Abstract
Due to their capacity to induce primary immune responses, dendritic cells (DC) are attractive vectors for immunotherapy of cancer. Yet the targeting of tumor Ags to DC remains a challenge. Here we show that immature human monocyte-derived DC capture various killed tumor cells, including Jurkat T cell lymphoma, malignant melanoma, and prostate carcinoma. DC loaded with killed tumor cells induce MHC class I- and class II-restricted proliferation of autologous CD8+ and CD4+ T cells, demonstrating cross-presentation of tumor cell-derived Ags. Furthermore, tumor-loaded DC elicit expansion of CTL with cytotoxic activity against the tumor cells used for immunization. CTL elicited by DC loaded with the PC3 prostate carcinoma cell bodies kill another prostate carcinoma cell line, DU145, suggesting recognition of shared Ags. Finally, CTL elicited by DC loaded with killed LNCap prostate carcinoma cells, which express prostate specific Ag (PSA), are able to kill PSA peptide-pulsed T2 cells. This demonstrates that induced CTL activity is not only due to alloantigens, and that alloantigens do not prevent the activation of T cells specific for tumor-associated Ags. This approach opens the possibility of using allogeneic tumor cells as a source of tumor Ag for antitumor therapies.
Dendritic cells (DC)3 are the only APCs able to induce primary immune responses (reviewed in Refs. 1 and 2). Circulating DC precursors home to tissues where they reside as immature, Ag-capturing cells with high endocytic and phagocytic activities. Following Ag uptake, DC migrate to the secondary lymphoid organs where they mature and become APCs able to select and activate naive Ag-specific CD4+ T cells. This permits diversification of the response and activation of Ag-specific effectors such as Ag-specific CTL and B cells as well as nonspecific effectors such as NK cells, macrophages, and eosinophils (3). The induction of tumor immunity can be viewed as a three-step process that includes 1) presentation of tumor-associated Ags (TAA) (3), 2) selection and activation of TAA-specific T cells as well as nonspecific effectors, and 3) homing of TAA-specific T cells to the tumor site and recognition of restriction elements leading to the elimination of tumor cells. Experiments in mice have demonstrated the development of both protective and therapeutic anti-tumor responses induced by TAA-loaded DC (reviewed in Ref. 4). Furthermore, pilot clinical trials in humans show the feasibility of DC administration and the ability of peptide-loaded DC to induce peptide-specific T cell responses in patients with lymphoma, malignant melanoma, and prostate carcinoma (5, 6, 7, 8, 9). However, the identification of an efficient Ag loading strategy remains one of the critical challenges in DC-based vaccination protocols. Several systems of TAA delivery to DC have been employed, including 1) defined peptides of known sequences (10, 11), 2) undefined acid-eluted peptides from autologous tumor (12), 3) whole tumor lysates (13), 4) retroviral and adenoviral vectors (14, 15), 5) tumor cell-derived RNA (16), and 6) fusion of DC with tumor cells (17). Although in all instances, induction of T cell responses and considerable anti-tumor effects have been observed, each of these methods has potential drawbacks (reviewed in Ref. 3). Foremost, the use of MHC class I binding peptides is associated with 1) HLA restriction, and 2) the limitation of induced immune responses to CD8+ T cells. Furthermore, it is unclear whether or which of the defined peptides represents tumor rejection Ags in vivo. Contrary to the peptide-based approach, the use of unfractionated tumor material may provide both MHC class I and MHC class II epitopes and does not require the identification of TAA. Recent studies demonstrated the ability of DC to capture apoptotic cells and elicit MHC class I-restricted secondary CTL responses (18, 19). Here we show that immature human DC can capture killed tumor cells and subsequently process and present their Ags to generate TAA-specific CTL.
Materials and Methods
Media and reagents
Complete culture medium (CM) consisted of RPMI 1640, 1% l
Cell purification
Purified B cells and CD4+ and CD8+ T cells were obtained from Ficoll-separated PBMC of healthy volunteers (with no history of blood transfusion) depleted of other cells using purified CD3 (UCHT1), CD4 (13B8.2), CD8 (B9.11), CD14 (RMO52), CD16 (3G8), CD19 (J4.119), CD56 (NKH-1), anti-HLA-DR (B8.12.2), and anti-glycophorin A (D2.10) mAbs (Beckman-Coulter, Miami, FL) and goat anti-mouse IgG Dynabeads (Dynal, Lake Success, NY). The purity of the enriched populations was >85%.
