Coordinate Expression of CC Chemokine Ligand 5 ... - CiteSeerX

0 downloads 0 Views 471KB Size Report
Spada, F. M., E. P. Grant, P. J. Peters, M. Sugita, A. Melian, D. S. Leslie,. H. K. Lee, E. .... Villalta, F., Y. Zhang, K. E. Bibb, J. C. Kappes, and M. F. Lima. 1998. The.
The Journal of Immunology

Coordinate Expression of CC Chemokine Ligand 5, Granulysin, and Perforin in CD8ⴙ T Cells Provides a Host Defense Mechanism against Mycobacterium tuberculosis1 Frank Stegelmann,* Max Bastian,* Kay Swoboda,* Rauf Bhat,* Viviane Kiessler,* Alan M. Krensky,† Martin Roellinghoff,* Robert L. Modlin,‡ and Steffen Stenger2* The ability of CD8ⴙ T cells to kill intracellular pathogens depends upon their capacity to attract infected cells as well as their secretion of cytolytic and antimicrobial effector molecules. We examined the Ag-induced expression of three immune effector molecules contained within cytoplasmic granules of human CD8ⴙ T cells: the chemokine CCL5, the cytolytic molecule perforin, and the antimicrobial protein granulysin. Macrophages infected with virulent Mycobacterium tuberculosis triggered the expression of CCL5 in CD8ⴙ T cells only in donors with previous exposure to the tuberculosis bacteria, not in naive donors. Functionally, CCL5 efficiently attracted M. tuberculosis-infected macrophages, but failed to exert direct antibacterial activity. Infected macrophages also triggered the expression of granulysin in CD8ⴙ T cells, and granulysin was found to be highly active against drugsusceptible and drug-resistant M. tuberculosis clinical isolates. The vast majority of CCL5-positive cells coexpressed granulysin and perforin. Taken together, this report provides evidence that a subset of CD8ⴙ T cells coordinately expresses CCL5, perforin and granulysin, thereby providing a host mechanism to attract M. tuberculosis-infected macrophages and kill the intracellular pathogen. The Journal of Immunology, 2005, 175: 7474 –7483.

H

ost defense against intracellular bacteria is traditionally believed to be dominated by CD4⫹ T cells that secrete IFN-␥, thereby activating antibacterial activity in infected host cells. Evidence is accumulating that CD8⫹ T cells complement protective immunity against infection with intracellular bacteria (1). CD8⫹ T cells mediate protection by the secretion of Th1 cytokines and lysis of infected host cells. This lysis will release the intracellular bacteria, thereby reducing the reservoir of infected cells and allow activated infiltrating macrophages to take up and kill the bacilli (2). Along with cytolytic molecules, including perforin, the CD8⫹ T cells may directly kill the intracellular bacteria by the delivery of the antimicrobial peptide granulysin (3). Granulysin is stored in cytoplasmic granules of conventional and unconventional lymphocyte subsets (3– 8), and the expression is induced in CD4⫹ T cells in response to mycobacterial Ags (9 –11). The expression of granulysin correlates with protection or clinical improvement in mycobacterial disease (7, 12) and tumors (13), but is also increased at sites of autoimmune disorders (14). Cytolytic granules contain not only classical mediators of cytotoxicity, but also chemokines such as CCL3 (MIP-1␣), CCL4

*Institut fu¨r Klinische Mikrobiologie, Immunologie und Hygiene der Friedrich Alexander Universitaet Erlangen-Nuernberg, Erlangen, Germany; †Department of Pediatrics, Stanford University School of Medicine, Stanford, CA 94305; and ‡Division of Dermatology, Department of Microbiology and Immunology, and Molecular Biology Institute, University of California School of Medicine, Los Angeles, CA 90095 Received for publication April 27, 2005. Accepted for publication September 18, 2005. The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact. 1 This work was supported by the German Research Foundation (SFB643, Graduiertenkolleg 592), the Interdisciplinary Center for Clinical Research in Erlangen, the Johannes und Frieda Marohn Stiftung, Erlangen, and the European Union (VIth framework, TB-VAC). 2 Address correspondence and reprint requests to Dr. Steffen Stenger, Institut fu¨r Klinische Mikrobiologie, Immunologie und Hygiene, Wasserturmstrasse 3, D-91054 Erlangen, Germany. E-mail address: [email protected]

Copyright © 2005 by The American Association of Immunologists, Inc.

(MIP-1␤), and CCL5 (RANTES) (15, 16). Initially described as a chemokine expressed late after T cell activation (17, 18), evidence has accumulated that CCL5 is a pleiotropic mediator of the immune system (19, 20) with a remarkable affinity for cytolytic lymphocyte subsets. It enhances cytolysis, degranulation (21), and Fas ligand expression (22) in CD8⫹ T cells and NK cells. Surprisingly, CCL5 has also been implicated in the direct killing of HIV (23), trypanosomes (24 –27), and Mycobacterium tuberculosis (28). In this study we addressed the hypothesis that a chemotactic agent (CCL5), a cytolytic molecule (perforin), and an antimicrobial peptide (granulysin) are coordinately expressed and may act in concert to provide a host defense mechanism during infection with M. tuberculosis.

Materials and Methods Cell culture reagents Cells were cultured in RPMI 1640 (Biochrom) supplemented with 10% heat-inactivated FCS (Sigma-Aldrich), glutamine (2 mM; Sigma-Aldrich), 10 mM HEPES, 13 mM NaHCO3, 100 ␮g/ml streptomycin, and 60 ␮g/ml penicillin (all purchased from Biochrom). Experiments involving the infection of cells with M. tuberculosis were performed in the absence of antibiotics, and FCS was replaced by pooled human serum (generated from the blood of healthy volunteers) to optimize the phagocytosis of bacteria (29).

Abs and reagents The following Abs and reagents were used: anti-CCL5, biotinylated antiCCL5, recombinant human CCL5, anti-CCR5 (all from R&D Systems), anti-CCR1 (Mobitec), anti-CCR7, anti-CCL5-PE, anti-perforin (Loxo), antiperforin-FITC, anti-CD3-FITC, anti-CD4-PerCP, anti-CD8-allophycocyanin, anti-CD8-PerCP (all from BD Biosciences), donkey anti-rabbit FITC, goat anti-mouse FITC, goat anti-mouse Cy2, goat-anti-mouse Cy5, streptavidin-Cy5, streptavidin-Cy3, streptavidin-APC (all from Dianova), and anti-IFN-␥-PE (Miltenyi Biotec). A polyclonal rabbit serum recognizing the 9- and 15-kDa forms of granulysin was used for flow cytometry and immunostaining (4). Granulysin was expressed in Escherichia coli as previously described (30). Chemicals were purchased from Sigma-Aldrich (BSA, Con A, ionomycin, isoniazid, PHA, saponin, strontium chloride, 0022-1767/05/$02.00

The Journal of Immunology Triton X-100, and Tween 80) or Roth (calcium chloride, glycerol, and magnesium chloride).

Purification of monocytes and lymphocytes PBMCs from healthy donors were isolated from buffy coats obtained from the Sta¨dtische Klinikum Fu¨rth by Ficoll-Paque density gradient centrifugation (Amersham Biosciences). Monocytes were isolated by adherence of PBMCs for 1 h in a cell culture flask. In selected experiments, monocytes and CD8⫹ T lymphocytes were purified by positive selection using CD14or CD8-conjugated microbeads exactly as suggested by the manufacturer (Miltenyi Biotec). Nonadherent cells were removed by rinsing the flask at least five times with PBS. The adherent cells were detached by treatment with EDTA (1 mM for 10 min). The resulting cell population contained at least 95% CD14⫹ cells as determined by flow cytometry. In selected experiments, nonadherent cells were used to isolate CD8⫹ T lymphocytes by MACS (positive selection) following the instructions of the supplier (Miltenyi Biotec).

