Cytotoxicity and Activation of NK Cell-Mediated Dendritic Cells: Direct ...

Download PDF
4 downloads 44204 Views 315KB Size Report
Comment
18 h of incubation, chromium release was measured. The results of ..... Nair, S. K., A. Heiser, D. Boczkowski, A. Majumdar, M. Naoe, J. S. Lebkowski,. J. Vieweg, and E. .... Infiltration of dendritic cells in relation to tumor invasion and lymph node.
The Journal of Immunology

Tumoricidal Potential of Native Blood Dendritic Cells: Direct Tumor Cell Killing and Activation of NK Cell-Mediated Cytotoxicity1 Marc Schmitz,2* Senming Zhao,* Yvonne Deuse,* Knut Scha¨kel,* Rebekka Wehner,* Hanka Wo¨hner,* Kristina Ho¨lig,† Florian Wienforth,* Andrea Kiessling,* Martin Bornha¨user,‡ Achim Temme,* Michael A. Rieger,* Bernd Weigle,* Michael Bachmann,* and E. Peter Rieber* Dendritic cells (DCs) are characterized by their unique capacity for primary T cell activation, providing the opportunity for DC-based cancer vaccination protocols. Novel findings reveal that besides their role as potent inducers of tumor-specific T cells, human DCs display additional antitumor effects. Most of these data were obtained with monocyte-derived DCs, whereas studies investigating native blood DCs are limited. In the present study, we analyze the tumoricidal capacity of M-DC8ⴙ DCs, which represent a major subpopulation of human blood DCs. We demonstrate that IFN-␥-stimulated M-DC8ⴙ DCs lyse different tumor cell lines but not normal cells. In addition, we show that tumor cells markedly enhance the production of TNF-␣ by M-DC8ⴙ DCs via cell-to-cell contact and that this molecule essentially contributes to the killing activity of M-DC8ⴙ DCs. Furthermore, we illustrate the ability of M-DC8ⴙ DCs to promote proliferation, IFN-␥ production, and tumor-directed cytotoxicity of NK cells. The M-DC8ⴙ DC-mediated enhancement of the tumoricidal potential of NK cells is mainly dependent on cell-to-cell contact. These results reveal that, in addition to their crucial role in activating tumor-specific T cells, blood DCs exhibit direct tumor cell killing and enhance the tumoricidal activity of NK cells. These findings point to the pivotal role of DCs in triggering innate and adaptive immune responses against tumors. The Journal of Immunology, 2005, 174: 4127– 4134.

D

endritic cells (DCs)3 are professional APCs that display an extraordinary capacity to induce, sustain, and regulate T cell responses (1–3). Because of their unique ability to activate naive T cells, DCs evolved as promising candidates for vaccination protocols in cancer therapy (4). Animal models demonstrated that tumor Ag-presenting DCs are capable of inducing protective and therapeutic antitumor responses (5, 6). Also in humans, clinical trials revealed promising immunologic and clinical effects of Ag-loaded DCs administered as a vaccine against cancer (7–9). Although there is convincing evidence that DCs mediate their antitumor effects by stimulating tumor-specific T lymphocytes, other tumor-directed mechanisms also play a role. One alternative mechanism was demonstrated by a report (10) illustrating the tumoricidal potential of rat spleen DCs in vitro, which was mediated by the NK cell receptor protein 1 (NKR-P1). Chapoval et al. (11) provided evidence that also human monocyte-derived DCs displayed potent growth inhibition activity on a wide spectrum of human tumor cell lines in vitro and that this effect was dependent *Institute of Immunology, †Institute of Transfusion Medicine, and ‡Department of Medicine I, Medical Faculty, Technical University of Dresden, Dresden, Germany Received for publication June 9, 2004. Accepted for publication January 26, 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 grants from the Wilhelm Sander-Stiftung (to M.S., E.P.R.) and the Medical Faculty, Technical University of Dresden (to M.S.). 2

Address correspondence and reprint requests to Dr. Marc Schmitz, Institute of Immunology, Medical Faculty, Technical University of Dresden, Fetscherstr. 74, 01307 Dresden, Germany. E-mail address: [email protected]

3 Abbreviations used in this paper: DC, dendritic cell; NKR-P1, NK cell receptor protein 1.

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

on contact between DCs and tumor cells. Furthermore, they showed that DC-mediated tumor growth inhibition could be enhanced by proinflammatory stimuli such as LPS and IFN-␥. Moreover, it has been reported that human monocyte-derived DCs exhibited potent lytic activity toward various human tumor cell lines of different tissues origin (12–16). Additional data revealed that the tumoricidal potential of human monocyte-derived DCs was mediated by effector molecules such as Fas ligand (13, 17), TNF (14, 15, 17), lymphotoxin-a1b2 (17), or TRAIL (17), dependent on the susceptibility of the tumor cells. An additional DC-mediated antitumor effect was described by recent data illustrating the capacity of DCs to enhance the tumoricidal potential of NK cells (18). NK cells are potent effectors of the innate immune system that are involved in the lysis of virus-infected cells and tumor cells. Their cytolytic function is regulated by the balance of activating and inhibitory signals that are transmitted by different membrane receptors after engagement by their ligands on the surface of potential target cells (19 –22). Fernandez et al. (23) reported that FMS-like tyrosine kinase-3 ligandexpanded or adoptively transferred DCs markedly enhanced NK cell-dependent antitumor effects in mice with MHC class I-negative tumors. They also documented that cell-to-cell contact between DCs and NK cells resulted in a substantial increase of NK cell cytotoxicity and IFN-␥ production in vitro. Additional studies also revealed that human DCs could trigger the effector functions of NK cells. Yu et al. (24) demonstrated the capability of CD34⫹ cell-derived DCs to significantly improve the tumor-directed cytotoxicity and IFN-␥ secretion of NK cells. In addition, it has been shown that the expansion, IFN-␥ production, and tumoricidal potential of NK cells are increased efficiently upon interaction with monocyte-derived DCs (25–27). Cross-talk is bidirectional because activated NK cells induce immature DC activation in terms 0022-1767/05/$02.00