Generation of DC
Immature monocyte-derived DC were generated from the adherent fraction of peripheral blood PBMC (20). Briefly, PBMC were suspended in CM and allowed to adhere to plastic dishes (Falcon six-well plates, Becton Dickinson, Franklin Lakes, NJ). After 2-h incubation at 37°C, the nonadherent cells were removed, and the adherent cells were cultured in CM with GM-CSF (100 ng/ml) and IL-4 (5 ng/ml). Cultures were fed every 2 days. Cells were routinely used on day 6, and DC recovery, as determined by immunofluorescence and flow cytometry, was >90% of CD1a+ CD14− cells.
Induction of tumor cell death
Cell death was induced by 16-h treatment with anti-Fas mAb (1 μg/1.0 × 106/ml; clone CH-11, Beckman-Coulter). The PC3 prostate cancer cell line required prior sensitization with cycloheximide (25 μg/ml) for 2 h followed by anti-Fas treatment. LNCap cell death was induced by gamma irradiation (80 Gy) 24 h after exposure to TNF-α (40 ng/ml). Alternatively, tumor cells were induced to die by gamma irradiation (100 Gy) followed by overnight culture in CM. Cell death was assessed by morphology, externalization of phosphatidylserine using FITC-labeled annexin V, and staining with the DNA-specific dyes, 7-aminoactinomycin D (7-AAD) and trypan blue.
Phagocytosis of apoptotic tumor cells
Killed tumor cells were washed with PBS and labeled with 7-AAD at a concentration of 20 μg/ml/106 cells for 30 min at 4°C. The labeled tumor cells were subsequently cocultured with CD1a-labeled immature DC at a 1/5 ratio (selected from preliminary experiments) at 4°C or 37°C. After 1 h cells were washed and treated with 0.05% trypsin/0.02% EDTA for 5 min to disrupt cell-cell binding. Phagocytosis was quantified by flow cytometry as the percentage of double-positive cells (CD1a+ 7-AAD+).
For confocal microscopy, cells were allowed to adhere on poly-l-lysine-coated slides (Baxter Diagnostic, Deerfield, IL) for 30 min at room temperature, fixed with 4% paraformaldehyde/PBS for 15 min, washed with PBS, and mounted. Confocal microscopy was performed using a Leica TCS-NT SP (Deerfield, IL) equipped with Ar, Kr, and He-Ne lasers and a spectrophotometer to separate the detection channels of FITC (510–550 nm) and 7-AAD (600–660 nm). Simultaneous acquisition of FITC and 7-AAD and transmission set for differential interference contrast (DIC) were performed using the 488- and 568-nm laser lines reflected with a 488/568 double dichroic mirror. Images were generated using the PL-APO ×100 objective, zoomed ×2, and measured in the 1024 × 1024 format.
Blocking of phagocytosis
Immature DC were incubated with cytocalasin D (20 μm) or CD36 (thrombospondin receptor, clone CB38, PharMingen, San Diego, CA; 5 μg/ml), CD51/CD61 (vitronectin receptor αvβ3, clone 23C6, PharMingen; 5 μg/ml), annexin V (clone RUU-WAC2A, Caltag, Burlingame, CA; 5 μg/ml), and isotype-matched controls mAbs for 30 min. DC were washed and used in the phagocytosis assay as described above.
T cell proliferation assay
CD1a-labeled DC were cocultured with unlabeled tumor cell bodies, sorted based on CD1a expression (purity, >98%), and cultured at graded doses in CM with 10% human AB serum, autologous PBMC (1 × 105 PBMC/well/200 μl), purified CD4+ T cells, or purified CD8+ T (5 × 104/well/200 μl) cells. In cultures with CD8+ T cells, soluble CD40 ligand (CD40L; 200 ng/ml) and IL-2 (5 U/ml) were added to induce DC maturation and to support T cell proliferation. Proliferation was determined after 5 days by uptake of tritiated thymidine (1 μCi/well for the last 16 h).
mAb blocking assays
For blocking experiments, DC were incubated with mAbs against HLA-DR (clone L243, Becton Dickinson; 5 μg/ml) and HLA-A,B,C (W 6/32, Dako, Carpenteria, CA; 5 μg/ml) for 30 min before the addition of T cells, and the mAbs were left throughout the culture period. Irrelevant mAbs were used as isotype controls.