Mycobacteria and mycobacterial extract For our experiments we used a virulent laboratory strain of M. tuberculosis (H37Rv) and clinical isolates from tuberculosis patients that were susceptible (Institut fu¨r Klinische Mikrobiologie) or resistant (Dr. E. Richter, Nationales Referenzzentrum fu¨r Mykobakterien, Borstel, Germany) to first-line tuberculosis drugs. Bacteria were grown in suspension with constant, gentle rotation in roller bottles (Corning) containing Middlebrook 7H9 broth (BD Biosciences) supplemented with 1% glycerol, 0.05% Tween 80, and 10% Middlebrook oleic acid/albumin/dextrose/catalase enrichment (BD Biosciences). Aliquots from logarithmically growing cultures were frozen in PBS containing 10% glycerol, and representative vials were thawed and enumerated for viable CFU on Middlebrook 7H11 plates. Staining of bacterial suspensions with a fluorochromic dye that differentiates between live and dead bacteria (BacLight; Mobitec) revealed a viability ⬎90%. Because clumping of mycobacteria is a common problem that can influence the validity and reproducibility of the experiments, we undertook two precautions to minimize clumping: 1) culture conditions (rotation and Tween 80) were chosen to support the growth of single-cell suspensions; and 2) before in vitro infection, M. tuberculosis bacilli were sonicated in a preheated (37°C) water bath for 5 min to disrupt small aggregates of bacteria. M. tuberculosis extract was generated by collecting the supernatant (after ultracentrifugation) of mycobacterial cells that were repeatedly sonicated to extract cell wall components and intracellular Ags (31).

Stimulation of lymphocytes with mycobacteria PBMCs (1 ⫻ 106) from healthy donors were incubated in the presence of purified protein derivative (PPD;3 10 ␮g/ml; Chiron Behring), M. tuberculosis extract (10 ␮g/ml), or M. tuberculosis (five bacteria per cell) in complete medium and 10% pooled human serum for 2, 4, or 7 days. Supernatants were harvested at all time points and stored at ⫺70°C until measurement of CCL5. Donors were scored as PPD⫹ if PBMCs released ⬎200 pg/ml IFN-␥ after overnight incubation with 10 ␮g/ml PPD. IFN-␥ (Perbio Sciences) and CCL5 release (R&D Systems) were measured by sandwich ELISA exactly as suggested by the manufacturer. The sensitivity of both ELISAs was 30 pg/ml in all experiments.

Flow cytometry For the combined labeling of cell surface markers and intracellular CCL5, PBMCs were incubated with 2 ␮M monensin (GolgiStop; BD Biosciences) for 16 h. Cells were harvested and resuspended in 100 ␮l of buffer (2% FCS, 1% NaN3, and PBS without Mg2⫹/Ca2⫹) and incubated with Abs directed against CD3, CD4, CD8, or CD56 (1 ␮l each) for 30 min at 4°C. Cells were fixed (2% paraformaldehyde, 20 min), permeabilized (0.5% saponin), and washed twice using PermWash buffer (BD Biosciences). Anti-CCL5-PE (7.5 ␮l) was added and incubated for 1 h at room temperature in the dark. After a final washing step, 0.5 ⫻ 105 cells were immediately acquired and analyzed by flow cytometry (FACSCalibur; BD Biosciences). For simultaneous detection of CCL5 and IFN-␥ in CD8⫹ T cells, PBMCs were stimulated with M. tuberculosis extract (10 ␮g/ml) for 4 days. Monensin (2 ␮M) was present during the final 16 h of incubation. Cells were labeled with CD8-PerCP (3 ␮l) and then fixed with 2% paraformaldehyde for 20 min. After permeabilization, cells were incubated 3 Abbreviations used in this paper: PPD, purified protein derivative; MOI, multiplicity of infection.

7475 with CCL5-biotin (3 ␮l), followed by incubation with streptavidin-allophycocyanin (0.5 ␮l) and IFN-␥-PE (5 ␮l). Granulysin labeling was performed by incubating fixed and permeabilized (see above) cells with a polyclonal anti-granulysin serum (3 ␮l of a 1/50 dilution) for 30 min, followed by detection with FITC-conjugated donkey anti-rabbit Abs (1/250). Triple labeling for CCL5, granulysin, and perforin was performed by staining monensin-treated cells with CCL5-PE as described above. In the second step, cells were labeled with antiperforin-FITC and anti-granulysin Abs. Granulysin was detected by incubation with donkey anti-rabbit biotin (1/500) and streptavidin-allophycocyanin (0.5 ␮l).

Confocal laser microscopy Purified CD8⫹ T cells were incubated with monensin overnight and airdried on poly-L-lysine-coated glass slides. Cells were then fixed with paraformaldehyde (2%) and permeabilized with 0.1% saponin. After blocking (10% goat serum, 1% BSA, 0.2% saponin, and 0.1% Triton X-100), cells were incubated with polyclonal granulysin Abs (1/2000 dilution) in incubation buffer (0.2% saponin and 0.5% BSA) or control serum at room temperature for 1 h. Then cells were labeled with donkey anti-rabbit FITC for 30 min at 4°C. CCL5 staining was performed using biotin-labeled antiCCL5 Abs (1 h; 1/100), followed by streptavidin-Cy3 (30 min; 1/200). For triple staining, mouse anti-perforin Abs were added (1/50 dilution) and detected by goat anti-mouse Cy5 (1/250) after staining of granulysin. Sections were mounted with Mowiol medium (Sigma-Aldrich) and analyzed with a TCS NT confocal microscope (Leica Microsystems).

Infection of monocytes Monocytes were infected with single-cell suspensions of M. tuberculosis at a multiplicity of infection (MOI) of 5. After 4 h of incubation, monocytes were harvested and centrifuged at 800 rpm for 8 min. This low-speed centrifugation selectively spins down monocytes, whereas extracellular bacteria remain in the supernatant. After three cycles of centrifugation, the majority of extracellular bacteria were removed, as determined by auramine-rhodamine stain (TB-Fluor; Merck). Infected cells were plated at a concentration of 1 ⫻ 106 cells/ml in medium supplemented with 10% pooled human serum. The efficiency of infection ranged from 18 –35%. Cell viability of infected monocytes was determined by trypan blue exclusion and was ⬎95% in all experiments.

Quantification of mycobacterial growth Incorporation of [3H]uracil (Amersham Biosciences) into mycobacterial RNA was used to determine extracellular growth of M. tuberculosis as described previously (32). Briefly, 2 ⫻ 106 mycobacteria were incubated in 96-well round-bottom plates and cultured with CCL5 or granulysin at 37°C for 4 days. [3H]Uracil (1 ␮Ci) was added for the final 12–18 h of incubation. Before harvesting onto glass-fiber filters, mycobacteria were killed by treatment with 4% paraformaldehyde for 30 min. [3H]Uracil incorporation was measured in a beta counter. Background radioactivity in wells containing only culture medium was ⬍300 cpm in all experiments. To measure the growth of intracellular M. tuberculosis, cells were lysed with 0.3% saponin to release bacteria. An aliquot of unlysed, infected cells was harvested and counted. This allowed exact quantification of cells as well as determination of cellular viability by trypan blue exclusion. Recovery of cells was ⬎80% in all experiments, with cell viability regularly exceeding 95% of total cells. Cell lysates were resuspended vigorously, transferred into screw caps, and sonicated in a preheated water bath for 5 min. Aliquots of the sonicate were diluted 10-fold in 7H9 medium. Four dilutions of each sample were plated in duplicate on 7H11 agar plates and incubated at 37°C in 5% CO2 for 21 days before determining the number of CFU.