4128 of up-regulation of MHC and costimulatory molecules, resulting in an enhancement of the allostimulatory capacity of DCs (28, 29). In addition, activated NK cells significantly increased the production of proinflammatory cytokines such as TNF-␣ and IL-12 by DCs. The NK cell-mediated activation of DCs is based on cell-to-cell contact and proinflammatory cytokines such as TNF-␣ (28, 29). Most of these data were obtained with DCs, which were differentiated for several days in the presence of various cytokines. However, only little is known about the tumoricidal potential of native human blood DCs. Recently, distinct human blood DC subsets have been described as HLA-DR-positive cells lacking the lineage markers CD3, CD14, CD19, and CD56. On the basis of the differential expression of CD11c and CD123, they have been subdivided into CD11c⫹CD123dim myeloid DCs and CD11c⫺CD123high plasmacytoid DCs (30 –34). Additional investigations revealed that CD11c⫹CD123dim myeloid DCs comprised three distinct subpopulations, which differ by the expression of CD16, CD1b/c, and BDCA-3 (35). In addition, these DC subsets differentially express Ig-like transcripts, C-type lectins, as well as costimulatory molecules, and display marked differences in their capacity to stimulate allogeneic T cell proliferation (35). The CD16⫹ DC subpopulation can be further subdivided into M-DC8⫺ and M-DC8⫹ DCs (36). M-DC8⫹ DCs represent a novel major subset of inflammatory human blood DCs, which we have identified previously (37, 38). These DCs exhibit a strong expression of the myeloid markers CD11c, CD13, and CD33, as well as of the costimulatory molecules CD40 and CD86. They differ from the two other myeloid blood DC subsets (CD1b/c⫹ DCs and BDCA-3⫹ DCs) and the CD11c⫺CD123high plasmacytoid DCs by the unique expression of the M-DC8 Ag, which represents a carbohydrate modification of P-selectin glycoprotein ligand-1 (38). In addition, M-DC8⫹ DCs characteristically express the anaphylatoxin receptors C3aR and C5aR and lack the cutaneous lymphocyte Ag (38). Functional data provide evidence that M-DC8⫹ DCs efficiently activate neoantigen-specific CD4⫹ T cells and tumor-reactive CD8⫹ CTLs (37, 38). In addition, M-DC8⫹ DCs were found to efficiently mediate Ab-dependent cellular cytotoxicity (39). To get more insight into the tumoricidal capacity of native human blood DCs, we investigated the ability of M-DC8⫹ DCs to directly kill tumor cells and to improve the activity of NK cells. In the present study, we found that M-DC8⫹ DCs stimulated with IFN-␥ lysed different tumor cell lines but not normal cells. In addition, we provided evidence that tumor cells markedly enhanced intracellular TNF-␣ production in DCs via cell-to-cell contact and that TNF-␣ essentially contributed to the cytotoxic effect of MDC8⫹ DCs. Furthermore, we demonstrated the ability of M-DC8⫹ DCs to stimulate proliferation, IFN-␥ secretion, and tumor-directed cytotoxicity of NK cells.

Materials and Methods Cell lines The pancreatic cancer cell line Capan-1, the breast cancer cell line MCF-7, the colorectal adenocarcinoma cell lines COLO 205 and HT-29, the chronic myelogenous leukemia cell line K-562, as well as the T cell leukemia cell line Jurkat, were obtained from American Type Culture Collection and were cultured according to the provider’ s instructions. The human lung fibroblast cell line NHLF was purchased from Cambrex and cultured as directed. HUVECs were kindly provided by Dr. A. Deussen (Institute of Physiology, Medical Faculty, Technical University of Dresden, Dresden, Germany). The generation and cultivation of the endothelial cells were performed as described previously (40). Briefly, cells were isolated by enzymatic digestion (40 mg of dispase/10 ml of PBS solution per vein) under sterile conditions. Subsequently, cells were cultivated to confluence on tissue culture dishes (Nunc) at 37°C in a humidified environment containing 5% CO2. Ready-to-use Endothelial Cell Growth Medium (Promo Cell) was used for cultivation.

TUMORICIDAL POTENTIAL OF BLOOD DCs Immunomagnetic isolation of M-DC8⫹ DCs and CD56⫹CD3⫺ NK cells Blood samples were obtained from healthy donors with informed consent. Isolation of M-DC8⫹ DCs was performed as described previously (37). Briefly, PBMCs were prepared by Ficoll-Hypaque (Biochrom) density centrifugation and incubated for 15 min at 4°C with M-DC8 hybridoma supernatant containing 10 ␮g/ml Ab. After washing with PBS, 1 ⫻ 108 cells were resuspended in 500 ␮l of PBS and labeled with 10 ␮l of rat antimouse IgM coupled with paramagnetic microbeads (Miltenyi Biotec) for another 15 min at 4°C. After washing, cells were sorted on separation columns (Miltenyi Biotec). PBS containing 1% human serum (CC pro) was used as running and elution buffer. The purity of the isolated M-DC8⫹ DCs was ⬎93% as determined by flow cytometric analysis. Complete medium used for DC culturing and functional assays consisted of RPMI 1640 medium (Biochrom) supplemented with 2 mM L-glutamine, 10 mM sodium pyruvate, 1% nonessential amino acids, 100 ␮g/ml penicillin, 100 ␮g/ml streptomycin (all from Biochrom), and 10% human serum. CD56⫹CD3⫺ NK cells were isolated from freshly prepared PBMCs by depletion of non-NK cells using immunomagnetic separation (Miltenyi Biotec), according to the manufacturer’s instructions. For depletion, PBMCs were indirectly magnetically labeled using a mixture of haptenconjugated anti-CD3, anti-CD14, anti-CD19, anti-CD36, and anti-IgE Abs. Subsequently, cells were incubated with anti-hapten mAb coupled to paramagnetic microbeads. The magnetically labeled cells were depleted by retaining them on a column in the magnetic field. The purity of the isolated CD56⫹CD3⫺ NK cells was ⬎93% as assessed by flow cytometric analysis.

Generation of monocyte-derived DCs Monocytes were isolated by immunomagnetic cell separation with an antiCD14 Ab coupled to paramagnetic microbeads (Miltenyi Biotec), according to the manufacturer’s instructions. To generate immature DCs, monocytes were cultured in the presence of 1000 U/ml GM-CSF and 1000 U/ml IL-4 (both from Strathmann Biotec) in RPMI 1640 medium supplemented with 10% human serum for 5 days. For some experiments, monocytederived DCs were maintained for 6 h in the presence of 200 U/ml IFN-␥ (Strathmann Biotec).

