PeproTech Inc.

This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fujii, S.-i.
Right arrow Articles by Kawano, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fujii, S.-i.
Right arrow Articles by Kawano, F.
(Journal of Leukocyte Biology. 2003;73:82-90.)
© 2003 by Society for Leukocyte Biology

Malignant counterpart of myeloid dendritic cell (DC) belonging to acute myelogenous leukemia (AML) exhibits a dichotomous immunoregulatory potential

Shin-ichiro Fujii, Kanako Shimizu, Fujimoto Koji and Fumio Kawano

The Center for Bone Marrow Transplantation and Immunotherapy, Institute for Clinical Research, Kumamoto National Hospital, Japan

Correspondence and current address of Shin-ichiro Fujii, M.D., Ph.D.: Laboratory of Cellular Physiology and Immunology, The Rockefeller University, 1230 York Avenue, New York, NY 10021-6399. E-mail: fujiis{at}mail.rockefeller.edu


arrow
ABSTRACT
 
Dendritic cells (DCs) play a central role in immune regulation. Some leukemic cells are argued to be malignant counterparts of DC because of their ability to differentiate into leukemic DC. We characterize DC-like leukemia homogenously expressing CD11c+CD86+ in acute myelogenous leukemia patients. They express the Wilms’ tumor-1 antigen and common DC phenotypes (i.e., fascin+, CD83+, and DR+) directly. Purified leukemic cells produce interleukin-12 (IL-12) simultaneously with Fas ligand (FasL) and IL-6, which may suppress T cell-mediated immunity. These cells can elicit strong allogeneic T cell responses as well as induce tumor-specific CD8+ cytotoxic T cells, suggesting that they effectively present tumor-associated antigens. In contrast, they drive primary T cells toward apoptosis mediated in a tumor-specific way by a Fas-FasL interaction. Taken together, DC-like leukemia uniquely influences immune surveillance in contradictory ways, some of which may be involved in the mechanism of immune escape.

Key Words: Fas ligand • apoptosis • cytotoxic T cells


arrow
INTRODUCTION
 
Dendritic cells (DCs) can be differentiated from normal monocytes or normal CD34-positive hematopoietic stem cells within 7–14 days in the presence of granulocyte macrophage-colony stimulating factor (GM-CSF) and tumor necrosis factor {alpha} or interleukin (IL)-4 [1 , 2 ]. DCs are potent initiators of T cell-mediated immune responses [3 4 5 ] and when pulsed with tumor antigens, can serve as effective cancer vaccines in vitro and in vivo [6 7 8 9 ]. Circulating DCs in peripheral blood (PB) are classified into three subsets, which have different phenotypes and functions [10 ].

The PBMC of some patients with acute myelogenous leukemia (AML) or chronic myelogenous leukemia (CML) with addition of cytokines can sometimes be induced to differentiate into a newly elucidated, malignant counterpart of myeloid DCs, based on phenotypic and functional characteristics, such as the ability to stimulate antigen-specific cytotoxic T cells (CTLs) [11 12 13 14 15 ]. As a result, these leukemic cells have been envisioned as useful therapeutic reagents for treatment in patients. We have previously characterized CML-derived DCs carrying the Philadelphia (Ph) chromosome and have successfully used them in therapeutic cancer vaccines. Objective responses were evident in treated patients, with regression of Ph1+ cells in PB, as measured by fluorescence in situ hybridization analysis, in conjunction with the expansion of specific T cells in vivo [16 ].

In contrast to other AML blast-derived DCs [12 13 14 ], the current study indicates that DC-like leukemia already expresses some DC characteristics without the requirement of ex vivo culture in the presence of cytokines. According to the French-American-British (FAB) classification, DC-like leukemic cells may be similar to blasts of mature monocytic leukemia [M5b, defined as follows: Differentiated monocytic leukemic cells with a large cerebriform nucleus are found in a higher proportion in the PB than in the bone marrow (BM)] [17 , 18 ]; however, these cells are unique in phenotype and function. These leukemic cells express markers of DCs, i.e., human leukocyte antigen (HLA) class II and various costimulatory molecules, such as B7-1, B7-2, and intercellular adhesion molecule-1, and have a similar T cell stimulatory capacity. Therefore, it is poorly understood why these cells evolve into a malignant disease.

By interacting with the other immune-regulatory cells, leukemic cells may escape from immune surveillance by various suppressive factors prohibiting protective/therapeutic antitumor immunity. Therefore, it is important to further characterize DCs of leukemic origin for possible use in clinical trials. Furthermore, the analysis of the characteristics of DC-like leukemia may provide important insight into DC biology in the cancer setting.


arrow
MATERIALS AND METHODS
 
Isolation of DC-like leukemic cells
Cell samples from four patients were collected at the time of AML diagnosis or within the next 2 days. PB and BM samples for this study were collected from patients and normal donors concurrently with clinical test samples after obtaining informed consent. PB mononuclear cells (PBMC) were isolated using Ficoll-Hypaque (Pharmacia, Uppsala, Sweden) density gradient separation, and the resulting cells were cultured or cryopreserved. Leukemic cells were enriched using anti-CD11c monoclonal antibody (mAb; Immunotech, Marseille, France) and immunomagnetic separation using CELLectionTM Pan Mouse immunoglobulin G (IgG) kits (Dynal, Oslo, Norway). For some experiments, further purification of CD11c+ cells was performed as follows: One million immunomagnetically selected CD11c+ cells were incubated for 30 min at 4°C with fluorescein isothiocyanate (FITC)-conjugated mAb to CD11c (mouse IgG, PharMingen, San Diego, CA). Mouse IgG1 (FITC+/RPE+/RPE-Cy5+, Dako A/S, Glostrup, Denmark) was used as the isotype control. The cells were again washed twice, resuspended in phosphate-buffered saline (PBS) containing 0.1% bovine serum albumin, and kept at 4°C before cell sorting on a FACS Vantage (Becton Dickinson, Duarte, CA).