Generation of CTL
DC were loaded with killed tumor cells, sorted, and cultured with autologous PBMC or purified CD8+ T cells at 0.5–1 × 105 DC/ml and 2 × 106 T cells/ml in a final volume of 2 ml. IL-7 (10 U/ml) was added in the first cycle. T cells were restimulated every week with tumor-loaded DC, and IL-2 (10 U/ml) was added in the second and third cycles. T cells were harvested after 3 wk, and CTL activity was tested in a standard chromium release assay using sensitized tumor cells, unloaded and peptide-loaded T2 cells, and K562 cells as targets.
Synthetic peptides and peptide binding to HLA-A2
The synthetic PSA peptides used were PSA1 (NH2-FLTPKKLQCV; aa 141–150) and PSA2 (NH2-KLQCVDLHV; aa 146–154; Biosynthesis, Lewisville, TX). HLA-A2 binding affinity of the PSA peptides was evaluated by up-regulation of HLA-A2 surface expression on T2 cells after overnight peptide loading in serum-free CM (50 μg peptide/ml/1 × 106 T2 cells). The T2 cells were washed and stained with FITC-conjugated HLA-A2-specific mAb (28, HLA-A2, One Lamda, Canoga Park, CA). The mean fluorescence intensity of HLA-A2 staining was analyzed by flow cytometry.
Results
Immature DC phagocytose bodies of killed tumor cells
We first determined the capacity of immature monocyte-derived DC to capture killed tumor cells (>80% of dead cells as determined by annexin V binding and propidium iodide staining). The capture of representative DNA-labeled (7-AAD) killed tumor cells (DU145) by CD1a-labeled DC was determined by flow cytometry as the percentage of FITC-CD1a+ DC that expressed the red dye 7-AAD (Fig. 1⇓A). Forward scatter/side scatter analysis revealed the increased granularity of the double-positive population, further confirming the capture of tumor cell bodies by DC (Fig. 1⇓A). As shown in Fig. 1⇓B, up to 45% of DC are able to capture cell bodies derived from various tumor cell lines, including the melanoma cell lines (A2058, A375.S2), prostate carcinoma cell lines (DU145, PC3, LNCap), and Jurkat T lymphoma. The extent of phagocytosis was independent of whether cell death was induced by Fas ligation or gamma irradiation (data not shown). The phagocytosis was also seen in an autologous setting where bodies of EBV-transformed B lymphoblastoid cells or nontransformed B cells were used (data not shown). Confocal microscopy examination of cells revealed that DC indeed internalize the killed tumor cells (Fig. 2⇓). The fate of internalized cell bodies within DC was analyzed after coculture and subsequent staining with FITC-labeled HLA-DR Ab. After 1 h of coculture DNA labeled cell bodies (red) were detected inside the DC in compartments distinct from MHC class II compartments (green, Fig. 2⇓b). However, after 4 h the two compartments were merged (Fig. 2⇓c). Together, these results demonstrate that immature DC have the ability to phagocytose cell bodies derived from different tumor cell lines.
Immature DC capture killed tumor-derived cell bodies. A, DU145 prostate carcinoma cells were induced to die by anti-Fas (1 μg/ml; CH-11; 16 h), and the generated bodies were labeled with 7-AAD and incubated with FITC-CD1a-labeled immature DC for 1 h at 37°C. The capture of killed cells by DC was quantitated by flow cytometry as the percentage of CD1a+ DC that become positive for 7-AAD (data represent CD1a+-gated events). Forward scatter/side scatter analysis of CD1a+ 7-ADD+ DC (R1) and CD1a+ 7-ADD− DC (R2) reveals the increased granularity of the double-positive population, confirming the engulfment of cell bodies by DC. B, The immature DC were incubated at either 4 or 37°C with killed tumor cells generated from the indicated tumor cell lines, and the percentage of double-positive cells was measured by flow cytometric assay. Results are representative of four experiments.
