IL-4 Influences Apoptosis of Mycobacterium-Reactive Lymphocytes in the Presence of TNF-␣1 Geok Teng Seah*† and Graham A. W. Rook2* 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-␣-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 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-␣-mediated apoptosis via TRAF2 depletion. This may be one mechanism by which IL-4 is associated with immunopatho- logical consequences in human tuberculosis. The Journal of Immunology, 2001, 167: 1230–1237. cell apoptosis is an important immune regulatory mech- losis, particularly in PPD-positive subjects (12). Thus, inappropri- anism associated with defective cell-mediated effector ate lymphocyte apoptosis is likely to negatively influence the pro- T functions in infectious diseases. Infection with Trypano- tective immune response. However, the mechanisms by which soma cruzi (1), Toxoplasma gondii (2), and Schistosoma mansoni inappropriate T cell apoptosis occurs in human tuberculosis are not (3) all result in down-regulation of cell-mediated immunity asso- clearly understood. ciated with CD4ϩ-T cell apoptosis. Pulmonary tuberculosis is as- Several groups have recently found that type 2 cytokine gene sociated with decreased lymphocyte proliferative responses to my- expression is elevated in human pulmonary tuberculosis patients cobacterial Ags (4), reduced IL-2 secretion, and IL-2 receptor and relates to disease extent (13, 14). Potential reasons for the expression (5). A significant proportion of patients (17–25%) is association of type 2 cytokines with disease in tuberculosis form unresponsive to purified protein derivative (PPD)3 (6, 7). There is the subject of the present work. In certain inflammatory situations, ϩ increased CD4 and ␥␦ T cell apoptosis when cells from tuber- TNF-␣-mediated pathology occurs only in the presence of IL-4; culosis patients are cultured with Mycobacterium tuberculosis (8). this has been shown both in murine Trichinella spiralis (15) and It has also recently been shown that T cell apoptosis occurs in mycobacterial (16) infections. We considered the possibility that areas of caseous necrosis within tuberculous granulomas (9). Mu- IL-4 produced in response to M. tuberculosis influences the sen- rine susceptibility to M. bovis bacillus Calmette-Gue´rin (BCG), sitivity of T lymphocytes to apoptosis by a TNF-␣-mediated path- associated with defects in T cell proliferative responses, correlates way. Duckett and Thompson observed that ligation of CD30 re- with T cell apoptosis in certain susceptible mouse strains (10). sults in signal-coupled depletion of TNFR-associated factor 2 Dysregulated lymphocyte apoptosis may thus be a reason for T cell (TRAF2) and sensitizes human embryonic kidney cells to apopto- anergy in tuberculosis patients (11). There is evidence that lym- sis in response to TNFR1 signal transduction (17). CD30 is a co- phocytes are much more important than soluble mediators such as stimulatory molecule chiefly expressed on activated T cells, and its TNF-␣ and IFN-␥ in mediating growth inhibition of M. tubercu- expression is IL-4 and/or CD28 dependent (18). CD30 signaling abrogates the TRAF2-mediated induction of NF-␬B by TNF-␣ *Department of Medical Microbiology, Windeyer Institute of Medical Sciences, (17). TRAF2 also plays a role in cytoprotective functions mediated Royal Free and University College Medical School, London, United Kingdom; and via stress-activated protein kinase cascades (19) and cellular in- † Department of Microbiology, National University of Singapore, Singapore hibitors of apoptosis (20). Hence, several mediators downstream of Received for publication November 27, 2000. Accepted for publication May TRAF2 that contribute to cytoprotection following TNFR1 liga- 21, 2001. tion would be affected by TRAF2 depletion. Because M. tubercu- The costs of publication of this article were defrayed in part by the payment of page ϩ charges. This article must therefore be hereby marked advertisement in accordance losis induces CD30 expression on human PBLs and CD30 cells with 18 U.S.C. Section 1734 solely to indicate this fact. are present in tuberculosis lesions (21), degradation of TRAF2 1 This work was funded by the National University of Singapore Overseas Graduate consequent to CD30 signaling may be a potential reason for the Scholarship (to G.T.S.). association of IL-4 with TNF-␣-mediated cytotoxicity in M. tu- 2 Address correspondence and reprint requests to Dr. Graham A. W. Rook, Depart- berculosis-stimulated lymphocytes. ment of Medical Microbiology, Windeyer Institute