To generate monocyte-derived DCs from healthy volunteers as a control, PBMCs were allowed to adhere to 75 cm2 tissue-culture flasks (Corning Inc., Corning, NY) for 2 h at 37°C. After washing with PBS four times, adherent cells at a concentration of 2 x 106/mL were cultured for 8 days in 75 cm2 tissue-culture flasks containing RPMI-1640 media (Gibco, Grand Island, NY) supplemented with 2 mM glutamine (Gibco), 10% heat-inactivated human AB serum, and the following human cytokines: recombinant (r) GM-CSF (Genzyme, Cambridge, MA) and rIL-4 (Genzyme), each at a final concentration of 1000 U/ml. Cultures were maintained for 7 days.

Phenotypic analysis of cells by flow cytometry
Leukemic cells from PB of patients and cultured DCs derived from normal, healthy donors were washed with PBS-human (h)IgG and were stained with the following mAb: FITC-CD11c and PE-conjugated mAb against CD1a, CD11c, CD44, CD80, CD86, CD40, GM-CSF receptor (R), IL-3R, interferon-{gamma} (IFN-{gamma})R, HLA-DR, and FITC- or PE-labeled isotype controls were purchased from Becton Dickinson. Stained leukemic cells were then washed with PBS-hIgG and analyzed on the flow cytometer using CELL Quest software (Becton Dickinson).

Evaluation of cytokine production by enzyme-linked immunosorbent assays (ELISAs)
CD11c+-sorted leukemic cells (5x105) were cultured in 1 ml medium alone or in medium supplemented with soluble (s)rCD40 ligand (provided by Immunex Research and Development, Seattle, WA). Culture medium was removed after 72 h, and the level of secreted cytokines evaluated by specific ELISA assays was: IL-10 (sensitivity 1 pg/mL), IL-6 (sensitivity 0.156 pg/mL), vascular endothelial growth factor (sensitivity 5 pg/mL), transforming growth factor (TGF)-ß (sensitivity 15.6 pg/mL), IL-12 p40 and p70 (15 pg/mL), and Fas ligand (FasL; sensitivity 0.1 ng/mL; FasL from MBL, Nagoya, Japan; all others from R & D Systems, Minneapolis, MN).

Allogeneic mixed lymphocyte reaction (Allo-MLR)
Naïve T cells were obtained from heparinized blood of unrelated healthy volunteers by density gradient centrifugation and removal of nylon wool-adherent cells. T cells (5x105 cells/mL) were cultured together with 30-Gy-irradiated stimulator cells (up to 2x105/mL), DC-like leukemic cells, normal DC, or 2-h plastic-adherent macrophages from the other subjects and were cultured for 6 days in 96-well U-bottomed culture microplates. For the final 16 h of culture, 1 µCi [3H]-thymidine (Amersham, Buckingham, UK) was added to each well.

Establishment of antileukemia CTL line and T cell cytotoxicity assays
For the generation of the CTL line, naïve T cells of PBMCs from patients in complete remission 2 months after chemotherapy were used. PBMC were suspended in 1 x 106 cells/ml and cocultured with 1 x 105/ml irradiated leukemic cells in the culture medium [45% RPMI-1640 medium, 45% AIM-V medium (Gibco-BRL), and 10% heat-inactivated human AB serum with 100 units/ml IL-2 and 0.1 mM minimum essential medium nonessential amino acids solution (Gibco-BRL); for another 7 days. Cells were resuspended in fresh medium every 3–4 days. On day 7, irradiated, autologous PBMC were added to restimulate T cell responders at a ratio of three PBMC:one T cell. Subsequently, on days 14, 21, and 28, this procedure was repeated. After 35 days of culture, the CTL line was assayed for its antileukemic activity. To inhibit CTL activity, 20 µg/mL anti-HLA class I (W6/32, IgG2a) or anti-HLA class II (H-DR-1, IgG2a) was added to target cell wells for 1 h before coculture with T cells, or 20 µg/ml anti-CD4 (Nu-Th/I, IgG1) or anti-CD8 (Nu-Ts/c, IgG2a) was added to effector cells before coculture with target cells. For negative-target cells, Epstein-Barr virus (EBV)-transformed lymphoblastoid cell line (LCL) or phytohemagglutinin (PHA) blasts (1 µg/mL) were used. T cell cytotoxicity assays were performed as described previously [4 ].

Immunohistochemistry and terminal deoxyribonucleotidyl transferase (TdT)-mediated dNTP nick end-labeling (TUNNEL) assay
Leukemic cells isolated by CD11c+ mAb and magnetic beads were analyzed by immunohistochemistry as well as by May-Giemsa and esterase staining. Single-parameter immunostaining of cytosmears of the chamber slides was performed using the following mAb: p55 (gift from Dr. Erik Langhoff, Massachusetts General Hospital, Boston), Wilms’ tumor-1 (WT-1), CD80, CD83, CD86, and Fas (PharMingen). The indirect immunoalkaline-phosphatase method was performed with an alkaline-phosphatase substrate kit I (Vector red, Vector Laboratories, Burlingame, CA), as described previously [4 ]. For evaluating the apoptosis of T cells by tumor cells, CD3+ T cells were isolated using anti-CD3 mAb and CELLectionTM mouse IgG kit (Dynal), cocultured with tumor cells at a 1:10 ratio for 24 h and were stained by TUNNEL assay. Detection of DNA fragments in situ was performed by an Oncor Apop Tag kit (Oncogene Research Products, Cambridge, MA) using TUNNEL assays. In this assay, TdT binds to exposed 3'-OH DNA termini in situ, which are preferentially found in apoptotic bodies of DNA fragments, where it catalyzes a template-independent addition of nucleotide triphosphates to the 3'-OH ends of double- or single-strand DNA. DNA fragments, which have been labeled with digoxigenin-nucleotide, are then allowed to bind an antidigoxigenin antibody conjugated to peroxidase. 3'-Diaminobenzidine tetrahydrochloride was used as a substrate for labeled samples to generate an insoluble chromogen at the site of DNA fragmentation. Double-immunostaining was used to allow for surface phenotyping in conjunction with the Oncor Apop Tag kit to further characterize the interaction of tumor cells and apoptotic T cells.