Immature DC internalize tumor cell bodies. Subcellular localization of tumor cell bodies, DU145 engulfed in DC. a, 7-AAD-labeled cell body (red) incubated for 1 h in the presence of FITC-anti-CD1a-labeled DC (green); b and c, unlabeled DC incubated with a 7-AAD-labeled body (red) for 1 h (b) and 4 h (c) were fixed, permeabilized, and stained for MHC class II (green) to reveal class II compartments. Each field represents two serial confocal sections (25 × 25 μm, 1 μm apart); the lower panel shows the corresponding DIC images, revealing nuclei localization. Examination of fluorescence micrographs and the corresponding DIC images in two representative DC further indicate that tumor cell bodies are cointernalized with the FITC-CD1a mAb in large juxta-nuclear phagocytic vesicles.
Phagocytosis of killed tumor cells is an active, receptor-mediated process
Previous studies demonstrated that thrombospondin receptor CD36 and vitronectin receptor (αvβ3, CD51/CD61) mediate the engulfment of cell bodies from HeLa cells or influenza-virus infected monocytes by immature DC (21, 22). Our results confirm that phagocytosis of killed tumor cells is an active process that is temperature dependent and requires intact cytoskeleton, as immature DC failed to capture Jurkat cell bodies when incubated at 4 or 37°C in the presence of cytocalasin D (Fig. 3⇓A). As shown in Fig. 3⇓A, both anti-CD36 and anti-αvβ3 Abs were able to partially inhibit, either alone or in combination, the engulfment of tumor bodies by immature DC. However, neither annexin V nor isotype-matched control mAbs blocked the uptake of tumor bodies by DC (Fig. 3⇓A). Finally, while immature DC homogeneously express CD36, only a fraction expressed vitronectin receptor (Fig. 3⇓B), suggesting a heterogeneity of DC with regard to the receptors used for capture of tumor bodies.
Immature DC use CD36 and vitronectin receptors to capture tumor bodies. A, FITC-CD1a-labeled immature DC were pretreated with cytochalasin D (20 μm) or 5 μg/ml of various mAb for 30 min and coincubated with 7-AAD-labeled killed tumor cells at 37°C for 1 h. The percentage of double-positive cells was analyzed by flow cytometry. Results show the percentage of inhibition compared with untreated cells. B, CD36 and vitronectin receptor expression on immature DC were measured by flow cytometry after staining of DC with anti-αvβ3 (clone 23C6, PharMingen) and anti-CD36 (clone CB38, PharMingen). Results are representative of two experiments.
Tumor body-loaded DC induce MHC class I- and class II-restricted proliferation of autologous CD4+ and CD8+ T cells
We next determined whether DC that captured killed allogeneic tumor cells can present tumor cell-derived Ags, whether alloantigens or tumor-specific Ags, as shown by induction of autologous T cells proliferation. To this end, DC were loaded with tumor cell bodies, sorted, and used as stimulators for autologous T cells, including unseparated PBMC, purified (>90%) CD4+ T cells, and purified (>90%) CD8+ T cells. As shown in Fig. 4⇓A, immature DC loaded with Jurkat lymphoma bodies induced the proliferation of autologous CD4+ T cells. Furthermore, DC loaded with DU145 prostate carcinoma cell bodies induced the proliferation of autologous CD8+ T cells (Fig. 4⇓B). Initial experiments indicated that the induction of DC maturation by exogenous CD40L and the addition of IL-2 were necessary to induce maximal proliferation of purified CD8+ T cells (data not shown). In addition, induced T cell proliferation was not due to autoreactivity, because DC loaded with killed nontransformed B cells autologous to DC and T cells resulted in no T cell proliferation (comparable counts per minute of <600 at 5000 DC/well to background were obtained). In blocking experiments CD4+ and CD8+ T cell proliferation induced by immature DC loaded with killed malignant melanoma, Colo 829, or prostate carcinoma, LNCap, cells was strongly inhibited by the presence of mAbs against MHC class II and class I molecules, respectively (Fig. 4⇓, C and D). Although macrophages were more potent in capturing EBV-transformed cell bodies than immature DC, they failed to elicit CD8+ T cell responses (data not shown). These results demonstrate that loading of immature DC with various killed tumor cells initiates both MHC class I- and class II-restricted T cells proliferation.