Migration assays The chemotactic stimulus (CCL5 or medium) was given into the lower chamber of a 96-well chemotaxis plate (5-␮m pores; Receptor Technologies) in a volume of 29 ␮l. Monocytes (1 ⫻ 105) were placed on the upper membrane of each well (80 ␮l; complete medium supplemented with 10% FCS, 0.9 mM CaCl2, and 0.5 mM MgCl2). After 1 h of incubation at 37°C, migrated cells were harvested from the lower chamber. To quantify the number of migrated cells, 0.5 ⫻ 105 calibrate beads (BD Biosciences) were added to each sample. Cells were acquired by flow cytometry, and acquisition was terminated in all samples when 5 ⫻ 103 calibrate beads had been detected. The migration factor was calculated by dividing the number of cells attracted by CCL5 by the number of cells in the medium control. Detection of intracellular M. tuberculosis by flow cytometry was performed as described previously (29). Briefly, mycobacteria were incubated

7476

COORDINATE EXPRESSION OF CCL5 AND GRANULYSIN IN CD8⫹ T CELLS

with an equal volume of SYTO9 (Mobitec) for 15 min at room temperature in the dark. After seven washing steps in PBS (500 ␮l), single-cell suspensions of M. tuberculosis were used for infection of purified monocytes. Infected cells were enumerated by flow cytometry at a wavelength of 488 nm (FITC channel).

Statistical analysis Data are presented as the mean ⫾ SEM, except where stated otherwise. Student’s t test was used to determine statistical significance between two differentially treated cultures. Differences were considered significant at p ⬍ 0.05.

Results

CD8⫹ T cells are the main source of CCL5 in human peripheral blood To identify CCL5-producing T cell subsets in humans, we performed intracellular flow cytometry of freshly isolated PBMCs from healthy donors. A median of 21% (n ⫽ 46) of the lymphocytes (range, 2–53%) constitutively expressed CCL5. CCL5 was preferentially expressed in CD8⫹ lymphocytes (median, 34%; range, 8 – 81%), and only a minority of CD4⫹ cells stained positively (median, 7%; range, 2–17%; Fig. 1, A and B). The majority of NK cells also expressed CCL5 (median, 76%; range, 53–92%; data not shown), suggesting preferential expression of CCL5 in cytolytic lymphocyte subsets. Accordingly, degranulation of cytoplasmic granules in PBMCs with 20 mM strontium overnight (33– 36) almost completely cleared intracellular stores of CCL5 (Fig. 1C, middle panel). The pharmacological inhibition of the export of proteins overnight (2 ␮M monensin) resulted in a significant increase in the mean fluorescence intensity of CCL5-positive cells (Fig. 1C, right panel), indicating that CCL5 is continually synthesized. However, the percentage of cells expressing CCL5 remained unchanged (51%). Therefore, CCL5 is produced and stored constitutively in a subset of CD8⫹ T cells in the peripheral blood of healthy human donors. Mycobacterial Ags up-regulate CCL5 in CD8⫹ T cells of PPD-positive donors Host defense mechanisms against invading microbes are tightly regulated and require the induction of effector molecules during infection. Therefore, we asked whether the expression of CCL5 in CD8⫹ T cells was increased after Ag-specific activation. M. tuberculosis, the causative agent of tuberculosis, was used as a model Ag, because CD8⫹ T cells are critical for protective immunity against this major human pathogen (37). The donors tested were classified according to their reactivity to PPD as being responsive or nonresponsive to the tuberculosis bacteria. PBMC from 12 PPD-reactive healthy donors were stimulated with M. tuberculosis extract for 2, 4, or 7 days and stained for CCL5. The expression of CCL5 within the population of CD8⫹ T cells was unchanged after 2 days, but a significant increase became apparent after 4 days, and levels remained higher than in freshly isolated cells throughout the 7-day observation period (Fig. 2A). This effect was Ag specific, because the expression of CCL5 was not induced in PPD-negative donors (Fig. 2A) or in the absence of Ag (not shown). Detection of CCL5 by flow cytometry does not allow for discrimination between increased storage and release, because monensin, an inhibitor of protein export, was present during the final 16 h of incubation for technical reasons. Therefore, the concentration of CCL5 was measured in the supernatants of purified CD8⫹ T cells cultured with M. tuberculosis extract-pulsed autologous macrophages. In PPD-positive donors (n ⫽ 5), the amount of CCL5 increased over time and reached a maximum after 4 days of incubation (4.1 ⫾ 0.5 ng/ml; Fig. 2B), confirming that CCL5 is

FIGURE 1. Resting CD8⫹ T cells produce and store CCL5 constitutively. A, Freshly isolated PBMCs were incubated in the presence of 2 ␮M monensin for 16 h and stained with Abs directed against CD3 (FITC), CD4 (PerCP), or CD8 (allophycocyanin) and CCL5 (PE). Cells were gated on the lymphocyte- (left panel), CD4⫹- (middle panel), or CD8⫹-positive population and analyzed for the expression of CCL5. Bright lines show the isotype controls. The figure shows the result of a representative donor (n ⫽ 46). B, Percentages of CCL5-expressing cells within CD3⫹, CD4⫹, and CD8⫹ T lymphocytes for all 46 donors. F, Median. C, Freshly isolated PBMCs were incubated with or without strontium (20 mM) or monensin (2 ␮M) overnight and analyzed for the expression of CCL5. The experiment was repeated four times with similar results.

secreted late after T cell activation (17). On the average, the release of CCL5 increased 3.7 ⫾ 1.1-fold after stimulation with M. tuberculosis extract. The majority of CCL5 release was Ag specific, because PPD-nonreactive donors (n ⫽ 5) responded significantly less (1.1 ⫾ 0.1 ng/ml) than PPD-reactive donors. CCL5 secretion in PPD-naive donors is most likely monocyte derived, because similar concentrations were measured in cultures of M. tuberculosis extract-pulsed monocytes in the absence of T cells (data not shown). During natural infection, it is of pivotal importance for T lymphocytes to sense, interact with, and destroy macrophages infected with virulent bacteria. We previously demonstrated that CD8⫹ T cells recognize Ags presented by M. tuberculosis-infected human macrophages (34) and now asked whether this interaction would result in the induced expression of CCL5. Purified CD8⫹ T cells

The Journal of Immunology

7477

FIGURE 2. CD8⫹ T cells up-regulate and release CCL5 in response to mycobacterial Ags and M. tuberculosis. A, PBMCs (2 ⫻ 106) were cultured in the presence of M. tuberculosis extract (10 ␮g/ml) for the times indicated. Monensin was present for the final 16 h of incubation. Cells were stained with anti-CD8-allophycocyanin, followed by intracellular staining of CCL5. Data were analyzed by gating on CD8⫹ T cells and calculating the percentage of CCL5-expressing cells within this population. The figure gives the results of one typical PPD-reactive donor of 12 (F) and one PPD-naive donor of 15 (E). The error bars show the SEM calculated from three independent experiments performed with cells from this donor. The differences between PPD-reactive and PPD-naive donors on days 4 and 7 are significant (ⴱ, p ⬍ 0.01). B, CD8⫹ T cells and monocytes were purified from PBMCs by positive immunomagnetic selection. CD8⫹ T cells were incubated with M. tuberculosis extract-pulsed monocytes overnight. At the indicated time points, supernatants were harvested and analyzed for CCL5 concentration by ELISA. The graph shows the average CCL5 concentration ⫾ SEM of five donors. Differences between unstimulated and M. tuberculosis extract-stimulated T cells are statistically significant (ⴱ, p ⬍ 0.001). C, Purified monocytes were infected overnight with M. tuberculosis (MOI, 5; efficacy of infection, 24 ⫾ 7%) and added to CD8⫹ T cells at a ratio of 1:1. After 2, 4, and 7 days, the percentage of CCL5-positive cells was determined by flow cytometry. The graph presents the average of five donors ⫾ SEM. Differences on days 4 and 7 are statistically significant (ⴱ, p ⬍ 0.01). D, Supernatants from parallel cultures to C were analyzed for CCL5. Differences between PPDpositive and PPD-naive donors were statistically significant at all time points (ⴱ, p ⬍ 0.01). E, PBMCs were stimulated with M. tuberculosis extract for 4 days and triple stained for CD8, CCL5, and IFN-␥. CD8⫹ T cells were gated, and the CCL5 and IFN-␥ staining of unstimulated (left panel) and stimulated cells (right panel) is shown. Quadrants were determined according to the isotype controls. The figure shows a representative result of three independent experiments.