DNA fragmentation and specific killing using the JAM test
Target cell (T cell) death by coculturing with effector cells (monocyte-derived DC or CD11c+-sorted leukemic cells) was quantified by measuring target-cell DNA fragmentation using the JAM test [19 ]. To determine the effect of T cells on leukemic cells, CD3+ T cells were labeled with [3H]-thymidine and challenged with exposure to unlabeled leukemic cells at the various, indicated effector cell/target cell (E/T) ratios. For this purpose, PBMC from patients in complete remission were activated in culture medium supplemented with 10 ng/mL OKT3 (Ortho Biotech, Raritan, NJ) and 500 U/mL IL-2 (Shionogi Pharm. Co., Osaka, Japan) for 10 days and then were positively isolated using CD3 mAb and CELLection mouse IgG kit (Dynal) followed by incubation with 1 µCi/mL [3H]-thymidine (Amersham) for 5 h, washing three times with PBS, and finally, resuspending the cells in culture medium. Cocultivation of cells was performed for 72 h at 37°C. T cell suspensions (100 µL; 2x104/well) were cocultured in 96-well plates, with or without 100 µL suspension of the relevant leukemic cells (from 2x105/mL to 4x106/mL). For blocking assays, leukemic cells (2x104/mL) were preincubated with 2.5 µg/mL anti-FasL mAb (NOK2, PharMingen) or relevant negative control IgG2a mAb (Dako), respectively, 1 h before cocultivation with the labeled T cells (2x105/mL) in the absence or presence of 0.25 µg/mL of the death-inducing anti-Fas mAb (clone CH11, MBL), which served as a positive control for Fas-mediated apoptosis. Cells were isolated automatically, transferred onto filter papers, and washed six times. The incorporated radioactivity from undegraded chromosomal DNA was then measured with a ß-scintillation counter. The reduction in incorporated radioactivity was used to calculate the percentage of specific target-cell killing [(cpm untreated cells-cpm cocultured cells)/cpm untreated cellsx100].

Statistical analysis
Statistical analysis in our experiments was performed using standard Student’s t-tests.


arrow
RESULTS
 
Characteristics of DC-like leukemic cells in patients
Four patients were clinically diagnosed with AML (M5b) according to the FAB classification, and tumor cell samples were collected at the time of diagnosis or within 48 h of diagnosis. Clinical features in these patients, including extremely high numbers of leukemic cells as well as each phenotype of leukemic cells before chemotherapy, are described in Table 1 . Leukemic cells in all the patients had no detectable chromosomal translocations. Systemic erythematous nodules containing tumors were evaluated in all patients (data not shown). We performed a CD45/side-scatter blast-gating method for detecting the frequency of leukemic blasts (Table 1) [20 ]. In addition, we identified these tumor cells as the population of CD1a-CD11c+ DC-like leukemia by using two-color analyses. Flow cytometric analysis showed that these leukemic cells were a homogeneous population and expressed HLA-DR, CD11c, CD40, CD86, and CD80, but not the CD14 marker without any in vitro culture, but did not have any T cell or B cell markers (Table 1 ; Fig. 1 ). The expression of CD11c on the surface of these cells provided a useful means to further enrich these leukemic cells for subsequent examination of their phenotype. In de novo leukemias shown here, more than 98% of CD11c+ cells also expressed DC phenotype. Positively isolated CD11c+-leukemic cells also expressed p55, CD83, Fas, and WT-1 markers without ex vivo culture (Fig. 2 ). CD11c+-isolated leukemic cells appeared as dispersed, spherical cells with short projections resembling myeloid DC.


View this table:
[in this window]
[in a new window]
  		Table 1. Phenotypic characteristics of leukemia patients



View larger version (30K):
[in this window]
[in a new window]
  		Figure 1. DC-like leukemia expresses DC surface molecules, CD11c, CD86, and HLA-DR. PBMC were isolated using Ficoll-Hypaque density gradient separation, stained with mAb directed against DC surface molecules, and analyzed by two-color flow cytometry. They expressed the HLA-DR, CD80, CD86, CD40, and CD44 but not the CD1a and CD14 molecules. They also expressed receptors for GM-CSF, IL-3, and IFN-{gamma}. Data are representative of four patients.



View larger version (94K):
[in this window]
[in a new window]
  		Figure 2. Microfluorographs of cytosmears of CD11c+ cells. CD11c+ leukemic cells were positively isolated from PBMC using anti-CD11c antibody and immunomagnetic separation using the CELLectionTM Pan Mouse IgG kit (Dynal). After immunostaining isolated leukemic cells for isotype control (a), 11c (b), fascin, p55 (c), CD80 (d), WT-1 (e), CD86 (f), CD83 (g), and Fas (h), areas of low cellularity were photographed. Note that leukemic cells express the FasR (CD95), WT-1, and CD83 as well as CD11c, CD80, and CD86 markers. A subset of cells express the p55+ antigen.

Stimulation of MLR by DC-like leukemia cells is comparable with normal DC but not macrophages
Normal monocyte-derived DC from healthy donors and DC-like leukemia cells from BM and PB in leukemia patients were prepared by isolation with anti-CD11c mAb and magnetic beads. Stimulatory activities for allogeneic lymphocytes were demonstrated by coculturing allogeneic T cells with 2-h plastic-adherent macrophages, normal DC, or DC-like leukemia cells. DC-like leukemia cells as well as normal DC showed stronger allo-MLR activities than plastic-adherent macrophages (Fig. 3 ). When we took two DC populations, i.e., malignant DCs versus healthy DCs, from the same patient (patient B) in the de novo stage and the complete remission state, both stimulated a strong MLR response (data not shown). The activity of allo-MLR by DC-like leukemia cells from PB or BM behaved similarly.



View larger version (20K):
[in this window]
[in a new window]
  		Figure 3. Stimulation of MLR by DC-like leukemic cells is comparable with normal DC but not macrophages. Normal monocyte-derived DC and DC-like leukemia cells were isolated by magnetic beads as described in Figure 2 . Two-hour plastic-adherent macrophages, normal DC, or DC-like leukemia cells were irradiated, cultured with allogeneic naïve T lymphocytes, and labeled with [3H]-thymidine for the final 16 h of culture. Alloantigen-presenting capacity of these leukemic cells or normal DC were compared with macrophages. Data are representative of four patients. Each point is the mean ± SD of three measurements.

