Processing and Release of IL-16 from CD41 But Not CD81 T Cells Is Activation Dependent1 David M. H. Wu, Yujun Zhang, Nereida A. Parada, Hardy Kornfeld, John Nicoll, David M. Center, and William W. Cruikshank2 IL-16 is synthesized as a precursor molecule of 68 kDa (pro-IL-16) that is processed by caspase-3, a member of the IL-1 converting enzyme (ICE) family. This cleavage results in a 13-kDa carboxy terminal peptide, which constitutes the bioactive secreted form of IL-16. We have previously reported constitutive IL-16 mRNA expression and pro-IL-16 protein in CD41 and CD81 T cells. Although bioactive IL-16 protein is present in unstimulated CD81 T cells, there is no bioactive IL-16 present in CD41 T cells. Along these lines, unstimulated CD81 T cells contain active caspase-3. In the current studies we investigated the regulation of IL-16 protein and mRNA expression in CD41 T cells and determined the kinetics of secretion following stimulation of the TCR. CD41 T cells release IL-16 protein following antigenic stimulation, and this release is accelerated in time by costimulation via CD28. However, CD3/CD28 costimulation did not alter IL-16 mRNA appearance or stability in either CD41 or CD81 T cells. The secretion of bioactive IL-16 from CD41 T cells correlated with the appearance of cleavage of pro-caspase-3 into its 20-kDa active form. Thus, resting CD81 T cells contain active caspase-3 that is capable of cleaving pro-IL-16, whereas CD41 T cells require activation for the appearance of active caspase-3. The mechanism of release or secretion of bioactive IL-16 is currently unknown, but does not correlate with cellular apoptosis. The Journal of Immunology, 1999, 162: 1287–1293. n CD81 T cells, pro-IL-16 is cleaved by caspase-3 (1, 2), resulting in the constitutive presence of stored bioactive 13kDa IL-16 (3– 6), as evidenced by the presence of IL-16specific bioactivity detected in cell lysates of unstimulated CD81 T cells (7). A noncytotoxic, noncytolytic release of IL-16 from CD81 T cells occurs within 4 h after stimulation with either histamine or serotonin (7, 8). The secretion of IL-16 from CD81 T cells following stimulation with vasoactive amines is not affected by inhibitors of protein synthesis. Furthermore, Northern blot analyses indicate that neither histamine nor serotonin induce changes in the level of IL-16 mRNA (7). In those studies that defined the role of CD81 T cells as a source of IL-16, we found that IL-16 mRNA was constitutively expressed not only in CD81 cells but in CD41 cells as well (7). However, in contrast to CD81 T cells, vasoactive amine stimulation of CD41 T cells did not result in IL-16 release; further, CD41 T cell lysates did not contain bioactive IL-16 protein (7). We have subsequently demonstrated that unstimulated CD41 T cells contain pro-IL-16 in quantities similar to that found in CD81 T cells (9). Although a number of cell types have recently been shown to produce bioactive IL-16, including epithelial cells (10), eosinophils (11), and mast cells (12), CD41 T lymphocytes have not been identified as a source of bioactive IL-16. The synthesis and secretion of IL-16 by CD41 T cells could
I
represent a positive amplification mechanism for further recruitment and activation of CD41 T cells. In the current studies, we investigated the synthesis and release of IL-16 by CD41 T lymphocytes after stimulation with specific Ag and anti-CD3 Ab and assessed the effect of costimulation of anti-CD3 with an activating anti-CD28 Ab. We observed that CD3 stimulation of peripheral blood CD41 T cells results in the secretion of bioactive IL-16 without detectable changes in the level of expression of IL-16 mRNA. Although costimulation through CD28 results in a more rapid secretion, with increased levels of bioactive IL-16 in cell supernatants, we could not detect any change in IL-16 mRNA levels, rate of transcription, or stability. Furthermore, we could not detect any changes in the level of total pro-IL-16 protein. Rather, we found that, unlike CD81 T cells, which contain active caspase-3 (1), CD41 T cells require activation to induce the cleavage of pro-caspase-3 into its 20kDa enzymatically active form. There is a temporal relationship between the appearance of active caspase-3 and bioactive IL-16 in antiCD3-stimulated CD41 T cells; in a similar fashion, CD28 costimulation augments the coincident appearance of both active caspase-3 and bioactive IL-16. These studies indicate that CD41 and CD81 T cells have distinct mechanisms and kinetics of caspase-dependent processing of IL-16. Whereas CD81 T cells do not require antigenic activation for the release of IL-16, the cleavage of pro-IL-16 and the release of bioactive IL-16 by CD41 T cells do require antigenic stimulation.
Pulmonary Center, Boston University School of Medicine, Boston, MA 02118
Materials and Methods
Received for publication July 16, 1998. Accepted for publication October 9, 1998. 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 Grants AI35680, AI37368, and HL32802 from the National Institutes of Health. W.W.C. is a recipient of an American Lung Association Career Investigator Award. 2
Address correspondence and reprint requests to Dr. William Cruikshank, Pulmonary Center, Housman R-304, Boston University School of Medicine, Boston, MA 02118. E-mail address:
[email protected] 3 Abbreviations used in this paper: CHX, cycloheximide; NWNA-T, nylon wool nonadherent human T cell; MIP, macrophage inflammatory protein.
Copyright © 1999 by The American Association of Immunologists
Reagents Cycloheximide (CHX)3 was obtained from Sigma (St. Louis, MO). Actinomycin D was obtained from Life Technologies (Grand Island, NY). Caspase-1 and -3 inhibitors were obtained from Bachem (King of Prussia, PA).