from PPD-reactive donors were incubated with autologous M. tuberculosis-infected macrophages, and the expression of CCL5 was assessed by flow cytometry after 2, 4, and 7 days (Fig. 2C). CCL5 expression was induced significantly (average increase, 41 ⫾ 12%; p ⬍ 0.01) in PPD-positive donors, but remained essentially unchanged in PPD-negative donors. CCL5 was also released by purified CD8⫹ T cells (n ⫽ 5; Fig. 2D). CCL5 concentrations were also slightly increased in cells derived from PPD-negative donors in agreement with T cell-independent release by macrophages (38). CCL5 release was below the level of detection (30 pg/ml) in all cultures in the absence of M. tuberculosis extract or M. tuberculosis (data not shown). Finally, we asked whether CD8⫹ T cells up-regulate both CCL5 and IFN-␥, a surrogate marker of Ag-specific responses to mycobacterial Ags. We found that the majority of cells producing IFN-␥ in response to M. tuberculosis extract (8%) also stained positively for CCL5 after 4 days of incubation (Fig. 2E). Polyclonal stimulation with PMA/ionomycin, anti-CD3, or Con A for 4 days also triggered the expression of CCL5 in CD8⫹ T cells in both PPD-positive and PPD-negative donors (Fig. 3). These results show that the expression of CCL5 is not limited to M. tuberculosis-specific cells, but applies to a significant subset of CD8⫹ T cells independent of the mode of activation.

CCL5 attracts macrophages infected with M. tuberculosis The induction of lytic and antimicrobial activity by CD8⫹ T cells requires direct contact with infected macrophages. We hypothesized that CD8⫹ T cell-derived CCL5 contributes to the formation of this cell-cell interaction by the attraction of M. tuberculosisinfected macrophages. We quantified the chemotactic activity of CCL5 (100 ng/ml) toward M. tuberculosis-infected (MOI, 5) and uninfected monocytes in a chemotaxis chamber. To take into account the variability of different donors, experiments were performed using monocytes from 11 donors. The chemoattractive activity of CCL5 was significantly higher toward infected macrophages compared with uninfected macrophages (Fig. 4). Fig. 4A shows the original data from one representative donor (migration of 733 uninfected macrophages vs migration of 1549 infected macrophages). Fig. 4B summarizes the migration factors of all 11 donors and shows unequivocally that infection with M. tuberculosis supports the migration of monocytes toward CCL5. Previous studies indicate that infection of macrophages leads to increased expression of CCR1 and CCR5, both receptors for CCL5 (39). To explore whether the CCL5-induced migration of infected monocytes was dependent on these chemokine receptors, the chemotaxis assays were performed after pretreating the infected cells

7478

COORDINATE EXPRESSION OF CCL5 AND GRANULYSIN IN CD8⫹ T CELLS

FIGURE 3. Mitogens trigger the expression of CCL5 in CD8⫹ T cells. PBMCs from healthy donors (2 ⫻ 106/ml) were stimulated with PMA (10 ng/ml) and ionomycin (500 ng/ml), anti-CD3 (2 ␮g/ml), or Con A (5 ␮g/ ml) for 4 days. Monensin was present during the final 16 h of incubation. Double staining was performed for CD8 (PerCP) and RANTES (allophycocyanin), and 100,000 cells were analyzed by flow cytometry. Quadrants were set according to the appropriate isotype controls. The experiment shows a representative result of three independent experiments.

with neutralizing Abs to CCR1 or CCR5 (Fig. 4C). Migration was strongly inhibited by Abs to CCR5 (migration factor, 4.7 vs 2.1), but not CCR1. Similarly, blocking Abs to CCR7, another chemokine receptor that binds to chemokine ligand 21, but not CCL5, had no effect on the migration of infected macrophages. These results demonstrate that the M. tuberculosis-induced enhanced migration toward CCL5 is mediated by CCR5. To prevent apoptosis of heavily infected monocytes, our culture conditions were chosen to limit the percentage of infected cells to 25– 44% of the monocytes. To discriminate between uninfected and infected cells, we labeled extracellular M. tuberculosis with a fluorescent dye before infection (29) and performed chemotaxis assays as described above. The percentage of infected cells was significantly ( p ⬍ 0.01) higher within the cells that had migrated compared with the initial inoculum (31 ⫾ 3 vs 51 ⫾ 5%; Fig. 4D). These experiments demonstrate that CCL5 attracts infected macrophages and may initiate a host mechanism to kill intracellular pathogens. Antimycobacterial activity of CCL5 and granulysin Because CD8⫹ T cells mediate direct antibacterial activity and CCL5 has been shown to kill viruses and parasites (23–28), we determined whether CCL5 inhibits the growth of extracellular M. tuberculosis. CCL5 (100 ng/ml) failed to modulate the metabolic activity of mycobacteria after 12, 24, 36, 60, or 72 h, as determined by incorporation of [3H]uracil in five independent experiments (Fig. 5A, F). In comparison, granulysin (40 ␮M), another molecule stored in cytolytic granules and previously shown to have antimicrobial activity (3, 4), efficiently decreased mycobacterial metabolism (Fig. 5A, f). Differences from control cultures were evident after 36 h (509 ⫾ 72 cpm in granulysin-treated cultures vs 1231 ⫾ 124 cpm in control cultures) and steadily increased over time. The antibacterial spectrum of granulysin was not limited to

FIGURE 4. CCL5 attracts M. tuberculosis-infected macrophages. A, Monocytes (4 ⫻ 105) were infected with M. tuberculosis (MOI, 5) for 24 h. CCL5 (100 ng/ml) or medium was put in the lower well of a chemotaxis chamber. Uninfected or infected macrophages (1 ⫻ 105) were placed on the upper membrane. All wells were set up in duplicate. After 1 h, migrated cells were harvested and quantified by flow cytometry. The figure shows a representative example of 11 experiments. B, The graph summarizes the migration factors of all 11 donors. Migration factors were calculated by dividing the number of cells attracted by CCL5 and by medium alone. The horizontal line shows the median. C, Migration assays were performed as described for A, except that infected macrophages were preincubated with neutralizing Abs to CCR1, CCR5, or CCR7 (all at 20 ␮g/ml). The graph shows the average migration factor ⫾ SEM calculated from three independent experiments. The difference between anti-CCR5-treated cultures and the controls is significant (ⴱ, p ⬍ 0.05). D, Mycobacteria were labeled with a fluorescent dye before infection and migration toward the CCL5 gradient was determined as described above. The percentage of infected macrophages was determined by flow cytometry in the total population and in the population that had migrated toward CCL5 after 1 h of incubation. The graph shows the results of five independent experiments. Differences in the number of infected macrophages before and after migration are significant (p ⬍ 0.01).

