Ex vivo induction of multiple myeloma–specific cytotoxic T lymphocytes

Multiple myeloma (MM) is currently incurable with conventional chemotherapy or high-dose chemotherapy with autologous stem cell support. Recent studies show that allogeneic stem cell transplantation may achieve prolonged progression-free survival compared with autografting¹  and that donor lymphocyte infusion achieves responses in patients with relapsed MM after bone marrow transplantation.^2,3  These reports suggest that a clinically significant immune-mediated allogeneic graft-versus-MM effect is inducible and provide the framework for development of immune-based therapeutic options.

Dendritic cells (DCs) play a pivotal role in tumor immunity.⁴  We have recently reported that DCs from MM patients may function normally in vitro,⁵  although others have shown decreased responsiveness to CD40 activation.⁶  Immature DCs are characterized by high-endocytic and -macropinocytic activity and low expression of accessory signals for T-cell activation; in contrast, mature DCs have enhanced T-cell stimulatory capacity in association with down-regulation of antigen uptake.⁷  Immature DCs in vivo efficiently phagocytose and process antigen derived from apoptotic bodies and generate peptide epitopes for cross-presentation on human leukocyte antigen (HLA) class I, which stimulates CD8⁺ T cells, after their maturation.^7-9  Importantly, several factors that abrogate DC function in vitro, including transforming growth factor β₁ (TGF-β₁), interleukin 10 (IL-10),⁶  vascular endothelial growth factor (VEGF),¹⁰  and IL-6,¹¹  are also implicated in MM pathogenesis in the bone marrow (BM). We therefore investigated in this study whether serum from BM of patients with MM inhibits both induction and maturation of DCs and delineated the potential role of cytokines mediating this inhibition.

In this study we also attempted to establish an effective DC-based method for generation of autologous cytotoxic T lymphocytes (CTLs) as the framework for vaccination therapy of MM. Although many DC-based vaccination strategies have been attempted, it remains controversial as to which antigen is optimal for DC loading. A single peptide directly binding to HLA molecules on DCs or an antigen expressed on tumor cells are candidate antigens, but these approaches are limited due both to HLA restriction of patients and selectivity of expression of antigen on tumor cells. For example, we established unique HLA-unrestricted CTLs against mucin-1 (MUC1) from patients with MM,^12,13  but the approach is limited since MUC1 is expressed on only 55% ± 12% MM cells.¹⁴  The specific idiotypic determinant of immunoglobulin (Ig) variable regions is an ideal target antigen to overcome these limitations, since it is uniquely expressed on the malignant B-cell clone. After a clinically successful initial report in follicular lymphoma,¹⁵  many studies of vaccination with idiotype-pulsed DCs have been performed in patients with MM.^16-19  Although idiotype-specific CTLs were inducible in these studies, their frequency was less than 50% and associated clinical responses were disappointing. Moreover, approaches using single antigens may also be limited due to cross-reactivity with normal cells or by evolution and escape of antigen-negative tumor cell clones.²⁰ 

In the current study, we generated mature DCs as antigen-presenting cells (APCs), which were pulsed with MM cell apoptotic bodies versus MM cell lysates to optimize antigen-presenting function and overcome limitations of single-antigen approaches. Our study shows that DCs pulsed with MM apoptotic bodies are significantly more effective at inducing CTLs against autologous MM cells than are DCs pulsed with MM cell lysates, providing the framework for related vaccination trials in MM.

Human RPMI 8226 and U266 MM cell lines, as well as natural killer (NK) cell–sensitive human leukemia cell line K562, were obtained from American Type Culture Collection (Rockville, MD). All cell lines were cultured in RPMI 1640 (Mediatech, Herndon, VA) containing 10% fetal bovine serum (FBS; Harlan, Indianapolis, IN), 2 mM L-glutamine, 100 U/mL penicillin, and 100 μg/mL streptomycin (GIBCO, Grand Island, NY).

