNAP (100 or 300 μM) in the presence of SNAP and the NO scavenger hemoglobin (Hb) (100 μM) or the diffusible cGMP analogue 8-Br cGMP (1 mM), as described in Materials and Methods. Results are expressed as percentage of FITC-annexin V+ cells and represent the mean ± SEM of three experiments.

b

, p < 0.005, significantly different from control.

We previously reported that the activation state of γδ T cells interferes with sensitivity to apoptosis mediated by CD95 cross-linking and that this feature depends on endogenous NOS-3 expression (28): resistance to apoptosis was abrogated when γδ T cells were challenged with anti-CD95 mAbs in the presence of NOS-2 inhibitors (28). Recently activated (less than 10 days) γδ T cells were completely resistant to IPP-induced apoptosis (Table II and Ref. 21). Resistance was abolished when endogenous NO generation was abrogated by treatment with the NOS inhibitor l-NIO (Table II). Upon chronic activation (more than 21 days) they acquired the ability to die when challenged with IPP and underwent apoptosis at a similar rate in the presence of l-NIO (Table II). Therefore, chronically activated cells depended on exogenous NO for protection, either generated by activated phagocytes or derived from pharmacological NO donors added in vitro (Fig. 1 and Table I).

Table II.

Endogenous NO is involved in the resistance to M. tuberculosis Ag-induced apoptosis of recently activated but not of late activated γδ T cellsa

Recently Activated γδ Cells (<10 days)Chronically Activated γδ Cells (>21 days)
l-NIO (300 μM)l-NIO (300 μM)
Untreated 5 ± 2 11 ± 2 13 ± 3 15 ± 5 
IPP 6 ± 3 54 ± 12b 64 ± 15b 65 ± 12 
Recently Activated γδ Cells (<10 days)Chronically Activated γδ Cells (>21 days)
l-NIO (300 μM)l-NIO (300 μM)
Untreated 5 ± 2 11 ± 2 13 ± 3 15 ± 5 
IPP 6 ± 3 54 ± 12b 64 ± 15b 65 ± 12 
a

Recently activated or chronically activated γδ cells were challenged with IPP (100 μg/ml) in the absence or the presence of the NOS-3 inhibitor l-NIO (300 μM), as described in Materials and Methods. Results are expressed as percentage of FITC-annexin V+ cells and represent the mean ± SEM of three experiments.

b

, p < 0.005, significantly different from control.

γδ T cell apoptosis triggered by M. tuberculosis-infected macrophages or by synthetic mycobacterial Ags, such as IPP, is mediated via CD95L expression and the ensuing activation of the CD95 receptor (26). We therefore evaluated whether NO protected from apoptosis interfering with CD95 expression by γδ cells or with CD95L induction after IPP recognition. The membrane expression of CD95 was unaffected by prolonged exposure to the NO donor SNAP, either alone or in combination with IPP (Fig. 2). CD95L synthesis was substantially enhanced by IPP recognition. The NO donor did not modify the CD95L synthesis by resting or IPP-stimulated γδ cells. Neither CD95 nor CD95L molecules are therefore limiting targets of NO.

FIGURE 2.

NO does not influence basal or IPP-induced expression of CD95 and of its ligand by γδ T cells. Human γδ cell clones expressed the CD95 receptor, as assessed by flow cytometry after staining with a specific mAb (see Materials and Methods). Basal (A) CD95-associated fluorescence (x-axis, filled histograms) was unaffected by the NO donor SNAP alone (B) or in combination with IPP (D), or by IPP alone (C). Open histograms represent the fluorescence background obtained in the presence of the second step reagent alone. CD95L expression, assessed by western blotting (see Materials and Methods) was enhanced by IPP addition (G and H). Basal and IPP-induced CD95L expression were unaffected by the simultaneous treatment with SNAP (F and H). Panels show the results of one of three separate experiments.

FIGURE 2.