Cytokine production by monocyte-derived DCs and DC-like leukemia cells
We assayed cytokine production from CD11c+-sorted leukemic cells versus normal DCs and found that IL-10 production was significantly lower in leukemic cells than normal, monocyte-derived DCs but that IL-12 production (p40 and p70) was comparable between the two DC types (Fig. 4 ). After stimulation with rsCD40L, IL-12 production was enhanced in both groups again to a comparable level of production. Elevations of IL-6 and FasL production were observed in leukemic cells. Overall, there was no significant difference in IL-12 or TGF-ß secretion (data not shown) in these two-cell populations. Lastly, VEGF was not produced by any DC evaluated (data not shown). The production of these cytokines was below the detection threshold for other AML blasts (data not shown; n=4; one AML-M1 patient, two M2 patients, one M4 patient).



View larger version (21K):
[in this window]
[in a new window]
  		Figure 4. DC-like leukemic cells produce IL-12 as well as IL-6 and Fas ligand. Normal monocyte-derived DC and DC-like leukemia cells were isolated by CD11c mAb using a FACS Vantage (Becton Dickinson). Cytokine production was evaluated after culture for 72 h. Leukemic cells preferentially produced IL-6 and sFasL. IL-12 production (p40 and p70) and TGF-ß (data not shown) were comparable with that produced by normal monocyte-derived DC, in the presence or absence of rsCD40L. IL-10 production by leukemic cells in the presence or absence of CD40L stimulation was less than that noted from monocyte-derived DC. Data are mean ± SEM of four patients. Statistical differences in cytokine production were noted in the amounts of IL-12 p70, IL-6, IL-10 (P<0.05), and Fas L (P<0.01) produced between DC-type leukemia cells and normal DC.

Cytotoxicity of leukemic cell-stimulated T cells against autologous tumor targets
T cells were established by coculturing monocyte-depleted PBMC with irradiated, autologous CD11c+-sorted leukemic cells in IL-2-containing medium, yielding a uniform (94.0±2.1%) responder of CD8+ T cells. Less than 5% of the cells were CD4+ T cells, and 0.5% of the cells was CD56+ (data not shown). Figure 5 shows the cytotoxic potential of this T cell effector population against autologous leukemic cell targets. At an E/T ratio of 25:1, 80% cytotoxicity was noted against leukemic cells (patient A), and only very low levels of killing were observed against autologous PHA-blast cells or autologous EBV-transformed LCL (Fig. 5) . However, culture supernatant harvested from the T cell line did not mediate cytotoxicity against leukemic targets in 4 h incubation (data not shown). CD8+ CTL lines were restricted by major histocompatibility complex (MHC) class I. The results suggest that specific cytotoxicity against autologous leukemic cells was directed with MHC-class I restriction and was mediated by the effector CD8+ T cell population. Directly isolated CD3+ T lymphocytes harvested from the PB of patients did not exhibit the cytotoxicity against tumor cells in in vitro assays (data not shown).



View larger version (18K):
[in this window]
[in a new window]
  		Figure 5. Cytotoxic assay against leukemic cells mediated by specific antileukemic T cells. Autologous T cell lines were established by coculturing T cells at a 10:1 ratio with autologous, irradiated leukemic cells for 35 days in the presence of IL-2 (100 U/ml). CTL lines were composed predominantly of CD8+ T cells (more than 80%). In patient A (Pt. A), the T cell line was tested against autologous-leukemic cells for patient A (a, left, {blacksquare}) and patients B, C, and D (b, {blacksquare}). In contrast, the T cell line was tested against autologous PHA blast for patient A (a, left, {square}) or EBV-transformed B-LCL for patients B, C, and D (b, {square}). In addition, blocking tests were performed as follows: Anticlasses I and II mAb were added to target cells (leukemic cells) and anti-CD4 mAb and anti-CD8 mAb were added to effector cells (T cell line) 1 h before coculturing. T cell-mediated lysis was inhibited by the specific anticlass I and anti-CD8 mAb (a, right).

Leukemic cells induce the death of autologous T lymphocytes as a result of the expression of FasL
T cell apoptosis induced by leukemic cells was evaluated by the TUNNEL assay and by immunohistochemical staining. CD3+ T lymphocytes cocultured with CD11c+-sorted leukemic cells of a leukemic cell/CD3+ T cell ratio greater than 10 exhibited a significant degree of apoptosis after 24 h culture (Fig. 6 ). For further assessing the tumor cell-mediated cytotoxic activity directed against naive T cells in a quantitative manner, we performed JAM test assays. After cocultivation of [3H]-thymidine-labeled T cells with unlabeled leukemic cells for 72 h, a significant decrease in T cell-specific [3H]-thymidine content was demonstrated, indicating the fragmentation of DNA in dying T cells (Fig. 7a ). Preincubation of target T cells with the Fas-blocking ZB4 mAb (0.25 µg/mL) or of leukemic cells with the FasL-neutralizing NOK-2 mAb (2.5 µg/mL; Fig. 7b ), respectively, significantly protected T cells from being killed by leukemic cells (15.3±4.7 vs. 13.2±1.1%, respectively). Isotype-matched control mAb used in the same concentrations had no protective effect (49.9±1.9%). Agonistic anti-Fas mAb (CH11)-induced T cell death registered 74.5 ± 0.9%, and spontaneous T cell death was 5.8 ± 2.1%. As with leukemic cell-mediated cytotoxicity, antibody treatment was unable to rescue all T cells from agonistic anti-Fas mAb-induced cell death, even when higher concentrations of the ZB4 mAb were applied. The CD11c- cell fraction from these patients as well as normal CD11c+ or CD11c- cells in healthy donors did not display this capacity (Fig. 7c) . Notably, the leukemic culture supernatant did not promote the death of T cells (data not shown).



View larger version (92K):
[in this window]
[in a new window]
  		Figure 6. Leukemic cells induce the death of T lymphocytes. Positively isolated T lymphocytes were cocultured with CD11c+-sorted leukemic cells by magnetic beads for 24 h, resulting in the formation of T cell/leukemic cell clusters. Within these clusters, apoptosis of T cells was observed as brown around DC-type leukemia cells (red) expressing CD86 (b), CD40 (c), and FasL (d) in two-color staining by immunostaining and TUNNEL assays.