Immature DC loaded with tumor cell bodies induce the proliferation of autologous CD4+ and CD8+ T cells. FITC-CD1a-labeled immature DC were cocultured with killed tumor cells (A, Jurkat lymphoma; B, DU145 prostate carcinoma) for 1 h at 37°C and sorted based on CD1a expression. The sorted cells were cocultured for 5 days with autologous purified CD4+ T cells (A) or CD8+ T cells (5 × 104 cells/well; B). After 4 days of incubation, tritiated thymidine was added, and thymidine incorporation was measured 16 h later. The experiments assessing the proliferation of CD8+ T cells were conducted in the presence of CD40L (200 ng/ml) to induce DC maturation and IL-2 (5 U/ml) to enhance T cell proliferation. Results are representative of three experiments, and each value represents the mean from triplicate wells. FITC-CD1a-labeled immature DC were cocultured with killed tumor cells (C, Colo 829 melanoma; D, LNCap prostate carcinoma) for 1 h at 37°C and sorted based on CD1a expression. The sorted cells were cocultured for 5 days with autologous purified CD4+ T cells (C) or CD8+ T cells (5 × 104 cells/well; D) in the presence of mAbs against MHC class I or II molecules before the onset of cultures and subsequently during the entire culture period. After 4 days of incubation, tritiated thymidine was added, and thymidine incorporation was measured 16 h later. Results are representative of two experiments, and each value represents the mean from triplicate wells.
Tumor cell body-loaded DC induce CTL to shared Ags
To establish whether the loading of DC with tumor cell bodies allows expansion of CTL with cytotoxic activity against the immunizing tumor cell line, CD8+ T cells were cocultured with autologous DC loaded with DU145 prostate carcinoma cell bodies (two cycles of 7 days). The cytotoxic activity of expanded CD8+ T cells was determined using either the DU145 prostate carcinoma cell line or NK-sensitive K562 cells as targets. As shown in Fig. 5⇓A, the expanded T cells could kill the prostate carcinoma cells used for immunization, but not the K562 target, thereby demonstrating CTL rather than NK activity. CTL with equivalent cytotoxic activity were also expanded against the immunizing tumor cell line when total PBMC, rather than purified CD8+ T cells, were used as responder cells (Fig. 5⇓A).
Tumor cell body-loaded DC induce CTL to shared Ags. A and B, FITC-CD1a-labeled immature DC were loaded with prostate carcinoma (DU145 or PC3)-derived bodies for 1 h at 37°C, sorted based on CD1a expression, and used as stimulator cells. The cultures were set in a 24-well plate by plating stimulator cells with autologous purified CD8+ T cells. After 7 days of culture, the cells were harvested, washed, and replated with the same stimulator cells for 7 additional days. The cytotoxic activity of expanded CD8+ T cells was assessed in a standard 51Cr release assay using sensitizing cells, nonsensitizing cells, and K562 as targets. A, CD8+ T cells primed by DC loaded with DU145 cell bodies display cytotoxic activity against DU145 prostate carcinoma used for priming, but not against NK-sensitive K562 cells. CTL with equivalent cytotoxic activity was also obtained with total PBMC. B, DC loaded with killed PC3 prostate carcinoma cells induce CTL that are able to lyse the sensitizing PC3 cells and the DU145 cells that were not used for sensitization, indicating the presence of shared Ags. The percent cytotoxicity is measured as a function of spontaneous and total release. Results are representative of three experiments, and each value represents the mean from triplicate wells.
