Tumor vaccines using dendritic cells (DCs) have been shown to induce antitumor CTL responses. The choice of the tumor antigen preparation used for DC loading is still an unresolved issue. We compared DCs pulsed with cell lysates, whole apoptotic tumor cells or their supernatants of the HLA-A2+ human pancreatic carcinoma cell line Panc-1 for their capacity to activate T cells. Monocyte-derived DCs from HLA-A2+ donors were pulsed with tumor antigen, matured subsequently, and cocultured with autologeous peripheral blood mononuclear cells. After three weekly restimulations with DCs, T-cell activation was assessed by intracellular IFN-γ staining and cytotoxicity assays. Compared with lysate, pulsing DCs with the supernatant of apoptotic tumor cells induced a higher frequency of activated CTLs and T-helper cells, as well as an enhanced MHC class I-restricted tumor cell lysis. No activation of natural killer (NK) or γδ T cells was detected. Pulsing DCs with whole apoptotic tumor cells induced an even more pronounced lytic effect. However, in this case, MHC class-I blocking was only partially effective, and unrelated cell lines were also killed. IFN-γ staining revealed activation of CTLs and T-helper cells, as well as NK and γδ T cells. Trans-well cultures of NK cells, apoptotic tumor cells, and DCs showed that NK cell activation was dependent on direct cell-to-cell contact with tumor cells and the presence of interleukin-12 produced by DCs. These results indicate that the choice of antigen preparation is a critical determinant in the induction of antitumor immunity. Tumor vaccines consisting of DCs and apoptotic tumor cells may be able to activate CTLs, as well as effector cells of the innate immune system.

Tumor vaccines aim at inducing CTL responses against tumor antigens to mount an immune response against tumors. The activation of CTL requires presentation of the antigen in the context of MHC molecules on the surface of antigen-presenting cells. DCs4 are the most effective antigen-presenting cells (1), and tumor vaccinations with DCs have been shown to induce CTL responses and tumor regression in some patients (2, 3, 4, 5, 6). In contrast to CTL, NK cells and γδ T cells, effector cells of the innate immune system, have the capacity to recognize and kill tumor cells in an antigen-independent manner. There is evidence that DC can link innate and acquired effector mechanisms (1) either by direct cell-to-cell contacts (7) or by secreting cytokines (8, 9). Activation of effector cells of the innate immune system by DC might be a prerequisite for antitumor immunity in MHC class-I-negative tumors (7) and, under certain circumstances, for successful CTL responses (10).

Compared with tumors, such as renal cell carcinoma and malignant melanoma, pancreatic carcinoma is considered to be weakly immunogenic. In addition, pancreatic cancer cells exhibit several mechanisms of immune evasion, such as the expression of the apoptosis-inducing molecule Fas-ligand (11), defective signaling via Fas (12), and secretion of transforming growth factor-β, a growth factor known to interfere with DC function (13). Nevertheless, as we have shown previously, CTL responses against pancreatic cancer cells can be induced in vitro using DCs (14). Therefore, vaccination with DCs might offer a therapeutic option for patients with pancreatic carcinoma.

A critical issue in optimizing DC vaccines is the choice of tumor antigen for DC loading. Clinical vaccination trials for patients with malignant melanoma have demonstrated that vaccinating against a single antigen can induce tumor-specific CTLs but carries the risk of promoting tumor antigen escape variants (6). A more recent trial showed that the generation of CTLs against three or more tumor antigens correlates with clinical response (15). However, for many tumors, no or only few antigenic epitopes are known. To circumvent this limitation, whole tumor cells containing a spectrum of known, as well as yet unknown antigens, might be used as an antigen source. The additional presence of epitopes for T-helper lymphocytes could be beneficial because MHC II-restricted activation of T-helper lymphocytes plays a pivotal role in the physiological immune response to pathogens and might be of considerable importance in the process of tumor rejection (16).

Effective cross-priming with antigens from tumor cells has been demonstrated with tumor cell lysates (17), apoptotic tumor cells (18, 19, 20, 21, 22), and released particles from tumor cells, such as apoptotic bodies (23) and tumor-derived exosomes (24, 25). In the present study, we compared DCs pulsed with cell lysates, apoptotic cells, or released particles from apoptotic cells obtained from pancreatic carcinoma cells for their ability to cross-prime CTLs, as well as to activate T-helper and effector cells of the innate immune system.

Media and Reagents.

RPMI 1640 (Biochrom, Berlin, Germany) supplemented with 2% human serum (BioWhittaker, Walkersville, MD), 2 mm L-glutamine (Life Technologies, Inc., Paisley, Scotland), 50 units/ml penicillin, and 50 μg/ml streptomycin (Sigma Chemical Co., Munich, Germany) was used to culture cells from PBMCs. Recombinant human cytokines: granulocyte macrophage colony-stimulating factor was purchased from Novartis (Basel, Switzerland), IL-4 from Promega (Madison, WI), IL-2 and IL-7 from Strathman Biotech (Hanover, Germany), and TNF-α from R&D Systems (Wiesbaden, Germany). Prostaglandin E2, CFSE, fluorochrome-labeled PKH-26, PI, and brefeldin A were obtained from Sigma Chemical Co. [3H]thymidine was purchased from Amersham Buchler (Freiburg, Germany) and Na2[51Cr]O4 from NEN Life Sciences (Zavantem, Belgium).