View larger version (16K):
[in this window]
[in a new window]
  		Figure 7. T cell apoptosis by leukemic cells expressing FasL in JAM test assays. (a) [3H]-Thymidine-labeled antitumor T cells were cocultured with CD11c+-sorted leukemic cells by magnetic beads at the indicated ratios for 72 h, at which time thymidine incorporated in nondegraded DNA was counted in a ß-counter. The percent-degraded DNA is plotted for T cells cultured with leukemia cells ({blacksquare}) and control, monocyte-derived DC ({square}). Statistical differences were noted in both groups (P<0.01). Data are mean ± SEM of four patients. (b) Leukemia-mediated apoptosis of specific T cells (2x105 cells/mL) was analyzed by coculturing with CD11c+ leukemic cells (2x106/mL) for 72 h for the role of Fas/FasL using blocking agents: preincubation of T cells with isotype mAb (2.5 µg/mL); preincubation of leukemic cells with anti-FasL mAb, NOK-2 (2.5 µg/mL); and preincubation of T cells with anti-Fas mAb and ZB4 mAb (0.25 µg/mL). Agonistic anti-Fas mAb (CH11)-induced T cell death and spontaneous T cell death were shown, respectively. Statistical differences were noted in the group (isotype group vs. others<0.05*). Data are the mean ± SEM of three determinations of two patients (patients A and B). (c) T cell apoptosis by CD11c+ or CD11c- population (2x106/mL) was summarized in four patients. CD11c- population was obtained after positive selection of CD11c+ cells. As described above, CD11c+ cells from PB in leukemic patients, monocyte-derived DC from healthy donors, or CD11c- cells were cultured with T cells for 72 h. Data were mean ± SEM.


arrow
DISCUSSION
 
DC-like leukemic cells expressing CD11c+CD86+ markers have a unique phenotype with many functional similarities to DC. Moreover, DC-like leukemic cells exhibit an impaired ability to regulate immune responses.

Clinical manifestations included a large number of tumor cells present in BM, PB, and lymph nodes as well as systemic erythematous nodules in the skin (data not shown). These leukemic cells as a whole fresh-blast population are a homogenous population and express CD11c+, CD86+, and MHC-class II (HLA-DR) molecules. Although CD1a-CD11c+DC are extremely rare, representing ~0.1% of PBMC in PB in healthy donors [13 , 21 ], high numbers and frequencies of these DC are detected in the PB in these leukemia patients. This suggests that most of them may be leukemic cells in the de novo AML stage. We found that these leukemic cells, isolated by CD11c- mAb and magnetic beads, homogenously exhibit DC morphology and express DC markers, i.e., CD40+, CD80+, CD86+, and furthermore, p55+, CD83+. Also, and since almost all of them expressed the WT-1 antigen, they must have originated from tumor cells (Figs. 1 and 2) . These cells also dimly expressed receptors for GM-CSF, IL-3, and IFN-{gamma} (Fig. 1) . Moreover, these isolated tumor cells could produce IL-12 when appropriately stimulated, and they could stimulate a strong allo-MLR (Figs. 3 and 4) . The data described by Mohty et al. [13 ], in which myeloid leukemic-DCs derived from leukemic blasts in culture showed a strong MLR as well as expression of costimulatory molecules, support our findings. Of note, as shown in Figure 2 , these cells again displayed a tumor expression marker, WT-1 [22 , 23 ], and a marker of an aggressiveness of neoplasm, CD44 (Fig. 1) [24 ]. As WT-1 is associated with a poor, long-term, clinical outcome, and WT-1 peptides sometimes play a role as the tumor antigens, this could be a useful marker of leukemic DCs. As shown in Figures 1 and 2 , by isolating CD11c, we collected a high number of CD1a-CD11c+DR+CD86+ cells expressing WT-1+ as a homogenous population. The CD1a-CD11c+ subset of DCs in PB is believed to be similar to the monocyte-derived DCs, also known to be located in the germinal center of lymph nodes [10 , 25 ].

It is generally accepted that tumor cells are poorly immunogenic, as they are deficient in expression of surface MHC classes I and II as well as costimulatory molecules [26 ]. However, DC-like leukemia cells do express these molecules and would be expected to serve as good antigen-presenting cells (APCs). Indeed, we were able to readily generate autologous, tumor-specific, cytotoxic CD8+ T lymphocytes reactive in in vitro cultures using IL-2. Little T cell-mediated cytotoxicity was noted against autologous PHA-T blasts or EBV-transformed LCL obtained from these patients during remission. This strongly suggests that these tumor cells are immunogenic and have the potential to elicit an antitumor immunity (Fig. 5) [27 ]. In our approach, additional cytokines in culture were not required for DC-like leukemia, which induce specific CTL against tumor cells.

We also demonstrated using TUNNEL and JAM assays that T cells were susceptible to Fas-mediated killing by coculturing with leukemic cells, which expressed FasL on their cell surface (Figs. 6 and 7) . T cell death was inhibited by inclusion of antagonistic FasR mAb (ZB4) or neutralizing anti-FasL mAb (NOK-2) [28 , 29 ]. Inhibition of the immune system by FasL-expressing tumors may also occur within the tumor microenvironment, i.e., in colon cancer or squamous cell carcinoma [29 , 30 31 32 ]. FasL expressions by leukemic cells, such as acute lymphocytic leukemia AML, and CML, have been associated with more aggressive, biologic behavior. This is most compelling in the case of transformation of CML to blast crisis [33 , 34 ]. DCs transfected with FasL cDNA, so-called "killer" DCs, can effectively deliver death signals rather than activation signals to T cells after an antigen-specific interaction [35 ]. Thus, the injection of antigen-pulsed killer DC into mice can induce antigen-specific immunosuppression. These phenomena may lend support for our data. Although leukemic-plasmacytoid DC derived from leukemic blasts may lead to immunosuppressive effects in patients [13 ], in contrast, our findings suggested that myeloid DC leukemia also showed the immunosuppressive effects in some leukemia patients. In the current study, the elevated number of tumor cells observed during disease progression may be at least partially a result of the ability of these DC-like leukemia cells to cause T cell apoptosis through the Fas-FasL system, thereby "deleting" relevant immune surveillance [36 37 38 ]. As shown in Figure 7c , the leukemic CD11c+ DC fraction killed T cells, suggesting a mechanism of immune escape for leukemic DC. The expression of FasL by tumors provides them with an offensive weapon with which to elude, by counterattack, the immune system [29 ].