NO does not influence basal or IPP-induced expression of CD95 and of its ligand by γδ T cells. Human γδ cell clones expressed the CD95 receptor, as assessed by flow cytometry after staining with a specific mAb (see Materials and Methods). Basal (A) CD95-associated fluorescence (x-axis, filled histograms) was unaffected by the NO donor SNAP alone (B) or in combination with IPP (D), or by IPP alone (C). Open histograms represent the fluorescence background obtained in the presence of the second step reagent alone. CD95L expression, assessed by western blotting (see Materials and Methods) was enhanced by IPP addition (G and H). Basal and IPP-induced CD95L expression were unaffected by the simultaneous treatment with SNAP (F and H). Panels show the results of one of three separate experiments.

Close modal

Along the CD95 apoptotic pathway, crucial biochemical events are the generation of intracellular signaling lipids, ceramides, and the activation of caspases, a family of enzymes involved in the proteolytic cleavage of key substrate (27). We first verified whether these events were indeed recruited after IPP recognition by γδ T cells. Apoptosis was assessed after treatment with ac-YVAD-CMK and ac-DVED-CHO, which are caspase inhibitors with different substrate specificity (27). Both compounds significantly reduced the percentage of γδ T cells undergoing apoptosis upon IPP recognition (from 36 ± 4% to 20 ± 3% and 12 ± 2%, respectively), implicating caspases as key elements in this process. The level of protection was similar to the gold standard, i.e., protection from the apoptosis induced by cross-linking of the CD95 receptor with the CH11 mAb.

Intracellular ceramides increased after recognition of IPP from 60 to a maximum of 110 pmol/106 cells, with a peak at around 60 min after IPP recognition (Fig. 3). Anti-CD95 mAb elicited ceramide accumulation to a similar extent, but with different kinetics: the peak was detectable 10 min after cross-linking of the CD95 receptor and ceramide intracellular concentration dropped to basal levels within 45 min (data not shown; see Ref. 36). Such differences are compatible with IPP-induced de novo synthesis of CD95L (Fig. 2).

FIGURE 3.

Ceramide intracellular accumulation is triggered by the mycobacterial Ag IPP and inhibited by NO. A, Ceramide levels in human γδ T cells were assessed by TLC at the indicated times after challenge with IPP, in the absence or in the presence of the NO donor, SNAP (300 μM). Results shown are mean ± SEM of three experiments run in triplicate. B, Ceramide accumulation visualized by autoradiography after 30- and 60-min incubation with IPP, either in the absence (lanes 1 and 3) or in the presence (lanes 2 and 4) of SNAP. C, Ceramide accumulation (expressed as percent of increase over the basal level, y-axis) was assessed after 30- and 60-min incubation with IPP (x-axis), either in the absence (dotted bars) or in the presence (open bars) of the cell-permeant cGMP analogue 8-Br cGMP (1 mM). Results shown are mean of three separate experiments.

FIGURE 3.

Ceramide intracellular accumulation is triggered by the mycobacterial Ag IPP and inhibited by NO. A, Ceramide levels in human γδ T cells were assessed by TLC at the indicated times after challenge with IPP, in the absence or in the presence of the NO donor, SNAP (300 μM). Results shown are mean ± SEM of three experiments run in triplicate. B, Ceramide accumulation visualized by autoradiography after 30- and 60-min incubation with IPP, either in the absence (lanes 1 and 3) or in the presence (lanes 2 and 4) of SNAP. C, Ceramide accumulation (expressed as percent of increase over the basal level, y-axis) was assessed after 30- and 60-min incubation with IPP (x-axis), either in the absence (dotted bars) or in the presence (open bars) of the cell-permeant cGMP analogue 8-Br cGMP (1 mM). Results shown are mean of three separate experiments.

Close modal

We then verified whether exogenous NO interfered with early events along the signaling cascade triggered by IPP. SNAP determined an important reduction of IPP-induced ceramide accumulation at all time points considered (Fig. 3). The same treatment dramatically reduced the enzymatic activity of caspase-3, as assessed by fluorometry (Fig. 4). 8-Br cGMP only marginally reduced ceramide accumulation in γδ T cells after challenge with the tubercular Ag (Fig. 3) and had virtually no effect on caspase-3 enzymatic activity (Fig. 4).