Analysis of DC-mediated tumor cell lysis M-DC8⫹ DCs were cultured for 6 h in the presence or absence of 200 U/ml IFN-␥, washed, and resuspended in complete medium. Cytotoxic activity of M-DC8⫹ DCs was determined against different human tumor or normal cell lines in an 18-h 51Cr release assay. For comparison, IFN-␥- or nonactivated monocyte-derived DCs were used as effector cells. Briefly, target cells (1 ⫻ 106) were labeled with 100 ␮Ci of 51Cr (PerkinElmer Life Sciences) for 1 h at 37°C and then washed four times with PBS. Labeled target cells were plated as triplicates in round-bottom 96-well plates at 5 ⫻ 103/well and were incubated with DCs for 18 h at a DC-to-tumor cell ratio of 40:1. In some assays, cocultivation of DCs and tumor cells was performed in the presence or absence of a neutralizing anti-TNF-␣ Ab or in the presence of an isotype-matched control Ab (both from BD Pharmingen) at a concentration of 10 ␮g/ml. Released 51Cr was determined in a beta counter (PerkinElmer Life Sciences). Maximal and minimal release were measured by treating labeled cells with 2% Triton X-100 (FERAK) or medium alone, respectively. The specific cytotoxicity was calculated according to this formula: percent-specific lysis ⫽ 100 ⫻ ((cpm experimental release ⫺ cpm spontaneous release)/(cpm maximal release ⫺ cpm spontaneous release)).

Evaluation of tumor cell-induced TNF-␣ production in M-DC8⫹ DCs Freshly isolated M-DC8⫹ DCs were cultured for 6 h in the presence of 200 U/ml IFN-␥, washed, and resuspended in complete medium. Subsequently, M-DC8⫹ DCs (2 ⫻ 105 cells/well) were plated alone or together with different human tumor or normal cell lines (1 ⫻ 104 cells/well) in roundbottom 96-well plates. In some assays, M-DC8⫹ DCs were transferred to 24-well plates at 5 ⫻ 105 cells/well and were incubated with the tumor cell lines COLO 205 and MCF-7 (5 ⫻ 104 cells/well). These experiments were performed in the presence or absence of a separating porous membrane. After 18 h, the tumor cell-mediated induction of TNF-␣ in M-DC8⫹ DCs was evaluated by FACS analysis.

Flow cytometric analysis Analysis of surface or intracytoplasmic molecules of M-DC8⫹ DCs and NK cells was performed using the following mAbs: FITC-conjugated anti-

The Journal of Immunology

4129

CD3, PE-conjugated anti-CD56, PE-conjugated anti-TNF-␣, PE-conjugated anti-IL-12, unlabeled anti-TRAIL, unlabeled anti-Fas ligand, unlabeled NKR-P1, and FITC- or PE-conjugated, isotype-specific anti-mouse Abs (all from BD Biosciences Pharmingen). Negative controls included directly labeled or unlabeled isotype-matched irrelevant Abs (BD Pharmingen). M-DC8⫹ hybridoma supernatant was used as described previously (38). For intracytoplasmic TNF-␣ staining, IFN-␥-stimulated M-DC8⫹ DCs were cultured alone or together with human tumor or normal cell lines in the presence of 1 ␮g/ml brefeldin A (Sigma-Aldrich). After 18 h at 37°C and 5% CO2, cells were harvested, fixed with freshly prepared ice-cold 4% paraformaldehyde for 15 min, and permeabilized using 0.1% Saponin in PBS. Subsequently, cells were stained for 15 min using a PE-conjugated anti-TNF-␣ Ab, washed twice, and analyzed by flow cytometry. Direct immunofluorescence staining of cell surface molecules was performed using the relevant mAbs, according to the provider’s instructions. For indirect immunofluorescence staining, cells were incubated with the relevant mAbs for 15 min at 4°C. After washing, FITC- or PE-conjugated, isotype-specific anti-mouse Abs were added for 15 min at 4°C. After the staining procedure, cells were washed twice and evaluated by FACS analysis. Flow cytometry was performed on a FACScan cytometer (BD Pharmingen).

NK cell-mediated, tumor-directed cytotoxicity Freshly isolated M-DC8⫹ DCs were cultured for 6 h in the presence or absence of 200 U/ml IFN-␥, washed, and resuspended in complete medium. Thereafter, M-DC8⫹ DCs were plated in round-bottom 96-well plates at 2 ⫻ 105/well and were incubated with autologous CD56⫹CD3⫺ NK cells (5 ⫻ 105 cells/well). For comparison, IFN-␥- or nonactivated monocyte-derived DCs were used as stimulator cells. In some experiments, 6 ⫻ 105 DCs/well and 1.5 ⫻ 106 NK cells/well were cocultured in 24-well plates in the presence or absence of a separating porous membrane. After 18 h, NK cells were separated from adherent DCs by resuspension, and tumor-directed cytotoxicity was determined. Cytotoxic activity of NK cells was analyzed against the tumor cell lines K-562 and Jurkat in a 4-h 51Crrelease assay. Briefly, tumor cells were labeled as described above. Labeled target cells were plated as triplicates in round-bottom 96-well plates at 5 ⫻ 103/well and incubated with NK cells at different E:T ratios. After 4 h, 51Cr release was determined as mentioned above.

Proliferation assay NK cells (1 ⫻ 105 cells/well) were incubated with increasing numbers of DCs for 4 days in round-bottom 96-well plates. One ␮Ci of [3H]thymidine (Hartmann Analytic) was added to each well for the last 18 h of culture. Cells were harvested, and incorporation was determined in a beta counter.

Cytokine assay DCs (4 ⫻ 104 cells/well) were cocultured with autologous NK cells (1 ⫻ 105 cells/well) in round-bottom 96-well plates. After 48 h, supernatants were collected, and IFN-␥ was quantified using a commercial ELISA kit (BD Pharmingen), according to the manufacturer’s instructions.

Results

Tumor-directed cytotoxicity of M-DC8⫹ DCs To determine whether human M-DC8⫹ DCs exhibit tumoricidal activity, DCs were isolated by immunomagnetic separation from freshly drawn blood of healthy donors. The purity of the isolated M-DC8⫹ DCs was ⬎93%. Other potential effector cells such as NK cells, CTLs, or monocytes represent ⬍1% of the contaminating cells. Following the observation that IFN-␥ activates monocytes and myeloid blood DCs (41), M-DC8⫹ DCs were cultured for 6 h in the presence or absence of IFN-␥. Subsequently, MDC8⫹ DCs were coincubated for an additional 18 h with the tumor cell lines Capan-1, MCF-7, COLO 205, and HT-29, as well as with normal human lung fibroblast cells and endothelial cells. Whereas unstimulated DCs displayed only moderate cytotoxic activity toward the tumor cell targets, M-DC8⫹ DCs stimulated with IFN-␥ lysed each of the tumor cell lines (Fig. 1A). Only a marginal cytotoxic effect was seen when normal human cells such as lung fibroblasts or endothelial cells were used as targets (Fig. 1A). In additional experiments, M-DC8⫹ DCs were compared with autologous monocyte-derived DCs that are regarded as mediators of tumor-directed cytotoxicity (12–16). The cytotoxic potential of MDC8⫹ DCs was similar to that observed with monocyte-derived

