IL-4 Influences Apoptosis of Mycobacterium-Reactive Lymphocytes in the Presence of TNF-α

Geok Teng Seah and Graham A. W. Rook
J Immunol August 1, 2001, 167 (3) 1230-1237; DOI: https://doi.org/10.4049/jimmunol.167.3.1230

Abstract

T cell apoptosis is associated with defective cell-mediated effector functions in several infectious diseases. In tuberculosis, there is evidence that T cell apoptosis may be cytokine mediated, but the mechanisms are not clearly understood. Type 2 cytokines have recently been associated with disease extent in human tuberculosis, but they have not previously been linked to apoptosis in mycobacterium-reactive T cells. This study presents evidence that PBLs from healthy donors respond to sonicated Mycobacterium tuberculosis Ags with increased IL-4 gene activation, CD30 expression, and apoptosis. The changes were significantly greater than those observed when cells were stimulated with Ags from nonpathogenic Mycobacterium vaccae. A hypothesis linking these observations was tested. CD30 expression and TNF-α-mediated lymphocyte apoptosis were both down-regulated by inhibiting IL-4 in this model. TNFR-associated factor 2 (TRAF2) expression was down-regulated in CD30+ cells, and addition of anti-TNF-α Ab significantly reduced apoptosis in the CD30+ but not the CD30− population. These observations support the hypothesis that increased IL-4 expression in M. tuberculosis-activated lymphocytes promotes CD30 expression, which sensitizes the lymphocytes to TNF-α-mediated apoptosis via TRAF2 depletion. This may be one mechanism by which IL-4 is associated with immunopathological consequences in human tuberculosis.

T cell apoptosis is an important immune regulatory mechanism associated with defective cell-mediated effector functions in infectious diseases. Infection with Trypanosoma cruzi (1), Toxoplasma gondii (2), and Schistosoma mansoni (3) all result in down-regulation of cell-mediated immunity associated with CD4+-T cell apoptosis. Pulmonary tuberculosis is associated with decreased lymphocyte proliferative responses to mycobacterial Ags (4), reduced IL-2 secretion, and IL-2 receptor expression (5). A significant proportion of patients (17–25%) is unresponsive to purified protein derivative (PPD)3 (6, 7). There is increased CD4+ and γδ T cell apoptosis when cells from tuberculosis patients are cultured with Mycobacterium tuberculosis (8). It has also recently been shown that T cell apoptosis occurs in areas of caseous necrosis within tuberculous granulomas (9). Murine susceptibility to M. bovis bacillus Calmette-Guérin (BCG), associated with defects in T cell proliferative responses, correlates with T cell apoptosis in certain susceptible mouse strains (10). Dysregulated lymphocyte apoptosis may thus be a reason for T cell anergy in tuberculosis patients (11). There is evidence that lymphocytes are much more important than soluble mediators such as TNF-α and IFN-γ in mediating growth inhibition of M. tuberculosis, particularly in PPD-positive subjects (12). Thus, inappropriate lymphocyte apoptosis is likely to negatively influence the protective immune response. However, the mechanisms by which inappropriate T cell apoptosis occurs in human tuberculosis are not clearly understood.

Several groups have recently found that type 2 cytokine gene expression is elevated in human pulmonary tuberculosis patients and relates to disease extent (13, 14). Potential reasons for the association of type 2 cytokines with disease in tuberculosis form the subject of the present work. In certain inflammatory situations, TNF-α-mediated pathology occurs only in the presence of IL-4; this has been shown both in murine Trichinella spiralis (15) and mycobacterial (16) infections. We considered the possibility that IL-4 produced in response to M. tuberculosis influences the sensitivity of T lymphocytes to apoptosis by a TNF-α-mediated pathway. Duckett and Thompson observed that ligation of CD30 results in signal-coupled depletion of TNFR-associated factor 2 (TRAF2) and sensitizes human embryonic kidney cells to apoptosis in response to TNFR1 signal transduction (17). CD30 is a costimulatory molecule chiefly expressed on activated T cells, and its expression is IL-4 and/or CD28 dependent (18). CD30 signaling abrogates the TRAF2-mediated induction of NF-κB by TNF-α (17). TRAF2 also plays a role in cytoprotective functions mediated via stress-activated protein kinase cascades (19) and cellular inhibitors of apoptosis (20). Hence, several mediators downstream of TRAF2 that contribute to cytoprotection following TNFR1 ligation would be affected by TRAF2 depletion. Because M. tuberculosis induces CD30 expression on human PBLs and CD30+ cells are present in tuberculosis lesions (21), degradation of TRAF2 consequent to CD30 signaling may be a potential reason for the association of IL-4 with TNF-α-mediated cytotoxicity in M. tuberculosis-stimulated lymphocytes.

In this study, an in vitro model was used to examine responses of lymphocytes from healthy donors to M. tuberculosis Ags. We show evidence that CD30 expression in this model is at least partly IL-4 dependent and that reduced TRAF2 expression following CD30 ligation may account for the TNF-mediated apoptosis in M. tuberculosis-stimulated lymphocytes.

Materials and Methods

Cell cultures

PBMCs were isolated from a panel of eight BCG-immunized healthy donors by density-gradient centrifugation. Cells were resuspended in RPMI 1640 supplemented with 2 mM l-glutamine, 100 U/ml penicillin, 100 μg/ml streptomycin (all from Life Technologies, Grand Island, NY), and 10% (v/v) autologous serum, then seeded into 24-well plates (Nunclon surface; Nalge Nunc International, Roskilde, Denmark) at a density of 5 × 105 cells/ml and incubated at 37°C in a humid 5% CO2 incubator. Only nonadherent cells (NAC) were harvested for further assays.