Next, we determined whether the loading of DC with one allogeneic tumor cell line allows expansion of CTL able to kill another allogeneic tumor cell line. DC were loaded with PC3 prostate carcinoma cell bodies (Fig. 5⇑B), and the expanded CTL were able to kill the PC3 cells used for sensitization as well as another prostate carcinoma cell line, DU145. In addition, the CTL activity elicited by DC loaded with DU145 prostate carcinoma cell bodies was able to kill PC3 cells (up to 23% specific lysis at a 30:1 E:T cell ratio). This suggests that DC loaded with killed tumor cells elicit CTL with cytotoxic activity against other tumor cells expressing shared Ags, whether alloantigens or tumor-specific Ags.
DC loaded with bodies of the prostate carcinoma LNCap cell line induce PSA-specific CTL
The next fundamental question was whether the allogeneic tumor cell-loaded DC can elicit tumor Ag-specific CTL. To address this question we used the allogeneic prostate carcinoma cell line LNCap that expresses a known tumor-specific Ag PSA (23, 24). As shown in Fig. 6⇓, immunofluorescence staining confirmed the cytoplasmic expression of PSA in LNCap cells (Fig. 6⇓, b and d), but not in PSA-negative control cells (Fig. 6⇓, a and c). Immature DC generated from HLA-A201 donors were cocultured with LNCap cell bodies, sorted, and used as stimulators of autologous CD8+ T cells. After three cycles of stimulation, the responder cells were harvested, and their cytotoxic activity and specificity were determined using LNCap cells, K562 cells, and PSA peptide-pulsed T2 cells as targets. The binding of PSA peptides to T2 cells was confirmed by flow cytometry (data not shown). As shown in Fig. 7⇓, the elicited CTL (three cycles of stimulation) were able to lyse LNCap cells, but not NK-sensitive K562 target cells. More importantly, the CTL generated were able to kill PSA-pulsed T2 cells, with up to 50% of specific lysis (5-fold increase), but neither T2 cells nor T2 cells loaded with control influenza matrix peptide showing TAA specificity. In addition, the elicited CTL showed increased killing of the PSA peptide-pulsed HLA-A201+ allogeneic melanoma cell line and autologous mature DC compared with unpulsed controls (data not shown).
PSA expression in the LNCap prostate carcinoma cell line. Immunofluorescence labeling of PSA using FITC-labeled mAb NCL-PSA (Novocastra Laboratories, Burlingame, CA) in the PSA-negative control PC3 cell line (a and c) and LNCap cell line (b and d). Each field represents the sum of two serial confocal sections (85 × 75 μm), and the lower panels show the corresponding DIC images.
DC loaded with bodies of the prostate carcinoma LNCap cell line induce PSA-specific CTL. FITC-CD1a-labeled immature DC, generated from HLA-A201-positive donors, were loaded with prostate carcinoma (LNCap)-derived bodies for 1 h at 37°C, sorted based on CD1a expression, and used as stimulator cells. The cultures were set in a 24-well plate by plating stimulator cells with autologous purified CD8+ T cells. The cytotoxic activity of expanded CD8+ T cells was assessed, after 3 cycles of stimulation, in a standard 51Cr release assay using sensitizing LNCap cells, T2 cells pulsed with PSA peptides or control peptide, and NK-sensitive K562 as targets. The percent cytotoxicity is measured as a function of spontaneous and total release. Results are representative of two experiments, and each value represents the mean from triplicate wells.
These results demonstrate that DC loaded with allogeneic tumor cell bodies (LNCap) are able to elicit immune responses to tumor-specific Ag (PSA) even in the presence of dominant alloantigens.
Discussion
Our study demonstrates that the killed allogeneic tumor cells can be used to load DC to elicit MHC class I-restricted CTL able to kill tumor cells as well as targets loaded with tumor-specific Ag-derived peptides. The current results therefore confirm and extend earlier findings of Albert (18, 19) that 1) DC that capture killed influenza virus-infected cells can activate lymphocytes to mount virus-specific CTL responses; and 2) DC that capture killed tumor cells overexpressing TAA can present Ag to tumor peptide-specific T cell lines. Our first observation is that bodies of an allogeneic prostate carcinoma cell line, PC3, given to DC permit the activation of T cells specific for Ags (alloantigens and/or tumor-specific Ags) expressed by another prostate carcinoma, DU145. The demonstration of CTL activity against PSA peptides using a third allogeneic prostate cancer cell line, LNCap, proves that the presence of alloantigens does not prevent expansion of CTL with activity against tumor-specific Ags.