BM specimens were acquired from patients with MM refractory or resistant to conventional chemotherapy (stage II or III) after obtaining informed consent and approval from the Dana-Farber Cancer Institute institutional review board for these studies. Sera from BM of MM patients were kept at –70°C until use. IL-6, IL-10, TGF-β₁, and VEGF levels in sera were measured by enzyme-linked immunosorbent assay (ELISA; R&D Systems, Minneapolis, MN) according to the manufacturer's instructions. Primary MM cells (> 90% CD138⁺) were purified from BM specimens using RosetteSep negative selection system (StemCell Technologies, Vancouver, BC, Canada) per the manufacturer's protocol.

Immature DCs were induced from healthy donor peripheral blood mononuclear cells (PBMCs) or BMMCs from MM patients as previously described⁵  with some modification. Briefly, mononuclear cells (MNCs) separated by Ficoll-Paque (Pharmacia, Piscataway, NJ) were suspended in RPMI 1640 medium containing 5% heat-inactivated human serum type AB (Sigma, St Louis, MO) or autologous serum and then incubated in 6-well culture plates (Becton Dickinson Labware, Franklin Lakes, NJ) for 2 hours at 37°C. After removing nonadherent cells, adherent cells (> 70% CD14⁺) were cultured in fresh medium containing 10% heat-inactivated human AB serum or autologous serum, recombinant human granulocyte-macrophage colony-stimulating factor (GM-CSF; 1000 U/mL; Immunex, Seattle, WA), and IL-4 (10 ng/mL; R&D Systems) for 5 days. In experiments to examine the effect of patient's serum on induction of DCs, 20% serum from the BM of patients was added instead of 10% AB serum, with or without neutralizing anti–IL-6 antibodies (Abs), anti-VEGF Abs (5 μg/mL; R&D System), or normal goat IgG (5 μg/mL; Oncogene Research Products, Boston, MA) as control. Recombinant human IL-6 (20 ng/mL) or VEGF (20 ng/mL; R&D Systems) in 10% AB serum was added to define its effects on DC production. Immature DCs were cocultured for 12 to 18 hours with irradiated (30 Gy) MM cell lines (2:1) or MM patient cells (5:1) and then cultured for 3 days in the presence of tumor necrosis factor α (TNF-α;10or 20 ng/mL; R&D Systems) to induce maturation. MM cell lysates, produced by repeating freezing and thawing 3 times, were pulsed to immature DCs at the same cell equivalent ratios as apoptotic bodies. MNCs from the same donors or patients were then cocultured with mature DCs at a ratio of 10:1 to 20:1 in AIM-V medium (Invitrogen, Carlsbad, CA) and stimulated with the apoptotic body–pulsed or cell lysate–pulsed DCs weekly. Seven days after the second stimulation, cytotoxicity of the CTLs was examined as described in “Cytotoxicity assay.”

Cell surface phenotype was analyzed by flow cytometry using anti-CD1a, -CD14, -CD80, and -CD83 antibodies (Beckman Coulter, Fullerton, CA) for DCs, as well as anti-CD3, -CD4, -CD8, and -CD56 antibodies (Beckman Coulter) for CTLs. Cells were washed with phosphate-buffered saline (PBS) and incubated with fluorescein isothiocyanate (FITC)–conjugated or phycoerythrin (PE)–conjugated antibody for 20 minutes on ice. After washing with cold PBS once, cells were fixed with 1% paraformaldehyde (Sigma) and analyzed using a flow cytometer (Coulter Epics XL; Beckman Coulter).

Dual staining with annexin-V–FITC and propidium iodide (PI) was used to detect apoptosis, as in our prior study.²¹  Briefly, 1 × 10⁶ MM cells were irradiated (30 Gy) and resuspended in 100 μL HEPES (N-2-hydroxyethylpiperazine-N′2-ethanesulfonic acid) buffer containing annexin-V–FITC and PI (annexin-V–FLUOS Staining Kit; Roche Diagnostics, Manheim, Germany) after incubation for 0, 2, 4, 6, 8, and 12 hours. Early apoptosis is defined by annexin-V–FITC⁺ and PI^– staining and detected by dual-fluorescence flow cytometry.