Crude whole-cell sonicates of M. tuberculosis H37Rv and M. vaccae NCTC 11659, denoted MtbS and MvacS, respectively, were prepared by the method described by Paul et al. (22). Briefly, mycobacteria grown on Sauton’s medium were harvested, washed twice with PBS (pH 6.8) and suspended in 50 ml of PBS. The suspensions were sonicated for 15 min, then centrifuged at 70,000 × g for 30 min. The supernatants were sterilized through a 0.22-μm filter, and the protein concentration was quantified. The sonicates were then stored in single-use aliquots at −70°C without adding other diluents or additives. To determine cellular responses to M. tuberculosis Ags, MtbS was added to the PBMC cultures at 50 μg/ml (final concentration) before incubation. Lymphocytes cultured in the presence of this concentration of MtbS have up-regulated surface expression of CD30 (21). To assess whether responses were M. tuberculosis specific, control wells were set up with the cells from the same donors treated either with 50 μg/ml MvacS or culture medium alone.

Affinity-purified anti-human Abs and recombinant human (rh) proteins (all from R&D Systems, Abingdon, U.K., unless otherwise stated) were added to the cultures in certain experiments. These were 10 μg/ml anti-human IL-4 Ab, 1 μg/ml anti-human TNF-α Ab (Insight Biotechnology, Middlesex, U.K.), 10 ng/ml rhTNF-α, 5 μg/ml rhTNF-soluble receptor 1 (TNF-sR1)/Fc chimera or TNFR2/Fc chimera, or 100 μg/ml rhCTLA-4/Fc chimera. Equivalent concentrations of isotype-matched Abs (from the same manufacturers) and BSA, fraction V, respectively, were used as negative controls for the added Abs or recombinant proteins in control wells.

Flow cytometry

Flow cytometry was performed on the FACSCalibur equipped with CellQuest software (version 3.0.1; both from BD Immunocytometry Systems, San Jose, CA). Lymphocytes were gated by light scatter characteristics (23), and the specificity of the lymphocyte gate was confirmed by ascertaining that >95% of gated cells were CD45brightCD14− (24). The CD3+ T cells in this gate represented 79.4 ± 2.9% of the cells in the lymphocyte gate (over 30 repeated experiments), which serves as an additional quality control for the lymphocyte gating. At least 50,000 gated events were acquired per experiment.

Cell surface markers

Fluorochrome-conjugated Abs and their specificities were as follows: CD3-PerCP, CD8-PerCP, CD25-PE, CD69-PE, γδ TCR-FITC and -PerCP (all from BD Biosciences, San Jose, CA); CD4-FITC and -CyChrome, CD28-CyChrome (both from BD PharMingen, San Diego, CA); TNFR1-PE, TNFR2-PE (both from R&D Systems); and CD30-FITC (DAKO, Glostrup, Denmark). Isotype-matched control Abs and autofluorescence controls (omitting the Ab conjugate) were used for each experiment. The NAC were harvested, pelleted, and washed, and 106 cells were resuspended in 100 μl of PBS. Conjugated Abs were mixed gently with the cells, and the tubes were incubated at 4°C for 30 min. The cells were then washed with 2-ml filtered staining buffer (sterile PBS/0.1% BSA), and the cell pellets were resuspended in 100 μl staining buffer and analyzed within 10 min.

Apoptosis assays

Cells that bind annexin V (BD PharMingen) but exclude 7-aminoactinomycin D (7-AAD; Sigma, St. Louis, MO) are early apoptotic cells (25). Such cells were identified by flow cytometry according to the manufacturer’s instructions. Following annexin V labeling, the cells were kept at 4°C with addition of 7-AAD (8 μl from stock solution of 0.1 mg/ml) to each tube 10 min before analysis. In certain experiments (see Fig. 8⇓A), the flow cytometry-adapted TUNEL assay was performed using the Flow-TACS kit (R&D Systems) according to the manufacturer’s instructions.

Lymphocyte proliferation assay

At appropriate intervals during culture, 1 μCi of [methyl-3H]thymidine (Amersham Pharmacia Biotech, Uppsala, Sweden) was added to certain microwells 12 h before the NAC were harvested with a Skatron AS cell harvester (Northumbria Biologicals, Cramlington, U.K.) onto glass microfibre filter discs (Whatman, Maidstone, U.K.). The discs were dried, immersed in vials containing scintillation fluid (Ecoscint A; National Diagnostics, Atlanta, GA), and the β emissions were measured over 10 min per vial using a Beckman LS5000CE scintillation counter (Beckman Coulter, Fullerton, CA).

Determination of TRAF2 expression

A total of 6 × 107 NAC were harvested from a 7-day culture of MtbS-stimulated PBMCs. Magnetic beads from the CELLection pan mouse IgG kit (Dynal, Oslo, Norway) were used with FITC-conjugated mouse anti-human CD30 mAb, clone Ber-H2 (DAKO) to sort the NAC into CD30+ and CD30− populations, according to the manufacturer’s protocol (Dynal). Dynabeads were detached from the cells by adding the releasing buffer supplied. Both CD30+ and CD30− cell populations were retained, and the extent of enrichment was determined by flow cytometry.

Total proteins were extracted from both CD30+ and CD30− populations (17), and protein quantification was performed using the Bio-Rad DC Protein Assay kit (Bio-Rad, Hercules, CA). SDS-PAGE was performed with equal quantities of protein from each population. Western blot for TRAF2 was performed using rabbit anti-human TRAF2 polyclonal IgG Ab (Santa Cruz Biotechnology, Santa Cruz, CA) at 1:500 dilution in blocking buffer, followed by HRP-conjugated donkey anti-rabbit IgG (Amersham Pharmacia Biotech) at 1/1000 dilution in blocking buffer as the secondary Ab. Blocking buffer consisted of 5% (w/v) nonfat dried milk dissolved in PBS-Tween 20 (0.1% v/v). The protein bands were detected using ECL Western blotting detection reagents, the bands were visualized on radiographic film (Hyperfilm ECL; both from Amersham Pharmacia Biotech), and their size was compared with the known size of TRAF2 (26).

RT-PCR

The protocol and primers used in performing quantitative RT-PCR to determine IL-4 gene expression in NAC have been described in an earlier publication (27). Briefly, synthetic RNA standards were constructed and serially diluted. These were used to generate standard curves relating the fluorescence intensity of amplicons to the initial RNA copy number. The unknown samples were then compared against the standards, in the same experiment. The primers used are specific for IL-4; amplification of IL-4 splice variant, which is expressed in parallel with IL-4 in human tuberculosis (13), is specifically excluded by these primers (27). All the cytokine mRNA copy numbers were normalized to β-actin expression to correct for potential differences in RNA extraction efficiency.