FIGURE 4.

Caspase-3 activity is induced in a time-dependent fashion by the mycobacterial Ag IPP and inhibited by NO. Caspase-3 activity was assessed at 120 min after challenge with IPP by fluorometry, as described in Materials and Methods. Values, expressed as picomoles of amc/min/mg of cell lysate, are reported on the y-axis. Enzyme activity was up-regulated by IPP, abated in the presence of SNAP (300 μM), whereas was substantially unaffected by the cGMP analogue, 8-Br cGMP (1 mM).

FIGURE 4.

Caspase-3 activity is induced in a time-dependent fashion by the mycobacterial Ag IPP and inhibited by NO. Caspase-3 activity was assessed at 120 min after challenge with IPP by fluorometry, as described in Materials and Methods. Values, expressed as picomoles of amc/min/mg of cell lysate, are reported on the y-axis. Enzyme activity was up-regulated by IPP, abated in the presence of SNAP (300 μM), whereas was substantially unaffected by the cGMP analogue, 8-Br cGMP (1 mM).

Close modal

γδ cell death at the site of early tubercular infection may bias the outcome of the immune response, possibly determining the evolution toward a full blown granulomatous disease (1, 7). M. tuberculosis causes the CD95-dependent apoptosis of γδ T lymphocytes (26). Generation of NO by activated phagocytes infected with intracellular microbes, including mycobacteria, is involved in microbe clearance (17, 18, 19, 20). The recent demonstration that NO acts as a survival factor for leukocytes undergoing apoptosis via the CD95-CD95L pathway (28, 29, 30) raises the possibility that, besides the well-characterized direct effect on microbe viability, NO limits microbe survival prolonging T cell life span at the infection site.

To investigate this possibility, we relied in vitro on two well-characterized NO-generating phagocytes, N9 murine microglial cells (33) and U937 human monocytic cells (30, 39).

Phagocytosis of heat-killed mycobacteria by IFN-γ-stimulated murine N9 cells resulted in NOS-2 induction and NO generation (Fig. 1). The gaseous messenger diffused through a semipermeable membrane and protected γδ cells from apoptosis induced by tubercular Ags (Fig. 1). NO requirement was demonstrated by the almost complete abrogation of apoptosis protection after inhibition of the NOS-2 enzymatic activity by means of the l-NAME compound (Fig. 1). l-Arginine inhibited the effect of NOS-inhibitors and protection was reconstituted by the synthetic NO donor SNAP (Table I).

To verify whether even cells less proficient at generating NO, such as human macrophages, were as well able to protect γδ cells from apoptosis, the experiment was repeated using U937 monocytic cells (30, 39). Also in the latter system, apoptosis was completely abrogated. It is interesting to note that the amount of NO generated was below the limit of detection of the assay (0.5 μM) (see also Ref. 39). However, apoptosis protection was substantially reversed when NOS-2 activity was prevented (Fig. 1).

Reversion in the presence of NOS inhibitors, including the more selective NOS-2 agent aminoguanidine, was less efficient when U937 cells were used, which is consistent with the possible role of other cross-reacting secreted soluble factors.

This result suggests that even low concentrations of NO, possibly better representative of those generated by human phagocytes, efficiently regulate activated lymphocyte apoptosis.

γδ T cells must be able to perform their regulatory and effector functions to proliferate when initially challenged with microbial Ags. We previously reported that recently activated γδ cells are indeed resistant to CD95-mediated activation induced cell death and that this resistance correlates with endogenous generation of NO. Accordingly, recently activated cells do not die when challenged with anti-CD95 mAbs (28) or with Mycobacterium Ags (Ref. 21 and Table II). Inhibition of their endogenous NOS activity reconstitutes their ability to die (Table II). Upon down-regulation of endogenous NOS-3 after chronic activation γδ cells become able to die upon challenge with the Ag, and this feature was not influenced by NOS inhibitors. In this second phase, cells are still fully sensitive to the action of NO if present in the extracellular milieu (Table I and Fig. 1) and very low concentration of the gas, like those generated by activated U937 cells, are sufficient to rescue them from apoptosis (Fig. 1).