drug-susceptible strains of M. tuberculosis, but also comprised clinical isolates that were resistant to the first-line antimycobacterial drugs isoniazid (five strains) or rifampin (four strains; Fig. 5B). Granulysin (40 ␮M) reduced the growth of extracellular bacteria

The Journal of Immunology

7479 tensity of positive cells (data not shown), demonstrating that granulysin, similar to what we demonstrated for CCL5, is stored and synthesized constitutively in a considerable fraction of CD8⫹ T cells. To determine whether granulysin expression is increased after activation by microbial Ags, we incubated purified CD8⫹ T cells from 14 PPD-reactive and 14 PPD-naive donors with M. tuberculosis extract or autologous M. tuberculosis-infected macrophages for 2, 4, or 7 days. Granulysin expression was induced after 4 and 7 days in response to M. tuberculosis extract (Fig. 6B, left panel) and infected macrophages (Fig. 6B, right panel) in PPD-positive donors. In contrast, the expression of granulysin in PPD-naive donors remained unchanged (Fig. 6B, E), showing that the expression of granulysin in CD8⫹ T cells is Ag specific. The number of granulysin-positive CD8⫹ T cells in cultures stimulated with mycobacterial Ags or control cultures from all individual donors is given in Fig. 6C and demonstrates the consistency of our observation. CD8⫹ T cells coexpress CCL5 and granulysin

FIGURE 5. Antimycobacterial activity of CCL5 and granulysin. A, Extracellular M. tuberculosis (2 ⫻ 106) were cultured alone (E) or in the presence of CCL5 (100 ng/ml; F) or granulysin (40 ␮M; f) for the times indicated. [3H]Uracil (1 ␮Ci) was added for the final 12 h of incubation, and uracil incorporation was measured in a beta counter. All wells were set up in triplicate. The graph presents a representative result of five independent experiments ⫾ SD calculated from triplicate determinations. Differences between untreated and granulysin-treated cultures are significant (ⴱ, p ⬍ 0.01). B, Bacteria (1.5 ⫻ 105) of clinical isolates of M. tuberculosis (F, susceptible; f, isoniazid resistant; Œ, rifampin resistant) were cultured alone (open symbols) or in the presence of granulysin (40 ␮M; closed symbols) for 72 h. The number of CFU was determined by plating serial 10-fold dilutions on 7H11 agar plates and counting colonies after 21 days of culture. The graph shows the average number of CFU for susceptible (n ⫽ 5), isoniazid-resistant (n ⫽ 5), and rifampin-resistant (n ⫽ 4) strains.

by ⬃50% compared with the untreated control. The effect of granulysin was bacteriostatic, because the number of viable bacilli at the end of the observation period exceeded that in the initial inoculum (1.3 ⫻ 105 vs 1.8 ⫾ 0.2 ⫻ 105). However, if granulysin is delivered into the infected cell via the granule exocytosis by Ag-activated CD8⫹ T cells, bactericidal activity is observed (3, 34). In vivo, additional antimicrobial peptides as well as the action of perforin and granzymes may complement the action of granulysin, thereby optimizing its biological activity against intracellular pathogens. Mycobacterial Ags up-regulate granulysin in CD8⫹ T cells of PPD-positive donors Next, we examined whether granulysin was expressed in CD8⫹ T cells of freshly isolated PBMCs. Granulysin expression in total PBMCs varied widely between 8 and 54% in 258 healthy donors. Phenotypic analysis of granulysin-expressing T cells revealed that the majority were CD8⫹ (Fig. 6A). Remarkably, a subset of CD4⫹ T cells was also found to store granulysin, albeit to a considerably lower extent. The second major population within the granulysinpositive cells in the peripheral blood is NK cells (data not shown). Treatment with strontium depleted granulysin and inhibition of protein export by monensin increased the mean fluorescence in-

Because not all CD8⫹ T cells express CCL5 (Fig. 1A) or granulysin (Fig. 6A), we investigated whether these molecules are jointly expressed or are located in distinct CD8⫹ T cell subsets. We quadruple-labeled PBMCs for CCL5, granulysin, CD4, and CD8 and analyzed the cells by flow cytometry. In all nine donors investigated, the majority of cells expressing granulysin also stained positively for CCL5 (Fig. 7A, left panel) and vice versa, even though the percentage of double-positive cells varied considerably between different donors (range, 13– 61%). Further analysis of the double-positive population (gate R1) revealed that the cells were predominantly CD8⫹ (Fig. 7A, middle and right panels), with an average CD8:CD4 ratio of 15:1. Confocal laser microscopy of CD8⫹ T cells confirmed coexpression of CCL5 and granulysin at the cellular level in the majority of cells (Fig. 7B). Remarkably, a considerable fraction of intracellular granules contained either CCL5 or granulysin, suggesting the existence of a functional specialization of intracellular compartments. A precondition for the existence of a host defense mechanism involving the attraction of infected macrophages and killing of the pathogen is the availability of the lytic molecule perforin, which is required for granulysin to enter the host cell (3). Therefore, we asked whether CD8⫹ T cells that contain CCL5 simultaneously express perforin and granulysin. We performed triple labeling of purified CD8⫹ T cells for CCL5, granulysin, and perforin and analyzed cells by flow cytometry (Fig. 8A). The vast majority of CCL5-positive cells expressed perforin (92%) and granulysin (96%), indicating that CD8⫹ T cells are equipped with the complete molecular program to attract macrophages and exert antibacterial activity. Confocal laser microscopy confirmed that the majority of CD8⫹ T cells store CCL5, perforin, and granulysin (Fig. 8B), but shows that there is a considerable heterogeneity at the granular level. All combinations of single-, double-, and triplepositive granules were detectable, with the striking exception that CCL5- and granulysin-expressing granules invariably coexpressed perforin. In summary, these findings show that a subset of CD8⫹ T cells contains chemokines, cytolytic molecules, and antibacterial peptides that are coordinately induced by microbial Ags. The combined action of these functionally distinct molecules provides a new defense mechanism by which an Ag-specific T cell can attract infected macrophages and kill the intracellular pathogen.

7480

COORDINATE EXPRESSION OF CCL5 AND GRANULYSIN IN CD8⫹ T CELLS

FIGURE 6. Induced expression of granulysin in response to M. tuberculosis. A, Freshly isolated PBMCs were stained with Abs directed against CD4 or CD8 and granulysin. Lymphocytes were gated according to forward/side scatter characteristics. The histograms present the granulysin staining within the total lymphocyte (left panel), CD4⫹ (middle panel), or CD8⫹ (right panel) population. The figure shows a typical result of 39 donors. B, Purified CD8⫹ T cells (2 ⫻ 106) from PPD-naive and reactive donors were cultured in the presence of autologous monocytes pulsed with M. tuberculosis extract (10 ␮g/ml; left panel) or infected with M. tuberculosis (MOI, 5; efficacy of infection, 21 ⫾ 9%; right panel). At the time points indicated, cells were labeled with granulysin Abs. The panels show the percentage of granulysin-expressing cells of one representative donor of 14. The differences between PPD-negative and PPD-positive donors on days 4 and 7 are significant (ⴱ, p ⬍ 0.01). C, Percentage of granulysin-positive CD8⫹ T cells of all 14 donors investigated on day 4 of incubation with M. tuberculosis extract-pulsed or M. tuberculosis-infected macrophages.