ELISA

Culture supernatants were harvested at different time points, and TNF-α concentrations were determined by sandwich ELISA using the Quantikine kit (R&D Systems) according to the manufacturer’s instructions. The lower detection limit was 4.4 pg/ml.

Statistics

At least three donors were studied in each experiment described. Cells from each donor (for the same type of experiment) were studied on different days. For each different treatment (e.g., MtbS or MvacS, with or without anti-TNF-α), cells from the same donor were cultured in triplicate wells; these were not pooled but analyzed as distinct samples in flow cytometry experiments. Unless otherwise specified in the legends, the means and 2 SD of results from these triplicates (showing data from one representative donor) are presented in the figures. The statistics shown within figures relate to the means of the triplicate data for that single donor. However, to reflect the interdonor (and interexperimental) variability, the mean data for each treatment group in repeated experiments using multiple donors is presented in the text as the overall probability that one treatment is significantly different from the other. The Student’s t test was used to compare mean results of different treatments, and p < 0.05 were considered significant. Where appropriate, the paired t test was used when the same cells were studied with or without addition of certain inhibitory reagents (e.g., anti-IL-4).

Results

Lymphocyte proliferation assays

The proliferative responses to MtbS and MvacS were equivalent at the time points tested over 7 days for each of the three donors tested, hence the two preparations had similar mitogenic capacities (Fig. 1⇓). The same concentrations of MtbS and MvacS (i.e., 50 μg/ml) were consistently used in all the subsequent experiments described.

FIGURE 1.
FIGURE 1.

Lymphocyte proliferation in response to MvacS, MtbS, or culture medium alone. The difference in proliferative responses to the two mycobacterial sonicates was not significant at any time point (p > 0.1 by Student’s t test for independent samples). Each data point represents the mean and 2 SD of results from triplicate wells in one experiment. The results are representative of three separate experiments showing similar results, performed with cells from different donors.

IL-4 mRNA expression

There is evidence that IL-4 mRNA expression in unstimulated PBMCs of tuberculosis patients is greater than in cells from matched healthy tuberculin-reactive controls (13). It has not previously been determined whether pathogenic and nonpathogenic mycobacteria differ in their capacity to induce IL-4 production. The IL-4 mRNA copy number was determined in samples of 1 × 106 NAC from PBMCs that had been stimulated with MtbS or MvacS. We found that 24 h cultures of MtbS-stimulated NAC had significantly higher IL-4 mRNA copy numbers than MvacS-stimulated (p = 0.0005) or unstimulated (p = 0.0004) cells (Fig. 2⇓). Thus MtbS and MvacS differentially induce IL-4 gene expression in healthy mycobacteria-reactive lymphocytes.

FIGURE 2.
FIGURE 2.

IL-4 gene expression in response to different mycobacterial sonicates at 24 h. The IL-4 mRNA copy number in 1 × 106 MtbS-stimulated NAC was initially determined at six time points to find the time point when maximal IL-4 mRNA was produced (24 h, data not shown). Subsequent assays were performed at 24 h poststimulation, using cells cultured with MtbS, MvacS, or no Ags (unstimulated). Means and 2 SD of data from three donors are shown, statistics by Student’s t test for independent samples. Where error bars are not shown, the error values fall within the symbols.

Kinetics of CD30 expression and lymphocyte apoptosis

There are previous reports that M. tuberculosis induces CD30 expression on lymphocytes (21). We confirmed these findings using MtbS-stimulated PBMCs. Over a period of 9 days, CD30 expression and lymphocyte apoptosis were monitored in NAC cultured in the presence of MtbS (Fig. 3⇓A). Day 7 was chosen as the optimal time point for all subsequent apoptosis and CD30 assays, as CD30 expression reached a peak on day 7 and subsequent time points showed >50% cell death.

FIGURE 3.
FIGURE 3.

A, Kinetics of CD30 expression and lymphocyte apoptosis in MtbS-stimulated cultures were studied in NAC at various time points over 9 days. Mean results and 2 SD of triplicate experiments performed using cells from one representative donor of three are shown. Where error bars are not shown, the error values fall within the symbols. B, Cells from six different donors were stimulated with MtbS or MvacS, and mean results (±2 SD) of day 7 lymphocyte apoptosis assays are shown (∗, p = 0.001 by Student’s t test for independent samples, n = 6).

Almost all (>98%) of the CD30+ lymphocytes were T cells (CD3+); this is in agreement with published reports (28, 29). More than 95% of CD30+ cells were CD4+, and only 3% were γδ T cells. Most were activated T cells, 96 and 77% expressing CD25 and CD69, respectively, in triplicate assays on cells from three different donors (not shown).

Apoptosis experiments

Effect of different mycobacterial preparations on lymphocyte apoptosis.

We examined the percentage of apoptotic cells among NAC cultured for 7 days in the presence of different mycobacterial sonicates (Fig. 3⇑B). The proportion of apoptotic lymphocytes was consistently higher among MtbS-stimulated than MvacS-stimulated lymphocytes (p = 0.001 MtbS vs MvacS, by Student’s t test for independent samples, n = 6).

Effect of blocking TNF-α activity.

The effect of neutralizing endogenously produced TNF-α was next examined. Either anti-human TNF-α Ab or isotype control Ab was included in some of the cultures from day 0. Fig. 4⇓A (inset) illustrates the dose-dependent effect of anti-TNF-α in decreasing lymphocyte apoptosis in MtbS-stimulated cultures. In cultures from each of the three donors studied, lymphocyte apoptosis was significantly reduced in MtbS-stimulated cells treated with anti-TNF-α (Fig. 4⇓A); the change in MvacS-stimulated cultures was not statistically significant (p = 0.002 (MtbS) and p = 0.15 (MvacS) by paired Student’s t test, data from three donors).

FIGURE 4.
FIGURE 4.