Isolation and Culture of Cells.

PBMCs were obtained from peripheral blood or buffy coats of healthy HLA-A2+ blood donors by Ficoll-Hypaque density gradient centrifugation. Monocyte-derived DCs were generated from the adherent fraction of PBMCs cultured in the presence of granulocyte macrophage colony-stimulating factor (1000 units/ml) and IL-4 (500 units/ml) for 6 days as described (14). NK cells were enriched from PBMCs by magnetically activated cell sorting using the isolation kit for untouched NK cells from Miltenyi Biotec (Bergisch Gladbach, Germany) according to the manufacturer’s protocol. The purity of isolated CD3 CD56+ NK cells was >95%. The human pancreatic carcinoma cell line Panc-1 (HLA-A2+) and the NK cell-sensitive cell line K562 were purchased from European Collection of Animal Cell Cultures (Salisbury, Great Britain); the gastric carcinoma cell line Kato-III (HLA-A2+) was obtained from American Type Culture Collection. Tumor cells were maintained in RPMI 1640 supplemented with 10% FCS (Biochrom), L-glutamine, penicillin, and streptomycin.

Preparation of Tumor Antigens.

Tumor antigens were obtained from Panc-1 tumor cells. Tumor cell lysates were generated by three rapid freeze-thaw cycles as described (14). To induce tumor cell apoptosis, 3 × 106 tumor cells were suspended in medium and exposed to 150 mJ/cm2 UV-B light or 43°C for 2 h. Subsequently, cells were cultured at 37°C on a shaker (30 rpm) to prevent cells from adhering to the culture dish. After 24–32 h, most tumor cells were early apoptotic, defined as annexin V positive and PI negative. By 48–72 h, all tumor cells had died staining positive for annexin V and PI. In our experiments, tumor cells were separated from their supernatant by centrifugation (10 min at 300 × g) after 24–32 h to obtain two fractions consisting of apoptotic tumor cells and low-density fragments, such as apoptotic bodies and other released tumor constituents.

Detection of Apoptosis.

To detect apoptosis after heat or UV-B exposure, 2 × 106 tumor cells were stained with FITC-conjugated annexin V (Bender Med Systems, Vienna, Austria), and 1 μg/ml PI was added shortly before analysis by flow cytometry.

Tumor Antigen Uptake by DCs.

Panc-1 tumor cells were stained with CFSE (1 μm) and DCs with PKH-26-FITC (2 μm) according to the manufacturer’s protocols. Tumor cells were exposed subsequently to UV-B light or 43°C. After 24 h, equal numbers of tumor cells (or their supernatants) were incubated with DCs. Phagocytosis was assessed at various time points by flow cytometry and fluorescence microscopy in two-well chamber slides (Nunc, Wiesbaden, Germany). Intracellular uptake of tumor cell particles was confirmed by confocal laser microscopy (LSM 410 Invert; Carl Zeiss, Jena, Germany). Digital images were obtained in 10 focal planes separated by 1 μm and processed with PhotoShop 3.0 (Adobe Systems, San Jose, CA).

Coculture of DCs with Autologeous PBMCs.

DCs were pulsed with tumor cell lysate, apoptotic tumor cells, or their supernatants (one tumor cell equivalent per DC) for 4 h, extensively washed, and incubated with TNF-α (1000 units/ml) and PGE2 (1 μg/ml) for 24 h to induce a mature phenotype (26). These DCs were cocultured with autologeous nonadherent PBMCs at a ratio of 1:20. Every 7 days, the cultures were restimulated with new DCs. One-third of the medium was replaced on days 3, 5, and at each restimulation by fresh culture medium containing IL-2 (25 units/ml) and IL-7 (10 ng/ml).

Flow Cytometry.

Surface antigen staining was performed as described previously (14). Fluorescence-labeled mAb against CD4, CD8, CD14, CD56, CD69, CD80, CD83, CD86, and HLA-DR was purchased from PharMingen (San Diego, CA), anti-CD3-PerCP from Becton Dickinson (San Jose, CA), and anti-Vγ9-FITC from Coulter Immunotech (Marseilles, France). For the detection of intracellular IFN-γ, cells were incubated with brefeldin A (1 μg/ml) during the last 4 h before harvest. Cells were stained with mAb against the surface markers of interest (CD3, CD4, CD8, or CD3, Vγ9, and CD56), fixed, permeabilized, and stained with anti-IFN-γ-phycoerythrin (Becton Dickinson).

Quantitation of Cytokine Secretion.

Supernatants of the cocultures were harvested, and concentrations of human cytokines IL-12 (p40/p70; Bender Med Systems), IFN-γ, IL-10, and IL-4 were quantified by ELISA (Becton Dickinson) as duplicates.

Cytotoxicity Assay and MHC Class-I Blocking.

