Abstract
Dendritic cells (DCs) in the periphery capture and process Ags, migrate to lymphoid organs, and initiate immune responses in T cells. IL-16, the soluble ligand for CD4, is a potent chemoattractant for CD4+ T cells, eosinophils, and monocytes and is mainly derived from activated T cells. Because migration is a fundamental property of DCs, we asked whether IL-16 induces chemotaxis in DCs and whether DCs are a source of IL-16. DCs were generated by culture of monocytes in IL-4 and GM-CSF for 6 days and subsequently highly purified employing magnetic beads. Migration was assayed by nitrocellulose and polycarbonate filter-based assays, and distinction of chemotaxis and chemokinesis was performed by a checkerboard analysis. Messenger RNA and protein data revealed constitutive expression and release of IL-16 by day-6 DCs. Gradients of rIL-16 induced a chemotactic response of DCs. Furthermore, the chemotactic activity of DC supernatant toward DCs themselves and T cells was mainly due to IL-16, because the addition of neutralizing Abs completely abrogated the migratory response. However, after induction of maturation by the addition of TNF-α and PGE2 DCs, neither expressed IL-16 mRNA nor produced IL-16 protein. We conclude that IL-16 may play a role in the trafficking of DCs and may be a major chemotactic signal from DCs toward themselves and toward T cells.
Interleukin-16 is a polypeptide cytokine that was originally described as a T cell-specific chemoattractant factor (1). It is synthesized by CD4+ and CD8+ T cells and released in response to Ag, mitogen, histamine, and serotonin (1); furthermore, IL-16 is produced by eosinophils and mast cells and by airway epithelial cells of asthmatics (1). It is cleaved from pro-IL-16 to the released 20-kDa IL-16, which is biologically active as a multimer (2, 3). As the natural soluble ligand for CD4, it has been shown to induce chemotaxis in CD4+ T cells as well as in monocytes and eosinophils (1). Besides its chemotactic properties, IL-16 induces the expression of IL-2Rα (CD25)2 and activates CD4+ T cells synergistically with IL-2 or IL-15 (4, 5). In contrast, preincubation of CD4+ T cells with IL-16 reduces the proliferative response to CD3/TCR ligation, probably via steric interaction with CD4-MHC binding (6).
Dendritic cells (DC)3 are specialized APC for the initiation of primary T cell immune responses (7, 8, 9). The current understanding is that resting DCs residing in interstitial or epithelial tissue are recruited to sites of pathogen challenge, acquire Ag, migrate from the tissue into afferent lymph, and enter draining lymph nodes, where they are found as interdigitating DCs clustering with T cells and inducing an immune response (8). Although migration is obviously a fundamental property of DCs, only little is known about signals mediating directed locomotion of DCs. Because DCs have been reported to express the CD4 Ag on the cell surface (10, 11), we were interested to explore whether IL-16 acts as a chemoattractant for DCs. We found that IL-16 exerts potent chemotactic activity and surprisingly noted that DCs themselves constitutively synthesize IL-16, which mainly accounts for the chemotactic activity of day-6 monocyte-derived DC supernatant toward DCs themselves and T cells.