Effect of neutralizing TNF-α activity on lymphocyte apoptosis. A, The inset graph shows the effect of different concentrations of anti-TNF-α on lymphocyte apoptosis in MtbS-stimulated cells (∗, p = 0.042; ∗∗, p = 0.019, compared with isotype control). The main figure illustrates the effect of 1 μg/ml anti-TNF-α added to cultures on day 0, with assays performed on day 7. B, The effect of 5 μg/ml rhTNFR-Fc chimera added to day 0 cultures on apoptosis in lymphocytes cultured in the presence of different mycobacterial sonicates. All the figures show means and 2 SD of triplicate assays from one representative experiment of three performed independently using cells from different donors. Statistics by Student’s t test for independent samples. Pooled statistics for all donors are given in the main text.

A different approach to inhibiting TNF-α activity was used by adding rhTNFR/Fc chimeric proteins to the cultures (Fig. 4⇑B). TNF-sR1/Fc, TNFR2/Fc, or neither (BSA control) was added to the PBMC cultures treated with MtbS, MvacS, or medium alone. A significantly lower proportion of lymphocytes from the MtbS-stimulated cultures underwent apoptosis in the presence of either TNF-sR1/Fc or TNFR2/Fc (p < 0.01 by paired Student’s t test, data from three donors) when assayed on day 7. This was not noted in the MvacS-stimulated cultures (p > 0.1 by paired Student’s t test, data from three donors). There was no significant difference between these two TNF-α inhibitors in reducing lymphocyte apoptosis.

TNF-α in the culture supernatants.

The concentration of TNF-α in the culture supernatants was studied at 0, 6, 12, 18 and 24 h, then at 2-day intervals until day 7 (data not shown). Cells cultured in medium alone produced undetectable levels of TNF-α at all time points. There was no significant difference in the kinetics or quantities of TNF-α in the supernatants of cells stimulated with MtbS or MvacS; mean levels of TNF-α were 2.30 and 2.32 ng/ml, respectively, at the peak (18 h). It was thus hypothesized that it is not the quantity of TNF-α produced but the susceptibility of the cells that determines the effect of TNF-α-mediated cytotoxicity.

The percentage of apoptotic lymphocytes in MtbS-stimulated but not in MvacS-stimulated cultures could be further increased by the addition of 10 ng/ml rhTNF-α to the NAC 2 h before assays (data not shown; p = 0.00030 (MtbS) and p = 0.35 (MvacS) by paired Student’s t test, data from five donors). It has been shown previously that this concentration of TNF-α is not toxic to human cells (30).

Effect of anti-IL-4.

Because the MtbS- and MvacS-treated NAC differed significantly in their levels of IL-4 mRNA production, but not in TNF-α production, it was hypothesized that the difference in lymphocyte susceptibility to TNF-α-mediated cytotoxicity may be attributable to IL-4. IL-4 activity was neutralized with anti-IL-4 in some cultures, and the effect was compared with an isotype control. Others have shown that anti-IL-4 neutralizing Abs do not affect proliferative responses to M. tuberculosis Ags (12). The decrease in lymphocyte apoptosis upon neutralizing IL-4 was seen only in MtbS-treated cultures, and the effect was only significant in the presence of rhTNF-α (Fig. 5⇓). This effect was consistent in cells of three donors studied (p < 0.01 in each case). Anti-IL-4 did not significantly affect the apoptosis of lymphocytes cultured with MvacS or with medium alone (p > 0.1 for all donors). However, even in the presence of anti-IL-4, the proportion of apoptotic lymphocytes was still higher in the presence of MtbS than MvacS. Hence, in this model, IL-4 contributes to but does not fully account for the differences in TNF-α-mediated lymphocyte apoptosis.

FIGURE 5.
FIGURE 5.

Effect of inhibiting IL-4 activity on TNF-α-mediated lymphocyte apoptosis, and influence of rhTNF-α. Either 10 μg/ml anti-human IL-4 or isotype control Ab was added to the culture medium from day 0, and 10 ng/ml rhTNF-α was added to half of the wells 2 h before harvesting NAC for apoptosis assays on day 7. Statistics by Student’s t test for independent samples. The figure shows mean results and 2 SD of triplicate assays using cells from one representative donor of three. Pooled statistics for all donors are given in the main text.

CD30 experiments

Effect of different mycobacterial preparations on CD30 expression.

The kinetics of CD30 expression in our experimental system have been illustrated (Fig. 3⇑A). The effect of MtbS in inducing CD30 expression was compared with that of MvacS and PHA to investigate whether the increased CD30 expression observed in MtbS-stimulated cells could be the result of nonspecific T cell activation. PHA-induced CD30 expression peaked earlier at 5% on day 5, but MtbS-treated cells expressed significantly higher levels of CD30 at the peak on day 7 (Fig. 6⇓). The mean level of CD30 expression in MtbS-stimulated cells from four donors was 12%, as compared with 5.1% in MvacS-treated cells (data not shown).

FIGURE 6.
FIGURE 6.

CD30 expression under different stimulation conditions, at various time points. The highest levels of CD30 expression during the period of observation were induced by MtbS (p < 0.002 on day 7 in comparison to all other Ags). The figure shows means and 2 SD of triplicate assays from one representative experiment of two performed using cells from different donors.

Effect of IL-4 and CD28 on CD30 expression.

The dependence of CD30 expression on both CD28 and IL-4 has been noted in earlier reports (18). Fig. 7⇓ shows experiments in which either 10 μg/ml anti-human IL-4 Ab or an isotype control Ab was included in the culture medium in some wells, and their effects on CD30 expression was investigated. The optimal concentration of anti-IL-4 was derived by titration (Fig. 7⇓A, inset). CD30 expression was significantly and consistently reduced by anti-IL4 in MtbS-stimulated but not MvacS-stimulated lymphocytes of four donors studied in separate experiments (p = 0.024 and p = 0.16, respectively, by paired Student’s t test, n = 4, Fig. 7⇓B).

FIGURE 7.
FIGURE 7.

Effect of inhibiting IL-4 on CD30 expression on day 7. A, The inset figure shows the dose-dependent effect of anti-IL-4 on CD30 expression on MtbS-stimulated lymphocytes. Either 10 μg/ml anti-IL4 or isotype control Ab was added to cultures from day 0, and CD30 expression was determined on day 7 by flow cytometry. The main figure shows one representative experiment (means and 2 SD of data from cells in triplicate wells) of four performed independently using cells from different donors. B, Each data point represents the mean of triplicate wells given the same treatment. Data from four donors are illustrated (p = 0.024 and p = 0.16 by paired Student’s t test, respectively, for MtbS- and MvacS-treated cells, n = 4).