After four weekly stimulations with DCs, the lytic activity of PBMCs was assessed in a 51Cr-release assay. A suspension of single target cells was incubated with 100 μCi Na2[51Cr]O4/106 cells for 1 h and washed five times. Tumor cells, 5 × 103/well, were incubated with PBMCs from the cocultures at E:T ratios ranging from 80:1 to 10:1 in round-bottomed, 96-well microtiter plates. Optimal incubation periods were determined for each target cell line by aiming for a low (experimental counts/spontaneous counts) ratio. The incubation period was 16 h for Panc-1 and Kato-III and 4 h for K562. Thereafter, 100 μl of supernatant of each well were collected, and radioactivity was measured with a gamma counter (Wallac Oy, Turku, Finland). Specific lysis was calculated by the formula:

\(specific\ ^{51}Cr-release\ {=}\ {[}(experimental\ counts\ {-}\ spontaneous\ counts)\)
\(/(maximal\ counts\ {-}\ spontaneous\ counts)\ {\times}\ 100\%{]}\)
. Where indicated, MHC class I molecules of the target cells were blocked with 10 μg/106 cells of the mAb W6/32 (Serotec, Oxford, United Kingdom).

Trans-well Cultures of DCs, Apoptotic Tumor Cells, and NK Cells.

Tissue culture inserts with a pore size of 0.3 μm (Nunc) were used to create two compartments within culture plates to separate NK cells, DCs, or apoptotic Panc-1 tumor cells from the other cell types. Anti-IL-12 p40/p70 (clone C11.5, 30 μg/ml) and anti-IL-15 (clone 34505.11, 50 μg/ml) mAb (R&D Systems) were used as indicated. After 48 h, NK cell activation was assessed by intracellular IFN-γ staining as described above.

Statistical Analysis.

Data are expressed as means ± SE. Statistical significance was determined by the Student t test for paired samples of original values. Differences were considered statistically significant for P < 0.05.

DCs Internalize Constituents of Apoptotic Tumor Cells.

Apoptosis of Panc-1 tumor cells was induced by UV-B light or hyperthermia as assessed by annexin V and PI staining (Fig. 1,A). After 24–48 h, when most tumor cells were early apoptotic (annexin V positive and PI negative), apoptotic tumor cells and their supernatants were separated and used for additional experiments. To study uptake of apoptotic tumor cells or released particles by immature DCs, tumor cells were stained with CFSE before the induction of apoptosis. PHK-26-PE-stained DCs were incubated with apoptotic tumor cells or their supernatants and analyzed by flow cytometry. Supernatant-pulsed DCs showed an increase of fluorescence intensity in the CFSE channel, indicating uptake of tumor-derived particles (Fig. 1,B, left graph). To determine the capacity of DCs to capture whole apoptotic tumor cells, DCs were mixed with tumor cells in equal numbers. Double-positive cells represented DCs engulfing tumor cells (Fig. 1,B, right graph). Intracellular uptake of tumor cells was confirmed by fluorescence and confocal microscopy (Fig. 1 C). Neither apoptotic tumor cells nor their supernatants were toxic to DCs or induced maturation as assessed by PI staining and the lack of CD83 expression, respectively (data not shown).

Antigen-loaded DCs Induce a Th1 Cytokine Profile in Cocultures with PBMCs.

Immature DCs were pulsed with lysate, apoptotic tumor cells, or their supernatants and were matured in the presence of TNF-α and PGE2. After 24 h, the DCs expressed high levels of costimulatory molecules and the maturation marker CD83. These DCs were cocultured subsequently with autologeous PBMCs at a ratio of 1:20 in the presence of low doses of IL-2 and IL-7, and new DCs were added every 7 days. IFN-γ was detected in increasing concentrations after each stimulation in the culture medium of cocultures of PBMCs with antigen-pulsed DCs but not in the cultures with unpulsed DCs or PBMCs without DCs. When the different antigen preparations were compared, the highest amount of IFN-γ was found when apoptotic tumor cells were used as the source of antigen with no significant difference between UV-B- and hyperthermia-induced apoptosis (Fig. 2). The supernatant of apoptotic tumor cells induced less IFN-γ but significantly more than tumor lysate (P < 0.01 for heat- and P = 0.02 for UV-B-induced apoptosis). IL-4 and IL-10 were below the detection limit under any of these culture conditions (data not shown).

DCs Loaded with Tumor Antigens Elicit CTLs Able to Kill Pancreatic Carcinoma Cells.

After four stimulations with DCs, the PBMCs were tested for their lytic activity against Panc-1 tumor cells in a 51Cr-release assay. The lytic activity correlated with the amount of IFN-γ found in the culture medium, the highest tumor lysis being detected when apoptotic tumor cells were used for DC pulsing followed by supernatant and lysate (Fig. 3). Again, no statistical difference could be seen between UV-B- and heat-treated tumor cells. Supernatant of apoptotic tumor cells induced higher cell lysis rates than lysate. The difference was statistically significant for hyperthermia-induced apoptosis (P < 0.01) but not for UV-B-induced apoptosis (P = 0.09). To determine the involvement of tumor antigen-specific CTLs in tumor killing, the experiments were also performed in the presence of the MHC class I-blocking antibody W6/32. Tumor cell lysis was effectively blocked in the experiments with antigens from cell lysate or tumor cell supernatant indicative of CTL-mediated tumor cell killing (Fig. 3). If CTL-mediated killing was defined as the difference between lysis without and with MHC I blocking, apoptotic tumor cell supernatants as antigen source were significantly superior to lysate (P < 0.01). Interestingly, PBMCs of the cocultures with apoptotic tumor cell-pulsed DCs still effectively killed tumor cells after MHC I blocking. Moreover, these cells were capable of lysing the unrelated gastric carcinoma cell line Kato-III, as well as the NK cell-sensitive cell line K562 (Fig. 4).