These results underline the role of NO in sustaining the function of T cells, both as an endogenous messenger (early after their activation) and as an intercellular mediator generated from activated scavenger cells in later phases and possibly during chronic infections.

T lymphocytes undergo apoptosis after TCR engagement only when chronically activated (activation-dependent apoptosis, AICD) (39, 40, 41, 42). This feature well applies to γδ cell apoptosis induced by mycobacterial Ags (26) and may justify the selective disappearance of γδ T cells from long-lasting chronic tubercular lesions and from the bronchoalveolar lavages of chronically infected patients (9, 43). Because AICD mostly, although possibly nonexclusively, depends on the interaction between CD95 and its ligand (39, 40, 41, 42), we verified at which level NO interfered with the CD95 pathway.

No down-regulation of the CD95 receptor at the cell membrane of activated γδ T cells was observed (Fig. 2). Furthermore, CD95L synthesis upon tubercular Ag recognition was not influenced by NO (Fig. 2). We were able to trace, after M. tuberculosis challenge, intracellular accumulation of ceramides, sphingomyelin breakdown products (38), and the activation of the proteolytic enzymes caspases (24) (Figs. 3 and 4). Both events are compatible with the activation of the CD95 pathway, selectively induced by M. tuberculosis in activated γδ cells (26). Both ceramide accumulation and caspase proteolytic activities abated as a consequence of exposure to NO (Figs. 3 and 4). Therefore, NO interferes with crucial events that link tubercular Ag recognition to the delivery of the signals necessary for lymphocytes to undergo apoptosis. The cGMP analogue affected ceramide accumulation less efficiently than the NO donor (Fig. 3) and did not influence caspase activity (Fig. 4), suggesting that diverse pathways are implicated in NO disruption of signaling events induced by tubercular Ags.

HSP that have been involved in γδ T cell activation (44) and contribute to resistance to apoptosis of macrophages infected by Toxoplasma gondii are over-expressed in infected APC (45). Of interest, HSP expression has also been implicated in resistance to NO-mediated cytolysis (46). In our system, purified HSP and the HSP-containing PPD fraction do not trigger γδ cell proliferation or apoptosis (data not shown). However, HSP expressed by M. tuberculosis-infected macrophages may be recognized, and possibly activate, γδ cells. Further studies are required to assess the relative contribution of macrophage HSP to NOS-2 induction with NO generation in infected macrophages and to the modulation of NO-induced signals.

The functional blockade of the NOS-2 enzyme skews the pattern of cytokines and chemokines secreted in vivo (47, 48). Our results indicate that the ability of M. tuberculosis-infected scavenger cells to generate NO influences the survival of bystander, M. tuberculosis-specific lymphocytes, by selective disruption of the CD95 signaling. γδ cells are selectively activated by mycobacterial Ags (49, 50, 51, 52, 53) and are early recruited in mycobacterial lesions in vivo (1, 2, 37, 44).

Although expressing both CD95 and CD95L molecules (27), recently activated γδ T cells endogenously generate high levels of NO and are protected by CD95L-induced suicide (28). Chronically activated γδ T cells progressively down-regulate, at least in vitro, the expression of the constitutive isoform of the NOS enzyme and become sensitive to CD95L-induced apoptosis (28). Chronic recognition of tubercular Ags may have opposite results, depending on the ability of infected macrophages to generate exogenous NO. Macrophage-generated NO would vicariate endogenous NO and protect γδ cells from M. tuberculosis-induced, CD95-CD95L mediated apoptosis.

The life span of mycobacteria-reactive T cells at the site of infection may influence the secretion of cytokines in situ (54, 55, 56) determining the outcome of the response. Our findings may contribute to explain the role of NOS-2 as a protective locus against tuberculosis (12), with NO acting as a molecular link between innate and adaptive immune response.

We thank C. Rugarli for fruitful discussions and continuous support, P. Ricciardi-Castagnoli for kindly providing N9 cells, M. G. Cifone for advice with the ceramide extraction procedure, R. van der Zee for recombinant mycobacterial HSPs, and T. Hercend for the TiγA Ab.