Thus, in the interaction between DC-like leukemic cells and T cells, we noted contradictory phenomena where the tumor can promote and silence specific T cells in vitro. These findings may imply that only T cells that escaped from being killed by tumor cells can survive as antitumor CTLs. These contradictory mechanisms may also be operating in vivo. Antitumor T cell generation in vivo may be suppressed in many ways, including tumor cell-induced T cell death or involving the FasL-mediated counterattack at the tumor sites in involved lymph nodes. As shown in Figure 4 , cytokines such as IL-6 and TGF-ß (data not shown) were also released from these leukemic cells, which may exert suppressive effects on CTLs and DCs in vivo [29 , 39 40 41 42 ]. IL-10 is known to be a key immunomodulatory cytokine capable of mediating supporting effects on CTL and immunosuppressive effects on macrophages, DCs, or CD4+ T cells [43 44 45 46 ]. However, as IL-10 produced by DC-like leukemic cells was less than that of ordinary DCs (Fig. 4) , it may not play an important role in this dichotomous immunoregulation.

Cumulatively, these data suggest a dichotomy in tumor functional immunogenicity. On one hand, DC-like leukemic cells can promote specific T cell induction, as they are like "professional" APC. Conversely, they can effectively delete responder T cells because of FasL expression. Further, unresponsive patients’ T cells might be elicited or expanded in situ by an encounter with the immunosuppressive milieu of the tumor at specific regional sites such as skin, PB, BM, lymph node, and spleen [43 , 44 ]. However, such tolerance may be reversed by adding rIL-2 for the induction of specific CTLs against tumor cells or by stimulation with allogeneic T cell responses, as is the case with activation-induced nonresponsiveness of T cells (Figs. 3 and 5) [47 , 48 ]. These findings suggest that despite the APC characteristics associated with certain tumors, the net results in situ may be a result of tumor-specific, peripheral tolerance. By further understanding the relevant, tumor-associated, suppressive mechanisms, these circuits may be abrogated clinically, thereby enhancing the underlying immunogenic nature of this malignancy, resulting in more effective immunotherapy.


arrow
ACKNOWLEDGEMENTS
 
We are grateful to Dr. Walter Storkus (University of Pittsburgh, PA) for peer-reviewing this manuscript. We thank Drs. Madhav Dhodapkar, Kara Bickham, and Caroline Smith (The Rockefeller Unversity, New York, NY) for helpful suggestions and reviewing.

Received June 1, 2002; revised October 10, 2002; accepted October 22, 2002.