Materials and Methods
Media, reagents, and Abs
2 was purchased from Calbiochem (Biotrade, Vienna, Austria). Monoclonal Abs directed against CD3 (HIT3a), CD4 (RPA-T4), CD14 (M5E2), CD19 (B43), and CD56 (B159) were obtained from PharMingen (Hamburg, Germany), and anti-CD83 (HB15A) was obtained from Immunotech (Instrumentation Laboratories, Vienna, Austria). Anti-CD1a mAb (OKT6) was kindly provided by Dr. N. Romani (Department of Dermatology, University Hospital Innsbruck, Innsbruck, Austria). Mouse IgG2a anti-IL-16 mAb 14.1 has been described (12
Cell culture
DCs were generated from PBMC as described (13, 14). In brief, PBMC were isolated from leukocyte-enriched buffy coats by density-gradient centrifugation through Histopaque-1077 (Sigma), resuspended in CM, and 5 × 107 PBMCs were allowed to adhere in 75-cm2 cell culture flasks for 45 min in a 37°C humidified 5% CO2 atmosphere. Nonadherent cells were removed, and adherent cells were cultured in 10 ml CM supplemented with 1 × 103 U/ml IL-4 and 1 × 103 U/ml GM-CSF. On day 2, 5 ml of CM supplemented with IL-4 and GM-CSF (both 1 × 103 U/ml), and on day 5, 5 ml of CM supplemented with 2 × 103 U/ml IL-4 and 2 × 103 U/ml GM-CSF were added. On day 6 of culture, DCs were harvested and highly purified DCs were obtained by magnetic cell separation. For this purpose, DCs were incubated with a mixture of mAbs against CD3, CD14, CD19, and CD56 followed by the addition of sheep anti-mouse IgG-coated magnetic beads (Dynabeads M-450; Dynal, Hamburg, Germany) following manufacturer’s instructions. The resulting DC population had a purity of >97% as determined by cytofluorometry and CD1a, CD3, CD14, CD19, CD56, and CD40 staining. Following this procedure, DCs were subjected to chemotaxis assay, protein and RNA isolation, or further cultured in RPMI supplemented with 0.5% BSA at a density of 2.5 × 105 DC per ml for 24 h, and the resulting supernatant was assayed for T cell and DC chemotaxis and IL-16 synthesis. Alternatively, maturation was induced by incubation of purified DCs in CM supplemented with IL-4, GM-CSF, TNF-α (1 × 103 U/ml), and PGE2 (10 μM) for 72 h. Subsequently, DCs were subjected to FACS analysis and RNA extraction.
Highly purified monocytes (>97%) were obtained by positive selection of CD14+ cells on ice. Cells were incubated with CD14 mAb followed by the addition of sheep anti-mouse IgG-coated magnetic beads (Dynal) and intensely washing of the CD14+ cell population.
Highly purified T cells (>98%) were obtained from the plastic-nonadherent cell population by negative selection of T lymphocytes employing anti-CD19-coated magnetic cell separation colloidal superparamagnetic microbeads (Miltenyi Biotec, Bergisch Gladbach, Germany) as recently described (15).
Northern analysis
Total RNA was prepared from 2 × 106 purified DCs or 8 × 106 freshly isolated purified monocytes, and 10 μg of each was gel-electrophoresed and blotted onto nylon membranes as recently described (16). The IL-16 probe was a 397-bp PCR product spanning the open reading frame encoding the C-terminal 130 aa of IL-16. The probe was radioactively labeled with [32P]dCTP employing the random primed labeling method according to manufacturer’s (Boehringer Mannheim, Vienna, Austria) instructions and hybridized as described (16). Control hybridizations were performed with the rat cDNA of the housekeeping gene GAPDH to ensure equal loading of RNA.
Western analysis
Cell extracts of 2 × 106 purified DCs were prepared by lysis in RIPA buffer (150 mM NaCl, 1% Nonidet P-40, 0.5% deoxycholate, 0.1% SDS, 50 mM Tris-HCl, pH 8.0, 0.2 mM PMSF, 1 μg/ml pepstatin, 0.5 μg/ml leupeptin) for 30 min at 4°C and subsequent centrifugation for 5 min at 16,000 × g. Protein content was estimated by a commercial Bradford assay (Bio-Rad, Vienna, Austria) with BSA standards, and 60 μg of cell lysates per lane were boiled in SDS sample buffer and subjected to electrophoresis through a 12.5% SDS-polyacrylamide gel. Proteins were electrophoretically transferred to nitrocellulose membrane (Schleicher & Schuell, Vienna, Austria) and probed with biotinylated rabbit anti-IL-16 pAb, followed by streptavidin-POD (Boehringer Mannheim) at a 1:2000 dilution. Visualization of the signal was by 0.07% 4-Cl-1-naphthol/15% methanol/0.05% H2O2/PBS.
ELISA
Sandwich ELISA was performed according to standard protocols with immobilized anti-IL-16 14.1 mAb (2 μg/ml in carbonate buffer) and biotinylated IL-16 pAb (1 μg/ml) as detection Ab. The secondary reagent streptavidin-POD (Boehringer Mannheim) was used according to manufacturer’s instructions. Tetramethylbenzidine (BM Blue POD substrate, Boehringer Mannheim) was used as substrate and measured at 450 nm in an ELISA reader. The standard curve was prepared with rhuIL-16 in a concentration range from 0.07 to 5 ng/ml.