1

This work was supported by grants from the Istituto Superiore di Sanità (progetto Tubercolosi) from the Associazione Italiana per la Ricerca sul Cancro (to A.A.M. and E.C.) and from the Consiglio Nazionale delle Ricerche (to E.C.). P.R. is the recipient of a “Mario e Valeria Rindi” fellowship of the Fondazione Italiana per la Ricerca sul Cancro.

4

Abbreviations used in this paper: NOS, NO synthase; CD95L, CD95 ligand; IPP, isopentenyl pyrophosphate; LAM, lipoarabinomannan; ac-YVAD-CMK, acetyl-Tyr-Val-Ala-Asp chloromethylketone; ac-DVED-CHO, acetyl-Asp-Glu-Val-Asp aldehyde; PI, propidium iodide; NOS-2+, NOS-2 positive cells; NOS-2, NOS-2 negative cells; PS, phosphatidylserine; BCG, bacillus Calmette-Guérin; l-NAME, l-Nω-arginine methyl ester; SNAP, S-nitrosoacetylpenicillamine; l-NIO, l-N-(iminoethyl)ornithine; amc, 7-amino-4-methylcoumarin; HSP, heat shock proteins.

1
Boismenu, R., W. L. Havran.
1997
. An innate view of γδ T cells.
Curr. Opin. Immunol.
9
:
57
2
Mak, T. W., D. A. Ferrick.
1998
. The γδ T-cell bridge: linking innate and acquired immunity.
Nat. Med.
4
:
764
3
Hiromatsu, K., Y. Yoshikai, G. Matsuzaki, S. Ohga, K. Muramori, K. Matsumoto, J. A. Bluestone, K. Nomoto.
1992
. A protective role of γδ T cells in primary infection with Listeria monocytogenes in mice.
J. Exp. Med.
175
:
49
4
Mombaerts, P., J. Arnoldi, F. Russ, S. Tonegawa, S. H. Kaufmann.
1993
. Different roles of αβ and γδ T cells in immunity against an intracellular bacterial pathogen.
Nature
365
:
53
5
Skeen, M. J., H. K. Ziegler.
1993
. Induction of murine peritoneal γδ T cells and their role in resistance to bacterial infection.
J. Exp. Med.
178
:
971
6
Ladel, C. H., C. Blum, A. Breher, K. Reifenberg, S. H. Kaufmann.
1995
. Protective role of γδ T cells and αβ T cells in tuberculosis.
Eur. J. Immunol.
25
:
2877
7
Dsouza, C. D., A. M. Cooper, A. A. Frank, R. J. Mazzaccaro, B. R. Bloom, I. M. Orme.
1997
. An anti-inflammatory role for γδ T lymphocytes in acquired immunity to Mycobacterium tuberculosis.
J. Immunol.
158
:
1217
8
Ueta, C., I. Tsuyuguchi, H. Kawasumi, T. Takashima, H. Toba, S. Kishimoto.
1994
. Increase of γδ T cells in hospital workers who are in close contact with tuberculosis patients.
Infect. Immun.
62
:
5434
9
Li, B. Q., M. D. Rossman, T. Imir, A. F. Onereyuboglu, C. W. Lee, R. Biancaniello, S. R. Carding.
1996
. Disease-specific changes in gamma delta T cell repertoire and function in patients with pulmonary tuberculosis.
J. Immunol.
157
:
4222
10
Boom, W. H..
1996
. The role of T-cell subsets in Mycobacterium tuberculosis infection.
Infect. Agents Dis.
5
:
73
11
Flynn, J. L., C. A. Scanga, K. E. Tanaka, J. Chan.
1998
. Effects of aminoguanidine on latent murine tuberculosis.
J. Immunol.
160
:
1796
12
MacMicking, J., R. J. North, R. La Course, J. S. Mudgett, S. K. Shah, C. F. Nathan.
1997
. Identification of nitric oxide synthase as a protective locus against tuberculosis.
Proc. Natl. Acad. Sci. USA
94
:
5243
13
Fenton, M. J., M. W. Vermeulen.
1996