Discussion ⫹

CD8 T cells are known to contribute to host defense by their ability to lyse infected targets and deliver antimicrobial effector molecules. However, once activated by Ag-expressing targets, these T cells require the capacity to attract other infected cells and kill the intracellular pathogen. Our results demonstrate that human CD8⫹ T cells induce the expression of CCL5 and granulysin upon recognition of M. tuberculosis-infected macrophages. CCL5 efficiently attracted mycobacteria-infected macrophages, but failed to exhibit antimicrobial activity. In contrast, granulysin exerted antimicrobial activity against M. tuberculosis, including drug-resistant clinical isolates. CCL5, granulysin, and perforin colocalized in a subset of CD8⫹ T cells, suggesting that chemotactic, cytolytic, and antimicrobial molecules provide a functional unit of the human T cell response to attract infected macrophages and kill intracellular pathogens. We show that CCL5 attracts M. tuberculosis-infected macrophages more efficiently than uninfected macrophages (Fig. 4). Increased migration was mediated via CCR5, because neutralizing Abs to CCR5 antagonized the effect (Fig. 4C), and infection of human macrophages with M. tuberculosis induced the expression of CCR5 (39). Intriguingly, CCR5 expression was more prominent in alveolar macrophages of tuberculosis patients than in control patients, suggesting that this mechanism is active during natural infection (39). CCR5-deficient mice retained the ability to form functional granulomas and control infection with M. tuberculosis

(40). However, this could reflect the fact that the biological activity of CCL5 can be mediated by the alternative receptors CCR1, CCR3, and CCR4. In the microenvironment of a tuberculous granuloma, this mechanism might be involved in the establishment of intimate contact between CD8⫹ T cells and infected cells that is required for the injection of effector molecules such as granulysin. To our knowledge this is the first report to demonstrate that bacterial Ags activate CD8⫹ T cells to increase the storage and release of CCL5. Mycobacteria-induced expression of CCL5 was Ag specific, because PPD-negative donors failed to respond (Fig. 2), and the effect was abrogated by blocking Abs to MHC class I (data not shown). Ag-specific release of ␤-chemokines by CD8⫹ T cells was also induced by viral Ags (15), suggesting that the release of ␤-chemokines provides hitherto underestimated defense mechanisms of CD8⫹ T cells against human pathogens of different origins. Support for the significance of this mechanism in vivo came from observations in AIDS patients, in which the number of CCL5-expressing CD8⫹ T cells in the peripheral blood was significantly higher than in healthy controls (41). Our initial findings on the induced expression of CCL5 and the efficient attraction of infected macrophages prompted us to investigate the expression of related granular effector molecules. The vast majority of CCL5-positive cells coexpressed the antimicrobial protein granulysin, and the expression of both molecules was induced by M. tuberculosis-infected human macrophages. CCL5 and granulysin were originally isolated from the same human cytotoxic

The Journal of Immunology

FIGURE 7. CCL5 and granulysin are coexpressed in CD8⫹ T cells A, PBMCs were incubated with monensin for 16 h and stained with antiCCL5-PE, anti-granulysin (detected with goat anti-mouse FITC), antiCD4-PerCP, and anti-CD8-allophycocyanin. Cells (5 ⫻ 105) were acquired and analyzed by flow cytometry. The left panel shows the double staining for granulysin and CCL5. Double-positive cells were gated (R1) and analyzed for the expression of CD4 (middle panel) or CD8 (right panel). The experiment was repeated with nine donors and yielded similar results. B, Purified CD8⫹ T cells were air-dried on glass slides, fixed, permeabilized, and stained with anti-CCL5-biotin and streptavidin-TRITC (red) and antigranulysin/donkey anti-rabbit FITC (green). The staining was analyzed by confocal laser microscopy. The pictures depict a representative area of the slide at a magnification of ⫻750.

T cell subtracted library in a search for genes expressed 3–5 days after activation (17, 42). These kinetics are similar to those described for other proteins that are associated with T cell terminal differentiation such as granzymes and perforin (43). This unusual late expression has been studied in detail and was shown to be mediated by a novel transcription factor of late-activated lympho-

FIGURE 8. CCL5-positive cells coexpress perforin and granulysin. A, Purified CD8⫹ T cells were incubated with monensin for 16 h and stained with anti-CCL5-PE, anti-granulysin (labeled with goat antirabbit biotin and streptavidin-allophycocyanin), antiperforin-FITC, or the respective control Abs. CCL5positive cells were gated (upper left panel; R2) and analyzed for the expression of granulysin and perforin (lower right panel). The graph presents a representative result of five experiments. B, Purified CD8⫹ cells were prepared as described above and stained with anti-CCL5-biotin/streptavidin-Cy3, anti-granulysin/ goat anti-rabbit-Cy2, and anti-perforin/goat antimouse-Cy5. Four hundred cells from three different donors were evaluated for perforin (green), granulysin (red), or CCL5 (blue)-positive cells by confocal laser microscopy at a magnification of ⫻850. Colors depict the following: red, CCL5; green, granulysin; blue, perforin; yellow, CCL5/granulysin; pink, CCL5/perforin; light blue, perforin/granulysin; and white, granulysin/perforin/CCL5.

7481 cyte-1, which is a member of the Kru¨ppel-like family of transcription factors regulating CCR5 expression (44). Granulysin regulation does not involve these factors, and its immediate upstream region appears to be of limited importance (45). Taken together, these findings illustrate that CCL5 and granulysin are available for immediate release upon T cell activation, and the supplies are replenished during Ag-specific responses. Even though CCL5, perforin, and granulysin colocalized at the cellular level (Figs. 7 and 8), there is evidence that CCL5 is stored in functionally and structurally unique intracellular compartments that lack perforin and granzymes (46). Our results using unselected CD8⫹ T cells also showed a fraction of granules that selectively contain CCL5, but not granulysin or perforin (Fig. 8). Remarkably, all granules containing CCL5 and granulysin unequivocally coexpressed perforin, suggesting the existence of a host defense pathway that includes the combined action of chemotactic, cytolytic, and antimicrobial molecules. There is an increasing amount of evidence of a close evolutionary and functional relationship between antimicrobial peptides and chemokines (47, 48). It is well established that defensins, classically considered as antimicrobial peptides of the innate immune system, are potent chemoattractants (49 –53). More recently, it was acknowledged that granulysin, the defensin of the adaptive immune response, also combines antibacterial and chemotactic activities (54). Finally, the chemokine CCL5 that is coexpressed with granulysin in a subset of CD8⫹ T cells (Fig. 7), executes microbial pathogens (23–28), even though we could not detect activity against virulent M. tuberculosis. Therefore, chemokines and antibacterial peptides should not be associated with a single function (e.g., directing cell movement or antimicrobial activity, respectively), but should be classified as multifunctional molecules that shape host defenses in overlapping ways. Our results provide one mechanism by which chemokines and antimicrobial peptides combine their distinct effector capacities to mediate an efficient immune response. These findings encourage the development of vaccination strategies that stimulate multifunctional CD8⫹ T cell

7482

COORDINATE EXPRESSION OF CCL5 AND GRANULYSIN IN CD8⫹ T CELLS

subsets that are armed with effector molecules to exert chemotactic, cytolytic, and antimicrobial activities.

Acknowledgments We appreciate the expert technical assistance of Kirstin Castiglione and Nives Schwerdtner, and we thank Dr. Elvira Richter (Nationales Referenzzentrum fu¨r Mykobakterien) for supplying drug-resistant mycobacteria. We are grateful to the Sta¨dtische Klinikum Fu¨rth for supplying buffy coats.

Disclosures

21.

22.

23.

24.

The authors have no financial conflict of interest.