Cytofluorometric analysis of DC surface phenotype
A total of 5 × 105 DCs were washed in PBS/2% FCS, resuspended in 250 μg/ml human IgG/PBS/2% FCS, and incubated for 20 min at 4°C. After pelleting, DCs were incubated alternatively with 10 μg/ml anti-CD1a, anti-CD4, anti-CD40, or anti-CD83 mAbs and the respective isotype-matched control Igs for 30 min at 4°C. After washing in PBS/2% FCS, a 1:40 dilution of FITC-anti-mouse IgG in PBS/2% FCS was incubated for 30 min at 4°C. Cells were immediately analyzed on a FACScan after the addition of 1 μg/ml PI. Analysis was performed on PI− cells (e.g., viable cell population) with Cellquest software (Becton Dickinson).
Cytofluorometric analysis of intracellular IL-16
Analysis for intracellular IL-16 was performed on either unstained DCs or surface-stained DCs (omitting the addition of PI) as described in the previous paragraph. 5 × 105 DCs were washed twice in PBS and resuspended in 2% formaldehyde/PBS for 10 min, washed once in PBS, and incubated with 0.1% saponin/10% goat serum/250 μg/ml rabbit IgG/PBS for 10 min at room temperature. After pelleting, 10 μg/ml biotinylated IL-16 pAb, or 14.1 anti-IL-16 mAb or the respective isotype-matched control Igs in PBS/2% FCS/0.1% saponin, were incubated for 30 min at room temperature, followed by washing in PBS/2% FCS/0.1% saponin. As indicated, the IL-16 Abs were preincubated with 10 μg/ml rhuIL-16 to confirm specificity of the Abs. Subsequently, DCs were incubated with either a 1:25 dilution of streptavidin-PE or 1:40 dilution of FITC-anti-mouse IgG in PBS/2% FCS/0.1% saponin for 30 min at room temperature, washed twice in PBS/2% FCS, and immediately analyzed on a FACScan.
IL-16 immunohistochemistry
Tissue sections of human normal skin were acquired from excess tissue obtained during elective plastic surgery. Serial sections were fixed in 2% paraformaldehyde in Tris buffer for 8 min at 4°C. Sections were incubated overnight at 4°C with biotinylated IL-16 pAb (dilution 1:50 in 1% BSA/0.1% saponin in Tris buffer). Blocking of specific binding sites of the pAb was performed with rhuIL-16 as described for the FACS analysis. Subsequently, streptavidin-POD was added at a dilution of 1:800 and incubated for 30 min, and signal developed with diaminobenzidine (Sigma). Slides were counterstained with hemalam. Identification of Langerhans cells was performed by anti-CD1a (Immunotech) according to standard protocols.
Chemotaxis assay: nitrocellulose
Migration of cells into cellulose nitrate to gradients of soluble attractants was measured using a 48-well microchemotaxis chamber (Neuroprobe, Bethesda, MD) in which a 5-μm (for lymphocytes) or a 8-μm (for DCs) pore-size filter (Sartorius, Göttingen, Germany) separates the upper and lower chamber. Then, 30-μl aliquots of chemoattractant solution or control medium (RPMI 1640/0.5% BSA) were added to the lower wells of the chamber, and the filter was carefully layered onto the wells and covered with a silicon gasket and the top plate. Then, 50 μl of cell suspension (1 × 106 cells/ml) were seeded in the upper chamber. For checkerboard analysis, cells were resuspended in RPMI 1640/0.5% BSA containing various concentrations of chemoattractants just before transferring them to the upper chamber. Lymphocytes or DCs were allowed to migrate toward attractants in the lower chambers for 2 or 4 h at 37°C in humidified atmosphere (5% CO2). After this incubation time, the nitrocellulose filters were dehydrated, fixed, and stained with hematoxylin-eosin. Migration depth of cells into the filter was quantified by microscopy, measuring the distance (μm) from the surface of the filter to the leading front of cells before any cells had reached the lower surface (leading-front assay). Data are expressed as chemotaxis index (CI), which is the ratio between the distances of directed and undirected migration of DCs into the nitrocellulose filters.