. Immunopathology of tuberculosis: roles of macrophages and monocytes.
Infect. Immun.
64
:
683
14
Venisse, A., J.-M. Berjeaud, P. Chaurand, M. Gilleron, G. Puzo.
1993
. Structural features of lipoarabinomannan from Mycobacterium bovis BCG.
J. Biol. Chem.
268
:
12401
15
Prinzis, S., D. Chatterjee, P. J. Brennan.
1993
. Structure and antigenicity of lipoarabinomannan from Mycobacterium bovis BCG.
J. Gen. Microbiol.
139
:
2649
16
Rojas, M., L. F. Barrera, G. Puzo, L. F. Garcia.
1997
. Differential induction of apoptosis by virulent Mycobacteriumtuberculosis in resistant and susceptible murine macrophages: role of nitric oxide and mycobacterial products.
J. Immunol.
159
:
1352
17
Liew, F. Y., S. Millott, C. Parkinson, R. M. Palmer, S. Moncada.
1990
. Macrophage killing of Leishmania parasite in vivo is mediated by nitric oxide from l-arginine.
J. Immunol.
144
:
4794
18
Flesch, I. E. A., S. H. E. Kaufmann.
1991
. Mechanisms involved in mycobacterial growth inhibition by γ interferon-activated bone marrow macrophages: role of reactive nitrogen intermediates.
Infect. Immun.
59
:
3213
19
Green, S. J., L. F. Scheller, M. A. Marletta, M. C. Seguin, F. V. Klotz, M. Slayter, B. J. Nelson, C. A. Nacy.
1994
. Nitric oxide: cytokine-regulation of nitric oxide in host resistant to intracellular pathogens.
Immunol. Lett.
43
:
87
20
Cooper, A. M., J. Flynn.
1995
. The protective immune response to Mycobacterium tuberculosis.
Curr. Opin. Immunol.
7
:
512
21
Nicholson, S., M. da Gloria Bonecini-Almeida, J. R. Lapa e Silva, C. Nathan, Q. Xie, R. Mumford, J. R. Weidner, J. Calaycay, J. Geng, N. Boechat, et al
1996
. Inducible nitric oxide synthase in pulmonary alveolar macrophages from patients with tuberculosis.
J. Exp. Med.
183
:
2293
22
Moncada, S., R. M. J. Palmer, E. A. Higgs.
1991
. Nitric oxide: physiology, pathophysiology and pharmacology.
Pharmacol. Rev.
43
:
109
23
Gross, S. S., M. S. Wolin.
1995
. Nitric oxide: pathophysiological mechanisms.
Annu. Rev. Physiol.
57
:
737
24
Chan, J., K. Tanaka, D. Carroll, J. Flynn, B. R. Bloom.
1995
. Effects of nitric oxide inhibitors on murine infection with Mycobacterium tuberculosis.
Infect. Immun.
63
:
736
25
Rhoades, E. R., I. M. Orme.
1997
. Susceptibility of a panel of virulent strains of Mycobacterium tuberculosis to reactive nitrogen intermediates.
Infect. Immun.
65
:
1189
26
Manfredi, A. A., S. Heltai, P. Rovere, C. Sciorati, G. Galati, C. Rugarli, R. Vaiani, E. Clementi, M. Ferrarini.
1998
. Mycobacterium tuberculosis exploits the CD95/CD95 ligand system of γδ T cells to cause apoptosis.
Eur. J. Immunol.
28
:
1798
27
Nagata, S..
1997
. Apoptosis by death factors.
Cell
88
:
355
28
Sciorati, C., P. Rovere, M. Ferrarini, S. Heltai, A. A. Manfredi, E. Clementi.
1997
. Autocrine nitric oxide modulates CD95-induced apoptosis in γδ T lymphocytes.
J. Biol. Chem.
272
:
23211
29
Mannick, J. B., X. Q. Miao, J. S. Stamler.
1997
. Nitric oxide inhibits Fas-induced apoptosis.
J. Biol. Chem.
272
:
24125
30