Antibodies An affinity-purified polyclonal rabbit anti-rIL-16 was prepared from rIL16-immunized rabbit sera as described previously (3, 6). Mouse monoclonal anti-human CD4 Abs, mouse anti-human CD8 Abs, FITC-conjugated anti-CD4 or anti CD8 mAbs, and phycoerythrin-conjugated anti-CD3 were 0022-1767/99/$02.00
1288
DIFFERENTIAL RELEASE OF IL-16 FROM CD41 AND CD81 T CELLS
purchased from Bio-Source International (Camarillo, CA). Monoclonal antihuman CD3 Ab and mouse anti-human CD28 Ab were obtained from PharMingen (San Diego, CA).
Cell preparation PBMCs were isolated by Ficoll-Paque (Pharmacia, Piscataway, NJ) density centrifugation of freshly drawn venous blood of healthy human volunteers. After centrifugation, cells were washed three times and resuspended at a concentration of 3.5 3 106/ml in culture medium 199 (Life Technologies) supplemented with 100 U/ml penicillin, 100 mg/ml streptomycin, 22 mM HEPES buffer, and 0.4% BSA. Nylon wool nonadherent human T cells (NWNA-Ts) were isolated by a modification (4) of the method of Julius et al. (13). For subset isolation, CD4- or CD8-depleted cell populations were prepared by magnetic bead negative selection (Biomag/Advanced Magnetics, Cambridge, MA) conjugated with goat anti-mouse IgG. NWNA-T lymphocytes were first labeled with mouse anti-human CD4 or mouse antihuman CD8 mAbs by incubating 1 3 106 cells with 5.0 mg of the appropriate Abs for 30 min at 4°C. Magnetic beads were washed with cold PBS (4°C) three times and mixed with the labeled lymphocytes for 30 min at 4°C in the dark. Nonadherent cells were collected in the supernatant. Immunostaining followed by FACScan analysis by flow cytometry (Becton Dickinson, FACS Division, Mountain View, CA) showed that CD4-depleted cell populations contained ,1% of CD41/CD31 lymphocytes and that CD8-depleted cell populations contained ,1% of CD81/CD31 cells.
Cell cultures Human PBMCs, CD41 T cells, or CD81 T cells were cultured for varying time intervals at 37°C in U-bottom microculture plates (Costar, Cambridge, MA) or P-60 culture plates in a humidified atmosphere of 5% CO2 at concentrations of 3.0 – 4.0 3 106/ml in the presence or absence of anti-CD3 (1–2.5 mg/ml). Costimulation studies involved the addition of soluble antiCD28 Ab (1 mg/ml). In certain experiments, CD42 or CD82 subsets were pretreated with a protein synthesis inhibitor (CHX, 20 mg/ml for 2 h) and washed twice before being added to plates containing anti-CD3 Ab. After incubation, the cell-free supernatants were harvested and assessed for IL-16 protein and bioactivity. In the experiments using caspase inhibitors, the inhibitors were added to the CD41 T cells at the same time that the cells were added to the culture wells containing immobilized anti-CD3 and antiCD28 Abs. CD41 T cells were incubated in the presence or absence of 100 mM of the tetrapeptide inhibitors Ac-DEVD-CHO or Ac-YVAD-CHO at 37°C in 5% CO2 for #24 h. The culture medium was collected at 6 and 24 h and assessed for IL-16 bioactivity by chemotaxis.
Chemotaxis assay The presence of IL-16 in cell-free supernatants was assessed by quantification of human NWNA-T lymphocyte migration using a modified Boyden chemotaxis chamber technique (4). A total of 52 ml of NWNA-T suspension (10 3 106 cells/ml) was placed in the upper compartments of the 48-well microchemotaxis chambers separated from 32 ml of test samples by 8-mm micropore nitrocellulose filters (Neuroprobe, Cabin John, MD) and was incubated at 37°C in a 5% CO2 atmosphere for 2 h. Test samples included cell-free supernatants diluted at various concentrations. The filters were fixed with hematoxylin, dehydrated, and mounted on glass slides. Cell migration was quantified by counting the total number of cells migrating beyond 50 mm into the filter by light microscopy, with baseline migration under control buffer conditions of 10 –15 cells. Six high-power fields in duplicate for each sample and the means (6SD) were calculated and expressed as percent values of control migration (100%). For each set of experiments, four to five individual experiments were performed. A paired Student’s t test was used for statistical analysis to identify differences between experimental and control conditions using the mean values for lymphocyte migration. Significance was established at the p , 0.05 level of confidence. To assess specificity for IL-16, neutralizing experiments were conducted by incubating test samples for 15 min with neutralizing concentrations of affinity-purified rabbit anti-human rIL-16 (rhuIL-16) antisera (5–10 mg/ml). At this concentration, the optimal chemotactic activity of 50 ng/ml rhuIl-16 is completely neutralized. There is no cross-neutralization by this Ab with the chemotactic activity induced by any of the lymphocyte chemoattractant chemokines tested thus far (14 –16). Preimmune rabbit serum did not recognize IL-16 by Western blot analysis (9) and had no effect on IL-16-induced, chemokine-induced, or random migration of human T cells (data not shown).