Chemotaxis assay: polycarbonate
For verification of results obtained from chemotaxis experiments with nitrocellulose filters, we tested DC chemotaxis in the Transwell system (Costar, Cambridge, MA) bearing 8-μm pore-size polycarbonate filters. Medium or soluble chemoattractants were put into the lower well of a 24-well tissue culture plate, and transwell cups, loaded with 5 × 104 DCs, were added. Cells migrated for 4 h toward chemoattractants, and after centrifugation they were stained with 2′,7′-bis(2-carboxyethyl)-5-(and -6)-carboxyfluorescein,acetoxymethyl ester (Molecular Probes, Leiden, the Netherlands; final concentration 6.25 μg/ml). After washing twice, fluorescence activity was determined using the CytoFluor 2350 fluorescence measurement system (Millipore, Bedford, MA).
Calcium flux assay
DCs were loaded with 1 μM Fluo-3/AM (Molecular Probes) in CM/0.02% Pluronic (Molecular Probes), incubated for 15 min at 20°C in the dark, then washed twice in CM, resuspended at 1 × 106 DCs per ml in CM, and further cultured at 37°C for 30 min. Subsequently 1 ng/ml rhuIL-16 was added and DCs analyzed by cytofluorometry at 488 nm excitation every 10 s for up to 10 min. Stimulation of DCs with 500 ng/ml ionomycin (Calbiochem, Biotrade) served as a positive control.
Anti-phosphotyrosine immunoblot
A total of 1 × 106 purified day-6 DCs were stimulated with 1 ng/ml rhuIL-16 for various time points. Whole-cell lysates were separated by 10% SDS-PAGE and electrotransferred to nitrocellulose paper. The nitrocellulose was blocked with 1% BSA in TBS, pH 7.5, for 20 min. Anti-phosphotyrosine-specific Ab labeled with HRP (RC-20; Transduction Laboratories, Lexington, KY) was used as the primary Ab and was added to the blot for 1 h. Subsequently the blot was washed and subjected to chemiluminescence analysis (Super Signal; Pierce). The autoradiograph was quantified by densitometry (Molecular Dynamics) with changes in total cellular tyrosine phosphorylation expressed as relative densitometric units.
Statistical analysis
Chemotaxis data are expressed as mean and SEM of the CI. Means were compared by Kruskal-Wallis ANOVA and by Mann-Whitney U test. A difference with p < 0.05 was considered to be significant. Statistical analyses were calculated using the StatView software package (Abacus Concepts, Berkeley, CA).
Results
Monocyte-derived DCs synthesize IL-16
Highly purified monocyte-derived DCs (day-6 DCs) were obtained by 6-day culture of monocytes in IL-4 and GM-CSF as described (13, 14) and by further depletion of contaminating CD3+, CD14+, CD19+, and CD56+ cells. The phenotype of the resulting day-6 DC population was determined by cytofluorometry after staining for CD1a, CD40, and CD83 (Fig. 1⇓, a–c) and revealed a purity of >97%. Purity was further ascertained by DCs staining negative for CD3, CD14, CD16, CD19, and CD56 (data not shown). Intracellular FACS staining with 14.1 mAb and a specific rabbit pAb revealed IL-16 reactivity, which could completely be inhibited by rhuIL-16 (Fig. 1⇓, e and f). Because the obtained day-6 DCs can be subgrouped into a population expressing CD1ahigh and CD1alow, we separately analyzed IL-16 reactivity and found that IL-16 is equally expressed in both populations (Fig. 1⇓, g and h). Analysis of IL-16 reactivity by intracellular FACS staining of the CD14+ fraction of freshly isolated PBMC (i.e., monocytes) provided no evidence for IL-16 synthesis (Fig. 1⇓i). As depicted in Fig. 2⇓A, Northern analysis revealed that monocytes hardly express any IL-16 mRNA, while day-6 DCs clearly express IL-16 mRNA. Western analysis of day-6 DC cell-extracts separated by SDS-PAGE revealed four molecular mass species of IL-16 resembling pro-IL-16 as shown in Fig. 2⇓B. No band was observed at 20 kDa, suggesting that mature IL-16 is not stored in DCs. Quantification of IL-16 in the supernatant of day-6 DCs by ELISA revealed that day-6 DCs constitutively release IL-16 and synthesize 3385 ± 422 pg/ml per 2.5×105 DCs per 24 h (n = 8). Full maturation of DCs can be induced by incubating day-6 DCs for 72 h in medium containing TNF-α and PGE2 as described by Rieser et al. (14). These mature DCs (day-9 DCs) hardly express any CD1a Ag (Fig. 1⇓j), while CD83 expression is markedly up-regulated (Fig. 1⇓k) as has been described (14). Intracellular FACS analysis revealed no IL-16 reactivity (Fig. 1⇓l), and Northern analysis provided no evidence for IL-16 mRNA expression (Fig. 2⇓A).