Hebestreit, H., B. Dibbert, I. Balatti, D. Braum, A. Schapowal, K. Blaser, H.-U. Simon.
1998
. Disruption of Fas receptor signaling by nitric oxide in eosinophils.
J. Exp. Med.
187
:
415
31
Ferrarini, M., S. Heltai, E. Toninelli, M. G. Sabbadini, C. Pellicciari, A. A. Manfredi.
1995
. Daudi lymphoma killing triggers the programmed death of cytotoxic Vγ9/Vδ2 T lymphocytes.
J. Immunol.
154
:
3704
32
Rovere, P., E. Clementi, M. Ferrarini, S. Heltai, C. Sciorati, M. G. Sabbadini, C. Rugarli, A. A. Manfredi.
1996
. CD95 engagement releases calcium from intracellular stores of long term activated, apoptosis prone γδ T cells.
J. Immunol.
156
:
4631
33
Betz Corradin, S., J. Mauel, S. Denis Donini, E. Quattrocchi, P. Ricciardi-Castagnoli.
1993
. Inducible nitric oxide synthase activity of cloned murine microglial cells.
Glia
7
:
255
34
Chambers, M. A., B. G. Marshall, A. Wangoo, A. Bune, H. T. Cook, R. J. Shaw, D. B. Young.
1997
. Differential responses to challenge with live and dead mycobacterium bovis bacillus Calmette-Guérin.
J. Immunol.
158
:
1742
35
Lembo, D., P. Ricciardi-Castagnoli, G. Alber, L. Ozmen, S. Landolfo, H. Bluthmann, Z. Dembic, S. V. Kotenko, J. R. Cook, S. Pestka, G. Garotta.
1996
. Mouse macrophages carrying both subunits of the human interferon-γ (IFN-γ) receptor respond to human IFN-γ but do not acquire full protection against viral cytopathic effect.
J. Biol. Chem.
271
:
32659
36
Green, L. C., D. A. Wagner, J. Glogouski, P. L. Skipper, J. S. Wishnok, S. R. Tarmenbeum.
1982
. Analysis of nitrate, nitrite and [15]nitrate in biological fluids.
Anal. Biochem.
126
:
131
37
Poccia, F., M. L. Gougeon, M. Bonneville, M. Lopez-Botet, A. Moretta, L. Battistini, M. Wallace, V. Colizzi, M. Malkovsky.
1998
. Innate T cell immunity to nonpeptidic antigens.
Immunol. Today
19
:
256
38
Cifone, M. G., R. De Maria, P. Roncaioli, M. R. Rippo., M. Azuma, L. L. Lanier, A. Santoni, R. Testi.
1993
. Apoptotic signaling through CD95 (Fas/Apo1) activates an acidic sphingomyelinase.
J. Exp. Med.
177
:
1547
39
Chinnaswamy, J., J.K. Actor, R.L. Hunter, Jr.
1998
. Induction of nitric oxide in human monocytes and monocyte cell lines by Mycobacterium tuberculosis.
Nitric Oxide
2
:
174
40
Brunner, T., J. R. Mogil, D. LaFace, N. J. Yoo, A. Mahboubi, F. Echeverri, S.J. Martin, W. R. Force, D. Y. Lynch, C. F. Ware, D. R. Green.
1995
. Cell autonomous Fas (CD95)/fas-ligand interaction mediated activation induced apoptosis in T-cell hybridomas.
Nature
373
:
441
41
Dhein, J., H. Walczak, C. Baumler, K. M. Debatin, P. H. Krammer.
1995
. Autocrine T-cell suicide mediated by APO-1-(Fas/CD95).
Nature
373
:
438
42
Ju, S., D. J. Panka, H. Cui, R. Ettinger, M. El-Khatib, D. H. Sherr, B. Z. Stanger, A. Marshak-Rothstein.
1995
. Fas (CD95)/FasL interaction required for programmed cell death after T cell activation.
Nature
373
:
444
43
Tazi, A, I. Fayac, P. Soler, D. Valeyre, J. P. Battesti, A. J. Hance.
1991
. γδ T lymphocytes are not increased in number in the granulomatous lesions of patients with tuberculosis or sarcoidosis.