References 1. Stenger, S. 2001. Cytolytic T cells in the immune response to Mycobacterium tuberculosis. Scand. J. Infect. Dis. 33: 483– 487. 2. Kaufmann, S. H. 1988. CD8⫹ T lymphocytes in intracellular microbial infections. Immunol. Today 9: 168 –174. 3. Stenger, S., D. A. Hanson, R. Teitelbaum, P. Dewan, K. R. Niazi, C. J. Froelich, T. Ganz, S. Thoma-Uszynski, A. Melian, C. Bogdan, et al. 1998. An antimicrobial activity of cytolytic T cells mediated by granulysin. Science 282: 121–125. 4. Pena, S. V., D. A. Hanson, B. A. Carr, T. J. Goralski, and A. M. Krensky. 1997. Processing, subcellular localization, and function of 519 (granulysin), a human late T cell activation molecule with homology to small, lytic, granule proteins. J. Immunol. 158: 2680 –2688. 5. Spada, F. M., E. P. Grant, P. J. Peters, M. Sugita, A. Melian, D. S. Leslie, H. K. Lee, E. van Donselaar, D. A. Hanson, A. M. Krensky, et al. 2000. Selfrecognition of CD1 by ␥/␦ T cells: implications for innate immunity. J. Exp. Med. 191: 937–948. 6. Dieli, F., M. Troye-Blomberg, J. Ivanyi, J. J. Fournie, A. M. Krensky, M. Bonneville, M. A. Peyrat, N. Caccamo, G. Sireci, and A. Salerno. 2001. Granulysin-dependent killing of intracellular and extracellular Mycobacterium tuberculosis by V␥9/V␦2 T lymphocytes. J. Infect. Dis. 184: 1082–1085. 7. Ochoa, M. T., S. Stenger, P. A. Sieling, S. Thoma-Uszynski, S. Sabet, S. Cho, A. M. Krensky, M. Rollinghoff, E. Nunes Sarno, A. E. Burdick, et al. 2001. T-cell release of granulysin contributes to host defense in leprosy. Nat. Med. 7: 174 –179. 8. Gansert, J. L., V. Kiessler, M. Engele, F. Wittke, M. Rollinghoff, A. M. Krensky, S. A. Porcelli, R. L. Modlin, and S. Stenger. 2003. Human NKT cells express granulysin and exhibit antimycobacterial activity. J. Immunol. 170: 3154 –3161. 9. Canaday, D. H., R. J. Wilkinson, Q. Li, C. V. Harding, R. F. Silver, and W. H. Boom. 2001. CD4⫹ and CD8⫹ T cells kill intracellular Mycobacterium tuberculosis by a perforin and Fas/Fas ligand-independent mechanism. J. Immunol. 167: 2734 –2742. 10. Worku, S., and D. F. Hoft. 2003. Differential effects of control and antigenspecific T cells on intracellular mycobacterial growth. Infect. Immun. 71: 1763–1773. 11. Toossi, Z., H. Mayanja-Kizza, A. Kanost, K. Edmonds, M. McHugh, and C. Hirsch. 2004. Protective responses in tuberculosis: induction of genes for interferon-␥ and cytotoxicity by Mycobacterium tuberculosis and during human tuberculosis. Scand. J. Immunol. 60: 299 –306. 12. Dieli, F., G. Sireci, N. Caccamo, C. Di Sano, L. Titone, A. Romano, P. Di Carlo, A. Barera, A. Accardo-Palumbo, A. M. Krensky, et al. 2002. Selective depression of interferon-␥ and granulysin production with increase of proliferative response by V␥9/V␦2 T cells in children with tuberculosis. J. Infect. Dis. 186: 1835–1839. 13. Kishi, A., Y. Takamori, K. Ogawa, S. Takano, S. Tomita, M. Tanigawa, M. Niman, T. Kishida, and S. Fujita. 2002. Differential expression of granulysin and perforin by NK cells in cancer patients and correlation of impaired granulysin expression with progression of cancer. Cancer Immunol. Immunother. 50: 604 – 614. 14. Raychaudhuri, S. P., W. Y. Jiang, S. K. Raychaudhuri, and A. M. Krensky. 2004. Lesional T cells and dermal dendrocytes in psoriasis plaque express increased levels of granulysin. J. Am. Acad. Dermatol. 51: 1006 –1008. 15. Wagner, L., O. O. Yang, E. A. Garcia-Zepeda, Y. Ge, S. A. Kalams, B. D. Walker, M. S. Pasternack, and A. D. Luster. 1998. ␤-Chemokines are released from HIV-1-specific cytolytic T-cell granules complexed to proteoglycans. Nature 391: 908 –911. 16. Caccamo, N., S. Milano, C. Di Sano, D. Cigna, J. Ivanyi, A. M. Krensky, F. Dieli, and A. Salerno. 2002. Identification of epitopes of Mycobacterium tuberculosis 16-kDa protein recognized by human leukocyte antigen-A*0201 CD8⫹ T lymphocytes. J. Infect. Dis. 186: 991–998. 17. Schall, T. J., J. Jongstra, B. J. Dyer, J. Jorgensen, C. Clayberger, M. M. Davis, and A. M. Krensky. 1988. A human T cell-specific molecule is a member of a new gene family. J. Immunol. 141: 1018 –1025. 18. Schall, T. J., K. Bacon, K. J. Toy, and D. V. Goeddel. 1990. Selective attraction of monocytes and T lymphocytes of the memory phenotype by cytokine RANTES. Nature 347: 669 – 671. 19. Taub, D. D., S. M. Turcovski-Corrales, M. L. Key, D. L. Longo, and W. J. Murphy. 1996. Chemokines and T lymphocyte activation. I. ␤ chemokines costimulate human T lymphocyte activation in vitro. J. Immunol. 156: 2095–2103. 20. Kim, J. J., L. K. Nottingham, J. I. Sin, A. Tsai, L. Morrison, J. Oh, K. Dang, Y. Hu, K. Kazahaya, M. Bennett, et al. 1998. CD8 positive T cells influence

25.

26.

27. 28.

29.

30.

31.

32.

33.

34.

35.

36.

37. 38.

39.

40.

41.

42.

43. 44.