FIGURE 1. M-DC8⫹ DC-mediated cytotoxicity toward different tumor cell lines. A, Freshly isolated M-DC8⫹ DCs were maintained for 6 h in the presence or absence of IFN-␥. Subsequently, M-DC8⫹ DCs were cocultured in round-bottom 96-well plates with the 51Cr-labeled tumor cell lines Capan-1, MCF-7, COLO 205, and HT-29, as well as the normal human cell lines NHLF and HUVEC (5 ⫻ 103 cells/well) at an E:T ratio of 40:1. After 18 h of incubation, chromium release was measured. The results of three different donors are presented as mean ⫾ SE of triplicate wells. B, 51Crlabeled tumor cells were incubated either with M-DC8⫹ DCs or monocytederived DCs at an E:T ratio of 40:1. After 18 h of incubation, chromium release was determined. The results of three different donors are presented as mean ⫾ SE of triplicate wells.

DCs (Fig. 1B). Stimulation of monocyte-derived DCs with IFN-␥ did not influence the cytotoxic activity toward the different tumor cell lines (data not shown). These results indicate that IFN-␥-stimulated human blood M-DC8⫹ DCs exert tumor-directed cytotoxicity, whereas normal cells are not lysed significantly. Tumor cells efficiently enhance TNF-␣ production of M-DC8⫹ DCs In previous studies, we demonstrated that LPS-stimulated M-DC8⫹ DCs produced large amounts of the cytotoxic effector molecule TNF-␣, which by far exceeded that of other myeloid

4130

TUMORICIDAL POTENTIAL OF BLOOD DCs

FIGURE 2. Tumor cell-mediated enhancement of TNF-␣ production by M-DC8⫹ DCs. Histograms demonstrate intracellular TNF-␣ expression of IFN-␥-stimulated M-DC8⫹ DCs (2 ⫻ 105 cells/well) cultured alone (A) or in the presence of the tumor cell lines Capan-1 (B), MCF-7 (C), COLO 205 (D), and HT-29 (E), as well as the normal human cell lines NHLF (F) and HUVEC (G) (1 ⫻ 104 cells/well) in round-bottom 96-well plates. After 18 h, intracellular TNF-␣ was determined by FACS analysis. The results of one of three representative donors performed with similar results are depicted. Values represent the mean fluorescent intensity (MFI) of cells staining positive for TNF-␣ (filled histogram) compared with the respective isotype control (open histogram).

blood DCs and monocytes (38). Following these findings, we analyzed whether the different tumor cell lines are able to induce TNF-␣ in M-DC8⫹ DCs and whether TNF-␣ contributes to the observed tumoricidal potential of this human blood DC subset. Therefore, freshly isolated M-DC8⫹ DCs were incubated for 6 h with IFN-␥ and subsequently cocultured for an additional 18 h with four different tumor cell lines and two normal cell lines. Remarkably, all tumor cell lines clearly augmented TNF-␣ production of IFN-␥-stimulated M-DC8⫹ DCs during cocultivation (Fig. 2). In contrast, only a marginal improvement of TNF-␣ production was detected during cocultivation of M-DC8⫹ DCs with the normal human cell lines NHLF or HUVEC. To evaluate the mechanisms underlying the tumor cell-mediated enhancement of TNF-␣ production, IFN-␥-activated DCs were cocultured with the tumor cell lines MCF-7 and COLO 205 in the presence or absence of a separating porous membrane. As dem-

onstrated in Fig. 3, the membrane clearly inhibited the tumor cellmediated improvement of TNF-␣ in M-DC8⫹ DCs, indicating that this ability is critically dependent on cell-to-cell contact. TNF-␣ efficiently contributes to the tumoricidal ability of M-DC8⫹ DCs To investigate the contribution of TNF-␣ to the tumoricidal potential of M-DC8⫹ DCs, cytotoxicity against the different tumor cell lines was determined in the presence of a neutralizing antiTNF-␣ Ab. Although the cytotoxic effect toward the tumor cell line Capan-1 was not reduced significantly by the anti-TNF-␣ Ab, the M-DC8⫹ DC- mediated killing of the tumor cell lines COLO 205, HT-29, and MCF-7 was almost completely abrogated. These data illustrate that TNF-␣ plays an important role in tumor-directed cytotoxicity mediated by M-DC8⫹ DCs (Fig. 4).

FIGURE 3. Contact-dependent, tumor cell-mediated enhancement of TNF-␣ production by M-DC8⫹ DCs. Histograms show intracellular TNF-␣ expression of IFN-␥-stimulated M-DC8⫹ DCs (5 ⫻ 105 cells/well) incubated alone (A) or with the tumor cell line MCF-7 (5 ⫻ 104 cells/well) in the absence (B) or presence (C) of a separating membrane or with the tumor cell line COLO 205 (5 ⫻ 104 cells/well) without (D) or with (E) a separating membrane in 24-well plates. After 18 h, intracellular TNF-␣ expression was determined by FACS analysis. The results of one of three representative donors performed with similar results are depicted. Values represent the MFI of cells staining positive for TNF-␣ (filled histogram) compared with the respective isotype control (open histogram).

The Journal of Immunology

4131

FIGURE 4. Contribution of TNF-␣ to tumor-directed cytotoxicity mediated by M-DC8⫹ DCs. IFN-␥-activated M-DC8⫹ DCs were cocultured in round-bottom 96-well plates with the 51Cr-labeled tumor cell lines Capan-1, MCF-7, COLO 205, and HT-29 at an E:T ratio of 40:1. Cocultivation was performed in the presence of a neutralizing anti-TNF-␣ Ab or an isotype-matched control Ab each at a concentration of 10 ␮g/ml or in the absence of Ab. After 18 h of incubation, chromium release was determined. The results of three different donors are presented as mean ⫾ SE of triplicate wells.