arrow
REFERENCES
 
    1
  1. Caux, C., Dezutter-Dambuyant, C., Schmitt, D., Banchereau, J. (1992) GM-CSF and TNF-alpha cooperate in the generation of dendritic Langerhans cells Nature 360,258-261[CrossRef][Medline]
    2. 2
  2. Sallusto, F., Lanzavecchia, A. (1994) Efficient presentation of soluble antigen by cultured human dendritic cells is maintained by granulocyte/macrophage colony-stimulating factor plus interleukin 4 and downregulated by tumor necrosis factor alpha J. Exp. Med. 179,1109-1118[Abstract/Free Full Text]
    3. 3
  3. Banchereau, J., Steinman, R. (1998) Dendritic cells and the control of immunity Nature 392,245-252[CrossRef][Medline]
    4. 4
  4. Steinman, R., Inaba, K. (1999) Myeloid dendritic cells J. Leukoc. Biol. 66,205-208[Abstract]
    5. 5
  5. Banchereau, J., Briere, F., Caux, C., Davoust, J., Lebecque, S., Liu, Y. J., Pulendran, B., Paluka, K. (2000) Immunobiology of dendritic cells Annu. Rev. Immunol. 18,767-811[CrossRef][Medline]
    6. 6
  6. Fujii, S-I., Fujimoto, K., Matsui, K., Kawakita, M., Takatsuki, K. (1998) Induction of antitumor cytotoxic activity using CD34+ cord blood cell-derived and irradiated tumor cell-primed dendritic cells Int. J. Hematol. 68,169-182[CrossRef][Medline]
    7. 7
  7. Fujii, S-I., Fujimoto, K., Shimizu, K., Ezaki, T., Kawano, F., Takatsuki, K., Kawakita, M., Matsuno, K. (1999) Presentation of tumor antigens by phagocytic dendritic cell clusters generated from human CD34+ hematopoietic progenitor cells: induction of autologous cytotoxic T lymphocytes against leukemic cells in acute myelogenous leukemia patients Cancer Res 59,2150-2158[Abstract/Free Full Text]
    8. 8
  8. Fujii, S-I., Hamada, H., Fujimoto, K., Shimomura, T., Kawakita, K. (1999) Activated dendritic cells from bone marrow cells of mice receiving cytokine-expressing tumor cells are associated with the enhanced survival of mice bearing syngeneic tumors Blood 93,4328-4335[Abstract/Free Full Text]
    9. 9
  9. Fujii, S-I., Shimizu, K., Fujimoto, K., Tetsuyuki, K., Tsukamoto, A., Sanada, I., Kawano, F. (2001) Treatment of post-transplanted, relapsed patients with hematological malignancies by transfusion of HLA-matched, allogeneic-dendritic cells (DCs) pulsed with irradiated tumor cells and primed T cells Leuk. Lymphoma 42,357-369[Medline]
    10. 10
  10. Ito, T., Inaba, M., Inaba, K., Toki, J., Sogo, S., Iguchi, T., Adachi, Y., Yamaguchi, K., Amakawa, R., Valladeau, J., Sealand, S., Fukuhara, S., Ikehara, S. (1999) A CD1a+/CD11c+ subset of human blood dendritic cells is a direct precursor of Langerhans cells J. Immunol. 163,1409-1419[Abstract/Free Full Text]
    11. 11
  11. Santiago-Schwarz, F., Coppock, D. L., Hindenburg, A. A., Kern, J. (1994) Identification of a malignant counterpart of the monocyte-dendritic cell progenitor in an acute myeloid leukemia Blood 84,3054-3062[Abstract/Free Full Text]
    12. 12
  12. Choudhury, A., Liang, J. C., Thomas, E. K., Flores-Romo, L., Sutaria, Q. S., Shinha, I., Champlin, R. E., Claxton, D. F. (1999) Dendritic cells derived in vitro from acute myelogenous leukemia cells stimulate autologous, antileukemic T-cell responses Blood 93,780-786[Abstract/Free Full Text]
    13. 13
  13. Mohty, M., Jarrosay, D., Lafage-pochitaloff, M., Zandotti, C., Briere, F., de Lamballeri, X. N., Isnardon, D., Sainty, D., Gaugler, B. (2001) Circulating blood dendritic cells from myeloid leukemia patients display quantative and cytogenetic abnormalities as well as functional impairment Blood 98,3750-3756[Abstract/Free Full Text]
    14. 14
  14. Harrison, B. D., Adams, J. A., Briggs, M., Bereton, M. L., Yin, J. A. L. (2001) Stimulation of autologous proliferative and cytotoxic T-cell responses by "leukemic dendritic cells" derived from blast cells in acute myeloid leukemia Blood 97,2764-2771[Abstract/Free Full Text]
    15. 15
  15. Choudhury, A., Gajewski, J. L., Liang, J. C., Popat, U., Claxton, D. F., Kliche, K. O., Andreeff, M., Champlin, R. E. (1997) Use of leukemic dendritic cells for the generation of antileukemic cellular cytotoxicity against Philadelphia chromosome-positive chronic myelogenous leukemia Blood 89,1133-1142[Abstract/Free Full Text]
    16. 16
  16. Fujii, S-I., Shimizu, K., Fujimoto, K., Shimomura, T., Taniguchi, O., Kinoshita, M., Kawano, F. (1999) Analysis of a vaccinated chronic myelogenous leukemia patient with leukemic dendritic cells following autologous peripheral blood stem cell transplantation Jpn. J. Cancer Res. 90,1117-1129[CrossRef][Medline]
    17. 17
  17. Gralnick, H. R., Galton, D. A. G., Catovsky, D., Sultan, C., Bennett, J. M. (1977) Classification of acute leukemia Ann. Intern. Med. 87,740-753
    18. 18
  18. Khalidi, H. S., Medeiros, L. J., Chang, K. L., Brynes, R. K., Slovak, M. L., Arber, D. A. (1998) The immunophenotype of adult acute myeloid leukemia; high frequency of lymphoid antigen expression and comparison of immunophenotype, French-American-British classification, and karyotypic abnormalities Am. J. Clin. Pathol. 109,211-220[Medline]
    19. 19
  19. Bohm, C., Hanski, M. L., Gratchev, A., Mann, B., Moyer, M. P., Riecken, E. O., Hanski, C. (1998) A modification of the JAM test is necessary for a correct determination of apoptosis induced by FasL+ adherent tumor cells J. Immunol. Methods 217,71-78[CrossRef][Medline]
    20. 20
  20. Lacombe, F., Durrieu, F., Briais, A., Dumain, P., Belloc, F., Bascans, E., Reiffers, J., Boisseau, M. R., Bernard, P. (1997) Flow cytometry CD45 gating for immunophenotyping of acute leukemia Leukemia 11,1878-1886[CrossRef][Medline]
    21. 21
  21. Freundenthal, P. S., Steinman, R. M. (1990) The distinct surface of human blood dendritic cells, as observed after an improved isolation method Proc. Natl. Acad. Sci. USA 87,7698-7702[Abstract/Free Full Text]
    22. 22
  22. Bergmann, L., Miething, C., Maurer, U., Brieger, J., Karakas, T., Weidmann, E., Hoelzer, D. (1997) High levels of Wilms’ tumor gene (WT1) mRNA in acute myeloid leukemias are associated with a worse long-term outcome Blood 90,1217-1225[Abstract/Free Full Text]
    23. 23
  23. Ohminami, H., Yasukawa, M., Fujita, S. (2000) HLA-class I-restricted lysis of leukemia cells by a CD8+ cytotoxic T-lymphocyte clone specific for WT-1 peptide Blood 95,286-293[Abstract/Free Full Text]
    24. 24
  24. Siegelman, M. H., DeGrendele, H. C., Pila, E. (1999) Activation and interaction of CD44 and hyaluronan in immunological systems J. Leukoc. Biol. 66,315-321[Abstract]
    25. 25
  25. Grouard, G., Durand, I., Filgueria, L., Banchereau, J., Liu, Y-J. (1996) Dendritic cells capable of stimulating T cells in germinal centers Nature 384,364-367[CrossRef][Medline]