CTLA-4 binds to CD80 and CD86 with 20- to 100-fold higher affinity than CD28, thus the CTLA-4/Fc chimeric protein acts as a competitive inhibitor of CD28 signaling. The effect of blocking CD28 signaling was also a significant reduction in CD30 expression (p = 0.0004 by paired Student’s t test on data from two independent experiments, data not shown). However, as IL-4 production is likely to be reduced by inhibition of CD28 costimulation (31), the effect of CD28 may not be independent of IL-4, although this was not specifically studied.

Colocalization of CD30 and apoptosis markers.

CD30 expression on apoptotic cells was next determined. CD30+ cells represented 32.9 ± 5% of all apoptotic lymphocytes (n = 4, data not shown). The proportion of CD30+ cells undergoing apoptosis was significantly higher (Fig. 8⇓A) than that in the CD30− population (p = 0.0054 by Student’s t test on data from three independent experiments). Incubating the MtbS-treated cells with anti-TNF-α (Fig. 8⇓A) caused a 10-fold reduction in the percentage of CD30+ lymphocytes undergoing apoptosis (p = 0.0017 in the experiment shown, p = 0.0054 by paired Student’s t test on data from three experiments). The percentage of apoptotic cells in the CD30− population was low and unchanged with inclusion of anti-TNF-α. Thus, TNF-α had a significant role in the apoptosis of CD30+ cells cultured with MtbS.

FIGURE 8.
FIGURE 8.

A, Apoptosis in CD30+ and CD30− lymphocyte populations after stimulation with MtbS. Apoptotic cells were identified by flow cytometry (TUNEL method), and TdT end-labeled cells were considered apoptotic. The figure shows means and 2 SD of triplicate assays from one representative experiment of three performed independently using cells from different donors (∗, p = 0.0026; †, p = 0.0017). B, TRAF2 expression in CD30+ and CD30− cells. NACs from four donors were studied in separate experiments, and one representative experiment is illustrated. Total proteins from CD30+ (lane 1) and CD30− (lane 2) populations were subjected to electrophoresis and Western blotting using TRAF2-specific Abs. The protein bands were detected on radiographic film. Lane 3, Negative control (BSA); lane 4, positive control.

TRAF2 expression on CD30+ and CD30− cells.

Increased susceptibility of certain cell types to TNF-mediated apoptosis following CD30 signaling has been attributed to depletion of TRAF2 upon CD30 signaling (17). To determine the relative expression of TRAF2 in the CD30+ and CD30− lymphocytes, MtbS-stimulated NAC were sorted into the two populations with magnetic beads. In the CD30+ population, 96% enrichment was achieved and only 0.3% of the selected CD30− population was CD30+. Interestingly, TRAF2 was strongly expressed in the CD30− population but was barely detectable in the CD30+ population (Fig. 8⇑B). Hence, CD30+ cells have down-regulated TRAF2. This provides a potential explanation for the TNF-α-mediated apoptosis of CD30+ cells.

In summary, these experiments demonstrated that MtbS-specific lymphocyte activation resulted in increased IL-4 gene expression, thereby increasing the CD30+ population. This population had reduced TRAF2 expression and displayed increased susceptibility to TNF-α-mediated apoptosis when compared with the CD30− cells.

Discussion

It has been hypothesized that a survival strategy of virulent mycobacteria may be to modulate the host immune response to minimize macrophage apoptosis (32, 33), and some suggest that this is achieved by reducing TNF-α bioactivity (34). T cell apoptosis in the same context has been relatively less investigated. There is histological evidence for large numbers of apoptotic CD3+ CD45RO+ T cells in caseous foci within tuberculous human lungs (9). Several groups have now suggested that depressed M. tuberculosis-induced T cell proliferative and IFN-γ responses noted in human and murine tuberculosis may be due to apoptosis of M. tuberculosis-responsive T cells (8, 10). Hirsch et al. (8) report that PBMCs of patients with newly diagnosed tuberculosis show an increase in both spontaneous and M. tuberculosis-induced apoptosis (in CD4 and non-CD4 T cells) when compared with healthy controls. Successful chemotherapy resulted in 50% reduction in apoptosis, coinciding with 3- to 8-fold increases in levels of mycobacteria-stimulated IL-2 and IFN-γ, respectively. In another study based on examination of pleural fluid from patients with pleural tuberculosis, the loss of IFN-γ-producing cells is limited specifically to M. tuberculosis-responsive cells (35). These studies provide evidence that T cell apoptosis is biologically significant in the disease process.

We focus on a more subtle aspect of mycobacteria-induced apoptosis by comparing responses of lymphocytes from healthy donors to sonicated preparations of pathogenic and nonpathogenic mycobacteria. The lymphocytes of our BCG-vaccinated donors clearly do not have a proliferative defect; the response to both MtbS and MvacS stimulation is equally robust. However, MtbS induces a relatively higher level of both IL-4 mRNA expression and lymphocyte susceptibility to TNF-mediated apoptosis. We demonstrate that it is the CD30+ population that is principally affected by TNF-mediated apoptosis in the presence of MtbS and provide evidence supporting the hypothesis that the apoptotic pathway is associated with reduced TRAF2 expression in this population. Because IL-4 is a major determinant of CD30 expression on lymphocytes, and neutralizing IL-4 in the MtbS-lymphocyte coculture reduced cellular susceptibility to TNF-mediated apoptosis, we thus propose that this is a novel pathway linking IL-4 to apoptosis of mycobacteria-reactive lymphocytes.