M-DC8⫹ DCs stimulate proliferation and IFN-␥ secretion of NK cells To evaluate the capacity of M-DC8⫹ DCs to stimulate NK cell proliferation and IFN-␥ production, M-DC8⫹ DCs were isolated from the blood of healthy donors and were cultured for 6 h in the presence or absence of IFN-␥. Thereafter, DCs were coincubated with autologous CD56⫹CD3⫺ NK cells, which were purified by depletion of non-NK cells using immunomagnetic separation. For comparison, monocyte-derived DCs were used as stimulator cells. As demonstrated in Fig. 5, both DC populations promote proliferation and IFN-␥ secretion of NK cells. Although monocyte-derived DCs were superior in stimulating NK cell proliferation (Fig. 5A), M-DC8⫹ DCs were particularly potent in inducing IFN-␥ production by NK cells (Fig. 5B). Preincubation with IFN-␥ for 6 h did not significantly influence the potential of both DC subsets to promote NK cell proliferation or IFN-␥ production. M-DC8⫹ DCs improve the tumoricidal potential of NK cells To determine whether M-DC8⫹ DCs enhance the tumoricidal activity of human CD56⫹CD3⫺ NK cells, freshly isolated M-DC8⫹ DCs were cultured for 6 h in the presence or absence of IFN-␥. Subsequently, DCs were coincubated with autologous purified CD56⫹CD3⫺ NK cells. In some experiments, monocyte-derived DCs were used as stimulator cells for comparison. After 18 h of cocultivation, NK cells were separated from adherent DCs by resuspension, and their cytotoxic potential toward the tumor cell lines K-562 and Jurkat was evaluated in a 4-h 51Cr release assay. Interestingly, whereas unstimulated DCs exhibited only little improvement of tumor-directed cytotoxicity by NK cells, IFN-␥stimulated M-DC8⫹ DCs markedly enhanced NK cell-mediated killing activity (Fig. 6, A and B). The M-DC8⫹ DC-mediated improvement of NK cell cytotoxicity was similar to that observed with autologous monocyte-derived DCs (Fig. 6C). To exclude a contribution of contaminating M-DC8⫹ DCs, which represent ⬍5% of the harvested NK cells to the cytotoxic effect, we determined the ability of these DCs to kill the tumor cell lines K-562 and Jurkat in a 4-h 51Cr release assay. M-DC8⫹ DCs exhibited only a marginal cytotoxic effect against these target cell lines even at an E:T ratio of 40:1 (data not shown).

FIGURE 5. M-DC8⫹ DCs stimulate proliferation and IFN-␥ secretion of NK cells. Freshly isolated M-DC8⫹ DCs and monocyte-derived DCs were maintained for 6 h in the presence or absence of IFN-␥. A, Both DC subsets were coincubated with NK cells (1 ⫻ 105 cells/well) in roundbottom 96-well plates. After 4 days, the proliferation of NK cells was determined. The results of one of three representative donors performed with similar results are depicted. Values of NK cell proliferation in the absence of DCs were subtracted from the total cpm. Values represent the mean of triplicate samples. B, M-DC8⫹ DCs or monocyte-derived DCs (4 ⫻ 104 cells/well) were coincubated with NK cells (1 ⫻ 105 cells/well) in round-bottom 96-well plates. After 48 h, supernatants were collected, and IFN-␥ concentration was analyzed by ELISA. The results of one of three representative donors performed with similar results are depicted. Values represent the mean ⫾ SE of duplicate samples.

To evaluate whether this DC-mediated antitumor effect was dependent on cell surface molecules or soluble factors, IFN-␥-activated DCs were cocultured with autologous NK cells in the presence or absence of a separating porous membrane. As shown in Fig. 7, the membrane clearly abrogated the ability of M-DC8⫹ DCs to stimulate NK cell activity, indicating that cell-to-cell contact is essential for the enhancement of NK cell-mediated, tumordirected cytotoxicity by M-DC8⫹ DCs.

4132

TUMORICIDAL POTENTIAL OF BLOOD DCs

FIGURE 7. Contact-dependent stimulation of tumor-directed, NK cellmediated cytotoxicity by M-DC8⫹ DCs. IFN-␥-stimulated M-DC8⫹ DCs (6 ⫻105 cells/well) were coincubated with purified CD56⫹CD3⫺ NK cells (1.5 ⫻ 106 cells/well) in 24-well plates in the presence or absence of a separating porous membrane. After 18 h, NK cells were harvested and were cocultured in round-bottom 96-well plates with the 51Cr-labeled tumor cell lines K562 and Jurkat (5 ⫻ 103 cells/well) at an E:T ratio of 20:1. After 4 h of incubation, chromium release was measured. The results of three different donors are presented as mean ⫾ SE of triplicate wells.

Discussion

FIGURE 6. Enhancement of tumor-directed NK cell-mediated cytotoxicity by M-DC8⫹ DCs. Unstimulated or IFN-␥-activated M-DC8⫹ DCs (2 ⫻ 105 cells/well) were coincubated with purified CD56⫹CD3⫺ NK cells (5 ⫻ 105 cells/well) in round-bottom 96-well plates. After 18 h, NK cells were separated from adherent DCs and were cocultured in round-bottom 96-well plates with the 51Cr-labeled tumor cell lines K562 (A) and Jurkat (B) (5 ⫻ 103 cells/well) at different E:T ratios. After 4 h of incubation, chromium release was measured. The results of three different donors are presented as mean ⫾ SE of triplicate wells. C, NK cells stimulated by M-DC8⫹ DCs or monocyte-derived DCs were coincubated with the tumor cell lines K562 and Jurkat at an E:T ratio of 20:1. After 4 h of incubation, chromium release was measured. The results of three different donors are presented as mean ⫾ SE of triplicate wells.

In the present study, we demonstrated that IFN-␥-stimulated MDC8⫹ DCs, which represent a major subpopulation of human blood DCs, exhibited cytotoxic activity toward several tumor cell lines. In contrast, normal human cells such as lung fibroblasts or endothelial cells were not lysed efficiently. When evaluating the effector molecules underlying the tumor-directed cytotoxicity mediated by M-DC8⫹ DCs, we focused on TNF-␣, which was previously shown to be produced in large amounts by activated MDC8⫹ DCs (38) and which is regarded as an important effector molecule of tumor cell killing (42, 43). We provide for the first time evidence that different tumor cell lines clearly augmented TNF-␣ production of IFN-␥-stimulated M-DC8⫹ DCs, which was critically dependent on cell-to-cell contact. In contrast, only a marginal enhancement of TNF-␣ production was detected when MDC8⫹ DCs were cocultivated with human fibroblasts or endothelial cells. This suggests that tumor-associated surface molecules are important for the observed increase of TNF-␣ production in M-DC8⫹ DCs. Furthermore, we illustrated the essential contribution of TNF-␣ to the cytotoxic effect of M-DC8⫹ DCs toward susceptible tumor cell lines. These results were in agreement with other reports demonstrating that TNF-␣ was responsible for the antitumor reactivity of monocyte-derived DCs (11, 14). When evaluating potential effector molecules that contribute to the MDC8⫹ DC-mediated killing of the tumor cell line Capan-1, we were not able to detect TRAIL, Fas ligand, or NKR-P1 on the surface of IFN-␥-stimulated M-DC8⫹ DCs by FACS analysis (data not shown). Although several reports have documented that human monocyte-derived DCs exhibit direct tumor cell killing (12–16), studies investigating the tumoricidal potential of native human blood DCs are limited. Fanger et al. (41) reported on the ability of IFN-␣ or -␥-stimulated CD11c⫹CD123dim myeloid blood DCs to specifically kill various TRAIL-sensitive cancer cells