Apoptotic Tumor Cells Induce Activation of NK and γδ T Cells.

To determine the activation pattern of different leukocyte subsets from PBMCs in the cocultures, we analyzed IFN-γ production of CTLs (CD3+ CD8+), T helper cells (CD3+ CD4+), NK cells (CD3 CD56+), and γδ T cells (CD3+ Vγ9+) by flow cytometry. As expected from the cytotoxicity assays, the activation pattern of leukocyte subsets was quite different if whole apoptotic tumor cells were chosen as the source of antigen compared with supernatant or lysate (Fig. 5). A portion (23% hyperthermia and 19% UV-B) of NK cells were activated in PBMCs cocultured with apoptotic tumor cell-pulsed DCs. Furthermore, 30% (hyperthermia) and 36% (UV-B) of γδ T cells stained positive for IFN-γ under the same conditions. In contrast, only a small number of NK and γδ T cells from cocultures with supernatant- or lysate-pulsed DCs produced IFN-γ. No significant differences in CTL activation could be seen between apoptotic tumor cell-pulsed versus supernatant-pulsed DCs. However, both conditions were clearly superior to lysate with a higher frequency of IFN-γ-producing CTLs and T-helper cells.

NK Cell Activation Is Dependent on Tumor Cell Contact and IL-12 Produced by DCs.

Because tumor cells, which were not internalized by DCs, were still present in the cocultures of PBMCs with tumor cell-pulsed DCs, we were interested if direct tumor cell contact with NK cells was necessary for NK cell activation or if DC directly triggered NK cell activation. Activation of NK cells (purity > 95%) was studied by intracellular IFN-γ staining in a trans-well culture system allowing only direct contact between two of the three cell types: DCs and apoptotic tumor cells, NK cells and apoptotic tumor cells, or DCs and NK cells. Only in cultures that allowed direct cell-to-cell contact between NK cells and apoptotic tumor cells, NK cell activation was observed (Fig. 6). However, no NK cell activation was seen in the absence of DCs (data not shown). Therefore, most likely, a soluble factor secreted by DCs was involved. Adding IL-12 was able to substitute for DCs. Moreover, IL-12-blocking antibodies inhibited activation when DCs were present (Fig. 6), whereas blocking IL-15 showed no effect. This indicated that NK cell activation depended on both direct cell contact with tumor cells, as well as DC-derived IL-12.

Cross-priming of CTLs with antitumor activity has been demonstrated with DCs pulsed with tumor cell lysates (17), apoptotic tumor cells (18, 19, 20, 21, 22), and apoptotic bodies (23), but to our knowledge, no quantitative comparison has been undertaken. In this study, we investigated the ability of monocyte-derived DCs pulsed with different antigen preparations from the pancreatic carcinoma cell line Panc-1 to induce an antitumor T-cell response in a cross-presentation in vitro model. The antigen preparations consisted of: (a) tumor cell lysates; (b) UV-B- or hyperthermia-induced apoptotic tumor cells; or (c) their supernatants, containing low-density particles released from the cells, such as apoptotic bodies. Immature DCs internalized tumor cells, as well as released particles, as shown by flow cytometry and confocal microscopy. After antigen loading, we activated the DCs with cytokines to induce a T-cell stimulatory phenotype for effective T-cell priming.

Our results showed that antigens from apoptotic tumor cells, whole cells as well as released particles, were more potent than tumor lysates in inducing T-cell priming and activation by DCs. This was evidenced by enhanced IFN-γ secretion and a higher frequency of activated CTLs and T-helper cells, as well as a higher rate of MHC class I-restricted tumor cell killing. This result is in agreement with a report from Hoffman et al.(19), who observed stronger CTL responses with apoptotic tumor cells, compared with cell lysates in a squamous cell carcinoma model. Enhanced CTL activation by antigens from apoptotic cells may be attributed to several mechanisms. After internalization, most particulate antigens requiring phagocytosis are digested into peptides associating with MHC class-II molecules in the endocytic compartments and are presented to T-helper cells (27). This is believed to be the predominant processing pathway of cell lysates. In contrary, scavenger receptor-mediated phagocytosis of apoptotic tumor cells allows antigens to gain access to MHC class-I compartments, resulting in cross-presentation of the antigen to CTLs (18, 28). In addition, enhanced CTL responses against tumors might be mediated by hsps expressed by stress-induced apoptotic tumor cells (29). Hsps have been shown to improve uptake of antigens by DCs (30) and can provide additional antigenic epitopes by peptides complexed to hsp (31). On the basis of this theoretical background and our own observations, we conclude that antigen preparations from apoptotic tumor cells represent a promising alternative to tumor lysate in DC-based tumor vaccines.

Tumor cell killing was highest when whole apoptotic tumor cells were used as the source of antigen. However, compared with experiments with supernatant or lysate, tumor cell killing was only partially MHC class I restricted. Moreover, unrelated cell lines were also effectively killed. The analysis of the activation pattern of different leukocyte subsets in these cocultures revealed not only activation of CTLs and T-helper cells but also of NK cells and γδ T cells. The latter two are effector cell types of the innate arm of the immune defense with the ability to kill tumor cells in an antigen-unspecific manner. In a recent study, vaccinating mice with DC cocultured with tumor cells induced CTL-mediated protective immunity, as well as NK cell activation (10). Interestingly, in that study, NK cell depletion abrogated tumor protection favoring the concept that combining innate and acquired effector mechanisms may enhance tumor immunogenicity.