It is clear from the current literature that other mechanisms may concurrently exist for the apoptotic removal of mycobacteria-activated T cells. CD4+ T cells in IFN-γ-deficient mice fail to undergo apoptosis during M. bovis BCG infection, suggesting a role for IFN-γ in the apoptotic process (36). Both TNF-α and Fas (CD95)-mediated pathways for lymphocyte apoptosis are likely to be significant in mycobacterial infections. Kremer et al. (10) report that BCG infection of susceptible C57BL/6 mice, but not the resistant C3H/HeJ mice, induces massive production of TNF-α in the serum and an increase in Fas and Fas ligand expression in T cells. Neutralizing anti-TNF-α Abs cause a significant reduction in CD3-induced T cell apoptosis of both CD4+ and CD8+ T cells from the C57BL/6 mice (10). We have not investigated the relevance of the Fas pathway in our experimental system, but it is known that the effects of TNF-α on activation-induced cell death (AICD) are often seen later than Fas effects (37). The potential interactions between all these pathways in the context of mycobacterial infections remain to be elucidated.

Given that TNF-α is clearly required for control of bacillary replication, granuloma organization, and preventing reactivation of persistent tuberculosis in the murine model (38), and yet high serum levels of TNF-α have been associated with detrimental outcomes in human tuberculosis (39), it remains unclear what host or immune factors influence the outcome of TNF-α activity. Based on our findings in this study, we postulate that because IL-4-dependent CD30 expression enhances lymphocyte susceptibility to TNF-α-mediated apoptosis, the host’s IL-4 response to the mycobacterial infection may be one such factor. This may be one possible reason for the positive association of IL-4 expression levels with extent of disease in tuberculosis patients (13, 14).

It has been proposed that in several chronic disease models including tuberculosis, immunopathology relates to selective apoptotic depletion of type 1 cells that are necessary for resolving the infection. Type 1 cells appear to be more susceptible than “Th0” or type 2 cells to Fas-mediated AICD following activation through the TCR-CD3 complex (40). Das et al. (41) also report that production of IL-2 and IFN-γ, but not IL-4, is significantly reduced following AICD in T cells from M. tuberculosis-infected mice and suggest that this indicates selective apoptotic loss of type 1 cells. Our experiments should not be taken to imply that the cells responding to MtbS are necessarily polarized type 2 cells, because this was not investigated, nor do we suggest that there is selective loss of type 2 cells as a consequence. There has been controversy over whether CD30 expression is a “marker” of cells with the type 2 phenotype. It now appears that the CD30+ population is likely to represent those CD4+ cells that can respond to IL-4 upon restimulation, and such cells may produce only type 2 cytokines or both IL-4 and IFN-γ (42). Hence, IL-4 responsiveness probably reflects a late stage of differentiation that may be marked by CD30 expression, independent of the Th phenotype (43). It is possible, as reported by Munk et al. (21), that CD30 expression by MtbsS-stimulated lymphocytes is not exclusive to “Th2” cells. However, given that apoptosis in a cell population is asynchronous, certain susceptible cells may undergo apoptosis very early after activation and are omitted from sampling by studies taking a “snapshot” at particular time points. This is a significant technical limitation to establishing the phenotype of an apoptotic population.

Our findings in this study have raised the intriguing possibility that certain Ags or epitopes present in MtbS (but not MvacS) up-regulate IL-4 and the associated CD30 expression on activated lymphocytes. Because CD30+ cells account for one-third of apoptotic lymphocytes in MtbS-stimulated cultures, CD30 expression is likely to be a significant factor influencing apoptosis in mycobacteria-reactive lymphocytes. It is potentially useful to identify the MtbS-specific antigenic determinants that are responsible for inducing the IL-4-dependent CD30 expression, as it would then be possible to promote recognition of such Ags in a Th1 context in vaccines. Such work is currently in progress.

It has been demonstrated that the immune response to M. tuberculosis Ags, but not to M. vaccae Ags, results in an excess of TNF-α-mediated lymphocyte apoptosis by a process that involves IL-4. Dysregulated T cell apoptosis may be a link between the increased IL-4 expression in tuberculosis patients and the T cell anergy noted in this condition by other reports. The provision of a potential mechanism for the association between type 2 cytokines and immunopathological responses in human tuberculosis supports the relevance of such cytokines to the disease process. It also highlights the need to consider whether current immunotherapeutic agents on trial affect such responses and whether type 2 priming before immunization may influence the protective efficacy of current vaccine candidates.

Acknowledgments

We thank Dr. Elaine Bayley for providing the M. tuberculosis and M. vaccae sonicates and for helpful discussions.

Footnotes

  • ↵1 This work was funded by the National University of Singapore Overseas Graduate Scholarship (to G.T.S.).

  • ↵2 Address correspondence and reprint requests to Dr. Graham A. W. Rook, Department of Medical Microbiology, Windeyer Institute of Medical Sciences, Royal Free and University College Medical School, 46 Cleveland Street, London W1T 4JF, U.K. E-mail address: g.rook{at}ucl.ac.uk

  • 3 Abbreviations used in this paper: PPD, purified protein derivative; BCG, bacillus Calmette-Guérin; TRAF2, TNFR-associated factor 2; MtbS, Mycobacterium tuberculosis sonicate; MvacS, Mycobacterium vaccae sonicate; NAC; nonadherent cells; 7-AAD, 7-aminoactinomycin D; rh, recombinant human; TNF-sR1, TNF-soluble receptor 1; AICD, activation-induced cell death.

  • Received November 27, 2000.
  • Accepted May 21, 2001.