The Journal of Immunology but not TRAIL-resistant tumor cells or normal cells, indicating that TRAIL may serve as an important effector molecule on this DC population. In contrast to CD11c⫹CD123dim blood DCs, IFNstimulated CD11c⫺CD123high plasmacytoid DCs exerted only minimal tumoricidal activity (41). Another report indicated that HLA-DR⫹CD4⫹ blood DCs, which were not further subdivided into CD11c⫹CD123dim myeloid DCs and CD11c⫺CD123high plasmacytoid DCs, induced apoptosis in susceptible cancer cells, which has been associated with the expression of the TNF family ligands lymphotoxin-a1b2, TNF, Fas ligand, and TRAIL (16, 17). Taken together, these findings reveal that among the different human blood DC subsets, myeloid DC subsets such as M-DC8⫹ DCs are able to exert tumor cell killing, whereas plasmacytoid DCs exhibit only marginal tumoricidal activity. Recent studies demonstrated the ability of human monocytederived DCs to efficiently promote NK cell proliferation, IFN-␥secretion, and cytotoxicity (25–27). More recently, Munz et al. (44) illustrated that various myeloid DC subsets contribute to different properties of NK cells. Thus, it was shown that monocytederived DCs and, to a lesser extent, CD34⫹-hemopoietic progenitor cell-derived dermal-interstitial DCs, markedly induce NK cell proliferation as well as NK cell-mediated cytotoxicity. In contrast, Langerhans cells were only poor mediators of NK cell proliferation and cytotoxicity (44). However, the role of human blood DCs as activators of NK cells is less well determined. In the present study, we analyzed the capacity of M-DC8⫹ DCs to stimulate proliferation and IFN-␥ secretion of NK cells in comparison to immature monocyte-derived DCs. We found that monocyte-derived DCs were superior in inducing NK cell proliferation, whereas M-DC8⫹ DCs efficiently improved IFN-␥ production, which by far exceeded that observed with monocyte-derived DCs. These data support the observation that DC subsets vary in their ability to stimulate different properties of NK cells. In additional experiments, we illustrated that IFN-␥-stimulated M-DC8⫹ DCs displayed potent improvement of NK cell-mediated killing activity, which was similar to that observed with monocytederived DCs. This is in line with two other reports demonstrating that CD11c⫹ myeloid blood DCs, as well as CD83⫹ human blood DCs, which were not further phenotypically characterized, efficiently augmented the cytotoxic potential of NK cells (25, 45). When evaluating the mechanisms underlying the enhancement of NK cell cytotoxicity by M-DC8⫹ DCs, we found that this effect was clearly abrogated in the presence of a separating membrane. In addition, intracellular IL-12 expression could not be detected in M-DC8⫹ DCs after stimulation with IFN-␥ or cocultivation with NK cells as determined by FACS analysis (data not shown). These data suggest that membrane-associated determinants were essential for the augmentation of NK cell-mediated antitumor effects. This finding obtained with M-DC8⫹ blood DCs is in accordance with several other reports demonstrating that the activation of NK cells by in vitro-differentiated DCs was critically dependent on cell-to-cell contact (23, 24, 26, 28, 29, 46). However, the surface molecules, which are involved in DC-NK interaction, remain to be determined. Additional data revealed that DC-derived soluble factors such as IL-12 (24, 26, 46, 47) and IL-18 (24), as well as so-far unknown soluble molecules (25), may also play an important role. The presence of DCs in various human carcinomas has been documented in several studies (48 –50). Moreover, it has been reported that the density of tumor-infiltrating DCs was associated with better prognosis, reduced tumor recurrence, and fewer metastases (51–54). These findings may be explained by a cross-talk between DCs and NK cells. Tumor-infiltrating NK cells produce IFN-␥, which leads to a stimulation of DCs. Activated DCs exhibit direct tumor cell killing and an improvement of NK cell-mediated

4133 cytotoxicity, which is dependent on cell-to-cell contact. In addition, the uptake of tumor-derived material, the efficient processing and presentation of tumor-associated Ags by DCs, and subsequent induction of tumor-reactive T cells in the lymph nodes may also contribute to the positive correlation between high rates of tumorinfiltrating DCs and better prognosis. In summary, our results emphasize the ability of a major subpopulation of native human blood DCs to mediate tumor-directed cytotoxicity, whereas normal cells are not significantly lysed. We documented for the first time the ability of tumor cells to induce large amounts of TNF-␣ in M-DC8⫹ DCs, which essentially contributes to the cytotoxic effect of M-DC8⫹ DCs toward susceptible tumor cell lines. In addition, we found that this subset of blood DCs promotes proliferation, IFN-␥ production, and tumor-directed cytotoxicity of NK cells. Taken together, these data revealed that in addition to their predominant role as inducers of tumor-reactive T cells, human blood DCs can also serve as potent effector and regulator cells of innate antitumor immunity.

Acknowledgments The technical assistance of Karin Gu¨nther, Ba¨rbel Lo¨bel, Livia Schulz, and Verona Schwarze is greatly appreciated.

Disclosures The authors have no financial conflict of interest.