Mechanisms leading to NK and γδ T-cell activation by tumor cells are only understood incompletely. Activation seems to be influenced by a balance of negative and positive signals. A positive signal can be mediated by the NKG2D-DAP10 receptor complex, expressed on the surface of NK cells and γδ T cells, which is known to interact with stress-induced ligands on tumor cells, such as MICA and Rae-1 (32, 33, 34, 35). Stimulation via NKG2D can override negative signals from inhibitory receptors and triggers degranulation and perforin-mediated apoptosis of the tumor cell. Another concept favors that direct cell-to-cell contact between DCs and NK cells enhances the cytolytic activity of NK cells (7). In our experiments, apoptotic tumor cells were present in cocultures of PBMCs with apoptotic tumor cell-pulsed DCs. This was not the case in the cocultures with supernatant- or lysate-pulsed DCs, because excess antigen was removed by washing. Therefore, it seemed likely that cell contacts of tumor cells with NK cells within the PBMCs played a role in the observed NK cell activation. When we analyzed the required components for this effect in a trans-well cultures system, we found that direct cell contact between NK cells and apoptotic tumor cells was required, as well as a soluble factor produced by DCs. A recent report also points into this direction, but the factor was not identified (36). In cytokine-blocking experiments, we found that NK cell activation was dependent on DC-derived IL-12. Therefore, DCs played a dual role in these cultures: (a) they cross-presented tumor antigen to T cells inducing tumor-specific CTLs; and (b) they activated innate effector cells via IL-12. A vaccination trial using DCs fused with tumor cells, achieved promising results in patients with renal cell carcinoma (4). Because in that study fusion was only effective partially, unfused tumor cells were injected together with DCs; possibly, activation of innate effector cells by direct contact with stressed tumor cells in the presence of DC-derived IL-12 played a role in the vaccination success observed in some patients.

In conclusion, pulsing DCs with antigen from apoptotic tumor cells seems to be a promising alternative to lysate and might be used for tumors when immunogenic epitopes are unknown. Furthermore, a vaccine containing apoptotic tumor cells in addition to DCs could offer advantages by stimulating effector cells of the innate immune system. Activation of NK cells and γδ T cells at the injection site could provide stimuli for local antigen-presenting cells, enhancing the effectiveness of the vaccine. However, the efficacy of this strategy needs further evaluation.

Fig. 1.

Immature DCs internalize apoptotic Panc-1 tumor cells and released particles. Annexin V-FITC and PI staining show apoptosis of tumor cells after exposure to UV-B light (A). DCs capture constituents of apoptotic tumor cells (B). CFSE-stained apoptotic tumor cells (dotplots) or their supernatants (histogram) were incubated with PHK-26-stained DCs at a ratio of 1:1 for 4 h and analyzed by flow cytometry. Controls were incubated on ice. Internalization of tumor cells was analyzed by fluorescence and confocal microscopy (C). From left to right, a DC engulfs a tumor cell fragment (green, top right) 30 min after coincubation. Three confocal plains confirm intracellular position of a tumor cell fragment (green).

Fig. 1.

Immature DCs internalize apoptotic Panc-1 tumor cells and released particles. Annexin V-FITC and PI staining show apoptosis of tumor cells after exposure to UV-B light (A). DCs capture constituents of apoptotic tumor cells (B). CFSE-stained apoptotic tumor cells (dotplots) or their supernatants (histogram) were incubated with PHK-26-stained DCs at a ratio of 1:1 for 4 h and analyzed by flow cytometry. Controls were incubated on ice. Internalization of tumor cells was analyzed by fluorescence and confocal microscopy (C). From left to right, a DC engulfs a tumor cell fragment (green, top right) 30 min after coincubation. Three confocal plains confirm intracellular position of a tumor cell fragment (green).

Close modal
Fig. 2.

IFN-γ production by PBMCs cocultured with tumor antigen-loaded DCs. DCs were pulsed with tumor lysate, supernatants of apoptotic tumor cells, or whole apoptotic tumor cells. After maturation with TNF-α and PGE2, the DCs were cocultured with autologeous nonadherent PBMCs. The graph shows concentrations of IFN-γ in the supernatants of the PBMCs after the fourth weekly stimulation with DCs (mean ± SE, n = 6).

Fig. 2.

IFN-γ production by PBMCs cocultured with tumor antigen-loaded DCs. DCs were pulsed with tumor lysate, supernatants of apoptotic tumor cells, or whole apoptotic tumor cells. After maturation with TNF-α and PGE2, the DCs were cocultured with autologeous nonadherent PBMCs. The graph shows concentrations of IFN-γ in the supernatants of the PBMCs after the fourth weekly stimulation with DCs (mean ± SE, n = 6).

Close modal
Fig. 3.

Tumor cell killing by PBMCs stimulated with tumor antigen-pulsed DCs. DCs were pulsed with tumor cell lysate, supernatants of apoptotic tumor cells, or whole apoptotic tumor cells. Subsequently, DCs were washed, activated with TNF-α and PGE2, and cocultured with autologeous PBMCs. Two days after the fourth weekly stimulation with DCs, the lytic activity of the PBMCs was assessed using 51Cr-labeled Panc-1 tumor cells as targets in a 51Cr-release assay (solid bars). To determine MHC class I restriction of tumor cell killing, the target cells were preincubated with the MHC class I-blocking antibody W6/32 (open bars). The specific lysis at an E:T ratio of 80:1 is shown (mean ± SE, n = 6).