Cytofluorometry. a–d, Surface markers on day-6 DCs; e, intracellular IL-16 expression in day-6 DCs detected by 14.1 mAb; f, intracellular IL-16 expression in day-6 DCs detected by polyclonal rabbit IgG; g and h, intracellular IL-16 expression in day-6 DCs staining low and high, respectively, for CD1a; i, intracellular IL-16 expression in CD14+ monocytes; j and k, surface markers on day-9 DCs; l, intracellular IL-16 expression in day-9 DCs. Solid lines indicate specific Abs (10 μg/ml), dotted lines indicate isotype-matched control Abs (10 μg/ml), and dashed lines indicate specific anti-IL-16 Abs (10 μg/ml) blocked with rhuIL-16 (10 μg/ml). Data are representative of at least three independent experiments.
A, Northern analysis. Total RNA was extracted from purified CD14+ monocytes and purified day-6 and day-9 DCs as indicated and analyzed for IL-16 mRNA expression by Northern analysis. Equal loading was ascertained by hybridization with the housekeeping gene GAPDH. B, Western analysis. A total of 400 ng of rhuIL-16 and 60 μg of cellular extracts derived from purified day-6 DCs were separated by SDS-PAGE, blotted, and probed with biotinylated rIgG pAb. Molecular mass standards (kDa) are indicated by arrows. The figure is representative of three independent experiments.
IL-16 is expressed in human epidermis
In vivo IL-16 biosynthesis was analyzed by immunohistochemistry on tissue sections of human skin. As depicted in Fig. 3⇓A, IL-16 is moderately expressed in human epidermis. IL-16 reactivity was specific because binding of the IL-16 pAb could be completely blocked by rhuIL-16 (Fig. 3⇓B). Langerhans cells were identified by expression of the CD1a Ag (Fig. 3⇓C). The homogeneity of immuno-staining for IL-16 suggests that all epidermal cell types including Langerhans cells may be a source of IL-16.
IL-16 immunohistochemistry. IL-16 synthesis was studied in serial tissue sections of human skin obtained routinely during elective plastic surgery. A, IL-16 reactivity in the epidermis detected by anti-IL-16 pAb. Magnification, ×400. B, Blocking of IL-16 reactivity by preincubation of IL-16 pAb with rhuIL-16. Magnification, ×400. C, Identification of Langerhans cells within the epidermis by CD1a immuno-staining. Magnification, ×400.
IL-16 is mainly responsible for the chemotactic activity of supernatant from day-6 DCs toward T lymphocytes
We analyzed chemotactic activity of day-6 DC supernatant toward T cells in a nitrocellulose filter-based assay and found that T cells efficiently migrate in response to factors released by DCs (Fig. 4⇓A). Therefore, we added 14.1 anti-IL-16 mAb, which has previously been shown to efficiently neutralize IL-16 bioactivity (17), and found that the chemotactic activity can be almost completely abrogated (Fig. 4⇓A). IL-15 has previously been reported to account for chemotactic activity of day-7 DCs obtained by culture of monocytes in IL-4 and GM-CSF for 6 days followed by addition of TNF-α and IL-1 (18). The same anti-IL-15 mAb used in that study had no influence on the chemotactic activity of day-6 DC supernatant as depicted in Fig. 4⇓A.