antigen-specific immune responses through the expression of chemokines. J. Clin. Invest. 102: 1112–1124. Taub, D. D., J. R. Ortaldo, S. M. Turcovski-Corrales, M. L. Key, D. L. Longo, and W. J. Murphy. 1996. ␤ chemokines costimulate lymphocyte cytolysis, proliferation, and lymphokine production. J. Leukocyte Biol. 59: 81– 89. Hadida, F., V. Vieillard, L. Mollet, I. Clark-Lewis, M. Baggiolini, and P. Debre. 1999. Cutting edge: RANTES regulates Fas ligand expression and killing by HIV-specific CD8 cytotoxic T cells. J. Immunol. 163: 1105–1109. Cocchi, F., A. L. DeVico, A. Garzino-Demo, S. K. Arya, R. C. Gallo, and P. Lusso. 1995. Identification of RANTES, MIP-1␣, and MIP-1␤ as the major HIV-suppressive factors produced by CD8⫹ T cells. Science 270: 1811–1815. Lima, M. F., Y. Zhang, and F. Villalta. 1997. ␤-Chemokines that inhibit HIV-1 infection of human macrophages stimulate uptake and promote destruction of Trypanosoma cruzi by human macrophages. Cell. Mol. Biol. 43: 1067–1076. Villalta, F., Y. Zhang, K. E. Bibb, J. C. Kappes, and M. F. Lima. 1998. The cysteine-cysteine family of chemokines RANTES, MIP-1␣, and MIP-1␤ induce trypanocidal activity in human macrophages via nitric oxide. Infect. Immun. 66: 4690 – 4695. Aliberti, J. C., F. S. Machado, J. T. Souto, A. P. Campanelli, M. M. Teixeira, R. T. Gazzinelli, and J. S. Silva. 1999. ␤-Chemokines enhance parasite uptake and promote nitric oxide-dependent microbiostatic activity in murine inflammatory macrophages infected with Trypanosoma cruzi. Infect. Immun. 67: 4819 – 4826. Tang, Y. Q., M. R. Yeaman, and M. E. Selsted. 2002. Antimicrobial peptides from human platelets. Infect. Immun. 70: 6524 – 6533. Saukkonen, J. J., B. Bazydlo, M. Thomas, R. M. Strieter, J. Keane, and H. Kornfeld. 2002. ␤-Chemokines are induced by Mycobacterium tuberculosis and inhibit its growth. Infect. Immun. 70: 1684 –1693. Engele, M., E. Stossel, K. Castiglione, N. Schwerdtner, M. Wagner, P. Bolcskei, M. Rollinghoff, and S. Stenger. 2002. Induction of TNF in human alveolar macrophages as a potential evasion mechanism of virulent Mycobacterium tuberculosis. J. Immunol. 168: 1328 –1337. Ernst, W. A., S. Thoma-Uszynski, R. Teitelbaum, C. Ko, D. A. Hanson, C. Clayberger, A. M. Krensky, M. Leippe, B. R. Bloom, T. Ganz, et al. 2000. Granulysin, a T cell product, kills bacteria by altering membrane permeability. J. Immunol. 165: 7102–7108. Beckman, E. M., A. Melian, S. M. Behar, P. A. Sieling, D. Chatterjee, S. T. Furlong, R. Matsumoto, J. P. Rosat, R. L. Modlin, and S. A. Porcelli. 1996. CD1c restricts responses of mycobacteria-specific T cells: evidence for antigen presentation by a second member of the human CD1 family. J. Immunol. 157: 2795–2803. Fortsch, D., M. Rollinghoff, and S. Stenger. 2000. IL-10 converts human dendritic cells into macrophage-like cells with increased antibacterial activity against virulent Mycobacterium tuberculosis. J. Immunol. 165: 978 –987. Neighbour, P. A., H. S. Huberman, and Y. Kress. 1982. Human large granular lymphocytes and natural killing ultrastructural studies of strontium-induced degranulation. Eur. J. Immunol. 12: 588 –594. Stenger, S., R. J. Mazzaccaro, K. Uyemura, S. Cho, P. F. Barnes, J. P. Rosat, A. Sette, M. B. Brenner, S. A. Porcelli, B. R. Bloom, et al. 1997. Differential effects of cytolytic T cell subsets on intracellular infection. Science 276: 1684 –1687. Serbina, N. V., C. C. Liu, C. A. Scanga, and J. L. Flynn. 2000. CD8⫹ CTL from lungs of Mycobacterium tuberculosis-infected mice express perforin in vivo and lyse infected macrophages. J. Immunol. 165: 353–363. Passmore, J. S., R. H. Glashoff, P. T. Lukey, and S. R. Ress. 2001. Granuledependent cytolysis of Mycobacterium tuberculosis-infected macrophages by human ␥␦⫹ T cells has no effect on intracellular mycobacterial viability. Clin. Exp. Immunol. 126: 76 – 83. Flynn, J. L., and J. Chan. 2001. Immunology of tuberculosis. Annu. Rev. Immunol. 19: 93–129. Sadek, M. I., E. Sada, Z. Toossi, S. K. Schwander, and E. A. Rich. 1998. Chemokines induced by infection of mononuclear phagocytes with mycobacteria and present in lung alveoli during active pulmonary tuberculosis. Am. J. Respir. Cell Mol. Biol. 19: 513–521. Fraziano, M., G. Cappelli, M. Santucci, F. Mariani, M. Amicosante, M. Casarini, S. Giosue, A. Bisetti, and V. Colizzi. 1999. Expression of CCR5 is increased in human monocyte-derived macrophages and alveolar macrophages in the course of in vivo and in vitro Mycobacterium tuberculosis infection. AIDS Res. Hum. Retroviruses 15: 869 – 874. Algood, H. M., and J. L. Flynn. 2004. CCR5-deficient mice control Mycobacterium tuberculosis infection despite increased pulmonary lymphocytic infiltration. J. Immunol. 173: 3287–3296. Jennes, W., S. Sawadogo, S. Koblavi-Deme, B. Vuylsteke, C. Maurice, T. H. Roels, T. Chorba, J. N. Nkengasong, and L. Kestens. 2002. Positive association between ␤-chemokine-producing T cells and HIV type 1 viral load in HIV-infected subjects in Abidjan, Cote d’Ivoire. AIDS Res. Hum. Retroviruses 18: 171–177. Jongstra, J., T. J. Schall, B. J. Dyer, C. Clayberger, J. Jorgensen, M. M. Davis, and A. M. Krensky. 1987. The isolation and sequence of a novel gene from a human functional T cell line. J. Exp. Med. 165: 601– 614. Song, A., T. Nikolcheva, and A. M. Krensky. 2000. Transcriptional regulation of RANTES expression in T lymphocytes. Immunol. Rev. 177: 236 –245. Song, A., Y. F. Chen, K. Thamatrakoln, T. A. Storm, and A. M. Krensky. 1999. RFLAT-1: a new zinc finger transcription factor that activates RANTES gene expression in T lymphocytes. Immunity 10: 93–103.

The Journal of Immunology 45. Kida, Y., K. Kuwano, Y. Zhang, and S. Arai. 2001. Acholeplasma laidlawii up-regulates granulysin gene expression via transcription factor activator protein-1 in a human monocytic cell line, THP-1. Immunology 104: 324 –332. 46. Catalfamo, M., T. Karpova, J. McNally, S. V. Costes, S. J. Lockett, E. Bos, P. J. Peters, and P. A. Henkart. 2004. Human CD8⫹ T cells store RANTES in a unique secretory compartment and release it rapidly after TcR stimulation. Immunity 20: 219 –230. 47. Durr, M., and A. Peschel. 2002. Chemokines meet defensins: the merging concepts of chemoattractants and antimicrobial peptides in host defense. Infect. Immun. 70: 6515– 6517. 48. Cole, A. M., T. Ganz, A. M. Liese, M. D. Burdick, L. Liu, and R. M. Strieter. 2001. Cutting edge: IFN-inducible ELR-CXC chemokines display defensin-like antimicrobial activity. J. Immunol. 167: 623– 627. 49. Carr, M. W., S. J. Roth, E. Luther, S. S. Rose, and T. A. Springer. 1994. Monocyte chemoattractant protein 1 acts as a T-lymphocyte chemoattractant. Proc. Natl. Acad. Sci. USA 91: 3652–3656.

7483 50. Territo, M. C., T. Ganz, M. E. Selsted, and R. Lehrer. 1989. Monocyte-chemotactic activity of defensins from human neutrophils. J. Clin. Invest. 84: 2017–2020. 51. Yang, D., O. Chertov, S. N. Bykovskaia, Q. Chen, M. J. Buffo, J. Shogan, M. Anderson, J. M. Schroder, J. M. Wang, O. M. Howard, et al. 1999. ␤-Defensins: linking innate and adaptive immunity through dendritic and T cell CCR6. Science 286: 525–528. 52. Yang, D., Q. Chen, O. Chertov, and J. J. Oppenheim. 2000. Human neutrophil defensins selectively chemoattract naive T and immature dendritic cells. J. Leukocyte Biol. 68: 9 –14. 53. Niyonsaba, F., K. Iwabuchi, A. Someya, M. Hirata, H. Matsuda, H. Ogawa, and I. Nagaoka. 2002. A cathelicidin family of human antibacterial peptide LL-37 induces mast cell chemotaxis. Immunology 106: 20 –26. 54. Deng, A., S. Chen, Q. Li, S. C. Lyu, C. Clayberger, and A. M. Krensky. 2005. Granulysin, a cytolytic molecule, is also a chemoattractant and proinflammatory activator. J. Immunol. 174: 5243–5248.