References 1. Hart, D. N. 1997. Dendritic cells: unique leukocyte populations which control the primary immune response. Blood 90:3245. 2. Banchereau, J., and R. M. Steinman. 1998. Dendritic cells and the control of immunity. Nature 392:245. 3. Banchereau, J., F. Briere, C. Caux, J. Davoust, S. Lebecque, Y. J. Liu, B. Pulendran, and K. Palucka. 2000. Immunobiology of dendritic cells. Annu. Rev. Immunol. 18:767. 4. Fong, L., and E. G. Engleman. 2000. Dendritic cells in cancer immunotherapy. Annu. Rev. Immunol. 18:245. 5. Mayordomo, J. I., T. Zorina, W. J. Storkus, L. Zitvogel, C. Celuzzi, L. D. Falo, C. J. Melief, S. T. Idstad, W. M. Kast, A. B. Deleo, and M. T. Lotze. 1995. Bone marrow-derived dendritic cells pulsed with synthetic tumour peptides elicit protective and therapeutic antitumour immunity. Nat. Med. 1:1297. 6. Nair, S. K., A. Heiser, D. Boczkowski, A. Majumdar, M. Naoe, J. S. Lebkowski, J. Vieweg, and E. Gilboa. 2000. Induction of cytotoxic T cell responses and tumor immunity against unrelated tumors using telomerase reverse transcriptase RNA transfected dendritic cells. Nat. Med. 6:1011. 7. Hsu, F. J., C. Benike, F. Fagnoni, T. M. Liles, D. Czerwinski, B. Taidi, E. G. Engleman, and R. Levy. 1996. Vaccination of patients with B-cell lymphoma using autologous antigen-pulsed dendritic cells. Nat. Med. 2:52. 8. Nestle, F. O., S. Alijagic, M. Gilliet, Y. Sun, S. Grabbe, R. Dummer, G. Burg, and D. Schadendorf. 1998. Vaccination of melanoma patients with peptide- or tumor lysate-pulsed dendritic cells. Nat. Med. 4:328. 9. Thurner, B., I. Haendle, C. Roder, D. Dieckmann, P. Keikavoussi, H. Jonuleit, A. Bender, C. Maczek, D. Schreiner, P. von den Driesch, et al. 1999. Vaccination with mage-3A1 peptide-pulsed mature, monocyte-derived dendritic cells expands specific cytotoxic T cells and induces regression of some metastases in advanced stage IV melanoma. J. Exp. Med. 190:1669. 10. Josien, R., M. Heslan, J. P. Soulillou, and M. C. Cuturi. 1997. Rat spleen dendritic cells express natural killer cell receptor protein 1 (NKR-P1) and have cytotoxic activity to select targets via a Ca2⫹-dependent mechanism. J. Exp. Med. 186:467. 11. Chapoval, A. I., K. Tamada, and L. Chen. 2000. In vitro growth inhibition of a broad spectrum of tumor cell lines by activated human dendritic cells. Blood 95:2346. 12. Vanderheyde, N., E. Aksoy, Z. Amraoui, P. Vandenabeele, M. Goldman, and F. Willems. 2001. Tumoricidal activity of monocyte-derived dendritic cells: evidence for a caspase-8-dependent, Fas-associated death domain-independent mechanism. J. Immunol. 167:3565. 13. Yang, R., D. Xu, A. Zhang, and A. Gruber. 2001. Immature dendritic cells kill ovarian carcinoma cells by a FAS/FASL pathway, enabling them to sensitize tumor-specific CTLs. Int. J. Cancer 94:407. 14. Manna, P. P., and T. Mohanakumar. 2002. Human dendritic cell mediated cytotoxicity against breast carcinoma cells in vitro. J. Leukocyte Biol. 72:312. 15. Joo, H. G., T. P. Fleming, Y. Tanaka, T. J. Dunn, D. C. Linehan, P. S. Goedegebuure, and T. J. Eberlein. 2002. Human dendritic cells induce tumor-specific apoptosis by soluble factors. Int. J. Cancer 102:20. 16. Janjic, B. M., G. Lu, A. Pimenov, T. L. Whiteside, W. J. Storkus, and N. L. Vujanovic. 2002. Innate direct anticancer effector function of human immature dendritic cells. I. Involvement of an apoptosis-inducing pathway. J. Immunol. 168:1823.

4134 17. Lu, G., B. M. Janjic, J. Janjic, T. L. Whiteside, W. J. Storkus, and N. L. Vujanovic. 2002. Innate direct anticancer effector function of human immature dendritic cells. II. Role of TNF, lymphotoxin ␣1␤2, Fas ligand, and TNFrelated apoptosis-inducing ligand. J. Immunol. 168:1831. 18. Moretta, A. 2002. Natural killer cells and dendritic cells: rendezvous in abused tissues. Nat. Rev. Immunol. 2:957. 19. Moretta, A., C. Bottino, M. Vitale, D. Pende, C. Cantoni, M. C. Mingari, R. Biassoni, and L. Moretta. 2001. Activating receptors and coreceptors involved in human natural killer cell-mediated cytolysis. Annu. Rev. Immunol. 19:197. 20. Farag, S. S., T. A. Fehninger, L. Ruggeri, A. Velardi, and M. A. Caligiuri. 2002. Natural killer cell receptors: new biology and insights into the graft-versus-leukemia effect. Blood 100:1935. 21. Raulet, D. H. 2004. Interplay of natural killer cells and their receptors with the adaptive immune response. Nat. Immunol. 5:996. 22. Moretta, L., C. Bottino, D. Pende, M. Vitale, M. C. Mingari, and A. Moretta. 2004. Different checkpoints in human NK-cell activation. Trends Immunol. 25:670. 23. Fernandez, N. C., A. Lozier, C. Flament, P. Ricciardi-Castagnoli, D. Bellet, M. Suter, M. Perricaudet, T. Tursz, E. Maraskovsky, and L. Zitvogel. 1999. Dendritic cells directly trigger NK cell functions: cross-talk relevant in innate antitumor immune responses in vivo. Nat. Med. 5:405. 24. Yu, Y., M. Hagihara, K. Ando, B. Gansuvd, H. Matsuzawa, T. Tsuchiya, Y. Ueda, H. Inoue, T. Hotta, and S. Kato. 2001. Enhancement of human cord blood CD34⫹ cell-derived NK cell cytotoxicity by dendritic cells. J. Immunol. 166:1590. 25. Nishioka, Y., N. Nishimura, Y. Suzuki, and S. Sone. 2001. Human monocytederived and CD83⫹ blood dendritic cells enhance NK cell-mediated cytotoxicity. Eur. J. Immunol. 31:2633. 26. Amakata, Y., Y. Fujiyama, A. Andoh, K. Hodohara, and T. Bamba. 2001. Mechanism of NK cell activation induced by coculture with dendritic cells derived from peripheral blood monocytes. Clin. Exp. Immunol. 124:214. 27. Ferlazzo, G., M. L. Tsang, L. Moretta, G. Melioli, R. M. Steinman, and C. Munz. 2002. Human dendritic cells activate resting natural killer (NK) cells and are recognized via the NKp30 receptor by activated NK cells. J. Exp. Med. 195:343. 28. Piccioli, D., S. Sbrana, E. Melandri, and N. M. Valiante. 2002. Contact-dependent stimulation and inhibition of dendritic cells by natural killer cells. J. Exp. Med. 195:335. 29. Gerosa, F., B. Baldani-Guerra, C. Nisii, V. Marchesini, G. Carra, and G. Trinchieri. 2002. Reciprocal activating interaction between natural killer cells and dendritic cells. J. Exp. Med. 195:327. 30. Rissoan, M. C., V. Soumelis, N. Kadowaki, G. Grouard, F. Briere, R. de Waal Malefyt, and Y. J. Liu. 1999. Reciprocal control of T helper cell and dendritic cell differentiation. Science 283:1183. 31. Siegal, F. P., N. Kadowaki, M. Shodell, P. A. Fitzgerald-Bocarsly, K. Shah, S. Ho, S. Antonenko, and Y. J. Liu. 1999. The nature of the principal type 1-interferon-producing cells in human blood. Science 284:1835. 32. Cella, M., D. Jarrossay, F. Facchetti, O. Alebardi, H. Nakajima, A. Lanzavecchia, and M. Colonna. 1999. Plasmacytoid monocytes migrate to inflamed lymph nodes and produce large amounts of type I interferon. Nat. Med. 5:919. 33. Liu, Y. J. 2001. Dendritic cell subsets and lineages, and their functions in innate and adaptive immunity. Cell 106:259. 34. Shortman, K., and Y. J. Liu. 2002. Mouse and human dendritic cell subtypes. Nat. Rev. Immunol. 2:151. 35. MacDonald, K. P. A., D. J. Munster, G. J. Clark, A. Dzionek, J. Schmitz, and D. N. J. Hart. 2002. Characterization of human blood dendritic cell subsets. Blood 100:4512. 36. Randolph, G. J., G. Sanchez-Schmitz, R. M. Liebman, and K. Scha¨kel. 2002. The CD16⫹ (Fc␥RIII⫹) subset of human monocytes preferentially becomes migratory dendritic cells in a model tissue setting. J. Exp. Med. 196:517.