A, T-lymphocyte chemotaxis. Dilutions of day-6 DC-derived supernatants with or without Abs against IL-16 or IL-15 (each 10 μg/ml) were added to the lower wells of a Boyden microchemotaxis chamber. RhuIL-16 (10 ng/ml) and RANTES (100 ng/ml) served as positive controls. Thereafter, 50-μl aliquots of purified T lymphocytes (1 × 106 cells/ml) were seeded in the upper wells and allowed to migrate into 5-μm pore-size nitrocellulose filters for 120 min (left). B and C, DC chemotaxis. Day-6 DCs migrated into 8-μm pore-sized filters for 4 h toward pure medium, rhuIL-16 (0.1 pg/ml to 10 ng/ml), day-6 DC supernatants (undiluted 3385 ± 422 pg/ml), and RANTES (100 ng/ml; positive control) or were supplemented with an Ab against rhuIL-16 (10 μg/ml). Migration depth was quantified microscopically (leading-front assay) and data are expressed as mean ± SEM of the CI. n = 6 (n = 3 for experiments using Abs). Statistical analyses: Mann-Whitney U test vs medium after Kruskal-Wallis ANOVA (p < 0.001); ∗, p < 0.05; ∗∗, p < 0.01.
IL-16 is chemotactic for DCs
It has previously been reported that DCs express the CD4 Ag (10, 11), which was confirmed in our experiments as shown in Fig. 1⇑d. Because CD4 is the receptor for IL-16, we wondered whether IL-16 might induce a chemotactic response. We employed a nitrocellulose filter-based leading-front chemotaxis assay and found that rhuIL-16 dose-dependently induces migration of DCs (Fig. 4⇑B). Experiments using polycarbonate filters revealed similar results, but with lower sensitivity (Table I⇓). Compared with other well-known DC chemoattractants at optimal concentrations such as monocyte chemotactic protein (MCP)-3 (CI = 1.48 ± 0.08) and C5a (CI = 1.52 ± 0.06), IL-16 exerted strong chemotaxis of DCs. A concentration of as low as 1 pg/ml rhuIL-16 resulted in significant migration in nitrocellulose, with the maximum response obtained with 1 ng/ml rhuIL-16 (CI = 1.45 ± 0.05; Fig. 4⇑B). The migratory response was consequently inhibited by addition of saturating amounts of 14.1 anti-IL-16 mAb. To differentiate between chemotaxis and chemokinesis, i.e., directed vs undirected migration, we performed a checkerboard assay. Increasing concentrations of rhuIL-16 were applied to the upper and lower compartment and the migratory responses determined after 4 h. Table II⇓ shows that migration was dependent on a concentration gradient of rhuIL-16, while even high concentrations of rhuIL-16 without a gradient did not induce DC movement, what resembles a chemotactic response. We furthermore tested the chemotactic activity of day-6 DC supernatant toward DCs and found efficient migration (Fig. 4⇑C). To evaluate the involvement of IL-16, we added 14.1 anti-IL-16 mAb and found that the chemotactic activity can be almost completely abrogated (Fig. 4⇑C).
DC chemotaxis through polycarbonate filters (8-μm pore size)a
Checkerboard analysis of IL-16-induced DC migrationa
IL-16 induces phosphorylation of tyrosine residues in DCs
As an independent measure of response of DCs toward IL-16, we evaluated intracellular Ca2+ flux and phosphorylation reactions. Although IL-16 did not induce a detectable Ca2+ flux as determined by fluorescent dyes and FACS analysis (data not shown), we noted increased total cellular tyrosine phosphorylation starting at 2 min and progressing through the 5 min time point (Fig. 5⇓A). Fig. 5⇓B shows increased tyrosine phosphorylation of a 40-kDa band in the IL-16- treated cells within the same time range.
Tyrosine phosphorylation by IL-16. DCs were either left untreated (□) or stimulated with 1 ng/ml rhuIL-16 (○) for the indicated period of time. Cellular lysates were analyzed for tyrosine phosphorylations as outlined in Materials and Methods. A shows total cellular tyrosine phosphorylation, and B shows tyrosine phosphorylation of a 40-kDa band.