RNA extraction and Northern hybridization NWNA-Ts or PBMCs were stimulated with plate-bound monoclonal antiCD3 (1–2.5 mg/ml) for 1, 2, 4, 6, or 24 h. Total cellular RNA was obtained from NWNA-T cell suspensions (20 –30 3 106 cells) using the standard TriReagent (Molecular Research Center, Cincinnati, OH) extraction method according to the manufacturer’s recommendations. A total of 10 –15 mg of total cellular RNA was fractionated by 1% agarose/17% formaldehyde gel electrophoresis and transferred to nylon membranes (Hybond N1, Amersham, Arlington Heights, IL) by capillary blotting. Shortwave UV light was used to fix the RNA on the membranes. The nylon membranes were treated for 20 min with prehybridization solution (Quick Hyb, Stratagene, La Jolla, CA) at 68°C and subsequently hybridized with 32 P-labeled IL-16 cDNA probe with 10 mg/ml herring sperm DNA for 1 h at 68°C. After hybridization, blots were washed twice at low stringency conditions of 23 SSC (300 nM NaCl, 30 mM sodium citrate, 0.5% sodium pyrophosphate) and 0.1% SDS at 25°C followed by wash at high stringency conditions of 0.13 SSC/0.1% SDS at 60°C and were subjected to autoradiography. Equal loading of RNA was assessed by ethidium bromide staining and by rehybridizing blots with the b-actin cDNA probe. IL-16 mRNA stability was determined in PBMCs in the presence or absence of anti-CD3 (2.5 mg/ml) and anti-CD28 Ab (1 mg/ml) for costimulation studies. New transcription was inhibited by the addition of actinomycin D (10 mg/ml) at 24 h poststimulation. Relative steady-state mRNA levels for IL-16 and b-actin were measured at 30 min and at 1, 2, and 3 h thereafter.
Western blot analysis Purified CD4 or CD8 cells were prepared by negative selection as described above. Cells under control, anti-CD3, or CD28 costimulatory conditions were washed once with PBS, released by scraping, and harvested by centrifugation. Cells were immediately lysed by sonication in a buffer containing PBS (pH 7.5), 1 mM PMSF, 10 mg/ml aprotonin, and 10 mg/ml leupeptin and subjected to electrophoresis through a 15% SDS polyacrylamide gel. Proteins were electrophoretically transferred to a nitrocellulose membrane and probed with goat-polyclonal anti-caspase-3 against the P20 subunit (Santa Cruz Biotechnology, Santa Cruz, CA). The secondary Ab, which was an anti-goat Ig labeled with horseradish peroxidase (Santa Cruz Biotechnology), was used at a dilution of 1/5000, and the signal was detected by chemiluminescence (Pierce, Rockford, IL).
Staining for apoptosis T cells harvested from the cultures were washed twice in PBS before staining with propidium iodide (1 mg/ml final concentration) for 5 min at room temperature. Apoptosis was determined using fluorescence microscopy and was quantitated by counting the percentage of stained nuclei.
Results Effects of anti-CD3 Ab and Ag stimulation on generation of IL-16 by CD41and CD81 T cells The current studies were undertaken to determine under what circumstances CD41 T cells could process pro-IL-16 and secrete bioactive IL-16. First, we investigated the effects of immobilized anti-CD3 Ab and Ag stimulation on the secretion of IL-16 in human T lymphocytes incubated as a mixed T cell population. Monocytes represented 5% of the total cell numbers, which was sufficient to facilitate T cell activation. The cells were stimulated with either anti-CD3 Ab for #24 h. Cells isolated from normal individuals were stimulated with anti-CD3 Ab (anti-e, 1 mg/ml). Because pro-IL-16 is recognized by current ELISAs but has no chemoattractant activity (1, 9), we identified the presence of bioactive IL-16 by chemotaxis of human T cells in the presence or absence of IL-16 neutralizing Ab. As shown in Fig. 1, anti-CD3 induced a significant increase in the overall T cell chemoattractant activity detected in the supernatants of mixed T cell cultures by 6 h, which increased through 12, 18, and 24 h. The contribution of IL-16 to the overall chemoattractant activity was determined by coincubating the supernatants with neutralizing anti-IL-16 Ab for 30 min before assessing cell motility. The IL-16 neutralizing Ab studies indicate that IL-16 contributed the majority of the chemoattractant
The Journal of Immunology
FIGURE 1. TCR stimulation via anti-CD3 Ab induces IL-16 secretion in mixed T cells. NWNA-T lymphocytes were stimulated with plate-bound anti-CD3 for 24 h. At 4, 6, 12, 18, and 24 h, the lymphocyte chemoattractant activity of cell-free supernatants was assessed in a Boyden chamber assay in the absence (hatched bars) and presence (filled bars) of neutralizing anti-IL-16 Ab. Chemotactic data are expressed as the mean 6 SD for three different experiments. An asterisk (p) denotes supernatants whose chemotactic activity was significantly decreased in the presence of neutralizing anti-IL-16 Ab (p , 0.05).