References

  1. ↵
    Lopes, M. F., V. F. da Veiga, A. R. Santos, M. E. Fonseca, G. A. Dos-Reis. 1995. Activation-induced CD4+ T-cell death by apoptosis in experimental Chagas’ disease. J. Immunol. 154: 744
    OpenUrlAbstract
  2. ↵
    Khan, I. A., T. Matsura, L. H. Kasper. 1996. Activation-mediated CD4+ T-cell unresponsiveness during acute Toxoplasma gondii infection in mice. Int. Immunol. 8: 887
    OpenUrlAbstract/FREE Full Text
  3. ↵
    Estaquier, J., M. Marguerite, F. Sahuc, N. Bessis, C. Auriault, J.-C. Ameisen. 1997. Interleukin-10-mediated T cell apoptosis during the T helper type-2 cytokine response in murine Schistosoma mansoni parasite infection. Eur. Cytokine Netw. 8: 153
    OpenUrlPubMed
  4. ↵
    Torres, M., T. Herrera, H. Villareal, E. A. Rich, E. Sada. 1998. Cytokine profiles for peripheral blood lymphocytes from patients with active pulmonary tuberculosis and healthy household contacts in response to the 30-kilodalton antigen of Mycobacterium tuberculosis. Infect. Immun. 66: 176
    OpenUrlAbstract/FREE Full Text
  5. ↵
    Toossi, Z., M. E. Kleinhenz, J. J. Ellner. 1986. Defective interleukin-2 production and responsiveness in human pulmonary tuberculosis. J. Exp. Med. 163: 1162
    OpenUrlAbstract/FREE Full Text
  6. ↵
    Holden, M., M. R. Dubin, P. H. Diamond. 1971. Frequency of negative intermediate-strength tuberculin sensitivity in patients with active tuberculosis. N. Engl. J. Med. 285: 1506
    OpenUrlCrossRefPubMed
  7. ↵
    Daniel, T. M., M. J. Oxtoby, E. Pinto, S. Moreino. 1981. The immune spectrum in patients with pulmonary tuberculosis. Am. Rev. Respir. Dis. 123: 556
    OpenUrlPubMed
  8. ↵
    Hirsch, C. G., Z. Toossi, G. Vanham, J. L. Johnson, P. Peters, A. Okwera, R. Mugerwa, P. Mugyenyi, J. J. Ellner. 1999. Apoptosis and T cell hyporesponsiveness in pulmonary tuberculosis. J. Infect. Dis. 179: 945
    OpenUrlAbstract/FREE Full Text
  9. ↵
    Fayyazi, A., B. Eichmeyer, A. Sorun, S. Schweyer, J. Herms, P. Schwarz, H. J. Radzun. 2000. Apoptosis of macrophages and T cells in tuberculosis associated caseous necrosis. J. Pathol. 191: 417
    OpenUrlCrossRefPubMed
  10. ↵
    Kremer, L., J. Estaquier, I. Wolowczuk, F. Biet, J.-C. Ameisen, C. Locht. 2000. Ineffective cellular immune response associated with T cell apoptosis in susceptible Mycobacterium bovis BCG-infected mice. Infect. Immun. 68: 4264
    OpenUrlAbstract/FREE Full Text
  11. ↵
    Toossi, Z., J. J. Ellner. 1996. Mechanisms of anergy in tuberculosis. T. M. Shinnick, ed. In Tuberculosis Vol. 215: 221 Springer-Verlag, Heidelberg.
    OpenUrlCrossRef
  12. ↵
    Zhang, M., J. Gong, D. V. Iyer, B. E. Jones, R. L. Modlin, P. F. Barnes. 1994. T cell cytokine responses in persons with tuberculosis and human immunodeficiency virus infection. J. Clin. Invest. 94: 2435
    OpenUrlCrossRefPubMed
  13. ↵
    Seah, G. T., G. M. Scott, G. A. W. Rook. 2000. Type-2 cytokine gene activation and its relationship to the extent of disease in patients with tuberculosis. J. Infect. Dis. 181: 385
    OpenUrlAbstract/FREE Full Text
  14. ↵
    van Crevel, R., E. Karyadi, F. Preyers, M. Leenders, B. Kullberg, R. H. H. Nelwan, J. W. M. van der Meer. 2000. Increased production of interleukin 4 by CD4+ and CD8+ T cells from patients with tuberculosis is related to the presence of pulmonary cavities. J. Infect. Dis. 181: 1194
    OpenUrlAbstract/FREE Full Text
  15. ↵
    Lawrence, C. E., J. C. M. Paterson, L. M. Higgins, T. T. MacDonald, M. W. Kennedy, P. Garside. 1998. IL-4-regulated enteropathy in an intestinal nematode infection. Eur. J. Immunol. 28: 2672
    OpenUrlCrossRefPubMed
  16. ↵
    Hernandez-Pando, R., L. Pavön, K. Arriaga, H. Orozco, V. Madrid Marina, G. Rook. 1997. Pathogenesis of tuberculosis in mice exposed to low and high doses of an environmental mycobacterial saprophyte before infection. Infect. Immun. 65: 3317
    OpenUrlAbstract/FREE Full Text
  17. ↵
    Duckett, C. S., C. B. Thompson. 1997. CD30-dependent degradation of TRAF2: implications for negative regulation of TRAF signalling and the control of cell survival. Genes Dev. 11: 2810
    OpenUrlAbstract/FREE Full Text
  18. ↵
    Gilfillan, M. C., P. J. Noel, E. R. Podack, S. L. Reiner, C. B. Thompson. 1998. Expression of the costimulatory receptor CD30 is regulated by both CD28 and cytokines. J. Immunol. 160: 2180
    OpenUrlAbstract/FREE Full Text
  19. ↵
    Rothe, M., V. Sama, V. M. Dixit, D. V. Goeddel. 1995. TRAF2-mediated activation of NF-κB by TNF receptor 2 and CD40. Science 269: 1424
    OpenUrlAbstract/FREE Full Text
  20. ↵
    Roy, N., Q. L. Deveraux, R. Takahashi, G. S. Salvesen, J. C. Reed. 1997. The c-IAP-1 and c-IAP-2 proteins are direct inhibitors of specific caspases. EMBO J. 16: 6914
    OpenUrlAbstract
  21. ↵
    Munk, M. E., P. Kern, S. H. Kaufmann. 1997. Human CD30+ cells are induced by Mycobacterium tuberculosis and present in tuberculosis lesions. Int. Immunol. 9: 713
    OpenUrlAbstract/FREE Full Text
  22. ↵
    Paul, R. C., J. L. Stanford, J. W. Carswell. 1975. Multiple skin testing in leprosy. J. Hyg. Camb. 75: 57
    OpenUrlCrossRef
  23. ↵