Discussion
DCs have been identified as the key cell population linking the innate with the adaptive immune system (7, 8, 9). The current understanding is that DCs resting in the periphery are activated by Ag-induced inflammatory cytokines, acquire and process Ag, migrate to lymphoid organs, cluster with T cells, and stimulate a specific immune response. Although migration is obviously a fundamental property of DCs, only little is known about signals mediating directed locomotion of DCs. Several molecules have been reported to provide a chemotactic signal to DCs: fMLP (representative of formyl peptides of bacterial origin), C5a, MCP-3, MCP-4, m-eotaxin, macrophage inflammatory protein-1α, RANTES, and EBV-induced molecule 1 ligand (19, 20, 21). We now report on the chemotactic activity of IL-16, a cytokine originating from T cells, eosinophils, and basophils, toward day-6 DCs. As an independent measure of response of DCs toward IL-16, we also demonstrate enhanced tyrosine phosphorylation of total cellular proteins, especially of a 40-kDa band, by IL-16. Further studies will aim at the characterization of IL-16-induced signal transduction in DCs.
Polycarbonate filter assays are used by many investigators, but there are several publications dealing with major limitations of this method (22, 23, 24, 25). Very important for our study was the fact that the main drawback of polycarbonate filters is their thickness, which does not allow any assessment of effects of absolute concentration compared with gradients. Therefore, the checkerboard in polycarbonate filter assay is considered invalid (25). These filters have a thickness of 10 μm, and cell numbers are counted on the lower surface of the filter through which cells migrate. It is still under debate whether concentration gradients can be formed and remain constant over time by a membrane of 10-μm thickness bearing pores of 8-μm size. In contrast, in the leading-front assay using nitrocellulose filters (thickness, 150 μm), the migration depth of cells into the filter is measured microscopically. Furthermore, the maintenance of a concentration gradient by a 150-μm nitrocellulose filter throughout the assay time of 4 h is undoubted. Limitations of our method include the known disadvantage that the leading front may be a small and unrepresentative sample of the cell population as well as the fact that the morphology of the migrated cells is more easily studied in polycarbonate filters (23). For reasons of comparison, we also demonstrate IL-16-induced migration of DCs through polycarbonate filters, showing similar results and confirming reliability of our method. Discrimination of chemotaxis and chemokinesis was performed by checkerboard analysis and revealed IL-16-induced migration as true chemotaxis.
Interestingly, our results also demonstrate that functional IL-16 is produced by monocyte-derived day-6 DCs generated by culture in medium supplemented with IL-4 and GM-CSF. Northern analysis revealed that purified monocytes do not express IL-16 mRNA, while highly purified day-6 DCs express IL-16 mRNA message. This translates into protein synthesis as determined by ELISA and Western analysis. The IL-16 species detected in cell extracts of day-6 DCs indicate that IL-16 is produced as pro-IL-16 as suggested for T lymphocytes (26). The absence of a 18- to 20-kDa band of mature IL-16 suggests that pro-IL-16 is rapidly released upon processing. Because day-6 DCs constitutively release bioactive IL-16 as determined by ELISA and chemotaxis bioassay, storage of the processed form of IL-16 is not likely. Constitutive IL-16 release from day-6 DCs is 9-fold larger on a per-cell basis than IL-16 release from Con A-stimulated PBMC, while PBMC do not constitutively secrete IL-16 (data not shown). A further confirmation for IL-16 synthesis is provided by intracellular FACS staining for IL-16 with the 14.1 anti-IL-16 mAb and an affinity-purified rabbit pAb, the reactivity of both of which could completely be blocked by rIL-16. It is noteworthy that we did not detect any difference in intracellular IL-16 expression in day-6 DCs staining bright for CD1a and those staining low. Intracellular staining of freshly isolated monocytes for IL-16 confirmed that monocytes do not synthesize IL-16. In contrast to day-6 DCs, mature day-9 DCs neither express IL-16 mRNA nor synthesize IL-16 protein. To demonstrate IL-16 synthesis in vivo, we studied IL-16 expression in human epidermis. We show that IL-16 is homogeneously expressed in the epidermis, suggesting that all epidermal cell types including Langerhans cells, which are premature dendritic cells resting in the skin, may be a source of IL-16.
Because our evidence suggests that DCs synthesize IL-16 and respond chemotactically toward IL-16, we tested whether DCs might attract themselves. Day-6 monocyte-derived DCs showed an efficient migratory response toward their own supernatant, mainly accountable to IL-16, as proven by complete inhibition in neutralization studies. This suggests that DCs might perpetuate their influx into inflammatory sites by secretion of IL-16.