of Medical Sciences, Royal Free and University College Medical School, 46 Cleveland Street, London W1T 4JF, U.K. In this study, an in vitro model was used to examine responses E-mail address: [email protected] of lymphocytes from healthy donors to M. tuberculosis Ags. We 3 Abbreviations used in this paper: PPD, purified protein derivative; BCG, bacillus show evidence that CD30 expression in this model is at least partly Calmette-Gue´rin; TRAF2, TNFR-associated factor 2; MtbS, Mycobacterium tuber- IL-4 dependent and that reduced TRAF2 expression following culosis sonicate; MvacS, Mycobacterium vaccae sonicate; NAC; nonadherent cells; 7-AAD, 7-aminoactinomycin D; rh, recombinant human; TNF-sR1, TNF-soluble re- CD30 ligation may account for the TNF-mediated apoptosis in M. ceptor 1; AICD, activation-induced cell death. tuberculosis-stimulated lymphocytes. Copyright © 2001 by The American Association of Immunologists 0022-1767/01/$02.00 The Journal of Immunology 1231 Materials and Methods harvester (Northumbria Biologicals, Cramlington, U.K.) onto glass micro- Cell cultures fibre filter discs (Whatman, Maidstone, U.K.). The discs were dried, im- mersed in vials containing scintillation fluid (Ecoscint A; National Diag- PBMCs were isolated from a panel of eight BCG-immunized healthy do- nostics, Atlanta, GA), and the ␤ emissions were measured over 10 min per nors by density-gradient centrifugation. Cells were resuspended in RPMI vial using a Beckman LS5000CE scintillation counter (Beckman Coulter, 1640 supplemented with 2 mM L-glutamine, 100 U/ml penicillin, 100 Fullerton, CA). ␮g/ml streptomycin (all from Life Technologies, Grand Island, NY), and 10% (v/v) autologous serum, then seeded into 24-well plates (Nunclon Determination of TRAF2 expression ϫ surface; Nalge Nunc International, Roskilde, Denmark) at a density of 5 ϫ 7 5 A total of 6 10 NAC were harvested from a 7-day culture of MtbS- 10 cells/ml and incubated at 37°C in a humid 5% CO2 incubator. Only stimulated PBMCs. Magnetic beads from the CELLection pan mouse IgG nonadherent cells (NAC) were harvested for further assays. kit (Dynal, Oslo, Norway) were used with FITC-conjugated mouse anti- Crude whole-cell sonicates of M. tuberculosis H37Rv and M. vaccae human CD30 mAb, clone Ber-H2 (DAKO) to sort the NAC into CD30ϩ NCTC 11659, denoted MtbS and MvacS, respectively, were prepared by and CD30Ϫ populations, according to the manufacturer’s protocol (Dynal). the method described by Paul et al. (22). Briefly, mycobacteria grown on Dynabeads were detached from the cells by adding the releasing buffer Sauton’s medium were harvested, washed twice with PBS (pH 6.8) and supplied. Both CD30ϩ and CD30Ϫ cell populations were retained, and the suspended in 50 ml of PBS. The suspensions were sonicated for 15 min, ϫ extent of enrichment was determined by flow cytometry. then centrifuged at 70,000 g for 30 min. The supernatants were sterilized Total proteins were extracted from both CD30ϩ and CD30Ϫ populations through a 0.22-␮m filter, and the protein concentration was quantified. The Ϫ (17), and protein quantification was performed using the Bio-Rad DC Pro- sonicates were then stored in single-use aliquots at 70°C without adding tein Assay kit (Bio-Rad, Hercules, CA). SDS-PAGE was performed with other diluents or additives. To determine cellular responses to M. tuber- ␮ equal quantities of protein from each population. Western blot for TRAF2 culosis Ags, MtbS was added to the PBMC cultures at 50 g/ml (final was performed using rabbit anti-human TRAF2 polyclonal IgG Ab (Santa concentration) before incubation. Lymphocytes cultured in the presence of Cruz Biotechnology, Santa Cruz, CA) at 1:500 dilution in blocking buffer, this concentration of MtbS have up-regulated surface expression of CD30 followed by HRP-conjugated donkey anti-rabbit IgG (Amersham Pharma- (21). To assess whether responses were M. tuberculosis specific, control cia Biotech) at 1/1000 dilution in blocking buffer as the secondary Ab. wells were set up with the cells from the same donors treated either with ␮ Blocking buffer consisted of 5% (w/v) nonfat dried milk dissolved in PBS- 50 g/ml MvacS or culture medium alone. Tween 20 (0.1% v/v). The protein bands were detected using ECL Western Affinity-purified anti-human Abs and recombinant human (rh) proteins blotting detection reagents, the bands were visualized on radiographic film (all from R&D Systems, Abingdon, U.K., unless otherwise stated) were ␮ (Hyperfilm ECL; both from Amersham Pharmacia Biotech), and their size added to the cultures in certain experiments.