MM cells were stained red with Vybrant Dil Cell-Labeling Solution (Molecular Probes, Eugene, OR) according to the manufacturer's protocol and then irradiated (30 Gy) using Gammacell 1000 (Atomic Energy of Canada, Mississauga, ON, Canada). Immature DCs were stained green with Vybrant DiO Cell-Labeling Solution (Molecular Probes) and then cocultured with irradiated red MM cells at a ratio of 1:1 for 0, 2, 4, 6, and 8 hours at 37°C or 4°C. Phagocytosis of apoptotic bodies by immature DCs was defined by the percentage of double-positive cells detected by dual-fluorescence flow cytometry, as described previously.²² 

To examine the function of DCs cocultured with apoptotic bodies as APCs, autologous T-cell proliferation was measured by ³[H] thymidine incorporation assay, as previously described.²³  Briefly, T cells (> 95% CD3⁺), obtained from PBMCs by high-affinity negative selection using T-cell enrichment columns (R&D Systems), were incubated in triplicate in 96-well round-bottom culture plates (2 × 10⁵ cells/well) for 5 days with irradiated (15 Gy) autologous DCs at various T-DC ratios. ³[H]thymidine (1.0 μCi [0.037 MBq]) was added to each well for the last 12 hours of 5-day cultures. Cells were then harvested onto glass filters using an automatic cell harvester (Cambridge Technology, Cambridge, MA) and counted using a MicroBeta Trilux counter (Wallac, Gaithersburg, MD).

Cytotoxicity of CTLs against MM cell lines or K562 cells was assessed with ⁵¹Cr-release assay, as previously described.¹²  Briefly, 5 × 10³ target cells labeled with Na₂CrO₄ (Perkin Elmer Life Sciences, Boston, MA) were cocultured with effector cells in triplicate at various effector-target ratios in 96-well round-bottom tissue culture plates (Corning Inc, Corning, NY) for 4 hours at 37°C. Supernatants were then harvested with Supernatant Collection System (Molecular Devices, Sunnyvale, CA), and radioactivity was counted using a Wizard 3 gamma counter (Wallac). Percent-specific lysis was calculated as follows: ([experimental release – spontaneous release]/[maximum release – spontaneous release]) × 100. Spontaneous release was measured by incubating target cells in the medium alone, and maximum release was obtained by adding 1% Triton X-100 (Sigma) to target cells. For target blocking assays, target cells were added after incubation with blocking antibody against HLA class I (clone W6/32; Research Diagnostics, Flanders, NJ), HLA class II (clone G46-6; BD PharMingen, San Diego, CA), or normal mouse IgG (Oncogene Research Products) at a concentration of 10 μg/mL for 30 minutes at room temperature.

Specific cytotoxicity against primary MM cells was assessed by colorimetric lactate dehydrogenase (LDH) assay (CytoTox 96 Non-Radioactive Cytotoxicity Assay; Promega, Madison, WI), per manufacturer's instruction and as described for the ⁵¹Cr-release assay.

Statistical significance of differences observed in expression of surface antigens, proliferation of autologous T cells, and cytotoxicity of CTLs was determined using the Mann-Whitney U test. The minimal level of significance was P < .05.

We first investigated the effects of serum from BM of MM patients on the induction of DCs especially focusing on VEGF and IL-6, which play important roles in MM pathogenesis.^24-26  Adherent cells (> 70% CD14⁺) from a healthy donor were cultured in the presence of GM-CSF (1000 U/mL), IL-4 (10 ng/mL), and either 20% human AB serum or BM sera from 10 MM patients for 5 days to obtain immature DCs. Cells were then cultured in the presence of TNF-α (10 ng/mL) to induce their maturation for 3 additional days. Phenotypic analysis included staining of DCs for CD14 (monocyte/DC precursor), CD1a and CD80 (immature and mature DC), and CD83 (mature DC). Sera from MM patients' BM were analyzed for VEGF, IL-6, TGF-β₁, and IL-10 by ELISA. As shown in Table 1, mean concentration of IL-6 and VEGF in sera from patients' BM is higher than that of AB serum used in this experiment, and DCs cultured with MM patients' BM sera express high levels of CD14 but low levels of CD1a, CD80, and CD83. There were no differences in concentrations of TGF-β₁ and IL-10 (data not shown). These results indicate that sera from BM of patients with MM, which contain high concentrations of IL-6 and VEGF, inhibit induction and maturation of DCs from adherent cells.