Am. Rev. Respir. Dis.
144
:
1373
44
Modlin, R. L., C. Pirmez, F. M. Hofman, V. Torigian, K. Uyemura, T. H. Rea, B. R. Bloom, M. B. Brenner.
1989
. Lymphocytes bearing antigen-specific γδ T cell receptors accumulate in human infectious disease lesions.
Nature
339
:
544
45
Hisaeda, H., T. Sakai, H. Ishikawa, Y. Maekawa, K. Yazutomo, R. A. Good, K. Himeno.
1997
. Heat shock protein 65 induced by γδ T cells prevents apoptosis of macrophages and contributes to host defense in mice infected with Toxoplasma gondii.
J. Immunol.
159
:
2375
46
Kim, Y. M., M. E. deVera, S.C. Watkins, T. R. Billiar.
1997
. Nitric oxide protects cultured rat hepatocytes from tumor necrosis factor-α induced apoptosis by inducing heat shock protein 70 expression.
J. Biol. Chem.
272
:
1402
47
Wei, X.-Q., I. G. Charles, A. Smith, J. Ure, G. Feng, F. Huang, D. Xu, W. Muller, S. Moncada, F. Y. Liew.
1995
. Altered immune responses in mice lacking inducible nitric oxide synthase.
Nature
375
:
408
48
Hogaboam, C. M., W. S. Chensue, M. L. Steinhauser, G. B. Huffnagle, N. W. Lukacs, R. M. Strieter, S. L. Kunkel.
1997
. Alteration of the cytokine phenotype in an experimental lung granuloma model by inhibiting nitric oxide.
J. Immunol.
159
:
5585
49
Janis, E. M., S. H. Kaufmann, R. H. Schwarts, D. M. Pardoll.
1989
. Activation of gamma delta T cells in the primary immune response to Mycobacterium tuberculosis.
Science
244
:
713
50
O’Brien, R. L., M. P. Happ, A. Dallas, E. Palmer, R. Kubo, W. K. Born.
1989
. Stimulation of a major subset of lymphocytes expressing T cell receptor γδ by an antigen derived from Mycobacterium tuberculosis.
Cell
57
:
667
51
Kabelitz, D., A. Bender, T. Prospero, S. Wesselborg, O. Janssen, K. Pechhold.
1991
. The primary response of human γδ+ T cells to Mycobacterium tuberculosis is restricted to Vγ9-bearing cells.
J. Exp. Med.
173
:
1331
52
Havlir, D. V., J. J. Ellner, K. A. Chervenak, W. H. Boom.
1991
. Selective expansion of human γδ T cells by monocytes infected with live Mycobacterium tuberculosis.
J. Clin. Invest.
87
:
729
53
Tsukaguchi, K., K. N. Balaji, W. H. Boom.
1995
. CD4+ αβ T cell and γδ T cell responses to Mycobacterium tuberculosis: similarities and differences in Ag recognition, cytotoxic effector function, and cytokine production.
J. Immunol.
154
:
1786
54
Ferrick, D. A., M. D. Schrensel, T. Mulvania, B. Hsieh, W. G. Ferlin, H. Lepper.
1995
. Differential production of interferon-γ and interleukin-4 in response to Th1 and Th2 stimulating pathogens by γδ T cells in vivo.
Nature
373
:
255
55
Hsieh, B., M. D. Schrenzel, T. Mulvania, H. D. Lepper, L. DiMolfetto-Lanson, D. A. Ferrick.
1996
. In vivo cytokine production in murine listeriosis: evidence for immunoregulation by γδ+ T cells.
J. Immunol.
156
:
232
56
Huber, S., A. Mortensen, G. Moulton.
1996
. Modulation of cytokine expression by CD4+ T cells during coxsackievirus B3 infections of BALB/c mice initiated by cells expressing the γδ+ T cell receptor.
J. Virol.
70
:
3039