    26. 26
  26. Kiertscher, S. M., Luo, J., Dubinett, S. M., Roth, M. D. (2000) Tumors promote altered maturation and early apoptosis of monocyte-derived dendritic cells J. Immunol. 164,1269-1276[Abstract/Free Full Text]
    27. 27
  27. Ochsenbein, A. F., Slerro, S., Odermatt, B., Pericin, M., Karrer, U., Hermans, J., Hemmi, S., Hengartner, H., Zinkernagel, R. M. (2001) Roles of tumour localization, second signals and cross priming in cytotoxic T-cell induction Nature 411,1058-1064[CrossRef][Medline]
    28. 28
  28. Tinhofer, I., Marschitz, I., Kos, M., Henn, T., Egle, A., Villunger, A., Greil, R. (1998) Differential sensitivity of CD4 and CD8 T lymphocytes to the killing efficacy of Fas (Apo-1/CD95) ligand tumor cells in B chronic lymphocytic leukemia Blood 91,4273-4281[Abstract/Free Full Text]
    29. 29
  29. O’Connell, J., Bennett, M. W., O’Sullivan, G. C., Collins, J. K., Shanahan, F. (1999) The Fas counterattack: cancer as a site of immune privilege Immunol. Today 20,46-52[CrossRef][Medline]
    30. 30
  30. Tanaka, M., Suda, T., Haze, K., Nakamura, N., Sato, K., Kimura, F., Motoyoshi, K., Mizuki, M., Tagawa, S., Ohga, S., Hatake, K., Drummond, A. H., Nagata, S. (1996) Fas ligand in human serum Nat. Med. 2,317-322[CrossRef][Medline]
    31. 31
  31. Gastman, B. R., Atarashi, Y., Reichert, T. E., Saito, T., Balkir, L., Rabinowich, H., Whiteside, T. L. (1999) Fas ligand is expressed on human squamous cell carcinomas of the head and neck, and it promotes apoptosis of T lymphocytes Cancer Res 59,5356-5364[Abstract/Free Full Text]
    32. 32
  32. Villunger, A., Egle, A., Marschitz, I., Kos, M., Bock, G., Ludwig, H., Geley, S., Kofler, R., Greil, R. (1997) Constitutive expression of Fas (Apo-1/CD95) ligand on multiple myeloma cells: a potential mechanism of tumor-induced suppression of immune surveillance Blood 90,12-20[Abstract/Free Full Text]
    33. 33
  33. Buzyn, A., Petit, F., Ostankovitch, M., Figueiredo, S., Varet, B., Guillet, J. G., Ameisen, J. C., Estaquier, J. (1999) Membrane-bound Fas (Apo-1/CD95) ligand on leukemic cells: a mechanism of tumor immune escape in leukemia patients Blood 94,3135-3140[Abstract/Free Full Text]
    34. 34
  34. Lickliter, J. D., Kratzke, R. A., Nguyen, P. L., Niehans, G. A., Miller, J. S. (1999) Fas ligand is highly expressed in acute leukemia and during the transformation of chronic myeloid leukemia to blast crisis Exp. Hematol. 27,1519-1527[CrossRef][Medline]
    35. 35
  35. Matsue, H., Matsue, K., Walters, M., Okumura, K., Yagita, H., Takashima, A. (1999) Induction of antigen-specific immunosuppression by CD95L cDNA-transfected ‘killer’ dendritic cells Nat. Med. 5,930-937[CrossRef][Medline]
    36. 36
  36. Alderson, M. R., Tough, T. W., Davis-Smith, T., Braddy, S., Falk, B., Schooley, K. A., Goodwin, R. G., Smith, C. A., Ramsdell, F., Lynch, D. H. (1995) Fas ligand mediates activation-induced cell death in human T lymphocytes J. Exp. Med. 181,71-77[Abstract/Free Full Text]
    37. 37
  37. Dhein, J., Walczak, H., Baumler, C., Debatin, K-M., Krammer, P. H. (1995) Autocrine T-cell suicide mediated by APO-1/(Fas/CD95) Nature 373,438-441[CrossRef][Medline]
    38. 38
  38. Suda, T., Nagata, S. (1994) Purification and characterization of the Fas-ligand that induces apoptosis J. Exp. Med. 179,873-879[Abstract/Free Full Text]
    39. 39
  39. De Saint-Vis, B., Fugier-Vivier, I., Massacrier, C., Gaillard, C., Vanbervliet, B., Ait-Yahia, S., Caux, C., Banchereau, J., Liu, Y. J., Lebecque, S. (1998) The cytokine profile expressed by human dendritic cells is dependent on cell subtype and mode of activation J. Immunol. 160,1666-1676[Abstract/Free Full Text]
    40. 40
  40. Menetrier-Caux, C., Montmain, G., Dieu, M. C., Bain, C., Favrot, M. C., Caux, C., Blay, J. Y. (1998) Inhibition of the differentiation of dendritic cells from CD34+ progenitors by tumor cells: role of interleukin-6 and macrophage colony-stimulating factor Blood 92,4778-4791[Abstract/Free Full Text]
    41. 41
  41. Gabrilovich, D. I., Chen, H. L., Girgis, K. R., Cunningham, H. T., Meny, G. M., Nadaf, S., Kavanaugh, D., Carbone, D. P. (1996) Production of vascular endothelial growth factor by human tumors inhibits the functional maturation of dendritic cells Nat. Med. 2,1096-1103[CrossRef][Medline]
    42. 42
  42. Chomarat, P., Banchereau, J., Davoust, J., Palucka, K. (2000) IL-6 switches the differentiation of monocytes from dendritic cells to macrophages Nat. Immunol. 1,510-514[CrossRef][Medline]
    43. 43
  43. Hermans, L. F., Daish, A., Yang, J., Ritchie, D. S., Ronchese, F. (1998) Antigen expressed on tumor cells fails to elicit an immune response, even in the presence of increased numbers of tumor-specific cytotoxic T lymphocyte Cancer Res 58,3909-3917[Abstract/Free Full Text]
    44. 44
  44. Lee, P. P., Yee, C., Savage, P. A., Fong, L., Brockstedt, D., Weber, J. S., Johnson, D., Swetter, S., Thompson, J., Greenberg, P. D., Roederer, M., Davis, M. M. (1999) Characterization of circulating T cells specific for tumor-associated antigens in melanoma patients Nat. Med. 5,677-685[CrossRef][Medline]
    45. 45
  45. Groux, H., Bigler, M., de Vries, J. E., Roncarolo, M. G. (1996) Interleukin-10 induces a long-term antigen-specific anergic state in human CD4+ T cells J. Exp. Med. 184,19-29[Abstract/Free Full Text]
    46. 46
  46. Fujii, S-I., Shimizu, K., Shimizu, T., Lotze, M. T. (2001) Interleukin-10 promotes the maintenance of antitumor CD8+ T-cell effector function in situ Blood 98,2143-2151[Abstract/Free Full Text]
    47. 47
  47. Pardoll, D. (2001) T cells and tumours Nature 411,1010-1012[Medline]
    48. 48
  48. Tham, E. L., Shrikant, P., Mescher, M. F. (2002) Activation-induced nonresponsiveness: a Th-dependent regulatory checkpoint in the CTL response J. Immunol. 168,1190-1197[Abstract/Free Full Text]




This Article
Right arrow Abstract Freely available
Right arrow Full Text (PDF)
Right arrow Alert me when this article is cited
Right arrow Alert me if a correction is posted
Services
Right arrow Similar articles in this journal
Right arrow Similar articles in PubMed
Right arrow Alert me to new issues of the journal
Right arrow Download to citation manager
Right arrow reprints & permissions
Citing Articles
Right arrow Citing Articles via Google Scholar
Google Scholar
Right arrow Articles by Fujii, S.-i.
Right arrow Articles by Kawano, F.
Right arrow Search for Related Content
PubMed
Right arrow PubMed Citation
Right arrow Articles by Fujii, S.-i.
Right arrow Articles by Kawano, F.