To define whether this inhibitory effect in BM serum was mediated via IL-6 (a major growth and survival factor for MM cells^24,25 ) and VEGF (which stimulates MM cell growth and migration²⁶  as well as BM angiogenesis²⁷ ) we repeated these experiments in the presence of neutralizing anti-VEGF Abs (5 μg/mL) and/or anti–IL-6 Abs (5 μg/mL). Neither Ab affected the induction and maturation of DCs in the presence of human AB serum (data not shown); however, the addition of either of these Abs reduced CD14 and increased CD1a expression on immature DCs in the presence of MM patients' BM sera (Figure 1A). Moreover, addition of these Abs increased expression of CD83 on mature DCs (Figure 1B). These results suggest that VEGF and IL-6 in MM patients' BM sera mediate, at least in part, an inhibitory effect on the induction and maturation of DCs.

We next examined the function of these DCs, induced in the presence of MM patients' BM sera with or without anti-VEGF Abs and anti–IL-6 Abs in an autologous T-cell proliferation assay. T cells (> 95% CD3⁺) were cocultured in AIM-V medium with DCs derived from irradiated (15 Gy) autologous adherent cells by culture for 5 days with GM-CSF (1000 U/mL), IL-4 (10 ng/mL), and 20% MM patients' BM sera or AB serum, followed by incubation with TNF-α (10 ng/mL) for 3 days. Proliferation was measured by ³[H] thymidine incorporation assay during the last 12 hours of 5-day cultures. As shown in Figure 1C, DCs cultured in the presence of MM patients' BM sera stimulated significantly less T-cell proliferation (stimulation index [SI] = 6.8 ± 1.7 at a T-DC ratio of 30:1) than DCs cultured in the presence of AB serum (SI = 36.3 ± 1.4, *P = .01). Importantly, addition of anti-VEGF and anti–IL-6 Abs neutralized these inhibitory effects on T-cell proliferation (SI = 27.6 ± 4.2, **P = .01), reflecting corresponding changes in DC phenotype.

To confirm this inhibitory effect of MM patients' BM sera, we next examined the effect of exogenous VEGF and IL-6 on DC induction. Healthy donor adherent cells were cultured in 10% human AB serum with GM-CSF and IL-4 for 5 days with or without VEGF (20 ng/mL) or IL-6 (20 ng/mL). TNF-α (10 ng/mL) was then added for 3 days and flow cytometric analysis was then performed. As shown in Figure 2A, CD14 expression increased, while both CD1a and CD83 decreased, on cells cultured with VEGF or IL-6 compared with cells cultured in media alone (*, ** P = .02, *** P = .046). We similarly studied the function of DCs induced in the presence of these cytokines using an autologous T-cell proliferation assay. Importantly, DCs generated in the presence of VEGF or IL-6 only weakly stimulated autologous T cells (SI = 45.0 ± 1.6 in control versus 31.9 ± 0.9 with VEGF and 20.7 ± 1.0 with IL-6 at a T-DC ratio of 30:1, * P = .02; Figure 2B). These results show that VEGF and IL-6 inhibit the maturation and function of DCs, confirming that these cytokines can account, at least in part, for immunosuppression in MM patients. Moreover, they further support the potential utility of generating anti-MM CTLs ex vivo.