activity that was present at the 6-h timepoint for anti-CD3 stimulation. IL-16 contributed approximately one-half of the total chemoattractant activity for anti-CD3 (Fig. 1). Although the total detectable chemoattractant activity increased from 6 to 24 h, the greatest contribution of IL-16, calculated as a percentage, was seen in the earlier timepoints. To determine the time course and subset specificity for IL-16 generation following CD3 stimulation, mixed T lymphocytes were then fractionated into CD41 or CD81 T cells by negative selection. Subset purity was determined to be 96% CD41 cells and 95% CD81 cells, with both populations containing 4 –5% monocytes as indicated by FACS analysis (data not shown). To determine the kinetics of IL-16 release, CD41 or CD81 cell cultures were stimulated with 1 mg/ml immobilized anti-CD3 Ab. After 6 h and then again after 24 h, culture supernatants were removed and cells were recultured on anti-CD3-coated plates in fresh media. As shown in Fig. 2, anti-CD3 stimulated the release of chemoattractant activity at the 6-h timepoint from both cell populations. Neutralizing antiIL-16 Ab had no effect on total chemoattraction from CD41 cell supernatants, indicating a lack of IL-16 contribution. Stimulating CD41 cells from 6 to 24 h did result in increased chemoattractant activity, which was partially (35%) inhibited by anti-IL-16 Ab (Fig. 2). Culturing these cells from 24 to 48 h in fresh media resulted in a further increase in overall activity, with the contribution of IL-16 increasing to ;43%. In contrast, the stimulation of CD81 T cells for the initial 6 h resulted in supernatants with a chemoattractant activity of 187% 6 8% of control cell migration, with all but an insignificant 8% attributed to IL-16 (Fig. 2). When the cells were washed and recultured from 6 to 24 h, the overall chemoattractant activity remained approximately the same; however, the contribution of IL-16 decreased to 40%. The same trend continued for the 24 – 48 h cultures, with the contribution of IL-16 decreasing to 15% (Fig. 2). These kinetic studies indicate that CD81 cells are the primary source of IL-16 immediately following stimulation, whereas the CD41 T cells become the primary source at 24 – 48 h after stimulation. The different kinetics of IL-16 release by these two T cell subpopulations also argue against the possibility that the IL-16 bioactivity measured in the CD41-enriched T cell supernatant was due to contaminating CD81 T cells.
1289
FIGURE 2. T cell subset specificity of anti-CD3-induced IL-16 secretion. NWNA-Ts were isolated into CD41 and CD81 subsets and stimulated with plate-bound anti-CD3 Ab. Supernatants were harvested and replaced with fresh media at 6, 24, and 48 h. Total lymphocyte chemoattractant activity was assessed in the supernatants of CD81 or CD41 T cells in the absence (hatched bars) or presence (filled bars) of neutralizing anti-IL-16 Ab. Migration data are expressed as the mean 6 SD for three different experiments. An asterisk (p) denotes supernatants whose chemotactic activity was significantly decreased in the presence of neutralizing anti-IL-16 Ab (p , 0.05).
Effects of CHX on IL-16 release from CD41 and CD81 T cells To determine whether the release of IL-16 was dependent upon the generation of de novo protein, T cell subsets were treated with the protein synthesis inhibitor CHX (20 mg/ml; sufficient to inhibit all [35S]methionine incorporation (data not shown)) for 2 h before stimulation with anti-CD3. At 6 h after stimulation, there was no significant chemoattractant activity in the supernatants of CD41 T cells treated with CHX, indicating that either the synthesis or secretion of all lymphocyte chemoattractant activities required new protein synthesis (Fig. 3). Similarly, at 24 h, all of the anti-CD3 induced chemoattractant activity was inhibited by CHX treatment. These findings are consistent with our previous reports of a lack of preformed chemoattractant activity contained in CD41 T cell lysates. In contrast to CD41 cells, treatment with CHX did not reduce the chemoattractant activity in CD81 cell supernatants at 6 h (Fig. 3). The addition of anti-IL-16 Ab to the chemotaxis assay resulted in the loss of all chemoattractant activity. All of the chemoattractant activity associated with either unstimulated CD81 cells or cells stimulated with anti-CD3 for #6 h is attributable to IL-16, which is consistent with the presence of preformed, bioactive, IL-16. CHX pretreatment did reduce the chemoattractant activity released by stimulated CD81 T cells at 24 h by ;50%; however, all of the CHX-insensitive chemoattractant activity was inhibited by anti-IL-16 Abs. The stimulation of CD81 T cells induces the release of preformed IL-16 by 6 and 24 h, whereas the additional chemoattractant activity produced by CD81 T cells over 24 h is not IL-16 and requires new protein synthesis. Chemokines are a likely source (14, 15) for the residual chemoattractant activity. In that regard, we identified RANTES and macrophage inflammatory protein (MIP)-1b as the other two chemoattractant factors by neutralizing Ab studies. No MIP-1a bioactivity was detected. A combination of anti-RANTES, anti-IL-16, and anti-MIP-1b Abs completely neutralized all chemoattractant activity (data not shown).
1290
DIFFERENTIAL RELEASE OF IL-16 FROM CD41 AND CD81 T CELLS
FIGURE 3. The effects of CHX on anti-CD3-induced IL-16 secretion from CD41 and CD81 T cells. NWNA-Ts were fractionated into CD41 and CD81 subsets and stimulated with plate-bound anti-CD3 for 6 and 24 h with and without prior incubation with CHX, as described in Materials and Methods. The T cell supernatants were then assessed in the absence or presence of neutralizing anti-IL-16 Ab as designated in the figure. Migration data are expressed as the mean 6 SD for three different experiments. An asterisk (p) denotes supernatants whose chemotactic activity was significantly decreased in the presence of neutralizing anti-IL-16 Ab (p , 0.05).