The loaded DC can also induce MHC class II-restricted CD4+ T cell proliferation. This is of particular importance, because CD4+ T cells can exert helper function for the induction and maintenance of tumor-specific CTL and have direct effector function against MHC class II-positive tumors as well as an indirect one by activation of other cells, e.g., macrophages. Thus, the use of allogeneic tumor cell lines may, in fact, provide help for the induction of tumor-specific immunity through CD4+ T cell responses, as mice immunized with allogeneic tumor cells expressing viral or tumor Ags develop protective immune responses (25, 26). In particular, mice immunized with B16 melanoma cells engineered to express allo-Ag were 60% tumor free upon tumor challenge compared with only 20% tumor-free mice immunized with wild-type B16 melanoma cells (26).
Our study indicates that only a fraction (up to 45%) of immature DC capture killed tumor cells. This partial capture does not seem to be due to a lack of substrate, as many labeled killed tumor cells remain unengulfed by DC. In addition, our studies confirm previous reports (21, 22) showing that engulfment of dead cells by DC is partially mediated by the thrombospondin receptor CD36 and the vitronectin receptors. Thus, the heterogeneity in DC surface receptor expression allowing for recognition/capture of killed tumor cells could be one of the factors. As both apoptosis and necrosis can occur during in vitro killing of tumor cell lines, future experiments will be performed to establish which bodies efficiently contribute to DC loading.
Major problems in tumor immunotherapy are 1) the limited number of well-defined TAA, and 2) the lack of evidence that the known TAA actually represent rejection Ags in vivo. Furthermore, the use of MHC class I binding peptides is associated with 1) the HLA restriction and 2) the limitation of induced immune responses to CD8+ T cells. In this context, the use of unfractionated antigenic material, in the form of killed allogeneic tumor cells, that provides both MHC class I and class II epitopes leading to a diversified immune response represents an attractive alternative. The possibility of using allogeneic tumor cells for targeting of TAA to DC to induce generation of TAA-specific CTL has considerable implications. First, we will be able to use allogeneic killed cells to load DC to induce both CD8+ and CD4+ T cell-mediated anti-tumor responses in patients. Furthermore, it may permit diversification of responses and recruitment of non-Ag-specific effectors, such as eosinophils, macrophages, and NK cells. These well-characterized Ag sources can permit a more rigorous clinical assessment than the ill-defined autologous tumor preparations that cannot permit standardization and are often limited in their quantity and purity. An important consequence of our novel results is that the tumor Ags do not need to be overexpressed as with transfected cells or transgenic animals. Finally, this strategy offers the possibility to identify novel shared tumor Ags that may escape detection using autologous tumors due to immunodominance of specific mutated tumor Ags unavailable from the analysis of T cell clones reacting with autologous tumors.
Acknowledgments
We thank Dr. Nina Bhardwaj for sharing data before publication, Dr. Tyler Curiel for discussion on the establishment of CTL assays, Dr. Joseph Fay for discussion, and Dr. John Fordtran for constant support.
Footnotes
-
↵1 Presented partly at the Fifth International Symposium on Dendritic Cells in Fundamental and Clinical Immunology, Pittsburgh, PA, September 23–28, 1998. This work was supported in part by Grant CA78846-01A1 as well as awards from the CapCURE Foundation and the Ligue Nationale de Recherche sur le Cancer, Axe Immunologie des Tumeurs.
-
↵2 Address correspondence and reprint requests to Dr. Karolina Palucka, Baylor Institute for Immunology Research, 3434 Live Oak, Dallas, TX 75204. E-mail address: ak.palucka{at}baylordallas.edu
-
↵3 Abbreviations used in this paper: DC, dendritic cells; TAA, tumor-associated Ags; CM, complete culture medium; 7-AAD, 7-aminoactinomycin D; PSA, prostate-specific Ag; DIC, differential interference contrast; CD40L, CD40 ligand.
- Received February 8, 2000.
- Accepted July 14, 2000.
- Copyright © 2000 by The American Association of Immunologists
References
In this issue