We next attempted to establish CTLs that recognize and lyse MM cells ex vivo. DCs can acquire antigen from apoptotic cells and present tumor antigen on HLA class I molecules, thereby inducing antigen-specific CTL responses,^8,28  and we therefore pulsed immature DCs with MM apoptotic bodies. MM cells were first irradiated (30 Gy) and stained with annexin V–FITC and PI. As shown in Figure 3A, the annexin-V–FITC⁺ and PI^– cell fraction gradually increased from 0 to 12 hours, with transition to annexin-V⁺/PI⁺ cells. Trypan blue staining showed fewer than 10% viable cells at 48 hours and no viable cells at 72 hours after irradiation (30 Gy; data not shown). We next monitored uptake of irradiation-induced apoptotic bodies by immature DCs in a phagocytosis assay. Immature DCs (CD14^–/+, CD1a^+/–, CD83^–) were prepared by culturing adherent mononuclear cells with GM-CSF and IL-4 for 5 days and stained green; MM cells were red-stained and irradiated. Green DCs and red MM cells were cultured at a ratio of 1:1 at either 37°C or 4°C. After 0, 2, 4, 6, and 8 hours, these cocultured cells were analyzed using flow cytometry with quantification of phagocytic uptake reflected as double-positive cells. As shown in Figure 3B, immature DCs take up irradiated MM cells within several hours at 37°C but not 4°C. Uptake of apoptotic cells was confirmed by fluorescence microscopy (data not shown). These data indicate that immature DCs effectively phagocytose MM apoptotic cells at 37°C.

We next examined the ability of DCs pulsed with MM apoptotic cells and DCs pulsed with MM cell lysates to induce MM CTLs. Immature DCs were derived from adherent cells of HLA-A2–positive healthy donors by culturing with GM-CSF, IL-4, and 10% autologous serum for 5 days and then cocultured with HLA-A2–positive U266 cells that had been irradiated 4 hours earlier at a ratio of 2:1. After overnight incubation, TNF-α (20 ng/mL) was added to media and incubated for 3 days to induce maturation of DCs (CD14^–, CD83⁺). T cells from the same donors were cocultured with these irradiated (15 Gy) mature DCs in AIM-V medium, and proliferation was measured by ³[H]thymidine incorporation for the last 12 hours of 5-day cultures. As shown in Figure 4A, DCs cocultured with MM apoptotic bodies stimulated significantly (* P = .02) greater autologous T-cell proliferation (SI = 23.0 ± 2.0 at T/DC of 360:1) than MM cell lysate–pulsed DCs (SI = 13.9 ± 1.6), DCs alone (SI = 9.3 ± 1.1), or apoptotic MM cells alone (SI = 5.6 ± 0.4). Importantly, CTLs established by 2 weekly stimulations with MM apoptotic body–pulsed DCs efficiently lysed those MM cells used as stimulators, with only weak cytotoxicity in CTLs generated using other stimulators (Figure 4B). These data indicate that MM apoptotic body–pulsed DCs induce functional CTLs against MM cells.

To characterize reactivity of CTLs induced by repeated stimulations with mature DCs pulsed with apoptotic MM bodies, we examined their cytotoxicity against cells other than those used for stimulation. CTLs induced in this manner from healthy donors were cocultured with ⁵¹Cr-labeled target cells for 4 hours, and radioactivity of supernatants was compared with maximal release. As shown in Figure 5A, CTLs efficiently lysed those MM cells used for pulsing DCs, but not unrelated MM cells and an NK cell–sensitive cell line. We next investigated the mechanism of recognition of target cells in a ⁵¹Cr-release assay by incubating target cells with neutralizing antibodies to HLA class I or class II. Cytotoxicity of CTLs against MM cells was significantly inhibited by anti–HLA class I Abs but not anti–HLA DR or control Abs (Figure 5B). These results indicate that cytotoxicity of CTLs is specific for target MM cells and restricted to HLA class I. Phenotypic analysis revealed that CTLs were mainly CD3⁺ and CD8⁺, with only a minor population of CD56⁺ cells (Figure 5C). This result confirmed that the cytotoxicity was mediated via CTLs but not NK cells.