Costimulation with CD28 increases IL-16 protein production To assess the effect of CD28 costimulation (16 –18) on IL-16 expression, we assayed supernatants from CD41 and CD81 subsets stimulated with anti-CD3, anti-CD28, or the combination of antiCD3 and anti-CD28 for 6 and 24 h. The major difference noted with CD28 costimulation was the appearance of chemoattractant activity attributable to IL-16 in CD41 T cell supernatants at 6 h (Fig. 4A). The majority of the chemoattractant activity seen at this timepoint was blocked with CHX treatment (Fig. 4A). There was only a slight increase in the proportion of IL-16 bioactivity induced by costimulation at the 24 h timepoint (Fig. 4B). As noted above for anti-CD3 stimulation, CHX treatment blocked the synthesis and/or release of almost all chemoattractant activity in costimulated cells at both timepoints. The same experiment was performed with purified CD81 T cells. In these studies, we did not find any significant increase in overall chemoattractant activity or in the proportion of IL-16 chemoattractant activity at 6 h after CD28 costimulation when compared with anti-CD3 alone (Fig. 4C). Again, treatment with CHX had no effect on the appearance of chemoattractant activity secreted over 6 h. However, CHX did reduce the overall chemoattractant activity at 24 h for both antiCD3 alone and costimulated supernatants (Fig. 4D). All of the CHX-insensitive chemoattractant activity appeared to be attributable to IL-16, as the addition of anti-IL-16 Ab blocked all residual activity. Regulation of IL-16 mRNA expression To directly investigate whether anti-CD3 stimulation could modulate IL-16 mRNA, human CD41 T cells were stimulated with immobilized anti-CD3 for 6 or 24 h before isolation of total RNA and Northern blot analysis. IL-16 mRNA was constitutively expressed and did not appear to increase after 6 or 24 h of anti-CD3 stimulation (Fig. 5). Furthermore, no change in mRNA following anti-CD3 activation could be identified in CD81 subsets (data not
FIGURE 4. Effect of CD28 costimulation on IL-16 secretion from CD41 and CD81 T cells. NWNA-Ts were isolated into CD41 and CD81 subsets and stimulated with plate-bound anti-CD3 Ab for 6 and 24 h with or without anti-CD28 costimulation and with or without prior incubation with CHX. Total lymphocyte chemoattractant activity was assessed in cellfree supernatants either in the absence (hatched bars) or presence (filled bars) of neutralizing anti-IL-16 Ab. The chemotactic activities generated by CD41 cells are shown in A (6 h of stimulation) and B (24 h of stimulation), whereas the activities generated by CD81 T cells are shown in C (6 h of stimulation) and D (24 h of stimulation). Migration data are expressed as the mean 6 SD for three different experiments. An asterisk (p) denotes supernatants whose chemotactic activity was significantly greater than that observed for anti-CD3 alone (p , 0.05). A pound sign (#) denotes migration, which was significantly decreased in the presence of neutralizing anti-IL-16 Ab (p , 0.05).
shown). We subsequently examined the effect of costimulation on IL-16 mRNA expression in CD41 T cells. CD3/CD28 costimulation had no significant effect on the levels of IL-16 mRNA at any
The Journal of Immunology
FIGURE 5. Effect of anti-CD3 stimulation on IL-16 mRNA in CD41 T cells. Northern hybridization analysis of total RNA extracted from antiCD3-stimulated CD41 T cells for 0 (lane 1), 6 (lane 2), or 24 (lane 3) h was performed using 32P-labeled cDNA probe specific for IL-16. RNA loading was assessed by ethidium bromide staining, and band intensities were normalized to b-actin mRNA levels. The positions of the 18S and 28S ribosomal RNAs are indicated to the right of the figure.
timepoint up to 6 h of stimulation (Fig. 6). Finally, there was no detectable effect of CD3/CD28 costimulation on the stability of IL-16 mRNA (Fig. 7). In these experiments, CD41 T cells were incubated in the presence or absence of anti-CD3/anti-CD28 for 24 h, followed by the addition of actinomycin D (10 mg/ml). In both conditions, IL-16 mRNA was detectable at 1 h, decreased at 2 h, and was not detectable by 3 h, indicating no change in mRNA stability. Taken together, these experiments suggest that IL-16 secretion in CD4 cells must be regulated at some critical step after transcription and translation. We hypothesized that this critical step might lie with the pro-IL-16-processing enzyme, caspase-3. Detection of caspase-3 in activated CD41 lymphocyte lysates We first investigated whether stimulated CD41 T lymphocytes contain active caspase-3 and then determined the time course of processing of pro-caspase-3 following anti-CD3 or anti-CD3/antiCD28 costimulation. In these experiments, purified CD41 lymphocytes were stimulated for 6 and 24 h with anti-CD3 or antiCD3/CD28 costimulation. Cell lysates at these two timepoints were subjected to SDS-PAGE followed by caspase-3 Western blot analysis. As seen in Fig. 8, no active caspase-3 is observed in unstimulated CD41 T cells. At 4 h, a timepoint when bioactive IL-16 is not observed in supernatants from anti-CD3-activated CD41 T cells, there is only a faint (20-kDa) detectable band correlating to active caspase-3. However, at this timepoint, a 20-kDa band of 10-fold the intensity (lane 3) compared with anti-CD3 stimulated alone (lane 2) is observed with anti-CD3/CD28 co-
FIGURE 6. Effect of anti-CD28 costimulation on IL-16 mRNA. Northern hybridization analysis of total RNA extracted from anti-CD3- and antiCD28-stimulated PBMCs for 0, 1, 2, 4, and 6 h was performed using a 32 P-labeled cDNA probe specific for IL-16. Equal RNA loading was assessed by hybridization with a b-actin probe. The positions of the 18S and 28S ribosomal RNAs are indicated to the left of the figure.