Furthermore, we show that the supernatant of highly purified day-6 DCs has potent chemotactic activity on T cells. Neutralizing IL-16 with mAb completely abrogated the migratory response of T cells, suggesting that IL-16 might account for almost all chemotactic activity of day-6 monocyte-derived DCs. This observation suggests that DCs are actively recruiting and attracting T cells via constitutive expression of IL-16. It has previously been reported that IL-15 might account for T cell attraction in mature DCs (18). Neutralization experiments with anti-IL-15 mAb did not have any inhibitory effect on chemotactic activity of day-6 DC supernatant, suggesting that DCs might use different cytokines at different maturation stages for this purpose, a notion that is further supported by the absence of IL-16 synthesis in day-9 DCs. Recently, a DC-derived C-C chemokine named DC-CK1 has been cloned and shown to be produced by a substantial proportion of monocyte-derived day-6 DCs and to be chemotactic on CD4+ and CD8+ T lymphocytes, especially naive T cells (27). No information has been provided on the relative contribution of DC-CK1 in DC-induced T cell chemotaxis. Our data suggest that IL-16 produced by DCs might provide a major contribution in the recruitment of CD4+ T cells by DCs.
IL-16 has been reported to be involved in the induction of airway hyperresponsiveness and up-regulation of IgE in a murine model of allergic asthma (28). Mice were sensitized with OVA, and IL-16 was subsequently detected in the bronchoalveolar lavage fluid, in airway epithelial cells, and in cellular infiltrates (28). Administration of anti-IL-16 mAb significantly reduced OVA-specific IgE-up-regulation and airway hyperresponsiveness (28). McWilliam et al. reported that DCs are recruited into the airway epithelium early during the acute inflammatory response to a broad spectrum of stimuli, including OVA (19). A recent report suggested that DCs are required for the development of chronic eosinophilic airway inflammation in response to OVA challenge (29). Therefore, we suggest that IL-16 released from airway epithelial cells might contribute to the attraction of DCs and DC-derived IL-16 might further promote the influx of DCs themselves, CD4+ T cells, and eosinophils, the characteristic feature of eosinophilic airway inflammation.
Recruitment of DCs to sites of pathogen challenge represents the first step following activation by the inflammatory stimulus, and migration toward lymphoid areas subsequent to Ag-uptake comprises the next step. It is currently anticipated that factors specifically produced by lymphoid areas might be responsible for the homing of DCs to lymphoid areas (8). IL-16 produced by T cells in lymph follicles might be a candidate cytokine for this function.
Because DCs play an essential role in the induction of immune responses, they are currently considered promising tools for immunotherapy (30). Most ongoing clinical trials evaluating the efficacy of administration of Ag-pulsed DCs in malignant disease employ IL-4/GM-CSF monocyte-derived DCs, as investigated in our study. The notion that IL-16 stimulates a directed migratory response of DCs might be exploited in the context of immunization strategies, especially because IL-16 is well known for its chemoattractant properties on T lymphocytes as effector cells of an immune response.
Acknowledgments
We thank Dr. H. Zoller, Department of Medicine, Innsbruck, for expert assistance with Western experiments, and Dr. N. Romani, Department of Dermatology, University Hospital Innsbruck, and Dr. G. Schuler, Department of Dermatology, University of Erlangen, Germany, for helpful discussions. The technical assistance of Sabine Jöbstl, Martin Lorentz, and Fahrettin Acikbas is appreciated.
Footnotes
-
↵1 This work was supported by Grants P12790 and P09977 of the Austrian Science Fund to H.T. and C.J.W., respectively, and National Institutes of Health Grant AI35680 to W.W.C.
-
↵2 Address correspondence and reprint requests to Dr. Herbert Tilg, Department of Medicine, University Hospital Innsbruck, Anichstrasse 35, 6020 Innsbruck, Austria. E-mail address: Herbert.Tilg{at}uibk.ac.at
-
↵3 Abbreviations used in this paper: DC, dendritic cells; CI, chemotaxis index; CM, complete medium; rhu, recombinant human; POD, peroxidase; PI, propidium iodide; MCP, monocyte chemotactic protein; pAb, polyclonal Ab.
- Received September 23, 1998.
- Accepted July 6, 1999.
- Copyright © 1999 by The American Association of Immunologists
References
In this issue