TUMORICIDAL POTENTIAL OF BLOOD DCs 37. Scha¨kel, K., E. Mayer, C. Federle, M. Schmitz, G. Riethmu¨ller, and E. P. Rieber. 1998. A novel dendritic cell population in human blood: one-step immunomagnetic isolation by a specific mAb (M-DC8) and in vitro priming of cytotoxic T lymphocytes. Eur. J. Immunol. 28:4084. 38. Scha¨kel, K., R. Kannagi, B. Kniep, Y. Goto, C. Mitsuoka, J. Zwirner, A. Soruri, M. von Kietzell, and E. P. Rieber. 2002. 6-Sulfo LacNAc, a novel carbohydrate modification of PSGL-1, defines an inflammatory type of human dendritic cells. Immunity 17:289. 39. Schmitz, M., S. Zhao, K. Scha¨kel, M. Bornha¨user, D. Ockert, and E. P. Rieber. 2002. Native human blood dendritic cells as potent effectors in antibody-dependent cellular cytotoxicity. Blood 100:1502. 40. Mattig, S., R. Knoefler, and A. Deussen. 2003. Modulation of adenine nucleotide concentrations in human plasma by erythrocytes and endothelial cells. Thromb. Res. 110:195. 41. Fanger, N. A., C. R. Maliszewksi, K. Schooley, and T. S. Griffith. 1999. Human dendritic cells mediate cellular apoptosis via tumor necrosis factor-related apoptosis-inducing ligand (TRAIL). J. Exp. Med. 190:1155. 42. Feinman, R., D. Hendriksen-DeStefano, M. Tsujimoto, and J. Vilcek. 1987. Tumor necrosis factor is an important mediator of tumor cell killing by human monocytes. J. Immunol. 138:635. 43. Pullyblank, A. M., P. J. Guillou, and J. R. T. Monson. 1995. Interleukin 1 and tumour necrosis factor ␣ may be responsible for the lytic mechanism during anti-tumour antibody-dependent cell-mediated cytotoxicity. Br. J. Cancer 72:601. 44. Munz, C., T. Dao, G. Ferlazzo, M. A. de Cos, K. Goodman, and J. W. Young. 2005. Mature myeloid dendritic cell subsets have distinct roles for activation and viability of circulating human natural killer cells. Blood 105:266. 45. Osada, T., H. Nagawa, J. Kitayama, N. H. Tsuno, S. Ishihara, M. Takamizawa, and Y. Shibata. 2001. Peripheral blood dendritic cells, but not monocyte-derived dendritic cells, can augment human NK cell function. Cell. Immunol. 213:14. 46. Borg, C., A. Jalil, D. Laderach, K. Maruyama, H. Wakasugi, S. Charrier, B. Ryffel, A. Cambi, C. Figdor, W. Vainchenker, et al. 2004. NK cell activation by dendritic cells (DCs) requires the formation of a synapse leading to IL-12 polarization in DCs. Blood 104:3267. 47. Alli, R. S., and A. Khar. 2004. Interleukin-12 secreted by mature dendritic cells mediates activation of NK cell function. FEBS Lett. 559:71. 48. Matsuda, H., M. Mori, S. Tsujitani, S. Ohno, H. Kuwano, and K. Sugimachi. 1990. Immunohistochemical evaluation of squamous cell carcinoma antigen and S-100 protein-positive cells in human malignant esophageal tissues. Cancer 65:2261. 49. Thurnher, M., C. Radmayr, R. Ramoner, S. Ebner, G. Bock, H. Klocker, N. Romani, and G. Bartsch. 1996. Human renal-cell carcinoma tissue contains dendritic cells. Int. J. Cancer 68:1. 50. Bell, D., P. Chomarat, D. Broyles, G. Netto, G. M. Harb, S. Lebecque, J. Valladeau, J. Davoust, K. A. Palucka, and J. Banchereau. 1999. In breast carcinoma tissue, immature dendritic cells reside within the tumor, whereas mature dendritic cells are located in peritumoral areas. J. Exp. Med. 190:1417. 51. Tsujitani, S., Y. Kakeji, A. Watanabe, S. Kohnoe, Y. Maehara, and K. Sugimachi. 1990. Infiltration of dendritic cells in relation to tumor invasion and lymph node metastasis in human gastric cancer. Cancer 66:2012. 52. Zeid, N. A., and H. K. Muller. 1993. S100-positive dendritic cells in human lung tumors associated with cell differentiation and enhanced survival. Pathology 25:338. 53. Ishigami, S., T. Aikou, S. Natsugoe, S. Hokita, H. Iwashige, M. Tokushige, and S. Sonoda. 1998. Prognostic value of HLA-DR expression and dendritic cell infiltration in gastric cancer. Oncology 55:65. 54. Iwamoto, M., H. Shinohara, A. Miyamoto, M. Okuzawa, H. Mabuchi, T. Hohara, G. Gon, M. Toyoda, and N. Tanigawa. 2003. Prognostic value of tumor-infiltrating dendritic cells expressing CD83 in human breast carcinomas. Int. J. Cancer 104:92.