1291
FIGURE 7. Effect of anti-CD28/anti-CD3 costimulation on IL-16 mRNA stability. CD41 T cells were incubated in the presence (lanes 1– 4) or absence (lanes 5– 8) of anti-CD28 Ab (1 mg/ml) and anti-CD3 Ab (2.5 mg/ml) for #4 h. Total RNA was extracted at time 0 (lanes 1 and 5), 1 h (lanes 2 and 6), 2 h (lanes 3 and 7), and 3 h (lanes 4 and 8) after the addition of actinomycin D (10 mg/ml). IL-16 mRNA was analyzed by Northern hybridization using a 32P-labeled cDNA probe specific for IL-16. The amounts of RNA loaded in each lane were compared by hybridization with a b-actin probe and ethidium bromide staining. The positions of the 18S and 28S ribosomal RNAs are indicated to the left of the figure.
stimulation. By 24 h, cells stimulated with anti-CD3 alone or costimulated with CD3/CD28 all express similar levels of active caspase-3 (Fig. 8). These data suggest that caspase-3 is activated after the stimulation of CD41 T cells with anti-CD3 or anti-CD3/ CD28 in a time course compatible with the appearance of the bioactive IL-16 detected in the cell supernatants. In an attempt to corroborate the requirement of caspase activation for the release of bioactive IL-16, CD41 T cells were incubated in the presence of peptides Ac-DEVD-CHO and Ac-YVADaldehyde, which are inhibitors of caspase-3 and caspase-1 activation, respectively (19 –21). CD41 T cells were exposed to the inhibitors for 1 h before stimulation with anti-CD3 alone or in combination with anti-CD28. After 24 h, supernatants were assessed for chemoattractant activity attributable to IL-16; cells were assessed for apoptosis by propidium iodide staining to demonstrate that the peptides were effective in an independent assay. Cells treated with Ac-DEVD-CHO were unable to generate detectable bioactive IL-16 in the supernatants after anti-CD3/anti-CD28 stimulation (Fig. 9). Similar effects were observed for stimulation by anti-CD3 alone (data not shown). The involvement of caspase-3 appears to be specific, as the inhibition of caspase-1 by the peptide Ac-YVAD-aldehyde did not alter the magnitude or the time course
FIGURE 8. Time course for the activation of caspase-3 in CD3- and CD28-costimulated CD41 T cells. CD41 T cells were stimulated for 4 or 16 h, at which time total cellular protein was subjected to SDS-PAGE followed by Western blot analysis for the active subunit of caspase-3 (P20). Cells were either left unstimulated (lane 1), stimulated with anti-CD3 for 4 (lane 2) or 16 (lane 4) h, or stimulated with anti-CD3 and anti-CD28 for 4 (lane 3) or 16 (lane 5) h. The presence of caspase-3 was detected using an Ab against its P20 subunit followed by a display of chemiluminescence.
1292
DIFFERENTIAL RELEASE OF IL-16 FROM CD41 AND CD81 T CELLS
FIGURE 9. Effects of caspase inhibitors on anti-CD3/anti-CD28induced IL-16 production in CD41 T cells. CD41 T cells were cultured in the absence or presence of 100 mM of the tetrapeptides DEVD-CHO or YVAD-CHO during stimulation with anti-CD3/anti-CD28 for 6 and 24 h. At each timepoint, cell-free supernatants were assessed for chemoattractant activity in the absence or presence of neutralizing anti-IL-16 Ab. Data have been normalized to control cell migration designated as 100% and represent the mean 6 SD from three independent experiments. The asterisk (p) denotes significantly different migration from that induced by anti-CD3/ anti-CD28 (p , 0.05).
required for expression of bioactive IL-16. It is interesting to note that treatment with the caspase-3 inhibitor did not appear to affect non-IL-16-induced migration for either CD3 alone or for CD3/ CD28-costimulated cells, suggesting that IL-16 is the only detectable chemoattractant processed by caspase-3. To confirm the bioactivity of the caspase-1 and -3 inhibitors, their effect on CD3stimulated apoptosis was assessed at the 24-h timepoint by propidium iodide staining. Based on apoptotic staining, cells treated with the inhibitors demonstrated 23% and 18% apoptotic staining for caspase-1 and -3 inhibitors, respectively, compared with 57% apoptotic staining for untreated cells. This represents a 60.7% and 69.5% decrease in apoptotic cells for the caspase-1 and -3 inhibitors, respectively, compared with untreated cells. Therefore, the data indicate that the caspase-1 inhibitor was functionally active in reducing apoptosis but had no effect on IL-16 release. These findings suggest that the release of bioactive IL-16 is coincident with and dependent upon activation of the enzyme that facilitates its cleavage, caspase-3, and not on transcriptional regulation of the IL-16 mRNA.