Fig. 3.

Tumor cell killing by PBMCs stimulated with tumor antigen-pulsed DCs. DCs were pulsed with tumor cell lysate, supernatants of apoptotic tumor cells, or whole apoptotic tumor cells. Subsequently, DCs were washed, activated with TNF-α and PGE2, and cocultured with autologeous PBMCs. Two days after the fourth weekly stimulation with DCs, the lytic activity of the PBMCs was assessed using 51Cr-labeled Panc-1 tumor cells as targets in a 51Cr-release assay (solid bars). To determine MHC class I restriction of tumor cell killing, the target cells were preincubated with the MHC class I-blocking antibody W6/32 (open bars). The specific lysis at an E:T ratio of 80:1 is shown (mean ± SE, n = 6).

Close modal
Fig. 4.

PBMCs from cocultures with apoptotic tumor cell-pulsed DCs effectively kill unrelated tumor cell lines. After four weekly stimulations of PBMCs with DCs pulsed with different antigen preparations from Panc-1 tumor cells, the killing of the unrelated HLA-A2+ gastric carcinoma cell line Kato-III (n = 6) and the NK-sensitive cell line K562 (n = 5) was assessed in a 51Cr-release assay. The specific lysis at an E:T ratio of 80:1 is shown (mean ± SE).

Fig. 4.

PBMCs from cocultures with apoptotic tumor cell-pulsed DCs effectively kill unrelated tumor cell lines. After four weekly stimulations of PBMCs with DCs pulsed with different antigen preparations from Panc-1 tumor cells, the killing of the unrelated HLA-A2+ gastric carcinoma cell line Kato-III (n = 6) and the NK-sensitive cell line K562 (n = 5) was assessed in a 51Cr-release assay. The specific lysis at an E:T ratio of 80:1 is shown (mean ± SE).

Close modal
Fig. 5.

Activation pattern of different lymphocyte subsets in PBMCs cocultured with tumor antigen-pulsed DCs. PBMCs were removed 12 h after the fourth stimulation with DCs, and intracellular IFN-γ was stained after incubation with brefeldin A. In a four-color flow cytometric analysis, CTLs were identified as CD3+ CD8+, T-helper cells as CD3+ CD4+, NK cells as CD3 CD56+, and γδ T cells as CD3+ vγ9+. Data represent means of percentage of IFN-γ-positive cells ± SE (n = 5).

Fig. 5.

Activation pattern of different lymphocyte subsets in PBMCs cocultured with tumor antigen-pulsed DCs. PBMCs were removed 12 h after the fourth stimulation with DCs, and intracellular IFN-γ was stained after incubation with brefeldin A. In a four-color flow cytometric analysis, CTLs were identified as CD3+ CD8+, T-helper cells as CD3+ CD4+, NK cells as CD3 CD56+, and γδ T cells as CD3+ vγ9+. Data represent means of percentage of IFN-γ-positive cells ± SE (n = 5).

Close modal
Fig. 6.

DC-mediated activation of NK cells is dependent on direct cell-to-cell contact between apoptotic tumor cells and NK cells, as well as the presence of IL-12. Purified NK cells (NK) were cocultured with DCs and UV light-induced apoptotic tumor cells in a trans-well culture system. On day 2, NK cell activation was assessed by intracellular IFN-γ staining. The culture conditions are shown in the top. Where indicated, IL-12 (p40/p70)-blocking antibodies were added to the medium. A representative experiment of six is shown.

Fig. 6.

DC-mediated activation of NK cells is dependent on direct cell-to-cell contact between apoptotic tumor cells and NK cells, as well as the presence of IL-12. Purified NK cells (NK) were cocultured with DCs and UV light-induced apoptotic tumor cells in a trans-well culture system. On day 2, NK cell activation was assessed by intracellular IFN-γ staining. The culture conditions are shown in the top. Where indicated, IL-12 (p40/p70)-blocking antibodies were added to the medium. A representative experiment of six is shown.

Close modal

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 is part of the thesis of C. Scholz at the Ludwig-Maximilians-University, Munich, Germany. M. Schnurr is supported by a grant from the university of Munich FöFoLe No. 216.

4

The abbreviations used are: DC, dendritic cell; NK, natural killer; PBMC, peripheral blood mononuclear cell; TNF, tumor necrosis factor; IL, interleukin; PI, propidium iodide; CFSE, carboxy-fluorescein diacetate succinimidyl ester; mAb, monoclonal antibody; PGE2, prostaglandin E2; hsp, heat shock protein.