We finally examined induction of CTLs that can specifically lyse autologous MM cells. Immature DCs were induced from adherent cells in BMMCs from MM patients by culturing in medium with 10% AB serum, GM-CSF, and IL-4 for 5 days. These immature DCs were cocultured overnight with irradiated MM cells (5:1) obtained from the same patient. TNF-α was then added for 3 days to induce DC maturation. BM or peripheral MNCs from the same patient were cocultured with mature DCs at a ratio of 10:1 to 20:1 in AIM-V medium. Seven days after second stimulation, cytotoxicity of CTLs against primary MM cells was analyzed using the LDH-release assay due to low ⁵¹Cr uptake by patient MM cells. As shown in Figure 6A, CTLs stimulated by mature DCs pulsed with patient MM apoptotic bodies effectively lysed autologous MM cells but not an unrelated MM cell line or NK cell–sensitive cells. In addition, CTLs from patient (Pt) 1 showed only slight cytotoxicity against RPMI 8226 cells, even though both produce λ light chain, indicating that CTLs do not have cross-reactivity with this isotype-matched cell line. Flow cytometric analysis revealed that CTLs consisted primarily of CD8⁺ T cells with only a small population of CD4⁺ T cells and CD56⁺ NK cells (Figure 6B). To further characterize cytotoxicity of CTLs induced from MM patients we next established CTLs from HLA-A2–positive patients by repeated stimulation with apoptotic body–pulsed DCs and examined for specific lysis in the presence or absence of neutralizing antibodies to HLA class I or class II. As shown in Figure 6C, cytotoxicity of CTLs against autologous primary MM cells was significantly inhibited by anti–HLA class I Abs but not anti–HLA DR or control Abs. These results indicate that cytotoxicity of CTLs is restricted to HLA class I. We also investigated their cytotoxicity against HLA-A2–positive irrelevant U266 cells, NK cell–sensitive K562 cells, autologous PBMCs, and primary MM cells using the LDH assay. As shown in Figure 6D, CTLs lysed autologous primary MM cells, but not U266, K562, or autologous PBMCs, showing that cytotoxicity of CTLs is specific for target MM cells.

In the present study, we show that sera from BM of patients with MM inhibit both induction and maturation of DCs, mediated at least in part by VEGF and IL-6, suggesting the potential utility of generating anti-MM CTLs ex vivo. We therefore establish a strategy to induce CTLs recognizing autologous MM cells that is widely applicable and does not depend upon expression of either selective tumor antigens or patient HLA type.

We first demonstrate that VEGF and IL-6 in sera from BM of MM patients can inhibit the induction of DCs from adherent mononuclear cells, evidenced by their weak APC function as stimulators of T-cell proliferation. Our results confirm and extend previous investigations. Specifically, differentiation of CD34⁺ cells to DCs is inhibited by exogenous IL-6,¹¹  or by IL-6 produced by renal cell and colon carcinoma cells,^29,30  due to macrophage rather than DC differentiation.³¹  VEGF similarly inhibits the maturation of DCs in mice.¹⁰  IL-6 induces growth, survival, and drug resistance in MM cells^24,25,32,33 ; VEGF triggers growth and migration of MM cells,²⁶  stimulates BM angiogenesis,²⁷  and augments IL-6 production in bone marrow stromal cells (BMSCs).³⁴  In addition to these effects on MM cells and BMSCs in the BM microenvironment, the present results further suggest that both cytokines contribute to immune deficits characteristic of MM. Conversely, suppression of VEGF and IL-6 may enhance anti-MM immunity. For example, our studies show that thalidomide and immunomodulatory derivatives (IMiDs) act not only directly on MM cells³⁵  but also in the BM microenvironment to inhibit up-regulation of IL-6 and VEGF secretion triggered by binding of MM cells to BMSCs.³⁶  Importantly, these drugs augment in vitro and in vivo NK-cell activity against MM cells.²³  Given that DCs directly trigger NK-cell function,³⁷  studies are ongoing to define the effects of these agents on DC induction, maturation, and function.