Discussion We have shown previously that mitogen-activated T cells secrete IL-16 (3, 4), and that CD81 T cells are a major source of this cytokine after 24 h of stimulation (3, 4, 7, 8). Interestingly, there is a pool of preformed, processed bioactive IL-16 in CD81 T cells that is released within 4 – 6 h after stimulation with histamine and within 2 h after stimulation with serotonin (7, 8). Although CD41 T cells contain constitutive IL-16 mRNA and pro-IL-16 protein (9), they contain no stored bioactive IL-16, nor do they release any IL-16 following stimulation with histamine or serotonin (7, 8). The present study demonstrates for the first time that CD41 T cells are a source of bioactive IL-16 that is secreted following stimulation via the TCR by either specific Ag or anti-CD3 Ab. Although CD81 T cells secrete IL-16 after anti-CD3 stimulation within 6 h, most of the released IL-16 appears to derive from the preformed bioactive pool, as demonstrated by studies with inhibitors of pro-
tein synthesis. However, in CD41 T cells, the kinetics of secretion of bioactive IL-16 differ from those observed in CD81 T cells. CD41 T cells require 24 h to release bioactive IL-16 following anti-CD3 stimulation. This process can be accelerated by costimulation with anti-CD3/anti-CD28, which results in the release of bioactive IL-16 by 6 h. CD3 activation of CD41 T cells does not result in increases in IL-16 mRNA or in changes in mRNA stability. Rather, we observed that anti-CD3 activation of CD41 T cells results in the activation of caspase-3 from its proenzyme state, presumably permitting the processing of constitutively expressed pro-IL-16. CD3/CD28 costimulation resulted in a more rapid activation of caspase-3, which correlated with an accelerated secretion of bioactive IL-16. In contrast to the CD81 T cell, CHX inhibits the appearance of bioactive IL-16 in the supernatants of CD41 T cells costimulated with anti-CD3/anti-CD28. This finding indicates that de novo protein synthesis is required for the secretion of IL-16. Studies designed to identify the required protein(s) are currently ongoing. Dependence of caspase-3 activation for the processing and release of IL-16 does not appear to be shared by any other detectable chemoattractant, as only IL-16-induced migration was affected. Of interest, in contrast to CD81 T cells, IL-16 does not represent a majority of the lymphocyte chemoattractant activity identified in CD41 T cell supernatants following CD3 or CD3/CD28 stimulation. However, the contribution of IL-16 increases proportionally to ;40% of the overall chemoattractant activity during the first 48 h poststimulation. All other chemoattractant activities did not appear following CHX treatment, suggesting new protein synthesis was required. In that regard, members of the CC chemokine family are likely candidates for this activity. The requirement of caspase activation for protein processing and secretion has been reported previously for both IL-1b (20) and IL-18 (21). Both of these ILs require activation of caspase-1. The efficiency for IL-1a and -1b processing and release is increased with induced apoptosis but not with necrosis (22). The mechanism for the release of IL-16 has not been determined; however, because IL-16 also lacks a signal peptide, there may be some similarities with IL-1a and -1b. Thus far, our data suggest that apoptosis is not required for the processing and release of IL-16 as the prevention of apoptosis with a caspase-1 inhibitor also did not effect the magnitude or the time course for the release of IL-16. However, we have noted that induction of apoptosis with staurosporin does result in an enhanced release of bioactive IL-16 (Y.Z., unpublished observations). These findings indicate that the processing and release of IL-16 by CD41 T cells is not contingent upon subsequent cell apoptosis, and further supports the concept that, although activation of caspase-3 enzymatic activity is essential for an apoptotic signal (23, 24), activation of caspase-3 alone is not sufficient for the induction of apoptosis (25). Based on the kinetics data, one could hypothesize that CD8 cells would represent a rapid release mechanism initiated by the mediators of early inflammation, such as vasoactive amines, and would contribute to the recruitment of CD41 T cells by the secretion of preformed IL-16. The ability of CD41 T cells to synthesize and secrete IL-16 may serve to function as a positive feedback mechanism for further cell recruitment and activation. A similar situation may occur with another proinflammatory cell, the eosinophil, which has been shown to generate and release IL-16 protein after activation (11). In summary, we have shown that CD41 T cells are a source of IL-16 following TCR activation or CD28 costimulatory conditions. The active form of caspase-3, an important mechanism for further processing of pro-IL-16 and the secretion of bioactive IL16, can be identified in activated CD41 T cells in a time course
The Journal of Immunology that is similar to that seen for IL-16 release. However, an apparent separation does exist between the release of IL-16 following caspase-3 activation, pro-IL-16 processing, and the induction of apoptosis. In addition, our data demonstrate that the mechanisms and requirements for the synthesis and release of IL-16 from CD41 and CD81 T cells are entirely different. Taken together, these studies suggest that Ag-activated CD41 T cells could be a source of IL-16, which might amplify the accumulation of unsensitized CD41 T cells at sites of Ag-induced inflammation.