1
Banchereau J., Briere F., Caux C., Davoust J., Lebecque S., Liu Y. J., Pulendran B., Palucka K. Immunobiology of dendritic cells.
Annu. Rev. Immunol.
,
18
:
767
-811,  
2000
.
2
Geiger J., Hutchinson R., Hohenkirk L., McKenna E., Chang A., Mule J. Treatment of solid tumours in children with tumour-lysate-pulsed dendritic cells.
Lancet
,
356
:
1163
-1165,  
2000
.
3
Hsu F. J., Benike C., Fagnoni F., Liles T. M., Czerwinski D., Taidi B., Engleman E. G., Levy R. Vaccination of patients with B-cell lymphoma using autologous antigen-pulsed dendritic cells.
Nat. Med.
,
2
:
52
-58,  
1996
.
4
Kugler A., Stuhler G., Walden P., Zoller G., Zobywalski A., Brossart P., Trefzer U., Ullrich S., Muller C. A., Becker V., Gross A. J., Hemmerlein B., Kanz L., Muller G. A., Ringert R. H. Regression of human metastatic renal cell carcinoma after vaccination with tumor cell-dendritic cell hybrids.
Nat. Med.
,
6
:
332
-336,  
2000
.
5
Nestle F. O., Alijagic S., Gilliet M., Sun Y., Grabbe S., Dummer R., Burg G., Schadendorf D. Vaccination of melanoma patients with peptide- or tumor lysate-pulsed dendritic cells.
Nat. Med.
,
4
:
328
-332,  
1998
.
6
Thurner B., Haendle I., Roder C., Dieckmann D., Keikavoussi P., Jonuleit H., Bender A., Maczek C., Schreiner D., von den Driesch P., Brocker E. B., Steinman R. M., Enk A., Kampgen E., Schuler G. 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
-1678,  
1999
.
7
Fernandez N. C., Lozier A., Flament C., Ricciardi-Castagnoli P., Bellet D., Suter M., Perricaudet M., Tursz T., Maraskovsky E., Zitvogel L. Dendritic cells directly trigger NK cell functions: cross-talk relevant in innate anti-tumor immune responses in vivo.
Nat. Med.
,
5
:
405
-411,  
1999
.
8
Geldhof A. B., Moser M., Lespagnard L., Thielemans K., De Baetselier P. Interleukin-12-activated natural killer cells recognize B7 costimulatory molecules on tumor cells and autologous dendritic cells.
Blood.
,
91
:
196
-206,  
1998
.
9
Rothenfusser S., Hornung V., Krug A., Towarowski A., Krieg A. M., Endres S., Hartmann G. Distinct CpG oligonucleotide sequences activate human gamma Δ T cells via interferon-α/-β.
Eur. J. Immunol.
,
31
:
3525
-3534,  
2001
.
10
Kim K. D., Choi S. C., Kim A., Choe Y. K., Choe I. S., Lim J. S. Dendritic cell-tumor coculturing vaccine can induce antitumor immunity through both NK and CTL interaction.
Int. Immunopharmacol.
,
1
:
2117
-2129,  
2001
.
11
Elnemr A., Ohta T., Yachie A., Kayahara M., Kitagawa H., Ninomiya I., Fushida S., Fujimura T., Nishimura G., Shimizu K., Miwa K. Human pancreatic cancer cells express non-functional Fas receptors and counterattack lymphocytes by expressing Fas ligand; a potential mechanism for immune escape.
Int. J. Oncol.
,
18
:
33
-39,  
2001
.
12
Elnemr A., Ohta T., Yachie A., Kayahara M., Kitagawa H., Fujimura T., Ninomiya I., Fushida S., Nishimura G. I., Shimizu K., Miwa K. Human pancreatic cancer cells disable function of Fas receptors at several levels in Fas signal transduction pathway.
Int. J. Oncol.
,
18
:
311
-316,  
2001
.
13
Bonham C. A., Lu L., Banas R. A., Fontes P., Rao A. S., Starzl T. E., Zeevi A., Thomson A. W. TGF-β 1 pretreatment impairs the allostimulatory function of human bone marrow-derived antigen-presenting cells for both naive and primed T cells.
Transpl. Immunol.
,
4
:
186
-191,  
1996
.
14
Schnurr M., Galambos P., Scholz C., Then F., Dauer M., Endres S., Eigler A. Tumor cell lysate-pulsed human dendritic cells induce a T-cell response against pancreatic carcinoma cells: an in vitro model for the assessment of tumor vaccines.
Cancer Res.
,
61
:
6445
-6450,  
2001
.
15
Banchereau J., Palucka A. K., Dhodapkar M., Burkeholder S., Taquet N., Rolland A., Taquet S., Coquery S., Wittkowski K. M., Bhardwaj N., Pineiro L., Steinman R., Fay J. Immune and clinical responses in patients with metastatic melanoma to CD34(+) progenitor-derived dendritic cell vaccine.
Cancer Res.
,
61
:
6451
-6458,  
2001
.
16
Bennett S. R., Carbone F. R., Karamalis F., Miller J. F., Heath W. R. Induction of a CD8+ cytotoxic T lymphocyte response by cross-priming requires cognate CD4+ T cell help.
J. Exp. Med.
,
186
:
65
-70,  
1997
.
17
Fields R. C., Shimizu K., Mule J. J. Murine dendritic cells pulsed with whole tumor lysates mediate potent antitumor immune responses in vitro and in vivo.
Proc. Natl. Acad. Sci. USA
,
95
:
9482
-9487,  
1998
.
18