    Stewart, C. C., and S. J. Stewart. 1994. Multiparameter analysis of leukocytes by flow cytometry. In Flow Cytometry, Vol. 41 (Part A). Z. Darzynkiewicz, J. P. Robinson, and H. A. Crissman, eds. Academic Press, London, p. 61.
  24. ↵
    Loken, M. R., D. A. Wells. 1994. Immunofluorescence of surface markers. M. G. Ormerod, ed. Flow Cytometry: A Practical Approach 67 Oxford University Press, New York.
  25. ↵
    Vermes, I., C. Haanen, H. Steffens-Nakken, C. Reutelingsperger. 1995. A novel assay for apoptosis: flow cytometric detection of phosphatidylserine expression on early apoptotic cells using fluorescein-labelled annexin V. J. Immunol. Methods 184: 39
    OpenUrlCrossRefPubMed
  26. ↵
    Rothe, M., S. C. Wong, W. J. Henzel, D. V. Goeddel. 1994. A novel family of putative signal transducers associated with the cytoplasmic domain of the 75 kDa tumour necrosis factor receptor. Cell 78: 681
    OpenUrlCrossRefPubMed
  27. ↵
    Seah, G. T., G. A. W. Rook. 1999. A sensitive, non-radioactive quantitative method for measuring IL-4 and IL-4δ2 mRNA in unstimulated cells from multiple clinical samples, using nested RT-PCR. J. Immunol. Methods 228: 139
    OpenUrlCrossRefPubMed
  28. ↵
    Ellis, T. M., P. E. Simms, D. J. Slivnick, H.-M. Jäck, R. I. Fisher. 1993. CD30 is a signal-transducing molecule that defines a subset of human activated CD45RO+ T cells. J. Immunol. 151: 2380
    OpenUrlAbstract
  29. ↵
    Alzona, M., H.-M. Jäck, R. I. Fisher, T. M. Ellis. 1994. CD30 defines a subset of activated human T cells that produce IFN-γ and IL-5 and exhibit enhanced B cell helper activity. J. Immunol. 153: 2861
    OpenUrlAbstract/FREE Full Text
  30. ↵
    Filley, E. A., H. A. Bull, P. M. Dowd, G. A. W. Rook. 1992. The effect of Mycobacterium tuberculosis on the susceptibility of human cells to the stimulatory and toxic effects of tumour necrosis factor. Immunology 77: 505
    OpenUrlPubMed
  31. ↵
    Tsuyuki, S., J. Tsuyuki, K. Einsle, M. Kopf, A. J. Coyle. 1997. Costimulation through B7-2 (CD86) is required for the induction of a lung mucosal T helper cell 2 (TH2) immune response and altered airway responsiveness. J. Exp. Med. 185: 1671
    OpenUrlAbstract/FREE Full Text
  32. ↵
    Kremer, L., J. Estaquier, E. Brandt, J. C. Ameisen, C. Locht. 1997. Mycobacterium bovis bacillus Calmette-Guérin infection prevents apoptosis of resting human monocytes. Eur. J. Immunol. 27: 2450
    OpenUrl
  33. ↵
    Keane, J., H. G. Remold, H. Kornfeld. 2000. Virulent Mycobacterium tuberculosis strains evade apoptosis of infected alveolar macrophages. J. Immunol. 164: 2016
    OpenUrlAbstract/FREE Full Text
  34. ↵
    Balcewicz-Sablinska, M. K., J. Keane, H. Kornfeld, H. G. Remold. 1998. Pathogenic Mycobacterium tuberculosis evades apoptosis of host macrophages by release of TNF-R2, resulting in inactivation of TNF-α. J. Immunol. 161: 2636
    OpenUrlAbstract/FREE Full Text
  35. ↵
    Hirsch, C. S., Z. Toossi, J. L. Johnson, H. Luzze, L. Ntambi, P. Peters, M. McHugh, A. Okwera, M. Joloba, P. Mugyenyi, et al 2001. Augmentation of apoptosis and interferon-γ production at sites of active Mycobacterium tuberculosis infection in human tuberculosis. J. Infect. Dis. 183: 779
    OpenUrlAbstract/FREE Full Text
  36. ↵
    Dalton, D. K., L. Haynes, C-Q. Chu, S. L. Swain, S. Wittmer. 2000. Interferon-γ eliminates responding CD4 T cells during mycobacterial infection by inducing apoptosis of activated CD4 T cells. J. Exp. Med. 192: 117
    OpenUrlAbstract/FREE Full Text
  37. ↵
    Zheng, L., G. Fisher, R. E. Miller, J. Peschon, D. H. Lynch, M. J. Lenardo. 1995. Induction of apoptosis in mature T cells by tumour necrosis factor. Nature 377: 348
    OpenUrlCrossRefPubMed
  38. ↵
    Mohan, V. P., C. A. Scanga, K. Yu, H. M. Scott, K. E. Tanaka, E. Tsang, M. C. Tsai, J. L. Flynn, J. Chan. 2001. Effects of tumor necrosis factor α on host immune response in chronic persistent tuberculosis: possible role for limiting pathology. Infect. Immun. 69: 1847
    OpenUrlAbstract/FREE Full Text
  39. ↵
    Bekker, L. G., G. Maartens, L. Steyn, G. Kaplan. 1998. Selective increases in plasma tumor necrosis factor-α and concomitant clinical deterioration after initiating therapy in patients with severe tuberculosis. J. Infect. Dis. 178: 580
    OpenUrlAbstract/FREE Full Text
  40. ↵
    Varadhachary, A. S., P. Salgame. 1998. CD95 mediated T cell apoptosis and its relevance to immune deviation. Oncogene 17: 3271
    OpenUrlCrossRefPubMed
  41. ↵
    Das, G., H. Vohra, B. Saha, J. N. Agrewala, G. C. Mishra. 1999. Apoptosis of Th1-like cells in experimental tuberculosis. Clin. Exp. Immunol. 115: 324
    OpenUrlCrossRefPubMed
  42. ↵
    Nakamura, T., R. K. Lee, S. Y. Nam, B. K. Al-Ramadi, P. A. Koni, K. Bottomly, E. R. Podack, R. A. Flavell. 1997. Reciprocal regulation of CD30 expression on CD4+ T cells by IL-4 and IFN-γ. J. Immunol. 158: 2090
    OpenUrlAbstract
  43. ↵
    Horie, R., T. Watanabe. 1998. CD30: expression and function in health and disease. Semin. Immunol. 10: 457
    OpenUrlCrossRefPubMed
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools

Jump to section