References 1. Zhang, Y., D. M. Center, D. M. Wu, W. W. Cruikshank, J. Juan, D. W. Andrews, and H. Kornfeld. 1998. Processing and activation of IL-16 by caspase-3. J. Biol. Chem. 273:1144. 2. Baier, M., N. Bannert, A. Werner, K. Lang, and R. Kurth. 1997. Molecular cloning, sequence, expression, and processing of the interleukin 16 precursor. Proc. Natl. Acad. Sci. USA 94:5273. 3. Cruikshank, W. W., D. M. Center, N. Nisar, B. Natke, A. C. Theodore, and H. Kornfeld. 1994. Molecular and functional analysis of a lymphocyte chemoattractant factor: association of biologic activity with CD4 expression. Proc. Natl. Acad. Sci. USA 91:5109. 4. Center, D. M., and W. W. Cruikshank. 1982. Modulation of lymphocyte migration by human lymphokines: Identification and characterization of chemoattractant activity for lymphocytes from mitogen-stimulated mononuclear cells. J. Immunol. 128:2563. 5. Cruikshank, W. W., and D. M. Center. 1982. Modulation of lymphocyte migration by human lymphokines: purification of lymphotactic factor (LCF). J. Immunol. 128:2569. 6. Cruikshank, W. W., J. L. Greenstein, A. C. Theodore, and D. M. Center. 1987. Lymphocyte chemoattractant factor induces CD4-dependent intracytoplasmic signaling in lymphocytes. J. Immunol. 146:2928. 7. Laberge, S., W. W. Cruikshank, H. Kornfeld, and D. M. Center. 1995. Histamineinduced secretion of lymphocyte chemoattractant factor from CD81 T cells is independent of transcription and translation: evidence for constitutive protein synthesis and storage. J. Immunol. 155:2902. 8. Laberge, S., W. W. Cruikshank, D. J. Beer, and D. M. Center. 1996. Secretion of IL-16 (lymphocyte chemoattractant factor) from serotonin-stimulated CD81 T cells in vitro. J. Immunol. 156:310. 9. Chupp, G. L., E. A. Wright, D. Wu, M. Vallen-Mashikian, W. W. Cruikshank, D. M. Center, H. Kornfeld, and J. S. Berman. 1998. The tissue and T cell distribution of precursor and mature IL-16. J. Immunol. 161:3114. 10. Laberge, S., P. Ernst, O. Ghaffar, W. W. Cruikshank, D. M. Center, and Q. Hamid. Increased expression of interleukin-16 in bronchial mucosa of subjects with atopic asthma. 1997. Am. J. Respir. Cell Mol. Biol. 17:193.
1293 11. Lim, K. G., H. C. Wan, P. T. Bozza, M. B. Resnick, D. T. Wong, W. W. Cruikshank, H. Kornfeld, D. M. Center, and P. F. Weller. 1996. Human eosinophils elaborate the lymphocyte chemoattractants IL-16 (lymphocyte chemoattractant factor) and RANTES. J. Immunol. 156:2566. 12. Rumsaeng, V., W. W. Cruikshank, B. Foster, C. Prussin, A. S. Kirshenbaum, T. A. Davis, H. Kornfeld, D. M. Center, and D. D. Metcalf. 1997. Human mast cells produce the CD41 T lymphocyte chemoattractant factor, IL-16. J. Immunol. 159:2904. 13. Julius, M. H., E. Simpson, and L. A. Herzenberg. 1973. A rapid method for the isolation of functional thymus-derived murine lymphocytes. Eur J. Immunol. 3:645. 14. Murphy, W. J., Z.-G. Tian, O. Asai, S. Funakoshi, P. Rotter, M. Henry, R. M. Strieter, S. L. Kunkel, D. L. Longo, and D. D. Taub. 1996. Chemokines and T lymphocyte activation: facilitation of human T cell trafficking in SCID mice. J. Immunol. 156:2104. 15. Baggioloni, M. 1998. Chemokines and leukocyte traffic. Nature 392:565. 16. Fraser, J. D., D. Straus, and A. Weiss. 1993. Signal transduction events leading to T-cell lymphokine gene expression. Immunol. Today 14:357. 17. Tagaya, Y., R. N. Bamford, A. P. DeFillipis, and T. A. Waldmann. 1996. IL-15: a pleiotropic cytokine with diverse receptor signaling pathways whose expression is controlled at multiple levels. Immunity 4:329. 18. Janeway, C. A., and K. Bottomly. 1994. Signals and signs for lymphocyte responses. Cell 76:275. 19. Margolin, N., S. A. Raybuck, K. P. Wilson, W. Chen, T. Fox, Y. Gu, and D. J. Livingston. 1997. Substrate and inhibitor specificity of interleukin-1b-converting enzyme and related caspases. J. Biol. Chem. 272:7223. 20. Molineaux, S. M., F. J. Casano, A. M. Rolando, E. P. Peterson, G. Limjuco, J. Chin, J. R. Griffin, G. Calaycay, J. F. Ding, T. T. Yamin, et al. 1993. IL-1-a and IL-1-b processing in murine macrophages requires a structurally conserved homologue of human IL-1b converting enzyme. Proc. Natl. Acad. Sci. USA 90: 1809. 21. Ghayur, T., S. Banerjee, M. Hugunin, D. Butler, L. Herzog, A. Carter, L. Quintal, L. Sekut, R. Talanian, M. Paskind, et al. 1997. Caspase-1 processes IFN-g-inducing factor and regulates LPS-induced IFN-g production. Nature 386:619. 22. Hogquist, K. A., M. A. Nett, E. R. Unanue, and D. D. Chaplin. 1991. Interleukin 1 is processed and released during apoptosis. Proc. Natl. Acad. Sci. USA 88:8485. 23. Janicke, R. U., M. L. Sprengart, M. R. Wati, and A. G. Porter. 1998. Caspase-3 is required for DNA fragmentation and morphological changes associated with apoptosis. J. Biol. Chem. 273:9357. 24. Woo, M., R. Hakem, M. S. Soengas, G. S. Duncan, A. Shahinian, D. Kagi, A. Hakem, M. McCurrach, W. Khoo, S. A. Kaufman, G. Senaldi, T. Howard, S. W. Lowe, and T. W. Mak. 1998. Essential contribution of caspase-3/CPP32 to apoptosis and its associated nuclear changes. Genes Dev. 12:806. 25. Wilhelm, S., H. Wagner, and G. Hacker. 1998. Activation of caspase-3-like enzymes in non-apoptotic T cells. Eur. J. Immunol. 28:891.