Albert M. L., Sauter B., Bhardwaj N. Dendritic cells acquire antigen from apoptotic cells and induce class I-restricted CTLs.
Nature (Lond.)
,
392
:
86
-89,  
1998
.
19
Hoffmann T. K., Meidenbauer N., Dworacki G., Kanaya H., Whiteside T. L. Generation of tumor-specific T-lymphocytes by cross-priming with human dendritic cells ingesting apoptotic tumor cells.
Cancer Res.
,
60
:
3542
-3549,  
2000
.
20
Berard F., Blanco P., Davoust J., Neidhart-Berard E. M., Nouri-Shirazi M., Taquet N., Rimoldi D., Cerottini J. C., Banchereau J., Palucka A. K. Cross-priming of naive CD8 T cells against melanoma antigens using dendritic cells loaded with killed allogeneic melanoma cells.
J. Exp. Med.
,
192
:
1535
-1544,  
2000
.
21
Jenne L., Arrighi J. F., Jonuleit H., Saurat J. H., Hauser C. Dendritic cells containing apoptotic melanoma cells prime human CD8+ T cells for efficient tumor cell lysis.
Cancer Res.
,
60
:
4446
-4452,  
2000
.
22
Shaif-Muthana M., McIntyre C., Sisley K., Rennie I., Murray A. Dead or alive: immunogenicity of human melanoma cells when presented by dendritic cells.
Cancer Res.
,
60
:
6441
-6447,  
2000
.
23
Henry F., Bretaudeau L., Hequet A., Barbieux I., Lieubeau B., Meflah K., Gregoire M. Role of antigen-presenting cells in long-term antitumor response based on tumor-derived apoptotic body vaccination.
Pathobiology
,
67
:
306
-310,  
1999
.
24
Zitvogel L., Regnault A., Lozier A., Wolfers J., Flament C., Tenza D., Ricciardi-Castagnoli P., Raposo G., Amigorena S. Eradication of established murine tumors using a novel cell-free vaccine: dendritic cell-derived exosomes.
Nat. Med.
,
4
:
594
-600,  
1998
.
25
Wolfers J., Lozier A., Raposo G., Regnault A., Thery C., Masurier C., Flament C., Pouzieux S., Faure F., Tursz T., Angevin E., Amigorena S., Zitvogel L. Tumor-derived exosomes are a source of shared tumor rejection antigens for CTL cross-priming.
Nat. Med.
,
7
:
297
-303,  
2001
.
26
Rieser C., Bock G., Klocker H., Bartsch G., Thurnher M. Prostaglandin E2 and tumor necrosis factor alpha cooperate to activate human dendritic cells: synergistic activation of interleukin 12 production.
J. Exp. Med.
,
186
:
1603
-1608,  
1997
.
27
Inaba K., Turley S., Yamaide F., Iyoda T., Mahnke K., Inaba M., Pack M., Subklewe M., Sauter B., Sheff D., Albert M., Bhardwaj N., Mellman I., Steinman R. M. Efficient presentation of phagocytosed cellular fragments on the major histocompatibility complex class II products of dendritic cells.
J. Exp. Med.
,
188
:
2163
-2173,  
1998
.
28
Albert M. L., Pearce S. F., Francisco L. M., Sauter B., Roy P., Silverstein R. L., Bhardwaj N. Immature dendritic cells phagocytose apoptotic cells via αvβ5 and CD36, and cross-present antigens to cytotoxic T lymphocytes.
J. Exp. Med.
,
188
:
1359
-1368,  
1998
.
29
Feng H., Zeng Y., Whitesell L., Katsanis E. Stressed apoptotic tumor cells express heat shock proteins and elicit tumor-specific immunity.
Blood
,
97
:
3505
-3512,  
2001
.
30
Todryk S., Melcher A. A., Hardwick N., Linardakis E., Bateman A., Colombo M. P., Stoppacciaro A., Vile R. G. Heat shock protein 70 induced during tumor cell killing induces Th1 cytokines and targets immature dendritic cell precursors to enhance antigen uptake.
J. Immunol.
,
163
:
1398
-1408,  
1999
.
31
Wells A. D., Malkovsky M. Heat shock proteins, tumor immunogenicity and antigen presentation: an integrated view.
Immunol. Today
,
21
:
129
-132,  
2000
.
32
Diefenbach A., Jensen E. R., Jamieson A. M., Raulet D. H. Rae1 and H60 ligands of the NKG2D receptor stimulate tumour immunity.
Nature (Lond.)
,
413
:
165
-171,  
2001
.
33
Diefenbach A., Jamieson A. M., Liu S. D., Shastri N., Raulet D. H. Ligands for the murine NKG2D receptor: expression by tumor cells and activation of NK cells and macrophages.
Nat. Immunol.
,
1
:
119
-126,  
2000
.
34
Cerwenka A., Bakker A. B., McClanahan T., Wagner J., Wu J., Phillips J. H., Lanier L. L. Retinoic acid early inducible genes define a ligand family for the activating NKG2D receptor in mice.
Immunity
,
12
:
721
-727,  
2000
.
35
Bauer S., Groh V., Wu J., Steinle A., Phillips J. H., Lanier L. L., Spies T. Activation of NK cells and T cells by NKG2D, a receptor for stress-inducible MICA.
Science
,
285
:
727
-729,  
1999
.
36
Nishioka Y., Nishimura N., Suzuki Y., Sone S. Human monocyte-derived and CD83(+) blood dendritic cells enhance NK cell-mediated cytotoxicity.
Eur. J. Immunol.
,
31
:
2633
-2641,  
2001
.