We have shown that TGF-β₁ is secreted by MM cells and can augment IL-6 secretion in BMSCs.³⁸  Others have reported that TGF-β₁ and IL-10 inhibit peripheral DC function in MM by down-regulating expression of CD80.⁶  We therefore examined the role of these cytokines in mediating the DC inhibition observed. Our results show that concentrations of TGF-β₁ and IL-10 in MM patients' BM sera are equivalent to those in AB serum or in normal peripheral blood,⁶  0 to 20 ng/mL and 0 to 30 ng/mL, respectively. Although it remains possible that TGF-β₁ and IL-10 or other humoral factors contribute to inhibiting DC function in MM, since even patients' serum with low levels of VEGF and IL-6 blocked DC maturation, generation of DCs ex vivo can overcome this inhibitory effect.

We next carried out in vitro studies to provide the framework for a vaccination therapy protocol using DCs and apoptotic tumor bodies to establish MM specific CTL ex vivo. Although DCs are the most effective APCs for vaccination strategy,⁴  it remains controversial as to which antigens are optimal for DC loading. DC pulsing with an HLA-binding single peptide from a tumor antigen is restricted by HLA type of patients and candidate MM antigens. Idiotype protein is uniquely produced by the malignant clone of MM cells and therefore a potential ideal antigen for vaccination. Based on clinical responses after idiotype vaccination in follicular lymphoma,¹⁵  several clinical trials of vaccination with idiotype-pulsed dendritic cells have been performed in patients with MM.^16-19  In these studies, idiotype-specific CTLs are inducible in vitro; however, clinical responses after injecting idiotype-pulsed immature DCs are disappointing,¹⁹  possibly due to inhibition of DC maturation by inhibitory VEGF and IL-6. In this regard, vaccination using DCs that have matured and cannot revert to immature DCs or monocytes³⁹  appears to be promising.⁴⁰ 

Compared with vaccinations using a single tumor antigen or molecule, strategies pulsing DCs with tumor cell apoptotic bodies have the potential advantage of presenting known, as well as unknown, tumor antigens. We therefore examined MM patient apoptotic body–pulsed DCs, as well as MM cell lysate–pulsed DCs, as stimulators of MM specific autologous CTLs. Uptake of apoptotic bodies by DCs is required for cross-presentation of antigens to induce HLA class I–restricted CD8⁺ T cells.⁸  We therefore pulsed apoptotic bodies to immature DCs, which phagocytose apoptotic bodies more efficiently than mature DCs.⁷  Apoptotic bodies, produced by irradiation of MM cells at doses (30 Gy) sufficient to prevent MM cell survival, were then cultured with DCs for 3 days. We determined that addition of apoptotic bodies to DCs were optimal at 4 hours after irradiation when the apoptotic cell population was increasing and there were few necrotic cells, which might either induce maturation of DC,⁴¹  resulting in ineffective uptake of apoptotic bodies,⁷  or stimulate both CD4⁺ and CD8⁺ T cells.⁴²  We show that MM apoptotic body–pulsed DCs are more effective at triggering autologous MM specific CTLs than MM lysate–pulsed DCs. Immature DCs are known to effectively phagocytose apoptotic bodies via various receptors including phosphtidylserine receptor, vitronectin receptor,⁴³  CD36,⁴⁴  and α_Vβ₅ integrin within a few hours after pulsing⁷  and then cross-present tumor antigen on their HLA class I molecule.⁸  In contrast, cell lysates are captured by DCs via macropinocytosis and mannose receptor and are then transferred to major histocompatibility complex (MHC) class II–rich vesicles.³⁹  These differences may account, at least in part, for the superiority of apoptotic body–pulsed DCs in our study and are consistent with other recent studies showing superiority of apoptotic bodies versus cell lysates in promoting cross-priming of CTLs in squamous-cell carcinoma²⁸  and pancreatic cancer,⁴⁵  as well as in virus-infected cells.⁸ 

In summary, in this study we provide the rationale for ex vivo generation of DCs, due to inhibitory factors in MM BM serum for their induction and maturation, and also develop a protocol to generate CTLs recognizing autologous MM cells using these DCs pulsed with MM apoptotic bodies ex vivo. This strategy is broadly applicable since it is not dependent upon expression of selective tumor antigens or patient HLA–type restriction, and therefore provides the framework for novel immunotherapy to improve patient outcome in MM.